To: All members of Congress
From: Americans for Safe Access Date: February 22, 2016 Re: The Dangers and Consequences of Misinformation on Marijuana (cannabis)
______________________________________________________________________
Forty-‐four states now allow patients under their physician's care to use medical cannabis (marijuana) in some form, and most of the rest of the states are discussing medical cannabis in their current legislative sessions. In addition, three cannabis related budget amendments and four bills have been introduced so far in the 115th Congress. We know that you rely on the Drug Enforcement Administration (DEA) to provide current and accurate information when you are making decisions about cannabis policy. Americans for Safe Access (ASA) has prepared this memo to inform you about four important changes in the DEA’s positions on medical cannabis that could have an impact on your policy making decisions this session. In August 2016, the DEA issued the “Denial of Petition to Initiate Proceedings to Reschedule Marijuana,” (see enclosure A) in response to Washington and Rhode Island’s attempts to reschedule cannabis. While the DEA did not move forward with rescheduling cannabis, the report did clarify four important misconception about medical cannabis: that cannabis was a “gateway drug”, and that it caused cognitive decline, psychosis and lung cancer based on new research (see enclosures F i-‐viii). This new report contradicted two previous reports from the DEA, “The Dangers and Consequences of Marijuana Abuse” and “Drugs of Abuse” that are often cited by policy makers. In response to the DEA’s new report, Americans for Safe Access (ASA) released “The DEA’s Denial of Existing Medical Cannabis Research” (see enclosure B) which called upon the Department of Justice (DOJ) and the DEA to update contradicting information in the DEA publications. On December 5th, 2016, utilizing rights granted to our members and patients under the Information Quality Act (IQA), ASA filed a petition with the DOJ (see enclosure C) requesting the DEA to remove or update several items from their website which contained this misinformation. ASA’s “IQA Request for Correction of Information Disseminated by DEA Regarding Marijuana (Cannabis)” identified 25 violations of the IQA on the DEA’s website and publications and included suggested changes based on the DEA’s finding from their August report. Despite the DOJ Information Quality Guidelines requirements for a 60-‐calendar day response, neither the DEA nor the DOJ has responded directly to ASA’s request (see enclosure D). The DEA has removed the document, “The Dangers and Consequences of Marijuana Abuse,” which contained the majority of the inaccurate statements outlined in ASA’s Request, but there is still more that needs to be corrected, and ASA is working to make sure it is. While we are waiting for the final response from DEA on their plans to update their information on their website, we know that policy makers are being asked to make decisions on cannabis policy now. It is our hope that the enclosed documents will provide a better understanding of the status of science on these important matters. Enclosures:
A. DEA Report: “Denial of Petition to Initiate Proceedings to Reschedule Marijuana,” (August 2016). B. ASA Report: “The DEA’s Denial of Existing Medical Cannabis Research” (August 2016) C. ASA’s “IQA Request for Correction of Information Disseminated by DEA Regarding Marijuana (Cannabis)” D. ASA's IQA Deadline Letter to DEA E. Relevant research studies from ‘Denial of Petition to Initiate Proceedings to Reschedule Marijuana” I. Pope, H. G., Gruber, A. J., Hudson, J. I., Huestis, M. A., & Yurgelun-‐Todd, D. (2001). Neuropsychological performance in long-‐term cannabis users. Archives of General Psychiatry, 58(10), 909–915. II. Degenhardt, L., Chiu, W. T., Conway, K., Dierker, L., Glantz, M., Kalaydjian, A., et al. (2009). Does the “gateway” matter? Associations between the order of drug use initiation and the development of drug dependence in the National Comorbidity Study Replication. Psychological Medicine, 39(1), 157–
167. http://doi.org/10.1017/S0033291708003425 III. Fergusson, D. M., Horwood, L. J., & Ridder, E. M. (2005). Tests of causal linkages between cannabis use and psychotic symptoms. Addiction, 100(3), 354–366. IV. Degenhardt, L., Hall, W., & Lynskey, M. (2003). Testing hypotheses about the relationship between cannabis use and psychosis. Drug Alcohol Depend, 71(1), 37–48. V. Kandel, D. B., & Chen, K. (2000). Types of marijuana users by longitudinal course. J Stud Alcohol, 61(3), 367–378. http://doi.org/10.15288/jsa.2000.61.367 VI. Lee, M. H. S., & Hancox, R. J. (2011). Effects of smoking cannabis on lung function. Expert Review of Respiratory Medicine, 5(4), 537–46– quiz 547. http://doi.org/10.1586/ers.11.40 VII. Kuepper, R., van Os, J., Lieb, R., Wittchen, H. U., Hofler, M., & Henquet, C. (2011). Continued cannabis use and risk of incidence and persistence of psychotic symptoms: 10 year follow-‐up cohort study. BMJ, 342,, 738–738. http://doi.org/10.1136/bmj.d738 VIII. Minozzi, S., Davoli, M., Bargagli, A. M., Amato, L., Vecchi, S., & Perucci, C. A. (2010). An overview of systematic reviews on cannabis and psychosis: Discussing apparently conflicting results. Drug and Alcohol Review, 29(3), 304–317. http://doi.org/10.1111/j.1465-‐3362.2009.00132.x Enclosure A: “Denial of Petition to Initiate Proceedings to Reschedule Marijuana” (August 2016) The Dangers and Consequences of Misinformation on Marijuana February 22, 2017
Xqn/! 92!
Htkfc{-!
Pq/!267!
Cwiwuv! 23-! 3127!
Rctv!KX!
Fgrctvogpv! qh! Lwuvkeg!
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Ftwi! Gphqtegogpv! Cfokpkuvcvkqp!
32! EHT! Ejcrvgt! KK! cpf! Rctv! 2412!
Fgpkcn! qh! Rgvkvkqp! Vq! Kpkvkcvg! Rtqeggfkpiu! Vq! Tguejgfwng! Octklwcpc=!
Rtqrqugf! Twngu! cpf! Crrnkecvkqpu! Vq! Dgeqog! Tgikuvgtgf! Wpfgt! vjg!
Eqpvtqnngf! Uwduvcpegu! Cev! Vq! Ocpwhcevwtg! Octklwcpc! Vq! Uwrrn{!
Tgugctejgtu! kp! vjg! Wpkvgf! Uvcvgu=! Rqnke{! Uvcvgogpv!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11112
Hov!5828
Uhov!5828
G<^HT^HO^23CWR3/UIO
23CWR3
64799!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
tr~||t}spx~}.wp.wt.WWb.q|xts.~.
wt.STP.x.t}r{~ts.xw.wx.{tt=.
Qpts.~}.wt.WWb.t
p{px~}.p}s.p{{.~wt.
Ftwi!Gphqtegogpv!Cfokpkuvtcvkqp!
t{t
p}.spp;.wt.STP.wp.r~}r{sts.wp.
wtt.x.}~.qp}xp{.t
xst}rt.wp.
32!EHT!Ejcrvgt!KK!
|pxyp}p.w~{s.qt.t|~
ts.u~|.brwts{t.
]Fqemgv!Pq/!FGC537_!
X=.P.s~r|t}.tpts.q.wt.STP.
psstx}v.wtt.|ptxp{.x}.stpx{.p{~.x.
Fgpkcn!qh!Rgvkvkqp!Vq!Kpkvkcvg!
t}r{~ts=.X}.w~;.|pxyp}p.r~}x}t.~.
Rtqeggfkpiu!Vq!Tguejgfwng!Octklwcpc! |tt.wt.rxtxp.u~.brwts{t.X.r~}~{.}st.
wt.RbP.qtrptI.
CIGPE[<!Sv.T}u~rt|t}.
[email protected]!jcu!c!jkij!rqvgpvkcn!hqt!
Ps|x}xpx~};.Stp|t}.~u.Yxrt=.
cdwug/!cwt.WWb.t
p{px~}.p}s.wt.
CEVKQP<!St}xp{.~u.txx~}.~.x}xxpt.
pssxx~}p{.spp.vpwtts.q.wt.STP.w~.
~rttsx}v.~.trwts{t.|pxyp}p=.
wp.|pxyp}p.wp.p.wxvw.~t}xp{.u~.pqt=.
6A7.Octklwcpc!jcu!pq!ewttgpvn{!ceegrvgf!
UWOOCT[<!Q.{tt.spts.Y{.@H;.A?@E.
ogfkecn!wug!kp!vtgcvogpv!kp!vjg!Wpkvgf!
wt.Sv.T}u~rt|t}.Ps|x}xpx~}.
Uvcvgu/!Qpts.~}.wt.tpq{xwts.ux
t<p.t.
6STP7.st}xts.p.txx~}.~.x}xxpt.
u~.|pzx}v.rw.stt|x}px~};.|pxyp}p.
wp.}~.rt}{.prrtts.|tsxrp{.t .
{t|pzx}v.~rttsx}v.~.trwts{t.
qtrptI.P.stpx{ts.x}.wt.WWb.t
p{px~};.
|pxyp}p=.Qtrpt.wt.STP.qt{xt
t.
wt.sv .rwt|x.x.}~.z}~}.p}s.
wp.wx.|pt.x.~u.pxr{p.x}tt.
~.|t|qt.~u.wt.q{xr;.wt.pvt}r.x. t~srxq{tJ.wtt.pt.}~.pstpt.put.
sxtJ.wtt.pt.}~.pstpt.p}s.t{{<.
q{xwx}v.qt{~.wt.{tt.t}.~.wt.
r~}~{{ts.sxt.~
x}v.tuuxrprJ.wt.sv.
txx~}t.wxrw.st}xts.wt.txx~};.
x.}~.prrtts.q.p{xuxts.ttJ.p}s.wt.
p{~}v.xw.wt.~x}v.
rxt}xuxr.t
xst}rt.x.}~.xst{.p
px{pq{t=.
s~r|t}px~}.wp.p.pprwts.~.wt.
6B7.Octklwcpc!ncemu!ceegrvgf!uchgv{!hqt!wug!
{tt=.
wpfgt!ogfkecn!uwrgtxkukqp/!P.tt};.wtt.
FCVGU<!Pv.@A;.A?@E=.
pt.}~.|pxyp}p.~sr.p~
ts.q.wt.
d=b=.U~~s.p}s.Sv.Ps|x}xpx~}.6USP7;.
HQT HWTVJGT KPHQTOCVKQP EQPVCEV<!
}~.x.|pxyp}p.}st.p.]t.Sv.
\xrwpt{.Y=.[tx;.^uuxrt.~u.Sx
tx~}.
P{xrpx~}.6]SP7.t
p{px~}.p.wt.USP.u~.
R~}~{;.Sv.T}u~rt|t}.
Ps|x}xpx~}J.\px{x}v.PsstI.GF?@. p}.x}sxrpx~}=.cwt.WWb.t
p{px~}.pt.
\~xtt.Sx
t;.bx}vuxt{s;.exvx}xp. wp.|pxyp}p.s~t.}~.wp
t.p.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
[email protected]{t•w~}tI.6A?A7.DHG¤EG@A=.
d}xts.bpt.~.p.rt}{.prrtts.|tsxrp{.
UWRRNGOGPVCT[ KPHQTOCVKQP<!
t.xw.t
tt.txrx~}=.P.wx.x|t;.wt.
Y{.@H;.A?@E.
z}~}.xz.~u.|pxyp}p.t.wp
t.}~.qtt}.
w~}.~.qt.~txvwts.q.trxuxr.qt}tux.
Stp.\=.apx|~}s~.p}s.\=.X}{ttI.
^}.]~
t|qt.B?;.A?@@;.~.tstrt~;. x}.t{{<r~}~{{ts.r{x}xrp{.xp{.wp.
cwt.W~}~•pq{t.[x}r~{}.S=.Rwputt.p}s.cwt.
rxt}xuxrp{{.t
p{pt.put.p}s.tuuxrpr=.
W~}~pq{t.Rwxx}t.^=.Vtv~xt;.txx~}ts.
cwt.p~.|p}spt.~u.cx{[email protected]}xts.
wt.Sv.T}u~rt|t}.Ps|x}xpx~}.6STP7. bpt.R~st;.btrx~}[email protected]@.d=b=R=.G@A6q77.
~.x}xxpt.{t|pzx}v.~rttsx}v.}st.wt. x.sx~xx
t=.R~}vt.tpq{xwts.~}{.~}t.
trwts{x}v.~
xx~}.~u.wt.R~}~{{ts.
rwts{t;.brwts{t.X;.u~.sv.~u.pqt.xw.
bqp}rt.Pr.6RbP7=.btrxuxrp{{;.~.
}~.rt}{.prrtts.|tsxrp{.t.x}.
tstrt~.txx~}ts.wt.STP.~.wp
t.
tp|t}.x}.wt.d}xts.bpt .p}s.{prz.~u.
|pxyp}p.p}s.t{pts.xt| .t|~
ts.u~|. prrtts.put.u~.t.= = =.}st.|tsxrp{.
brwts{t.X.~u.wt.RbP.p}s.trwts{ts.p.
t
xx~}= [email protected]=b=R=.G@A6q7=.
|tsxrp{.rp}}pqx.x}.brwts{t.XX=.
P{w~vw.wt.WWb.t
p{px~}.p}s.p{{.~wt.
h~.tstrt~.ttts.wp.wt.STP.
t{t
p}.spp.{tps.~.wt.r~}r{x~}.wp.
t|~
t.|pxyp}p.p}s.t{pts.xt|.u~|.
|pxyp}p.|.t|px}.x}.rwts{t.X;.x.
brwts{t.X.qpts.~}.wtx.ptx~}.wpI.
w~{s.p{~.qt.}~ts.wp;.x}.
xt.~u.d}xts.
[email protected]}}pqx.wp.prrtts.|tsxrp{.t.x}.
bpt.~q{xvpx~}.}st.x}t}px~}p{.sv.
wt.d}xts.bptJ.
r~}~{.tpxt;.|pxyp}p.rp}}~.qt.{prts.
6A7.Rp}}pqx.x.put.u~.t.}st.|tsxrp{.
x}.p.rwts{t.{t.txrx
t.wp}.rwts{t.
t
xx~}J.
XX=.cwx.x.t{px}ts.x}.stpx{.x}.
6B7.Rp}}pqx.u~.|tsxrp{.~t.wp.p.
prr~|p}x}v.s~r|t}.x{ts._t{x|x}p.
t{px
t{.{~.~t}xp{.u~.pqt;.ttrxp{{.
]~t.atvpsx}v.ctp.R~}xstpx~}= .
x}.r~|px~}.xw.~wt.brwts{t.XX.sv=.
Prr~sx}v{;.p}s.p.t.u~w.x}.stpx{.x}.
X}.prr~sp}rt.xw.wt.RbP.trwts{x}v.
~
xx~};.put.vpwtx}v.wt.}trtp.spp;. wt.prr~|p}x}v.WWb.p}s.STP.s~r|t};.
wtt.x.}~.p~.qpx.}st.wt.RbP.u~.
wt.STP.ttts.p.rxt}xuxr.p}s.|tsxrp{.
t
p{px~}.p}s.rwts{x}v.tr~||t}spx~}.
wt.STP.~.vp}.~.tstrt~ .txx~}.
u~|.wt.Stp|t}.~u.Wtp{w.p}s.W|p}.
~.x}xxpt.{t|pzx}v.~rttsx}v.~.
bt
xrt.6WWb7=.cwt.WWb.r~}r{sts.wp.
trwts{t.|pxyp}p=.cwt.txx~}.x;.
|pxyp}p.wp.p.wxvw.~t}xp{.u~.pqt;.wp. wttu~t;.wttq.st}xts=.
}~.prrtts.|tsxrp{.t.x}.wt.d}xts.bpt;.
p}s.{prz.p}.prrtpq{t.{t
t{.~u.put.u~.t. bx}rtt{;.
t
t}.}st.|tsxrp{.t
xx~}=.cwttu~t;.
Rwrz.a~t}qtv;.
wt.WWb.tr~||t}sts.wp.|pxyp}p.
Cevkpi!Cfokpkuvtcvqt=.
t|px}.x}.brwts{t.X=.cwt.rxt}xuxr.p}s.
|tsxrp{.t
p{px~}.p}s.rwts{x}v.
Pprw|t}I.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
FGRCTVOGPV!QH!LWUVKEG!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11113
Hov!5812
Uhov!5813
Rtgnkokpct{!Pqvg!Tgictfkpi!Vtgcv{!
Eqpukfgtcvkqpu!
R~
t.[tt.u~|.WWb.~.STP.
b||pxx}v.wt.brxt}xuxr.p}s.\tsxrp{.
T
p{px~}.p}s.brwts{x}v.atr~||t}spx~}.
u~.\pxyp}p=.
d=b=.Stp|t}.~u.Wtp{w.p}s.W|p}.
bt
xrt.6WWb7¥Qpx.u~.wt.
atr~||t}spx~}.u~.\px}px}x}v.\pxyp}p.
x}.brwts{t.X.~u.wt.R~}~{{ts.bqp}rt.
Pr.
d=b=.Stp|t}.~u.Yxrt¥Sv.
T}u~rt|t}.Ps|x}xpx~}.6STP7;.
brwts{t.~u.R~}~{{ts.bqp}rtI.
\px}px}x}v.\pxyp}p.x}.brwts{t.X.~u.wt.
R~}~{{ts.bqp}rt.Pr;.Qprzv~}s;.
Spp;.p}s.P}p{xI.Txvw.Upr~.
Stt|x}px
t.~u.R~}~{.p}s.Ux}sx}v.
_p}.[email protected]=b=R=.G@A6q7.
SptsI.Y{.@H;.A?@E=.
Ejwem!Tqugpdgti-!
Cevkpi!Cfokpkuvtcvqt-!Rtgnkokpct{!Pqvg!
Tgictfkpi!Vtgcv{!Eqpukfgtcvkqpu/!
P.wt.R~}~{{ts.bqp}rt.Pr.
6RbP7.tr~v}xt;.wt.d}xts.bpt.x.p.
p.~.wt.bx}v{t.R~}
t}x~}.~}.
]pr~xr.Sv;.@[email protected]tutts.~.wtt.p.
wt.bx}v{t.R~}
t}x~}.~.wt.tp7=.A@.
d=b=R=.G?@6F7=._pxt.~.wt.bx}v{t.
R~}
t}x~}.pt.~q{xvpts.~.|px}px}.
px~.r~}~{.~
xx~}.t{pts.~.wt.
sv.wp.pt.r~
tts.q.wt.tp=.
\p}.~u.wt.~
xx~}.~u.wt.RbP.tt.
t}prts.q.R~}vt.u~.wt.trxuxr.
~t.~u.t}x}v.d=b=.r~|{xp}rt.
xw.wt.tp=.P|~}v.wtt.x.p.
rwts{x}v.~
xx~};[email protected]=b=R=.
G@@6s76@7=.btrx~}.G@@6s76@7.~
xst.
wp;.wtt.p.sv.x.qytr.~.r~}~{.
}st.wt.bx}v{t.R~}
t}x~};.wt.STP.
Ps|x}xp~.6q.st{tvpx~}.u~|.wt.
P~}t.Vt}tp{7.|.xt.p}.~st.
r~}~{{x}v.rw.sv.}st.wt.
rwts{t.wt.stt|.|~.p~xpt.~.
rp.~.rw.jtpl.~q{xvpx~};.
xw~.tvps.~.wt.ux}sx}v.txts.
q[email protected]=b=R=.G@@6p7.~[email protected]}s.
xw~.tvps.~.wt.~rtst.
trxqts.q[email protected]=b=R=.G@@6p7.p}s.
6q7l= .
\pxyp}p.x.p.sv.{xts.x}.wt.
bx}v{t.R~}
t}x~}=.cwt.bx}v{t.
R~}
t}x~}.t.wt.t|.rp}}pqx .~.
tut.~.|pxyp}p= cw;.wt.STP.
d}st.wt.bx}v{t.R~}
t}x~};.rp}}pqx.{p} .
|tp}.p}.{p}.~u.wt.vt}.Ecppcdku/ .Pxr{t.
@6r7=.cwt.bx}v{t.R~}
t}x~}.stux}t.rp}}pqx .~.
x}r{st.wt.u{~tx}v.~.uxx}v.~.~u.wt.
rp}}pqx.{p}.6tr{sx}v.wt.tts.p}s.{tp
t.
wt}.}~.prr~|p}xts.q.wt.~7.u~|.wxrw.wt.
tx}.wp.}~.qtt}.tprts;.q.wpt
t.}p|t.
wt.|p.qt.stxv}pts= .Pxr{t.@6q7=.cwx.
stux}xx~}.~u.rp}}pqx .}st.wt.bx}v{t.
R~}
t}x~}.x.{xvw{.{t.x}r{x
t.wp}.wt.RbP.
stux}xx~}.~u.|pxwp}p; .wxrw.x}r{st.p{{.p.
~u.wt.rp}}pqx.{p}.trt.u~.wt.|pt.p{z;.
tx{xts.tts;.~x{.u~|.wt.tts;.p}s.rtpx}.
stx
px
t.wt[email protected]=b=R=.G?A6@E7=.Rp}}pqx.
p}s.rp}}pqx.tx}.pt.x}r{sts.x}.wt.{x.~u.sv.
x}.brwts{t.X.p}s.brwts{t.Xe.~u.wt.bx}v{t.
R~}
t}x~}=.X}.r~}p.~.wt.RbP;.wt.sv.{xts.
x}.brwts{t.Xe.~u.wt.bx}v{t.R~}
t}x~}.pt.p{~.
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
\pxyp}p.|tt.wt.wtt.rxtxp.u~.
{prx}v.p.qp}rt.x}.brwts{t.X.~u.wt.RbP.
}st[email protected]=b=R=.G@A6q76@7=.P.sxrts.x}.
wt.t}r{~ts.p}p{t;.|pxyp}p.wp.p.wxvw.
~t}xp{.u~.pqt;.}~.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt;.p}s.p.{prz.~u.prrtts.put.u~.t.
}st.|tsxrp{.t
xx~}=.Prr~sx}v{;.
WWb.tr~||t}s.wp.|pxyp}p.qt.
|px}px}ts.x}.brwts{t.X.~u.wt.RbP=.
T}r{~ts.pt.~.s~r|t}.tpts.q.
USP .R~}~{{ts.bqp}rt.bpuu.6x}.
t~}t.~.txx~}.ux{ts.x}.A??H.q.\=.
Qp}.Z||.p}s.x}.A?@@.q.V~
t}~.
[x}r~{}.S=.Rwputt.p}s.Rwxx}t.^=.Vtv~xt7.
wp.u~|.wt.qpx.u~.wt.tr~||t}spx~}=.
_p}.~.wt.tt.x}.wt.txx~};.USP.
q~ps{.t
p{pts.|pxyp}p;.p}s.sxs.}~.
u~r.x.t
p{px~}.~}.pxr{p.px}.~u.
|pxyp}p.~.r~|~}t}.~.stx
px
t.~u.
|pxyp}p=.
USP .Rt}t.u~.Sv.T
p{px~}.p}s.
attprw .rt}.t
xt.~u.wt.p
px{pq{t.
t
xst}rt.p}s.wt.q{xwts.r{x}xrp{.sxt.
~}.|pxyp}p.st|~}pts.wp.x}rt.~.
A??E.rxt}xuxr.p}s.|tsxrp{.t
p{px~}.p}s.
rwts{x}v.tr~||t}spx~}.t~}sx}v.~.p.
t
x~.STP.txx~};.ttprw.xw.
|pxyp}p.wp.~vtts=.W~t
t;.wt.
p
px{pq{t.t
xst}rt.x.}~.uuxrxt}.~.
stt|x}t.wp.|pxyp}p.wp.p}.prrtts.
|tsxrp{.t=.cwttu~t;.|~t.ttprw.x.
}ttsts.x}~.|pxyp}p .tuutr;.x}r{sx}v.
~t}xp{.|tsxrp{.t.u~.|pxyp}p.p}s.x.
stx
px
t=.Qpts.~}.wt.rt}.t
xt;.t.
xst}xuxts.t
tp{.|tw~s~{~vxrp{.rwp{{t}vt.
x}.wt.|pxyp}p.sxt.q{xwts.x}.wt.
{xtpt=.ft.tr~||t}s.wt.qt.psstts.
x}.ut.r{x}xrp{.sxt.xw.|pxyp}p.~.
t}t.wp.
p{xs.rxt}xuxr.spp.pt.vt}tpts.
x}.sxt.t
p{px}v.|pxyp}p .put.p}s.
tuuxrpr.u~.wtptxr.t=.U~.tp|{t;.t.
tr~||t}s.wp.sxt.}tts.~.u~r.~}.
r~}xt}.ps|x}xpx~}.p}s.t~srxq{t.
s~x}v.~u.|pxyp}p;.~t}xp{{.w~vw.wt.
t.~u.ps|x}xpx~}.|tw~s.~wt.wp}.
|~zx}v=.P.||p.~u.~.t
xt.~u.wt.
q{xwts.{xtpt.~}.wt.r{x}xrp{.t.~u.
|pxyp}p;.x}r{sx}v.tr~||t}spx~}.u~.
ut.sxt;.x.pprwts.~.wx.s~r|t}=.
USP.p}s.wt.]px~}p{.X}xt.~u.Wtp{w .
]px~}p{.X}xt.~}.Sv.Pqt.6]XSP7.p{~.
qt{xt
t.wp.~z.r~}x}t.~.qt.}ttsts.~.
t}t.~.q.wt.utstp{.v~
t}|t}.u~.
wt.tuuxrxt}.r~}sr.~u.r{x}xrp{.ttprw.
Stp|t}.~u.Wtp{w.p}s.W|p}.bt
xrt;.
x}v.|pxyp}p=.R~}rt}.wp
t.qtt}.pxts.
^uuxrt.~u.wt.btrtp.Pxp}.btrtp.
pq~.wtwt.wt.txx}v.utstp{.tv{p~.
u~.Wtp{w;.^uuxrt.~u._q{xr.Wtp{w.p}s.
t|.x.u{txq{t.t}~vw.~.t~}s.~.
brxt}rt;.fpwx}v~};.SR.A?A?@=.
x}rtpts.x}tt.x}.ttprw.x}~.wt.
Y}t.AD;.A?@D=.
{xts.x}.brwts{t.X.~u.wt.bx}v{t.R~}
t}x~}.p}s.
~t}xp{.wtptxr.t.~u.|pxyp}p.p}s.
pt.qytr.~.wt.p|t.r~}~{.p.brwts{t.X.sv. cwt.W~}~pq{t.Rwrz.a~t}qtv.
|pxyp}p<stx
ts.sv=.WWb.t{r~|t.p}.
p.t{{.p.pssxx~}p{.r~}~{=.Pxr{t.A;.p=.D.
Prx}v.Ps|x}xp~;.Sv.T}u~rt|t}.
~~}x.~.r~}x}t.~.t{~t.wtt.
cwt.R~.uwt.ptsI.U~.tp|{t;.jpxr{t.
Ps|x}xpx~};.d=b=.Stp|t}.~u.Yxrt;. r~}rt}.xw.STP=.
B@.ppvpw.C.~u.wt.bx}v{t.R~}
t}x~}l.txt.
GF?@.\~xtt.Sx
t;.bx}vuxt{s;.eP.
bw~{s.~.wp
t.p}.tx~}.tvpsx}v.
x|~.p}s.t~.t|x.wp.~{s.}~.qt.
AA@DA=.
wtt.tr~||t}spx~};.{tpt.r~}pr.
~qpx}ts.xu.wt.qp}rt.tt.{prts.x}.RbP.
rwts{t.XXX.w~vw.e=.X}.pssxx~};.wt.~p.p}s.
R~x}}t._=.\~~s;.brxt}rt._~{xr.P}p{;.
Stp.\=.a~t}qtvI.
jtr~szttx}vl.txt|t}.~u.Pxr{t.@H.w~vw.
_p}.~.wt.R~}~{{ts.bqp}rt.Pr. R~}~{{ts.bqp}rt.bpuu;.Rt}t.u~.Sv.
A@.~u.wt.bx}v{t.R~}
t}x~}.~{s.qt.pxuxts.~}{. 6RbP;[email protected]=b=R=.µ G@@6q7;.6r7;.p}s.6u77;.wt.
T
p{px~}.p}s.attprw;.USP;.p[email protected]¤.
q.{prx}v.wt.qp}rt.x}.RbP.rwts{t.X.~.XX= .
B@DA=.
Stp|t}.~u.Wtp{w.p}s.W|p}.bt
xrt.
Kf/!}[email protected]}t}p{.rxpx~}.~|xts7=.
6WWb7.x.tr~||t}sx}v.wp.|pxyp}p.
bx}rtt{.~;.
P.STP.wp.pts.x}.t
p{px}v.x~.|pxyp}p.
r~}x}t.~.qt.|px}px}ts.x}.brwts{t.X.~u.
trwts{x}v.txx~};.R~}vt.tpq{xwts.~}{.
Zpt}.Q=.Stbp{
~;.\S;.\_W;.\br.
wt.RbP=.
~}t.rwts{t;.rwts{t.X;.u~.sv.~u.pqt.xw.
Cevkpi!Cuukuvcpv!Ugetgvct{!hqt!Jgcnvj=.
cwt.U~~s.p}s.Sv.Ps|x}xpx~}.6USP7.
}~.rt}{.prrtts.|tsxrp{.t.x}.tp|t}.x}.
T}r{~tI.Qpx.u~.wt.atr~||t}spx~}.u~.
wp.r~}xstts.wt.pqt.~t}xp{.p}s.
wt.d}xts.bpt p}s.{prz.~u.prrtts.put.u~.
\px}px}x}v.\pxyp}p.x}.brwts{t.X.~u.wt.
stt}st}rt<~srx}v.rwpprtxxr.~u.
t.= = =.}st.|tsxrp{.t
xx~}= [email protected]=b=R=.
G@A6q7= .FE.Ua.C?DDA.6A?@@7J.EE.Ua.A??BG.6A??@7=.
R~}~{{ts.bqp}rt.Pr.
|pxyp}p=.
Ps|x}xp~.x.~q{xvpts.}st.trx~}.
G@@6s7.~.r~}~{.|pxyp}p.x}.wt.
rwts{t.wp.wt.stt|.|~.
p~xpt.~.rp.~.wt.d=b=.
~q{xvpx~}.}st.wt.bx}v{t.
R~}
t}x~}=.X.wp.qtt}.tpq{xwts.x}.
x~.|pxyp}p.trwts{x}v.
~rttsx}v.wp.{prt|t}.~u.
|pxyp}p.x}.txwt.rwts{t.X.~.
rwts{t.XX.~u.wt.RbP.x.}trtp.p.
t{{.p.uuxrxt}.~.pxu.~.
x}t}px~}p{.~q{xvpx~} .}st.wt.
bx}v{t.R~}
t}x~}=.PQTON!
=.FGC-!DDH.
U=As.FBD;[email protected]=R=.Rx=.@HFF7=.P.wt.
d}xts.bpt.R~.~u.Ptp{.u~.wt.
SR.Rxrx.wp.pts;.t
tp{.
txt|t}.x|~ts.q.wt.bx}v{t.
R~}
t}x~}.~{s.}~.qt.|t.xu.
rp}}pqx.p}s.rp}}pqx.tx}.tt.
{prts.x}.RbP.rwts{t.XXX;.Xe;.~.
e= Kf/!cwttu~t;.x}.prr~sp}rt.xw.
trx~}.G@@6s76@7;.STP.|.{prt.
|pxyp}p.x}.txwt.rwts{t.X.~.
rwts{t.XX=.
Qtrpt.rwts{t.X.p}s.XX.pt.wt.
~}{.~xq{t.rwts{t.x}.wxrw.
|pxyp}p.|p.qt.{prts;.u~.~t.
~u.t
p{px}v.wx.rwts{x}v.txx~};.x.
x.tt}xp{.~.}stp}s.wt.
sxuutt}rt.qttt}.wt.rxtxp.u~.
{prt|t}.~u.p.qp}rt.x}.rwts{t.X.
p}s.w~t.u~.{prt|t}.x}.rwts{t.XX=.
cwtt.rxtxp.pt.t.u~w.x}[email protected]=b=R=.
G@[email protected]}s.6q76A7;.ttrx
t{=.P.
x}sxrpts.wttx};.qp}rt.x}.q~w.
rwts{t.X.p}s.rwts{t.XX.wpt.wt.
rwpprtxxr.~u.p.wxvw.~t}xp{.u~.
pqt= .fwtt.wt.sxx}rx~}.{xt.x.
wp.rwts{t.X.sv.wp
t.}~.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bpt .p}s.p.{prz.~u.prrtts.
put.u~.t.~u.wt.sv.= = =.}st.
|tsxrp{.t
xx~}; .wx{t.rwts{t.XX.
sv.s~.wp
t.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt= .
Prr~sx}v{;.x}.
xt.~u.trx~}.
G@@6s76@7;.wx.rwts{x}v.txx~}.}.
~}.wtwt.|pxyp}p.wp.p.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bpt=.Xu.x.s~t.}~;.STP.|;.
p}.~.trx~}.G@@6s7;.st}.wt.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
6479;!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
txx~}.p}s.ztt.|pxyp}p.x}.rwts{t.
X=.
P.x}sxrpts;.wtt.trx~}.G@@6s76@7.
p{xt.~.p.sv.wp.x.wt.qytr.~u.
p.trwts{x}v.txx~};.wt.STP.
Ps|x}xp~.|.xt.p}.~st.
r~}~{{x}v.wt.sv.}st.wt.rwts{t.
wt.stt|.|~.p~xpt.~.rp.~.
d}xts.bpt.~q{xvpx~}.}st.wt.
bx}v{t.R~}
t}x~};.xw~.tvps.~.
wt.ux}sx}v.txts.q.trx~}.G@@6p7.
~[email protected]}s.xw~.tvps.~.wt.
~rtst.trxqts.q.trx~}.
G@@6p7.p}s.6q7=.cw;.x}rt.wt.~}{.
stt|x}px
t.xt.x}.t
p{px}v.wt.
tt}.rwts{x}v.txx~}.x.wtwt.
|pxyp}p.wp.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt;.STP.}tts.}~.r~}xst.wt.
ux}sx}v.~u.trx~}.G@@6p7.~.G@A6q7.wp.
wp
t.}~.qtpx}v.~}.wp.stt|x}px~};.
p}s.STP.{xztxt.}tts.}~.u~{{~.wt.
~rtst.trxqts.q.trx~}.
G@@6p7.p}s.6q7.xw.ttr.~.rw.
xt{t
p}.ux}sx}v=.btrxuxrp{{;.STP.
}tts.}~.t
p{pt.wt.t{px
t.pqt.
~t}xp{.~u.|pxyp}p.~.wt.t{px
t.
tt}.~.wxrw.pqt.~u.|pxyp}p.|p.
{tps.~.wxrp{.~.rw~{~vxrp{.
st•t}st}rt=.
P.t{px}ts.qt{~;.wt.|tsxrp{.p}s.
rxt}xuxr.t
p{px~}.p}s.rwts{x}v.
tr~||t}spx~}.xts.q.wt.btrtp.
~u.Wtp{w.p}s.W|p}.bt
xrt.
r~}r{st.wp.|pxyp}p.wp.}~.
rt}{.prrtts.|tsxrp{.t.x}.
tp|t}.x}.wt.d}xts.bpt;.p}s.wt.
STP.Ps|x}xp~.{xztxt.~.
r~}r{st=.U~.wt.tp~}.y.
x}sxrpts;.}~.uwt.p}p{x.qt~}s.
wx.r~}xstpx~}.x.txts=.
]~}twt{t;.qtrpt.~u.wt.xsttps.
q{xr.x}tt.x}.}stp}sx}v.p{{.wt.
upr.t{px}v.~.wt.wp|.p~rxpts.
xw.|pxyp}p;.STP.x.q{xwx}v.wtt.
wt.t}xt.|tsxrp{.p}s.rxt}xuxr.
p}p{x.p}s.rwts{x}v.t
p{px~}.
xts.q.wt.btrtp;.p.t{{.p.
STP .pssxx~}p{.p}p{x=.
RQ!11111
Hto!11114
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
647;1!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Dcuku!hqt!vjg!Tgeqoogpfcvkqp!hqt!
Ockpvckpkpi!Octklwcpc!kp!Uejgfwng!K!qh!
vjg!Eqpvtqnngf!Uwduvcpegu!Cev!
^}.]~
t|qt.B?;.A?@@;.V~
t}~.
[x}r~{}.S=.Rwputt.~u.aw~st.X{p}s.p}s.
Rwxx}t.^=.Vtv~xt.~u.fpwx}v~}.
q|xts.p.txx~}.~.wt.Sv.
T}u~rt|t}.Ps|x}xpx~}.6STP7.
ttx}v.wp.~rttsx}v.qt.x}xxpts.
~.ttp{.wt.{t.p}s.tv{px~}.wp.
{prt.|pxyp}p x}.brwts{t.X.~u.wt.
R~}~{{ts.bqp}rt.Pr.6RbP7=.cwt.
txx~}.r~}t}s.wp.rp}}pqx.wp.p}.
prrtts.|tsxrp{.t.x}.wt.d}xts.
bpt;.x.put.u~.t.}st.|tsxrp{.
t
xx~};.p}s.wp.p.t{px
t{.{~.
pqt.~t}xp{.r~|pts.~.~wt.
brwts{t.XX.sv=.cwt.txx~}.tt.
wp.|pxyp}p.p}s.t{pts.xt| .qt.
trwts{ts.x}.brwts{t.XX.~u.wt.RbP=.
X}.Y}t.A?@B;.wt.STP.Ps|x}xp~.
ttts.wp.wt.d=b=.Stp|t}.~u.
Wtp{w.p}s.W|p}.bt
xrt.6WWb7.
~
xst.p.rxt}xuxr.p}s.|tsxrp{.
t
p{px~}.~u.wt.p
px{pq{t.x}u~|px~}.
p}s.p.rwts{x}v.tr~||t}spx~}.u~.
|pxyp}p;.x}.prr~sp}rt.xw.wt.
~
xx~}[email protected]=b=R=.G@@6q7=.
X}.prr~sp}rt.x[email protected]=b=R=.G@@6q7;.
STP.wp.vpwtts.x}u~|px~}.t{pts.~.
wt.r~}~{.~u.|pxyp}p.6Ecppcdku!
ucvkxc7 }st.wt.RbP=._p}.~.A@.
d=b=R=.G@@6q7;.wt.btrtp.~u.WWb.x.
txts.~.r~}xst.x}.p.rxt}xuxr.p}s.
|tsxrp{.t
p{px~}.txvw.upr~.
stt|x}px
t.~u.r~}~{.}st.wt.RbP=.
U~{{~x}v.r~}xstpx~}.~u.wt.txvw.
upr~;.xu.x.x.p~xpt;.wt.btrtp.
|.|pzt.wtt.ux}sx}v.~.tr~||t}s.
rwts{x}v.p.qp}rt.x}.wt.RbP=.cwt.
ux}sx}v.t{pt.~.p.qp}rt .pqt.
~t}xp{;.{tvxx|pt.|tsxrp{.t;.p}s.
put.~.stt}st}rt.{xpqx{x=.
Ps|x}xpx
t.t~}xqx{xxt.u~.
t
p{px}v.p.qp}rt.u~.r~}~{.}st.
wt.RbP.pt.tu~|ts.q.wt.U~~s.p}s.
Sv.Ps|x}xpx~}.6USP7;.xw.wt.
r~}rt}rt.~u.wt.]px~}p{.X}xt.~}.
Sv.Pqt.6]XSP7;.p.strxqts.x}.wt.
\t|~p}s|.~u.d}stp}sx}v.6\^d7.
~u.\p•rw.G;[email protected]?.Ua.HD@G¤A?7=.
X}.wx.s~r|t};.USP.tr~||t}s.
wt.r~}x}ts.r~}~{.~u.|pxyp}p.x}.
]~t.wp.|pxwp}p .x.wt.t{{x}v.~xvx}p{{.
ts.x}.wt.R~}~{{ts.bqp}rt.Pr.6RbP7=.cwx.
s~r|t}.t.wt.t{{x}v.wp.x.|~t.r~||~}.
x}.rt}.pvt;.|pxyp}p= .
cwt.RbP.stux}t.|pxyp}p.p.wt.u~{{~x}vI.
P{{.p.~u.wt.{p}.Rp}}pqx.bpx
p.[=;.wtwt.
v~x}v.~.}~J.wt.tts.wtt~uJ.wt.tx}.
tprts.u~|.p}.p.~u.rw.{p}J.p}s.t
t.
r~|~}s;.|p}uprt;.p{;.stx
px
t;.|xt;.
~.tppx~}.~u.rw.{p};.x.tts.~.tx}=.brw.
t|.s~t.}~.x}r{st.wt.|pt.p{z.~u.rw.
{p};.uxqt.~srts.u~|.rw.p{z;.~x{.~.rpzt.
|pst.u~|.wt.tts.~u.rw.{p};.p}.~wt.
r~|~}s;.|p}uprt;.p{;.stx
px
t;.|xt;.
~.tppx~}.~u.rw.|pt.p{z.6trt.wt.
tx}.tprts.wttu~|7;.uxqt;.~x{;.~.rpzt;.~.wt.
tx{xts.tts.~u.rw.{p}.wxrw.x.x}rppq{t.~u.
vt|x}px~}[email protected]=b=R=.G?A6@E77=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
brwts{t.X.~u.wt.RbP=._p}.~.A@.
d=b=R=.G@@6r7;.wt.txvw.upr~.
tpx}x}v.~.wt.rwts{x}v.~u.
|pxyp}p.pt.r~}xstts.qt{~=.
~}.wp.x.z}~}.pq~.wt.p}vt.~u.
wtt.r~}xt}.pr~.p{{.r{x
pts.
px}=.
Stt|x}x}v.wt.pqt.~t}xp{.~u.p.
qp}rt.x.r~|{t.xw.|p}.
2/!Kvu!Cevwcn!qt!Tgncvkxg!Rqvgpvkcn!hqt!
sx|t}x~};.p}s.}~.x}v{t.t.~.
Cdwug!
pt|t}.~
xst.p.r~|{tt.
rwpprtxpx~}=.cw;.}~.x}v{t.
d}st.wt.ux.upr~.wt.btrtp.
|tpt.~u.pqt.~t}xp{.x.xstp{=.
|.r~}xst.|pxyp}p .prp{.~.
brxt}xuxrp{{;.p.r~|twt}x
t.
t{px
t.~t}xp{.u~.pqt=.cwt.RbP.
t
p{px~}.~u.wt.t{px
t.pqt.
s~t.}~.stux}t.wt.t|.pqt= .
~t}xp{.~u.p.qp}rt.rp}.x}r{st.
W~t
t;.wt.RbP .{tvx{px
t.wx~.
r~}xstpx~}.~u.wt.u~{{~x}v.t{t|t}I.
vvt.wt.u~{{~x}v.x}.stt|x}x}v.
atrt~.qx}sx}v.puux}x;.tr{x}xrp{.
wtwt.p.pxr{p.sv.~.qp}rt.
wp|pr~{~v;.tx}u~rx}v.tuutr;.
wp.p.~t}xp{.u~.pqtI .
sxrx|x}px
t.x|{.tuutr;.
p=.cwtt.x.t
xst}rt.wp.x}sx
xsp{.
stt}st}rt.~srx}v.~t}xp{;.
pt.pzx}v.wt.sv.~.sv.r~}px}x}v.
wp|pr~zx}txr;.~t.~u.
rw.p.qp}rt.x}.p|~}.uuxrxt}.
ps|x}xpx~};.~xrx;.spp.~}.prp{.
~.rtpt.p.wpps.~.wtx.wtp{w.~.~.
pqt;.r{x}xrp{.pqt.~t}xp{.sxt;.
wt.put.~u.~wt.x}sx
xsp{.~.~.wt.
p}s.q{xr.wtp{w.xz=.X|~p}{;.
r~||}x=.
pqt.rp}.tx.x}stt}st}{.u~|.
q=.cwtt.x.p.xv}xuxrp}.sx
tx~}.~u.
~{tp}rt.~.wxrp{.stt}st}rt.
wt.sv.~.sv.r~}px}x}v.rw.p.
qtrpt.x}sx
xsp{.|p.pqt.sv.x}.
qp}rt.u~|.{tvxx|pt.sv.
s~t.~.pt}.wp.s~.}~.x}srt.
rwp}}t{=.
wtt.wt}~|t}p=.Pssxx~}p{{.
r=.X}sx
xsp{.pt.pzx}v.wt.sv.~.
t
xst}rt.~u.r{p}stx}t.~{px~}.p}s.
sv.r~}px}x}v.rw.p.qp}rt.~}.
x{{xrx.puuxrzx}v.~u.p.qp}rt.rp}.wts.
wtx.~}.x}xxpx
t.pwt.wp}.~}.wt.
{xvw.~}.q~w.wt.st|p}s.u~.p.
qpx.~u.|tsxrp{.ps
xrt.u~|.p.
qp}rt.p.t{{.p.wt.tpt.~u.
prxx~}t.{xrt}ts.q.{p.~.
~qpx}x}v.p.qp}rt=.P}x|p{.p}s.
ps|x}xt.rw.sv.x}.wt.r~t.~u.
w|p}.{pq~p~.spp.p}s.
wx.~utx~}p{.prxrt=.
s=.cwt.sv.~.sv.r~}px}x}v.rw. txst|x~{~vxrp{.spp.pt.p{{.ts.x}.
p.qp}rt.pt.}t.sv.~.t{pts.x}. stt|x}x}v.p.qp}rt .pqt.
~t}xp{=.\~t~
t;.txst|x~{~vxrp{.
wtx.prx~}.~.p.sv.~.sv.p{tps.
{xts.p.wp
x}v.p.~t}xp{.u~.pqt.~. spp.rp}.x}sxrpt.prp{.pqt=.
cwt.txx~}t.r~|pt.wt.tuutr.~u.
|pzt.x.{xzt{.wp.wt.sv.x{{.wp
t.
|pxyp}p.~.rt}{.r~}~{{ts.
wt.p|t.~t}xp{x.u~.pqt.p.rw.
brwts{t.XX.qp}rt.p}s.|pzt.
sv;.w.|pzx}v.x.tp~}pq{t.~.
ttpts.r{px|.pq~.wtx.r~|ppx
t.
p|t.wp.wtt.|p.qt.xv}xuxrp}.
tuutr=.R~|px~}.qttt}.|pxyp}p.
sx
tx~}.u~|.{tvxx|pt.rwp}}t{;.
p}s.wt.sx
tt.pp.~u.brwts{t.XX.
xv}xuxrp}.t.r~}p.~.~.xw~.
qp}rt.x.sxuuxr{;.qtrpt.~u.wt.
|tsxrp{.ps
xrt;.~.wp.x.wp.p.
wp|pr~{~vxrp{{.sxx|x{p.prx~}.~u.
qp}xp{.rppqx{x.~u.rtpx}v.
qp}rt.~u.brwts{t.XX.~u.wt.RbP=.
wpps.~.wt.wtp{w.~u.wt.t.~.~.
U~.tp|{t;.brwts{t.XX.qp}rt.
wt.put.~u.wt.r~||}x=.
x}r{st.x|{p}<{xzt.sv.6g/i/-!
X}.wt.st
t{~|t}.~u.wx.rxt}xuxr.
r~rpx}t;.|tw{wt}xspt;.p}s.
p}s.|tsxrp{.t
p{px~}.u~.wt.~t.
p|wtp|x}t7;.~x~xs.6g/i/-!~r~s~}t;.
~u.rwts{x}v;.wt.btrtp.p}p{ts.
ut}p}{7;.tspx
t.6g/i/-!t}~qpqxp{;.
r~}xstpq{t.spp.t{pts.~.wt.
p|~qpqxp{7;.sx~rxpx
t.p}twtxr.
qp}rt .pqt.~t}xp{=.cwt.spp.
6g/i/-!_R_7;.p}s.}pp{{.~rrx}v.
x}r{st.p.sxrx~}.~u.wt.t
p{t}rt.
{p}.r~|~}t}.6g/i/-!r~rp.{tp
t.p}s.
p}s.utt}r.~u.t;.wt.p|~}.~u.wt. ~.p7=.cwt.|trwp}x|67.~u.
qp}rt.p
px{pq{t.u~.x{{xrx.t;.wt.
prx~}.~u.wt.pq~
t.brwts{t.XX.
tpt.~u.~qpx}x}v.~.|p}uprx}v.wt.
qp}rt.pt.w~{{.sxuutt}.u~|.
qp}rt;.wt.tpx~}.~.p.~u.wt. ~}t.p}~wt;.p}s.wt.pt.sxuutt}.u~|.
qp}rt.~}.wt.tt; .p}s.t
xst}rt. tpws~rp}}pqx}~{.6cWR7.p}s.
t{t
p}.~.p<xz.~{px~}=.
|pxyp}p.p.t{{=.U~.tp|{t;.
X|~p}{;.wt.txx~}t.stux}t.
brwts{t.XX.x|{p}.xrp{{.
|pxyp}p.p.x}r{sx}v.p{{.Ecppcdku!
u}rx~}.q.x}rtpx}v.|~}~p|x}tvxr.
r{x
pts.px}=.Sxuutt}.|pxyp}p.
~}t.
xp.p}.x}rtpt.x}.s~p|x}t.p}s.
p|{t.stx
ts.u~|.
px~.r{x
pts. }~tx}twx}t.6brw|x.t.p{=;.A?@B7=.X}.
px}.|p.wp
t.
t.sxuutt}.rwt|xrp{. r~}p;.~x~xs.p}p{vtxr.u}rx~}.
xp.
r~}xt};.w.wt.p}p{x.x.qpts.
|<~x~xs.trt~.pv~}x.tuutr=.
cwtt.sxuutx}v.|trwp}x|67.~u.prx~}.
R~|twt}x
t.Sv.Pqt._t
t}x~}.p}s.
t{.x}.
p{.sxuutt}.qtwp
x~p{.p}s.
R~}~{.Pr.~u.@HF?;.W=a=.at=.]~=.H@¤@CCC;.H@.
ps
tt.tuutr.~ux{t;.|pzx}v.
R~}v=;.bt[email protected]@HF?7.tx}ts.x}.d=b=R=R=P=]=.
CDEE;.CE?B=.
r~|px~}.pr~.wt.p}vt.~u.
RQ!11111
Hto!11115
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
wp|pr~{~vxrp{{.sx
tt.R¤XX.
qp}rt.x}p~xpt=.
X}.pssxx~};.|p}.qp}rt.
rwts{ts.}st.wt.RbP.pt.t
xtts.
p}s.t
p{pts.xwx}.wt.r~}t.~u.
r~||trxp{.sv.st
t{~|t};.x}v.
spp.q|xts.x}.wt.u~|.~u.p.}t.
sv.p{xrpx~}.6]SP7=.P.}t.
p}p{vtxr.sv.|xvw.qt.r~|pts.~.p.
rt}{.rwts{ts.p}p{vtxr.sv.p.
p.~u.wt.pt|t}.~u.x.t{px
t.
pqt.~t}xp{=.W~t
t;.qtrpt.wt.
txx~}t.wp
t.}~.xst}xuxts.p.trxuxr.
x}sxrpx~}.u~.wt.t.~u.|pxyp}p;.
xst}xux}v.p}.p~xpt.r~|pp~.
qpts.~}.x}sxrpx~}.rp}}~.qt.s~}t=.
p=.cwtt.x.t
xst}rt.wp.x}sx
xsp{.
pt.pzx}v.wt.qp}rt.x}.p|~}.
uuxrxt}.~.rtpt.p.wpps.~.wtx.
wtp{w.~.~.wt.put.~u.~wt.
x}sx
xsp{.~.~.wt.r~||}x=.
T
xst}rt.w~.wp.~|t.x}sx
xsp{.
pt.pzx}v.|pxyp}p.x}.p|~}.
uuxrxt}.~.rtpt.p.wpps.~.wtx.
wtp{w.p}s.~.wt.put.~u.~wt.
x}sx
xsp{.p}s.wt.r~||}x=.P.{pvt.
}|qt.~u.x}sx
xsp{.t.|pxyp}p=.
WWb.~
xst.spp.~}.wt.tt}.~u.
|pxyp}p.pqt.w~vw.]XSP.p}s.wt.
bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~}.6bP\WbP7=.
Prr~sx}v.~.wt.|~.trt}.spp.u~|.
bP\WbP .A?@A.]px~}p{.b
t.~}.
Sv.dt.p}s.Wtp{w.6]bSdW7;.wxrw.
tx|pt.wt.}|qt.~u.x}sx
xsp{.
w~.wp
t.ts.p.qp}rt.xwx}.p.
|~}w.x~.~.wt.s.6strxqts.p.
rt}.t 7;.|pxyp}p.x.wt.|~.
r~||~}{.ts.x{{xrx.sv.p|~}v.
P|txrp}.pvts.@A.tp.p}s.~{st;.
xw.p}.tx|pts.@G=H.|x{{x~}.
P|txrp}.wp
x}v.ts.|pxyp}p.
xwx}.wt.|~}w.x~.~.wt.A?@A.
]bSdW=.R~|pts.~.A??C;.wt}.p}.
tx|pts.@C=E.|x{{x~}.x}sx
xsp{.
t~ts.x}v.|pxyp}p.xwx}.wt.
|~}w.x~.~.wt.s;.wt.tx|pts.
pt.x}.A?@A.w~.p}.x}rtpt.~u.
p~x|pt{.C=B.|x{{x~}.x}sx
xsp{=.
cwt.A?@B.\~}x~x}v.wt.Ut.6\cU7.
t.~u.Gw;.@?w;.p}s.@Aw.vpst.
st}.p{~.x}sxrpt.wp.|pxyp}p.x.
wt.|~.xst{.ts.x{{xrx.qp}rt.x}.
wx.pvt.v~=.btrxuxrp{{;.rt}.
|~}w.t.p.p.F=?.trt}.~u.Gw.
vpst;.@G=?.trt}.~u.@?w;.vpst.
p}s.AA=F.trt}.~u.@Aw.vpst=.
Pssxx~}p{{;.wt.A?@@.ctp|t}.
Tx~st.Spp.bt.6cTSb7.t~ts.wp.
x|p.|pxyp}p.pqt.prr~}ts.u~.
@G=@.trt}.~u.}~}<x
pt.qp}rt<.
pqt.tp|t}.uprx{x.ps|xx~};.
xw.AC=B.trt}.~u.w~t.ps|xts.
t~x}v.spx{.t=.W~t
t;.~u.wtt.
ps|xx~}.u~.x|p.|pxyp}p.
pqt;.wt.rx|x}p{.yxrt.t|.
tutts.D@=E.trt}.~.tp|t}=.
bP\WbP .Sv.Pqt.fp}x}v.
]t~z.6SPf]7.p.p.}px~}p{.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
~qpqx{x.
t.~u.d=b=.w~xp{.xw.
t|tvt}r.stp|t}.6TS7.p}s.p.
stxv}ts.~.~qpx}.x}u~|px~}.~}.TS.
xx.x}.wxrw.|pxyp}p.p.
|t}x~}ts;.prr~}x}v.u~.BE=C.trt}.
~u.x{{xrx.sv.t{pts.TS.
xx=.cwtt.
pt.~|t.{x|xpx~}.t{pts.~.SPf].
spp.~}.TS.
xx;.wxrw.pt.sxrts.
x}.stpx{.x}.Upr~.C;.X.Wx~.p}s.
Rt}._pt}.~u.PqtJ .Upr~.D;.
cwt.br~t;.Spx~};.p}s.bxv}xuxrp}rt.
~u.PqtJ .p}s.Upr~.E;.fwp;.xu.p};.
axz.cwtt.x.~.wt._q{xr.Wtp{w= .
cwtt.upr~.r~}px}.stpx{ts.
sxrx~}.~u.wtt.spp=.
P.}|qt.~u.xz.rp}.~rr.xw.q~w.
prt.p}s.rw~}xr.t.~u.|pxyp}p=.
Stpx{ts.sxrx~}.~u.wt.xz.pt.
psstts.x}.Upr~.A;.brxt}xuxr.
T
xst}rt.~u.x._wp|pr~{~vxrp{.Tuutr;.
xu.Z}~}; .p}s.Upr~.E;.fwp;.xu.p};.
axz.cwtt.x.~.wt._q{xr.Wtp{w= .
q=.cwtt.x.xv}xuxrp}.sx
tx~}.~u.wt.
qp}rt.u~|.{tvxx|pt.sv.
rwp}}t{=.
cwtt.x.p.{prz.~u.t
xst}rt.~u.
xv}xuxrp}.sx
tx~}.~u.|pxyp}p.u~|.
{tvxx|pt.sv.rwp}}t{;.q.wx.x.
{xzt{.st.~.wt.upr.wp.|pxyp}p.x.
|~t.xst{.p
px{pq{t.u~|.x{{xrx.
~rt.pwt.wp}.w~vw.{tvxx|pt.
rwp}}t{=.\pxyp}p.x.}~.p}.USP<.
p~
ts.sv.~sr;.p.p}.]SP.~.
qx~{~vxr.{xrt}t.p{xrpx~}.6Q[P7.wp.
}~.qtt}.p~
ts.u~.|pztx}v.x}.wt.
d}xts.bpt=.]|t~.pt.p}s.wt.
Sxxr.~u.R~{|qxp.wp
t.pt<{t
t{.
|tsxrp{.|pxyp}p.{p.wp.p{{~.u~.
|pxyp}p.t.xwx}.wp.pt=.cwtt.
pt<{t
t{.sv.rwp}}t{.s~.}~.wp
t.
uuxrxt}.r~{{trx~}.~u.spp.t{pts.~.
|tsxrp{.tp|t};.x}r{sx}v.tuuxrpr.
p}s.put=.
\pxyp}p.x.ts.q.ttprwt.u~.
}~}r{x}xrp{.ttprw.p.t{{.p.r{x}xrp{.
ttprw.}st.x}
txvpx~}p{.}t.sv.
6X]S7.p{xrpx~}J.wx.ttt}.wt.
~}{.{tvxx|pt.sv.rwp}}t{.x}.wt.
d}xts.bpt=.W~t
t;.|pxyp}p.ts.
u~.ttprw.ttt}.p.
t.|p{{.
r~}xqx~}.~u.wt.~p{.p|~}.~u.
|pxyp}p.p
px{pq{t.x}.wt.d}xts.bpt;.
p}s.w.~
xst.{x|xts.x}u~|px~}.
pq~.sx
tx~}=.X}.pssxx~};.wt.{prz.~u.
xv}xuxrp}.sx
tx~}.~u.x}
txvpx~}.
{xt.x.{xzt{.qtrpt.~u.wt.
xsttps.p
px{pqx{x.~u.x{{xrx.
|pxyp}p.~u.tp{.~.vtpt.p|~}.~u.
st{p <cWR=.cwt.spp.~xvx}px}v.u~|.
wt.STP.~}.txt.pxxr.
st|~}pt.wt.|pv}xst.~u.wt.
p
px{pqx{x.u~.x{{xrx.|pxyp}p=.STP .
bt|.~.atxt
t.X}u~|px~}.u~|.
Sv.T
xst}rt.6bcaXST7.~
xst.
x}u~|px~}.~}.~p{.s~|txr.sv.
txt=.bcaXST.t~.p.~p{.
s~|txr.txt.~u.DFB;@HD.zv.~u.
|pxyp}p.x}.A?@@;.wt.|~.trt}.tp.
xw.r~|{tt.spp.wp.x.rt}{.
RQ!11111
Hto!11116
Hov!5812
Uhov!5813
647;2!
q{xrp{{.p
px{pq{t.6STP.S~|txr.
Sv.btxt;.}=s=7=.
r=.X}sx
xsp{.pt.pzx}v.wt.qp}rt.
~}.wtx.~}.x}xxpx
t.pwt.wp}.~}.
wt.qpx.~u.|tsxrp{.ps
xrt.u~|.p.
prxx~}t.{xrt}ts.q.{p.~.
ps|x}xt.rw.qp}rt=.
Qtrpt.wt.USP.wp.}~.p~
ts.p}.
]SP.~.Q[P.u~.p.|pxyp}p.sv.
~sr.u~.p}.wtptxr.x}sxrpx~};.
wt.~}{.p.p}.x}sx
xsp{.rp}.pzt.
|pxyp}p.~}.wt.qpx.~u.|tsxrp{.
ps
xrt.w~vw.{tvxx|pt.rwp}}t{.p.
wt.utstp{.{t
t{.x.q.pxrxpx}v.x}.
ttprw.}st.p}.X]S.p{xrpx~}=.cwp.
pxs;.}|t~.pt.p}s.wt.Sxxr.~u.
R~{|qxp.wp
t.pts.pt<{t
t{.
|tsxrp{.|pxyp}p.{p.p{{~x}v.u~.
x}sx
xsp{.~.t.|pxyp}p.}st.
rtpx}.rxr|p}rt=.W~t
t;.spp.
pt.}~.t.p
px{pq{t.~.stt|x}t.wt.
}|qt.~u.x}sx
xsp{.x}v.|pxyp}p.
}st.wtt.pt<{t
t{.|tsxrp{.
|pxyp}p.{p=.atvps{t;.prr~sx}v.~.
wt.A?@A.]bSdW.spp;.@G=H.|x{{x~}.
P|txrp}.ps{.rt}{.t.
|pxyp}p.6bP\WbP;.A?@B7=.Qpts.~}.
wt.{pvt.}|qt.~u.x}sx
xsp{.
t~x}v.rt}.t.~u.|pxyp}p.p}s.
wt.{prz.~u.p}.USP<p~
ts.sv.
~sr.x}.wt.d}xts.bpt;.~}t.rp}.
p|t.wp.x.x.{xzt{.wp.wt.|py~x.
~u.x}sx
xsp{.x}v.|pxyp}p.s~.~.~}.
wtx.~}.x}xxpx
t.pwt.wp}.~}.wt.
qpx.~u.|tsxrp{.ps
xrt.u~|.p.{xrt}ts.
prxx~}t=.
s=.cwt.qp}rt.x.~.t{pts.x}.x.
prx~}.~.p.qp}rt.p{tps.{xts.p.
wp
x}v.p.~t}xp{.u~.pqt.~.|pzt.x.
{xzt{.wp.x.x{{.wp
t.wt.p|t.
~t}xp{.u~.pqt.p.rw.qp}rt;.
w.|pzx}v.x.tp~}pq{t.~.p|t.wp.
wtt.|p.qt.xv}xuxrp}.sx
tx~}.u~|.
{tvxx|pt.rwp}}t{;.xv}xuxrp}.t.
r~}p.~.~.xw~.|tsxrp{.ps
xrt;.
~.wp.x.wp.p.qp}xp{.rppqx{x.~u.
rtpx}v.wpps.~.wt.wtp{w.~u.wt.t.
~.~.wt.put.~u.wt.r~||}x=.
USP.wp.p~
ts.~.sv.~sr.
r~}px}x}v.rp}}pqx}~xs.r~|~}s.wp.
pt.rp{{.t{pts.~.wt.prx
t.
r~|~}t}.x}.|pxyp}p=.cwtt.~.
|pztts.~sr.pt.r~}~{{ts.}st.
wt.RbP=.^}rt.p.trxuxr.sv.~sr.
r~}px}x}v.rp}}pqx}~xs.qtr~|t.
p~
ts;.wp.trxuxr.sv.~sr.
|p.qt.|~
ts.u~|.brwts{t.X.~.p.
sxuutt}.brwts{t.6XX¤e7.}st.wt.
RbP=.Ux{;.\px}~{¥vt}txrp{{.
z}~}.p.s~}pqx}~{¥x.p.brwts{t.XXX.
sv.~sr.r~}px}x}v.}wtxr.
st{p <cWR=.\px}~{;.wxrw.x.
u~|{pts.x}.tp|t.~x{.x}.~u.vt{px}.
rp{t;.p.ux.{prts.x}.brwts{t.XX.
}st.wt.RbP.u~{{~x}v.x.p~
p{.q.
wt.USP=.\px}~{.p.{pt.trwts{ts.
~.brwts{t.XXX.}st.wt.RbP.qtrpt.
~u.{~.}|qt.~u.t~.~u.pqt.
t{px
t.~.|pxyp}p=.S~}pqx}~{.x.
G<^HT^HO^23CWR3/UIO
23CWR3
647;3!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
{xts.x}.brwts{t.X.}st.wt.RbP=.USP.
p~
ts.\px}~{.x}[email protected]~.wt.
tp|t}.~u.}ptp.p}s.
~|xx}v.
p~rxpts.xw.rp}rt.rwt|~wtp.x}.
pxt}.w~.upx{ts.~.t~}s.
pstpt{.~.r~}
t}x~}p{.p}x<t|txr.
tp|t}=.X}.@HHA;.USP.p~
ts.
\px~}p{.u~.p}~txp.p~rxpts.xw.
txvw.{~.x}.pxt}.xw.prxts.
x||}~stuxrxt}r.}s~|t.6PXSb7=.
btr~}s{;.x}.@HGD;.USP.p~
ts.
Rtp|t;.p.sv.~sr.r~}px}x}v.wt.
brwts{t.XX.qp}rt.}pqx{~}t;.u~.wt.
tp|t}.~u.}ptp.p}s.
~|xx}v.
p~rxpts.xw.rp}rt.rwt|~wtp=.
Qtxst.wt.~.rp}}pqx}~xs<r~}px}x}v.
sv.~sr.USP.p~
ts.u~.
|pztx}v;.~wt.}pp{{.~rrx}v.
rp}}pqx}~xs.p}s.wtx.stx
px
t.
6u•~|.Ecppcdku7.p}s.wtx.}wtxr.
tx
p{t}.xw.x|x{p.rwt|xrp{.
rt.p}s.wp|pr~{~vxrp{.prx
x.
pt.x}r{sts.x}.wt.RbP.p.brwts{t.X.
qp}rt=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
3/!Uekgpvkhke!Gxkfgpeg!qh!Kvu!
Rjctoceqnqikecn!Ghhgevu-!kh!Mpqyp!
d}st.wt.tr~}s.upr~;.wt.btrtp.
|.r~}xst.wt.rxt}xuxr.t
xst}rt.~u.
|pxyp}p .wp|pr~{~vxrp{.tuutr=.
Pq}sp}.rxt}xuxr.spp.pt.p
px{pq{t.
~}.wt.}t~rwt|x;.~xr~{~v;.p}s.
wp|pr~{~v.~u.|pxyp}p=.cwx.
trx~}.x}r{st.p.rxt}xuxr.t
p{px~}.
~u.|pxyp}p .}t~rwt|xJ.
wp|pr~{~vJ.p}s.w|p}.p}s.p}x|p{.
qtwp
x~p{;.rt}p{.}t
~.t|;.
r~v}xx
t;.rpsx~
pr{p;.p~}~|xr;.
t}s~rx}~{~vxrp{;.p}s.x||}~{~vxrp{.
t|.tuutr=.cwt.~
t
xt.tt}ts.
qt{~.t{xt.~}.wt.|~.rt}.
ttprw.{xtpt.~}.rp}}pqx}~xs=.
Pgwtqejgokuvt{!cpf!Rjctoceqnqi{!qh!
Octklwcpc!
\pxyp}p.x.p.{p}.wp.r~}px}.
}|t~.}pp{.r~}xt};.rw.p.
rp}}pqx}~xs;.wp.wp
t.p.
pxt.~u.
wp|pr~{~vxrp{.prx~}=.cwt.txx~}.
stux}t.|pxyp}p.p.x}r{sx}v.p{{.
Ecppcdku!r{x
pts.px}=.Sxuutt}.
|pxyp}p.p|{t.stx
ts.u~|.
px~.
r{x
pts.px}.|p.wp
t.
t.
sxuutt}.rwt|xrp{.r~}xt}.
x}r{sx}v.st{p <cWR.p}s.~wt.
rp}}pqx}~xs.6Pt}sx}~.t.p{=;.A?@@7=.
P.p.r~}tt}rt;.|pxyp}p.~sr.
u~|.sxuutt}.px}.x{{.wp
t.sxuutt}.
qx~{~vxrp{.p}s.wp|pr~{~vxrp{.~ux{t=.
Prr~sx}v.~.T{b~w{.p}s.b{pst.
6A??D7.p}s.Pt}sx}~.t.p{=.6A?@@7;.
|pxyp}p.r~}px}.p~x|pt{.DAD.
xst}xuxts.}pp{.r~}xt};.
x}r{sx}v.p~x|pt{.@??.
r~|~}s.r{pxuxts.p.rp}}pqx}~xs=.
Rp}}pqx}~xs.x|px{.tx.x}.
Ecppcdku-!p}s.q{xwts.spp.vvt.
wp.|~.|py~.rp}}pqx}~xs.
r~|~}s.~rrx}v.}pp{{.wp
t.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
qtt}.xst}xuxts.rwt|xrp{{=.]t.p}s.
|x}~.rp}}pqx}~xs.p}s.~wt.}t.
r~|~}s.pt.r~}x}~{.qtx}v.
rwpprtxts.6_~{{p~.t.p{=;.A?@@7=.b~.
up;.~}{.~.rp}}pqx}~xs.
6rp}}pqxvt~{.p}s.x.r~t~}sx}v.
prxs7.wp
t.qtt}.~qpx}ts.u~|.p.}~}<.
Ecppcdku!~rt=.P.b~w.Puxrp}.
Jgnkejt{uwo!6J/!wodtcewnkigtwo7.
prr|{pt.wtt.r~|~}s.
6Pt}sx}~.t.p{=;.A?@@7=.cwt.rwt|x.
~u.|pxyp}p.x.strxqts.x}.|~t.stpx{.
x}.Upr~.B;.cwt.bpt.~u.Rt}.
brxt}xuxr.Z}~{tsvt.atvpsx}v.wt.
Sv.~.^wt.bqp}rt= .
cwt.xt.~u.rp}}pqx}~xs.prx~}.x.p.
wt.rp}}pqx}~xs.trt~=.R{~}x}v.~u.
rp}}pqx}~xs.trt~;.ux.u~|.p.
qpx}.xt.6\psp.t.p{=;.@HH?7.p}s.
wt}.u~|.w|p}.qpx}.xt.6Vtps.t.
p{=;.@HH@7;.wp.
txuxts.wt.xt.~u.prx~}=.
c~.rp}}pqx}~xs.trt~;.RQ p}s.
RQ ;.tt.rwpprtxts.6Qpxp.t.p{=;.
A?@AJ._x~|t{{x;.A??D7=.T
xst}rt.~u.p.
wxs.rp}}pqx}~xs.trt~.tx;.q.x.
wp.}~.qtt}.xst}xuxts.6Qpxp.t.p{=;.
A?@A7=.
cwt.rp}}pqx}~xs.trt~;.RQ p}s.
RQ ;.qt{~}v.~.wt.up|x{.~u.V<~tx}<.
r~{ts.trt~;.p}s.tt}.p.xrp{.
t
t}.p}|t|qp}t<p}}x}v.s~|px}.
rt=.Rp}}pqx}~xs.trt~.{x}z.~.
p}.x}wxqx~.V<~tx}.6V 7;.rw.wp.
pst}{pt.rr{pt.prx
x.x.x}wxqxts.
wt}.p.{xvp}s.qx}s.~.wt.trt~=.
cwx;.x}.};.t
t}.wt.r~}
tx~}.~u.
Pc_.~.wt.tr~}s.|tt}vt;.rr{xr.
P\_.6rP\_7=.Tp|{t.~u.x}wxqx~.
r~{ts.trt~.x}r{st.~x~xs;.
|rpx}xr.rw~{x}tvxr;.p{wp <.
pst}~trt~;.s~p|x}t.6S 7;.p}s.
t~~}x}.6D<Wc 7=.
Rp}}pqx}~xs.trt~.prx
px~}.
x}wxqx.]<.p}s._>`<t.rp{rx|.
rwp}}t{.p}s.prx
pt.x}ps{.
trxux}v.~px|.rwp}}t{.6\przxt.
t.p{=;[email protected]xrwt{{.t.p{=;.@HHF7=.]<.
t.rp{rx|.rwp}}t{.x}wxqxx~}.
strtpt.}t~p}|xt.t{tpt.u~|.
t
tp{.xt=.cw;.rp{rx|.rwp}}t{.
x}wxqxx~}.|p.qt.wt.|trwp}x|.q.
wxrw.rp}}pqx}~xs.x}wxqx.
prt{rw~{x}t;.}~tx}twx}t;.p}s.
v{p|pt.t{tpt.u~|.trxuxr.ptp.~u.
wt.qpx}=.cwtt.tuutr.|p.ttt}.p.
~t}xp{.rt{{{p.|trwp}x|.
}st{x}v.rp}}pqx}~xs .
p}x}~rxrtx
t.p}s.rw~prx
t.tuutr.
6P|t•x;.@HHH7=.
RQ trt~.pt.u~}s.x|px{.x}.
wt.rt}p{.}t
~.t|;.q.pt.p{~.
tt}.x}.txwtp{.xt=.RQ !
trt~.pt.{~rpts.|px}{.x}.wt.qpp{.
vp}v{xp;.wx~rp|;.p}s.rttqt{{|.
~u.wt.qpx}.6W~{t.t.p{=;.A??C7=.cwt.
{~rp{xpx~}.~u.wtt.trt~.|p.
t{px}.rp}}pqx}~xs.x}tutt}rt.xw.
|~
t|t}.r~~sx}px~}.p}s.tuutr.~}.
|t|~.p}s.r~v}xx~}=.Pssxx~}p{{;.
RQ!11111
Hto!11117
Hov!5812
Uhov!5813
RQ trt~.pt.u~}s.x}.wt.x||}t.
t|.p}s.}|t~.~wt.txwtp{.
xt.6_t~rt{{x.p}s.Sx.\p~;.A??H7=.
W~t
t;.wt.r~}rt}px~}.~u.RQ !
trt~.x.r~}xstpq{.{~t.x}.
txwtp{.xt.wp}.x}.wt.rt}p{.
}t
~.t|.6Wtzt}wp}.t.p{=;.@HH?.
p}s.@HHA7=.
RQ trt~.pt.u~}s.x|px{.x}.
wt.x||}t.t|;.q.pt.p{~.tt}.
x}.wt.rt}p{.}t
~.t|.p}s.~wt.
txwtp{.xt=.X}.wt.x||}t.
t|;.RQ trt~.pt.u~}s.
ts~|x}p}{.x}.Q.{|w~rt.p}s.
}pp{.zx{{t.rt{{.6Q~pq~{p.t.p{=;.
@HHB7=.RQ trt~.|p.|tsxpt.
rp}}pqx}~xs .x||}~{~vxrp{.tuutr.
6Vp{xtvt.t.p{=;.@HHD7=.Pssxx~}p{{;.RQ !
trt~.wp
t.qtt}.{~rp{xts.x}.wt.
qpx};.x|px{.x}.wt.rttqt{{|.p}s.
wx~rp|.6V~}v.t.p{=;.A??E7=.cwt.
sxxqx~}.~u.RQ trt~.w~vw~.
wt.q~s.x.{t.tt}x
t.wp}.wt.
sxxqx~}.~u.RQ trt~.6_t~rt{{x.
p}s.Sx.\p~;.A??H7=.W~t
t;.q~w.RQ !
p}s.RQ trt~.pt.tt}.x}.
}|t~.xt.~u.wt.q~s=.
Rp}}pqx}~xs.trt~.wp
t.
t}s~vt}~.{xvp}s=.X}[email protected]}s.@HHD;.
~.t}s~vt}~.rp}}pqx}~xs.trt~.
pv~}x;.p}p}sp|xst.p}s.pprwxs~}{.
v{rt~{.6A<PV7;.ttrx
t{;.tt.
xst}xuxts.6Sx.\p~;.A??E7=.
P}p}sp|xst.x.p.{~.tuuxrpr.pv~}x.
6Qtx
~vt{.p}s.Rwx{st;.A???7.p}s.A<PV.
x.p.wxvw.tuuxrpr.pv~}x.6V~}x~tz.t.
p{=;.A???7=.Rp}}pqx}~xs.t}s~vt}~.
{xvp}s.pt.tt}.x}.rt}p{.p.t{{.p.
txwtp{.xt=.P.r~|qx}px~}.~u.
pzt.p}s.ws~{x.t|x}pt.wt.
prx~}.~u.wt.t}s~vt}~.{xvp}s=.cwt.
t}s~vt}~.rp}}pqx}~xs.t|.x.p.
{~rp{{.prx
t.xv}p{x}v.t|.wp;.~.
wt{.t~t.w~|t~px;.x.prx
pts.
~}.st|p}s .x}.t~}t.~.rwp}vt.~.
wt.{~rp{.w~|t~px.6_t~rt{{x.p}s.Sx.
\p~;.A??H7=.cwt.t}s~vt}~.
rp}}pqx}~xs.t|;.x}r{sx}v.wt.
t}s~vt}~.rp}}pqx}~xs.p}s.wt.
rp}}pqx}~xs.trt~;.st|~}pt.
qp}xp{.{pxrx.x}.t~}t.~.
t
tp{.wx~{~vxrp{.p}s.pw~{~vxrp{.
x|{x.6_t~rt{{x.p}s.Sx.\p~;.A??H7=.
cwx.{pxrx.x.pxr{p{.t
xst}.x}.
wt.rt}p{.}t
~.t|=.
St{p <cWR.p}s.rp}}pqxsx~{.6RQS7.p•t.
~.pq}sp}.rp}}pqx}~xs.tt}.x}.
|pxyp}p=.\pxyp}p .|py~.
rw~prx
t.rp}}pqx}~xs.x.st{p <cWR.
6fprwt{.t.p{=;.A??A7=.X}.@HEC;.Vp~}x.
p}s.\trw~{p}.ux.strxqts.st{p <.
cWR .rt.p}s.u}rx~}=.X}.@HEB;.
\trw~{p}.p}s.bw
~.ux.strxqts.
RQS .rt=.cwt.wp|pr~{~vxrp{.
prx~}.~u.RQS.wp
t.}~.qtt}.u{{.
sxts.x}.w|p}=.
St{p <cWR.p}s.RQS.wp
t.
px}v.
puux}x.p}s.tuutr.p.wt.rp}}pqx}~xs.
trt~=.St{p <cWR.sx{p.x|x{p.
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
wxrw.~u.wt.~.qp.wt.p}x|p{.
tt.|~t.~ut}=.
X}.pssxx~}.~.w|p}.6[x{t.t.p{=;.
A??HJ.[x{t.t.p{=;.A?@@7;.x.wp.qtt}.}~ts.
wp.p}x|p{;.x}r{sx}v.|~}zt.
6\r\pw~};.A??H7;.|xrt.6\r\pw~}.t.
p{=;.A??G7;.p}s.p.6V~{s.t.p{=;.@HHA7;.
pt.pq{t.~.sxrx|x}pt.rp}}pqx}~xs.
u~|.~wt.sv.~.{prtq~=.\~t~
t;.
wt.|py~.prx
t.|tpq~{xt.~u.st{p <.
cWR;.@@<ws~<st{p <cWR;.p{~.
vt}tp{xt.6u~{{~x}v.~p{.
ps|x}xpx~}7.~.wt.x|{.rt.
t{xrxts.q.st{p <cWR.6Q~}t.p}s.
ftx|p};.@HG@7=.ct}<~.~wt.
rp}}pqx}~xs.u~}s.x}.|pxyp}p.p{~.
u{{.qxt.u~.st{p <cWR=.
W~t
t;.RQS.s~t.}~.qxt.u~.
st{p <cWR.x}.p.6ep}}.t.p{=;.A??G7=.
Sxrx|x}px
t.x|{.tuutr.~u.
st{p <cWR.pt.wp|pr~{~vxrp{{.
trxuxr.u~.|pxyp}p.r~}px}x}v.
P}x|p{.Qtwp
x~p{.Tuutr.
rp}}pqx}~xs.6Qp{t.p}s._tr~;.
@HHAJ.Q~}t.p}s.ftx|p};.@HG@J.
bt{u<Ps|x}xpx~}.
fx{t.t.p{=;.@HHB;.@HHD7=.cwt.
bt{u<ps|x}xpx~}.x.p.|tw~s.wp.
sxrx|x}px
t.x|{.tuutr.~u.wt.
ptt.wt.pqx{x.~u.p.sv.~.~srt.
rp}}pqx}~xs.v~.ptp.~.~
xst.
tpsx}v.tuutr=.cwt.tt}rt.~u.
}xt.tuutr.qtrpt.x|{p};.
tpsx}v.tuutr.x}rtpt.wt.
wp{{rx}~vt};.~x~xs;.
{xzt{xw~~s.~u.qtwp
x~p{.t~}t.~.
R~}sxx~}ts._{prt._tutt}rt.
qt}~sxptx}t;.qpqxpt;.]\SP.
~qpx}.pssxx~}p{.sv=.P}x|p{.t{u<.
R~}sxx~}ts.{prt.tutt}rt.6R__7.x. p}pv~}x;.p}s.p}xrw~xr.s~.}~.
ps|x}xpx~}.~u.p.sv.x.~ut}.tu{.
u{{.qxt.u~.st{p <cWR=.
p.{t.xv~~.|tw~s.wp}.t{u<.
x}.tsxrx}v.tpsx}v.tuutr.x}.
ps|x}xpx~}.u~.stt|x}x}v.wtwt. Egpvtcn!Pgtxqwu!U{uvgo!Ghhgevu!
w|p};.p}s.x.x}sxrpx
t.~u.pqt.
~.}~.p.sv.wp.tpsx}v.~txt=.
{xpqx{x=.P.v~~s.r~t{px~}.x.~ut}.
W|p}._wx~{~vxrp{.p}s._rw~{~vxrp{.
X}.wx.qtwp
x~p{.t;.p}x|p{.t}s.
~qt
ts.qttt}.w~t.sv.wp.
Tuutr.
x|t.x}.~.sxx}r.t}
x~}|t}I.~}t.
wt.|~}zt.t{u<ps|x}xt.p}s.
w~t.sv.wp.w|p}.pqt.6Qp{t. wtt.wt.t
x~{.trtx
ts.p.sv.
_rw~prx
t.Tuutr.
p}s.~}t.wtt.wt.trtx
ts.p.{prtq~=.
p}s.Qxvt{~;.A??B7=.X}xxp{{;.
Qt{~.x.p.{x.~u.wt.r~||~}.
Xu.wt.sv.x.tx}u~rx}v;.p}x|p{.x{{.
ttprwt.r~{s.}~.tpq{xw.t{u<.
qytrx
t.t~}t.~.rp}}pqx}~xs.
rw~~t.~.t}s.|~t.x|t.x}.wt.
ps|x}xpx~}.~u.rp}}pqx}~xs;.
6Psp|.p}s.\px};[email protected]~}p{t;.
t}
x~}|t}.pxts.xw.wt.sv;.
x}r{sx}v.st{p <cWR;.x}.p}x|p{.
A??FJ.W~{{xt.@HGE;[email protected]}xt.~u.
|~st{=.W~t
t;.t{u<ps|x}xpx~}.~u. pwt.wp}.xw.wt.{prtq~;.wt}.
\tsxrx}t;.@HGA7=.Prr~sx}v.~.
tt}ts.xw.q~w.~x~}.
st{p <cWR.rp}.}~.qt.tpq{xwts.x}.
\p{s~}ps~.6A??A7;.wtt.t~}t.~.
=x|{p}t~{=.
p.
pxt.~u.p}x|p{.|~st{.}st.
|pxyp}p.pt.{tppq{t.~.|p}.
P}x|p{.w~.R__.~.st{p <cWR;.q. w|p}.p}s.pt.~ut}.p~rxpts.xw.
trxuxr.px}x}v.ppsxv|.6Yx}~
p.t.
~}{.p.wt.{~t.s~t.tts.6?=?FD¤.
p{=;.A??B;.A??C;.A??D7=.
sv<ttzx}v.p}s.sv<pzx}v=.Wxvw.
bxt{.|~}zt;.xw.p}s.xw~.
@=?.|v>zv;.x}ptx~}tp{.6x==77.6Qpxsp. {t
t{.~u.~xx
t.rw~prx
t.tuutr.
x~.t~t.~.~wt.sv.~u.pqt;.
t.p{=;.A??C7=.ba@C@[email protected]}s.}p{~~}t.
pt.p~rxpts.xw.x}rtpts.|pxyp}p.
t{u<ps|x}xt.st{p <cWR.}st.
p}pv~}xt.wx.tuutr.6Qpxsp.t.p{=;.
t;.pqt;.p}s.stt}st}rt.6brwtt.t.
trxuxr.r~}sxx~}=.U~.x}p}rt;.cp}sp. A??C7=.P.p.pxp{.pv~}x;.ba@C@F@E.
p{=;.A??HJ.itxvt.t.p{=;.A?@?7=.
t.p{=.6A???7.~qt
ts.wp.wt}.xt{. rp}.x}srt.R__.p.s~t.~u.?=AD;.?=D;.A.
[email protected]x}wxqxx~};.t{ppx~};.
|~}zt.pt.x}xxp{{.px}ts.~.t{u<.
p}s.B.|v>zv.6Rwtt.t.p{=;.A???7=.X}.
x}rtpts.~rxpqx{x;.p}s.p{zpx
t}t=.
ps|x}xt.x}p
t}~.r~rpx}t;.wt.
z}~rz~.|xrt;.w~t.xw~.I<~•x~xs.
6A7.X}rtpts.|tx|t}.p}s.ptxt;.
x{{.r~}x}t.~.qp<t.st{p <cWR.p. trt~.s~.}~.st
t{~.R__.~.st{p <. p}s.t
t}.twx{ppx~}.p.wxvw.s~t=.
wt.p|t.pt.p.wt.~{s.xw.
cWR.6Vw~{p}s.t.p{=;.A??A7=.
6B7.T}wp}rts.t}~.trtx~};.
r~rpx}t=.cwt.s~t.tt.}~pq{.
wxrw.rp}.vt}tpt.p}.x}rtpts.
Sv.Sxrx|x}px~}.bsxt.
r~|ppq{t.~.w~t.s~t.ts.q.
ptrxpx~}.~u.|xr;.p;.p}s.~rw=.
w|p}.w~.|~zt.|pxyp}p=.
Sv.sxrx|x}px~}.x.p.|tw~s.
6C7.Wtxvwt}ts.x|pvx}px~};.wxrw.
ba@C@F@E;.p.RQ rp}}pqx}~xs.trt~.
wtt.p}x|p{.x}sxrpt.wtwt.p.t.
rp}.{tps.~.p.qytrx
t.t}t.~u.
pv~}x<p}pv~}x;.rp}.q{~rz.wx.
sv.~srt.wxrp{.~.rwxr.
x}rtpts.rtpx
x=.
tpsx}v.tuutr=.^wt.sxt.w~.
trtx~}.x|x{p.~.w~t.~srts.
6D7.X}xxp{.sxx}t;.}ptp;.
wp.}p'¶
t.xt{.|~}zt.rp}.qt.
q.p.z}~}.sv.~u.pqt=.X}.wx.t;.
prwrpsxp;.uprxp{.u{wx}v;.s.|~w;.
rrtu{{.px}ts.~.t{u<ps|x}xt.
p}.p}x|p{.{tp}.~.t.~}t.qp.wt}. p}s.t|~=.
st{p <cWR.x}p
t}~{.6Yx}~
p.t.
x.trtx
t.wt.z}~}.sv.~u.pqt.p}s.
6E7.Sx~vp}xts.wx}zx}v;.x}pqx{x.~.
p{=;.A??B7=.cwt.|px|p{.t~}sx}v.pt. p}~wt.qp.wt}.x.trtx
t.{prtq~=.c~. r~}
tt.{~vxrp{{;.x|t.sx~x~};.p}s.
x.C.Iv>zv.t.x}ytrx~};.wxrw.x.A¤B.
stt|x}t.wtwt.wt.t.sv.x.{xzt.
w~<t|.|t|~.x|px|t}=.
x|t.vtpt.wp}.~qt
ts.x}.t
x~. wt.z}~}.sv.~u.pqt;.p.rwp{{t}vt.
6F7.Ppxp.p}s.x|pxts.ysv|t};.
tx~}.xw.wt.t.sv.st|~}pt. wxrw.rp}.x|tst.sx
x}v.pqx{x.~.
sxt.x}v.r~rpx}t<ttxt}rts.
puux}x.u~.RQ p}s.RQ trt~;.q.
qtwp
t.p.p.tpz.pv~}x.u~.RQ !
trt~=.cwt.xst}xuxrpx~}.~u.
}wtxr.rp}}pqx}~xs.{xvp}s.wp.
t{trx
t{.qx}s.~.RQ trt~.q.s~.
}~.wp
t.wt.xrp{.st{p <cWR<{xzt.
rw~prx
t.~txt.vvt.wp.
wt.prx
px~}.~u.RQ <trt~.|tsxpt.
rp}}pqx}~xs .rw~~xr.tuutr.
6Wp}.t.p{=;.@HHH7=.RQS.wp.{~.
puux}x.u~.q~w.RQ p}s.RQ trt~.
6\trw~{p|.t.p{=;.A??F7=.Prr~sx}v.~.
\trw~{p|.t.p{=.6A??F7;.RQS.wp.
p}pv~}xxr.tuutr.p.RQ trt~.p}s.
~|t.x}
tt.pv~}xxr.~txt.p.RQ !
trt~=.fwt}.rp}}pqx}~xs.pt.vx
t}.
qprt{.~.p;.RQ trt~.s~}<.
tv{pt.p}s.wt.qx}sx}v.~u.wt.tr~}s.
|tt}vt.t|.r~{ts.~.RQ !
trt~;.Vc_vp}|pb;.strtpt.
6Qtx
~vt{.t.p{=;.A??@7=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
647;4!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
|~}zt=.]p{t~}t;.p.|<~x~xs.
p}pv~}x;.pxp{{.p}pv~}xt.wtt.
tpsx}v.tuutr.~u.st{p <cWR.
6Yx}~
p.t.p{=;.A??C7=.
Pssxx~}p{{;.spp.st|~}pt.wp.
}st.trxuxr.r~}sxx~};.~st}.t{u<.
ps|x}xt.rp}}pqx}~xs=.ap.x{{.t{u<.
ps|x}xt.st{p <cWR.wt}.p{xts.
x}prttq~
t}xr{p{.6x=r=
=7;.q.
~}{.p.wt.{~t.s~t.tts.6?=?@¤.
?=?A.Iv>x}ux~}7.6Qpxsp.t.p{=;.A??C7=.
ba@C@[email protected]}s.wt.~x~xs.p}pv~}x.
}p{~~}t.rp}.p}pv~}xt.wx.tuutr=.
W~t
t;.|~.sxt.x}
~{
t.~st}.
t{u<ps|x}xpx}v.wt.}wtxr.
rp}}pqx}~xs.fX].DDA@A;.p.RQ trt~.
pv~}x.xw.p.}~}<rp}}pqx}~xs.
rt.6Stxp}p.t.p{=;.A??FJ.Up~t.t.
p{=;.A??FJ.\pt{{~p.t.p{=;.@HHGJ.
\t}sxpqp{.t.p{=;.A??E7=.
P
tx
t.tuutr;.pwt.wp}.
tx}u~rx}v.tuutr;.~rr.x}.p.wp.
trtx
ts.wxvw.s~t.~u.fX].DDA@A.
6Rwpt~}.t.p{=;.@HHG7.~.st{p <cWR.
6bp}s~<_t}p.t.p{=;.@HHF7;.x}sxrpx}v.p.
~xq{t.rxxrp{.s~t<stt}st}.tuutr=.
X}.q~w.sxt;.ba@C@F@E.t
tts.
wtt.p
tx
t.tuutr=.
RQ!11111
Hto!11118
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
647;5!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
{tps.~.p}.x}rtpt.x}.xz<pzx}v.
qtwp
x~=.
6G7.X{{x~};.st{x~};.p}s.
wp{{rx}px~}.wp.x}t}xu.xw.wxvwt.
s~t=.
6H7.T|~x~}p{.{pqx{x;.x}r~}vx.~u.
puutr;.sw~xp;.pvxpx~};.pp}~xp;.
r~}ux~};.s~x}t;.p}s.p}xr.
pprz;.wxrw.pt.|~t.r~||~}.x}.
x}ttxt}rts.~.wxvw<s~ts.t=.
P.xw.|p}.rw~prx
t.sv;.p.
t~} .|tsxrp{;.rwxpxr;.p}s.sv<.
pzx}v.wx~.rp}.x}u{t}rt.wt.
x}sx
xsp{ .t~}t.~.|pxyp}p=.
S~t.tutt}rt.~.|pxyp}p.~rr.x}.
wp.|pxyp}p.t.tut.wxvwt.
r~}rt}px~}.~u.wt.x}rxp{.
rw~prx
t.qp}rt.6@=HD.trt}.
st{p <cWR7.~
t.{~t.r~}rt}px~}.
6?=EB.trt}.st{p <cWR7.6Rwpx.p}s.
Qzt;.@HHC7=.]~}twt{t;.utt}.
|pxyp}p.t.6M@??.x|t.~u.t7.
tt.pq{t.~.xst}xu.p.sv.tuutr.u~|.
{~<s~t.st{p <cWR.qtt.wp}.
~rrpx~}p{.t.6K@?.x|t.~u.t7.
wx{t.p{~.ttxt}rx}v.utt.tspx
t.
tuutr.u~|.|pxyp}p.6Zxz.p}s.st.fx;.
@HHH7=.
cwt.txx~}t.r~}t}s.wp.|p}.~u.
|pxyp}p .}pp{{.~rrx}v.
rp}}pqx}~xs.|xxvpt.wt.rw~prx
t.
tuutr.~u.st{p <cWR;.p}s.wttu~t.wp.
|pxyp}p.{prz.uuxrxt}.pqt.
~t}xp{.~.pp}.brwts{t.X.
{prt|t};.qtrpt.\px}~{;.wxrw.x.x}.
brwts{t.XXX;.r~}px}.~}{.st{p <cWR=.
cwx.wt~.wp.}~.qtt}.st|~}pts.
x}.r~}~{{ts.sxt=.\~t~
t;.wt.
r~}rt.~u.pqt.~t}xp{.t}r~|pt.
p{{.~txt.~u.p.qp}rt;.x}r{sx}v.
x.rwt|x;.wp|pr~{~v;.p}s.
wp|pr~zx}txr;.p.t{{.p.pvt.
pt}.p}s.sx
tx~}.wx~=.cwt.
pqt.~t}xp{.~u.p.qp}rt.x.
p~rxpts.xw.wt.ttpts.~.~psxr.
t.~u.p.qp}rt.x}.}~}|tsxrp{.
xpx~}.u~.wt.rw~prx
t.tuutr.
wt.qp}rt.~srt=.cwtt.
rw~prx
t.tuutr.x}r{st.tw~xp;.
trtp{.p}s.~wt.r~v}xx
t.
sx~x~};.wp{{rx}px~};.p}s.|~~s.
rwp}vt=.W~t
t;.p.pts.pq~
t;.wt.
pqt.~t}xp{.}~.~}{.x}r{st.wt.
rw~prx
t.tuutr;.q.p{~.x}r{st.
~wt.ptr.t{pts.~.p.qp}rt=.
STP .ux}p{.q{xwts.{t.t}x{ts.
atrwts{x}v.~u.wt.U~~s.p}s.Sv.
Ps|x}xpx~}.P~
ts._~sr.
R~}px}x}v.b}wtxr.S~}pqx}~{.j6¤7<.
st{p <6p}7<ctpws~rp}}pqx}~{l.x}.
btp|t.^x{.p}s.T}rp{pts.x}.b~u.
Vt{px}.Rp{t.U~|.brwts{t.XX.~.
brwts{t.XXX .6EC.Ua.BDHAG;.Y{.A;.
@HHH7.trwts{ts.\px}~{.u~|.
brwts{t.XX.~.brwts{t.XXX=.cwt.WWb.
pt|t}.~u.wt.pqt.~t}xp{.p}s.
qtt}.rwts{x}v.tr~||t}spx~}.
r~|pts.\px}~{.~.|pxyp}p.~}.
sxuutt}.ptr.t{pts.~.pqt.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
~t}xp{=.\py~.sxuutt}rt.x}.
u~|{px~};.p
px{pqx{x;.p}s.pvt.
qttt}.|pxyp}p.p}s.wt.sv.
~sr;.\px}~{;.r~}xqt.~.wtx.
sxuutx}v.pqt.~t}xp{=.
W~{{xt.p}s.Vx{{txt.6@HFB7.
tx|pts.wp.st{p <cWR.q.|~zx}v.x.
A=E.~.B.x|t.|~t.~t}.wp}.st{p <.
cWR.x}vtts.~p{{=.cwt.x}t}t.
rw~prx
t.sv.tuutr.prwxt
ts;.
pxs{.q.|~zx}v.x.vt}tp{{.
r~}xstts.~.~srt.wt.tuutr.stxts.
q.wt.pqt=.cwx.tuutr.t{px}.w.
pqt.~ut}.tut.~.ps|x}xt.
rtpx}.sv.q.x}wp{px~};.
x}p
t}~{;.~.x}p}pp{{.pwt.
wp}.~p{{=.brw.x.wt.rpt.xw.
r~rpx}t;.~x|;.wt~x};.wt}rr{xsx}t;.
|twp|wtp|x}t;.p}s.st{p <cWR.
u~|.|pxyp}p.6?=@¤H=D.trt}.st{p <.
cWR.p}vt7.~.wpwxw.6@?¤B?.trt}.
st{p <cWR.p}vt7.6ft~}.p}s.
fpwq};.@HH?7=.cw;.wt.st{pts.
~}t.p}s.{~}vt.spx~}.~u.prx~}.u~.
\px}~{.|p.qt.r~}xqx}v.upr~.
{x|xx}v.wt.pqt.~.ptp{.~u.\px}~{.
p.p.sv.~u.pqt.t{px
t.~.|pxyp}p=.
cwt.u~|{px~}.~u.\px}~{.x.p.upr~.
wp.r~}xqt.~.sxuutt}xp{.
rwts{x}v.~u.\px}~{.p}s.|pxyp}p=.
U~.tp|{t;.tprx~}.p}s.xuxrpx~}.
~u.s~}pqx}~{.u~|.wt.t}rp{pts.
tp|t.~x{.|xt.~u.\px}~{.x.wxvw{.
r~|{t.p}s.sxuuxr{=.Pssxx~}p{{;.wt.
tt}rt.~u.tp|t.~x{.|xt.x}.wt.
u~|{px~}.|p.tr{st.wt.|~zx}v.
~u.\px}~{<{prts.rxvptt=.
Pssxx~}p{{;.wtt.x.p.sp|pxr.
sxuutt}rt.qttt}.prp{.pqt.p}s.
x{{xrx.puuxrzx}v.~u.\px}~{.p}s.
|pxyp}p=.Stxt.\px}~{ .
p
px{pqx{x.x}.wt.d}xts.bpt;.wtt.
wp
t.qtt}.}~.xv}xuxrp}.t~.~u.
pqt;.sx
tx~};.~.q{xr.wtp{w.
~q{t|.st.~.\px}~{=.Q.
r~|px~};.@G=H.|x{{x~}.P|txrp}.
ps{.t~.rt}{.x}v.|pxyp}p.
6bP\WbP;.A?@B7=.
X}.pssxx~};.USP .p~
p{.~u.p}.
]SP.u~.\px}~{.p{{~ts.u~.\px}~{.~.
qt.trwts{ts.~.brwts{t.XX;.p}s.
qtt}{.~.brwts{t.XXX.~u.wt.RbP=.
X}.r~}r{x~};.|pxyp}p.p}s.\px}~{.
sxuut.~}.p.xst.
pxt.~u.upr~.wp.
r~}xqt.~.tprw.qp}rt .pqt.
~t}xp{=.cwtt.sxuutt}rt.pt.|py~.
tp~}.sxx}vxwx}v.wt.wxvwt.pqt.
~t}xp{.u~.|pxyp}p.p}s.wt.sxuutt}.
rwts{x}v.stt|x}px~}.~u.|pxyp}p.
p}s.\p•x}~{=.
X}.t|.~u.wt.txx~}t .r{px|.wp.
sxuutt}.rp}}pqx}~xs.tt}.x}.
|pxyp}p.|xxvpt.wt.rw~prx
t.
tuutr.~u.st{p <cWR;.~}{.wtt.~u.wt.
rp}}pqx}~xs.tt}.x}.|pxyp}p.tt.
x|{p}t~{.ps|x}xtts.xw.
st{p <cWR.~.tp|x}t.w~.wt.
r~|qx}px~}.~u.wtt.rp}}pqx}~xs.
rw.p.RQS;.rp}}pqxrw~|t}t.6RQR7.
RQ!11111
Hto!11119
Hov!5812
Uhov!5813
p}s.rp}}pqx}~{.6RQ]7.x}u{t}rt.st{p <.
cWR .rw~prx
t.tuutr=.Sp{~}.t.p{=.
6@HFE7.~qt
ts.wp.|~zts.
ps|x}xpx~}.~u.{prtq~.|pxyp}p.
rxvptt.r~}px}x}v.x}ytrx~}.~u.?=@D.
|v>zv.RQS.r~|qx}ts.xw.?=?AD|v>zv.
~u.st{p <cWR;.x}.p.FI@.px~.~u.RQS.~.
st{p <cWR;.xv}xuxrp}{.strtpts.
px}v.~u.prt.qytrx
t.tuutr.p}s.
wxvw .wt}.r~|pts.~.|~zx}v.
st{p <cWR.p{~}t=.X}.r~}p;.X{p}.t.p{=.
6A??D7.rp{r{pts.wt.}pp{{.
~rrx}v.r~}rt}px~}.~u.RQR.p}s.
RQS.x}.p.qprw.~u.|pxyp}p.rxvptt.
xw.txwt.@=G.trt}.~.B=E.trt}.
st{p <cWR.r~}rt}px~}.q.txvw=.U~.
tprw.t}vw.~u.st{p <cWR.x}.
|pxyp}p.rxvptt;.wt.r~}rt}px~}.
~u.RQR.p}s.RQS.tt.r{pxuxts.x}.
v~.~u.txwt.{~.~.wxvw=.cwt.s.
pxts.wt.p|~}.~u.RQR.p}s.RQS.
xwx}.tprw.t}vw.~u.st{p <cWR.
|pxyp}p.rxvptt;.xw.
ps|x}xpx~}.r~}xx}v.~u.txwt.{~.
RQR.6qttt}.?=@¤?=A.trt}.RQR.
r~}rt}px~}.q.txvw7.p}s.{~.RQS.
6qttt}.?=@¤?=C.trt}.RQS.
r~}rt}px~}.q.txvw7;.wxvw.RQR.
6M?=D.trt}.RQR.r~}rt}px~}.q.
txvw7.p}s.{~.RQS;.~.{~.RQR.p}s.
wxvw.RQS.6M@=?.trt}.RQS.
r~}rt}px~}.q.txvw7=.^
tp{{;.p{{.
r~|qx}px~}.r~ts.xv}xuxrp}{.
vtpt.wp}.{prtq~.~}.px}v.~u.
qytrx
t.tuutr;.p}s.wtt.p.}~.
xv}xuxrp}.sxuutt}rt.qttt}.p}.
r~|qx}px~}=.
cwt.~p{.ps|x}xpx~}.~u.p.
r~|qx}px~}.~u.txwt.@D;.B?;.~.E?.|v.
RQS.xw.B?.|v.st{p <cWR.sx~{
ts.
x}.{xxs.6x}.p.px~.~u.p.{tp[email protected].
~.st{p <cWR7.tsrts.wt.qytrx
t.
tuutr.~srts.q.st{p <cWR.p{~}t.
6Zp}x~{.t.p{=;.@HFC7=.Pssxx~}p{{;.
~p{{.ps|x}xtx}v.p.{xxs.|xt.
r~|qx}x}v.@.|v>zv.RQS.xw.?=D.|v>zv.
~u.st{p <cWR.6p[email protected].~.st{p <.
cWR7.strtpts.r~t.~u.p}xt.p}s.
|pxyp}p.sv.tuutr.~}.wt.Pssxrx~}.
attprw.Rt}t.X}
t}~.6PaRX7.
r~|pts.~.st{p <cWR.p{~}t.6ipsx.
t.p{=;@HGA7=.[p{;.~p{.ps|x}xpx~}.
~u.txwt.@A=D;.AD;.~.D?.|v.RQ].
r~|qx}ts.xw.AD.|v.st{p <cWR.
sx~{
ts.x}.{xxs.6px~.~u.p.{tp.@IA.
RQ].~.st{p <cWR7.xv}xuxrp}{.
x}rtpts.qytrx
t.px}v.~u.
svvts; .s~; .sx; .p}s.
s}z; .r~|pts.~.st{p <cWR.p{~}t.
6Zp}x~{.t.p{=;.@HFD7=.
T
t}.w~vw.~|t.sxt.vvt.wp.
RQS.|p.strtpt.~|t.~u.st{p <cWR .
rw~prx
t.tuutr;.wt.px~.~u.RQS.
~.st{p <cWR.ps|x}xtts.x}.wtt.
sxt.pt.}~.tt}.x}.|pxyp}p.
ts.q.|~.t~{t=.U~.tp|{t;.x}.
~}t.s;.ttprwt.ts.|~zts.
|pxyp}p.xw.px~.~u.RQS.~.st{p <.
cWR.}pp{{.tt}.x}.|pxyp}p.
G<^HT^HO^23CWR3/UIO
23CWR3
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
{p}.|ptxp{.p}s.wt.u~}s.~.wp.
px}v.wt.p|~}.~u.RQS.prp{{.wps.
}~.tuutr.~}.st{p <cWR .rw~prx
t.
tuutr.6X{p} t.p{=;.A??D7=.Qtrpt.|~.
|pxyp}p.rt}{.p
px{pq{t.~}.wt.
tt.wp wxvw.p|~}.~u.st{p
xw.{~.p|~}.~u.RQS.p}s.~wt.
rp}}pqx}~xs;.|~.x}sx
xsp{.t.
|pxyp}p.xw {~.{t
t{.~u.RQS.
tt}.6\tw|tsxr.t.p{=;.A?@?7=.cw;.
p} ~xq{t.|xxvpx~}.~u.st{p <cWR .
rw~prx
t tuutr.q.RQS.x{{.}~.
~rr.u~.|~.|pxyp}p.t=.X}.
r~}p;.~}t.s.x}sxrpts.wp.
p}~wt.rp}}pqx}~xs.tt}.x}.
|pxyp}p;.RQ];.|p.t}wp}rt.st{p
cWR .rw~prx
t.tuutr.6Zp}x~{.t.
Qtwp
x~p{.X|px|t}.
\pxyp}p x}srt.
px~.
rw~prx
t tuutr.wp.rp}.{tps.~.
qtwp
x~p{.x|px|t}=.\pxyp}p .
prt.tuutr rp}.xv}xuxrp}{.x}tutt.
xw.p.t~} .pqx{x.~.{tp}.x}.wt.
r{p~~|.~.~.~tpt.|~~.
twxr{t=.
Prt.ps|x}xpx~}.~u.|~zts.
|pxyp}p.x|px.tu~|p}rt.~}.
{tp}x}v;.p~rxpx
t ~rtt;.p}s.
rw~|~~ qtwp
x~p{.t.6Q{~rz.t.
p{=;.@HHA7=.ap|ptzt.t.p{=.6A??Ep7.
w~ts.wp.prt.ps|x}xpx~}.~u.AD?.
v>zv.~u.st{p
|~zts.|pxyp}p.s~t<stt}st}{.
x|px.r~v}xx~}.p}s.|~~.r~}~{;.
x}r{sx}v.|~~.x|{x
x.p}s.
przx}v.x|px|t}.6ap|ptzt.t.p{=;.
A??Eq7=.bx|x{p{;.ps|x}xpx~}.~u.AH?.
v>zv.st{p <cWR.x}.p.|~zts.|pxyp}p.
rxvptt.t{ts.x}.x|pxts.
trtp{.|~~.tts.p}s.prrprI.
c~.zx{{.wxrw.pt.rxxrp{.~.sx
x}v.
pqx{x.6Zwp{t.t.p{=;.@HHH7=.[p{;.
ps|x}xpx~}.~u.B=HD.trt}.st{p
cWR.x} p.|~zts.|pxyp}p.rxvptt.
}~.~}{.x}rtpts.sxtx{xqx|.
|tpt;.q p{~.x}rtpts.wt.{pt}r.
x}.p.pz.~u.x|{pts.
twxr{t.qpzx}v.p.
p.pt.r~|ppq{t.~.p}.x}rtpt.x}.
~x}v sxp}rt.~u.ux
t.utt.p.E?.|w.
6[xv~x t.p{=;.@HHG7=.W~t
t;.prt.
ps|x}xpx~}.~u.|pxyp}p.r~}px}x}v.
A=@.trt}.st{p <cWR.s~t.}~.
~srt.wp}v~
t.tuutr .6Rwpx;.
X}.pssxx~}.~.|tpx}v.wt.prt.
tuutr.x||tsxpt{.u~{{~x}v.|pxyp}p.
ps|x}xpx~};.ttprwt.wp
t.
r~}srts sxt ~ stt|x}t w~
{~}v qtwp
x~p{.x|px|t}.{p.put.
pqx}t}rt=.b~|t ~u.|pxyp}p .prt.
tuutr |p }~ u{{ t~{
t }x{ p
{tp.~}t.sp.put.wt.prt.
rw~prx
t tuutr.wp
t.qxsts=.
Wtxw|p} t p{= 6@HH?7 w~ts wp
x|px|t}.~}.|t|~.pz.tx.
u~.AC w~.put.|~zx}v.|pxyp}p.
rxvptt r~}px}x}v A=DF trt}
st{p <cWR=.W~t
t;.Up}.t.p{=.6@HHG7.
w~ts.wp.wt.|~}x}v.put.t~t.
~.@=G.trt}.~.B=E.trt}.|~zts.
st{p <cWR; qytr wps |x}x|p{
txsp{.p{tpx~}.x}.qytrx
t.~.
tu~|p}rt.|tpt=.
P }|qt ~u upr~ |p x}u{t}rt
|pxyp}p qtwp
x~p{.tuutr.x}r{sx}v.
wt.spx~} ~u.t.6rw~}xr.~.w~.
t|7; utt}r ~u t 6spx{; ttz{;
~.~rrpx~}p{{7;.p}s.p|~}.~u.t.
6wtp
.~.|~stpt7=.attprwt.p{~.
wp
t tp|x}ts w~ {~}v qtwp
x~p{
x|px|t}.{p.u~{{~x}v.rw~}xr.
|pxyp}p.t=.cwtt.sxt.ts.t{u<.
t~ts wx~xt ~u p spx~};
utt}r;.p}s.p|~}.~u.p.
|pxyp}p.t;.p}s.ps|x}xtts.p.
pxt ~u tu~|p}rt p}s r~v}xx
t
|tpt.p.sxuutt}.x|t.~x}.
u~{{~x}v.|pxyp}p.pqx}t}rt=.X}.
rw~}xr |pxyp}p t; qtwp
x~p{
x|px|t}.|p.tx.u~..~.AG.
sp.~u.pqx}t}rt=.b~{~xy.t.p{=.6A??A7.
st|~}pts wp put @F w~ ~u
pqx}t}rt;[email protected]{.wtp
.rw~}xr.
|pxyp}p.t.tu~|ts.~t.~}.
|t|~ p}s pt}x~} pz wp} BB
}~}<x}v.r~}~{.~[email protected]
;.w~<.
t|.t=.P}~wt.s.}~ts.wp.
wtp
; utt} |pxyp}p t;
pqx}t}.u~.p.{tp.AC.w~;.
tu~|ts.xv}xuxrp}{.~t.wp}.wt.
r~}~{ ~}
tqp{ |t|~ p}s
rw~|~~ tts.t.6\tx}x.t.
p{=;.A??E7=.Pssxx~}p{{;.put.p.{tp.@.
ttz ~u pqx}t}rt; ~}v ps{
utt}.|pxyp}p.t;.pvts.@G¤AG;.
w~ts.stuxrx.x}.rw~|~~.tts;.
px}ts pt}x~}; p}s r~v}xx
t
x}wxqxx~}.6[xspw{.p}s._xrt;.A?@A7=.
Ps{ wtp
;.rw~}xr.|pxyp}p.t.
w~ts stuxrx ~} |t|~ t put F
sp.~u.t
xts.pqx}t}rt.6_~t.t.
p{=;.A??A7=.W~t
t;.wt}.wtt.p|t.
x}sx
xsp{ tt pvpx} tts put AG
sp.~u.pqx}t}rt;.wt.sxs.}~.w~.
xv}xuxrp}.|t|~.stuxrx=.cwt.pw~.
r~}r{sts; rp}}pqx<p~rxpts
r~v}xx
t.stuxrx.pt.t
txq{t.p}s.
t{pts.~ trt}.rp}}pqx.t~t;.
pwt wp} xt
txq{t p}s t{pts ~
r|{px
t.{xutx|t.t= W~t
t;.
~wt.ttprwt.t~ts.
}t~rw~{~vxrp{ stuxrx x} |t|~;
ttrx
t.u}rx~}x}v;.rw~|~~.
tts p}s |p}p{.sttx.x}.wtp
.
|pxyp}p t pqx}t} u~ AG sp
6Q~{{p.t.p{=;.A??A7=.Uwt|~t;.p.
u~{{~<.s ~u.wtp
.|pxyp}p.
t }~ts strxx~}<|pzx}v stuxrx
put.AD.sp.~u.t
xts.pqx}t}rt=.
6Q~{{p.t.p{=;.A??D7=.W~t
t;.|~stpt.
|pxyp}p t sxs }~ w~ strxx~}<
|pzx}v.stuxrx.put.AD.sp.~u.
pqx}t}rt;.vvtx}v.wt.p|~}.~u.
X}.wx.~px~}.wt.t|.x.ts.
x}trwp}vtpq{.u~.
|pxyp}p.t.|p.x|pr.wt.spx~}.
~u.txsp{.x|px|t}=.
cwt.tuutr ~u.rw~}xr.|pxyp}p.t.
s~ }~ tt| ~ tx put |~t wp}
@.~.B.|~}w.~u.pqx}t}rt=.Put.B.
|~}w.~u.pqx}t}rt;.p}.stuxrx.
~qt
ts.x} X`;.x||tsxpt.|t|~;.
st{pts.|t|~;.p}s.x}u~|px~}<.
~rtx}v.tts.u~{{~x}v.wtp
.
|pxyp}p.t.r~|pts.~.t<sv.t.
r~t.tt }~.{~}vt.ppt}.6Uxts.t.
p{=;.A??D7=.\pxyp}p.sxs.}~.ptp.~.
wp
t.{px}v.tuutr.~}.tu~|p}rt.~u.p.
r~|twt}x
t }t~rw~{~vxrp{
qpt.wt} DC.|~}~v~xr.|p{t.
x}.6~}t.~u.w~|.ts.|pxyp}p;.
~}t ~u.w~|.sxs.}~7.tt.r~|pts.@¤.
A?.tp.put.rtpx~}.~u.|pxyp}p.t.
6[~}.t.p{=;.A??C7=.bx|x{p{;.u~{{~x}v.
pqx}t}rt.u~.p.tp.~.|~t;.q~w.{xvw.
p}s.wtp
.ps{.|pxyp}p.t.sxs.}~.
w~.stuxrx.~}.r~t.~u.
tqp{.
|t|~.r~|pts.~.}~}<x}v.r~}~{.
6cpx t p{=; A?@@7= Prr~sx}v ~ p trt}
|tp<p}p{x.{~~zx}v p.}~}<prt.p}s.
{~}v<{px}v.tuutr.~u.|pxyp}p.t.~}.
}t~r~v}xx
t.tu~|p}rt;.p}.
stuxrx.tt}.xwx}.wt.ux.|~}w.
u~{{~x}v.pqx}t}rt.pt.vt}tp{{.}~.
tt}.put.pq~.@.|~}w.~u.
pqx}t}rt.6brwtx}t.p}s.S}};.A?@A7=.
P}~wt.ptr.wp.|p.qt.p.rxxrp{.
upr~.x}.wt.x}t}x.p}s.txt}rt.~u.
x|px|t}.t{x}v.u~|.rw~}xr.
|pxyp}p.t.x.wt.pvt.~u.ux.t=.
X}sx
xsp{.xw.p.sxpv}~x.~u.
|pxyp}p.|xt.~.stt}st}rt.w~.
tt.ttzx}v.tp|t}.u~.qp}rt.
t;.w~.x}xxpts.|pxyp}p.t.qtu~t.
wt.pvt.~u.@D.tp;.w~ts.stuxrx.x}.
tu~|p}rt.~}.pz.ptx}v.
px}ts.pt}x~};.x|{t.r~}~{;.
p}s.vt}tp{ ttrx
t.u}rx~}x}v.
r~|pts.~.}~}<x}v.r~}~{=.cwtt.
stuxrx.tt.}~.tt}.x}.x}sx
xsp{.
w~ x}xxpts.|pxyp}p.t.put.wt.
pvt.~u @D.tp.6U~}t.t.p{=;.A?@@7=.
bx|x{p{;.wtp
;.rw~}xr.|pxyp}p.
t.w~ qtvp}.x}v.|pxyp}p.qtu~t.
wt.pvt.~u.@E.tp.wps.vtpt.
strt|t}.x}.ttrx
t.u}rx~}x}v.
pz.wp}.wtp
;.rw~}xr.|pxyp}p.
t.w~ pts.x}v.put.wt.pvt.~u.
@E.tp.p}s.}~}<x}v.r~}~{.6Vqt.
t.p{=;.A?@A7=.Pssxx~}p{{;.x}.p.
~trx
t {~}vxsx}p{.qxw.r~w~.
s.~u.@;?BF.x}sx
xsp{;.|pxyp}p.
stt}st}rt.~.rw~}xr.|pxyp}p.t.
p.p~rxpts.xw.p.strtpt.x}.X`.
p}s.vt}tp{ }t~rw~{~vxrp{.
tu~|p}rt.r~|pts.~.t<|pxyp}p.
t~t.{t
t{.x}.ps~{trt}.~}t.
t.6\txt t.p{=;.A?@A7=.cwt.str{x}t.x}.
ps~{trt}<~}t t .X`.txts.
t
t}.put.tsrx~}.~.pqx}t}rt.~u.
|pxyp}p.t.u~.p.{tp.@.tp=.X}.
r~}p;.wt ps{<~}t.rw~}xr.
|pxyp}p t w~ts }~ xv}xuxrp}
rwp}vt x}.X`.r~|pts.~.t<t~t.
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
{t
t{ wtwt.wt.tt.rt}.t.
~.pqx}t}.u~.p.{tp.@.tp.6\txt.t.
X}.pssxx~}.~.wt.pvt.~u.~}t.~u.t;.
~|t.t
xst}rt.vvt.wp.wt.p|~}.
~u.|pxyp}p.ts.|p.t{pt.~.wt.
x}t}x.~u.x|px|t}=.X}.wt.pq~
t.
s.q.Vqt.t.p{=.6A?@A7;.wtt.
tp{<~}t t wps vtpt stuxrx
wp}.{pt<~}t.t;.wt.tp{<~}t.
t.t~ts.x}v.|pxyp}p.xrt.p.
~ut} p}s x}v.wtt.x|t.p.|rw.
|pxyp}p.t.ttz.wp}.wt.{pt<~}t.
t= \txt.t.p{=.6A?@A7.w~ts.wp.
wt.stuxrx.x}.X`.tt}.x}.ps~{trt}<.
~}t.t.x}rtpts.xw.wt.p|~}.
~u.|pxyp}p.ts=.\~t~
t;.wt}.
r~|px}v.r~t u~.|tpt.~u.X`;.
x||tsxpt.|t|~;.st{pts.|t|~;.
p}s.x}u~|px~}<~rtx}v tts.~.
t<sv<t.{t
t{;.wt.rt};.wtp
;.
rw~}xr.|pxyp}p.t.w~ts.stuxrx.
x}.p{{.wtt.|tpt.wx{t.rt};.
~rrpx~}p{.|pxyp}p.t.sxs.}~.
6Uxts.t.p{=;.A??D7=.
P~rxpx~}.~u.\pxyp}p.dt.fxw.
_rw~x.
cwx.p}p{x.t
p{pt.~}{.wt.
t
xst}rt.u~ p.sxtr.{x}z.qttt}.x~.
|pxyp}p.t.p}s.wt.qtt}.
st
t{~|t} ~u.rw~x=.cw;.wx.
sxrx~}.s~t.}~.r~}xst.xt.
rw.p.wtwt.|pxyp}p .p}xt}.
tuutr.pt.x|x{p.~.rw~xr.
|~|.x}.wtp{w.x}sx
xsp{.~.
tprtqpt.rw~xr.|~|.x}.
x}sx
xsp{.p{tps.sxpv}~ts.xw.
rwx~wt}xp=.
Tt}x
t.ttprw.wp.qtt}.
r~}srts.~ x}
txvpt.wtwt.
t~t.~ |pxyp}p.x.p~rxpts.xw.
wt.st
t{~|t}.~u.rwx~wt}xp.~.
~wt.rw~t=.P{w~vw.|p}.sxt.
pt.|p{{.p}s.x}utt}xp{;.~wt.sxt.
x}.wt {xtpt.t.w}sts.~.
w~p}s.~u.qytr=.P.tt};.wt.
p
px{pq{t.spp s~.}~.vvt.p.rppx
t.
{x}z.qttt}.|pxyp}p.t.p}s.wt.
st
t{~|t} ~u.rw~x.6\x}~x.t.
p{=;.A?@?7=.]|t~.{pvt;.{~}vxsx}p{.
sxt.w~ wp.qytr.w~.ts.
|pxyp}p.s~.}~.wp
t.p.vtpt.
Qtwp
x~p{.Tuutr.~u._t}pp{.T~t.
x}rxst}rt.~u rw~xr.sxpv}~t.
r~|pts.~.w~t.w~.s~.}~.t.
bsxt.xw.rwx{st}.p.sxuutt}.
|pxyp}p.6Utv~}.t.p{=;.A??DJ.
pvt.~u.st
t{~|t}.pt.ts.~.
Ztt.t.p{=;.A?@@J.ep}.^.t.p{=;.
tp|x}t.wt.x|pr.~u.t}pp{.
|pxyp}p.t~t.~}.tu~|p}rt.x}.p.
fwt}.p}p{x}v.wt.p
px{pq{t.
txt.~u.r~v}xx
t.pz=.W~t
t;.|p}.
t
xst}rt.~u.wt.r~}}trx~}.qttt}.
tv}p}.~|t}.w~.t~ts.
rw~x.p}s.|pxyp}p;.x.x.rxxrp{.~.
|pxyp}p.t.tt.|~t.{xzt{.~.p{~.
stt|x}t wtwt.wt.qytr.x}.wt.
t~.t.~u.p{r~w~{;.~qprr~;.p}s.
sxt.pt.pxt}.w~.pt.p{tps.
r~rpx}t.6V~{srw|xs.t.p{=;.A??G7=.
sxpv}~ts xw.rw~x.~.x}sx
xsp{.
cw;.xw ~t}xp{.t~t.~.
w~ st|~}pt.p.{x|xts.}|qt.~u.
|{x{t.sv;.x.x.sxuuxr{.~.
|~|.p~rxpts.xw.rw~x.
stt|x}t wt.trxuxr.x|pr.~u.
xw~.p{xux}v.u~.p.sxpv}~x.~u.wt.
t}pp{.|pxyp}p t~t=.
sx~st=.U~.tp|{t;.x}tps.~u.x}v.
\~.sxt.ptx}v.wt.qtwp
x~p{. p.sxpv}~x.~u.rw~x;.~|t.
ttprwt.t{xts.~}.}~}<p}sps.
tuutr ~u t}pp{ |pxyp}p t~t
|tw~s.~u.ttt}x}v.|~|.~u.
x}r{sts.~|t}.w~;.x}.pssxx~}.~.
rw~x.x}r{sx}v.rwx~wt}xr.
x}v.|pxyp}p; p{~.t~ts.x}v.
r{t 6\pt||p}x.t.p{=;.A??C7;.
p{r~w~{.p}s.~qprr~=.W~t
t;.~|t.
qr{x}xrp{.rw~xr |~| .6ep}.
t
xst}rt.vvt.p}.p~rxpx~}.
Vpt{.t.p{=;.A?@A7;.t<rw~xr.
qttt}.wtp
.t}pp{.|pxyp}p.
t~t.p}s.stuxrx.x}.~|t.r~v}xx
t. r{x}xrp{.wxvw.xz .6ep}.st.\tt.t.p{=;.
s~|px}=.X}.q~w.C<tp<~{s.p}s.E<tp<. A?@A7;.p}s.|~|.t{pts.~.
rw~x.
{}tpqx{x .6Vxuuxw<.
~{s.rwx{st};.wtp
.t}pp{.|pxyp}p.
[t}stx}v t.p{=;.A?@A7=.cwtt.v~x}v.
t.x.}tvpx
t{.p~rxpts.xw.
s~.}~.r~}u~|.~.wt.rxtxp.x}.wt.
tu~|p}rt.~}.pz.ptx}v.|t|~;.
Sxpv}~xr.p}s.bpxxrp{.\p}p{.
tqp{.tp~}x}v;.p}s.p}xpx
t.
6Sb\¤D7.~.wt.X}t}px~}p{.
tp~}x}v.6Uxts.p}s.fpzx}~};.@HGFJ.
R{pxuxrpx~}.~u.Sxtpt.6XRS¤@?7.u~.p.
V~{srw|xs.t.p{=;.A??G7=.Pssxx~}p{{;. sxpv}~x ~u.rw~x=.cw;.wtt.
wtp
.t}pp{.|pxyp}p.t.x.
v~x}v.pt.}~.p~xpt.u~.t.x}.
p~rxpts.xw.stuxrx.x}.|tpt.~u.
t
p{px}v.|pxyp}p .x|pr.~}.wt.
px}ts.pt}x~}.x}.rwx{st}.p.wt.
st
t{~|t} ~u.prp{.rw~x=.
pvt.~u.E.tp.p}s.@B¤@E.tp.6Uxts.
Prr~sx}v{;.wx.p}p{x.x}r{st.~}{.
t.p{=;[email protected]xts;.A??A7=.X}.H<.~.@A<.
w~t.sxt wp.t.qytr.
tp<~{s.rwx{st};.t}pp{.|pxyp}p.
sxpv}~ts xw.p.rw~xr.sx~st=.
t~t.x.}tvpx
t{.p~rxpts.xw.
X}.wt {pvt.s.t
p{px}v.wt.{x}z.
ttrx
t.u}rx~}x}v pz.wp.txt. qttt}.rw~x.p}s.sv.t;.AFC.~u.
x|{t.r~}~{;.
xp{.p}p{x;.p}s.
wt.p~x|pt{.CD;D??.btsxw.
r~}rx.x}.wt.s.~{px~}.
w~wtx 6Uxts.t.p{=;.@HHG7=.
6K?=?@.trt}7.trtx
ts.p.sxpv}~x.~u.
rwx~wt}xp.xwx}.wt.@C<tp.tx~s.
u~{{~x}v.|x{xp x}srx~}.u~|.@HEH.
[email protected]}stp~}.t.p{=;.@HGF7=.^u.wt.
r~}rx.sxpv}~ts xw.rw~x;.
F=F.trt}.6A@.~u.wt.AFC.r~}rx.
xw.rw~x7.wps.ts.|pxyp}p.
|~t wp} D?.x|t.p.x}srx~};.wx{t.
FA.trt}.6@HF.~u.wt.AFC.r~}rx.
xw.rw~x7.wps.}t
t.ts.
|pxyp}p=.P{w~vw.wxvw.|pxyp}p.t.
x}rtpts.wt.t{px
t.xz.u~.
rwx~wt}xp.~ E=?;.wt.pw~.}~t.
wp.qp}xp{.|pxyp}p.t.wx~.
prr~}.u~.~}{.p.|x}~x.~u.p{{.
rpt .~u.rw~x.6P}stp~}.t.p{=;.
@HGF7=.X}tps;.wt.qt.tsxr~.u~.
wtwt.p.r~}rx.~{s.st
t{~.
rw~x.p.p.}~}<rw~xr.
rwxpxr.sxpv}~x.~}.x}srx~}=.
cwt.pw~.r~}r{sts.wp.|pxyp}p.
t.x}rtpts.wt.xz.u~.rw~x.~}{.
p|~}v.x}sx
xsp{.tsx~ts.~.
st
t{~ wt.sx~st=.X}.pssxx~};.p.BD<.
tp.u~{{~..~.wx.s.t~ts.
t x|x{p.t{.6\p}xt<Vprxp.t.
p{=;.A?@A7=.X}.wx.u~{{~..s;.BDC.
r~}rx.st
t{~ts.rwx~wt}xpJ.~u.
wtt.BDC r~}rx;.BA.ts.|pxyp}p.
|~t wp} D?.x|t.p.x}srx~}.6H.
trt};.p}.~ss.px~.~u.E=B7;.wx{t.ADD.
wps.}t
t.ts.|pxyp}p.6FA.trt}7=.
Pssxx~}p{{;.wt.r~}r{x~}.wp.wt.
x|pr.~u |pxyp}p.|p.|p}xut.~}{.
x}.x}sx
xsp{.{xzt{.~.st
t{~.
rw~xr.sx~st wp.qtt}.w~}.x}.
|p}.~wt.t.~u.sxt=.U~.
tp|{t;.p{w~vw.t
xst}rt.w~.wp.
|pxyp}p.t.|p.trtst.wt.
tt}px~}.~u.|~|.x}.x}sx
xsp{.
{pt.sxpv}~ts xw.rw~x.
6brwx||t{|p}}.t.p{=;.A?@@7;.|~.
t~.r~}r{st.wp.~s~|p{.
|~|.~u.rwx~wt}xp.ptp.x~.
~.|pxyp}p.t.6brwxuu|p}.t.p{=;.
A??D7=.bx|x{p{; p.t
xt.~u.wt.vt}t<.
t}
x~}|t}.x}tprx~}.|~st{.u~.
|pxyp}p.p}s.rw~x.r~}r{sts.wp.
~|t.t
xst}rt.~.|pxyp}p.t.
p.p.upr~.wp.|p.x}u{t}rt.wt.
st
t{~|t} ~u.rw~x;.q.~}{.x}.
w~t.x}sx
xsp{ xw.rw~xr.
{xpqx{x.6_t{p~<ctp}.t.p{=;.A?@A7=.
P.x|x{p.r~}r{x~}.p.sp}.
wt}.wt.t
p{t}rt.~u.rwx~wt}xp.
p.|~st{ts.pvpx}.|pxyp}p.t.
pr~.txvw.qxw.r~w~.x}.Pp{xp.x}.
x}sx
xsp{ q~} qttt} wt tp @HC?
[email protected]}wps.t.p{=;.A??B7=.
P{w~vw |pxyp}p.t.x}rtpts.~
t.
x|t.x}.ps{.q~}.sx}v.wt.u~<.
strpst.tx~s;.wtt.p.}~.p.
r~t~}sx}v.x}rtpt x}.sxpv}~t.u~.
rw~x.x}.wtt.x}sx
xsp{=.cwt.
pw~.r~}r{st.wp.|pxyp}p.|p.
trxxpt.rwx~wt}xr.sx~st.~}{.
x}.w~t x}sx
xsp{.w~.pt.
{}tpq{t.
~.st
t{~x}v.rw~x=.cw;.
|pxyp}p.s~t.}~.ptp.~.
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
x}srt.rwx~wt}xp.x}.wt.|py~x.~u.
x}sx
xsp{.w~.wp
t.xts.~.r~}x}t.
~.t.|pxyp}p=.W~t
t;.x}.
x}sx
xsp{.xw.p.vt}txr.
{}tpqx{x.
u~.rw~x; |pxyp}p.t.|p.
x}u{t}rt wt.st
t{~|t}.~u.rw~x=.
bx}v{t.|~zts.~.~p{.s~t.~u.st{p
cWR.~srt.prwrpsxp.p}s.|p.
x}rtpt.q{~~s tt.6Rpx~x.t.p{=;.
@HGGJ.Qt}~x.p}s.Y~}t;.@HFD7=.b~|t.
t
xst}rt.p~rxpt.wt.prwrpsxp.
~srts.q.st{p <cWR.xw.trxpx~}.
~u.wt.|pwtxr.p}s.sttx~}.~u.wt.
pp|pwtxr.}t
~.t|.
6\p{x}~zp.t.p{=;.A?@A7=.Sx}v.
rw~}xr.|pxyp}p.x}vtx~};.p.~{tp}rt.
~.prwrpsxp.st
t{~.6\p{x}~zp.t.
W~t
t;.~{~}vts st{p
x}vtx~}.~srt.qpsrpsxp.p}s.
w~t}x~}.6Qt}~x p}s.Y~}t;.
@HFD7=._{p}<stx
ts rp}}pqx}~xs.p}s.
t}s~rp}}pqx}~xs.t{xrx.w~t}x~}.
p}s.qpsrpsxp.
xp.prx
px~}.~u.
txwtp{{<{~rpts RQ@.trt~.
6fpv}t.t.p{=;.@HHG7=.btrxuxrp{{;.wt.
|trwp}x|.~u.wx.tuutr.x.w~vw.
t}pxr.RQ@.trt~<|tsxpts.
x}wxqxx~}.~u.}~tx}twx}t.t{tpt.
u~|.txwtp{.|pwtxr.}t
t.
t|x}p{;.xw.~xq{t.pssxx~}p{.
sxtr
p~sx{px~}.
xp.prx
px~}.~u.
pr{p.rp}}pqx}~xs.trt~.6_prwt.
t.p{=;.A??E7=.X}.w|p};.~{tp}rt.rp}.
st
t{~ ~.~w~pxr.w~t}x~}.
6Y~}t;.A??AJ.bxs}t;.A??A7.~xq{.
t{pts.~ {p|p.
~{|t.tp}x~};.q.
~{tp}rt.s~t.}~.st
t{~.~.wt.x}t.
w~t}x
t.tuutr.6Qt}~x.p}s.
Y~}t;.@HFD7=.Pssxx~}p{{;.
t{tr~rpsx~vpwxr.rwp}vt.pt.
|x}x|p{;.t
t}.put.{pvt.r|{px
t.
s~t.~u.st{p <cWR.pt.ps|x}xtts=.
6Qt}~x p}s Y~}t;.@HFD7=.
\pxyp}p |~zx}v.q.x}sx
xsp{;.
pxr{p{ w~t xw ~|t stvtt ~u
r~~}p.pt ~.rttq~
pr{p.
sxtpt;.~t.xz.rw.p.x}rtpts.
rpsxpr ~z; rptrw~{p|x}t p}s
rpq~wt|~v{~qx}; |~rpsxp{.
x}uprx~};.p}s.~p{.w~t}x~}.
6Qt}~x p}s Y~}t; @HG@J W~{{xt;
@HGGJ.\x{t|p}.t.p{=;.A??@J.
\p{x}~zp.t.p{=;.A?@A7=.
Put prt t~t ~ |pxyp}p;
p}xt}.q~}rw~sx{px~}.x.wt.|~.
xrp{.txp~.tuutr.6V~}v.t.p{=;.
@HGC7= P trt} A?<tp {~}vxsx}p{
s.xw.~
t.D;???.x}sx
xsp{.
r~{{trts.x}u~|px~}.~}.wt.p|~}.~u.
|pxyp}p t p}s {|~}p u}rx~}
spp.p.tp.?;.A;.D;.@?;.p}s.A?.6_{trwt.
t.p{=;.A?@A7=.P|~}v.wt.|~t.wp}.D;???.
x}sx
xsp{ w~ pxrxpts x} wt
s;.p{|~.G??.~u.wt|.t~ts.
~|t}.w~ts.wp.rw~}xr.|pxyp}p.
t.sxs.}~.xv}xuxrp}{.p{t.
r~}rt}px~}.~u.t~t~}t;.
{tx}xx}v w~|~}t; u~{{xr{t.x|{px}v.
w~|~}t;.~{prx};.~ r~x~{.6Q{~rz.t.
p{=;.@HH@7=.Pssxx~}p{{;.rw~}xr.
|pxyp}p.t.sxs.}~.puutr.t|.
{t
t{ ~u.w~~x};.w~x}t;.p}s.
xx~s~w~}x}t.6Q~}}t;.A?@B7=.
W~t
t;.x}.p.s~q{t<q{x}s;.{prtq~<.
r~}~{{ts; p}s~|xts.r{x}xrp{.xp{.~u.
WXe<~xx
t.|t};.|~zx}v.|pxyp}p.
s~t<stt}st}{.x}rtpts.{p|p.
{t
t{ ~u.vwt{x}.p}s.{tx};.p}s.
strtpts.{p|p.{t
t{.~u.txst.hh.
6axvv.t.p{=;.A?@A7=.
cwt.tuutr ~u.|pxyp}p.~}.ut|p{t.
t~srx
t.t| u}rx~}p{x.sxuut.
qttt}.w|p}.p}s.p}x|p{=.X}.
|~}zt;.st{p <cWR.ps|x}xpx~}.
tts.~
{px~}.6Prw.t.p{=;.@HG@7.
p}s.tsrts.~vtt~}t.{t
t{.
6P{|xt.t.p{=;.@HGB7=.W~t
t;.x}.
~|t}; |~zts |pxyp}p sxs }~ p{t
w~|~}t.{t
t{.~.wt.|t}p{.rr{t.
6\t}st{~}.p}s.\t{{~;.@HGC7=.Q~}.
p}s.S~q.6A??A7.vvt.wp.wt.
st
t{~|t} ~u.~{tp}rt.x}.w|p}.
|p.qt.wt.rpt.~u.wt.sxrtp}rxt.
qttt}.p}x|p{.p}s.w|p}.w~|~}p{.
t~}t.~.rp}}pqx}~xs=.
cwt.tt}rt.~u.st{p
tsrt.qx}sx}v.~u.wt.r~xr~t~xs;.
stp|twp~}t;.x}.wx~rp|p{.xt.
u~|.pst}p{tr~|xts.p;.vvtx}v.
p}.x}tprx~}.xw.wt.v{r~r~xr~xs.
trt~.6T{sxsvt.t.p{=;.@HH@7=.
P{w~vw prt.st{p <cWR.tt}rt.
t{tpt.r~xr~t~}t; ~{tp}rt.
st
t{~.x} p.xw.rw~}xr.
ps|x}xpx~}.6T{sxsvt.t.p{=;.@HH@7=.
b~|t.sxt.~.p.~xq{t.
p~rxpx~}.qttt}.utt};.{~}v<t|.
|pxyp}p.t.p}s.x}rtpts.xz.~u.
txr{p.vt|.rt{{.|~.6cpqt.t.
p{=;.A?@@7=.^}.wt.~wt.wp}s;.trt}.
spp.vvt.wp.rp}}pqx}~xs.pv~}x.
|p.wp
t.wtptxr.
p{t.x}.wt.
tp|t}.~u.~pt.rp}rt;.p.t.~u.
Ttx|t}p{.|pxyp}p.
rprx}~|p.x} wxrw.v~w.x.
ps|x}xpx~}.~.w|p}.s~t.}~.
x|{pts.q.p}s~vt}=.attprw.xw.
r~}xt}{.p{t.|p}.t}s~rx}t.
~pt rp}rt.rt{{.w~.wp.wt.
pp|tt= X}.p}.tp{.s;.|p{t.
|xts.RQ pv~}x;.fX]¤DDA@A¤A;.
qytr w~.ttx|t}p{{.trtx
ts.
x}srt.p~~x.x}.~pt.rp}rt.
|~zts.|pxyp}p.w~ts.p.xv}xuxrp}. rt{{;.p.t{{.p.strtpt.wt.
sttx~} x}.{tx}xx}v.w~|~}t.p}s.
ttx~}.~u.p}s~vt}.trt~.p}s.
p.xv}xuxrp}.x}rtpt.x}.r~x~{.6R~}t.t. ~pt<trxuxr.p}xvt}.6bpupp.t.
p{=;.@HGE7=.W~t
t;.~.{pt.sxt.
w~ts }~ rwp}vt x} w~|~}t= \p{t
qytr ttx|t}p{{.t~ts.~.
|~zts.st{p <cWR.6@G.|v>|pxyp}p.
Rp}}pqx}~xs puutr.wt.x||}t.
rxvptt7 ~ ~p{ st{p
t|.x}.|p}.sxuutt}.p=.
wtt x|t.t.sp.u~.B.sp.p}s.~}.
b}wtxr; }pp{;.p}s.t}s~vt}~.
wt.|~}x}v.~u.wt.u~w.sp7.w~ts.
rp}}pqx}~xs.~ut}.rpt.sxuutt}.
}~ rwp}vt x} {p|p
tuutr.x}.p.s~t<stt}st}.qxwpxr.
pst}~r~xr~~xr.w~|~}t.6PRcW7;.
|p}}t.6R~u~s.p}s.hp|p|p;.A??DJ.
r~x~{;.~{prx};.{tx}xx}v.w~|~}t;. cp}ptr.p}s.R~}p}x}tr;.A?@?7=.
bsxt x} w|p} p}s p}x|p{ vx
t
~ t~t~}t {t
t{ 6Sp t p{=; @HGH7=
r~}u{xrx}v.t{.pq~.rp}}pqx}~xs.
bx|x{p{;.p s.xw.HB.|t}.p}s.DE.
rt} |pxyp}p.t.q.}~.~qprr~.
t.p wt.x|t.~u.pt|t}=._{trwt.
t.p{[email protected]~}s.wp.wt.~rrpx~}p{.
t.~u.|pxyp}p.x.}~.p~rxpts.xw.
strtpts.{|~}p.u}rx~}=.
W~t
t;.~|t.t{x|x}p.t
xst}rt.
vvt.wp.wtp
.|pxyp}p.t.|p.
qt.p~rxpts.xw.}tvpx
t.{|~}p.
tuutr.6_{trwt.t.p{=;.A?@A7=.[~}v<t|.
t.~u.|pxyp}p.rp}.{tps.~.rw~}xr.
r~vw.p}s.x}rtpts.|;.p.t{{.p.
p}.x}rtpts.utt}r.~u.rw~}xr.
q~}rwxx p}s wp}vxx=.X}.pssxx~};.
{|~}p.u}rx~}.t.t
tp{.wp.
{pvt<pxp ~qrx~}.rp}.~rr.xw.
rw~}xr.|pxyp}p.|~zx}v;.p.rp}.
rt{{{p.x}u{p||p~ wx~pw~{~vxrp{.
pq}~|p{xxt.x}.q~}rwxp{.txwt{x|.
6Psp|.p}s.\px}[email protected]~{{xt.
T
xst}rt.tvpsx}v.|pxyp}p.
|~zx}v.{tpsx}v.~.rp}rt.x.
x}r~}xt};.p.~|t.sxt.vvt.p.
~xx
t r~t{px~} wx{t ~wt s~ }~
6[tt.p}s.Wp}r~;.A?@@J.cpwzx};.A??D7=.
bt
tp{.{}v.rp}rt.rpt.wp
t.qtt}.
t~ts.x}.~}v.|pxyp}p.t.xw.
}~.~qprr~.|~zx}v.wx~.~.~wt.
xv}xuxrp}.xz.upr~.6U}v.t.p{=;.
@HHH7=.\pxyp}p.t.|p.s~t<.
stt}st}{.x}tpr.xw.|pvt}xr.
t}xx
x;.rxvptt.|~zx}v;.p}s.
p{r~w~{.t.~.x}rtpt.wt.xz.~u.wtps.
p}s }trz rp}rt 6iwp}v t p{=; @HHH7=
W~t
t;.x}.p.{pvt.s.xw.@;ED?.
qytr;.p.~xx
t.p~rxpx~}.p.}~.
u~}s.qttt}.|pxyp}p.p}s.{}v.
rp}rt 6cpwzx} t.p{=;.A??E7=.cwx.
ux}sx}v t|px}ts.t;.tvps{t.~u.wt.
tt}.~u |pxyp}p.t;.wt}.
r~}~{{x}v.u~.~qprr~.t.p}s.~wt.
~t}xp{.r~}u~}sx}v.
pxpq{t=.
^
tp{{; }t.t
xst}rt.vvt.wp.wt.
tuutr ~u |pxyp}p |~zx}v ~}
txp~ u}rx~}.p}s.rprx}~vt}xrx.
sxuut.u~|.w~t.~u.~qprr~.|~zx}v.
6[tt.p}s.Wp}r~;.A?@@7=.
647;9!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
tuutr.~}.x||}t.u}rx~}x}v.x}.
qytr.xw.r~|~|xts.x||}t.
t|=.Pqp|.t.p{=.6A??B7.
x}
txvpts.|pxyp}p .tuutr.~}.
x||}~{~vxrp{.u}rx~}x}v.x}.EA.PXSb.
pxt}.pzx}v.~tpt.x}wxqx~=.
bqytr.trtx
ts.~}t.~u.wt.u~{{~x}v.
wtt.x|t.p.spI.P.|~zts.|pxyp}p.
rxvptt.r~}px}x}v.B=HD.trt}.st{p <.
cWR;.p}.~p{.pq{t.r~}px}x}v.st{p <.
cWR.6A=D.|v.~•p{.s•~}pqx}~{7;.~•.p}.~•p{.
{prtq~=.cwt.t{.w~ts.}~.rwp}vt.
x}.RSC9.p}s.RSG9.rt{{.r~};.WXe.
a]P.{t
t{;.~.~tpt.x}wxqx~.{t
t{.
qttt}.v~=.cw;.wt.t.~u.
rp}}pqx}~xs.w~ts.}~.w~<t|.
ps
tt.
x~{~vxr.tuutr.x}.x}sx
xsp{.
xw.r~|~|xts.x||}t.t|=.
W~t
t;.wtt.w|p}.spp.r~}p.
xw.spp.vt}tpts.x}.x||}~stuxrxt}.
|xrt;.wxrw.st|~}pts.wp.
t~t.~.st{p <cWR.kp!xkxq!
tt.x||}t.u}rx~};.x}rtpt.
WXe.r~<trt~.ttx~};.p}s.pr.p.
p.r~upr~.~.t}wp}rt.WXe.t{xrpx~}.
6a~w.t.p{=;.A??D7=.
x}r{sx}v.rp}}pqx}~xs=.^
tp{{;.
px~.Ecppcdku!px}.r~}px}.|~t.
wp}.DAD.xst}xuxts.}pp{.r~}xt}=.
P|~}v.w~t.r~}xt};.wt.|~.
x|~p}.~}t.pt.[email protected]~.AA7.rpq~}.
tt}~xs.u~}s.x}.wt.{p};.p.t{{.p.
wtx.rpq~{xr.prxs;.p}p{~vt;.p}s.
p}u~|px~}.~sr;.z}~}.p.
rp}}pqx}~xs.6Pvt{{.t.p{=;.@HGC;.@HGEJ.
\trw~{p|;[email protected]t}sx}~.t.p{=;.
A?@@7=.cw.up;.|~t.wp}.@??.
r~|~}s.r{pxuxts.p.rp}}pqx}~xs.
wp
t.qtt}.rwpprtxts.6T{b~w{.p}s.
b{pst;.A??DJ.apsp};.T{b~w{.t.p{=;.
A??HJ.Pt}sx}~.t.p{=.A?@@7=.
Rp}}pqx}~xs.x|px{.tx.x}.
Ecppcdku-!p}s.q{xwts.spp.vvt.
wp.|~.|py~.rp}}pqx}~xs.
r~|~}s.~rrx}v.}pp{{.wp
t.
qtt}.rwt|xrp{{.xst}xuxts=.]t.p}s.
|x}~.rp}}pqx}~xs.p}s.~wt.}t.
r~|~}s.pt.r~}x}~{.qtx}v.
rwpprtxts.6_~{{p~.t.p{=;.A?@@7=.b~.
up;.~}{.~.rp}}pqx}~xs.
6rp}}pqxvt~{.p}s.x.r~t~}sx}v.
prxs7.wp
t.qtt}.~qpx}ts.u~|.p.pqp.!
Ecppcdku!~rt=.P.b~w.Puxrp}.
4/!Vjg!Uvcvg!qh!Ewttgpv!Uekgpvkhke!
Jgnkejt{uwo!)J!wodtcewnkigtwo*!
Mpqyngfig!Tgictfkpi!vjg!Ftwi!qt!
prr|{pt.wtt.r~|~}s.
Qvjgt!Uwduvcpeg!
6Pt}sx}~.t.p{=.A?@@7=.
P|~}v.wt.rp}}pqx}~xs.u~}s.x}.
d}st.wt.wxs.upr~;.wt.btrtp.
|pxyp}p;.st{p <cWR.6p{t}pt.}p|t.
|.r~}xst.wt.pt.~u.rt}.
st{p <cWR7.p}s.st{p<G<.
rxt}xuxr.z}~{tsvt.tvpsx}v.
|pxyp}p=.cw;.wx.trx~}.sxrt. tpws~rp}}xqx}~{.6st{p <cWR;.
p{t}pt.}p|t.st{p <cWR7.~srt.
wt.rwt|x;.w|p}.
|pxyp}p .rwpprtxxr.rw~prx
t.
wp|pr~zx}txr;.p}s.|tsxrp{.t.~u.
tuutr=.Qtrpt.st{p <cWR.x.|~t.
|pxyp}p=.
pq}sp}.wp}.st{p <cWR;.|pxyp}p .
Ejgokuvt{!
rw~prx
x.x.{pvt{.pxqts.~.
\pxyp}p.x.~}t.~u.wt.r~||~}.
wt.u~|t=.^}{.p.ut.
pxtxt.~u.
}p|t.~u.Ecppcdku!ucvkxc![=.x}.wt.
|pxyp}p.p}p{ts.r~}px}.st{p <cWR.
up|x{.Rp}}pqprtpt=.
p.xv}xuxrp}.p|~}.6Wx
t{.t.p{=;.
Ecppcdku!x.~}t.~u.wt.~{st.
@HEE7=.St{p <cWR.x.p}.~xrp{{.prx
t.
r{x
pts.r~;.~
xsx}v.p.~rt.~u.
tx}~.qp}rt;.x}~{q{t.x}.pt;.
uxqt;.u~~s;.~x{;.p}s.sv=.Q~p}x.x{{. p}s.tt|t{.{xxs.~{q{t=.
stqpt.wtwt.Ecppcdku!w~{s.qt.
Rwt|xrp{{;.st{p <cWR.x.6Epa<p}7<.
r~}xstts.p.p.x}v{t.6cwt._{p}.[x;.
Ep;F;G;@?p<tpws~<E;E;H<x|tw{<B<.
A?@?7.~.wtt.trxt;.k/g/-!R=.ucvkxc-!R=. t}{<EW<sxqt}~<jq;slp}<{<~{;.~.
kpfkec-!p}s.R=.twfgtcnku!6Wx{{xv;.A??D7=.
6¤7<st{pH<6p}7<tpws~rp}}pqx}~{=.
btrxuxrp{{;.|pxyp}p.x.st
t{~ts.p. cwt.6¤7<p}.x~|t.~u.st{p <cWR.x.
px
p.p}s.x}sxrp.r{x
pts.
pxtxt.
wp|pr~{~vxrp{{.E¤@??.x|t.|~t.
~t}.wp}.wt.697<p}.x~|t.6Stt.
6px}7.~.
px~.wqxs=.
cwt.txx~}.stux}t.|pxyp}p.p.
t.p{=;.@HGC7=.
^wt.rp}}pqx}~xs.tt}.x}.
x}r{sx}v.p{{.Ecppcdku!r{x
pts.
|pxyp}p.x}r{st.RQS;.RQR;.p}s.RQ]=.
px}=.Sxuutt}.|pxyp}p.p|{t.
RQS;.p.|py~.rp}}pqx}~xs.~u.|pxyp}p;.
stx
ts.u~|.
px~.r{x
pts.px}.
x.x}~{q{t.x}.pt.p}s.{xxs<~{q{t=.
|p.wp
t.
t.sxuutt}.rwt|xrp{.
r~}xt}.x}r{sx}v.st{p <cWR.p}s. Rwt|xrp{{;.RQS.x.A<j6@a;Ea7<B<|tw{<.
E<~<@<t}<A<{rr{~wt<A<t}<@<{l<D<.
~wt.rp}}pqx}~xs.6Pt}sx}~.t.p{=;.
t}{qt}t}t<@;B<sx~{=.RQS.s~t.}~.
A?@@7=.P.p.r~}tt}rt;.|pxyp}p.
~sr.u~|.sxuutt}.px}.x{{.wp
t. wp
t.rp}}pqx}~{<{xzt.rw~prx
x.
6Psp|.p}s.\px};[email protected]t{{.t.p{=;.
sxuutt}.put;.qx~{~vxrp{;.
@HGC;[email protected]~{{xt;.@HGE7=.RQR.x.
wp|pr~{~vxrp{;.p}s.~xr~{~vxrp{.
p}~wt.|py~.rp}}pqx}~xs.x}.
~ux{t=.cw;.p{{.Ecppcdku!px}.
|pxyp}p=.Rwt|xrp{{;.RQR.x.A<.
rp}}~.qt.r~}xstts.~vtwt.qtrpt.
|tw{<A<6C<|tw{t}<B<t}{7<F<.
~u.wt.
px}v.rwt|xrp{.r~}xt}.
t}{<D<rw~|t}~{=.RQ];.p.|py~.
qttt}.px}=.
\pxyp}p.r~}px}.}|t~.
|tpq~{xt.~u.st{p <cWR;.x.p{~.p.
}pp{{.~rrx}v.r~}xt}.
|x}~.}pp{{<~rrx}v.rp}}pqx}~xs.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11123
Hov!5812
Uhov!5813
xw.tpz.rw~prx
x=.Rwt|xrp{{;.
RQ].x.E;E;H<x|tw{<B<t}{<.
qt}~jrlrw~|t}<@<~{=.
Sxuutt}.|pxyp}p.p|{t.stx
ts.
u~|.
px~.r{x
pts.px}.|p.
sxuut.x}.rwt|xrp{.r~}xt}.
x}r{sx}v.st{p <cWR.p}s.~wt.
rp}}pqx}~xs.6Pt}sx}~.t.p{=.A?@@7=.
P.p.r~}tt}rt;.|pxyp}p.~sr.
u~|.sxuutt}.px}.|p.wp
t.sxuutt}.
put;.qx~{~vxrp{;.wp|pr~{~vxrp{;.p}s.
~xr~{~vxrp{.~ux{t=.X}.pssxx~}.~.
sxuutt}rt.qttt}.r{x
pts.px};.
wt.r~}rt}px~}.~u.st{p <cWR.p}s.
~wt.rp}}pqx}~xs.x}.|pxyp}p.|p.
p.xw.v~x}v.r~}sxx~}.p}s.
~rtx}v.put.wp
t=.X}.pssxx~}.~.
vt}txr.sxuutt}rt.p|~}v.Ecppcdku!
trxt;.wt.{p}.p.r~{{trts¥u~.
tp|{t;.u{~t;.{tp
t;.p}s.t|¥.
rp}.x}u{t}rt.|pxyp}p .~t}r;.
p{x;.p}s.x.6Psp|.p}s.\px};.
@HHEJ.Pvt{{.t.p{=;.@HGCJ.\trw~{p|;.
@HFB7=.P{{.wtt.
pxpx~}.~srt.
|pxyp}p.xw.~t}rxt;.p.x}sxrpts.
q.rp}}pqx}~xs.r~}t};.~}.p
tpvt.
u~|.p.{~.p.@¤A.trt}.~.p.wxvw.
p.@F.trt}=.
^
tp{{;.wtt.
pxpx~}.x}.wt.
r~}rt}px~}.~u.rp}}pqx}~xs.p}s.
~wt.rwt|xrp{.r~}xt}.x}.
|pxyp}p.r~|{xrpt.wt.x}ttpx~}.
~u.r{x}xrp{.spp.x}v.|pxyp}p=.cwt.
{prz.~u.r~}xt}.r~}rt}px~}.~u.
st{p <cWR.p}s.~wt.qp}rt.x}.
|pxyp}p.u~|.sx
tt.~rt.|pzt.
x}ttx}v.wt.tuutr.~u.sxuutt}.
|pxyp}p.r~}xt}.sxuuxr{=.X}.
pssxx~}.~.sxuutt}.rp}}pqx}~xs.
r~}rt}px~}.wp
x}v.sxuutt}.
wp|pr~{~vxrp{.p}s.~xr~{~vxrp{.
~ux{t;.wt.}~}<rp}}pqx}~xs.
r~|~}t}.x}.|pxyp}p;.rw.p.~wt.
tt}~xs.p}s.u{p
~}~xs;.|xvw.p{~.
r~}xqt.~.wt.~
tp{{.
wp|pr~{~vxrp{.p}s.~xr~{~vxrp{.
~ux{t.~u.
px~.|pxyp}p.px}.p}s.
~sr.stx
ts.u~|.w~t.px}=.
cwt.t|.|pxyp}p.x.~ut}.ts.~.
tut.~.p.|xt.~u.wt.sxts.u{~tx}v.
~.p}s.{tp
t.u~|.Ecppcdku/!
\pxyp}p.x}.wx.{x|xx}v.stux}xx~}.x.
~}t.~u.wtt.|py~.stx
px
t.~{s.p.
tppt.x{{xrx.~sr;.wxrw.p{~.
x}r{st.wpwxw.p}s.wpw.~x{=.Prr~sx}v.
~.wt.STP;.Ecppcdku!ucnkxc!x.wt.
x|p.trxt.~u.Ecppcdku!rt}{.
|pztts.x{{tvp{{.x}.wt.d}xts.bpt=.
\pxyp}p.rp}.
p.x}.rp}}pqx}~xs.
r~}t}.p}s.~t}r.6Pvt{{.t.p{=;.
@HGC;.@HGEJ.\trw~{p|.@HFB;.Rprx}x.t.
p{=;.A?@A7=.X}.wt.p{.|xt.~u.{tp
t.
p}s.t|.sxxqts.p.|pxyp}p;.wt.
r~}rt}px~}.~u.st{p <cWR.p
tpvt.
~
t.@A.trt}.q.txvw=.W~t
t;.
trxp{{.v~}.p}s.t{trts.|pxyp}p.
rp}.r~}px}.@D.trt}.~.vtpt.st{p <.
cWR.6Pt}sx}~.t.p{=;.A?@@7=.cw;.p.
@<vp|.|pxyp}p.rxvptt.|xvw.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
r~}px}.st{p <cWR.x}.p.p}vt.u~|.p.
{x{t.p.B.|x{{xvp|.~.p.|rw.p.@D?.
|x{{xvp|.~.|~t=.Pssxx~}p{{;.p.
trt}.t|pxr.t
xt.p}s.|tp<.
p}p{x.u~}s.wp.|pxyp}p .st{p <.
cWR.r~}t}.wp.x}rtpts.xv}xuxrp}{.
u~|.@HFH¤A??H.6Rprx}x.t.p{=;.A?@A7=.X}.
pssxx~}.~.|~zx}v.|pxyp}p;.
x}sx
xsp{.x}vt.|pxyp}p.w~vw.
u~~s.|pst.xw.qt.~.~x{.x}uts.
xw.|pxyp}p.p}s.x.tpr=.cwtt.
|pxyp}p.qt.pt.vt}tp{{.|pst.q.
pssx}v.|pxyp}p.~.qt.p}s.wtpx}v.
x=.cwt.t{p}.qt.x.wt}.ts.~.
r~~z.p.
pxt.~u.u~~s=.cwtt.pt.}~.
q{xwts.sxt.|tpx}v.wt.
r~}rt}px~}.~u.rp}}pqx}~xs.x}.wtt.
|pxyp}p.u~~s.~sr=.
Wpwxw.r~}x.~u.wt.sxts.p}s.
r~|tts.rp}}pqx}~xs<xrw.tx}~.
|ptxp{.~u.Ecppcdku!p}s.r~|t.x}.p.
pxt.~u.u~|.6g/i/!qp{{.p}s.rpzt7=.
X}sx
xsp{.|p.qtpz.~uu.xtrt;.{prt.
x.x}~.p.xt.p}s.|~zt.x=.STP.t~.
wp.rp}}pqx}~xs.r~}t}.x}.wpwxw.
p
tpvt.x.trt}.6STP;.A??D7=.fxw.
wt.st
t{~|t}.p}s.r{x
px~}.~u.
|~t.wxvw.~t}r.Ecppcdku!px};.wt.
p
tpvt.rp}}pqx}~xs.r~}t}.x}.wpwxw.
x{{.{xzt{.x}rtpt=.
Wpw.~x{.x.~srts.q.~{
t}.
tprx~}.~u.wt.rp}}pqx}~xs.u~|.
{p}.|ptxp{=.cwt.tpr .r~{~.p}s.
~s~.
p;.stt}sx}v.~}.wt.~{
t}.
t.ts=.Wpw.~x{.x.p.
xr~.q~}<.
~.p|qt<r~{~ts.{xxs.r~}px}x}v.
p~x|pt{.D?.trt}.rp}}pqx}~xs=.
^}t.~.~.s~.~u.wt.{xxs.{prts.
~}.p.rxvptt.~ts{.~srt.wt.
tx
p{t}.~u.p.x}v{t.|pxyp}p.
rxvptt.6STP;.A??D7=.
X}.r~}r{x~};.|pxyp}p.wp.
w}sts.~u.r{x
p.r~}px}x}v.
pxpq{t.r~}rt}px~}.~u.st{p <cWR;.
rp}}pqx}~xs;.p}s.~wt.r~|~}s=.
cw;.|pxyp}p.x.}~.p.x}v{t.rwt|xrp{.
xw.p.r~}xt}.p}s.t~srxq{t.
rwt|xrp{.~ux{t.~.tsxrpq{t.p}s.
r~}xt}.r{x}xrp{.tuutr=.P.vxsp}rt.
u~.x}s;.t}x{ts.Dqvcpkecn!Ftwi!
Rtqfwevu- ~
xst.x}u~|px~}.~}.wt.
p~
p{.~u.q~p}xrp{.sv.~sr=.c~.
x}
txvpt.|pxyp}p.u~.|tsxrp{.t.x}.
p.|p}}t.prrtpq{t.p.~.u~.
|pztx}v.p~
p{.}st.p}.]SP;.
r{x}xrp{.sxt.}st.p}.X]S.~u.
r~}xt}.qprwt.~u.p.pxr{p.
|pxyp}p.~sr.u~.pxr{p.sxtpt.
x}sxrpx~}.w~{s.qt.r~}srts=.X}.
pssxx~};.x}u~|px~}.p}s.spp.
tvpsx}v.wt.|pxyp}p.~sr .
rwt|x;.|p}uprx}v.p}s.r~}~{;.
wp|pr~{~v;.p}s.p}x|p{.~xr~{~v.
spp;.p|~}v.~wt.|.qt.~
xsts.
cwx.vxsp}rt.x.p
px{pq{t.~}.wt.X}t}t.p.
jvvr<00yyy/hfc/iqx0Ftwiu0fghcwnv/jvo!}st.
Vxsp}rt.6Sv7=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
647;;!
p}s.|tt.wt.txt|t}.u~.}t.sv. xv}xuxrp}.{~.~u.st{p <cWR.x}.xst<.
p~
p{[email protected]@C=D?7=.
tp|.|~zt;.
pxpx~}.x}.x}sx
xsp{.
|~zx}v.qtwp
x~;.rp}}pqx}~xs.
Jwocp!Rjctoceqmkpgvkeu!
~{x;.x}r~|{tt.pq~x~}.~u.
\pxyp}p.rp}.qt.pzt}.x}.p.
pxt.~u. x}wp{ts.|~zt;.p}s.|tpq~{x|.x}.wt.
u~|{px~}.q.|{x{t.~t.~u.
{}v=.P}.x}sx
xsp{ .ttxt}rt.p}s.
ps|x}xpx~}=.X}sx
xsp{.|~zt.
trw}xt.xw.|~zx}v.|pxyp}p.p{~.
|pxyp}p.p.p.rxvptt;.txvwx}v.
stt|x}t.wt.s~t.pq~qts.6Wt|x}v.
qttt}.?=D.p}s.@=?.vp|;.~.x}.p.xt=.
t.p{=;[email protected]~wp}~}.t.p{=;.@HGH7=.
Pssxx~}p{{;.x}sx
xsp{.pzt.
Put.|~zx}v;.st{p <cWR.
t}~.
|pxyp}p.~p{{.x}.u~~s.~.p.p}.
{t
t{.str{x}t.trxx~{.xwx}.
tpr.x}.twp}~{.~.~wt.~{
t}=.
|x}t;.p}s.r~}x}t.~.v~.s~}.~.
\~t.trt}{;.prrt.~.
p~xt.
pq~.D.~.@?.trt}.~u.wt.tpz.{t
t{.
~
xst.p}~wt.|tp}.u~.pqt.~.
xwx}.p}.w~.6Pvt{{.t.p{=;.@HGE;.
x}wp{t.|pxyp}p;.
Wtx.t.p{=;@HHAp;.@HHAq7=.
cwt.pq~x~};.|tpq~{x|;.p}s.
wp|pr~zx}txr.~ux{t.~u.st{p <cWR;. Rjctoceqmkpgvkeu!hqt!Qtcn!
Cfokpkuvtcvkqp!qh!Ecppcdkpqkfu!
rp}}pqx}~xs;.p}s.sv.~sr.
r~}px}x}v.st{p <cWR.
p.xw.~t.
Put.~p{.ps|x}xpx~}.~u.st{p <.
~u.ps|x}xpu~}.p}s.u~|{px~}.
cWR.~.|pxyp}p;.wt.~}t.~u.tuutr.
6Psp|.p}s.\px};[email protected]t{{.t.p{=;. p.xwx}.B?.~.H?.|x}t;.tprwt.
@HGC;.@HGE7=.
x.tpz.put.A.~.B.w~.p}s.wt}.
t|px}.u~.C.[email protected]~.
Rjctoceqmkpgvkeu!qh!Uoqmgf!
6V~t}wt|t};.A??BJ.Psp|.p}s.
Cfokpkuvtcvkqp!qh!Ecppcdkpqkfu!
\px};[email protected]t{{.t.p{=;.@HGC;.@HGE7=.
Rwpprtxpx~}.~u.wt.
St.~.wt.st{p.x}.~}t.~u.tuutr;.
wp|pr~zx}txr.~u.st{p <cWR.p}s.
t.wp
t.sxuuxr{.x}.xpx}v.~p{.
~wt.rp}}pqx}~xs.u~|.|~zts.
st{p <cWR.s~t.r~|pts.~.|~zx}v.
|pxyp}p.x.sxuuxr{.qtrpt.p.qytr . |pxyp}p=.^p{.qx~p
px{pqx{x.~u.st{p <.
|~zx}v.qtwp
x~.sx}v.p}.ttx|t}. cWR;.wtwt.t.~.x}.|pxyp}p;.x.
pxt.6Pvt{{.t.p{=;[email protected]|x}v.t.
{~.p}s.tt|t{.
pxpq{t;.p}vx}v.
p{=;[email protected]tx.t.p{=;.@HHAp7=.Tprw.
qttt}.D.p}s.A?.trt}.6Pvt{{.t.p{=;.
uu.st{x
t.p.sxrtt.s~t.~u.st{p <.
@HGC;.@HGE7=.U~{{~x}v.~p{.
cWR=.P}.ttxt}rts.|pxyp}p.|~zt. ps|x}xpx~}.~u.psx~prx
t<{pqt{ts.
rp}.xpt.p}s.tv{pt.wt.s~t.~.
st{p <cWR;.st{p <cWR.{p|p.{t
t{.
~qpx}.wt.stxts.prt.rw~{~vxrp{.
pt.{~.t{px
t.~.{p|p.{t
t{.put.
tuutr.p}s.|x}x|xt.}stxts.tuutr=. |~zx}v.~.x}p
t}~.ps|x}xpx~}=.
U~.tp|{t;.}st.}pp{xxr.
X}t<.p}s.x}p<qytr.
pxpqx{x.
r~}sxx~};.t.w~{s.|pxyp}p.|~zt. ~rr.t
t}.xw.ttpts.s~x}v.}st.
x}.wtx.{}v.u~.p}.tt}sts.tx~s.~u. r~}~{{ts.r~}sxx~}=.cwt.{~.p}s.
x|t.wxrw.rpt.~{~}vts.pq~x~}.
pxpq{t.~p{.qx~p
px{pqx{x.~u.st{p <.
p}s.x}rtpt.rw~prx
t.tuutr=.cwt. cWR.x.p.r~}tt}rt.~u.x.ux<p.
tuutr.~u.ttxt}rt.x}.wt.rw~{~vxrp{. wtpxr.t{x|x}px~}.u~|.q{~~s.p}s.
t~}t.|p.t{px}.w.st{p <cWR.
tpxr.pq~x~}.u~|.~|prw.p}s.
t}~.q{~~s.{t
t{.r~t{pt.~~{.
q~t{=.
xw.x}t}x.~u.tuutr.p}s.x}~xrpx~}.
Ecppcdkpqkf!Ogvcdqnkuo!cpf!Gzetgvkqp!
{t
t{.6Pvt{{.t.p{[email protected]}t.t.p{=.
Rp}}pqx}~xs.|tpq~{x|.x.r~|{t=.
@HGDJ.Wtx.t.p{=;.@HHAp7=._uu.p}s.
x}wp{px~}.
~{|t.w~{s.qt.tr~sts. St{p <cWR.x.|tpq~{xts.
xp.
x}.sxt.p.wt.r~}rt}px~}.6s~t7.~u. |xr~~|p{.ws~{px~}.~.q~w.prx
t.
p}s.x}prx
t.|tpq~{xt.6[t|qtvt.t.
rp}}pqx}~xs.ps|x}xtts.rp}.
p.p.
p{=;.@HF?;.@HFAp;[email protected]t{{.t.p{=;.
sxuutt}.pvt.~u.|~zx}v=.
b|~zts.|pxyp}p.t{.x}.
@HGEJ.W~{{xt;.@HGG7=.cwt.x|p.
pq~x~}.~u.st{p <cWR.x}.wt.u~|.~u. prx
t.|tpq~{xt.~u.st{p <cWR.
p}.pt~~{.xwx}.tr~}s=._rw~prx
t. u~{{~x}v.~p{.x}vtx~}.x.@@<ws~<.
tuutr.~rr.x||tsxpt{.u~{{~x}v.
st{p <cWR=.cwx.|tpq~{xt.x.
pq~x~};.xw.|t}p{.p}s.qtwp
x~p{. p~x|pt{.tx~t}.~.st{p <cWR.
tuutr.|tppq{t.u~..~.E.w~.
x}.~srx}v.|pxyp}p<{xzt.qytrx
t.
6V~t}wt|t};.A??BJ.W~{{xt.@HGE;.
tuutr.6Pvt{{.t.p{=;.@HGE;.[t|qtvt.
@HGG7=.St{p <cWR.x.st{x
tts.~.wt.
p}s.aqx};.@HFD7=.Put.~p{.
qpx}.pxs{.p}s.tuuxrxt}{.p.ttrts. ps|x}xpx~};.|tpq~{xt.{t
t{.|p.
~u.p.
t.{xxs.~{q{t.sv=.
trtts.wp.~u.st{p <cWR.p}s.w.
cwt.qx~p
px{pqx{x.~u.wt.st{p <cWR;. r~}xqt.vtp{.~.wt.
u~|.|pxyp}p.x}.p.rxvptt.~.xt;.
wp|pr~{~vxrp{.tuutr.~u.~p{.st{p <.
rp}.p}vt.u~|.@.~.AC.trt}.xw.wt. cWR.~.|pxyp}p=.
uprx~}.pq~qts.pt{.trttsx}v.@?.~.
_{p|p.r{tpp}rt.~u.st{p <cWR.
A?.trt}.6Pvt{{.t.p{=;@HGEJ.
p~x|pt.wtpxr.q{~~s.u{~.p.
W~{{xt;.@HGG7=.cwt.t{px
t{.{~.p}s. pq~.HD?.|{>|x}.~.vtpt=.cwt.pxs.
pxpq{t.qx~p
px{pqx{x.t{.u~|.
sxptpp}rt.~u.st{p <cWR.u•~|.q{~~s.
RQ!11111
Hto!11124
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
64811!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
x.{pvt{.st.~.tsxxqx~}.~.~wt.
xt.x}.wt.q~s;.pwt.wp}.~.
|tpq~{x|.6Pvt{{.t.p{=;.@HGC;.@HGE7=.
\tpq~{x|.x}.|~.xt.x.t{px
t{.
{~.~.pqt}=.b{~.t{tpt.~u.st{p <.
cWR.p}s.~wt.rp}}pqx}~xs.u~|.
xt.p}s.qtt}.|tpq~{x|.
t{.x}.p.{~}v.t{x|x}px~}.wp{u<{xut=.
cwt.t|x}p{.wp{u<{xut.~u.st{p <cWR.
p}vt.u~|.p~x|pt{.A?.w~.~.
p.{~}v.p.@?.[email protected];.w~vw.
t~ts.tx|pt.
p.p.ttrts.
xw.p}.{~{.r{tpts.qp}rt.p}s.
wt.t.~u.pp.xw.
pxpq{t.
t}xx
xxt.6W}.p}s.Y~}t;.@HG?7=.
[t|qtvt.t.p{=.6@HF?7.stt|x}ts.wt.
wp{u<{xut.~u.st{p <cWR.~.p}vt.u~|.AB.
~.AG.w~.x}.wtp
.|pxyp}p.t.~.
E?.~.F?.w~.x}.}px
t.t=.X}.
pssxx~}.~.@@<ws~<st{p <cWR;.
~|t.x}prx
t.rpq~.|tpq~{xt.wp
t.
t|x}p{.wp{u<{x
t.~u.D?.w~.~.E.sp.
~.|~t=.cwt.{pt.qp}rt.t
t.p.
{~}v<t|.|pzt.x}.x}t.t.u~.
tp{xt.|pxyp}p.t=.
cwt.|py~x.~u.wt.pq~qts.st{p <.
cWR.s~t.x.t{x|x}pts.x}.utrt;.p}s.
pq~.BB.trt}.x}.x}t=.St{p <cWR.
t}t.t}t~wtpxr.rxr{px~}.p}s.
}stv~t.ws~{px~}.p}s.~xspx~}.
~.@@<}~<H<rpq~<st{p <cWR=.cwt.
v{r~}xst.x.trtts.p.wt.|py~.
x}t.|tpq~{xt.p{~}v.xw.pq~.@G.
}~}<r~}yvpts.|tpq~{xt=.Utt}.
p}s.x}utt}.|pxyp}p.t.
|tpq~{xt.st{p <cWR.x|x{p{.
6Pvt{{.t.p{=;.@HGE7=.
~t}xp{.t.u~.tpx}v.|{x{t.
r{t~x;.}t~pwxr.px};.ptxt.
tx~}.p}s.rprwtxp=.W~t
t;.
pxst.u~|.wt.spp.~srts.q.R\Ra;.
}~.pt<{t
t{.|tsxrp{.|pxyp}p.{p.
wp
t.~srts.rxt}xuxr.spp.~}.
|pxyp}p .put.p}s.tuutrx
t}t=.
USP.p~
t.|tsxrp{.t.~u.p.sv.
u~{{~x}v.p.q|xx~}.p}s.t
xt.~u.p}.
]SP.~.Q[P=.cwt.USP.wp.}~.
p~
ts.p}.sv.~sr.r~}px}x}v.
|pxyp}p.u~.|pztx}v=.T
t}.~;.
t{.~u.|p{{.r{x}xrp{.t{~p~.
sxt.wp
t.qtt}.q{xwts.x}.wt.
rt}.|tsxrp{.{xtpt=.\p}.sxt.
strxqt.w|p}.ttprw.xw.
|pxyp}p.x}.wt.d}xts.bpt.}st.
USP<tv{pts.X]S.p{xrpx~}=.
W~t
t;.USP.p~
p{.~u.p}.]SP.x.
}~.wt.~}{.|tp}.w~vw.wxrw.p.
sv.rp}.wp
t.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt=.X}.vt}tp{;.p.sv.|p.wp
t.p.
rt}{.prrtts.|tsxrp{.t .x}.
tp|t}.x}.wt.d}xts.bpt.xu.wt.
sv.|tt.p.ux
t<p.t=.Tpq{xwts.
rpt.{p.6P{{xp}rt.u~.Rp}}pqx.
cwtptxr.
=.STP;[email protected]=Bs.@@B@;.
@@BD.6S=R=.Rx=.@HHC77.wt{s.wt.
Ps|x}xp~.~u.STP .p{xrpx~}.~u.
wt.ux
t<p.t.~.stt|x}t.wtwt.
p.sv.wp.p.rt}{.prrtts.
|tsxrp{.t= .cwt.u~{{~x}v.strxqt.
wt.ux
t.t{t|t}.wp.rwpprtxt.
rt}{.prrtts.|tsxrp{.t .u~.p.
svI .
x=.wt.sv .rwt|x.|.qt.z}~}.
p}s.t~srxq{t=.
Uvcvwu!qh!Tgugctej!Kpvq!vjg!Ogfkecn!
cwt.qp}rt .rwt|x.|.qt.
Wugu!hqt!Octklwcpc!
rxt}xuxrp{{.tpq{xwts.~.t|x.x.~.
bpt<{t
t{.q{xr.x}xxpx
t;.
qt.t~srts.x}~.s~pvt.wxrw.rp}.
x}r{sx}v.{p.p}s.tutt}sp.x}.~. qt.p}spsxts=.cwt.{xx}v.~u.wt.
~u.wt.|tsxrp{.t.~u.|pxyp}p;.wp
t.
qp}rt.x}.p.rt}.tsxx~}.~u.~}t.~u.
vt}tpts.x}tt.x}.wt.|tsxrp{.
wt.~uuxrxp{.r~|t}sxp;.p.stux}ts.q.
r~||}x.p}s.wt.}tts.u~.wxvw.
trx~}[email protected].~u.wt.U~~s;.Sv.p}s.
p{x.r{x}xrp{.x}
txvpx~}.p.t{{.p.
R~|txr.Pr;[email protected]=b=R=.BA@V7;.x.
r~|twt}x
t.put.p}s.tuutrx
t}t. uuxrxt}.~.|tt.wx.txt|t}= .
spp=.X}.~st.~.psst.wt.}tts.u~.
xx=.wtt.|.qt.pstpt.put.
wxvw.p{x.r{x}xrp{.x}
txvpx~};.wt. sxt=.
cwtt.|.qt.pstpt.
pt.~u.Rp{xu~}xp.tpq{xwts.wt.Rt}t.
wp|pr~{~vxrp{.p}s.~xr~{~vxrp{.
u~.\tsxrx}p{.Rp}}pqx.attprw.
sxt;.s~}t.q.p{{.|tw~s.tp~}pq{.
6R\Ra;.yyy/eoet/weuf/gfw7.x}.A???.
p{xrpq{t;.~}.wt.qpx.~u.wxrw.x.
x}.t~}t.~.rxt}xuxr.t
xst}rt.u~.
r~{s.upx{.p}s.t~}xq{.qt.
wtptxr.~xqx{xxt.~u.rp}}pqx .
r~}r{sts;.q.tt.p{xuxts.q.
p}s.{~rp{.{tvx{px
t.x}xxpx
t.x}.up
~.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
~u.r~|px~}pt.t .6Vp};.A??D7=.
bpt.{tvx{px~}.tpq{xwx}v.wt.R\Ra. t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
rp{{ts.u~.wxvw.p{x.|tsxrp{.ttprw. sv;.wp.wt.qp}rt.x.put.u~.
tpx}v.p.trxuxr;.tr~v}xts.sx~st= .
wp.~{s.t}wp}rt.}stp}sx}v.~u.
xxx=.wtt.|.qt.pstpt.p}s.t{{<.
wt.tuuxrpr.p}s.ps
tt.tuutr.~u.
|pxyp}p.p.p.wp|pr~{~vxrp{.pvt}; . r~}~{{ts.sxt.~
x}v.tuuxrpr=.
cwtt.|.qt.pstpt;.t{{<.
q.tts.wt.~ytr.w~{s.}~.qt.
r~}ts.p.t}r~pvx}v.~.p}rx~}x}v. r~}~{{ts;.t{{<stxv}ts;.t{{<.
r~}srts;.p}s.t{{<s~r|t}ts.
wt.~rxp{.~.trtpx~}p{.t.~u.
|pxyp}p= .cwt.R\Ra.u}sts.|p}.~u. sxt;.x}r{sx}v.r{x}xrp{.
x}
txvpx~};.q.tt.p{xuxts.q.
wt.q{xwts.sxt.~}.|pxyp}p .
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
X}.wx.~px~}.wt.t|.rp}}pqx.x.
x}trwp}vtpq{t.xw.|pxyp}p=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
DF.Ua.@?CHH;.@?D?C¤?E.6\p•rw.AE;.@HHA7=.
RQ!11111
Hto!11125
Hov!5812
Uhov!5813
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
sv;.~}.wt.qpx.~u.wxrw.x.r~{s.qt.
upx{.p}s.t~}xq{.r~}r{sts.q.
rw.tt.wp.wt.qp}rt.x{{.
wp
t.wt.x}t}sts.tuutr.x}.tpx}v.p.
trxuxr;.tr~v}xts.sx~st= .
x
=.wt.sv.|.qt.prrtts.q.
p{xuxts.tt=.
cwt.sv.wp.p.]t.Sv.
P{xrpx~}.6]SP7.p~
ts.q.wt.
U~~s.p}s.Sv.Ps|x}xpx~};.
p}.~.wt.U~~s;.Sv.p}s.
R~|txr.Pr;[email protected]=b=R=.BDD=.^;.p.
r~}t}.~u.wt.}px~}p{.r~||}x.~u.
tt;.p{xuxts.q.rxt}xuxr.px}x}v.
p}s.ttxt}rt.~.t
p{pt.wt.put.
p}s.tuutrx
t}t.~u.sv;.prrt.wt.
put.p}s.tuutrx
t}t.~u.wt.qp}rt.
u~.t.x}.tpx}v.p.trxuxr;.tr~v}xts.
sx~st=.P.|ptxp{.r~}u{xr.~u.~x}x~}.
p|~}v.tt.tr{st.p.ux}sx}v.~u.
r~}t}= .p}s.
=.wt.rxt}xuxr.t
xst}rt.|.qt.
xst{.p
px{pq{t=.
X}.wt.pqt}rt.~u.]SP.p~
p{;.
x}u~|px~}.r~}rt}x}v.wt.rwt|x;.
wp|pr~{~v;.~xr~{~v;.p}s.
tuutrx
t}t.~u.wt.qp}rt.|.qt.
t~ts;.q{xwts;.~.~wtxt.
xst{.p
px{pq{t;.x}.uuxrxt}.stpx{.~.
t|x.tt;.p{xuxts.q.rxt}xuxr.
px}x}v.p}s.ttxt}rt.~.t
p{pt.wt.
put.p}s.tuutrx
t}t.~u.sv;.~.
upx{.p}s.t~}xq{.r~}r{st.wt.
qp}rt.x.put.p}s.tuutrx
t.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st= .
\pxyp}p.s~t.}~.|tt.p}.~u.wt.
ux
t.t{t|t}.}trtp.u~.p.sv.~.
wp
t.p.rt}{.prrtts.|tsxrp{.
t= .
Ux{;.wt.rwt|x.~u.|pxyp}p;.p.
stux}ts.x}.wt.txx~};.x.}~.
t~srxq{t.x}.t|.~u.rtpx}v.p.
p}spsxts.s~t=.cwt.txx~}.stux}t.
|pxyp}p.p.x}r{sx}v.p{{.Ecppcdku!
r{x
pts.px}=.Sxuutt}.|pxyp}p.
p|{t.stx
ts.u~|.
px~.r{x
pts.
px}.|p.wp
t.
t.sxuutt}.rwt|xrp{.
r~}xt}.x}r{sx}v.st{p <cWR.p}s.
~wt.rp}}pqx}~xs.6Pt}sx}~.t.p{=;.
A?@@7=.P.p.r~}tt}rt;.|pxyp}p.
~sr.u~|.sxuutt}.px}.x{{.wp
t.
sxuutt}.put;.qx~{~vxrp{;.
wp|pr~{~vxrp{;.p}s.~xr~{~vxrp{.
~ux{t=.cw;.wt}.r~}xstx}v.p{{.
Ecppcdku!px}.~vtwt;.qtrpt.~u.
wt.
px}v.rwt|xrp{.r~}xt};.
t~srx}v.r~}xt}.p}spsxts.
s~t.x.}~.~xq{t=.Pssxx~}p{{;.
|~zx}v.|pxyp}p.rt}{.wp.}~.
qtt}.w~}.~.p{{~.st{x
t.~u.
r~}xt}.p}s.t~srxq{t.s~t=.
W~t
t;.xu.p.trxuxr.Ecppcdku!px}.x.
v~}.p}s.~rtts.}st.xr{.
r~}~{{ts.r~}sxx~};.wt.{p}.
rwt|x.|p.qt.zt.r~}xt}.
t}~vw.~.~srt.t~srxq{t.p}s.
p}spsxts.s~t=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
P.~.wt.tr~}s.p}s.wxs.rxtxpJ.
wtt.pt.}txwt.pstpt.put.sxt.
}~.pstpt.p}s.t{{<r~}~{{ts.
sxt.~
x}v.|pxyp}p .tuuxrpr=.c~.
~.wt.txx~}t .ptx~}.wp.
|pxyp}p.wp.prrtts.|tsxrp{.t;.wt.
txx~}t.rxt.wt.P|txrp}.\tsxrp{.
P~rxpx~} .6P\P7.A??H.t~.
t}x{ts.dt.~u.Rp}}pqx.u~.\tsxrx}p{.
_~t= .cwt.txx~}t.r{px|.wt.
P\P.t~.x.t
xst}rt.wt.P\P.
prrt.|pxyp}p .put.p}s.tuuxrpr=.
W~t
t;.wt.A??H.P\P.t~.r{pxuxt.
wp.wt.t~.w~{s.}~.qt.
xtts.p.
p}.t}s~t|t}.~u.pt<qpts.|tsxrp{.
rp}}pqx.~vp|;.wt.{tvp{xpx~}.~u.
|pxyp}p;.~.wp.rxt}xuxr.t
xst}rt.~}.
wt.wtptxr.t.~u.rp}}pqx.|tt.
wt.p|t.p}s.rt}.p}sps.u~.p.
trxx~}.sv.~sr= .
Rt}{;.}~.q{xwts.sxt.
r~}srts.xw.|pxyp}p.|tt.wt.
rxtxp.~u.p}.pstpt.p}s.t{{<.
r~}~{{ts.tuuxrpr.s=.cwt.rxtxp.
u~.p}.pstpt.p}s.t{{<r~}~{{ts.
s.u~.~t.~u.stt|x}x}v.wt.
put.p}s.tuuxrpr.~u.p.w|p}.sv.pt.
stux}ts.}st.wt.R~st.~u.Utstp{.
atv{px~}.6RUa7.x}[email protected]@C=@AE=.X}.
~st.~.pt.wx.t{t|t};.USP.
r~}srts.p.t
xt.~u.r{x}xrp{.sxt.
q{xwts.p}s.p
px{pq{t.x}.wt.q{xr.
s~|px}.qtu~t.Utqp;.A?@B=.bsxt.
tt.xst}xuxts.w~vw.p.tprw.~u.
_q\ts u~.pxr{t.q{xwts.u~|.
x}rtx~}.~.Utqp.A?@B;.u~.
p}s~|xts.r~}~{{ts.xp{.x}v.
|pxyp}p.~.pt.|pxyp}p .tuuxrpr.
x}.p}.wtptxr.x}sxrpx~}=.
Pssxx~}p{{;.wt.t
xt.x}r{sts.
sxt.xst}xuxts.w~vw.p.tprw.~u.
qxq{x~vpwxr.tutt}rt.x}.t{t
p}.
t|pxr.t
xt.p}s.xst}xuxts.
sxt.tt}x}v.~xvx}p{.ttprw.x}.
p}.{p}vpvt=.bt{trts.sxt.}ttsts.
~.qt.{prtq~<r~}~{{ts.p}s.s~q{t<.
q{x}sts=.Pssxx~}p{{;.sxt.}ttsts.~.
t}r~|p.ps|x}xtts.|pxyp}p.
{p}.|ptxp{=.cwtt.p.}~.
txt|t}.u~.p}.trxuxr.~t.~u.
ps|x}xpx~};.}~.p}.pvt.{x|x.~}.
s.qytr=.bsxt.tt.tr{sts.
wp.ts.{prtq~.|pxyp}p.
{t|t}ts.q.wt.pssxx~}.~u.
trxuxr.p|~}.~u.cWR.~.~wt.
rp}}pqx}~xs=.Pssxx~}p{{;.sxt.
ps|x}xtx}v.|pxyp}p.{p}.tpr.
tt.tr{sts=.
cwt._q\ts.tprw.xt{sts.p.~p{.~u.
DEE.pqpr.~u.rxt}xuxr.pxr{t=.^u.
X}.wx.~px~}.wt.t|.rp}}pqx.x.ts.
x}trwp}vtpq{.u~.|pxyp}p=.
cwt.u~{{~x}v.tprw.ptv.p.ts;.
6rp}}pqx.^a.|pxyp}p7.P]S.6wtptxr.t.^a.
wtp7.P]S.6aRc.^a.p}s~|xts.r~}~{{ts.xp{.
^a.t|pxr.t
xt .^a.r{x}xrp{.xp{.^a.
r{x}xrp{.xp{7.]^c.6|pxyp}p.pqt j\twl.^a.
pssxrx
t.qtwp
x~.^a.qp}rt.t{pts.
sx~st7= .
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
wtt.pqpr;.p.u{{<t.t
xt.p.
r~}srts.xw.GD.pt.~.pt.
t{xvxqx{x=.^u.wt.sxt.xst}xuxts.
w~vw.wt.tprw.~u.wt.tutt}rt.p}s.
wt.DEE.pqpr.u~|.wt._q\ts.
tprw;.~}{.@@.sxt.|t.p{{.wt.
rxtxp.u~.t{trx~}.6Pqp|.t.p{=;.
A??FJ.R~t<Q{~~|.t.p{=;.A?@AJ.
Rpu~s.p}s.\tx;[email protected]{{x.t.p{=;.
A??HJ.Wp}t.t.p{=;.A??DJ.Wp}t.t.p{=;.
A??FJ.\tx.t.p{=;.@HG?J.cpwzx}.t.p{=;.
@HFCJ.fpt.t.p{=;.A?@?J.fx{t.t.p{=;.
A??GJ.fx{t.t.p{=;.A?@B7=.cwtt.@@.
sxt.tt.q{xwts.qttt}.@HFC.
p}s.A?@B=.ct}.~u.wtt.sxt.tt.
r~}srts.x}.wt.d}xts.bpt.p}s.~}t.
s.p.r~}srts.x}.Rp}psp=.cwt.
xst}xuxts.sxt.tp|x}t.wt.tuutr.~u.
|~zts.p}s.
p~xts.|pxyp}p.u~.wt.
x}sxrpx~}.~u.rw~}xr.}t~pwxr.px};.
pxrx.t{pts.~.\{x{t.br{t~x.
6\b7;.ptxt.x|{px~}.x}.w|p}.
x||}~stuxrxt}r.
x.6WXe7.pxt};.
v{pr~|p;.p}s.pw|p=.P{{.sxt.ts.
ps{.qytr=.
cwt.@@.xst}xuxts.sxt.tt.
x}sx
xsp{{.t
p{pts.~.stt|x}t.xu.
wt.rrtu{{.|tt.prrtts.
rxt}xuxr.p}sps=.btrxuxrp{{;.wt.
tt.t
p{pts.~}.s.stxv}.
x}r{sx}v.qytr.t{trx~}.rxtxp;.
p|{t.xt;.q{x}sx}v.trw}xt;.s~x}v.
ppsxv|;.~r~|t.|tpt;.p}s.wt.
pxxrp{.p}p{x.~u.wt.t{=.cwt.
p}p{x.t{xts.~}.q{xwts.sxt;.
w.x}u~|px~}.p
px{pq{t.pq~.
~~r~{;.~rtst;.p}s.t{.tt.
{x|xts.~.s~r|t}.q{xwts.p}s.
xst{.p
px{pq{t.x}.wt.q{xr.s~|px}=.
cwt.t
xt.u~}s.wp.p{{.@@.sxt.wp.
tp|x}ts.tuutr.~u.x}wp{ts.|pxyp}p.
s~.}~.rt}{.~
t.tuuxrpr.~u.
|pxyp}p.x}.p}.wtptxr.x}sxrpx~}.
qpts.~}.p.}|qt.~u.{x|xpx~}.x}.
wtx.s.stxv}J.w~t
t;.wt.|p.
qt.r~}xstts.~~u.~u.r~}rt.sxt=.
_~~u.~u.r~}rt.sxt.~
xst.
t{x|x}p.t
xst}rt.~}.p.~~ts.
w~wtx.x}
~{
x}v.p.sv .tuutr=.U~.
sv.}st.st
t{~|t};.wt.tuutr.
~ut}.t{pt.~.p.w~<t|.r{x}xrp{.
~r~|t.qtx}v.x}
txvpts=._~~u.~u.
r~}rt.sxt.~ut}.t
t.p.wt.{x}z.
qttt}.tr{x}xrp{.sxt.p}s.s~t.
p}vx}v.r{x}xrp{.sxt=.cw;.~~u.~u.
r~}rt.sxt.vt}tp{{.pt.}~.
uuxrxt}.~.~
t.tuuxrpr.~u.p.sv.
qtrpt.wt.~
xst.~}{.t{x|x}p.
x}u~|px~}.pq~.wt.tuutr.~u.p.sv=.
X}.pssxx~}.~.wt.{prz.~u.q{xwts.
pstpt.p}s.t{{<r~}~{{ts.tuuxrpr.
sxt.~
x}v.tuuxrpr;.wt.rxtxp.u~.
pstpt.put.sxt.wp.p{~.}~.
qtt}.|t=.X|~p}{;.x}.x.sxrx~}.
~u.wt.ux
t<p.t.ts.~.stt|x}t.
wtwt.p.sv.wp.p.rt}{.
prrtts.|tsxrp{.t; .STP.pxs;.]~.
sv.rp}.qt.r~}xstts.put.x}.wt.
pqpr=.bput.wp.|tp}x}v.~}{.wt}.
RQ!11111
Hto!11126
Hov!5812
Uhov!5813
64812!
ysvts.pvpx}.wt.x}t}sts.t.~u.wt.
sv;.x.z}~}.tuutrx
t}t;.x.z}~}.
p}s.~t}xp{.xz;.wt.t
tx.~u.wt.
x{{}t.~.qt.tpts;.p}s.wt.p
px{pqx{x.
~u.p{t}px
t.t|tsxt .6DF.Ua.@?D?C7=.
fwt}.stt|x}x}v.wtwt.p.sv.
~sr.x.put.p}s.tuutrx
t.u~.p}.
x}sxrpx~};.USP.tu~|.p}.tt}x
t.
xz<qt}tux.p}p{x.~.stt|x}t.
wtwt.wt.xz.~ts.q.wt.sv.
~sr .xst.tuutr.pt.~txvwts.q.
wt.sv.~sr .~t}xp{.qt}tux.u~.
p.pxr{p.x}sxrpx~}=.cw;.r~}p.
~.wt.txx~}t .ptx~}.wp.
|pxyp}p.wp.prrtts.put;.x}.wt.
pqt}rt.~u.p}.prrtts.wtptxr.
x}sxrpx~}.wxrw.rp}.qt.txvwts.
pvpx}.|pxyp}p .xz;.|pxyp}p.
s~t.}~.pxu.wt.t{t|t}.u~.wp
x}v.
pstpt.put.sxt.rw.wp.
tt.|p.r~}r{st.wp.x.x.put.u~.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
cwt.u~w.~u.wt.ux
t.t{t|t}.u~.
stt|x}x}v.rt}{.prrtts.
|tsxrp{.t .txt.wp.wt.}px~}p{.
r~||}x.~u.tt;.p{xuxts.q.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
sv;.prrt.wt.put.p}s.
tuutrx
t}t.~u.wt.qp}rt.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
P.|ptxp{.r~}u{xr.~u.~x}x~}.p|~}v.
tt.tr{st.p.ux}sx}v.~u.
r~}t}=.\tsxrp{.prxx~}t.w~.
pt.}~.tt.x}.t
p{px}v.sv.pt.
}~.p{xuxts.~.stt|x}t.wtwt.p.
sv.x.vt}tp{{.tr~v}xts.p.put.p}s.
tuutrx
t.~.|tt.]SP.txt|t}.6DF.
Ua.@?CHH¤@?D?D7=.
cwtt.x.}~.t
xst}rt.wp.wtt.x.p.
r~}t}.p|~}v.p{xuxts.tt.wp.
|pxyp}p.x.put.p}s.tuutrx
t.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
P.sxrts.pq~
t;.wtt.pt.}~.
pstpt.rxt}xuxr.sxt.wp.w~.
|pxyp}p.x.put.p}s.tuutrx
t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
X}.pssxx~};.wtt.x.}~.t
xst}rt.wp.p.
r~}t}.~u.p{xuxts.tt.wp
t.
prrtts.wt.put.p}s.tuutrx
t}t.~u.
|pxyp}p.u~.t.x}.tpx}v.p.trxuxr;.
tr~v}xts.sx~st=.P{w~vw.|tsxrp{.
prxx~}t.pt.}~.p{xuxts.q.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
sv;.t.p{~.}~t.wp.wt.P\P .
t~;.t}x{ts.dt.~u.Rp}}pqx.u~.
\tsxrx}p{._~t; .s~t.}~.prrt.
wp.|pxyp}p.rt}{.wp.prrtts.
|tsxrp{.t=.Uwt|~t;.qpts.~}.wt.
pq~
t.stux}xx~}.~u.p.p{xuxts.tt ;.
w~.x.p}.x}sx
xsp{.p{xuxts.q.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
p.sv;.pt<{t
t{.|tsxrp{.|pxyp}p.
{p.s~.}~.~
xst.t
xst}rt.~u.p.
r~}t}.p|~}v.p{xuxts.tt.wp.
|pxyp}p.x.put.p}s.tuutrx
t.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
G<^HT^HO^23CWR3/UIO
23CWR3
64813!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
P.~.wt.uxuw.p.~u.wt.t;.wxrw.
txt.wp.x}u~|px~}.r~}rt}x}v.wt.
rwt|x;.wp|pr~{~v;.~xr~{~v;.
p}s.tuutrx
t}t.~u.|pxyp}p.~.qt.
t~ts.x}.uuxrxt}.stpx{;.wt.
rxt}xuxr.t
xst}rt.tvpsx}v.p{{.~u.wtt.
ptr.x.}~.p
px{pq{t.x}.uuxrxt}.
stpx{.~.p{{~.pstpt.rxt}xuxr.
rx}=.btrxuxrp{{;.wt.rxt}xuxr.
t
xst}rt.tvpsx}v.|pxyp}p .
rwt|x.x}.t|.~u.p.trxuxr.
Ecppcdku!px}.wp.r~{s.~srt.
p}spsxts.p}s.t~srxq{t.s~t.x.
}~.rt}{.p
px{pq{t=.
P{t}pt{;.p.sv.rp}.qt.r~}xstts.
~.wp
t.p.rt}{.prrtts.|tsxrp{.
t.xw.t
tt.txrx~} [email protected]=b=R=.
G@A6q76A76Q77;.p.p{{~ts.}st.wt.
x{px~}.u~.p.brwts{t.XX.sv=.ht;.
p.pts.pq~
t;.rt}{.|pxyp}p.
s~t.}~.wp
t.p}.prrtts.|tsxrp{.t;.
t
t}.}st.r~}sxx~}.wtt.x.t.x.
t
tt{.txrts=.
X}.r~}r{x~};.~.spt;.ttprw.~}.
|pxyp}p .|tsxrp{.t.wp.}~.
~vtts.~.wt.~x}.wtt.|pxyp}p.
x.r~}xstts.~.wp
t.p.rt}{.
prrtts.|tsxrp{.t .~.p.rt}{.
prrtts.|tsxrp{.t.xw.t
tt.
txrx~}= .
t.~u.x{{xrx.sv;.x}r{sx}v.|pxyp}p;.
x.x}rtpx}v=.cwt.A?@A.]bSdW.
tx|pt.wp.AB=H.|x{{x~}.x}sx
xsp{.
~
t.@A.tp.~u.pvt.6H=A.trt}.~u.wt.
d=b=.~{px~}7.rt}{.t.x{{xrx.
sv;.wxrw.x.p}.x}rtpt.~u.C=G.
|x{{x~}.x}sx
xsp{.u~|.A??C.wt}.
@H=@.|x{{x~}.x}sx
xsp{.6F=H.trt}.~u.
wt.d=b=.~{px~}7.tt.rt}.x{{xrx.
sv.t=.]bSdW.t~.|pxyp}p.p.
wt.|~.r~||~}{.ts.x{{xrx.sv;.
xw.@G=H.|x{{x~}.x}sx
xsp{.6F=B.
trt}.~u.wt.d=b=.~{px~}7.
rt}{.x}v.|pxyp}p.x}.A?@A=.cwx.
ttt}.p}.x}rtpt.~u.C=B.|x{{x~}.
x}sx
xsp{.u~|.A??C;.wt}.@C=E.
|x{{x~}.x}sx
xsp{.6E=@.trt}.~u.wt.
d=b=.~{px~}7.tt.rt}.|pxyp}p.
t=.
cwt.|py~x.~u.x}sx
xsp{.w~..
|pxyp}p.p.{tp.~}rt.x}.wtx.{xutx|t.
s~.}~.rt}{.t.|pxyp}p=.cwt.
A?@A.]bSdW.tx|pt.wp.@@@=A.
|x{{x~}.x}sx
xsp{.6CA=G.trt}.~u.wt.
d=b=.~{px~}7.wp
t.ts.|pxyp}p.p.
{tp.~}rt.x}.wtx.{xutx|t=.Qpts.~}.
wx.tx|pt.p}s.wt.tx|pt.u~.wt.
}|qt.~u.x}sx
xsp{.rt}{.x}v.
|pxyp}p;.p~x|pt{.@E=H.trt}.
~u.w~t.w~.wp
t.xts.|pxyp}p.p.
5/!Kvu!Jkuvqt{!cpf!Ewttgpv!Rcvvgtp!qh!
{tp.~}rt.x}.wtx.{xutx|t.rt}{.t.
Cdwug!
|pxyp}pJ.r~}
tt{;.GB=@.trt}.s~.
d}st.wt.u~w.upr~;.wt.btrtp. }~.rt}{.t.|pxyp}p=.X}.t|.~u.
wt.utt}r.~u.|pxyp}p.t;.p}.
|.r~}xst.wt.wx~.p}s.rt}.
pt}.~u.|pxyp}p.pqt=.P.
pxt.~u. tx|pts.C?=B.trt}.~u.x}sx
xsp{.
~rt.~
xst.spp.}trtp.~.pt. w~.ts.|pxyp}p.x}.wt.p.|~}w.
pqt.pt}.p}s.t}s.~u.|pxyp}p=. ts.|pxyp}p.~}.A?.~.|~t.sp.
xwx}.wt.p.|~}w=.cwx.p|~}.
cwt.spp.x}sxrp~.~u.|pxyp}p.t.
x}r{st.wt.]bSdW;.\cU;.SPf];.p}s. r~t~}s.~.p}.tx|pts.F=E.|x{{x~}.
x}sx
xsp{.w~.ts.|pxyp}p.~}.p.
cTSb=.cwt.u~{{~x}v.qxtu{.strxqt.
spx{.~.p{|~.spx{.qpx=.
tprw.spp.~rt;.p}s.||pxt.wt.
b~|t.rwpprtxxr.~u.|pxyp}p.
spp.u~|.tprw.~rt=.
t.pt.t{pts.~.pvt;.vt}st;.p}s.
Pcvkqpcn!Uwtxg{!qp!Ftwi!Wug!cpf!
rx|x}p{.yxrt.t|.x}
~{
t|t}=.X}.
Jgcnvj!)PUFWJ* .
~qt
x}v.t.p|~}v.sxuutt}.pvt.
Prr~sx}v.~.A?@A.]bSdW spp;.wt. r~w~;.wt.|py~x.~u.x}sx
xsp{.w~.
|~.trt}.tp.xw.r~|{tt.spp;.wt. rt}{.t.|pxyp}p.pt.w~}.~.qt.
qttt}.wt.pvt.~u.@G¤AD;.xw.@G=F.
trt}.~u.wx.pvt.v~.rt}{.x}v.
|pxyp}p=.X}.wt.AE.p}s.~{st.pvt.
v~;.D=B.trt}.~u.x}sx
xsp{.
rt}{.t.|pxyp}p=.Pssxx~}p{{;.
x}.x}sx
xsp{.pvts.@A.tp.p}s.~{st;.
|p{t.t~ts.|~t.rt}.|pxyp}p.
t.wp}.ut|p{t=.
]bSdW.x}r{st.p.txt.~u.tx~}.
px|ts.p.ptx}v.wt.t
p{t}rt.~u.
stt}st}rt.p}s.pqt.~u.sxuutt}.
qp}rt.x}.wt.p.@A.|~}w= X}.
A?@A;.|pxyp}p.p.wt.|~.r~||~}.
x{{xrx.sv.t~ts.q.x}sx
xsp{.xw.
p.tp.stt}st}rt.~.pqt=.P}.
tx|pts.C=B.|x{{x~}.x}sx
xsp{.|tt.
wt.]bSdW.rxtxp.u~.|pxyp}p.
stt}st}rt.~.pqt.x}.A?@A=.cwt.
tx|pts.pt.p}s.}|qt.~u.
x}sx
xsp{.xw.|pxyp}p.stt}st}rt.
~.pqt.wp.t|px}ts.x|x{p.u~|.
A??A.~.A?@A=.X}.pssxx~}.~.spp.~}.
stt}st}rt.p}s.pqt;.]bSdW.
x}r{st.tx~}.px|ts.p.ptx}v.
tp|t}.u~.p.qp}rt.t.~q{t|= .
X}.A?@A;.p}.tx|pts.HDF;???.t~}.
trtx
ts.tp|t}.u~.|pxyp}p.t.
sx}v.wtx.|~.trt}.tp|t}.x}.
wt.tp.x~.~.wt.
t=.
]bSdW.~
xst.}px~}p{.tx|pt.~u.wt.
t
p{t}rt.p}s.x}rxst}rt.~u.x{{xrx.sv;.p{r~w~{.
p}s.~qprr~.t.x}.wt.d}xts.bpt=.]bSdW.x.p}.
p}}p{.s.r~}srts.q.bP\WbP=._x~.~.
A??A;.wt.sppqpt.p.z}~}.p.wt.]px~}p{.
W~tw~{s.b
t.~}.Sv.Pqt.6]WbSP7=.
]bSdW.x{xt.p.}px~}p{{.ttt}px
t.p|{t.
~u.d}xts.bpt.rx
x{xp};.}~}<x}xx~}p{xts.
~{px~}.pvts.@A.tp.p}s.~{st=.cwt.
t.
tr{st.w~|t{t.t~{t.w~.s~.}~.t.wt{t;.
prx
t.|x{xp.t~}}t{;.p}s.txst}.~u.
x}xx~}p{.v~.pt.rw.p.ypx{.p}s.
w~xp{=.cwt.
t.xst}xuxt.wtwt.p}.
x}sx
xsp{.ts.p.sv.xwx}.p.trxuxr.x|t.
tx~s;.q.s~t.}~.xst}xu.wt.p|~}.~u.wt.sv.
ts.~}.tprw.~rrpx~}=.]bSdW.stux}t.rt}.
t .p.wp
x}v.ts.wt.qp}rt.xwx}.wt.|~}w.
x~.~.wt.s=.
[email protected]<00yyy/ucojuc/iqx0fcvc0!
PUFWJ/curz/!
cwtt.tx~}.pt.ts.~.r{pxu.t~}.
p.stt}st}.~}.~.pqx}v.trxuxr.qp}rt.
twpqx{xpx~}.uprx{x.6~pxt}.~.x}pxt}7;.
|t}p{.wtp{w.rt}t;.t|tvt}r.~~|;.x
pt.
s~r~ .~uuxrt;.x~}.~.ypx{;.~.p.t{u<wt{.v~;.
rw.p.P{r~w~{xr.P}~}|~.~.]pr~xr.
P}~}|~= .6]bSdW;.A?@B7=.
\~}x~x}v.wt.Ut.x.p.}px~}p{.
t.wp.
prz.sv.t.t
p{t}rt.p}s.t}s.p|~}v.
ps~{trt}.x}.wt.d}xts.bpt=.\cU.x.t~ts.
p}}p{{.q.wt.X}xt.u~.b~rxp{.attprw.p.wt.
d}x
tx.~u.\xrwxvp}.}st.p.vp}.u~|.]XSP=.
T
t.x}v;.\cU.
t.Gw;.@?w;.p}s.@Aw.
vpst.x}.p}s~|{.t{trts.d=b=.rw~~{=.\cU.wp.
qtt}.r~}srts.x}[email protected]~.@Aw.vpst.p}s.
x}rt.@[email protected]~.Gw.p}s.@?w.vpst=.cwt.\cU.
t.tt}.spp.x}.t|.~u.t
p{t}rt.p|~}v.
wt.p|{t.x}t
xtts=.U~.A?@A;.wt.{pt.tp.
xw.r~|{tt.spp;.wt.p|{t.xt.tt.@D;A??¥.
Gw.vpstJ.@B;B??¥@?w.vpstJ.p}s.@B;A??¥.
@Aw.vpst=.X}.p{{;.p.~p{.~u.pq~.C@;F??.st}.
~u.BGH.rw~~{.pxrxpts.x}.wt.A?@B.\cU=.
[email protected]<00yyy/oqpkvqtkpivjghwvwtg/qti0
kpfgz/jvon/!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
qpts.~}.rxtxp.trxuxts.x}.wt.Fkcipquvke!cpf!
Uvcvkuvkecn!Ocpwcn!qh!Ogpvcn!Fkuqtfgt-!Cw.tsxx~}.
6Sb\¤Xe7=.cwt.tx~}.t{pts.~.stt}st}rt.pz.
pq~.wtp{w.p}s.t|~x~}p{.~q{t|.p~rxpts.
xw.qp}rt.t;.}rrtu{.pt|.~.r.
s~}.~}.t;.~{tp}rt;.xwspp{;.tsrx}v.~wt.
prx
xxt.~.t.qp}rt;.t}sx}v.p.{~.x|t.
t}vpvx}v.x}.prx
xxt.t{pts.~.qp}rt.t;.~.
x}v.wt.qp}rt.x}.vtpt.p}xxt.~.u~.
{~}vt.x|t.wp}.x}t}sts=.cwt.tx~}.~}.pqt.
pz.pq~.~q{t|.p.~z;.w~|t;.p}s.rw~~{J.
~q{t|.xw.up|x{.~.uxt}sJ.wxrp{.sp}vtJ.
p}s.~q{t.xw.wt.{p.st.~.qp}rt.t=.
Stt}st}rt.x.r~}xstts.~.qt.p.|~t.t
tt.
qp}rt.t.~q{t|.wp}.pqt.qtrpt.x.
x}
~{
t.wt.rw~{~vxrp{.p}s.wx~{~vxrp{.tuutr.
~u.~{tp}rt.p}s.xwspp{= .6]bSdW;.A?@B7=.
Tx|pt.= = =.tut.~.tp|t}.trtx
ts.u~.
x{{xrx.sv.~.p{r~w~{.t;.~.u~.|tsxrp{.~q{t|.
p~rxpts.xw.wt.t.~u.x{{xrx.sv.~.p{r~w~{=.
cwx.x}r{st.tp|t}.trtx
ts.x}.wt.p.tp.p.
p}.{~rpx~};.rw.p.p.w~xp{.6x}pxt}7;.
RQ!11111
Hto!11127
Hov!5812
Uhov!5813
Oqpkvqtkpi!vjg!Hwvwtg!)OVH*
.
Prr~sx}v.~.\cU; pt.~u.
|pxyp}p.p}s.x{{xrx.sv.t.str{x}ts.
u~.p{{.wtt.vpst.u~|.A??D.w~vw.
A??F=.W~t
t;.px}v.p~}s.A??G;.
pt.~u.p}}p{.t.~u.x{{xrx.sv.p}s.
|pxyp}p.x}rtpts.w~[email protected]~.p{{.
wtt.vpst=.\pxyp}p.t|px}ts.wt.
|~.xst{.ts.x{{xrx.sv.sx}v.p{{.
x|t.tx~s=.cwt.t
p{t}rt.~u.p}}p{.
p}s.p.|~}w.|pxyp}p.t.x}.@?w.
p}s.@Aw.vpst.x}[email protected].vtpt.wp}.
x}.A??D=.cpq{t.@.{x.wt.{xutx|t;.
p}}p{;.p}s.|~}w{.t
p{t}rt.pt.~u.
px~.sv.u~.Gw;.@?w;.p}s.@Aw.
vpst.x}.A?@B=.
G<^HT^HO^23CWR3/UIO
23CWR3
A;CEA;HCG.6jRXlI.A;@@A;GEG.~.A;G@B;?AG7.
xx.x}
~{
ts.sv.|xt.~.pqt=.
Sx}v.wt.p|t.tx~s;.SPf].
X|~p}{;.|p}.upr~.rp}.
tx|pt.wp.@;ADA;D??.6RXI.HFE;@EH.~.
x}u{t}rt.wt.tx|pt.~u.TS.
xx;.
@;DAG;[email protected]v.t{pts.TS.
xx.
x}r{sx}v.t}s.x}.~
tp{{.t.~u.p.
x}
~{
ts.x{{xrx.sv=.cw;.~
t.wp{u.~u.
qp}rt.p.t{{.p.t}s.x}.wt.
p{{.sv<t{pts.TS.
xx.p~rxpts.
tp~}.u~.TS.pvt=.U~.x}p}rt;.~|t. xw.sv.|xt.~.pqt.x}
~{
ts.p}.
sv.t.|p.
xx.TS.u~.{xut<.
x{{xrx.sv=.U~.TS.
xx.x}
~{
x}v.
wtpt}x}v.xt.wx{t.~wt.|p.
x{{xrx.sv;.DE=B.trt}.x}
~{
ts.
xx.~.ttz.rpt.u~.st~xuxrpx~}.
|{x{t.sv.wx{t.CB=F.trt}.
qtrpt.wt.}ttsts.rtxuxrpx~}.qtu~t. x}
~{
ts.p.x}v{t.sv=.
t}tx}v.tp|t}=.Pssxx~}p{{;.
\pxyp}p.p.x}
~{
ts.x}.CDD;EEG.
SPf].spp.s~.}~.sxx}vxw.wt.sv. TS.
xx.6RXI.BF?;HHD.~.DC?;BC?7;.wx{t.
t~}xq{t.u~.wt.TS.
xx.u~|.~wt.
r~rpx}t.p.x}
~{
ts.x}.D?D;AAC.6RXI.
sv.wp.|p.wp
t.qtt}.ts.
BAC;AEA.~.EGE;[email protected].
xx;.wt~x}.
r~}r~|xp}{=.P.pts.x}.p.SPf].
p.x}
~{
ts.x}.ADG;CGA.6RXI.A?D;?CE.~.
t~;.bx}rt.|pxyp}p>wpwxw.x.
B@@;[email protected].
xx.p}s.x|{p}.
utt}{.tt}.x}.r~|qx}px~}.xw. x}r{sx}v.p|wtp|x}t.p}s.
~wt.sv;.wt.tp~}.u~.wt.TS.
xx.
|twp|wtp|x}t.tt.x}
~{
ts.x}.
|p.qt.|~t.t{t
p}.~.wt.~wt.
@DH;GC?.6RXI.@??;@HH.~.A@H;[email protected].
sv67.x}
~{
ts.x}.wt.tx~st= .
xx=.^wt.x{{xrx.sv;.rw.p._R_;.
\S\P;.VWQ.p}s.[bS.tt.|rw.{t.
U~•.A?@@;.SPf] tx|pt.p.~p{.
utt}{.p~rxpts.xw.TS.
xx=.
~u.D;?EF;BFC.6HD.trt}.r~}uxst}rt.
cwt.}|qt.~u.TS.
xx.x}
~{
x}v.
x}t
p{.jRXlI.C;E@E;FDB.~.D;D@F;HHD7.
|pxyp}p.wp.x}rtpts.q.EA.trt}.
sv<t{pts.TS.
xx.u~|.wt.t}xt.
x}rt.A??C=.
d}xts.bpt=.^u.wtt;.p~x|pt{.
\pxyp}p<t{pts.TS.
xx.tt.|~.
utt}.p|~}v.~}v.ps{.p}s.
SPf].x.p.}px~}p{.~qpqx{x.
t.~u.wt.
|x}~=.X}sx
xsp{.}st.wt.pvt.~u.@G.
d=b=.w~xp{.xw.TS.stxv}ts.~.~qpx}.
x}u~|px~}.~}.sv.t{pts.TS.
xx=.SPf].x.
prr~}ts.u~.@B=A.trt}.~u.wtt.
~}~ts.q.bP\WbP=.cwt.SPf].t|.
|pxyp}p<t{pts.
xx;.wttp.wx.
~
xst.x}u~|px~}.~}.wt.wtp{w.r~}tt}rt.~u.
pvt.v~.prr~}ts.u~.p~x|pt{.
sv.t.x}.wt.d}xts.bpt;.p.|p}xutts.q.
@=A.trt}.~u.TS.
xx.x}
~{
x}v.
sv<t{pts.
xx.~.TS=.cwt.TS.spp.u~|.p.
ttt}px
t.p|{t.~u.w~xp{.t|tvt}r.
r~rpx}t;.p}s.{t.wp}.@.trt}.~u.TS.
stp|t}.pt.txvwts.~.~srt.}px~}p{.
xx.x}
~{
x}v.wt~x}=.W~t
t;.wt.
tx|pt=.X|~p}{;.SPf].spp.p}s.tx|pt;.
pvt.v~.xw.wt.|~.|pxyp}p<.
px}v.x}.A??C;.pt.}~.r~|ppq{t.~.w~t.u~.
x~.tp.qtrpt.~u.
p.rwp}vt.x}.wt.
t{pts.TS.
xx.p.qttt}.AD.p}s.AH.
|tw~s~{~v.ts.~.r~{{tr.wt.spp=.Uwt|~t;.
tp.~{s=.ht;.qtrpt.~{px~}.
tx|pt.u~.A??C.pt.wt.ux.~.qt.qpts.~}.p.
sxuut.qttt}.pvt.v~;.p.
tstxv}ts.p|{t.~u.w~xp{;.wxrw.t}sts.x}.
p}spsxts.|tpt.u~.~{px~}.
A?@@=.
A?@@J.jvvr<00yyy/ucojuc/iqx0fcvc0fcyp/curz/! xt.x.tu{.~.|pzt.r~|px~}=.U~.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Ftwi!Cdwug!Yctpkpi!Pgvyqtm!
)FCYP* .
XgtFcvg!Ugr>22@3125!
32<1;!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11128
Hov!5812
Uhov!5813
64814!
|pxyp}p;.wt.pt.~u.TS.
xx.t.
@??;???.~{px~}.tt.wxvwt.u~.
pxt}.pvts.@G.~.A?.6CCB=G.TS.
xx.
t.@??;???7.p}s.u~.pxt}.pvts.A@.~.
AC.6CCE=H.TS.
xx.t.@??;???7=.
fwx{t.SPf].~
xst.tx|pt.u~.
TS.
xx.p~rxpts.xw.wt.t.~u.
|tsxrp{.|pxyp}p.u~.A??H¤A?@@;.wt.
p{xsx.~u.wtt.tx|pt.x.
tx~}pq{t=.Qtrpt.wt.sv.x.}~.
p~
ts.q.wt.USP;.t~x}v.|tsxrp{.
|pxyp}p.|p.qt.x}r~}xt}.p}s.
t{xp}.~}.p.}|qt.~u.upr~.x}r{sx}v.
wtwt.wt.pxt}.t{u<t~.wt.
|pxyp}p.t.p.|tsxrx}p{;.w~.wt.
tpx}v.wtp{w.rpt.~
xst.tr~s.wt.
|pxyp}p.t;.p}s.{p{.w~.wt.
bP\WbP.r~st.x}tt.wt.t~=.
P{{.~u.wtt.ptr.x{{.
p.vtp{.
qttt}.pt.xw.|tsxrp{.|pxyp}p.
{p.p}s.pt.xw~.|tsxrp{.
|pxyp}p.{p=.cw;.t
t}.w~vw.
tx|pt.pt.t~ts.u~.|tsxrp{.
|pxyp}p.t{pts.TS.
xx;.|tsxrp{.
|pxyp}p.tx|pt.rp}}~.qt.ptts.
xw.p}.prrtpq{t.prrpr.p.wx.
x|t;.p.USP.wp.}~.p~
ts.
|pxyp}p.tp|t}.~u.p}.|tsxrp{.
r~}sxx~}=.cwtt.spp.w~.wt.
sxuuxr{.x}.t
p{px}v.pqt.~u.p.
~sr.wp.x.}~.rt}{.p~
ts.
q.USP;.q.pw~xts.u~.|tsxrp{.t;.
p{qtx.x}r~}xt}{;.p.wt.pt.{t
t{=.
cw;.t.qt{xt
t.wt.{xzt{xw~~s.~u.wt.
tpx}v.wtp{w.rpt.~
xst.~.
bP\WbP.r~st.pxqx}v.wt.TS.
xx.
~.|tsxrp{.|pxyp}p .
t.
|pxyp}p .~.qt.
t.{~=.^
tp{{;.wt.
p
px{pq{t.spp.pt.x}pstpt.~.
rwpprtxt.x.pqt.p.wt.r~||}x.
{t
t{=.
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/116>0IRJ@
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
64815!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
6/!Vjg!Ueqrg-!Fwtcvkqp-!cpf!
Ukipkhkecpeg!qh!Cdwug!
_x|p.|pxyp}p.pqt.prr~}ts.
d}st.wt.uxuw.upr~;.wt.btrtp.
u~.@G=@.trt}.~u.p{{.A?@@.cTSb .
|.r~}xst.wt.r~t;.spx~};.p}s.
ps|xx~}=.X}sx
xsp{.ps|xts.u~.
xv}xuxrp}rt.~u.|pxyp}p.pqt=.
x|p.|pxyp}p.pqt.tt.}tp{.
Prr~sx}v.[email protected]p.u~|.]bSdW.
wtt<pt.6FB=C.trt}7.|p{t;.p}s.
p}[email protected]p.u~|.\cU;.|pxyp}p.
p{|~.wp{u.6CD=A.trt}7.tt.wxt=.
t|px}.wt.|~.tt}x
t{.ts.
cwt.p
tpvt.pvt.p.ps|xx~}.p.AC.
x{{tvp{.sv.x}.wt.d}xts.bpt;.xw.
tp.~{s;.p}s.B@=@.trt}.~u.
CA=G.trt}.~u.d=b=.x}sx
xsp{.~
t.pvt.
x}sx
xsp{.ps|xts.u~.x|p.
@A.6@@@=A.|x{{x~}7.p}s.CD=D.trt}.~u.
|pxyp}p.pqt.tt.}st.wt.pvt.~u.
@Aw.vpst.wp
x}v.ts.|pxyp}p.p.
@G=.cwt.t~ts.utt}r.~u.|pxyp}p. {tp.~}rt.x}.wtx.{xutx|t=.P{w~vw.wt.
t.p.AC=B.trt}.t~x}v.spx{.
|py~x.~u.x}sx
xsp{.~
t[email protected]=@.
t=.P{|~.p{{.6HE=G.trt}7.x|p.
trt}7.w~.wp
t.t
t.ts.|pxyp}p.
|pxyp}p.t.x{xts.wt.qp}rt.
x}.wtx.{xutx|t.s~.}~.t.wt.sv.
q.|~zx}v=.Pssxx~}p{{;.HA=H.trt}. |~}w{;.@G=H.|x{{x~}.x}sx
xsp{.6F=B.
t~ts.x}v.|pxyp}p.u~.wt.ux.
trt}.~u.wt.d=b=.~{px~}7.t~.
wp.wt.ts.|pxyp}p.xwx}.wt.p.
x|t.qtu~t.wt.pvt.~u.@G=.
B?.sp=.P}.tp|x}px~}.~u.t.p|~}v.
P}.x|~p}.ptr.~u.cTSb.
px~.pvt.r~w~.w~vw.]bSdW.
ps|xx~}.spp.u~.|pxyp}p.x.~u.wt.
st|~}pt.wp.|~}w{.t.~rr.
tutp{.~rt.u~.tp|t}=.
x|px{.p|~}v.r~{{tvt<pvts.
btrxuxrp{{;.x|p.|pxyp}p.
x}sx
xsp{;.xw.t.s~x}v.~uu.
ps|xx~}.tt.{t.{xzt{.wp}.p{{.
wp{.put.pvt.AD=.Pssxx~}p{{;.
~wt.ps|xx~}.~.txwt.qt.t{u<.
]bSdW.spp.w~.wt.}|qt.~u.
tutts.~.tutts.q.p}.x}sx
xsp{.u~. x}sx
xsp{.t~x}v.p<|~}w.t.~u.
tp|t}=.X}tps;.wt.rx|x}p{.yxrt.
|pxyp}p.wp.x}rtpts.q.C=B.|x{{x~}.
t|.tutts.|~t.wp}.wp{u.6D@=E.
x}sx
xsp{.x}rt.A??C=.Spp.u~|.\cU.
trt}7.~u.x|p.|pxyp}p.
w~.wp.p}}p{.t
p{t}rt.~u.
ps|xx~}=.
|pxyp}p.t.str{x}ts.u~.p{{.wtt.
bx}rt.A??B;.wt.trt}.~u.ps|xx~}. vpst.u~|.A??D.w~vw.A??F;.wt}.
qtvp}.~.xt.w~vw.A?@B=.
u~.x|p.|pxyp}p.pqt.x}rtpts.
Pssxx~}p{{;.x}.A?@B;.@=@.trt}.~u.Gw.
u~|.@D=D.trt}.~u.p{{.ps|xx~}.x}.
vpst;.C=?.trt}.~u.@?w.vpst;.p}s.
A??B.~.@G=@.trt}.x}.A?@@=.cwx.
E=D.trt}.~u.@Aw.vpst.t~ts.
x}rtpt.x.{t.wp}.wt.x}rtpt.tt}.
u~.ps|xx~}.u~.x|p.~x~xs.~wt. spx{.t.~u.|pxyp}p;.stux}ts.p.t.
~}.A?.~.|~t.sp.xwx}.wt.p.B?.
wp}.wt~x};.wxrw.x}rtpts.u~|.A=G.
sp=.
trt}.x}.A??B.~.F=B.trt}.x}.A?@@=.
cwt.A?@@.SPf].spp.w~.wp.
X}.r~}p;.wt.ps|xx~}.u~.x|p.
|pxyp}p.t.p.|t}x~}ts.x}.
r~rpx}t.pqt.str{x}ts.u~|.H=G.trt}.
CDD;EEG.TS.
xx;.wxrw.p|~}.~.
x}.A??B.~.A=?.trt}.x}.A?@@=.
p~x|pt{.BE=C.trt}.~u.p{{.x{{xrx.
sv<t{pts.TS.
xx= .
cwt.cTSb.t|.x.p.~u.bP\WbP .Sv.
cTSb.spp.u~.A?@@.w~.wp.@G=@.
p}s.P{r~w~{.bt
xrt.X}u~|px~}.bt|.6^uuxrt.~u.
trt}.~u.p{{.ps|xx~}.tt.u~.
P{xts.brxt}rt;.bP\WbP7=.cwt.cTSb.t~.
x|p.|pxyp}p.pqt= Qttt}.
tt}.x}u~|px~}.~}.wt.st|~vpwxr.p}s.
A??B.p}s.A?@@;.wtt.p.p.A=E.trt}.
qp}rt.t.rwpprtxxr.~u.wt.@=G.|x{{x~}.
p}}p{.ps|xx~}.~.tp|t}.u~.p{r~w~{.p}s.
x}rtpt.x}.wt.}|qt.~u.cTSb.
sv.pqt.x}.uprx{xxt.wp.t~.~.x}sx
xsp{.
ps|xx~}.u~.x|p.|pxyp}p.t=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Vtgcvogpv!Grkuqfg!Fcvc!Ugv!)VGFU*
.
pt.ps|x}xpx
t.spp.t|=.btrxuxrp{{;.
cTSb.x}r{st.uprx{xxt.{xrt}ts.~.rtxuxts.q.wt.
pt.~.~
xst.qp}rt.pqt.tp|t}.p}s.x.
txts.q.wt.pt.~.~
xst.cTSb.r{xt}<{t
t{.
spp=.Uprx{xxt.wp.t~.cTSb.spp.pt.w~t.
trtx
x}v.bpt.p{r~w~{.p}s.sv.pvt}r.u}s.u~.
wt.~
xx~}.~u.p{r~w~{.p}s.sv.tp|t}.
t
xrt=.bx}rt.cTSb.x.qpts.~}{.~}.t~.u~|.
wtt.uprx{xxt;.cTSb.spp.s~.}~.ttt}.wt.~p{.
}px~}p{.st|p}s.u~.qp}rt.pqt.tp|t}.~.
wt.t
p{t}rt.~u.qp}rt.pqt.x}.wt.vt}tp{.
~{px~}=.cwt.x|p.v~p{.u~.cTSb.x.~.
|~}x~.wt.rwpprtxxr.~u.tp|t}.=tx~st.
u~.qp}rt.pqt=.X|~p}{;.cTSb.x.p}.
ps|xx~}<qpts.t|;.wtt.ps|xp}rt.~.
tp|t}.x.r~}ts.p.p}.p}~}|~.p{{=.U~.
x}p}rt;.p.vx
t}.x}sx
xsp{.w~.x.ps|xts.~.
tp|t}.xrt.xwx}.p.vx
t}.tp.~{s.qt.
r~}ts.p.~.ps|xx~}=.cwt.|~.trt}.tp.
xw.r~|{tt.spp.x.A?@@=.
A?@@J.jvvr<00yyy/ucojuc/iqx0fcvc0
FCUKU/curzAst?v$VGFU/!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
\p}.upr~.rp}.x}u{t}rt.wt.tx|pt.~u.TS.
xx;.x}r{sx}v.t}s.x}.wt.tp~}.u~.TS.pvt=.
U~.x}p}rt;.~|t.sv.t.|p.
xx.TS.u~.{xut<.
wtpt}x}v.xt.wx{t.~wt.|p.
xx.~.ttz.
rpt.u~.st~xuxrpx~}.qtrpt.wt.}ttsts.
rtxuxrpx~}.qtu~t.t}tx}v.tp|t}=.
Pssxx~}p{{;.SPf].spp.s~.}~.sxx}vxw.wt.
sv.t~}xq{t.u~.wt.TS.
xx.u~|.~wt.sv.
wp.|p.wp
t.qtt}.ts.r~}r~|xp}{=.P.pts.
x}.p.SPf].t~;.bx}rt.|pxyp}p>wpwxw.x.
utt}{.tt}.x}.r~|qx}px~}.xw.~wt.sv;.
wt.tp~}.u~.wt.TS.
xx.|p.qt.|~t.t{t
p}.~.
wt.~wt.sv67.x}
~{
ts.x}.wt.tx~st= .
P}.x|~p}.ptr.~u.cTSb.ps|xx~}.spp.
u~.|pxyp}p.x.~u.wt.tutp{.~rt.u~.tp|t}=.
btrxuxrp{{;.x|p.|pxyp}p.ps|xx~}.tt.
{t.{xzt{.wp}.p{{.~wt.ps|xx~}.~.txwt.qt.
t{u<tutts.~.tutts.q.p}.x}sx
xsp{.u~.
tp|t}=.X}tps;.wt.rx|x}p{.yxrt.t|.
tutts.|~t.wp}.wp{u.6D@=E.trt}7.~u.x|p.
|pxyp}p.ps|xx~}=.
RQ!11111
Hto!11129
Hov!5812
Uhov!5813
P~x|pt{.E@=D.trt}.~u.x|p.
|pxyp}p.ps|xx~}.x}.A?@@.tt.u~.
x}sx
xsp{.}st.wt.pvt.~u.AD.tp=.
7/!YJCV-!kh!Cp{-!Tkum!Vjgtg!Ku!vq!vjg!
Rwdnke!Jgcnvj!
d}st.wt.xw.upr~;.wt.btrtp.
|.r~}xst.wt.xz.~ts.~.wt.
q{xr.wtp{w.q.|pxyp}p=.Upr~.@;.C;.
p}s.D.x}r{st.p=.sxrx~}.~u.wt.xz.
~.wt.q{xr.wtp{w.p.|tpts.q.
t|tvt}r.~~|.tx~st.p}s.sv.
tp|t}.ps|xx~}=.Pssxx~}p{{;.
Upr~.A.x}r{st.p.sxrx~}.~u.
|pxyp}p .rt}p{.}t
~.t|;.
r~v}xx
t;.rpsx~
pr{p;.p~}~|xr;.
txp~;.p}s.x||}t.t|.tuutr=.
Upr~.E.u~rt.~}.wt.wtp{w.xz.~.
wt.x}sx
xsp{.t.x}.t|.~u.wt.xz.
u~|.prt.p}s.rw~}xr.t.~u.
|pxyp}p;.p.t{{.p.wt.vptp.
w~wtx= .
Tkumu!Htqo!Cewvg!Wug!qh!Octklwcpc!
Prt.t.~u.|pxyp}p.x|px.
rw~|~~.tu~|p}rt;.x}r{sx}v.
r~|{t.pz.tu~|p}rt;.wxrw.
|pzt.~tpx}v.|~~.
twxr{t.~.
wtp
.tx|t}.put.x}v.|pxyp}p.
x}ps
xpq{t.6ap|ptzt.t.p{=;.A??CJ.
ap|ptzt.t.p{=;.A??Ep7=.P.|tp<.
p}p{x.r~}srts.q.[x.t.p{=.6A?@@7.
w~ts.p}.p~rxpx~}.qttt}.
|pxyp}p.t.q.wt.sx
t.p}s.p.
xv}xuxrp}{.x}rtpts.xz.~u.
x}
~{
t|t}.x}.p.rp.prrxst}=.
Pssxx~}p{{;.x}.p.|x}~x.~u.
x}sx
xsp{.w~.t.|pxyp}p;.~|t.
~t}xp{.t~}t.x}r{st.sw~xp.
p}s.rw~{~vxrp{.sxt;.x}r{sx}v.
~{~}vts.p}xt.tprx~}.6Wp}t.t.
p{=;.@HHH7=.
Tkumu!Htqo!Ejtqpke!Wug!qh!Octklwcpc!
P.sxx}rx
t.|pxyp}p.xwspp{.
}s~|t.u~{{~x}v.{~}v.t|.~.
rw~}xr.t.wp.qtt}.xst}xuxts=.cwt.
xwspp{.}s~|t.x}sxrpt.wp.
|pxyp}p.~srt.wxrp{.
stt}st}rt.wp.x.|x{s;.w~<{x
ts;.
p}s.r~|ppq{t.~.~qprr~.xwspp{.
6Qs}t.t.p{=;.A??G7=.\pxyp}p.
xwspp{.}s~|t.x.strxqts.x}.
stpx{.qt{~.}st.Upr~.F=.
cwt.u~{{~x}v.pt.w~.wt.Sb\¤e.
6A?@B7.~u.wt.P|txrp}._rwxpxr.
P~rxpx~}.strxqt.wt.r~}tt}rt.
~u.Ecppcdku pqtI.
X}sx
xsp{.xw.rp}}pqx.t.
sx~st.|p.t.rp}}pqx.w~vw~.
wt.sp.~
t.p.tx~s.~u.|~}w.~.
tp;.p}s.w.|p.t}s.|p}.w~.
p.sp.}st.wt.x}u{t}rt=.^wt.|p.
t.{t.utt}{;.q.wtx.t.rpt.
trt}.~q{t|.t{pts.~.up|x{;.
Ecppcdku!x.wt.t|.ts.x}.wt.Sb\¤e.~.
tut.~.|pxyp}p=.X}.wt.u~{{~x}v.trt.wt.t|.
Ecppcdku!x.x}trwp}vtpq{t.u~.wt.t|.octklwcpc/!
G<^HT^HO^23CWR3/UIO
23CWR3
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
rw~~{;.~z;.~.~wt.x|~p}.
prx
xxt.6ttpts.pqt}rt.p.
~zJ.}tv{tr.~u.up|x{.~q{xvpx~}7=.
_tx~sxr.rp}}pqx.t.p}s.x}~xrpx~}.
rp}.}tvpx
t{.puutr.qtwp
x~p{.p}s.
r~v}xx
t.u}rx~}x}v.p}s.w.x}tutt.
xw.~x|p{.tu~|p}rt.p.~z.~.
rw~~{;.~.{prt.wt.x}sx
xsp{.p.
x}rtpts.wxrp{.xz.wt}.
tu~|x}v.prx
xxt.wp.r~{s.qt.
wxrp{{.wpps~.6sx
x}v.p.rpJ.
{px}v.rtpx}.~J.tu~|x}v.
|p}p{.~z.prx
xxt;.x}r{sx}v.
~tpx}v |prwx}t7=.Pv|t}.xw.
~t.~.pt}.~
t.wt.t.~u.
rp}}pqx.x}.wt.w~|t;.~.x.t.x}.wt.
tt}rt.~u.rwx{st};.rp}.ps
tt{.
x|pr.up|x{.u}rx~}x}v.p}s.pt.
r~||~} utpt.~u.w~t.xw.rp}}pqx.
t.sx~st=.[p;.x}sx
xsp{.xw.
rp}}pqx.t.sx~st.|p.r~}x}t.
x}v.|pxyp}p.stxt.z}~{tsvt.~u.
wxrp{.~q{t|.6rw~}xr.r~vw.
t{pts.~ |~zx}v7.~.rw~{~vxrp{.
~q{t|.6trtx
t.tspx~}.~.
tprtqpx~}.~u ~wt.|t}p{.wtp{w.
~q{t|7.p~rxpts.xw.x.t=.
|pxyp}p.t.{tpsx}v.~.wt~x}.
pssxrx~}= Pssxx~}p{{;.x}.p}~wt.
{~}vxsx}p{ s.~u.A;CCE.ps~{trt};.
|pxyp}p.stt}st}rt.p.}r~||~}.
q.wt} x.sxs.~rr;.wt.r~||~}.
tsxr~ ~u.|pxyp}p.stt}st}rt.
tt.wt.u~{{~x}vI._pt}p{.stpw;.
stx
ts.~rx~<tr~}~|xr.p;.p}s.
qpt{x}t.x{{xrx.sv.t.~wt.wp}.
|pxyp}p.6
~}.bs~.t.p{=;.A??A7=.
fwt}.tp|x}x}v.wt.p~rxpx~}.
qttt}.|pxyp}p.p}s.x{{xrx.sv;.
u~rx}v.~} sv.t.
t.pqt.~.
stt}st}rt; sxuutt} pt} t|tvt=
U~.tp|{t; p.s.tp|x}x}v.wt.
~xq{t.rpp{.t{px~}wx.~u.wt.
vptp.w~wtx u~}s.p.r~t{px~}.
qttt}.|pxyp}p.t.x}.ps~{trt}.
p}s.~wt.x{{xrx.sv.t.x}.tp{.
ps{w~~s.p}s; psyx}v.u~.pvt<{x}zts.
ttxt}rt;.sxs.}~.tuutr.wx.
r~t{px~} 6ep}.V}s.p}s.atqt{{~};.
A?@?7=.W~t
t;.wt} tp|x}x}v.wt.
p~rxpx~} x} t| ~u st
t{~|t} ~u
sv.pqtJ.pvt<{x}zts.t~.p}s.
~rxp{ ~{t.|~stpts.wt.r~t{px~}.
qttt}.|pxyp}p.t.x}.ps~{trt}.
p}s.~wt.x{{xrx.sv.pqt=.bx|x{p{;.
Octklwcpc!cu!c!Icvgyc{!Ftwi!
Stvt}wps.t.p{=.6A??H7.tp|x}ts.wt.
Zp}st{.6@HFD7.~~ts.}tp{.C?.
st
t{~|t} ~u.sv.stt}st}rt.p}s.
tp.pv~.wt.w~wtx.wp.|pxyp}p. u~}s.p}.p~rxpx~}.wp.sxs.}~.
x.p.vptp.sv .wp.{tps.~.wt.t. ~.wt.vptp.w~wtx=.
~.pqt.~u.~wt.x{{xrx.sv=.bx}rt.wp. btrxuxrp{{;.sv.stt}st}rt.p.
x|t; txst|x~{~vxrp{ ttprw.t{~ts. xv}xuxrp}{ p~rxpts xw wt t ~u
wx.t|xt=.^
tp{{;.ttprw.s~t.}~. ~wt.x{{xrx.sv.x~.~.|pxyp}p.
~.p sxtr.rpp{.t{px~}wx.
t=.
qttt}.tv{p |pxyp}p.t.p}s.
X}ttx}v{;.wt.~st.~u.x}xxpx~}.~u.
~wt.x{{xrx.sv.t=.cwt.sxt.
sv.t.tt|.~.stt}s.~}.wt.
tp|x}x}v.wt.vptp.w~wtx.pt.
t
p{t}rt ~u.t.~u.tprw.sv;.wxrw.
{x|xts=.Ux;.x}.vt}tp{;.sxt.trx.
pxt.q.r~}=.Qpts.~}.wt.f~{s.
x}sx
xsp{.x}u{t}rts.q.p.|xps.~u.
Wtp{w.^vp}xpx~}.6fW^7.f~{s.
~rxp{;.qx~{~vxrp{; p}s.tr~}~|xr.upr~. \t}p{.Wtp{w.b
t.wp.x}r{st.spp.
wp.r~}xqt.~.tt}x
t.sv.pqt.
u~|[email protected]t}.r~}xt;.wt.~st.~u.
6Wp{{.4.[}zt;.A??D7=.btr~}s;.|~.
sv.t.x}xxpx~}.
pxt.q.r~}.
sxt.wp.t.wt.w~wtx.wp.
p}s.t{pt ~.t
p{t}rt.~u.sv.t.x}.
|pxyp}p.t.rpt.pqt.~u.x{{xrx.
tprw.r~}.6Stvt}wps.t.p{=;.A?@?7=.
sv.t.wt.stt|x}px
t.|tpt.
btrxuxrp{{;.x}.wt.r~}xt.xw.wt.
pwt.wp}.
{~t t
p{t}rt.~u.|pxyp}p.t;.t.
Sb\¤D.rxtxp.u~.sv.pqt.~.
~u.~wt.x{{xrx.sv.qtu~t.|pxyp}p.
stt}st}rt.~} p}.x{{xrx.sv.6Sb\¤D;.
p.r~||~}= cwx.tt}rt.~u.
A?@B7=.R~}tt}{; p{w~vw.p}.
x}xxpx~}.x {t.r~||~}.x}.r~}xt.
x}sx
xsp{.w~.ts.|pxyp}p.|p.. xw.wxvwt t
p{t}rt.~u.|pxyp}p.
~wt.x{{xrx.sv;.wt.x}sx
xsp{.|p.
t=.P.s.~u.H;AGA.w~tw~{s.x}.wt.
}~.tv{p{.t.sv;.~.wp
t.p.
d}xts.bpt.u~}s.wp.|pxyp}p.t.
sxpv}~x ~u.sv.pqt.~.stt}st}rt=. ~ut} trtsts.wt.t.~u.~wt.x{{xrx.
[x{t.t
xst}rt.~.wt.
svJ.w~t
t; x~.}~}<|pxyp}p.
w~wtx wp.x}xxpx~}.~u.|pxyp}p.
sv.stt}st}rt.p.p{~.utt}{.
t.{tps.~.p}.pqt.sx~st.xw.
r~t{pts.xw wxvwt.{t
t{.~u.x{{xrx.
~wt x{{xrx qp}rt= U~ tp|{t;
sv.pqt 6Stvt}wps.t.p{=;.A??H7=.
~}t {~}vxsx}p{.s.~u.F?G.
Pssxx~}p{{;.x}.p.{pvt.AD<tp.
ps~{trt} st|~}pts.wp.tp{.
{~}vxsx}p{ s.~u.@;ADE.]t.
~}t |pxyp}p t sxs }~ {tps ~
itp{p}s.rwx{st};.wt.pw~.r~}r{sts.
~q{t|pxr.sv.t.6Zp}st{.4.Rwt};.
wp.|pxyp}p t.r~t{pts.~.p}.
A???7=.bx|x{p{; ]prt.t.p{=.6@HFD7.
x}rtpts.xz.~u.pqt.~u.~wt.sv;.
tp|x}ts ext}p|<tp ~{sxt w~
x}r{sx}v.r~rpx}t.p}s.wt~x}.
tt}x
t{.pqts.|pxyp}p.p}s.
6Utv~} t.p{=;.A??D7=.
P{w~vw |p}.x}sx
xsp{.xw.p.
wt~x} wx{t.wt.tt.x}.wt.|x{xp;.
sv pqt sx~st |p wp
t ts
p}s u~}s p {prz ~u r~t{px~} ~u p
|pxyp}p.p.~}t.~u.wtx.ux.x{{xrx.
rpp{.t{px~}wx.st|~}px}v.
sv;.wx.upr.s~t.}~.r~tr{.{tps.
~.wt.t
tt.x}utt}rt.wp.|~.
x}sx
xsp{.w~.ts.|pxyp}p.x{{.
x}wtt}{.v~.~}.~..~.qtr~|t.
tv{p.t.~u.~wt.x{{xrx.sv=.
btrxuxrp{{;.spp.u~|.wt.A?@@.]bSdW.
t.x{{pt.wx.xt.6bP\WbP;.
A?@A7=.]bSdW.spp tx|pt.@?F=G.
|x{{x~}.x}sx
xsp{.wp
t.p.{xutx|t.
wx~.~u.|pxyp}p.t;.wxrw.
x}sxrpt.t ~}.p.{tp.~}t.~rrpx~};.
r~|pts.~.p~x|pt{.BE.|x{{x~}.
x}sx
xsp{.wp
x}v p.{xutx|t.wx~.~u.
r~rpx}t.t p}s.p~x|pt{.C.
|x{{x~}.x}sx
xsp{.wp
x}v.p.{xutx|t.
wx~.~u.wt~x}.t=.]bSdW.spp.s~.
}~.~
xst x}u~|px~}.pq~.tprw.
x}sx
xsp{ .trxuxr.sv.wx~=.
W~t
t;.t
t}.xu.~}t.~x.wp.t
t.
r~rpx}t.p}s.wt~x}.t.t
x~{.ts.
|pxyp}p;.wt.]bSdW.spp.w~.wp.
|pxyp}p.t.p.{tp.~}rt.x}.p.{xutx|t.
s~t.}~.tsxr.wp.p}.x}sx
xsp{.x{{.
p{~.t p}~wt.x{{xrx.sv.p.{tp.
Ux}p{{;.p t
xt.~u.wt.vptp.
w~wtx q.ep}z~
t.p{=.6A?@A7.
}~t.wp.qtrpt.wt.vptp.
w~wtx ~}{ psstt.wt.~st.~u.
sv.t.x}xxpx~};.wt.vptp.
w~wtx s~t.}~.trxu.p}.
|trwp}xxr.r~}}trx~}.qttt}.sv.
pvt .u~{{~x}v.t~t.~.
|pxyp}p p}s s~t }~ tt}s ~ wt
xz.u~.pssxrx~}=.cwx.r~}rt.
r~}p.xw.wt.r~}rt.~u.p.r~||~}.
{xpqx{x.~.pssxrx~}.wp.x}
~{
t.
|trwp}x|.p}s.qx~qtwp
x~p{.
rwpprtxxr.tpx}x}v.~.wt.t}xt.
r~t ~u.sv.pqt.xz.p}s.sx~st=.
d}st.wt.t
t}w.upr~;.wt.
btrtp.|.r~}xst.|pxyp}p .
rwxr.~.wx~{~vxrp{.stt}st}rt.
{xpqx{x=.
_rwxr.~.rw~{~vxrp{.stt}st}rt.
wp.qtt} w~}.x}.t~}t.~.
|pxyp}p rw~prx
t tuutr=
_rw~prx
t t~}t.~.|pxyp}p.pt.
{tppq{t.~.|p}.w|p}.p}s.pt.
p~rxpts xw sv<ttzx}v p}s sv<
pzx}v.6\p{s~}ps~;.A??A7=.\~t~
t;.
wxvw.{t
t{ ~u.rw~prx
t.tuutr;.
}~pq{ ~xx
t tx}u~rt|t}; pt
p~rxpts.xw.x}rtpts.|pxyp}p.
t;.pqt; p}s.stt}st}rt.6brwtt.t.
p{=; A??HJ itxvt t p{=; A?@?7=
Txst|x~{~vxrp{.spp.~.wtt.
ux}sx}v.w~vw A?@A.]bSdW.pxxr.
wp w~ wp ~u x}sx
xsp{ tp @A ~
~{st.w~ ts.|pxyp}p.x}.wt.p.
|~}w;.p} tx|pts.C?=B.trt}.ts.
|pxyp}p ~} A? ~ |~t sp xwx}
wt.p.|~}w=.cwx.tpt.~.
p~x|pt{ F=E |x{{x~}.x}sx
xsp{.
pvts @A ~ ~{st w~ ts |pxyp}p ~}
p.spx{.~.p{|~.spx{.qpx=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64817!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
Pssxx~}p{{;.wt.A?@B.\cU.spp.t~.
wt.t
p{t}rt.~u.spx{.|pxyp}p.t;.
stux}ts.p.t.~}.A?.~.|~t.sp.
xwx}.wt.p.B?.sp;.x}.Gw;.@?w;.
p}s.@Aw.vpst.x.@=@.trt};.C=?.
trt};.p}s.E=D.trt};.ttrx
t{=.
c~{tp}rt.x.p.pt.~u.psppx~}.
wtt.t~t.~.p.sv.x}srt.
rwp}vt.wp.t{.x}.p.sx|x}x~}.~u.
~}t.~.|~t.~u.wt.sv .tuutr.~
t.
x|t.6P|txrp}.Prpst|.~u._px}.
\tsxrx}t;.P|txrp}._px}.b~rxt.p}s.
P|txrp}.b~rxt.~u.Pssxrx~}.
\tsxrx}t.r~}t}.s~r|t};.A??@7=.
c~{tp}rt.rp}.st
t{~.~.~|t;.q.}~.
p{{;.~u.|pxyp}p .tuutr=.btrxuxrp{{;.
~{tp}rt.s~t.}~.tt|.~.st
t{~.x}.
t~}t.~.|p}.~u.|pxyp}p .
rw~prx
t.tuutr=.cwx.{prz.~u.
~{tp}rt.|p.t{pt.~.
t{tr~wx~{~vxrp{.spp.st|~}px}v.
wp.rw~}xr.st{p <cWR.ps|x}xpx~}.
s~t.}~.puutr.x}rtpts.}t~}p{.uxx}v.
x}.wt.
t}p{.tv|t}p{.ptp;.p.tvx~}.
z}~}.~.{p.p.rxxrp{.~{t.x}.sv.
tx}u~rt|t}.p}s.tps.6f.p}s.
Ut}rw;.A???7=.X}.wt.pqt}rt.~u.~wt.
pqt.x}sxrp~;.rw.p.tpsx}v.
~txt;.wt.tt}rt.~u.~{tp}rt.~.
wxrp{.stt}st}rt.s~t.}~.
stt|x}t.wtwt.p.sv.wp.pqt.
~t}xp{=.
W~t
t;.w|p}.rp}.st
t{~.
~{tp}rt.~.|pxyp}p .rpsx~
pr{p;.
p~}~|xr;.p}s.qtwp
x~p{.tuutr.6Y~}t.
t.p{=;.@HG@7=.c~{tp}rt.~.~|t.~u.
|pxyp}p .qtwp
x~p{.tuutr.tt|.~.
st
t{~.put.wtp
.|pxyp}p.t;.q.
}~.put.~rrpx~}p{.|pxyp}p.t=.U~.
x}p}rt;.u~{{~x}v.prt.ps|x}xpx~}.
~u.|pxyp}p;.wtp
.|pxyp}p.t.sxs.
}~.twxqx.x|px|t}.x}.przx}v.p}s.
pt}x~}.pz;.p.tt.tt}.x}.
~rrpx~}p{.|pxyp}p.t.6ap|ptzt.
t.p{=;.A??H7=.Uwt|~t;.p.
}t~wx~{~vxrp{.pt|t}.
ps|x}xtts.w~vw.p}.
t{tr~t}rtwp{~vpw.6TTV7.wxrw.
|tpt.t
t}<t{pts.~t}xp{.6Ta_7.
r~}srts.x}.wt.p|t.qytr.p.wt.
t
x~.s;.u~}s.p.r~t~}sx}v.
tuutr.x}.wt._@?? r~|~}t}.~u.Ta_=.
btrxuxrp{{;.r~t~}sx}v.~.
tu~|p}rt.~}.przx}v.p}s.pt}x~}.
pz;.wtp
.|pxyp}p.t.w~ts.}~.
rwp}vt.x}._@??.p|{xst.u~{{~x}v.
prt.|pxyp}p.ps|x}xpx~};.
p{w~vw.~rrpx~}p{.t.w~ts.p.
strtpt.x}._@??.p|{xst.
6cwt}xt}.t.p{=;.A?@A7=.P.~xq{t.
|trwp}x|.}st{x}v.~{tp}rt.~.
|pxyp}p .tuutr.|p.qt.wt.s~}<.
tv{px~}.~u.rp}}pqx}~xs.trt~.
6Wx
~}t}.t.p{=;[email protected]~}p{t.t.p{=;.
cwt._@??.r~|~}t}.~u.Ta_.x.w~vw.~.
t{pt.~.wt.
xp{.~rtx}v.~u.x|{x.p}s.rp}.qt.
|~s{pts.q.pt}x~}=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
A??DJ.a~sxvt.st.U~}trp.t.p{=;.@HHCJ.
^
xts~.t.p{=;.@HHB7=.
X|~p}{;.wp|pr~{~vxrp{.
~{tp}rt.p{~}t.s~t.}~.x}sxrpt.p.
sv .wxrp{.stt}st}rt.{xpqx{x=.X}.
~st.u~.wxrp{.stt}st}rt.~.tx;.
t
xst}rt.~u.p.xwspp{.}s~|t.x.
}ttsts=._wxrp{.stt}st}rt.x.p.pt.
~u.psppx~};.|p}xutts.q.p.sv<.
r{p.trxuxr.xwspp{.}s~|t.
~srts.q.pq.rtpx~};.pxs.
s~t.tsrx~};.strtpx}v.q{~~s.{t
t{.
~u.wt.sv;.p}s>~.ps|x}xpx~}.~u.p}.
p}pv~}x.6kdkf7=.\p}.|tsxrpx~}.}~.
p~rxpts.xw.pqt.~.pssxrx~}.rp}.
~srt.wxrp{.stt}st}rt.p}s.
xwspp{.|~|.put.rw~}xr.t=.
Sxr~}x}px~}.~u.wtp
;.rw~}xr.
|pxyp}p.t.wp.qtt}.w~}.~.{tps.
~.wxrp{.stt}st}rt.p}s.xwspp{.
|~|.6P|txrp}._rwxpxr.
P~rxpx~}.Sb\¤e;[email protected]s}t.p}s.
Wvwt;.A??EJ.Wp}t.t.p{=;.@HHH7=.X}.
wtp
;.rw~}xr.|pxyp}p.t;.wt.
|~.r~||~}{.t~ts.xwspp{.
|~|.pt.{tt.sxuuxr{xt;.
strtpts.ptxt.~.txvw.{~;.
xxpqx{x;.p}vt;.p}xt.~.
}t
~}t;.p}s.t{t}t=.b~|t.
{t.r~||~}{.t~ts.xwspp{.
|~|.pt.sttts.|~~s;.
tpx}v;.wpzx}t;.wxrp{.
sxr~|u~;.p}s.rwx{{.6Qs}t.p}s.
Wvwt;.A??EJ.Wp}t.t.p{=;.@HHH7=.cwt.
~rrt}rt.~u.|pxyp}p.xwspp{.
|~|.x}.{xvw.~.}~}<spx{.
|pxyp}p.t.wp.}~.qtt}.
tpq{xwts=.cwt.P|txrp}._rwxpxr.
P~rxpx~} .Sb\¤[email protected]}r{st.p.
{x.~u.|~|.~u.rp}}pqx.
xwspp{= .\~.|pxyp}p.
xwspp{.|~|.qtvx}.xwx}.AC¤.
CG.w~.~u.sxr~}x}px~};.tpz.
xwx}.C¤E.sp;.p}s.{p.u~.@¤B.ttz=.
\pxyp}p.xwspp{.}s~|t.wp.
qtt}.t~ts.x}.ps~{trt}.p}s.ps{.
ps|xts.u~.qp}rt.pqt.tp|t}=.
Qpts.~}.r{x}xrp{.strxx~};.wx.
}s~|t.ptp.~.qt.|x{s.r~|pts.
~.r{pxrp{.p{r~w~{.p}s.qpqxpt.
xwspp{.}s~|t;.wxrw.rp}.
x}r{st.|~t.tx~.|~|.rw.p.
pvxpx~};.pp}~xp;.p}s.txt=.
\{x{t.sxt.r~|px}v.|pxyp}p.
p}s.~qprr~.xwspp{.|~|.x}.
w|p}.st|~}pt.wp.wt.|pv}xst.
p}s.x|t.r~t.~u.wt.~.xwspp{.
}s~|t.pt.x|x{p.6Qs}t.t.p{=;.
A??GJ.ep}st.t.p{=;.A??D;.A??G7=.
9/!Yjgvjgt!vjg!Uwduvcpeg!Ku!cp!
Koogfkcvg!Rtgewtuqt!qh!c!Uwduvcpeg!
Cntgcf{!Eqpvtqnngf!Wpfgt!Vjku!Ctvkeng!
d}st.wt.txvw.upr~.p}p{x;.wt.
btrtp.|.r~}xst.wtwt.
|pxyp}p.x.p}.x||tsxpt.tr~.~u.
p.r~}~{{ts.qp}rt=.\pxyp}p.x.}~.
p}.x||tsxpt.tr~.~u.p}~wt.
r~}~{{ts.qp}rt=.
RQ!11111
Hto!11131
Hov!5812
Uhov!5813
Tgeqoogpfcvkqp!
Put.r~}xstpx~}.~u.wt.txvw.upr~.
sxrts.pq~
t;.USP.tr~||t}s.wp.
|pxyp}p.t|px}.x}.brwts{t.X.~u.wt.
RbP=.]XSP.r~}r.xw.wx.
rwts{x}v.tr~||t}spx~}=.\pxyp}p.
|tt.wt.wtt.rxtxp.u~.{prx}v.p.
qp}rt.x}.brwts{t.X.~u.wt.RbP.
}st[email protected]=b=R=.G@A6q76{7I.
6@7.\pxyp}p.wp.p.wxvw.~t}xp{.u~.
pqtI.
P.}|qt.~u.upr~.x}sxrpt.
|pxyp}p .wxvw.pqt.~t}xp{;.
x}r{sx}v.wt.{pvt.}|qt.~u.
x}sx
xsp{.tv{p{.x}v.|pxyp}p;.
|pxyp}p .xsttps.t;.p}s.wt.
p.p|~}.~u.|pxyp}p.p
px{pq{t.u~.
x{{xrx.t=.P~x|pt{.@G=H.|x{{x~}.
x}sx
xsp{.x}.wt.d}xts.bpt.6F=B.
trt}.~u.wt.d=b=.~{px~}7.ts.
|pxyp}p.|~}w{.x}.A?@A=.
Pssxx~}p{{;.p~x|pt{.C=B.|x{{x~}.
x}sx
xsp{.|t.sxpv}~xr.rxtxp.u~.
|pxyp}p.stt}st}rt.~.pqt.x}.wt.
tp.x~.~.wt.A?@A.]bSdW.
t=.
P.A?@B.
t.x}sxrpt.wp.q.@Aw.
vpst;.BE=C.trt}.~u.st}.t~.
x}v.|pxyp}p.xwx}.wt.p.tp;.
p}s.AA=F.trt}.t~.x}v.|pxyp}p.
|~}w{=.X}.A?@@;.CDD;EEG.TS.
xx.
tt.|pxyp}p<t{pts;.ttt}x}v.
BE=C.trt}.~u.p{{.x{{xrx.sv<t{pts.
tx~st=._x|p.|pxyp}p.t.
prr~}ts.u~.@G=@.trt}.~u.
ps|xx~}.~.sv.tp|t}.~vp|.
x}.A?@@=.Pssxx~}p{{;.|pxyp}p.wp.
s~t<stt}st}.tx}u~rx}v.tuutr;.p.
st|~}pts.q.spp.w~x}v.wp.
w|p}.tut.t{px
t{.wxvwt.s~t.~.
{~t.s~t=.Uwt|~t;.|pxyp}p.
t.rp}.t{.x}.rw~{~vxrp{.
st•t}st}rt=.
6A7.\pxyp}p.wp.}~.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bptI.
USP.wp.}~.p~
ts.p.|pztx}v.
p{xrpx~}.u~.p.|pxyp}p.sv.
~sr.u~.p}.x}sxrpx~}=.cwt.
~~}x.u~.rxt}x.~.r~}sr.
r{x}xrp{.ttprw.xw.|pxyp}p.tx;.
p}s.wtt.pt.prx
t.X]S.u~.|pxyp}pJ.
w~t
t;.|pxyp}p.s~t.}~.wp
t.p.
rt}{.prrtts.|tsxrp{.t.u~.
tp|t}.x}.wt.d}xts.bpt;.}~.s~t.
|pxyp}p.wp
t.p}.prrtts.|tsxrp{.t.
xw.t
tt.txrx~}=.
P.sv.wp.p.rt}{.prrtts.
|tsxrp{.t .xu.p{{.~u.wt.u~{{~x}v.ux
t.
t{t|t}.wp
t.qtt}.pxuxtsI.
p=.wt.sv .rwt|x.x.z}~}.p}s.
t~srxq{tJ.
q=.wtt.pt.pstpt.put.sxtJ.
r=.wtt.pt.pstpt.p}s.t{{<.
r~}~{{ts.sxt.~
x}v.tuuxrprJ.
s=.wt.sv.x.prrtts.q.p{xuxts.
ttJ.p}s.
t=.wt.rxt}xuxr.t
xst}rt.x.xst{.
p
px{pq{t=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
jDF.Ua.@?CHH;.\p•rw.AE;.@HHAl.
\pxyp}p.s~t.}~.|tt.p}.~u.wt.
t{t|t}.u~.wp
x}v.p.rt}{.
prrtts.|tsxrp{.t= .
Ux;.USP.q~ps{.t
p{pts.
|pxyp}p;.p}s.sxs.}~.u~r.x.
t
p{px~}.~}.pxr{p.px}.~u.
|pxyp}p.~.r~|~}t}.~.stx
px
t.
~u.|pxyp}p=.bx}rt.sxuutt}.px}.|p.
wp
t.sxuutt}.rwt|xrp{.r~}xt};.
|pxyp}p;.p.xst}xuxts.x}.wx.txx~};.
s~t.}~.wp
t.p.z}~}.p}s.
t~srxq{t.rwt|x;.wxrw.~{s.
qt.}ttsts.~.~
xst.p}spsxts.
s~t=.
btr~}s;.wtt.pt.}~.pstpt.put.
sxt.~}.|pxyp}p.x}.wt.|tsxrp{.
{xtpt.x}.t{px~}.~.p.trxuxr;.
tr~v}xts.sx~st=.cwxs;.wtt.pt.}~.
q{xwts.pstpt.p}s.t{{.r~}~{{ts.
sxt.~
x}v.tuuxrpr.~u.|pxyp}p=.
U~w;.wtt.x.}~.t
xst}rt.wp.
p{xuxts.tt.prrt.|pxyp}p.u~.
t.x}.tpx}v.p.trxuxr;.tr~v}xts.
sx~st=.[p{;.wt.rxt}xuxr.t
xst}rt.
tvpsx}v.|pxyp}p .rwt|x.x}.
t|.~u.p.trxuxr.Ecppcdku!px}.wp.
r~{s.~srt.p}spsxts.p}s.
t~srxq{t.s~t.x.}~.rt}{.
p
px{pq{t;.~.wt.rxt}xuxr.t
xst}rt.~}.
|pxyp}p.x.}~.xst{.p
px{pq{t=.
P{t}pt{;.p.brwts{t.XX.sv.rp}.qt.
r~}xstts.~.wp
t.p.rt}{.
prrtts.|tsxrp{.t.xw.t
tt.
txrx~} [email protected]=b=R=.G@A6q76A76Q77=.
ht.p.pts.pq~
t;.wt.{prz.~u.prrtts.
|tsxrp{.t.u~.p.trxuxr;.tr~v}xts.
sx~st.tr{st.wt.t.~u.|pxyp}p.
t
t}.}st.r~}sxx~}.wtt.x.t.x.
t
tt{.txrts=.
X}.r~}r{x~};.~.spt;.ttprw.~}.
|pxyp}p .|tsxrp{.t.wp.}~.
st
t{~ts.~.wt.~x}.wtt.|pxyp}p.
x.r~}xstts.~.wp
t.p.rt}{.
prrtts.|tsxrp{.t .~.p.rt}{.
prrtts.|tsxrp{.t.xw.t
tt.
txrx~}= .
6B7.cwtt.x.p.{prz.~u.prrtts.put.
u~.t.~u.|pxyp}p.}st.|tsxrp{.
t
xx~}I.
cwtt.pt.rt}{.}~.USP<p~
ts.
|pxyp}p.sv.~sr=.\pxyp}p.
s~t.}~.wp
t.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt.~.p.rt}{.prrtts.|tsxrp{.
t.xw.t
tt.txrx~}=.cw;.USP.
wp.}~.stt|x}ts.wp.|pxyp}p.x.
put.u~.t.}st.|tsxrp{.t
xx~}=.
X}.pssxx~};.USP.rp}}~.r~}r{st.
wp.|pxyp}p.wp.p}.prrtpq{t.{t
t{.~u.
put.t{px
t.~.x.tuutrx
t}t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st.
xw~.t
xst}rt.wp.wt.qp}rt.x.
r~}p|x}px~}.utt;.p}s.pp}rt.~u.p.
r~}xt}.p}s.tsxrpq{t.s~t=.
X}
txvpx~}.x}~.wt.|tsxrp{.t.~u.
|pxyp}p.w~{s.x}r{st.x}u~|px~}.
p}s.spp.tvpsx}v.wt.rwt|x;.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
64818!
Qp{t.a[.p}s.Qxvt{~.VT=.Vxst{x}t.p}s.
|p}uprx}v;.p}s.trxuxrpx~}.~u.
|tw~s~{~vxrp{.t
xt.r~}rt}x}v.sv.
|pxyp}p=.Pssxx~}p{{;.p.~rtst.u~.
pqt.{xpqx{x.pt|t}=.Sv.p}s.
st{x
tx}v.p.r~}xt}.s~t.~u.
P{r~w~{.St•t}st}rt=.A??BJ.F?I.b@B¤bC?=.
|pxyp}p.w~{s.p{~.qt.st
t{~ts=.
Qp}t;.V=J.[xrz~;.e=J.p}s.cw~|~};.c=.
cwttu~t;.USP.r~}r{st.|pxyp}p.
Qtwp
x~p{.wp|pr~zx}txr.~u.
s~t.}~.rt}{.wp
t.p}.prrtts.
|pxyp}p=._rw~wp|pr~{~v.@HGD;.
{t
t{.~u.put.u~.t.}st.|tsxrp{.
GD6@7;.D@¤DE=.
Qpxp.];.Sx.c\;.Qpx.\;.\prrp~}t.\=.
t
xx~}=.
cwt.t}s~rp}}pqx}~xs.t|I.p}.
~
t
xt=.U~}=Qtwp
=]t~rx=.A?@AJ.
EIH=.
Pqp|.SX;.Wx{~}.YU;.[txt.aY;.bwpst.bQ;.
Qt}~x.][;.Y~}t.ac=.Rpsx~
pr{p.
T{qtxz.cP;.Pttzp.Uc;.Qt}~x.][;.
tuutr.~u.~{~}vts.st{p<H<.
Qts.Q\;.Z~t{.Q;.Pqtv.YP;.Sttz.bV;.
tpws~rp}}pqx}~{.x}vtx~}=.R{x}.
\xrwt{{.cU;.\{{xvp}.Z;.Qprrwtx._;.
_wp•|pr~{.cwt•[email protected]•J.@G6B7IAGF¤HF=.
\rR}t.Y\;.brwp|qt{p}.\=.bw~<t|.
Qt}~x.][;.Y~}t.ac=.Rpsx~
pr{p.p}s.
tuutr.~u.rp}}pqx}~xs.x}.pxt}.xw.
|tpq~{xr.r~}xstpx~}.x}.~{~}vts.
WXe¤@.x}utrx~}I.p.p}s~|xts;.{prtq~.
rp}}pqx}~xs.ps|x}xpx~}.x}.|p}=.Y.
r~}~{{ts.r{x}xrp{.xp{=.P}}.X}t}.\ts=.
R{x}._wp|pr~{=.@[email protected]v<btJ.A@6G¤H.
A??B.Pv.@HJ.@BH6C7IADG¤EE=.
b{7IA@Cb¤AABb=.
Pqp|.SX;.Yp.RP;.bwpst.bQ;.ex~~.W;.
Q{~rz.aX;.Upx}~.a;.brw{trwt.YP=.Tuutr.
atsp.W;._t.b;.Zt{{.\T;.a~q~wp|.
~u.rw~}xr.|pxyp}p.t.~}.t~t~}t;.
\R;.p}s._ttt}.Z[=.A??F=.Rp}}pqx.x}.
{tx}xx}v.w~|~}t;.u~{{xr{t.x|{px}v.
px}u{.WXe<p~rxpts.t}~.
w~|~}t;.~{prx}.p}s.r~x~{.x}.|t}.
}t~pwI.p.p}s~|xts.{prtq~<.
p}s.~|t}=.Sv.P{r~w~{.Stt}s=.
r~}~{{ts.xp{=.Pgwtqnqi{!EG6F7I.D@D¤.
@HH@PvJ.AG6A7I@A@¤G=.
DA@=.
Q{~rz.aX;.Upx}~.a;.Qp
t|p}.Z=.Prt.
Psp|;.[Q=;.p}s.\px};.Q=a=.Rp}}pqxI.
tuutr.~u.|pxyp}p.~}.r~v}xx~}I.
_wp|pr~{~v.p}s.~xr~{~v.x}.p}x|p{.
t{px~}wx.~.rw~}xr.tuutr.p}s.
p}s.w|p}=.Pssxrx~}.@HHE;.
|~zx}v.trw}xt=._wp|pr~{.Qx~rwt|.
H@6@@7I@DGD¤@E@C=.
Qtwp
=.@HHA.]~
J.CB6B7IH?F¤@F=.
Pvt{{;.b=;.Stt;.f=[=;.p}s.fx{{t;.a=T=;.
Q~{{p.ZX;.Q~}.Z;.T{stw.S;.cpt.Z;.p}s.
ts=.cwt.Rp}}pqx}~xsI.Rwt|xrp{;.
Rpst.Y[=.S~t<t{pts.}t~r~v}xx
t.
_wp|pr~{~vxr;.p}s.cwtptxr.
tuutr.~u.|pxyp}p.t=.]t~{~v.A??A.
Ptr=.]t.h~zI.Prpst|xr._t;.
DHI@BBF¤@BCB=.
@HGC=.
Q~{{p.ZX;.T{stw.SP;.\p~rwxz.YP;.p}s.Rpst.
Pvt{{;.b=J.Wp{{sx};.\=J.[x}svt};.Y=T=J.
Y[=.]tp{.qpt.~u.up{.strxx~}<.
^w{~};.P=J.fxs|p};.\=J.Vx{{txt;.W=J.
|pzx}v.x}.pqx}t}.|pxyp}p.t=.
p}s.W~{{xt;.[=._wp|pr~zx}txr.p}s.
]t~X|pvt.A??D.AEICG?¤CHA=.
|tpq~{x|.~u.st{p.@<.
Q~}}t.d=.Rw~}xr.rp}}pqx.pqt;.st{p<H<.
tpws~rp}}pqx}~{.p}s.~wt.
tpws~rp}}pqx}~{.p}s.w~xs.
rp}}pqx}~xs.xw.t|wpx.~}.|p}=.
u}rx~}=._wp|pr~rwxp[email protected]}J.
_wp|pr~{.at
.@HGE;.BG6@7;.A@¤CB=.
CE6@7IBD¤E=.
P{|xt.aV;.b|xw.RV;.Prw.aW=.cwt.tuutr.
Q~pq~{p.\;.ax}p{sx.\;.Rpp~}._;.Rpx{{~}.
~u.|pxyp}p.tpr.p}s.st{p.H<.
R;.St{trw.Q;.bwxt.S;.[t.U.V;.Rpt{{p.
tpws~rp}}pqx}~{.~}.{tp{.u}rx~}.
_=.Rp}}pqx}~xs<trt~.ttx~}.x}.
x}.wt.wt.|~}zt=.Utx{.btx{=.@HGB.
w|p}.{tz~rt=.T.Y.Qx~rwt|=.@HHB.
UtqJ.BH6A7IA@A¤F=.
\p[email protected]@C6@7I@FB¤G?=.
P|tx;.P=.cwt.tuutr.~u.rp}}pqx}~xs.~}.wt.
qpx}=._~vt.x}.]t~qx~{~v.@HHH;.DG. Qpxsp.S;.X~t.b;._tv~x}x.b;.bp{p.\=.St{pH<.
tpws~rp}}pqx}~{<x}srts.
6C7;.B@D¤BCG=.
r~}sxx~}ts.{prt.tutt}rt.p}s.
P|txrp}.Prpst|.~u._px}.\tsxrx}t;.
x}prttq~
t}xr{p.t{u<.
P|txrp}._px}.b~rxt.p}s.P|txrp}.
ps|x}xpx~}.x}.p=.T.Y._wp|pr~{=.
b~rxt.~u.Pssxrx~}.\tsxrx}t.
A??C.Str.BJ.D?E6@7IEB¤H=.
R~}t}.S~r|t}=.Stux}xx~}.t{pts.
~.wt.t.~u.~x~xs.u~.wt.tp|t}.~u. Qtx
~vt{.Rb;.Rwx{st.ba=.Rp}}pqx}~xs.
pv~}x.xv}p{.p}srx~}.x}.p.qpx}I.
•px}=.A??@=.
r~|px~}.~u.rp}}pqx}~xs.pv~}x.x}.
P}stp~}.b;.P{{tqtrz._;.T}v~|.P;.
trt~.qx}sx}v;.V<~tx}.prx
px~};.
asqtv.d=.Rp}}pqx.p}s.rwx~wt}xp=.
p}s.pst}{{.rr{pt.x}wxqxx~}=.Y.
P.{~}vxsx}p{.s.~u.btsxw.
_wp|pr~{.T.cwt=.A???=.^rJ.
r~}rx=.[p}rt[email protected].
AHD6@7IBAG¤BE=.
A6GDFC7I@CGB¤E=.
Qtx
~vt{.Rb;.Vxuux}.V;.Sx.\p~.e;.\px}.
Pt}sx}~.V;.Rwxp}tt.V;.cpv{xp{pt{p<.
Qa=.T
xst}rt.u~.p.}t.V.~tx}<.
brpupx.^=.Rp}}pqx}~xsI.~rrt}rt.p}s.
r~{ts.rp}}pqx}~xs.trt~.x}.|~t.
|tsxrx}p{.rwt|x=.R.\ts.Rwt|=.
qpx}=.\~{._wp|pr~{[email protected]{J.
A?@@J.@G6F7I@?GD¤HH=.
E?6@7I@DD¤EB=.
Prw.aW;.b|xw.RV;.bx{t<Zw~s.c\;.
Q~}.cc;.S~q.Pb=.T}s~rx}t.tuutr.~u.
_pttx}.RY=.Tuutr.~u.st{p.H<.
|pxyp}p=.Y.R{x}._wp|pr~{=.A??A.]~
J.
tpws~rp}}pqx}~{.sx}v.wt.
CA6@@.b{7IH?b¤HEb=.
u~{{xr{p.wpt.~u.wt.wt.|~}zt.
Q~}t.aV;.ftx|p}.P=.Sxrx|x}px
t.
6\prprp.|{pp7=.Y.R{x}.T}s~rx}~{.
x|{.~txt.~u.st{p.H<.
\tpq=.@[email protected]}J.DA6@7ID?¤D=.
tpws~rp}}pqx}~{I.|trwp}xxr.
Qp{t;.a=[=;._tr~;.f=a=;.st{p <.
sxt=.Y.R{x}._wp|pr~{=.@[email protected]v<btJ.
ctpws~rp}}pqx}~{.sxrx|x}px~}.x}.
A@6G¤H.b{7IAAFb¤ABCb=.
p.p.p.|~st{.u~.rp}}pqx.x}~xrpx~}=.
Qs}t.PY;.Wvwt.Ya;.\~~t.QP;.ep}st.
]t~rx=.4.Qx~qtwp
=.at
=.@HHA;.@E6@7;.
a=.at
xt.~u.wt.
p{xsx.p}s.
DD¤EA=.
Tghgtgpegu!
RQ!11111
Hto!11132
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64819!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
xv}xuxrp}rt.~u.rp}}pqx.xwspp{.
}s~|t=.P|.Y._rwxp=.A??C.]~
J.
@E@6@@7I@HEF¤FF=.
Qs}t.PY;.Wvwt.Ya=.cwt.rp}}pqx.
xwspp{.}s~|t=.R.^x}.
_rwxp.A??E.\pJ.@H6B7IABB¤G=.
Qs}t.PY;.ep}st.aV;.Wvwt.Ya;.
cw~t}~}.YS;.Qpt.i=.R~|px~}.~u.
rp}}pqx.p}s.~qprr~.xwspp{I.
t
tx.p}s.r~}xqx~}.~.t{pt=.Y.
bq=Pqt.ctp=.A??G.StrJ.BD6C7IBEA¤.
G=.
Rpx~x.a\;.U~{x}.af;.Qps.Ye;.Uxrw|p}.
\f=.Tuutr.~u.|pxyp}p.~}.wt.pz<.
t{xrxts.wx~{~vxrp{.t~}t=.Sv.
P{r~w~{.Stt}[email protected]{J.A@6B7I@GB¤F=.
Rprx}x.U;.Pxt{{~.R;.Sx.cp}}p.V=.X}rtpx}v.
st{p<H<tpws~rp}}pqx}~{.6%<H¤cWR7.
r~}t}.x}.wtqp{.rp}}pqx.~
t.x|tI.
t|pxr.t
xt.p}s.|tp<p}p{x=.
R.Sv.Pqt.at
=.A?@A.\pJ.D6@7IBA¤.
C?=.
Rwpx.[S=.bqytrx
t.p}s.qtwp
x~p{.tuutr.~u.
|pxyp}p.wt.|~}x}v.put.|~zx}v=.
_rw~wp|pr~{~v.6Qt{=7.@HH?J.
@??6B7IBAG¤BB=.
Rwpx.[S;.Qzt.ZP=._tutt}rt.u~.wxvw<.
t.{~<~t}r.|pxyp}p=.
_wp|pr~{.Qx~rwt|.Qtwp
=.@HHC.]~
J.
CH6B7IECB¤F=.
Rwpt~}.U;.b~qxt._;._trw.PY;.cwxtq~.
\W=.X}
~{
t|t}.~u.rt}p{.rp}}pqx}~xs.
6RQX7.trt~.x}.wt.tpq{xw|t}.~u.
{prt.r~}sxx~}x}v.x}.p=.
_rw~wp|pr~{~v.6Qt{[email protected].
@BD6C7IBAC¤BA=.
Rwtt.YU;.Zt}sp{{.SP;.\pst}.RP=.
Rp}}pqx}~xs.trt~.p}s.tps.x}.wt.
pI.p.r~}sxx~}ts.{prt.tutt}rt.
s=._rw~wp|pr~{~v.6Qt{7=.A???.
Y{J.@D@6@7IAD¤B?=.
R~}t.TY;.Y~w}~}.aT;.\~~t.YS;.a~prwt.YS=.
Prt.tuutr.~u.|~zx}v.|pxyp}p.~}.
w~|~}t;.qytrx
t.tuutr.p}s.
tu~|p}rt.x}.|p{t.w|p}.qytr=.
_wp|pr~{.Qx~rwt|.Qtwp
[email protected]}J.
AC6E7I.@FCH¤DC=.
R~t<Q{~~|.Y;.f~{u~}.c;.Vp|.P;.Yx}.b;.
\pr~t.cS;.Qt}{t.W;.p}s.V~p.Q=.
A?@A=.b|~zts.rp}}pqx.u~.pxrx.x}.
|{x{t.r{t~xI.p.p}s~|xts;.
{prtq~<r~}~{{ts.xp{[email protected]@?7I.
@@CB¤@@D?=.
R~}rx{.~}.brxt}rt.p}s._q{xr.Wtp{w.at~.
B=.dt.~u.rp}}pqx.u~.|tsxrx}p{.
~t=.P|txrp}.\tsxrp{.
P~rxpx~};.X}tx|.\ttx}v;.W~~};.
ctpJ.]~
t|qt.A??H=.
Rpu~s.fY;.p}s.\tx.YR=.@HFH=.Tuutr.~u.
tpws~rp}}pqx}~{.~}.ptxp{.p}s.
x}p~r{p.wtt}x~}=.Kpvgtpcvkqpcn!
lqwtpcn!qh!enkpkecn!rjctoceqnqi{!cpf!
dkqrjctoce{!{F6D7I.@H@¤@HE=.
R~u~s.Y[;.h.p|p|p.c=.Rp}}pqx}~xs.p}s.
wt.x||}t.t|I.~t}xp{.u~.wt.
tp|t}.~u.x}u{p||p~.sxtptN.Y.
]t~x||}~{=.A??D.btJ.@EE6@¤A7IB¤.
@G=.at
xt=.
Sp{~}.fb;.\p.a;.[t|qtvt.[;.a~ssp.QT;.
U~}t.aQ=.X}u{t}rt.~u.rp}}pqxsx~{.~}.
st{p<H<tpws~rp}}pqx}~{.tuutr=.
R{x}=_wp•|pr~{=.cwt•=.@HFE.\p•J.
@H6B7IB??¤H=.
Sp.T\;._x{~t.]b;.Ps{t.fW;.]pvt{.YT;.
[p}vt.fa=.cwt.tuutr.~u.H<t}t<.
tpws~rp}}pqx}~{.~}.w~|~}t.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
t{tpt.p}s.x||}t.u}rx~}=.Y.bt~xs.
Qx~rwt|[email protected]@¤E7IAEB¤F?=.
Stvt}wps.[;.Wp{{.f;.[}zt.\=.ctx}v.
w~wtt.pq~.wt.t{px~}wx.
qttt}.rp}}pqx.t.p}s.rw~x=.
Sv.P{r~w~{.Stt}s=.A??B.Y{.A?J.
F@6@7IBF¤CG=.
Stvt}wps.[;.Rwx.fc;.R~}p.Z;.Sxtzt.
[;.V{p}.\;.Zp{psyxp}.P;.\txzp}vp.
Z;.bp|~}.];.bt}st}.Y;.Zt{t.aR=.
S~t.wt. vptp .|ptN.P~rxpx~}.
qttt}.wt.~st.~u.sv.t.x}xxpx~}.
p}s.wt.st
t{~|t}.~u.sv.
stt}st}rt.x}.wt.]px~}p{.R~|~qxsx.
bs.at{xrpx~}=._rw~{=\ts.A??H.
Yp}J.BH6@7I@DF¤EF=.
Stvt}wps.[;.Sxtzt.[;.Rwx.fc;.\tsx}p<.
\~p.\T;.]t|pz.h;.bp|~}.];.
P{~}~.Y;.P}vt|tt.\;.P}w~}.YR;.
Quupt.a;.t.p{=.T
p{px}v.wt.sv.
t.vptp .wt~.x}v.r~<.
}px~}p{.sppI.r~}xt}r.p}s.
p~rxpx~}.~u.wt.~st.~u.x}xxpx~}.~u.
sv.t.p|~}v.pxrxp}.x}.wt.fW^.
f~{s.\t}p{.Wtp{w.b
t=.Sv.
P{r~w~{=St•t}s=.A?@?.P••.@J.@?G6@¤.
A7IGC¤HF=.
Stxp}p.b;.Up~t.[;.bp}~.\b;.R~.V;.
_~r.T;.Upssp._;.Upp.f=.bpx}.p}s.
rwts{t.stt}st}.sxuutt}rt.x}.wt.
prxxx~};.|px}t}p}rt.p}s.tx}rx~}.
~u.x}p
t}~.rp}}pqx}~xs.t{u<.
ps|x}xpx~}.x}.p=.
]t~wp|pr~{~v=.A??F.UtqJ.
DA6A7IECE¤DC=.
Stt;.f=.[=;.\px};.Q=.a=;.\p;.T=.[=.
Rp}}pqx}~xs.tt~x~|tI.
wp|pr~{~vxrp{.tuutr=.X}.b|xw;.S=.U=.
6Ts=7.RaR.Wp}sq~~z.~u.tt~x~|tI.
sv.x}.rw~wp|pr~{~v;.B@F¤BAE.
6Q~rp.ap~};.U[;.RaR._t7;@HGC=.
Sx.\p~;.e=.P.qxtu.wx~.~u.rp}}pqx}~xs.
p}s.t}s~rp}}pqx}~xs.wp|pr~{~v.p.
x}xts.q.wt.~z.~u.Qxxw.rxt}x=.
ct}s=_wp|pr~{=brx.A??E.\pJ.
AF6B7I@BC¤C?=.
STP.bpxxr.p}s.Upr=.6}=s=7.STP.
S~|txr.Sv.btxt=.jvvr<00
yyy/lwuvkeg/iqx0fgc0tguqwteg.egpvgt0
uvcvkuvkeu/ujvon0!6prrtts.Pv.D;.
A?@C7.
Sv.T}u~rt|t}.Ps|x}xpx~};.Ftwiu!qh!
Cdwug-!A??D=.
Sv.T}u~rt|t}.Ps|x}xpx~}=.
Uqwtegdqqm!qh!Etkokpcn!Lwuvkeg!Uvcvkuvkeu-!
A??B=.
Sb\¤e=.Sxpv}~xr.p}s.bpxxrp{.\p}p{.~u.
\t}p{.Sx~st;.Uxuw.Tsxx~}=.
P|txrp}._rwxpxr.P~rxpx~}=.
fpwx}v~};.SRI.P|txrp}._rwxpxr.
_q{xwx}v;.A?@B=.
T{sxsvt.YR;.\w.[[;.[p}suxt{s._f=.
Rp}}pqx}~xs.p}s.wt.wx~rp|p{.
v{r~r~xr~xs.trt~I.trt}.ux}sx}v.
p}s.~xq{t.xv}xuxrp}rt=.bt~xs=.@HH@.
\pJ.DE6D7IAAE¤B@=.at
xt=.
T{{x.aY;.c~t~uu.f;.epxsp.U;.ep}.St}.
Qp}st.V;.V~}p{t.Y;.V~p.Q;.Qt}{t.
W;.p}s.Pzx}~}.YW=.A??H=.b|~zts.
|tsxrx}p{.rp}}pqx.u~.}t~pwxr.px}.
x}.WXeI.p.p}s~|xts;.r~~
t.r{x}xrp{.
xp{=.Pgwtqru{ejqrjctoceqnqi{<!qhhkekcn!
rwdnkecvkqp!qh!vjg!Cogtkecp!Eqnngig!qh!
Pgwtqru{ejqrjctoceqnqi{!BC6B7I.EFA¤.
EG?=.
T{b~w{.\P;.b{pst.S=.Rwt|xrp{.r~}xt}.
~u.|pxyp}pI.cwt.r~|{t.|xt.~u.
RQ!11111
Hto!11133
Hov!5812
Uhov!5813
}pp{.rp}}pqx}~xs=.[xut.brxt}rt=.
A??DJFGIDBH¤CG=.
Up}.ae;.Wtxw|p}.bY;.Q}zt.TQ;._xrz~w.
fQ=.Prt.p}s.txsp{.tuutr.~u.
|pxyp}p.x}.w|p}=._wp|pr~{.
Qx~rwt|.Qtwp
[email protected]vJ.E?6C7IFFF¤GC=.
Up~t.[;.bp}~.\b;.P{tp.b;.P}vx.U;.
Upssp._;.Upp.f=.Rp}}pqx}~xs.t{u<.
ps|x}xpx~}.x}.pI.t.sxuutt}rt.
p}s.wt.x}u{t}rt.~u.~
pxp}.u}rx~}=.
Q=Y._wp|pr~{=.A??F.]~
J.@DA6D7IFHD¤.
G?C=.
Utv~}.S\;.W~~~s.[Y;.axsst.T\=ct.
~u.rpp{.{x}zpvt.qttt}.rp}}pqx.t.
p}s.rw~xr.|~|=.Pssxrx~}=.
A??D.\p•J.@??6B7IBDC¤EE=.
U~}t.\P;.Q~{{p.ZX;.R}wp._Y;.P{|txsp.__;.
Y}vt|p}.U;.[pp}ytxp.aaJ.Qtp}.aP;.
[prtsp.P[=.Rp}}pqx.t.qtu~t.pvt.@D.
p}s.qtt}.ttrx
t.u}rx~}x}v=.
Q=Y._rwxp.A?@@.Y}J.@HG6E7ICCA¤F=.
Uxts;._=.P=;.fpzx}~};.Q=.BE<.p}s.CG<|~}w.
}t~qtwpqx~p{.u~{{~<.~u.rwx{st}.
t}pp{{.t~ts.~.|pxyp}p;.
rxvptt.p}s.p{r~w~{=.Y=.St
=.Qtwp
=.
_tsxp=.@HGF;.G;.B@G¤BAE=.
Uxts;._=.P=;.fpzx}~};.Q=;.Vp;.a=.P.
u~{{~<.s.~u.pt}x~}p{.qtwp
x~.
x}.E<tp<~{s.rwx{st}.t~ts.
t}pp{{.~.|pxwp}p;.rxvptt.p}s.
p{r~w~{=.]t~~xr~{=.ctp~{=.@HHA;.@C;.
AHH¤B@@=.
Uxts;._=.P=;.fpzx}~};.Q=;.Vp;.a=.
Sxuutt}xp{.tuutr.~}.r~v}xx
t.
u}rx~}x}v.x}.H<.~.@A<.tp.~{s.
t}pp{{.t~ts.~.rxvptt.p}s.
|pxwp}p=.]t~~xr~{=.ctp~{=.@HHG;.
A?6B7;.AHB¤B?E=.
Uxts._P=.Ps~{trt}.t}pp{{.t~ts.~.
|pxyp}pI.tp|x}px~}.~u.uprt.~u.
r~|{t.qtwp
x~.p}s.r~|px~}.
xw.wt.x}u{t}rt.~u.x}.t~.rxvptt=.
Y=R{x}=_wp|pr~{=.A??A.]~
J.CA6@@.
b{7IHFb¤@?Ab=.
Uxts._P;.fpzx}~}.Q;.Vp.a=.
]t~r~v}xx
t.r~}tt}rt.~u.
|pxwp}p¥p.r~|px~}.xw.t<sv.
tu~|p}rt=.]t~~xr~{=.ctp~{=.A??D.
\p•J.AF6A7IAB@¤H=.
U}v;.\=;.Vp{{pvwt;.R=;.\prwp;.\=.[}v.
p}s.pt~<sxvtx
t.rp}rt.x}.~}v.
|pxyp}p.|~zt=.c|~x.@HHH;.GD.6A7;.
@C?¤@CA=.
Vp{xtvt.b;.\p.b;.\prwp}s.Y;.S~~.
S;.Rpxtt.S;.Rpp~}._;.Q~pq~{p.\;.
bwxt.S;.[t.U.V;.Rpt{{p._=.Ttx~}.
~u.rt}p{.p}s.txwtp{.rp}}pqx}~xs.
trt~.x}.w|p}.x||}t.xt.p}s.
{tz~rt.q~{px~}=.T.Y.
Qx~rwt|[email protected][email protected]@7IDC¤E@=.
Vp~}x;.h=;.\trw~{p|;.a=.X~{px~};.rt;.
p}s.pxp{.}wtx.~u.p}.prx
t.
r~}xt}.~u.wpwxw=.Y=.P|=.Rwt|=.b~r=.
@HEC;.GE;.@ECE¤@HCF=.
Vtps;.R=.\=;.\~{{ttp;.R=;.epp;.V=;.
_p|t}xt;.\=.\~{tr{p.r{~}x}v.~u.p.
w|p}.rp}}pqx}~xs.trt~.wxrw.x.
p{~.ttts.x}.tx=.Qx~rwt|.Y=.@HH@;.
AFH;.@AH¤BC=.
Vw~{p}s.b;.\pwt.Wf;.bx|~}x}.U;.Ux{{x~{.
S;.Zxtuut•.Q[;.\p{s~}ps~.a=.
\~x
px~}p{.tuutr.~u.rp}}pqx}~xs.pt.
|tsxpts.q.|<~x~xs.p}s.zpp<.
~x~xs.trt~=.Y.]t~rx=.A??A.Utq.@J.
AA6B7I@@CE¤DC=.
V~{s.[W;.Qp{t.a[;.Qpt.a[;.Qx.Sc;.
\px}.Qa=.P.r~|px~}.~u.wt.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
sxrx|x}px
t.x|{.~txt.~u.
st{p.H<tpws~rp}}pqx}~{.p}s.R_.
DD;HC?.x}.p.p}s.wt.|~}zt=.Y.
_wp|pr~{.T.cwt[email protected]vJ.
AEA6A7ICFH¤GE=.
V~{srw|xs.[;.axrwps~}.VP;.fx{{u~s.Y;.
Sp.][=._t}pp{.|pxyp}p.t~t.
p}s.x}t{{xvt}rt.t.tu~|p}rt.p.pvt.
E=.Y=P|=Prps=Rwx{s=Ps~{tr=_rwxp=.
A??G.\p•J.CF6B7IADC¤EB=.
V~}v.W.Y;.cpwzx}.S_;.bx||~}.\b;.
Rp{
ptt.Q;.bwpx~.QY=.Prt.p}s.
qprt.q~}rwxp{.tuutr.~u.~p{.
rp}}pqx}~xs=.R{x}._wp|pr~{.cwt=.
@HGC.Yp}J.BD6@7IAE¤BA=.
V~}v.Y_;.^}px
x.Tb;.{wxv~.W;.[x.`a;.
cpv{xput~._P;.Qr~.P;.dw{.Va=.
Rp}}pqx}~xs.RQA.trt~I.
X||}~wx~rwt|xrp{.{~rp{xpx~}.x}.p.
qpx}=.Qpx}.at=.A??E.Utq.BJ.@?F@6@7I@?¤.
AB.
V~}x~tz.f;.[}}.R;.Up}.g;.]p{p.b;.
[}st{{.S;.Wxzx}.af=.T}s~rp}}pqx}~xs.
A<.pprwxs~}{.v{rt~{.x.p.u{{.pv~}x.
w~vw.w|p}.t.A.rp}}pqx}~xs.
trt~I.p}pv~}x|.q.p}p}sp|xst=.
\~{._wp|pr~{=.A???.\pJ.DF6D7I.@?CD¤.
D?=.
V~}p{t.a=.Prt.p}s.}~}<prt.tuutr.~u.
rp}}pqx.~}.qpx}.u}rx~}x}v.p}s.
}t~rw~{~vxrp{.tu~|p}rt=.
]t~rw~{=at
=.A??F.btJ.@F6B7IBCF¤.
E@=.
V~}p{t.b;.Rtqtxp.\;.Ut}p}st<ax.Y=.
Rp}}pqx}~xs.~{tp}rt.p}s.stt}st}rtI.
p.t
xt.~u.sxt.x}.{pq~p~.p}x|p{=.
_wp|pr~{=Qx~rwt|=Qtwp
=.A??D.Y}J.
G@6A7IB??¤@G=.
Vp}.X=.U~t~s.q.Xv~.Vp};.\=S=;.
Sxtr~;.Rt}t.u~.\tsxrx}p{.Rp}}pqx.
attprw.6R\Ra7=.]t~wp|pr~{~v=.
A??D.Y}J.CG6G7I.@?EF=.
Vxuuxw<[t}stx}v.\U;.fxv|p}.Yc;._x}rt.
p}.[P;.Wxyqtv.bR;.Wxx}z.PR;.
^|t{.Y;.etw{.UR;.
p}.^Y;.bppq.W;.
e~{{tqtvw.fP=.Rp}}pqx.t.p}s.
{}tpqx{x.u~.rw~x.x}.tp{.
ps~{trt}rt<p.caPX[b.s=.Pssxrx~}.
[email protected]=.
V~t}wt|t}.U=._wp|pr~zx}txr.p}s.
wp|pr~s}p|xr.~u.rp}}pqx}~xs=.
R{x}._wp|pr~zx}t=.A??BJ.CA6C7IBAF¤E?=.
Vqt.bP;.bpvp.ZP;.Spw{vt}.\Z;.aprx}t.
\;.[zp.bT=.Pvt.~u.~}t.~u.|pxyp}p.
t.p}s.ttrx
t.u}rx~}=.
_rw~{=Pssxr=Qtwp
[email protected]J.
AE6B7ICHE¤D?E=.
Wp{{.fS;.[}zt.\=.X.rp}}pqx.p.vptp.
svN.ctx}v.w~wtt.pq~.wt.
t{px~}wx.qttt}.rp}}pqx.t.p}s.
wt.t.~u.~wt.x{{xrx.sv=.Sv.
P{r~w~{.at
=.A??D.Yp}J.AC6@7IBH¤CG=.
Wp}t.\;.V}st~}.Tf;.apqzx}.Y;.Wp.R[;.
e~qv.bZ;.R~|t.bS;.p}s.Up{x}.af=.
A??F=.S~}pqx}~{.p}s.|pxyp}p.x}.WXe<.
~xx
t.|pxyp}p.|~zt=.
Rp{~xr=x}pzt;.|~~s;.p}s.{tt=.Lqwtpcn!
qh!ceswktgf!koowpg!fghkekgpe{!
u{pftqogu!)2;;;*!CD.6D7I.DCD¤DDC=.
Wp}t.\;.apqzx}.Y;.V}st~}.T;.p}s.U~{x}.
af=.A??D=.S~}pqx}~{.p}s.|pxyp}p.x}.
WXe697.|pxyp}p.|~ztI.prt.tuutr.
~}.rp{~xr.x}pzt.p}s.|~~s=.
Ru{ejqrjctoceqnqi{!@G@6@7I.@F?¤@FG=.
Wp}t.\;.fps.Pb;.R~|t.bS;.Up{x}.af;.
Uxrw|p}.\f=.Pqx}t}rt.|~|.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
u~{{~x}v.|~zts.|pxyp}p.x}.w|p}=.
_rw~wp|pr~{~v.6Qt{7.@HHH;.
@C@6C7IBHD¤C?C=.
Wp};.[=;.Qtt;.P=;.crwx{xq~};.b=;.bwx{~pw;.
b=;.V~{st}qtv;.S=;.W~~x;.\=;.
_ttt;.a=V=;.a~~;.a=.P=;.\trw~{p|;.
a=;._xst;.T=.Wd¤B?GI.p.trxuxr.pv~}x.
u~•.RQ6A7;.p.•t•x•wt•p{.Rp}}pqx}~xs.
trt~=._~r=.]p{=.Prps=.brx=.dbP.
@HHH;.HE;.@CAAG¤BB=.
Wtxw|p}.bY;.Wtx.\P;.Qt}}x}vuxt{s.YT;.
R~}t.TY=.Prt.p}s.txsp{.tuutr.~u.
|pxyp}pI.~ux{t.~u.{p|p.cWR.{t
t{;.
wx~{~vxrp{;.qytrx
t;.p}s.
tu~|p}rt.|tpt=._wp|pr~{.
Qx~rwt|.Qtwp
=.@HH?.]~
J.BF6B7IDE@¤D=.
Wtzt}wp|;.\=.Rp}}pqx}~xs.trt~.
{~rp{xpx~}.x}.qpx}I.at{px~}wx.~.
|~~.p}s.tps.t|=.X}I.Zp{x
p;.
_=f=;.p}s.bp|~};.W=W=;.ts=.cwt.
}t~qx~{~v.~u.sv.p}s.p{r~w~{.
pssxrx~}=.P}}.]h.Prps.brx.@HHA;.EDC;.
@H¤BA=.
Wtzt}wp|;.\=;.[}};.P=.Q=;.[x{t;.\=.S=;.
Y~w}~};.\=.a=;.\t{
x};.[=.b=;.st.R~p;.
Q=.a=;.axrt;.Z=.R=.Rp}}pqx}~xs.trt~.
{~rp{xpx~}.x}.Qpx}=._~r=.]p{=.Prps=.
brx=.db.P=.@HH?;.GF;.@HBA¤@HBE=.
Wt|x}v;.a=X=J.W~~zt;.f=S=J.p}s.Y~}t;.a=c=.
ctpws~rp}}pqx}~{.r~}t}.p}s.
sxuutt}rt.x}.|pxyp}p.|~zx}v.
qtwp
x~=._rw~wp|pr~{~v.@HGE;.
H?6A7I@E?¤@EA=.
Wx{{xv;.Z=f=.Vt}txr.t
xst}rt.u~.trxpx~}.
x}.Rp}}pqx.6Rp}}pqprtpt7=.Vt}txr.
at~rt.p}s.R~.T
~{x~}.DAI.@E@¤.
@G?;.A??D=.
Wx
~}t};.Y=;.V~~sx};.a=.b=;.[x;.R=.c=;.ct;.
V=.T=;.i~vwqx;.b=.b=;.\~t;.R=;._xzt;.e=.
f=;.e~{z~;.]=.S=;.Wtx;.\=P=;.X}}x;.
a=.Q=.at
txq{t.p}s.tvx~}p{{.t{trx
t.
s~}tv{px~}.~u.qpx}.rp}}pqx}~xs.
RQ@.trt~.x}.rw~}xr.spx{.rp}}pqx.
|~zt=.\~{=._rwxp=.A?@A6Y}7;.
@F6E7;.ECB¤ECH=.
Wx
t{;.a=.[=;.\~wt;.f=.P=;.W~uu|p};.U=.f=.
X~{px~}.~u.p}<7 <.
tpws~rp}}pqx}~{.u~|.|pxwp}p=.Y=.
P|=.Rwt|=.b~r=.@HEE;.GG;.@GBA¤@GBB=.
W~{{xt.[T;.Vx{{txt.WZ=.St{p<G<.p}s.st{p<.
H<tpws~rp}}pqx}~{.r~|px~}.x}.
|p}.q.~p{.p}s.x}p
t}~.
ps|x}xpx~}=.R{x}=_wp|pr~{=cwt=.
@HFB.\pJ@C6B7IBDB¤F=.
W~{{xt;.[=T=.Wtp{w.ptr.~u.rp}}pqx=.
_wp|pr~{~vxrp{.at
=.@HGE;.BG;.@¤A?=.
W~{{xt;.[=T=.Rp}}pqx=.6[xtpt.t
xt7=.
Prp._rwxp.brp}s.6b{7.@HGG;.FG;.
@?G¤@@G=.
W~{t.PR;.Qtx
~vt{.Rb;.Rwx{st.ba;.
Stps{t.bP;.Wp|~}.aT;._px}~.[Y=.
Rp}}pqx}~xs.wx~{~v.p}s.
wp|pr~{~vI.B?.tp.~u.~vt=.
]t~wp|pr~{~v=.A??CJ.CF.b{.
@IBCD¤DG=.
Wtx;.\=P=;.bp|~};.P=W=;.W~{xrz;.Q=.Y=;.
Qt}}x}vuxt{s;.Y=T=;.R~}t;.T=.Y=.
Rwpprtxpx~}.~u.wt.pq~x~}.wpt.
~u.|pxyp}p.|~zx}v=.R{x}=._wp|pr~{=.
cwt•=.@HHAp;.DA;.B@¤C@=.
Wtx;.\=P=J.Qt}}x}vuxt{s;.Y=T=J.p}s.R~}t;.
T=Y=.Q{~~s.Rp}}pqx}~xs=.@=.Pq~x~}.
~u.cWR.p}s.u~|px~}.~u.@@¤^W¤cWR.
p}s.cWR.R^^W.sx}v.p}s.put.
|~zx}v.|pxyp}p=.Y.P}p{.c~xr~{.
@HHAq;.@E6D7;.AFE¤AGA=.
RQ!11111
Hto!11134
Hov!5812
Uhov!5813
6481;!
W}.RP;.Y~}t.ac=.c~{tp}rt.p}s.
sx~xx~}.~u.tpws~rp}}pqx}~{.x}.
|p}=.Y._wp|pr~{.T.cwt=.@HG?.^rJ.
A@D6@7IBD¤CC=.
X{p}.PQ;.Vt
x}.P;.R~{t|p}.\;.T{~w{.\P;.
st.fW=.]t~wx~{~vxrp{.p}s.
qytrx
t.~ux{t.~u.|pxyp}p.xw.
px}v.r~}rt}px~}.~u.rp}}pqx}~xs=.
Qtwp
=_wp|pr~{=.A??D.btJ.@E6D¤.
E7ICGF¤HE=.
X}xt.~u.\tsxrx}t=.Sx
xx~}.~u.Wtp{w.
brxt}rt._~{xr=.\pxyp}p.p}s.Wtp{wI.
at~.~u.p.bs.q.p.R~||xtt.~u.wt.
X}xt.~u.\tsxrx}t;.Sx
xx~}.~u.Wtp{w.
brxt}rt._~{xr=.fpwx}v~};.SRI.
]px~}p{.Prpst|._t;.@HGA=.
X}xt.~u.\tsxrx}t;.Sx
xx~}.~u.
]t~rxt}rt.p}s.Qtwp
x~p{.Wtp{w=.
\pxyp}p.p}s.\tsxrx}tI.Ptx}v.wt.
brxt}rt.Qpt=.fpwx}v~}.SRI.]px~}p{.
Prpst|._t;.@HHH=.
Y~wp}~};.T=J.Wp{{sx};.\=\=J.Pvt{{;.b=J.
W~{{xt;.[=T=J.p}s.Vx{{txt;.W=Z=.
ct|x}p{.t{x|x}px~}.{p|p.wp{u<{xut.~u.
st{p.@<tpws~rp}}pqx}~{.6st{p.@¤.
cWR7.x}.wtp
.t.~u.|pxyp}p=.T.Y.
R{x}._wp•|pr~{.@HGH;.BF6B7;.AFB¤AFF=.
Y~w}~};.[=.S=;.^ \p{{t;._=.\=;.\xtrw;.a=.
P=;.Qprw|p};.Y=.V=;.4.brw{t}qtv;.Y=T=.
6A?@C7=.Oqpkvqtkpi!vjg!Hwvwtg!pcvkqpcn!
uwtxg{!tguwnvu!qp!ftwi!wug<!2;863124<!
Qxgtxkgy-!mg{!hkpfkpiu!qp!cfqnguegpv!
ftwi!wug/!P}}.Pq~I.X}xt.u~.b~rxp{.
attprw;.cwt.d}x
tx.~u.\xrwxvp};.
GC••=.
Y~}t;.a=c=J.Qt}~x;.]=[=J.p}s.Wt|x}v;.a=X=.
R{x}xrp{.t{t
p}rt.~u.rp}}pqx.~{tp}rt.
p}s.st•t}st}rt=.Y.R{x}._wp•|pr~{.@HG@;.
A@;@CBb¤@DAb=.
Y~}t.ac=.Rpsx~
pr{p.t|.tuutr.~u.
|pxyp}p=.Y.R{x}._wp|pr~{=.A??A.
]~
JCA6@@.b{7IDGb¤EBb=.
Yx}~
p.i;.V~{sqtv.ba;.Wtxw|p}.bY;.
cp}sp.V=.bt{u<ps|x}xpx~}.~u.
rp}}pqx}~xs.q.ttx|t}p{.p}x|p{.
p}s.w|p}.|pxyp}p.|~zt=.
_wp|pr~{.Qx~rwt|.Qtwp
=.A??D.Y}J.
[email protected]¤AHH=.
Yx}~
p.i;.cp}sp.V;.atswx.VW;.V~{sqtv.
ba=.bt{u<ps|x}xpx~}.~u.st{pH<.
tpws~rp}}pqx}~{.6cWR7.q.sv.
}px
t.xt{.|~}zt=.
_rw~wp|pr~{~v.6Qt{7=.A??B.btJ.
@EH6A7I.@BD¤C?=.
Yx}~
p.i;.cp}sp.V;.\}p._;.V~{sqtv.
ba=.cwt.~x~xs.p}pv~}x.}p{t~}t.
tsrt.wt.tx}u~rx}v.tuutr.~u.St{p.H.
tpws~rp}}pqx}~{.6cWR7.x}.xt{.
|~}zt=._rw~wp|pr~{~v=6Qt{=7.
A??C.P••J{.FB6@¤A7I@GE¤HC=.
Zp}st{;.S=.bpvt.x}.ps~{trt}.x}
~{
t|t}.
x}.sv.t=.brxt}rt.@HFDJ.@H?IH@A¤H@C=.
Zp}st{.SQ;.Rwt}.Z=.ct.~u.|pxyp}p.t.
q.{~}vxsx}p{.r~t=.Y.bs.P{r~w~{=.
A???.\pJ.E@6B7IBEF¤FG=.
Zp}x~{.XV;.bwxpzpp.X;.Zpx}zx.];.
_utut•|p}.P;.Rp•{x}x.TP=.Rp}}pqxsx~{.
x}tutt.xw.wt.tuutr.~u.st{p.H<.
tpws~rp}}pqx}~{.x}.|p}=.T=Y.
_wp•|pr~{[email protected]•JAG6@7I@FA¤F=.
Zp}x~{.XV;.bwxpzpp.X;.cpzpwpwx.a];.
Z}~qt{.T;.\.aT=.Tuutr.~u.st{pH<.
tpws~rp}}pqx}~{.p}s.rp}}pqx}~{.x}.
|p}=._wp|pr~{~v.@HFDJ.@B6E7ID?A¤@A=.
Zxz.Y\;.st.fx.W=.at~}t.~.~p{.st{pH<.
tpws~rp}}pqx}~{.x}.utt}.p}s.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64821!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
x}utt}.|pxyp}p.t=._wp|pr~{.
Qx~rwt|.Qtwp
=.@HHH.\pJ.EB6@7I@BF¤CA=.
Ztt.a;.
p}.^Y;.[xtq.a;.fxrwt}.Wd;.
W~u{t.\;.Wt}t.R=.R~}x}ts.
rp}}pqx.t.p}s.xz.~u.x}rxst}rt.p}s.
txt}rt.~u.rw~xr.|~|I.@?.
tp.u~{{~<.r~w~.s=.Q\Y.A?@@J.
BCAIsFBG=.
Zwp{t.X;.W||t.\;.\x{{t.R;.bt}t<.
d}ttvt.Q;.V}wt.e;.ftrws~}.W;.
Qpxp.WY;.U{txrwwprzt.ff=.Tuutr.~u.
rp}}pqx.t.~}.r~v}xx
t.u}rx~}.p}s.
sx
x}v.pqx{x=.Y.R{x}._rwxp=.@HHH.
Y}J.E?6E7IBHD¤H=.
[tt.\W;.Wp}r~.aY=.Tuutr.~u.|~zx}v.
rp}}pqx.~}.{}v.u}rx~}=.Tt.
at
=atx=\ts.A?@@.PvJ.D6C7IDBF¤CE=.
[t|qtvt.[=;.bx{qttx};.b=.S=;.Pt{~s;.Y=;.
Z~x};.X=.Y=.\pxwp}pI.sxt.~}.wt.
sx~xx~}.p}s.|tpq~{x|.~u.st{p<H<.
tpws~rp}}pqx}~{.x}.|p}=.brxt}rt.
@HF?;.@F?;.@BA?¤@BAA=.
[t|qtvt.[=;.ftx;.Y=.[=;.fpp}pqt;.P=.\=;.
Vp{p}t;.X=.\=;.fp;.a=.Y=;.Rps~};._=e=.
St{p<H<tpws~rp}}pqx}~{I.t|~p{.
r~t{px~}.~u.wt.rw~{~vxrp{.tuutr.
p}s.q{~~s.{t
t{.put.
px~.~t.~u.
ps|x}xpx~}=.]t.T}v=.Y=.\ts=.@HFAp;.
AGE6@B7;.EGD¤EGG=.
[t|qtvt;.[=;.Rpqtt;.a=.T=;.a~t;.W=.\=.
@@<Ws~<7 <tpws~rp}}pqx}~{I.
wp|pr~{~v;.sx~xx~}.p}s.
|tpq~{x|.~u.p.|py~.|tpq~{xt.~u.
|pxwp}p.x}.|p}=.brxt}rt.@HFAq;.@FF;.
EA¤EB=.
[t|qtvt.[=;.aqx}.P=.cwt.wx~{~vxr.
sx~xx~}.~u.|pxwp}p.x}.|p};.[xut.
brx=.@HFD;.@F;.@EBF¤CA=.
[x.\¤R=;.Qps;.Y=.T=;.Sx\pvvx~;.R=.Y=;.[psx;.
P=.a=;.c~}v;.Z=.h=;.[x;.V=.\pxyp}p.t.
p}s.|~~.
twxr{t.rpwt=.
Txst|x~{~vxr.at
xt=.A?@A;.BC;.ED¤FA=.
[xv~x.P;.Vp~.R_;.a~qx}~}.YW=.Tuutr.~u.
|pxyp}p.~}.tx{xqx|;.rw~|~~.
tu~|p}rt;.p}s.x|{pts.sx
x}v=.
Qtwp
._wp|pr~{=.@HHG.]~
J.H6F7IDHH¤.
E?H=.
[x{t.YP;.Zt{{.cW;.Wp.[a=.btppt.p}s.
r~|qx}ts.tuutr.~u.wt.rp}}pqx}~xs.
pv~}x.}pqx{~}t.p}s.St{p6H7<cWR.x}.
w|p}.sxrx|x}px}v.St{p6H7<cWR=.
Sv.P{r~w~{.Stt}s=A?@@.Y{.@J.@@E6@¤.
B7IGE¤HA=.
[x{t.YP;.Zt{{.cW;._x}z.SY;.Wp.[a=.
bqxx~}.~ux{t.~u.St{pH<.
tpws~rp}}pqx}~{;.xp~{p|;.
ws~|~w~}t;.p}s.|tw{wt}xspt.
x}.w|p}.sxrx|x}px}v.St{pH<.
tpws~rp}}pqx}~{=.
_rw~wp|pr~{~v.6Qt{=7.A??H.PJ.
A?B.6A7IAC@¤D?=.
[xspw{.Z\;._xrt.Yb=.X}rtpts.|pxyp}p.
t.p}s.vt}st.tsxr.~~t.r~v}xx
t.
u}rx~}x}v.x}.ps~{trt}.p}s.t|tvx}v.
ps{=.Y.X}.]t~rw~{[email protected]{J.
@G6C7IEFG¤GG=.
[~}.\Y;.Qp.Y[;._p}x~}.\b;.c~~|t.a;.
Txt}.b;.gxp}.W;.cp}v.\c=.
]t~rw~{~vxrp{.r~}tt}rt.~u.
tv{p.|pxyp}p.tI.P.x}.s=.
_rw~{.\ts=.A??C.^rJ.BC6F7I@ABH¤D?=.
\przxt.Z;.[px.h;.ftt}q~tz.a;.\xrwt{{.a=.
Rp}}pqx}~xs.prx
pt.p}.x}ps{.
trxux}v.~px|.r~}srp}rt.p}s.
x}wxqx.`<t.rp{rx|.rt}.x}.PcA?.
rt{{.p}utrts.xw.p.qpx}.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
rp}}pqx}~xs.trt~=.Y.]t~rx=.@HHD.
^rJ.@D6@?7IEDDA¤E@=.
\p{s~}ps~.a=.bs.~u.rp}}pqx}~xs.
stt}st}rt.x}.p}x|p{=._wp|pr~{.cwt=.
A??A.PvJ.HD6A7I.@DB¤EC=.
\p{x}~zp.Q;.Qpp}~zp<Zrz~.\;.
brw{xrzt.T=.cxwpxr.q{~~s.tt.
t~}t.~.rp}}pqx}~xsI.S~.t.
}stp}s.wt.|trwp}x|N.Q=Y.
_wp•|pr~{[email protected]••J.@ED6F7IA?FB¤GG=.
\p}xt<Vprxp.T;.ip||x.b;.Sp{|p}.R;.
Wt||x}v~}.c;.P}stp~}.b;.P{{tqtrz.
_=.Rp}}pqx;.rwx~wt}xp.p}s.~wt.
}~}<puutrx
t.rw~tI.BD.tp.~u.
u~{{~<.~u.p.~{px~}<qpts.r~w~=.
_rw~{=\[email protected]}J.CA6E7I@BA@¤G=.
\pt||p}x.X;.[ptx.P;._prx}x.\;.[~
trxr.
\;._{prxsx.VU;._tvx.V=.Sxpv}~xr.p}s.
|~|p~{~vxrp{.utpt.x}.rw~}xr.
rw~xr.pxt}.prr~sx}v.~.rp}}pqx.
t.p=.Y._rw~prx
t.Sv=.A??C.
Y}J.BE6A7IABD¤C@=.
\pxyp}p.brwts{x}v._txx~}J.St}xp{.~u.
_txx~}J.at|p}sJ.Ux}p{.^st; .DF.
Utstp{.atvxt.DH.6AE.\prw.@HHA7;.=.
@?CHH¤@?D?G=.
\pt{{~p.\R;.R~.V;.Up~t.[;.Vtp.V[;.
Upp.f=.bt{u<ps|x}xpx~}.~u.wt.
rp}}pqx}~xs.trt~.pv~}x.fX].
DD;A@A¤A.x}.sv<}px
t.|xrt=.
]t~rxt}[email protected]{J.GD6A7IBAF¤B?=.
\psp;.[=P=;.[~{px;.b=Y=;.Q~}tx};.\=Y=;.
h~}v;.P=R=;.Q~}}t;.c=X=.brt.~u.p.
rp}}pqx}~xs.trt~.p}s.u}rx~}p{.
ttx~}.~u.wt.r{~}ts.rS]P=.]pt.
@HH?;.BCE;.DE@¤DEC=.
\r\pw~}.[a=.Ppt}.puux}x.tx|pt.~u.
x|~}pqp}.x}.r~|qx}px~}.xw.
p}p}sp|xst.p}s.rwt|xrp{.p}p{~v.~u.
p}p}sp|xst.x}.wt.|~}zt.
sxrx|x}px}v.St{pH<.
tpws~rp}}pqx}~{=.
_rw~wp|pr~{~v.6Qt{=7.A??H.PJ.
A?B6A7IA@H¤AG=.
\r\pw~}.[a;.Vx}qv.QR;.[p|q.aY=.
Rp}}pqx}~xs.pv~}x.sxuutt}xp{{.
qxt.u~.wt.sxrx|x}px
t.x|{.
tuutr.~u.St{p6H7<tpws~rp}}pqx}~{.
x}.RDFQ[>EY.|xrt=._rw~wp|pr~{~v.
6Qt{=7.A??G.Y{J.{HG6C7ICGF¤HD=.
\trw~{p|;.a=.Rp}}pqx}~xs.rwt|x=.X}.
\trw~{p|;.a=.6TS=7.\pxyp}p;.=A¤GG.
6]t.h~z;.]h;.Prpst|xr._t;.X}r=7;.
@HFB=.
\trw~{p|.a;._tt.\;.\x{{~<a~sxvt.
T;.Wp}.[^=.Rp}}pqxsx~{¥trt}.
ps
p}rt=.Rwt|=Qx~sx
t=.A??F.
PvJC6G7I.@EFG¤HA=.
\trw~{p|.a;.bw
~.h=.Wpwxw<XI.cwt.
rt.~u.Rp}}pqxsx~{=.ctpwts~}=.
@[email protected]?FB¤FG=.
\tw|tsxr.i;.Rwp}s•p.b;.b{pst.S;.St}wp|.W;.
U~t.b;._pt{.Pb;.a~.bP;.Zwp}.XP;.
T{b~w{.\P=._~t}r.ct}s.~u.% <cWR.
p}s.~wt.rpx|pqx}~xs.x}.r~}uxrpts.
rp}}pqx.tppx~}.u~|.@HHB.~.A??G=.
Y.U~t}xr.brx=.A?@?.btJ.DD6D7I.@A?H¤.
@A@F=.
\txt.\W;.Rpx.P;.P|q{t.P;.Wpx}v~}.W;.
W~.a;.Zttut.ab;.\rS~}p{s.Z;.fps.
P;._~{~}.a;.\~uux.cT=._txt}.
rp}}pqx.t.w~.}t~rw~{~vxrp{.
str{x}t.u~|.rwx{sw~~s.~.|xs{xut=.
_~r=]p{=Prps=brx.d=b=P.A?@A.^r.AJ.
@?H6C?7ITAEDF¤TAEEC=.
\txyt.YW;.Stzzt.];.Z~tt.\f;.`tt._Y;.
p}.Qt
tt}.]Y;.\txyt.RY=.Rp}}pqx.p}s.
RQ!11111
Hto!11135
Hov!5812
Uhov!5813
r~v}xx
t.tu~|p}rt.x}.rw~xI.P.
r~<trx~}p{.s.x}.pxt}.xw.
}~}<puutrx
t.rw~xr.x{{}t.p}s.wtx.
}puutrts.xq{x}v=._rw~{=\ts.A?@A.
P••J.CA6C7IF?D¤@E=.
\t}st{~}.YW;.\t{{~.]Z=.Tuutr.~u.
|pxyp}p.~}.}t~t}s~rx}t.w~|~}t.
x}.w|p}.|p{t.p}s.ut|p{t=.]XSP.at.
\~}~v•[email protected]¤@@C=.
apsp}.\\;.T{~w{.\P;.b{pst.S;.Pw|ts.
bP;.Zwp}.XP;.a~.bP=.Qx~{~vxrp{{.
prx
t.rp}}pqx}~xs.u~|.wxvw<~t}r.
Rp}}pqx.px
p=.Y.]p._~s=.A??H.\p.
AAJ.FA6D7IH?E¤@@=.
ap|ptzt.YV;.Qtvwp.V;.
p}.[pp.\;.
S||t.^W=.S~t.t{pts.xz.~u.|~~.
twxr{t.rpwt.put.rp}}pqx.t=.Sv.
P{r~w~{.St•t}s=.A??C.Utq.FJ.FB6A7I.@?H¤.
@H=.
ap|ptzt.YV;.Zpt.V;.
p}.a_;.cwt}xt}.
T[;.brw}txst•.T;.\~t{{t•.\a=.Wxvw<.
~t}r.|pxyp}p.x|px.ttrx
t.
u}rx~}.p}s.x}wxqx~.|~~.r~}~{=.
]t~rw~wp|pr~{~v.A??E.^rJ.
B@6@?7IAAHE¤B?B=.
ap|ptzt.YV;.Zpt.V;.cwt}xt}.T[;.
c~t}}t.bf=;.\~t{{t.\a=.
]t~r~v}xx
t.tu~|p}rt.sx}v.
prt.cWR.x}~xrpx~}.x}.wtp
.p}s.
~rrpx~}p{.rp}}pqx.t=.
Y=_rw~wp|pr~{=.A??H.\pJ.
AB6B7IAEE¤FF=.
ap|ptzt.YV;.\~t{{t.\a;.
p}.axt}qttz.
_;.cwt}xt}.T[;.brw}txst.T;.Zpt.V=.
R~v}xx~}.p}s.|~~.r~}~{.p.p.
u}rx~}.~u.% <cWR.r~}rt}px~}.x}.
t|.p}s.~p{.u{xsI.[x|x.~u.
x|px|t}=.Sv.p}s.P{r~w~{.
St•t}st}rt=.A??EJ.GDI@@@C¤@AA=.
atrwts{x}v.~u.wt.U~~s.p}s.Sv.
Ps|x}xpx~}.P~
ts._~sr.
R~}px}x}v.b}wtxr.S~}pqx}~{.j6¤7<.
st{p <6p}7<ctpws~rp}}pqx}~{l.x}.
btp|t.^x{.p}s.T}rp{pts.x}.b~u.
Vt{px}.Rp{t.U~|.brwts{t.XX.~.
brwts{t.XXXJ.Ux}p{.a{t; .EC.Utstp{.
atvxt[email protected]{.@HHH7;.=BDHAG¤.
BDHB?=.
axvv._Z;.epxsp.U;.a~x.bb;.b~zx}.[b;.
V~p.Q;.Vp}.X;.T{{x.aY=.P.x{~.s.
~u.wt.tuutr.~u.rp}}pqx.~}.ptxt.
w~|~}t.x}.WXe<x}utrts.ps{.|t}=.
Qpx}.at[email protected]}.@@J.@CB@ICE¤DA=.
a~sxvt.st.U~}trp.U;.V~xx;.\=P=;.
Ut}p}st<ax;.Y=Y=;._p{~|~;.c=;.ap|~;.
Y=P=.S~}tv{px~}.~u.p.qpx}.
rp}}pqx}~xs.qx}sx}v.xt.put.rw~}xr.
st{p.H<tpws~rp}}pqx}~x{.tp|t}=.
_wp|pr~{=.Qx~rwt|=.Qtwp
=.@HHC;.CF.6@7;.
BB¤C?=.
a~w.\S;.cpwzx}.S_;.fwxpzt.Z\;.Rw~x.
a;.Qp{sx}.VR=.ctpws~rp}}pqx}~{.
tt.x||}t.u}rx~}.p}s.
t}wp}rt.WXe.t{xrpx~}.x}.wt.w_Q[¤.
bRXS.|~t=.[xut.brx=.A??D.Pv.@HJ.
FF6@C7I@F@@¤AA=.
a~.TQ=.cp|x}v.cWRI.~t}xp{.rp}}pqx.
}tv.p}s.w~rp}}pqx}~xs<tt}~xs.
t}~pvt.tuutr=.Q.Y._wp|pr~{=.A?@@.
PvJ.@EB6F7I@BCC¤EC=.
bpupp.b;.Pup.U;.Pswp|x.e\;.\zwp.W=.
Rp}}pqx}~xs.trt~.p.p.}~
t{.pvt.
u~.wt.tp|t}.~u.~pt.rp}rt=.
Rp}rt.at=.A??D.\p[email protected]@EBD¤C@=.
bp}s~<_txp.\=R=;.c~;.Z=;.St{p;.T=a=;.
W~w|p};.P=V=;.U~•rt;.\=;.fp{zt•;.Y=\=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
T}s~vt}~.rp}}pqx}~xs.p.p}.p
tx
t.
~.r~}t<tpsx}v.t|.x}.wt.p=.
]t~rx=.[t=;.AAB;.@AD¤@AG;.@HHF=.
\t}sxpqp{.e;.ix||t.P;.\p{s~}ps~.a=.
X}
~{
t|t}.~u.zpp>s}~wx}.t|.
x}.fX].DD;A@A¤A.t{u<ps|x}xpx~}.x}.
|xrt=.]t~rw~wp|pr~{~v=.A??E.
bt•J.B@6H7I@HDF¤EE=.
\tx.YR;.Rpu~s.fY;.P{tp}st._R;.
P}st.P[;.p}s.Vt{qp.bb=.@HG?=.Tuutr.
~u.|pxwp}p.~}.x}p~r{p.p}s.q{~~s.
tt.x}.v{pr~|p=.Qrjvjcnoqnqi{!GF.
6B7I.AAA¤AAG=.
\tx}x.[;.Zxp}xs~.P;.\p{tupzx.b;.p}s.
_ppwp}p~~{~._=.
]t~rw~{~vxrp{.stuxrx.x}.{~}v<t|.
utt}.rp}}pqx.t=.]t~{~v.A??E.
EEIFBF¤FBH=.
\x}~x.b;.Sp
~{x.\;.Qpvpv{x.P\;.P|p~.[;.
etrrwx.b;._trrx.RP=.P}.~
t
xt.~u.
t|pxr.t
xt.~}.rp}}pqx.p}s.
rw~xI.Sxrx}v.ppt}{.
r~}u{xrx}v.t{=.Sv.P{r~w~{.at
=.
A?@?.\pJ.AH6B7IB?C¤@F=.
\x{t|p}.\P;.[tx.aP;.\pr{t.\;.
bwt~~s.YQ;.p}s.\{{t.YT=.cxvvtx}v.
|~rpsxp{.x}uprx~}.q.|pxyp}p=.
Rxr{px~}=.A??@J.@?BIAG?D¤AG?H=.
]prt.T_;.\tt.P[;.a~wqtv.Y\;.\p{t~}.
U=.Pssxrts.p}s.}~}pssxrts.sv.t=.
P.r~|px~}.~u.sv.pvt.pt}=.
Prw.Vt}._rwxp[email protected]@7IFF¤G?=.
^
xts~;.P=;.V{~p;.Y=;.Wtzt}wp|;.\=.
Rw~}xr.rp}}pqx}~xs.ps|x}xpx~}.
p{t.rp}}pqx}~xs.trt~.qx}sx}v.x}.
p.qpx}I.P.p}xpx
t.
p~psx~vpwxr.s=.Qpx}.at=.@HHB;.
E@E;.AHB¤B?A=.
_prwt._;.Qpzpx.b;.Z}~.V=.cwt.
t}s~rp}}pqx}~xs.t|.p.p}.t|tvx}v.
pvt.~u.wp|pr~wtp=._wp|pr~{=.
at
=.A??E.btJ.DG6B7IBGH¤CEA=.
_t{p~<ctp}.Y\;.bpt<_x}x{{p._;.Rwpsx.];.
Rt~<_pr~~.Q=.Vt}t<t}
x~}|t}.
x}tprx~}.}st{x}v.wt.tuutr.~u.
rp}}pqx.x}.ux.tx~st.rw~x=.R.
_wp|.St=.A?@AJ.@G6BA7ID?AC¤BD=.
_t~rt{{x._[;.Sx.\p~.e=.P}.x}~srx~}.
~.wt.t}s~rp}}pqx}~xs.t|I.u~|.wt.
tp{.~.wt.{pt.r~}rt=.Qt.
_pr=at=R{x}=T}s~rx}~{=\tpq=.A??H.
UtqJ.AB6@7I@¤@D=.
_x~|t{{x.S=.cwt.t}s~rp}}pqx}~xs.t|I.P.
sv.sxr~
t.ttrx
t=.R.^x}.
X}
txv.Sv=.A??D.Y{J.E6F7IEFA¤H=.
_{trwt.\Y;.exx}vw~uu.T;.Zp{wp}.a;.
axrw|p}.Y;.bpuu~s.\;.bxs}t.b;.[x}.U;.
Ztt.b=.P~rxpx~}.qttt}.
|pxyp}p.t~t.p}s.{|~}p.
u}rx~}.~
t.A?.tp=.YP\[email protected]}.
@@J.B?F6A7I@FB¤G@=.
_~{{p~.U;.cpv{xp{pt{p<brpupx.^;.P{{pp.\;.
\}~.T;.Sx.\p~.e;.St._t~rt{{x.[;.
Pt}sx}~.V=.Qx~prx
t.t}{~v~.
rp}}pqx}~xs.u~|.uxqt.wt|.6Rp}}pqx.
px
p7=.Y.]p._~s=.A?@@.bt.
ABJFC6H7IA?@H¤AA=.
_~t.WV.Y;.Vqt.PY;.Ws~}.YX;.Wtx.
\P;.hvt{}<c~ss.S=.R~v}xx
t.
|tpt.x}.{~}v.t|.rp}}pqx.t=.Y.
R{x}._wp|pr~{=.A??A.]~
J.CA6@@.
b{7IC@b¤CFb=.at
xt=.
brwtt.YU;.Vp}.YS;.S}rp}.PT;.bp~.RT;.
Wpqt.Ya;.Ypr~q.c;.Qrw~{.ZZ=.
bqytrx
t.tuutr.~.rp}}pqx.pt.
p~rxpts.xw.t;.pqt.p}s.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
stt}st}rt.put.psyx}v.u~.vt}txr.
p}s.t}
x~}|t}p{.x}u{t}rt=.Sv.
P{r~w~{.Stt}s=.A??H.]~
.@J.@?D6@¤.
A7IFE¤GA=.
brwxuu|p}.Y;.]pzp|p.Q;.Tp{tx}t.\;.
[pQxt.Y=.b|~|.~u.rwx~.
trtst.rp}}pqx.t=._rwxp.at=.
A??D.\p•.B?J.@BC6@7IBF¤CA=.
brwtx}t.P\;.S}}.\T=.atxsp{.tuutr.~u.
rp}}pqx.t.~}.}t~r~v}xx
t.
tu~|p}rt.put.~{~}vts.pqx}t}rtI.
P.|tp<p}p{x=.T=R{x}.
_rw~wp|pr~{=.A?@A.^rJ.A?6D7ICA?¤.
H=.
bxs}t.b=.Rpsx~
pr{p.r~}tt}rt.~u.
|pxyp}p.t=.Y.R{x}._wp|pr~{=.A??A.
]~
J.CA6@@.b{7IECb¤F?b=.
b~{~xy.];.btwt}.ab;.a~uu|p}.aP;.Qpq~.
c;.Zpsst}.a;.\x{{t.\;.Rwxxp}t}.Z;.
\ratt.Q;.et}stx.YJ.\pxyp}p.
ctp|t}._~ytr.attprw.V~=.
R~v}xx
t.u}rx~}x}v.~u.{~}v<t|.wtp
.
rp}}pqx.t.ttzx}v.tp|t}=.YP\P=.
A??A.\p•.EJ.AGF6H7I.@@AB¤B@=.
bx{x}v.Y;.[tx.b;.W~zx}.a;.fwxt.R=.
Rp}}pqx.t.x~.~.ux.~}t.
rw~x.tsxr.pts.
}t~r~v}xx~}.p.@?<tp.u~{{~<=.
brwx~w.at=.A??D.Y}[email protected]@7I@BD¤F=.
bqp}rt.Pqt.p}s.\t}p{.Wtp{w.bt
xrt.
Ps|x}xpx~};.Ftwi!Cdwug!Yctpkpi!
Pgvyqtm-!3122<!Pcvkqpcn!Guvkocvgu!qh!
Ftwi.Tgncvgf!Gogtigpe{!Fgrctvogpv!
Xkukvu/!WWb._q{xrpx~}.]~=.6b\P7.@B¤.
CFE?;.SPf].btxt.S¤BH=.a~rz
x{{t;.
\SI.bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~};.A?@B=.
bqp}rt.Pqt.p}s.\t}p{.Wtp{w.bt
xrt.
Ps|x}xpx~};.Tguwnvu!htqo!vjg!3123!
Pcvkqpcn!Uwtxg{!qp!Ftwi!Wug!cpf!
Jgcnvj<!Uwooct{!qh!Pcvkqpcn!Hkpfkpiu-!
]bSdW.btxt.W¤CE;.WWb._q{xrpx~}.
]~=.6b\P7.@B¤CFHD=.a~rz
x{{t;.\SI.
bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~};.A?@B=.
bqp}rt.Pqt.p}s.\t}p{.Wtp{w.bt
xrt.
Ps|x}xpx~};.Rt}t.u~.Qtwp
x~p{.
Wtp{w.bpxxr.p}s.`p{x=.Vtgcvogpv!
Grkuqfg!Fcvc!Ugv!)HGFU*<!31123122/!
Pcvkqpcn!Cfokuukqpu!vq!Uwduvcpeg!Cdwug!
Vtgcvogpv!Ugtxkegu/!QWbXb.btxt.b¤ED;.
WWb._q{xrpx~}.]~=.6b\P7.@B¤CFFA=.
a~rz
x{{t;.\SI.bqp}rt.Pqt.p}s.
\t}p{.Wtp{w.bt
xrt.Ps|x}xpx~};.
A?@B=.
cpx.aY;.\przx}}~}.P;.Rwxt}t}.W=.
Rp}}pqx.t.p}s.r~v}xx
t.u}rx~}I.G<.
tp.pytr~.x}.p.~}v.ps{.r~w~=.
Pssxrx~}.A?@@.StrJ.@?E6@A7IA@HD¤A?B=.
cp}ptr.a;.R~}p}x}tr.Rb=.
Rp}}pqx}~xs.p}s.wt.x||}t.t|I.
p}.~
t
xt=.X||}~qx~{~v=.A?@?.PvJ.
A@D6G7IDGG¤HF=.
cp}sp.V;.\}p._;.V~{sqtv.ba=.bt{u<.
ps|x}xpx~}.qtwp
x~.x.|px}px}ts.q.
wt.rw~prx
t.x}vtsxt}.~u.|pxyp}p.
x}.xt{.|~}zt=.]p.]t~rx=.A???.
]~
J.B6@@7I@?FB¤C=.
cpwzx}.S_=.b|~zts.|pxyp}p.p.p.rpt.~u.
{}v.x}y=.\~}p{sx.Prw.Rwt.Sx=.
A??D.Y}J.EB6A7IHB¤@??=.
cpwzx}.S_;.bwpx~.QY;.p}s.Up}z.X\=.@HFC=.
Prt.tuutr.~u.|~zts.|pxyp}p.p}s.
~p{.st{pH<tpws~rp}}pqx}~{.~}.
trxuxr.pxp.r~}srp}rt.x}.pw|pxr.
qytr=.Vjg!Cogtkecp!tgxkgy!qh!
tgurktcvqt{!fkugcug!@?H.6C7I.CA?¤CAG=.
RQ!11111
Hto!11136
Hov!5812
Uhov!5813
64822!
cpwzx};.S_;.iwp}v;.iU;.Vtt}{p}s;.b;.
R~t};.f;.\prz;.c\;.\~vt}t};.W=.
\pxyp}p.dt.p}s.[}v.Rp}rtI.at{.
~u.p.Rpt<R~}~{.bs=.Pqpr.1P.
FFF;.P|txrp}.cw~prxr.b~rxt.|ttx}v;.
\p.AC;.A??E=.
cwt._{p}.[x.6A?@?7=.etx~}.@=._q{xwts.
~}.wt.X}t}tJ.jvvr<00
yyy/vjgrncpvnkuv/qti0!6prrtts.
btt|qt.A?;.A?@B7.
cwt}xt}.T[;.Zpt.VU;.c~t}}t.bf=;.
\~t{{t.\a;.bp|qtw.P;.Q{p}rwps.\\;.
ap|ptzt.YV=.]t~wx~{~vxrp{.
u}rx~}x}v.~u.~rrpx~}p{.p}s.wtp
.
rp}}pqx.t.sx}v.cWR.x}~xrpx~}=.
_rw~wp|pr~{~v.6Qt{=7.A?@A.
\p•JAA?6A7IBC@¤D?=.
cpqt.Q;.bxvs~}.PY;.btt}t.P\;.b~|.
bb;.\rV{}}.ZP=.\pxyp}p.t.p}s.
txr{p.vt|.rt{{.|~=.Rp}rt=.A?@@.
Utq.@DJ.@@FI.GCG¤GDB=.
cxrwt{{.f;.Q~}.b;.\przxt.Z=.
Rp}}pqx}~xs.x}wxqx.]<.p}s._>`<t.
rp{rx|.rwp}}t{.x}.r{ts.p.
wx~rp|p{.}t~}=.Y.]t~wx~{=.
@HHF.Y{J.FG6@7ICB¤D?=.
p}.st.\tt.UY;.et{w~.T;.\txyt.RY;.
\prwxt{t}.\f;.st.W[=.Rp}}pqx.t.x}.
pxt}.p.r{x}xrp{.wxvw.xz.~u.
rw~xI.X|pr.~}.~s~|p{.
|~|.p}s.p}xx~}.~.rw~x=.
R._wp|.St=.A?@AJ.@G6BA7ID?BE¤CC=.
p}.Vpt{.fP;.fxv|p}.Yc;.\~}w~t.Z;.
Zpw}.ab;.
p}.^Y;.Q~z.\_;.e~{{tqtvw.
fP=.Rp}}pqx.t.p}s.qr{x}xrp{.
~xx
t.rw~xr.ttxt}rt.x}.tp{.
ps~{trt}rtI.Ux}sx}v.u~|.p.Srw.
t=.Pssxrx~}[email protected].@?F6A7IBG@¤.
F=.
ep}.V}s.Z;.atqt{{~}.RY=.P.[xut<r~t.
_ttrx
t.~}.wt.Vptp.
W~wtx= .Y.Wtp{w.b~r.Qtwp
=.A?@?.
bt•J.D@6B7IACC¤DH=.
p}.^.Y;.Qpz.\;.Wp}t}.\;.Qxy{.ae;.st.
Vppu.a;.ets~.W=.Rp}}pqx.t.p}s.
rw~xI.P.{~}vxsx}p{.~{px~}<.
qpts.s=.P|.Y.Txst|x~{=.A??A.Pv.
@DJ.@DE6C7IB@H¤AF=.
ep}st.aV;.Qs}t.PY;.\~~t.QP;.Wvwt.
Ya=.P.r~<s.r~|px~}.~u.rp}}pqx.
p}s.~qprr~.xwspp{=.P|.Y.Pssxr=.
A??D.Yp}¤UtqJ.@C6@7IDC¤EB=.
ep}st.aV;.Qs}t.PY;.Wvwt.Ya;.[xv~x.
P=.P.xwx}<qytr.r~|px~}.~u.
xwspp{.|~|.sx}v.pqx}t}rt.
u~|.rp}}pqx;.~qprr~;.p}s.q~w.
qp}rt=.Sv.P{r~w~{.Stt}s=.A??G.
Yp}[email protected]@¤B7ICG¤DC=.
ep}}.aT;.Vp|pvt.cU;.fp}t.YP;.\pwp{{.
T\;.cp{~.][;.\px}.Qa;.fx{t.Y[=.
Sx
tvt}.tuutr.~u.rp}}pqxsx~{.~}.wt.
sxrx|x}px
t.x|{.p}s.{prt.
r~}sxx~}x}v.tuutr.~u.St{p6H7.
tpws~rp}}pqx}~{=.Sv.
P{r~w~{=St•t}s=.A??G.P••[email protected]@¤.
B7I@H@¤G=.
ep}z~
.\\;.cpt.aT;.Zxx{{~
p.V_;.
Zxxrx.[;.at}~{s.\S;.Zttz.\Y;.
R~}p.Z_;.\pwt.Qb;.Xpr~}~.fV;.
Qxt.[;.]tp{t.\R;.R{pz.SQ;.axst}~.
cP=.R~||~}.{xpqx{x.~.pssxrx~}.p}s.
vptp.w~wtx I.cwt~txrp{;.
t|xxrp{.p}s.t
~{x~}p.ttrx
t=.
SvP{r~w~{.Stt}[email protected]}J.@AB.
b{[email protected]¤@F=.
~}.bs~.Z;.[xtq.a;._uxt.W;.W~u{t.\;.
fxrwt}.Wd=fwp.tsxr.x}rxst}.t.
G<^HT^HO^23CWR3/UIO
23CWR3
64823!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
~u.rp}}pqx.p}s.~vtx~}.~.pqt.p}s.
stt}st}rtN.P.C<tp.~trx
t.
tp|x}px~}.~u.xz.upr~.x}.p.
r~||}x.p|{t.~u.ps~{trt}.p}s.
~}v.ps{=.Sv.P{r~w~{.Stt}s=.
A??A.bt•[email protected]@7ICH¤EC=.
fprwt{.ba;.T{b~w{.\P;.a~.bP;.P|qt.Y;.
st.fx.W=.R~|px~}.~u.wt.qytrx
t.
tuutr.~u.St{p.6H7<tpws~rp}}pqx}~{.
p}s.|pxyp}p.x}.w|p}=.
_rw~wp|pr~{~v.6Qt{7=.A??A.Y}J.
@E@6C7IBB@¤H=.
fpv}t.YP;.epvp.Z;.Z}~.V=.Rpsx~
pr{p.
prx~}.~u.rp}}pqx}~xs.p}s.wtx.
vt}tpx~}.sx}v.w~rz=.Y.\~{.\ts=.
@HHG.]~
¤StrJ.FE6@A7IGAC¤BE=.
fpt.\P;.fp}v.c;.bwpx~.b;.a~qx}~}.P;.
Srt.c;.W}w.c;.Vp|p.P;.Qt}}t.
VY;.p}s.R~{{t.Y_=.A?@?=.b|~zts.rp}}pqx.
u~.rw~}xr.}t~pwxr.px}I.P.
p}s~|xts.r~}~{{ts.xp{=.EOCL<!
Ecpcfkcp!Ogfkecn!Cuuqekcvkqp!lqwtpcn!?!
lqwtpcn!fg!nCuuqekcvkqp!ogfkecvg!
ecpcfkgppg!@[email protected]¤TF?@=.
ft~}.Sa;.fpwq}._=.Rt}.pt}.~u.
sv.pqt.wp.x}
~{
t.|~zx}v=.]XSP=.
at=.\~}~v=.@HH?J.HHI.D¤@@=.
fx{t.Y[;.Qpt.a[;.Qx.Sc;.Qp{t.a[;.
\px}.Qa=.Sxrx|x}px
t.x|{.
tuutr.~u.st{p.H<tpws~rp}}pqx}~{.
p}s.st{p.H<@@<tpws~rp}}pqx}~{.x}.
p.p}s.wt.|~}zt=.
]t~wp|pr~{~v[email protected]J.
BA6C7IBDH¤ED=.
fx{t.Y[;.Wuu|p}.Yf;.Qp{t.a[;.\px}.Qa=.
_wp|pr~{~vxrp{.trxuxrx.~u.wt.
sxrx|x}px
t.x|{.tuutr.~u.st{p.
H<tpws~rp}}pqx}~{.x}.wt.
|~}zt=.Sv.P{r~w~{.Stt}s=.@HHD.
]~
J.C?6@7IG@¤E=.
fx{t.Q;.\pr~t.c;.Strw.a;.V~p.Q;.
bpzpx.b;.p}s.S~}pvwt.W=.A?@B=.[~<S~t.
ep~xts.Rp}}pqx.bxv}xuxrp}{.
X|~
t.]t~pwxr._px}=.Vjg!lqwtpcn!
qh!rckp<!Qhhkekcn!lqwtpcn!qh!vjg!Cogtkecp!
Rckp!Uqekgv{/!
fx{t.Q;.\pr~t.c;.c~sxz~
.P;.\x{{|p}.
Y;.Qt}{t.W;.V~p.Q;.p}s.Uxw|p}.b=.
A??G=.P.p}s~|xts;.{prtq~<r~}~{{ts;.
r~~
t.xp{.~u.rp}}pqx.rxvptt.x}.
}t~pwxr.px}=.Vjg!lqwtpcn!qh!rckp<!
Qhhkekcn!lqwtpcn!qh!vjg!Cogtkecp!Rckp!
Uqekgv{!H6E7I.D?E¤DA@=.
f.g;.Ut}rw.TS=.Tuutr.~u.rw~}xr.st{pH<.
tpws~rp}}pqx}~{.~}.p.|xsqpx}.
s~p|x}t.}t~}I.p}.
t{tr~wx~{~vxrp{.pt|t}=.
]t~wp|pr~{~v=.A???.Yp}.AGJ.
BH6B7IBH@¤G=.
hrt{.\;.Q~p.T;.[q|p}.SX;.b~{~xy.];.
Qtt.fY;.R~~}.b\;.R~}._;.cpzpvx.
\Y;.U~|x~.P;.f~~s.bY;.t.p{=.cwt.x|pr.
~u.rp}}pqx.t.~}.r~v}xx
t.u}rx~}x}v.
x}.pxt}.xw.rwx~wt}xpI.p.|tp<.
p}p{x.~u.txx}v.ux}sx}v.p}s.}t.
spp.x}.p.ux<tx~st.p|{t=.
brwx~w=Q{{=.A?@A.\pJ.BG6A7IB@E¤B?=.
itxvt.Yb;.Wpqtxrz.QR;.R~{t.a_;.Twx}vt.
\P;.R~{t.cY;.Wtx.YZ;.W~ut.RY;.
bp{{x}v.\R;.h~}v.bT=;.awtt.bW=.
bqytrx
t.tuutr.~.|pxyp}p.
p~rxpts.xw.|pxyp}p.t.x}.
r~||}x.p}s.r{x}xrp{.qytr=.Sv.
P{r~w~{.Stt}s=.A?@?.Y}.@J.@?H6@¤.
B7I@E@¤E=.
iwp}v.iU;.\~vt}t}.W;.bx.\a;.cpwzx}.
S_;.h.V_;.\pwp{{.Ya;.W.cR;.brwp}.
b_=.\pxyp}p.t.p}s.x}rtpts.xz.~u.
p|~.rt{{.rprx}~|p.~u.wt.wtps.p}s.
}trz=.Rp}rt.Txst|x~{.Qx~|pzt.
_t
[email protected]@A7I.@?F@¤G=.
ipsx.Pf;.bwxpzpp.X;.Ux}zt{upq.T;.
Zp}x~{.XV=.Prx~}.~u.rp}}pqxsx~{.~}.wt.
p}xt.p}s.~wt.tuutr.~srts.q.
st{p.H¤cWR.x}.}~|p{.qytr=.
_rw~wp|pr~{~v.6Qt{=7.@HGAJ.
FE6B7IACD¤D?=.
Vjg!Ogfkecn!Crrnkecvkqp!qh!Octklwcpc<!
C!Tgxkgy!qh!Rwdnkujgf!Enkpkecn!Uvwfkgu!
\p•rw.@H;.A?@D.
_tpts.qI.
d=b=.U~~s.p}s.Sv.Ps|x}xpx~}.
Rt}t.u~.Sv.T
p{px~}.p}s.attprw.
6USP>RSTa7.
R~}~{{ts.bqp}rt.bpuu.6Rbb7.
VCDNG QH EQPVGPVU!
@=.X}~srx~} ========================================================================================================================================================================================
A=.\tw~s ==============================================================================================================================================================================================
A=@ Stux}t.wt.^qytrx
t.~u.wt.at
xt =======================================================================================================================================
A=A Stux}t.\pxyp}p ================================================================================================================================================================
A=B Stux}t.Pstpt.p}s.ft{{<R~}~{{ts.R{x}xrp{.bsxt ===================================================================================================
A=C btprw.\tsxrp{.[xtpt.Sppqpt.p}s.Xst}xu.at{t
p}.bsxt ==================================================================================
A=D at
xt.p}s.P}p{t.`p{xux}v.R{x}xrp{.bsxt ================================================================================================================
B=.at{.p}s.Sxrx~} ======================================================================================================================================================================
B=@.]t~pwxr._px} ======================================================================================================================================================================
B=@=@ ]t~pwxr._px}.P~rxpts.xw.WXe<bt}~.]t~pw ===================================================================================
B=@=A Rt}p{.p}s._txwtp{.]t~pwxr._px} ===================================================================================================================
B=A Ptxt.bx|{px~}.x}.WXe ==================================================================================================================================================
B=B bpxrx.x}.\{x{t.br{t~x =============================================================================================================================================
B=C Pw|p ====================================================================================================================================================================================
B=D V{pr~|p =================================================================================================================================================================================
B=E R~}r{x~} =============================================================================================================================================================================
B=E=@ R~}r{x~}.u~.Rw~}xr.]t~pwxr._px} ================================================================================================================
B=E=A R~}r{x~}.u~.Ptxt.bx|{px~}.x}.WXe ============================================================================================================
B=E=B R~}r{x~}.u~.bpxrx.x}.\b ================================================================================================================================
B=E=C R~}r{x~}.u~.Pw|p ===============================================================================================================================================
B=E=D R~}r{x~}.u~.V{pr~|p ===========================================================================================================================================
B=F Stxv}.Rwp{{t}vt.u~.Ut.bsxt ===================================================================================================================================
B=F=@ bp|{t.bxt ==================================================================================================================================================================
B=F=A \pxyp}p.S~t.bp}spsxpx~} ==================================================================================================================================
B=F=B Prt.
=.Rw~}xr.cwtptxr.\pxyp}p.dt ===========================================================================================================
B=F=C b|~zx}v.p.p.a~t.~u.Ps|x}xpx~} =======================================================================================================================
B=F=D Sxuuxr{.x}.Q{x}sx}v.~u.Sv.R~}sxx~} ==================================================================================================================
B=F=E _x~.\pxyp}p.Ttxt}rt =========================================================================================================================================
B=F=F X}r{x~}.p}s.Tr{x~}.Rxtxp =================================================================================================================================
B=F=G ]|qt.~u.Ut|p{t.bqytr =========================================================================================================================================
Pt}sx.6cpq{t7 ==================================================================================================================================================================================
[x.~u.UxvtI.
Uxv[email protected]}xuxrpx~}.~u.bsxt.u~|._q\ts.btprw =============================================================================================================
[x.~u.cpq{tI.
cpq{[email protected]}s~|xts;.r~}~{{ts;.s~q{t<q{x}s.xp{.tp|x}x}v.|~zts.|pxyp}p.x}.tp|t}.~u.}t~pwxr.px} ========
cpq{t.AI.ap}s~|xts;.r~}~{{ts;.s~q{t<q{x}s.xp{.tp|x}x}v.|~zts.|pxyp}p.x}.tp|t}.~u.ptxt.x|{px~}.x}.
WXe>PXSb =====================================================================================================================================================================================
cpq{t.BI.ap}s~|xts;.r~}~{{ts;.s~q{t<q{x}s.px{. tp|x}x}v.|~zts.|pxyp}p.x}.tp|t}. ~u.pxrx.x}.\{x{t.
br{t~x =======================================================================================================================================================================================
cpq{t.CI.ap}s~|xts;.r~}~{{ts;.s~q{t<q{x}s.px{.tp|x}x}v.|~zts.|pxyp}p.x}.tp|t}.~u.x}p~r{p.tt.x}.
V{pr~|p ======================================================================================================================================================================================
cpq{t.DI.ap}s~|xts;.r~}~{{ts;.s~q{t<q{x}s.px{.tp|x}x}v.|~zts.|pxyp}p.x}.tp|t}.~u.pw|p =========================
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11137
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
F@.
FB.
FB.
FB.
FC.
FC.
FF.
FF.
FF.
FF.
G@.
GD.
GG.
GH.
H@.
H@.
H@.
HA.
HA.
HA.
HB.
HB.
HB.
HC.
HD.
HE.
HE.
HF.
HG.
HG.
@?B.
FE.
@?B.
@?G.
@@@.
@@A.
@@C.
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Gzgewvkxg!Uwooct{!
\pxyp}p.x.p.brwts{t.X.qp}rt.
}st.wt.R~}~{{ts.bqp}rt.Pr.
6RbP7=.brwts{t.X.x}sxrpt.p.wxvw.
~t}xp{.u~.pqt;.}~.rt}{.
prrtts.|tsxrp{.t.x}.wt.d}xts.
bpt;.p}s.p.{prz.~u.prrtts.put.u~.
t.}st.|tsxrp{.t
xx~}=.c~.spt;.
|pxyp}p.wp.}~.qtt}.qytr.~.p}.
p~
ts.}t.sv.p{xrpx~}.6]SP7.
wp.st|~}pt.x.put.p}s.tuuxrpr.
u~.p.trxuxr.x}sxrpx~}.}st.wt.U~~s.
Sv.p}s.R~|txr.Pr.6USRP7=.
]t
twt{t;.p.~u.^r~qt.A?@C;.
t}<wtt.pt.p}s.wt.Sxxr.~u.
R~{|qxp.wp
t.pts.pt<{t
t{.
|tsxrp{.|pxyp}p.{p.wp.p{{~.u~.
|pxyp}p.t.xwx}.wp.ptJ.x|x{p.
qx{{.pt.t}sx}v.x}.~wt.pt=.
cwt.tt}.t
xt.p.}stpzt}.
q.wt.U~~s.p}s.Sv.Ps|x}xpx~}.
6USP7.~.p}p{t.wt.r{x}xrp{.sxt.
q{xwts.x}.wt.|tsxrp{.{xtpt.
x}
txvpx}v.wt.t.~u.|pxyp}p.x}.p}.
wtptxr.ptp=.Ux;.t.sxr.wt.
r~}t.u~.wx.rxt}xuxr.t
xt=.]t;.
t.strxqt.wt.|tw~s.ts.x}.wx.
t
xt.~.xst}xu.pstpt.p}s.t{{<.
r~}~{{ts.sxt.t
p{px}v.wt.put.
p}s.tuuxrpr.~u.|pxyp}p.u~.pxr{p.
wtptxr.t=.
cwt.USP.r~}srts.p.t|pxr.
tprw.u~.q{xwts.sxt.x}.wt.
|tsxrp{.{xtpt.wp.|tt.wt.
strxqts.rxtxp.u~.s.stxv}.p}s.
~r~|t.|tpt.x~.~.Utqp.
A?@B=.fwx{t.}~.p.~u.~.t|pxr.
t
xt;.t.wp
t.r~}x}ts.~.~x}t{.
u~{{~.wt.{xtpt.qt~}s.wp.spt.u~.
qtt}.sxt=.bsxt.tt.
r~}xstts.~.qt.t{t
p}.~.wx.t
xt.
xu.wt.x}
txvp~.ps|x}xtts.
|pxyp}p.~.pxt}.xw.p.sxpv}~ts.
|tsxrp{.r~}sxx~}.x}.p.t{{<r~}~{{ts;.
s~q{t<q{x}s;.{prtq~<r~}~{{ts.
r{x}xrp{.xp{=.^u.wt.t{t
t}.sxt.wp.
|t.wt.rxtxp.u~.t
xt;.ux
t.sxuutt}.
wtptxr.ptp.tt.x}
txvptsI.
Ux
t.sxt.tp|x}ts.rw~}xr.
}t~pwxr.px}.
c~.sxt.tp|x}ts.ptxt.
x|{px~}.x}.w|p}.
x||}~stuxrxt}r.
x.6WXe7.
pxt}.
c~.sxt.tp|x}ts.v{pr~|p.
^}t.s.tp|x}ts.pxrx.p}s.
px}.x}.|{x{t.r{t~x.6\b7.
^}t.s.tp|x}ts.pw|p=.
U~.tprw.~u.wtt.t{t
t}.r{x}xrp{.
sxt;.x}u~|px~}.x.~
xsts.
tvpsx}v.wt.qytr.sxts;.wt.sv.
r~}sxx~}.tts.6x}r{sx}v.s~t.p}s.
|tw~s.~u.ps|x}xpx~}7;.~wt.sv.
ts.q.qytr.sx}v.wt.s;.wt.
wx~{~vxrp{.p}s.qytrx
t.|tpt.
r~{{trts;.wt.~r~|t.~u.wtt.
|tpt.r~|px}v.tp|t}.xw.
|pxyp}p.~.{prtq~;.p}s.wt.t~ts.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
p}s.~qt
ts.ps
tt.t
t}=.cwt.
r~}r{x~}.sp}.q.wt.x}
txvp~.
pt.wt}.strxqts;.p{~}v.xw.~t}xp{.
{x|xpx~}.~u.wtt.r~}r{x~}.qpts.
~}.wt.s.stxv}=.P.qxtu.||p.~u.
tprw.s .ux}sx}v.p}s.{x|xpx~}.x.
~
xsts.p.wt.t}s.~u.wt.trx~}=.
cwt.t{t
t}.r{x}xrp{.sxt.wp.|t.
wt.rxtxp.p}s.tt.t
p{pts.x}.wx.
t
xt.w~ts.~xx
t.xv}p{.wp.
|pxyp}p.|p.~srt.p.stxpq{t.
wtptxr.~r~|t;.}st.wt.trxuxr.
ttx|t}p{.r~}sxx~}.tts=.]~pq{;.
x.x.qt~}s.wt.r~t.~u.wx.t
xt.~.
stt|x}t.wtwt.wtt.spp.
st|~}pt.wp.|pxyp}p.wp.p.
rt}{.prrtts.|tsxrp{.t.x}.wt.
d}xts.bpt=.W~t
t;.wx.t
xt.
r~}r{st.wp.wtt.t{t
t}.r{x}xrp{.
sxt.t
t.p.~~u<~u<r~}rt.
sxt;.qpts.~}.wt.{x|xpx~}.~u.wtx.
s.stxv};.p.strxqts.x}.wt.s.
||pxt=._~~u<~u<r~}rt.sxt.
~
xst.t{x|x}p.t
xst}rt.~}.p.
~~ts.w~wtx.tvpsx}v.p.sv .
tuutr=.U~.sv.}st.st
t{~|t};.
wt.tuutr.~ut}.t{pt.~.p.w~<t|.
r{x}xrp{.~r~|t.qtx}v.x}
txvpts=.
_~~u<~u<r~}rt.sxt.t
t.p.wt.
{x}z.qttt}.tr{x}xrp{.sxt.p}s.
s~t.p}vx}v.r{x}xrp{.sxt=.cwttu~t;.
~~u<~u<r~}rt.sxt.pt.}~.
uuxrxt}.~.st|~}pt.tuuxrpr.~u.p.
sv.qtrpt.wt.~
xst.~}{.
t{x|x}p.x}u~|px~}.pq~.wt.
tuutr.~u.p.sv=.W~t
t;.wt.sxt.
t
xtts.~srts.~xx
t.t{;.
vvtx}v.|pxyp}p.w~{s.qt.uwt.
t
p{pts.p.p}.psy}r.tp|t}.u~.
}t~pwxr.px};.ptxt.x|{px~}.
x}.WXe.pxt};.p}s.pxrx.x}.\b.
pxt}=.
cwt.|px}.{x|xpx~}.xst}xuxts.x}.wt.
t{t
t}.sxt.tx}v.wt.|tsxrp{.
p{xrpx~}.~u.|pxyp}p.pt.{xts.
qt{~I.
cwt.|p{{.}|qt.~u.qytr.
t}~{{ts.x}.wt.sxt;.wxrw.{x|x.wt.
pxxrp{.p}p{t.~u.put.p}s.
tuuxrpr=.
cwt.t
p{px~}.~u.|pxyp}p.~}{.
put.prt.ps|x}xpx~}.x}.wt.sxt;.
wxrw.{x|x.wt.pqx{x.~.stt|x}t.
tuuxrpr.u~{{~x}v.rw~}xr.
ps|x}xpx~}=.
cwt.ps|x}xpx~}.~u.|pxyp}p.
xrp{{.w~vw.|~zx}v;.wxrw.
t~t.x{{.pxt}.~.r~|qts.
|ptxp{.p}s.x}~srt.~q{t|.xw.
stt|x}x}v.wt.s~t.st{x
tts=.
cwt.~t}xp{.u~.qytr.~.
xst}xu.wtwt.wt.trtx
ts.
|pxyp}p.~.{prtq~;.wxrw.qtpz.wt.
q{x}s.~u.wt.sxt=.
cwt.|p{{.}|qt.~u.rp}}pqx}~xs.
}p'¶
t.qytr;.wxrw.{x|x.wt.pqx{x.
~.stt|x}t.put.p}s.~{tpqx{x.x}.
wtt.qytr=.
RQ!11111
Hto!11138
Hov!5812
Uhov!5813
64824!
cwt.{~.}|qt.~u.ut|p{t.qytr;.
wxrw.|pzt.x.sxuuxr{.~.vt}tp{xt.
wt.s.ux}sx}v.~.qytr.~u.q~w.
vt}st=.
cw;.wx.t
xt.sxrt.wt.
u~{{~x}v.|tw~s~{~vxrp{.rwp}vt.wp.
|p.qt.|pst.x}.~st.~.t~{
t.wtt.
{x|xpx~}.p}s.x|~
t.wt.stxv}.~u.
ut.sxt.wxrw.tp|x}t.wt.put.
p}s.tuuxrpr.~u.|pxyp}p.u~.trxuxr.
wtptxr.x}sxrpx~}I.
Stt|x}t.wt.p~xpt.}|qt.
~u.qytr.sxts.qpts.~}.
tr~||t}spx~}.x}.
px~.USP.
Iwkfcpegu!hqt!Kpfwuvt{!tvpsx}v.wt.
r~}sr.~u.r{x}xrp{.xp{.u~.trxuxr.
|tsxrp{.x}sxrpx~}=.
Ps|x}xt.r~}xt}.p}s.
t~srxq{t.s~t.~u.|pxyp}p.qpts.
~}.tr~||t}spx~}.x}.wt.USP.
Iwkfcpeg!hqt!Kpfwuvt{<!Dqvcpkecn!Ftwi!
Rtqfwevu!)3115*/ .
T
p{pt.wt.tuutr.~u.|pxyp}p.
}st.wtptxr.r~}sxx~}.u~{{~x}v.
q~w.prt.p}s.rw~}xr.ps|x}xpx~}=.
R~}xst.p{t}px
t.~.|~zts.
|pxyp}p.6g/i/-!
p~xpx~}7=.
Psst.p}s.x|~
t.wt}t
t.
~xq{t.wt.sxuuxr{.x}.q{x}sx}v.~u.
|pxyp}p.p}s.{prtq~.tp|t}.x}.
r{x}xrp{.sxt=.
T
p{pt.wt.tuutr.~u.x~.
ttxt}rt.xw.|pxyp}p.xw.tvps.
~.wt.put.p}s.~{tpqx{x.~u.
|pxyp}p=.
bx
t.u~.vt}st.qp{p}rt.x}.wt.
qytr.ts.x}.sxt=.
X}.r~}r{x~};.wt.t{t
t}.r{x}xrp{.
sxt.r~}srts.~.spt.s~.}~.|tt.
wt.rxtxp.txts.q.wt.USP.~.
stt|x}t.xu.|pxyp}p.x.put.p}s.
tuutrx
t.x}.trxuxr.wtptxr.ptp=.
W~t
t;.wt.sxt.rp}.t
t.p.~~u<.
~u<r~}rt.sxt.p}s.~.uwt.
ttprw.x}~.wt.t.~u.|pxyp}p.x}.
wtt.wtptxr.x}sxrpx~}=.
Pssxx~}p{{;.wt.r{x}xrp{.~r~|t.spp.
p}s.ps
tt.t
t}.~ux{t.t~ts.x}.
wtt.q{xwts.sxt.rp}.qt}tuxrxp{{.
x}u~|.w~.ut.ttprw.x}.wx.ptp.
x.r~}srts=.Ux}p{{;.p{xrpx~}.~u.wt.
tr~||t}spx~}.{xts.pq~
t.q.
x}
txvp~.wt}.stxv}x}v.ut.
sxt.r~{s.vtp{.x|~
t.wt.
p
px{pq{t.r{x}xrp{.spp.wp.rp}.qt.ts.
~.stt|x}t.xu.|pxyp}p.wp.
p{xspts.
p}s.t{xpq{t.|tsxrp{.p{xrpx~}=.
2/!Kpvtqfwevkqp!
X}.t~}t.~.rxxt}.txx~}.
q|xts.~.wt.Sv.T}u~rt|t}.
Ps|x}xpx~}.6STP7.ttx}v.STP.
~.trwts{t.|pxyp}p;.wt.STP.
Ps|x}xp~.ttts.wp.wt.d=b=.
Stp|t}.~u.Wtp{w.p}s.W|p}.
cwx.Vxsp}rt.x.p
px{pq{t.~}.wt.x}t}t.p.
jvvr<00yyy/hfc/iqx0Ftwiu0fghcwnv/jvo!}st.
Vxsp}rt.6Sv7=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64825!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
bt
xrt.6WWb7.~
xst.p.rxt}xuxr.p}s.
|tsxrp{.t
p{px~}.~u.wt.p
px{pq{t.
x}u~|px~}.p}s.p.rwts{x}v.
tr~||t}spx~}.u~.|pxyp}p;.x}.
prr~sp}rt.x[email protected]=b=R=.G@@6q7=.cwt.
btrtp.~u.WWb.x.txts.~.r~}xst.
x}.p.rxt}xuxr.p}s.|tsxrp{.t
p{px~}.
txvw.upr~.stt|x}px
t.~u.r~}~{.
}st.wt.R~}~{{ts.bqp}rt.Pr.
6RbP7=.Ps|x}xpx
t.t~}xqx{xxt.
u~.t
p{px}v.p.qp}rt.u~.r~}~{.
}st.wt.RbP.pt.tu~|ts.q.wt.
U~~s.p}s.Sv.Ps|x}xpx~}.6USP7;.
xw.wt.r~}rt}rt.~u.wt.]px~}p{.
X}xt.~}.Sv.Pqt.6]XSP7=._p.~u.
wx.t
p{px~}.x}r{st.p}.pt|t}.
~u.wtwt.|pxyp}p.wp.p.rt}{.
prrtts.|tsxrp{.t.x}.wt.d}xts.
bpt=.cwx.pt|t}.}trtxpts.p.
t
xt.~u.wt.p
px{pq{t.spp.u~|.
q{xwts.r{x}xrp{.sxt.~.stt|x}t.
wtwt.wtt.x.pstpt.rxt}xuxr.
t
xst}rt.~u.|pxyp}p .tuutrx
t}t=.
d}st.btrx~}.A?A.~u.wt.RbP;.
|pxyp}p.x.rt}{.r~}~{{ts.p.p.
brwts{t.X.qp}[email protected]=b=R=.G@A7=.
brwts{t.X.x}r{st.w~t.qp}rt.
wp.wp
t.p.wxvw.~t}xp{.u~.pqt;.
wp
t.}~.rt}{.prrtts.|tsxrp{.t.
x}.tp|t}.x}.wt.d}xts.bpt;.p}s.
{prz.prrtts.put.u~.t.}st.
|tsxrp{.t
xx~}[email protected]=b=R=.
µ G@A6q76@76P7¤6R77=.
P.sv.~sr.wxrw.wp.qtt}.
p~
ts.q.USP.u~.|pztx}v.x}.wt.
d}xts.bpt.x.r~}xstts.~.wp
t.p.
rt}{.prrtts.|tsxrp{.t= .
\pxyp}p.x.}~.p}.USP<p~
ts.sv.
~sr;.p.p.]t.Sv.P{xrpx~}.
6]SP7.~.Qx~{~vxr.[xrt}t.p{xrpx~}.
6Q[P7.u~.|pxyp}p.wp.}~.qtt}.
p~
ts.q.USP=.W~t
t;.USP.
p~
p{.~u.p}.]SP.x.}~.wt.~}{.
|tp}.w~vw.wxrw.p.sv.rp}.wp
t.
p.rt}{.prrtts.|tsxrp{.t.x}.wt.
d}xts.bpt=.
X}.vt}tp{;.p.sv.|p.wp
t.p.
rt}{.prrtts.|tsxrp{.t .x}.wt.
d}xts.bpt.xu.wt.sv.|tt.p.ux
t<.
p.t=.Tpq{xwts.rpt.{p.6Cnnkcpeg!
hqt!Ecppcdku!Vjgtcrgwvkeu!
=.FGC-!@D.
U=Bs.@@B@;.@@BD.6S=R=.Rx=.@HHC77.wt{s.
wt.Ps|x}xp~.~u.STP .p{xrpx~}.
~u.wt.ux
t<p.t.~.stt|x}t.
wtwt.p.sv.wp.p.rt}{.
prrtts.|tsxrp{.t= .cwt.u~{{~x}v.
strxqt.wt.ux
t.t{t|t}.wp.
rwpprtxt.rt}{.prrtts.
|tsxrp{.t .u~.p.svI .
x=.wt.sv .rwt|x.|.qt.z}~}.
p}s.t~srxq{t=.
cwt.qp}rt .rwt|x.|.qt.
rxt}xuxrp{{.tpq{xwts.~.t|x.x.~.
qt.t~srts.x}~.s~pvt.wxrw.rp}.
qt.p}spsxts=.cwt.{xx}v.~u.wt.
qp}rt.x}.p.rt}.tsxx~}.~u.~}t.~u.
wt.~uuxrxp{.r~|t}sxp;.p.stux}ts.q.
DF.Ua.@?CHH;.@?D?C¤?E.6\p•rw.AE;.@HHA7=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
trx~}.A?@6y7.~u.wt.U~~s;.Sv.p}s.
R~|txr.Pr;[email protected]=b=R=.BA@6y7;.x.
uuxrxt}.~.|tt.wx.txt|t}= .
xx=.wtt.|.qt.pstpt.put.
sxt=.
cwtt.|.qt.pstpt.
wp|pr~{~vxrp{.p}s.~xr~{~vxrp{.
sxt;.s~}t.q.p{{.|tw~s.tp~}pq{.
p{xrpq{t;.~}.wt.qpx.~u.wxrw.x.
r~{s.upx{.p}s.t~}xq{.qt.
r~}r{sts;.q.tt.p{xuxts.q.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
sv;.wp.wt.qp}rt.x.put.u~.
tpx}v.p.trxuxr;.tr~v}xts.sx~st= .
xxx=.wtt.|.qt.pstpt.p}s.t{{<.
r~}~{{ts.sxt.~
x}v.tuuxrpr=.
cwtt.|.qt.pstpt;.t{{<.
r~}~{{ts;.t{{<stxv}ts;.t{{<.
r~}srts;.p}s.t{{<s~r|t}ts.
sxt;.x}r{sx}v.r{x}xrp{.
x}
txvpx~};.q.tt.p{xuxts.q.
rxt}xuxr.px}x}v.p}s.ttxt}rt.~.
t
p{pt.wt.put.p}s.tuutrx
t}t.~u.
sv;.~}.wt.qpx.~u.wxrw.x.r~{s.qt.
upx{.p}s.t~}xq{.r~}r{sts.q.
rw.tt.wp.wt.qp}rt.x{{.
wp
t.wt.x}t}sts.tuutr.x}.tpx}v.p.
trxuxr;.tr~v}xts.sx~st= .
x
=.wt.sv.|.qt.prrtts.q.
p{xuxts.tt=.
cwt.sv.wp.p.]t.Sv.
P{xrpx~}.6]SP7.p~
ts.q.wt.
U~~s.p}s.Sv.Ps|x}xpx~};.
p}.~.wt.U~~s;.Sv.p}s.
R~|txr.Pr;[email protected]=b=R=.BDD=.^;.p.
r~}t}.~u.wt.}px~}p{.r~||}x.~u.
tt;.p{xuxts.q.rxt}xuxr.px}x}v.
p}s.ttxt}rt.~.t
p{pt.wt.put.
p}s.tuutrx
t}t.~u.sv;.prrt.wt.
put.p}s.tuutrx
t}t.~u.wt.qp}rt.
u~.t.x}.tpx}v.p.trxuxr;.tr~v}xts.
sx~st=.P.|ptxp{.r~}u{xr.~u.~x}x~}.
p|~}v.tt.tr{st.p.ux}sx}v.~u.
r~}t}= .p}s.
=.wt.rxt}xuxr.t
xst}rt.|.qt.
xst{.p
px{pq{t=.
X}.wt.pqt}rt.~u.]SP.p~
p{;.
x}u~|px~}.r~}rt}x}v.wt.rwt|x;.
wp|pr~{~v;.~xr~{~v;.p}s.
tuutrx
t}t.~u.wt.qp}rt.|.qt.
t~ts;.q{xwts;.~.~wtxt.
xst{.p
px{pq{t;.x}.uuxrxt}.stpx{.~.
t|x.tt;.p{xuxts.q.rxt}xuxr.
px}x}v.p}s.ttxt}rt.~.t
p{pt.wt.
put.p}s.tuutrx
t}t.~u.sv;.~.
upx{.p}s.t~}xq{.r~}r{st.wt.
qp}rt.x.put.p}s.tuutrx
t.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st= .
^}t.p.~.p.wt.ux
t<p.t.u~.
wp
x}v.rt}{.prrtts.|tsxrp{.
t .x.w~vw.q|xx~}.~u.p}.]SP.
~.Q[P.wxrw.x.p~
ts.q.USP=.
W~t
t;.USP.p~
p{.~u.p}.]SP.~.
Q[P.x.}~.txts.u~.p.sv.~.p.
wt.ux
t<p.t=.
cwx.t
xt.u~rt.~}.USP .p}p{x.
~u.~}t.t{t|t}.~u.wt.ux
t<p.t.u~.
stt|x}x}v.wtwt.p.sv.wp.
RQ!11111
Hto!11139
Hov!5812
Uhov!5813
rt}{.prrtts.|tsxrp{.t =.
btrxuxrp{{;.wt.tt}.t
xt.ptt.
wt.Bs.rxtx~}.wp.psstt.wtwt.
|pxyp}p.wp.pstpt.p}s.t{{<.
r~}~{{ts.sxt.~
x}v.tuuxrpr =.
cw;.wx.t
xt.t
p{pt.q{xwts.
r{x}xrp{.sxt.wp.wp
t.qtt}.
r~}srts.x}v.|pxyp}p.x}.qytr.
w~.wp
t.p.
pxt.~u.|tsxrp{.
r~}sxx~}.q.ptx}v.wt.pstpr.~u.
wt.||pxts.s.stxv}.p}s.wt.
s.spp=.cwt.|tw~s~{~v.u~.
t{trx}v.wt.sxt.wp.tt.t
p{pts.
x.st{x}tpts.qt{~=.
USP .t
p{px~}.p}s.r~}r{x~}.
tvpsx}v.wt.t|px}x}v.u~.rxtxp.u~.
wtwt.|pxyp}p.wp.p.rt}{.
prrtts.|tsxrp{.t; .p.t{{.p.wt.
txvw.upr~.tpx}x}v.~.wt.
rwts{x}v.~u.|pxyp}p;.pt.~xst.wt.
r~t.~u.wx.t
xt=.P.stpx{ts.
sxrx~}.~u.wtt.upr~.x.r~}px}ts.
x}.USP .rxt}xuxr.p}s.|tsxrp{.
t
p{px~}.~u.|pxyp}p=.
3/!Ogvjqfu!
cwt.|tw~s.u~.t{trx}v.wt.sxt.
~.x}r{st.x}.wx.t
xt.x}
~{
ts.wt.
u~{{~x}v.t;.wxrw.pt.strxqts.x}.
stpx{.x}.wt.qtrx~}.qt{~I.
@=.Stux}t.wt.~qytrx
t.~u.wt.t
xt=.
A=.Stux}t.|pxyp}p .x}.~st.~.
uprx{xpt.wt.|tsxrp{.{xtpt.tprw.
u~.sxt.wp.ps|x}xtts.wt.
qp}rt;.
B=.Stux}t.pstpt.p}s.t{{<.
r~}~{{ts.sxt .x}.~st.~.uprx{xpt.
wt.tprw.u~.t{t
p}.spp.p}s.
{xtpt;.
C=.btprw.|tsxrp{.{xtpt.sppqpt.
p}s.xst}xu.t{t
p}.pstpt.p}s.t{{<.
r~}~{{ts.sxt;.p}s.
D=.at
xt.p}s.p}p{t.wt.pstpt.
p}s.t{{<r~}~{{ts.r{x}xrp{.sxt.~.
stt|x}t.xu.wt.st|~}pt.tuuxrpr.
~u.|pxyp}p.u~.p}.wtptxr.
x}sxrpx~}=.
3/2 Fghkpg!vjg!Qdlgevkxg!qh!vjg!Tgxkgy!
cwt.~qytrx
t.~u.wx.t
xt.x.~.
pt.wt.s.stxv}.p}s.t{x}v.
spp.u~|.r{x}xrp{.sxt.q{xwts.x}.
wt.|tsxrp{.{xtpt.wp.tt.
r~}srts.xw.|pxyp}p.6p.stux}ts.
qt{~7.p.p.tp|t}.u~.p}.
wtptxr.x}sxrpx~};.x}.~st.~.
stt|x}t.xu.wt.|tt.wt.rxtxp.~u.
pstpt.p}s.t{{<r~}~{{ts.sxt.
~
x}v.tuuxrpr =.
3/3 Fghkpg!Octklwcpc!
X}.wx.t
xt;.wt.t|.|pxyp}p .
tut.~.wt.u{~tx}v.~.~.{tp
t.~u.
wt.Ecppcdku!{p}=.cwtt.tt.}~.
txrx~}.~}.wt.~t.~u.
ps|x}xpx~}.ts.u~.|pxyp}p.x}.wt.
sxt=.
bsxt.wxrw.ps|x}xtts.
x}sx
xsp{.rp}}pqx}~xs.6wtwt.
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
wt}.wt.sxt.qtvx};.p}s.x.spts.
w~vw~.wt.s=.X}.~|t.rpt;.
t{.~u.wt.s.~.t{x}v.
q{xrpx~}.rxpx~}.pt.q|xts.~.
wt.ftq.xt.put.wt.s.t}s.
6jvvru<00enkpkecnvtkcnu/iqx0ev30cdqwv.ukvg0
dcemitqwpf7=.
EnkpkecnVtkcnu/iqx!p.tprwts.u~.p{{.
sxt.ps|x}xtx}v.|pxyp}p=.cwt.
t{.~u.wx.tprw.tt.ts.~.
r~}ux|.wp.}~.r~|{tts.sxt.xw.
q{xwts.spp.tt.|xts.x}.wt.
3/4 Fghkpg!Cfgswcvg!cpf!Ygnn.!
{xtpt.tprw=.Sx}v.wt.{xtpt.
Eqpvtqnngf!Enkpkecn!Uvwfkgu!
tprw;.tutt}rt.u~}s.x}.t{t
p}.
cwt.rxtxp.u~.p}.pstpt.p}s.
sxt.p}s.t|pxr.t
xt.tt.
3/5
Ugctej!Ogfkecn!Nkvgtcvwtg!
t{{<r~}~{{ts.s .u~.~t.~u.
t
p{pts.u~.pssxx~}p{.t{t
p}.
stt|x}x}v.wt.put.p}s.tuuxrpr.~u.p. Fcvcdcugu!cpf!Kfgpvkh{!Tgngxcpv!Uvwfkgu! rxpx~}=.P{{.{p}vpvt.tt.x}r{sts.
w|p}.sv.x.stux}ts.}st.wt.R~st.
ft.xst}xuxts.p}s~|xts;.s~q{t<.
x}.wt.tprw=.cwt._q\ts.tprw.
~u.Utstp{.atv{px~}.6RUa7.x}[email protected]. q{x}s;.{prtq~<r~}~{{ts.r{x}xrp{.
xt{sts.p.~p{.~u.DEE.pqpr= ^u.
B@C=@AE=.cwt.t{t|t}.~u.p}.pstpt.
sxt.r~}srts.xw.|pxyp}p.~.
wtt.pqpr;.p.u{{<t.t
xt.p.
p}s.t{{<r~}~{{ts.s.p.strxqts.
pt.|pxyp}p .tuuxrpr.x}.p}.
r~}srts.xw.GD.pt.~.pt.
x}[email protected]@[email protected]}.qt.||pxts.
wtptxr.x}sxrpx~}=.c~.x|p.
t{xvxqx{x=.U~|.wx.t
p{px~};.~}{.
p.u~{{~I.
|tsxrp{.{xtpt.sppqpt.tt.
t{t
t}.~u.GD.sxt.|t.wt.E.RUa.
@=.cwt.|px}.~qytrx
t.|.qt.~.
tprwts.u~.p{{.sxt.~ts.~.wt.
t{t|t}.u~.x}r{x~}.p.pstpt.p}s.
pt.p.wtptxrp{{.t{t
p}.
sppqpt.x~.~.Utqp.A?@BI .
t{{<r~}~{{ts.sxt=.
~r~|t=.
RwdOgf<!_q\ts.x.p.sppqpt.~u.
Uxv[email protected]{~7.~
xst.p}.~
t
xt.
A=.cwt.s.|.qt.{prtq~<.
q{xwts.|tsxrp{.p}s.rxt}xuxr.sxt. ~u.wt.~rt.ts.~.xst}xu.sxt.
r~}~{{ts=.
wp.x.|px}px}ts.q.wt.d=b=.]px~}p{.
u~|.wt._q\ts.tprw=.cwt.t{t
t}.
B=.cwt.qytr.|.p{xu.p.wp
x}v. [xqp.~u.\tsxrx}t.6][\7.p.]XW.p.p.
sxt.t
xtts.tt.q{xwts.
wt.|tsxrp{.r~}sxx~}.qtx}v.sxts=.
p.~u.wt.T}t.t|.~u.x}u~|px~}. qttt}[email protected]}s.A?@B=.ct}.~u.wtt.
C=.cwt.s.stxv}.t|x.p.
p{xs.
txt
p{=._q\ts.r~|xt.|~t.wp}. sxt.tt.r~}srts.x}.wt.d}xts.
r~|px~}.xw.p}.p~xpt.r~}~{. AC.|x{{x~}.rxpx~}.u~.qx~|tsxrp{.
bpt.p}s.~}t.s.p.r~}srts.x}.
r~}sxx~}=.
{xtpt.u~|.\TS[X]T;.{xut.rxt}rt.
Rp}psp=.cwtt.t{t
t}.sxt.tp|x}ts.
D=.cwt.pxv}|t}.~u.qytr.~.
y~}p{;.p}s.~}{x}t.q~~z.6jvvr<00
wt.tuutr.~u.|~zts.p}s.
p~xts.
tp|t}.p}s.r~}~{.v~.|.qt.
yyy/pedk/pno/pkj/iqx0rwdogf7=.
|pxyp}p.u~.wt.x}sxrpx~}.~u.rw~}xr.
p}s~|xts=.
EnkpkecnVtkcnu/iqx<!
}t~pwxr.px};.pxrx.t{pts.~.
E=.cwtt.x.|x}x|xpx~}.~u.qxp.
R{x}xrp{cxp{=v~
.x.p.sppqpt.~u.
|{x{t.r{t~x.6\b7;.ptxt.
w~vw.wt.t.~u.p.s~q{t<q{x}s.s. q{xr{.p}s.x
pt{.~ts.
x|{px~}.x}.pxt}.xw.w|p}.
stxv}=.
r{x}xrp{.sxt.wp.x.|px}px}ts.q.
x||}~stuxrxt}r.
x.6WXe7;.
F=.cwt.s.t~.r~}px}.p.u{{.
wt.][\=.X}u~|px~}.pq~.wt.r{x}xrp{. v{pr~|p;.p}s.pw|p=.P{{.x}r{sts.
~~r~{.p}s.x|p.spp=.
sxt.x.~
xsts.q.wt.b~}~.~.
sxt.ts.ps{.pxt}.p.qytr=.
G=.P}p{x.~u.wt.s.spp.x.
_x}rxp{.X}
txvp~.~u.wt.s=.
P{{.sxt.r~}srts.x}.wt.d}xts.
p~xpt{.r~}srts=.
X}u~|px~}.pq~.wt.sxt.x.
bpt.tt.r~}srts.}st.p}.X]S.p.
P.}~ts.pq~
t;.wt.rt}.t
xt.
q|xts.~.wt.ftq.xt.6tvxtts 7. _wpt.A.x}
txvpx~}=.
tp|x}t.~}{.w~t.spp.p
px{pq{t.x}.
wt.q{xr.s~|px}.p}s.w.t{xt.~}.
fwx{t.}~.p.t|pxr.t
xt;.t.wp
t.
cwt.u~{{~x}v.tprw.ptv.p.ts;.
u~{{~ts.wt.trt}.q{xwts.{xtpt.~}.
6rp}}pqx.^a.|pxyp}p7.P]S.6wtptxr.t.^a.
r{x}xrp{.sxt.q{xwts.x}.wt.
wtp7.P]S.6aRc.^a.p}s~|xts.r~}~{{ts.xp{.
|tsxrp{.{xtpt=._q{xwts.sxt.q. |pxyp}p.t.u~.~xq{t.wtptxr.~t.
p}s;.p.~u.Yp}p.A?@D;.t.u~}s.~}{.~}t.}t.
^a.t|pxr.t
xt .^a.r{x}xrp{.xp{.^a.
wtx.}pt.pt.||pxt.wp.s~.}~.
s.wp.~{s.|tt.~.rxtxp.6]pup{x.t.p{=;.
r{x}xrp{.xp{7.]^c.6|pxyp}p.pqt .j\twl.^a.
x}r{st.wt.{t
t{.~u.stpx{.txts.q.
A?@B7=.cwx.s.tp|x}ts.wt.tuutr.~u.|~zts.
pssxrx
t.qtwp
x~.^a.qp}rt.t{pts.
sxt.q|xts.~.USP.x}.p}.]SP=.
|pxyp}p.~}.R~w} .sxtpt=.
sx~st7 =.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
ttx|t}p{.qp}rt.~.|pztts.
sv.~sr7.~.|pxyp}p.tpr.
tt.tr{sts.u~|.wx.t
xt=.
Pssxx~}p{{;.sxt.~u.ps|x}xtts.
}tp{.{p}.|ptxp{.~.{prtq~.
|pxyp}p.6|pxyp}p.xw.p{{.
rp}}pqx}~xs.tprts7.wp.wps.
qtt}{.qtt}.{t|t}ts.q.wt.
pssxx~}.~u.trxuxr.p|~}.~u.cWR.~.
~wt.rp}}pqx}~xs.tt.p{~.tr{sts.
6Rwp}v.t.p{=;.@HFH7=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
fwx{t.wt.|py~x.~u.wt.t{t|t}.
stux}x}v.p}.pstpt.p}s.t{{<.
r~}~{{ts.s.rp}.qt.pxuxts.
w~vw.p.q{xwts.pt.6t{t|t}.
1@¤E7;.wtt.pt.~.t{t|t}.wp.
rp}}~.qt.|t.q.p.s.q{xwts.x}.
wt.|tsxrp{.{xtptI.t{t|t}.1F.
6p
px{pqx{x.~u.p.s.t~.xw.u{{.
~~r~{.p}s.x|p.spp7.p}s.t{t|t}.
1G.6p.stt|x}px~}.~u.wtwt.wt.spp.
p}p{x.p.p~xpt7=.cw;.u~.
~t.~u.wx.t
xt;.~}{.t{t|t}.
1@¤E.x{{.qt.ts.~.p{xu.p.s.p.
qtx}v.pstpt.p}s.t{{<r~}~{{ts=.
64826!
RQ!11111
Hto!1113;
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
c~.p{xux}v.sxt;.wxrw.
ptts.|pxyp}p.u~.v{pr~|p;.tt.
t
x~{.t
xtts.x}.wt.@HHH.
X}xt.~u.\tsxrx}t.6X^\7.t~.
t}x{ts.\pxyp}p.p}s.\tsxrx}tI.
Ptx}v.wt.brxt}rt.Qpt = ft.sxs.
~.~}.p}p{x.~u.wtt.~.sxt.
p}s.r~}rts.xw.wt.r~}r{x~}.x}.
wt.X^\.t~=.cw;.p.stpx{ts.
sxrx~}.~u.wt.~.v{pr~|p.sxt.
x.}~.x}r{sts.x}.wt.tt}.t
xt=.
cwt.tt}.t
xt.~}{.sxrt.H.~u.
wt.xst}xuxts.@@.sxt=.U~.p.||p.
~u.wt.s.stxv}.u~.p{{.t{t
t}.
p{xux}v.sxt;.tt.cpq{t.@¤D.
6{~rpts.x}.wt.Pt}sx7=.
Qpts.~}.wt.t{trx~}.rxtxp.u~.
t{t
p}.sxt.strxqts.x}.btrx~}.A=B.
6Stux}t.Pstpt.p}s.ft{{<R~}~{{ts.
R{x}xrp{.bsxt7;.p.}|qt.~u.r{x}xrp{.
sxt.wp.x}
txvpts.|pxyp}p;.p.
stux}ts.x}.wx.t
xt;.tt.tr{sts.
u~|.wx.t
xt=.bsxt.wp.tp|x}ts.
wt.tuutr.~u.|pxyp}p.x}.wtp{w.
qytr.tt.tr{sts.qtrpt.wt.sxs.
}~.t.p.pxt}.~{px~}.xw.p.
|tsxrp{.r~}sxx~}.6U{~|.t.p{=;.@HFDJ.
U~{x}.t.p{=;[email protected]~{x}.t.p{=;.@HGGJ.
Wx{{.t.p{=;.@HFCJ.\x{tx}.t.p{=;.@HFCJ.
\x{tx}.t.p{=;[email protected]~stp{|.t.p{=;.
[email protected]{{prt.t.p{=;.A??FJ.Vtt}p{s.
p}s.bxt;.A???7=.P.@HFD.s.q.
cpwzx}.t.p{=.p.tr{sts.qtrpt.x.
wps.p.x}v{t<q{x}s;.pwt.wp}.s~q{t<.
q{x}s;.s.stxv}=.c~.~wt.sxt.
tt.tr{sts.qtrpt.wt.x|p.
~r~|t.|tpt.ptts.put.pwt.
wp}.p.wtptxr.~r~|t.6Vtt}qtv.
t.p{=;[email protected]p|.t.p{=;.A??B7=.
X}.Yp}p.@HHF;.wt.fwxt.W~t.^uuxrt.~u.
]px~}p{.Sv.R~}~{._~{xr.6^]SR_7.ttts.
wp.wt.X^\.r~}sr.p.t
xt.~u.wt.rxt}xuxr.
t
xst}rt.~.pt.wt.~t}xp{.wtp{w.qt}tux.p}s.
xz.~u.|pxyp}p.p}s.x.r~}xt}.rp}}pqx}~xs=.
X}u~|px~}.u~.wx.s.p.vpwtts.w~vw.
rxt}xuxr.~zw~;.xt.
xx.~.rp}}pqx.qt .
r{q.p}s.WXe>Prxts.X||}~stuxrxt}r.
b}s~|t.6PXSb7.r{x}xr;.p}p{x.~u.wt.t{t
p}.
rxt}xuxr.{xtpt;.p}s.tt}x
t.r~}{px~}.xw.
qx~|tsxrp{.p}s.~rxp{.rxt}x=.cwt.t~.p.
ux}p{xts.p}s.q{xwts.x}.@HHH=.
3/6 Tgxkgy!cpf!Cpcn{|g!Swcnkh{kpi!
Enkpkecn!Uvwfkgu!
`p{xuxts.r{x}xrp{.sxt.wp.
t
p{pts.|pxyp}p.u~.wtptxr.
~t.tt.tp|x}ts.x}.t|.~u.
pstpr.~u.s.stxv}.x}r{sx}v.
|tw~s.~u.sv.ps|x}xpx~};.s.
xt;.p}s.qytr.x}r{x~}.p}s.
tr{x~}.rxtxp=.Pssxx~}p{{;.wt.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11141
Hov!5812
Uhov!5813
|tpt.p}s.|tw~s.~u.p}p{x.ts.
x}.wt.sxt.~.pt.wt.tp|t}.
tuutr.tt.tp|x}ts=.
4/!Tguwnvu!cpf!Fkuewuukqp!
cwt.t{t
t}.p{xux}v.sxt.x}.wx.
t
xt.ptts.p.
pxt.~u.wtptxr.
x}sxrpx~}=.X}.~st.~.qtt.uprx{xpt.
p}p{x.p}s.sxrx~}.~u.wt.sxt;.
wt.u~{{~x}v.trx~}.v~.wt.sxt.
q.wtptxr.ptp=.fxwx}.tprw.
trx~};.tprw.x}sx
xsp{.s.x.
||pxts.x}.t|.~u.x.stxv};.
~r~|t.spp.p}s.x|~p}.{x|xpx~}=.
cwx.x}u~|px~}.x.p{~.~
xsts.x}.wt.
Pt}sx.x}.pq{p.u~|.u~.tprw.
s=.
4/2 Pgwtqrcvjke!Rckp!
Ux
t.p}s~|xts;.s~q{t<q{x}s;.
{prtq~<r~}~{{ts._wpt.A.r{x}xrp{.
sxt.wp
t.qtt}.r~}srts.~.tp|x}t.
wt.tuutr.~u.x}wp{ts.|pxyp}p.|~zt.
~}.}t~pwxr.px}.p~rxpts.xw.
WXe<t}~.}t~pw.6Pqp|.t.p{=;.
A??FJ.T{{x.t.p{=;.A??H7.p}s.rw~}xr.
}t~pwxr.px}.u~|.|{x{t.rpt.
6fx{t.t.p{=;.A??GJ.fpt.t.p{=;.A?@?J.
fx{t.t.p{=;.A?@B7=.cpq{t.@.~u.wt.
Pt}sx.||pxt.wtt.sxt=.
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/118>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64827!
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
B=@=@ ]t~pwxr._px}.P~rxpts.
xw.WXe<bt}~.]t~pw.
c~.sxt.tp|x}ts.wt.tuutr.~u.
|pxyp}p.~.tsrt.wt.px}.x}srts.q.
WXe<t}~.}t~pw=.
Pqp|.t.p{=.6A??F7.r~}srts.wt.
ux.s.t}x{ts;.Rp}}pqx.x}.px}u{.
WXe<p~rxpts.t}~.}t~pwI.P.
p}s~|xts.{prtq~<r~}~{{ts.xp{ =.
cwt.qytr.tt.D?.ps{.pxt}.xw.
}r~}~{{ts.WXe<p~rxpts.t}~.
}t~pw;.w~.wps.p.{tp.E.
ttxt}rt.xw.|~zx}v.|pxyp}p=.
cwt.qytr.tt.{x.x}~.~.pp{{t{.
v~.~u.AD.qytr.tprw=.\~t.wp}.
EG3.~u.qytr.tt.rt}.|pxyp}p.
t;.q.p{{.x}sx
xsp{.tt.txts.
~.sxr~}x}t.x}v.|pxyp}p.x~.~.
wt.s=.\~.qytr.tt.pzx}v.
|tsxrpx~}.u~.px}.sx}v.wt.s;.
xw.wt.|~.r~||~}.|tsxrpx~}.
qtx}v.~x~xs.p}s.vpqpt}x}=.d~}.
t}.x}~.wt.s;.qytr.wps.p}.
p
tpvt.spx{.px}.r~t.~u.p.{tp.B?.~}.
p.?¤@??.
xp{.p}p{~v.rp{t.6ePb7=.
bqytr.tt.p}s~|xts.~.trtx
t.
txwt.|~zts.|pxyp}p.6B=DE3.
cWR 7.~.|~zts.{prtq~.rxvptt.
wtt.x|t.t.sp.u~.D.sp;.x}v.p.
p}spsxts.rts.|~zx}v.~rtstI.
[email protected].tr~}s.x}wp{t;.6A7.@?.tr~}s.
w~{sx}v.|~zt.x}.wt.{}v;.6B7.C?.
tr~}s.twp{t.p}s.qtpwx}v.}~|p{{.
qttt}.uu=.cwt.pw~.sxs.}~.
trxu.w~.|p}.uu.wt.qytr.
|~zts.p.tprw.|~zx}v.tx~};.q.
wt.pts.wp.~}t.rxvptt.p.
|~zts.t.|~zx}v.tx~}=.
_x|p.~r~|t.|tpt.x}r{sts.
spx{.ePb.px}v.~u.rw~}xr.px}.p}s.
wt.trt}pvt.~u.qytr.w~.t~ts.
p.t{.~u.|~t.wp}.B?3.tsrx~}.x}.
px}.x}t}x=.cwt.pqx{x.~u.|~zts.
|pxyp}p.~.x}srt.prt.p}p{vtxp.p.
ptts.x}v.q~w.wt|p{.wtp.|~st{.
p}s.rppxrx}.t}xxpx~}.|~st{;.
wx{t.p}x<wtp{vtxp.p.ptts.
xw.qw.p}s.
~}.Ut.wpx.x|{x=.
cwt.x||tsxpt.p}p{vtxr.tuutr.~u.
|~zts.|pxyp}p.p.ptts.x}v.p.
?¤@??.~x}.ePb.p.C?<|x}t.x}t
p{.
wtt.x|t.qtu~t.p}s.wtt.x|t.put.
wt.ux.p}s.{p.|~zx}v.tx~};.
wxrw.p.s~}t.~.r~t~}s.~.wt.
x|t.~u.tpz.{p|p.rp}}pqx}~xs.{t
t{=.
]~pq{;.}~.p{{.qytr.r~|{tts.wt.
x}srts.px}.~x~}.~u.wt.s.6}.L.
@@.x}.|pxyp}p.v~;.H.x}.{prtq~.
v~7.qtrpt.~u.wtx.x}pqx{x.~.
~{tpt.wt.x|{x=.cw~vw~.wt.
s;.qytr.p{~.r~|{tts.wt.
_~ux{t.~u.\~~s.bpt.6_^\b7.
cwt.sv.s~t.x.t~ts.p.trt}pvt.~u.cWR.
tt}.x}.wt.|pxyp}p.pwt.wp}.|x{{xvp|.~u.
cWR.tt}.x}.tprw.rxvptt.qtrpt.~u.wt.
sxuuxr{.x}.stt|x}x}v.wt.p|~}.~u.cWR.
st{x
tts.q.x}wp{px~}.6tt.sxrx~}.x}.wt.
trx~}.t}x{ts.B=F=A.\pxyp}p.S~t.
bp}spsxpx~} 7=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
tx~}}pxt;.p.t{{.p.qytrx
t.ePb.
|tpt.~u.p}xt;.tspx~};.
sx~xt}px~};.pp}~xp;.r~}ux~};.
sxx}t;.p}s.}ptp=.
P.p.t{;.wt.|tsxp}.spx{.px}.p.
tsrts.BC3.q.|~zts.|pxyp}p.
r~|pts.[email protected].{prtq~.6r!L.?=?B7=.
Uxu<~.trt}.~u.qytr.w~.
|~zts.|pxyp}p.t~ts.p.MB?3.
tsrx~}.x}.px}.r~|pts.~.AC3.x}.
wt.{prtq~.v~.6r!L.?=?C7=.P{w~vw.
|pxyp}p.tsrts.ttx|t}p{{<.
x}srts.wtp{vtxp.6r!!%?=?D7.sx}v.
wt.ux.|~zx}v.tx~};.|pxyp}p.
sxs.}~.p{t.t~}t.~.prt{.
px}u{.x|{x=.
cwtt.tt.}~.tx~.PT.p}s.}~.
tx~st.~u.wtt}x~};.w~t}x~};.
~.prwrpsxp.txx}v.|tsxrp{.
x}t
t}x~}=.]~.qytr.xwst.u~|.
wt.s.u~.sv.t{pts.tp~}=.
bqytr.x}.wt.|pxyp}p.v~.
t~ts.wxvwt.px}v.~}.wt.qytrx
t.
|tpt.~u.p}xt;.tspx~};.
sx~xt}px~};.r~}ux~};.p}s.sxx}t.
r~|pts.~.wt.{prtq~.v~=.cwtt.
p.~}t.rpt.~u.t
tt.sxx}t.x}.p.
|pxyp}p<tpts.qytr=.Q.wt.t}s.~u.
wt.s;.qytr.tpts.xw.
|pxyp}p.p}s.{prtq~.t~ts.p.
tsrx~}.x}.~p{.|~~s.sxqp}rt.p.
|tpts.q._^\b=.
cwt.pw~.r~}r{st.wp.|~zts.
|pxyp}p.tuutrx
t{.tsrts.rw~}xr.
}t~pwxr.px}.u~|.WXe<p~rxpts.
t}~.}t~pw.xw.~{tpq{t.xst.
tuutr=.W~t
t;.{x|xpx~}.~u.wx.
s.x}r{stI.|px}t}p}rt.~u.qytr.
~}.~wt.p}p{vtxr.|tsxrpx~}.wx{t.
qtx}v.tts.xw.|pxyp}p.p}s.p.{prz.
~u.x}u~|px~}.pq~.wt.}|qt.~u.
uu.sx}v.tprw.x}wp{px~}.~u.|~zt=.
cwtt.{x|xpx~}.|pzt.x.sxuuxr{.~.
r~}r{st.wp.|pxyp}p.wp.p}p{vtxr.
~txt.~}.x.~}.p}s.wp.wt.prp{.
PT.ttxt}rts.sx}v.wt.s.x}.
t~}t.~.|pxyp}p.pt.~{tpq{t=.
W~t
t;.wt.s.~srts.~xx
t.
t{.vvtx}v.wp.|pxyp}p.w~{s.
qt.sxts.uwt.p.p}.psy}r.
tp|t}.u~.}r~}~{{ts.WXe<.
p~rxpts.t}~.}t~pw=.
T{{x.t.p{=.6A??H7.r~}srts.p.|~t.
trt}.s.t}x{ts.b|~zts.
|tsxrx}p{.rp}}pqx.u~.}t~pwxr.px}.
x}.WXeI.P.p}s~|xts;.r~~
t.
r{x}xrp{.xp{ =.cwt.qytr.tt.AG.
WXe<~xx
t.ps{.|p{t.pxt}.xw.
x}prpq{t.}t~pwxr.px}.wp.p.
tupr~.~.wt.tuutr.~u.p.{tp.~.
sv.pzt}.u~.p}p{vtxr.~t=.
d~}.t}.x}~.wt.s;.qytr.wps.
p.|tp}.r~t.~u.MD.~}.wt._px}.X}t}x.
qrp{t.~u.wt.Strx~.Sxuutt}xp{.
brp{t.6SSb7=.bqytr.tt.p{{~ts.~.
r~}x}t.pzx}v.wtx.rt}.~x}t.~u.
px}.|tsxrpx~};.wxrw.x}r{sts.
~x~xs;.}~}<}pr~xr.p}p{vtxr;.
p}xsttp};.p}s.p}xr~}
{p}=.
RQ!11111
Hto!11142
Hov!5812
Uhov!5813
64828!
_t
x~.ttxt}rt.xw.|pxyp}p.p.
}~.txts.u~.pxrxpx~}.x}.wt.
s;.q.AF.~u.AG.qytr.6HE37.
t~ts.t
x~.ttxt}rt.xw.
|pxyp}p=.W~t
t;.~u.wtt.AF.
ttxt}rts.qytr;.EB3.6}.L.@G7.
t~ts.}~.|pxyp}p.t.xwx}.wt.
p.tp=.
cwt.s.~rtst.r~|pts.wt.
tuutr.~u.wt.pvt.s~t.~u.|pxyp}p.
p}s.{prtq~.sx}v.~.tp|t}.
tx~s.{px}v.D.sp;.xw.A.ttz.
pw~.tx~s=.cwt.|pxyp}p.
t}vw.p
px{pq{t.tt.@3;.A3;.C3;.
E3;.~.G3.cWR.r~}rt}px~}.q.
txvw=.bqytr.|~zts.|pxyp}p.~.
{prtq~.rxvptt.u~.x|t.t.sp;.
p~x|pt{.H?¤@A?.|x}t.pp;.
x}v.p.p}spsxts.rts.|~zx}v.
~rts[email protected].tr~}s.|~zt.
x}wp{px~};.6A7.@?.tr~}s.w~{s.~u.|~zt.
x}.{}v;.6B7.C?.tr~}s.twp{t.p}s.
}~|p{.qtpwx}v.qttt}.uu=.cwt.
x}
txvp~.sxs.}~.~
xst.p.
strxx~}.~u.wt.}|qt.~u.uu.pzt}.
p.p}.|~zx}v.tx~}=.P{{.qytr.
prxrts.wt.|~zx}v.~rtst.x}v.
{prtq~.|pxyp}p.x~.~.t.tx~}=.
^}.wt.ux.sp.~u.tprw.t.tx~s;.
s~t.xpx~}.~rrts.w~vw~.wt.
u~.|~zx}v.tx~}.rwts{ts.u~.
wp.sp;.xw.p.px}v.t}vw.~u.C3.
cWR.r~}rt}px~}=.bqytr.tt.
p{{~ts.~.xpt.~.p.t~}p{xts.
pvt.s~t ;.wxrw.p.stux}ts.p.wt.
s~t.wp.~
xsts.wt.qt.px}.t{xtu.
xw~.x}~{tpq{t.ps
tt.tuutr=.cwx.
s~t.xpx~}.p.prr~|{xwts.q.
p{{~x}v.qytr.~.txwt.x}rtpt.wt.
s~t.x}rt|t}p{{.6~.E3.~.G3.cWR7.
~.x|~
t.p}p{vtxp;.~.~.strtpt.wt.
s~t.x}rt|t}p{{.6~.@3.~.A3.cWR7.
xu.PT.tt.x}~{tpq{t=.U~.wt.}t.C.
sp.~u.tprw.t.tx~s;.wt.qytr.
|~zts.wtx.pvt.s~t.sx}v.tprw.~u.
wt.u~.spx{.|~zx}v.tx~}=.c~.
|px}px}.wt.q{x}s;.{prtq~.|pxyp}p.
p.ttt}ts.p.r~}px}x}v.@3<G3.
cWR;.t
t}.w~vw.x.sxs.}~.r~}px}.p}.
rp}}pqx}~xs=.
cwt.x|p.~r~|t.|tpt.p.
wt.rwp}vt.x}.px}.|pv}xst.~}.wt.
SSb.p.wt.t}s.~u.tprw.t.tx~s.
r~|pts.~.qpt{x}t;.xw.p.r{x}xrp{{.
xv}xuxrp}.{t
t{.~u.p}p{vtxp.r~}xstts.
~.qt.p.tsrx~}.x}.px}.~u.p.{tp.B?3=.
Pssxx~}p{.|tpt.x}r{sts.wt.
_^\b;.wt.bxrz}t.X|pr._~ux{t.
6bX_7;.wt.Qxtu.b|~|.X}
t}~.
6QbX7.p}s.wt.dZd.bxst.Tuutr.apx}v.
brp{t.p}s.p.qytrx
t.wxvw}t>
tspx~}.ePb=.
Sx}v.wt.|pxyp}p.tp|t}.ttz;.
@H.qytr.xpts.~.wt.A3¤C3.cWR.
s~t.wx{t.wt.E3¤G3.s~t.p.
tutts.q.G.qytr.p}s.@.qytr.
rw~t.[email protected]~t=.X}.r~}p;.sx}v.
wt.{prtq~.tp|t}.ttz;.p{{.AG.
qytr.xpts.~.wt.wxvwt.~xq{t.
G<^HT^HO^23CWR3/UIO
23CWR3
64829!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
s~t.~u.G3.cWR .wp.r~}px}ts.}~.
prp{.rp}}pqx}~xs;.vvtx}v.wp.
{prtq~.tp|t}.~
xsts.{x{t.
p}p{vtxr.t{xtu=.
cwt.stvtt.~u.px}.tsrx~}.p.
xv}xuxrp}{.vtpt.put.ps|x}xpx~}.
~u.|pxyp}p.r~|pts.~.{prtq~.
6|tsxp}.rwp}vt.~u.B=B.~x}.~}.SSb;.
r!?!?=?@E7=.cwt.|tsxp}.rwp}vt.u•~|.
qpt{x}t.x}.ePb.px}.r~t.p.§@F.
u~.|pxyp}p.tp|t}.r~|pts.~.§C.
u~.{prtq~.tp|t}.6r!K.?=??@7=.P.
{pvt.~~x~}.~u.qytr.w~.tt.
tpts.xw.|pxyp}p.6?=CE7.t~ts.p.
MB?3.tsrx~}.x}.px};.r~|pts.~.
{prtq~.6?=@G7=.Pssxx~}p{{;.wt.pw~.
t~.x|~
t|t}.x}.~p{.|~~s.
sxqp}rt;.wxrp{.sxpqx{x;.p}s.
p{x.~u.{xut.p.|tpts.~}._^\b;.
bX_;.p}s.QbX.rp{t.put.q~w.{prtq~.
p}s.|pxyp}p.tp|t}.6spp.}~.
~
xsts.x}.pt7=.
X}.t|.~u.put;.wtt.tt.}~.
p{tpx~}.x}.WXe.sxtpt.pp|tt.x}.
t~}t.~.|pxyp}p.~.{prtq~=.cwt.
pw~.t~.wp.|pxyp}p.{ts.~.p.
vtpt.stvtt.~u.dZd.t~}t.p.t{{.
p.PT.rw.p.sxuuxr{.x}.
r~}rt}px~};.upxvt;.{ttx}t.~.
tspx~};.x}rtpts.spx~}.~u.{tt;.
tsrts.p{x
px~}.p}s.wx.r~|pts.
~.{prtq~.6spp.}~.~
xsts.x}.pt7=.
c~.qytr.xwst.u~|.wt.s.
qtrpt.~u.|pxyp}p<t{pts.PTI.~}t.
qytr.st
t{~ts.p}.x}prpq{t.
|~zx}v<t{pts.r~vw.sx}v.|pxyp}p.
ps|x}xpx~}.p}s.wt.~{t.|pxyp}p<.
}p'¶
t.qytr.x}.wt.s.ttxt}rts.
p}.x}rxst}.~u.prt.rp}}pqx<x}srts.
rw~x= .
cwt.pw~.r~}r{st.wp.|~zts.
|pxyp}p.tuutrx
t{.tsrts.rw~}xr.
}t~pwxr.px}.u~|.WXe<p~rxpts.
t}~.}t~pw=.cwt.{x|xpx~}.~u.
wx.s.x}r{stI.p.{prz.~u.x}u~|px~}.
pq~.wt.}|qt.~u.uu.sx}v.tprw.
x}wp{px~}.~u.|~ztJ.p.{prz.~u.
x}u~|px~}.pq~.wt.trxuxr.x|x}v.~u.
wt.qytrx
t.pt|t}.p}s.
r~{{trx~}.~u.PT.t{px
t.~.x}xxpx~}.~u.
wt.|~zx}v.tx~}J.p}s.wt.x}r{x~}.
~u.~}{.~}t.|pxyp}p<}p'¶
t.qytr=.
cwtt.{x|xpx~}.|pzt.x.sxuuxr{.~.
r~}r{st.wp.wt.prp{.PT.
ttxt}rts.sx}v.wt.s.x}.
t~}t.~.|pxyp}p.pt.~{tpq{t=.X.x.
ttrxp{{.r~}rt}x}v.wp.wt.~}{.
|pxyp}p<}p'¶
t.qytr.{tu.wt.s.
qtrpt.~u.tx~.rwxpxr.t~}t.
P.wt.x|t.~u.wt.s;.wt.u~{{~x}v.rxtxp.
u~|.wt.Sxpv}~xr.p}s.bpxxrp{.\p}p{.~u.
\t}p{.Sx~st.6Sb\¤Xe¤ca;.A???7.tt.ts.~.
sxpv}~t.qp}rt<x}srts.rw~xr.sx~stI.
_~|x}t}.wp{{rx}px~}.~.st{x~}J.
Wp{{rx}px~}.p}s>~.st{x~}.wp.st
t{~.
sx}v;.~.xwx}.~}t.|~}w.~u;.x}~xrpx~}.~.
xwspp{J.cwt.sxqp}rt.x.}~.qtt.prr~}ts.
u~.q.p.rw~xr.sx~st.wp.x.}~.qp}rt.
x}srts=.cwt.sxqp}rt.s~t.}~.~rr.
tr{x
t{.sx}v.wt.r~t.~u.p.st{xx|=.
XgtFcvg!Ugr>22@3125!
32<55!Cwi!22-!3127
Lmv!349112
~.|pxyp}p.t~t.p.p}p{vtxr.
s~t=.W~t
t;.wt.s.~srts.
~xx
t.t{.vvtx}v.wp.
|pxyp}p.w~{s.qt.sxts.uwt.p.
p}.psy}r.tp|t}.u~.}r~}~{{ts.
WXe<p~rxpts.t}~.}t~pw=.
~x}.ePb.u~.p{{~s}xp;.p}s.rwp}vt.x}.
wt|p{.px}.wtw~{s=.bqytrx
t.
|tpt.tt.p{~.t
p{pts.xw.
}x~{p.?¤@??.~x}.ePb.u~.p}.sv.
tuutr;.v~~s.sv.tuutr;.qps.sv.tuutr;.
wxvw;.s}z;.x|pxts;.~}ts;.{xzt.wt.
sv.tuutr;.tspts;.r~}uts;.
B=@=A Rt}p{.p}s._txwtp{.
}ptpts;.stxt.|~t.~u.wt.sv;.
]t~pwxr._px}.
p}x~;.s~};.w}v;.p}s.qx~{p.?¤.
cwtt.sxt.tp|x}ts.wt.tuutr.~u.
@??.~x}.ePb.u~.ps>wp;.p}x~>.
|pxyp}p.~}.rw~}xr.}t~pwxr.px}=. t{pts;.yxt>rp{|;.qps>v~~s;.
fx{t.t.p{=.6A??G7.tp|x}ts.rw~}xr. pp}~xs>t{u<pts;.utpu{>}pupxs=.
}t~pwxr.px}.u~|.|{x{t.rpt.
]t~r~v}xx
t.pt|t}.|tpts.
x}.wt.s.t}x{ts;.P.ap}s~|xts;.
pt}x~}.p}s.r~}rt}px~};.{tp}x}v.
_{prtq~<R~}~{{ts;.R~~
t.cxp{.~u.
p}s.|t|~;.p}s.ux}t.|~~.tts=.
Rp}}pqx.Rxvptt.x}.]t~pwxr.
\pxyp}p.~srts.p.tsrx~}.x}.
_px} =.cwt.qytr.tt.BA.pxt}.
px}.r~|pts.~.{prtq~;.p.|tpts.
xw.p.
pxt.~u.}t~pwxr.px}.
q.wt.px}.ePb;.wt._VXR.p}s.~}.px}.
r~}sxx~};.x}r{sx}v.AA.xw.r~|{t.
strx~.x}.wt.]_b;.x}r{sx}v.wp.
tvx~}p{.px}.}s~|t;.E.xw.x}p{.
6R!K.=??@7;.q}x}v.6R!K.=??@7;.prwx}v.6R!
r~s.x}y;.C.xw.|{x{t.r{t~x;.B. K.=??@7;.t}xx
t.6R!L.=?B7;.tuxrxp{.6R!
xw.sxpqtxr.}t~pw;.A.xw.
[email protected]}s.stt•.•px}.6R!K.=??@7=.]~pq{;.
x{x~x}vx}p{.}tp{vxp;.p}s.@xw.
wtt.tt.}~.pssxx~}p{.qt}tux.u~|.
{|q~prp{.{t~pw=.P{{.qytr.
wt.F3.cWR.t}vw.~u.|pxyp}p.
t~ts.p.px}.x}t}x.~u.p.{tp.B?.
r~|pts.~.wt.B=D3.cWR.t}vw;.
~}.p.?¤@??.ePb.p}s.tt.p{{~ts.~.
tt|x}v{.qtrpt.~u.r|{px
t.sv.
r~}x}t.pzx}v.wtx.tv{p.
tuutr.~
t.x|t=.cwtt.tt.}~.rwp}vt.
|tsxrpx~}.sx}v.wt.s.tx~s;.
x}.p{{~s}xp.~.wt|p{.px}.
wxrw.x}r{sts.~x~xs;.
t~}x
x.u~{{~x}v.ps|x}xpx~}.~u.
p}xsttp};.p}xr~}
{p};.p}s.
txwt.s~t.~u.|pxyp}p=.
]bPXS=.P{{.qytr.tt.txts.~.
\pxyp}p.p.q~w.t}vw.~srts.
wp
t.ttxt}rt.xw.|pxyp}p.q.
x}rtpt.~}.|tpt.~u.p}.sv.
r~{s.}~.t.p}.rp}}pqx}~xs.u~.B?.
tuutr;.v~~s.sv.tuutr;.wxvw;.~}ts;.
sp.qtu~t.s.tx~}=.
x|px|t};.tspx~};.r~}ux~};.p}s.
cwt.s.r~}xts.~u.wtt.t.
w}vt=.cwt.F3.cWR.|pxyp}p.
tx~}.xw.p}.x}t
p{.~u.B¤[email protected].
x}rtpts.p}xt.r~t.p}s.qps.sv.
qttt}.tx~}=.ctp|t}.r~}sxx~}. tuutr.6{pt.x}.tx~}7.r~|pts.~.
tt.wxvw<t}vw.|pxyp}p.6F3.st{p<. {prtq~=.]txwt.t}vw.~u.|pxyp}p.
H¤cWR7;.{~<t}vw.|pxyp}p.6B=D3.
puutrts.wt.|tpt.~u.|~~s=.^}.
st{p<H¤cWR7;.p}s.{prtq~.rxvptt;.
}t~r~v}xx
t.|tpt;.q~w.wt.B=D3.
ps|x}xtts.w~vw.p.p}spsxts.
cWR.p}s.F3.cWR.|pxyp}p.~srts.
rts<uu.~rtstI.6@7.{xvw.wt.
x|px|t}.x}.{tp}x}v.p}s.|t|~;.
rxvptt .6B?.tr~}s7;.6A7.vt.tps .
wx{t.~}{.wt.F3.cWR.|pxyp}p.
6D.tr~}s7;.6B7.x}wp{t .6D.tr~}s7;.6C7. x|pxts.pt}x~}.p}s.rw~|~~.
w~{s.|~zt.x}.{}v .6@?.tr~}s7;.6D7. tts;.r~|pts.~.{prtq~=.cwtt.tt.
twp{t; .p}s.6E7.px.qtu~t.ttpx}v.
}~.ps
tt.rpsx~
pr{p.xst.tuutr.
wt.uu.rr{t.6C?.tr~}s7=._pxrxp}. p}s.}~.qytr.s~ts.~.qtrpt.~u.
~~z.A.uu.put.qpt{x}t.
p}.ps
tt.t
t}.t{pts.~.|pxyp}p=.
cwt.pw~.r~}r{st.wp.|pxyp}p.
|tpt|t};.B.uu.p}.w~.{pt;.
|p.qt.tuutrx
t.p.p|t{x~px}v.
p}s.C.uu.p}.w~.put.wp;.u~.p.
}t~pwxr.px}.p.s~t.wp.x}srt.
r|{px
t.s~t.~u.H.uu.t.t.
|x{s.r~v}xx
t.tuutr;.q.wp.|~zx}v.
tx~}=.
x.}~.p}.~x||.~t.~u.
W~{.pt|t}.tx~s.tt.
ps|x}xpx~}=.cwt.{x|xpx~}.~u.wx.
rwts{ts.qtu~t.p}s.put.tprw.t.~u.
uu.p}s.u~.A.pssxx~}p{.w~.sx}v. s.x}r{stI.x}r{x~}.~u.qytr.
xw.|p}.u~|.~u.}t~pwxr.px}.
wt.tr~
t.tx~s=._{p|p.
p}s.|px}t}p}rt.~u.qytr.~}.~wt.
rp}}pqx}~xs.tt.|tpts.p.
p}p{vtxr.|tsxrpx~}.wx{t.qtx}v.tts.
qpt{x}t;.D.|x}t.put.wt.ux.uu.
xw.|pxyp}p=.cwtt.{x|xpx~}.|pzt.
p}s.pvpx}.p.B.w~.put.wt.{p.uu.
x.sxuuxr{.~.r~}r{st.wp.|pxyp}p.
rr{t=.
cwt.x|p.~r~|t.|tpt.p.
wp.p}p{vtxr.~txt.~}.x.~}.p}s.
~}p}t~.px}.t{xtu;.p.|tpts.q. wp.wt.prp{.PT.ttxt}rts.sx}v.
p.?¤@??.~x}.ePb.u~.rt}.px}=.
wt.s.x}.t~}t.~.|pxyp}p.pt.
_px}.}{tpp}}t.p.|tpts.~}.p. ~{tpq{t=.cwt.pw~.r~|pts.px}.
?¤@??.~x}.ePb;.p}s.stvtt.~u.px}.
r~t.t{.q.wt.t.~u.px}.
t{xtu.p.|tpts.~}.p.F<~x}._pxt}. r~}sxx~};.xw.}~.xv}xuxrp}.
V{~qp{.X|tx~}.~u.Rwp}vt.6_VXR7.
sxuutt}rt.u~}sJ.w~t
t;.wt.p|{t.
rp{t=.btr~}sp.|tpt.x}r{sts.wt. xt.~u.wx.s.p.|p{{.w.p.t.
]t~pwxr._px}.brp{t.6]_b7;.p.?¤@??.
XX.t~.|p.wp
t.qtt}.tt}=.cw;.x.
RQ!11111
Hto!11143
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
x.sxuuxr{.~.stt|x}t.xu.p}.pxr{p.
qt.~u.}t~pwxr.px}.r~}sxx~}.
~{s.qt}tux.trxuxrp{{.u~|.
|pxyp}p.ps|x}xpx~}=.W~t
t;.wt.
s.~srts.~xx
t.t{.
vvtx}v.wp.|pxyp}p.w~{s.qt.
sxts.uwt.p.p}.psy}r.tp|t}.
u~.}r~}~{{ts.}t~pwxr.px}=.
cwt.tr~}s.s;.r~}srts.q.fpt.
t.p{=.6A?@?7.x}.Rp}psp.x.t}x{ts;.
b|~zts.rp}}pqx.u~.rw~}xr.
}t~pwxr.px}I.p.p}s~|xts.
r~}~{{ts.xp{ =.cwt.qytr.tt.A@.
ps{.pxt}.xw.}t~pwxr.px}.
rpts.q.p|p.~.vt.
r~|~}sts.xw.p{{~s}xp.~.
wtp{vtxp;.p}s.p.px}.x}t}x.r~t.
vtpt.wp}.C.~}.p.@?.~x}.ePb=.P{{.
qytr.|px}px}ts.wtx.rt}.
p}p{vtxr.|tsxrpx~}.p}s.wt.tt.
p{{~ts.~.t.prtp|x}~wt}.u~.
qtpzw~vw.px}=.Txvwtt}.qytr.
wps.t
x~.ttxt}rt.xw.|pxyp}p.
q.}~}t.~u.wt|.wps.ts.|pxyp}p.
xwx}.p.tp.qtu~t.wt.s=.
cwt.s.stxv}.ts.p.u~<tx~s.
r~~
t.stxv};.tx}v.|pxyp}p.
6A=D3;.E=?3.p}s.H=C3.cWR7.p}s.
{prtq~.|pxyp}p=.cwt.A=D3.p}s.E=?3.
s~t.~u.|pxyp}p.tt.x}r{sts.~.
x}rtpt.rrtu{.q{x}sx}v=.Tprw.
tx~s.p[email protected].x}.spx~};.
qtvx}}x}v.xw.D.sp.~}.wt.s.sv.
u~{{~ts.q.p.H<sp.pw~.tx~s=.
S~t.tt.st{x
tts.p.AD.|v.~u.
|pxyp}p.wp.p.|~zts.x}.p.x}v{t.
x}wp{px~}.x}v.p.xp}x|.xt=.cwt.
ux.s~t.~u.tprw.tx~s.p.t{u<.
ps|x}xtts.x}v.p.p}spsxts.uu.
~rts[email protected]}wp{t.u~.D.tr~}s;.6A7.
w~{s.wt.|~zt.x}.wtx.{}v.u~.@?.
tr~}s;.p}s.6B7.twp{t=.bqtt}.
s~t.tt.t{u<ps|x}xtts.x}.wt.
p|t.|p}}t.u~.p.~p{.~u.wtt.x|t.
spx{.p.w~|t.~}.p}.~pxt}.qpx.u~.
wt.ux.ux
t.sp.~u.tprw.tx~s=.
cwt.x|p.|tpt.p.p}.@@<~x}.
px}.x}t}x.rp{t;.p
tpvts.~
t.wt.D.
sp.tp|t}.tx~s;.wxrw.p.
ps|x}xtts.~}rt.spx{.u~.tt};.
~;.{tp.p}s.p
tpvt.px}.x}t}x.
sx}v.wt.t
x~.AC.w~=.
btr~}sp.|tpt.x}r{sts.p}.prt.
px}.?¤@??.~x}.ePb;.px}.p{x.
ptts.xw.wt.\rVx{{._px}.
`tx~}}pxt;.{tt.ptts.xw.wt.
[tts.b{tt.T
p{px~}.`tx~}}pxt;.
|~~s.ptts.xw.wt._^\b;.p{x.
~u.{xut.ptts.x}v.wt.T`¤DS.wtp{w.
~r~|t.x}|t}=.bqytrx
t.
|tpt.x}r{sts.?¤@??.~x}.ePb.
rp{t.u~.wxvw;.t{pts;.tts.p}s.
wp=.
^
t.wt.ux.wtt.w~.put.
|~zx}v.|pxyp}p;.px}v.~u.px};.
wxvw;.t{ppx~};.t;.wpx}t.p}s.
wtp.pt.tt.tr~sts=.Sx}v.wt.ux
t.
sp.~u.tprw.s.tx~s;.pxrxp}.
tt.r~}prts.spx{.~.ps|x}xt.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
tx~}}pxt.~}.px}.x}t}x;.{tt;.
|tsxrpx~}.p}s.PT=.bqytr.t}ts.
~}.wt.uxuw.sp.~.r~|{tt.
tx~}}pxt.~}.px}.p{x;.|~~s;.
p{x.~u.{xut.p}s.pt|t}.~u.
~t}r=.P.wt.t}s.~u.wt.s;.
pxrxp}.r~|{tts.ux}p{.ps
tt.
t
t}.t~.p}s.~t}r.pt|t}=.
cwt.p
tpvt.spx{.px}.x}t}x.p.
xv}xuxrp}{.{~t.~}.H=C3.cWR.
|pxyp}p.6D=C7.wp}.~}.{prtq~.
|pxyp}[email protected]!L.?=?AB7=.cwt.H=C3.
cWR.t}vw.p{~.~srts.|~t.
s~x}t;.qtt.{tt;.xw.{t.
p}xt.p}s.sttx~};.r~|pts.~.
•{prtq~.6p{{.r!K.?=?D7=.W~t
t;.wtt.
tt.}~.xv}xuxrp}.sxuutt}rt.~}.
_^\b.r~t.~.~}.ePb.r~t.u~.wxvw;.
wp;.t{pts.~.tts.qttt}.cWR.
s~t=.
cwt.|~.utt}.sv<t{pts.
ps
tt.t
t}.t~ts.x}.wt.v~.
trtx
x}v.H=C3.cWR.|pxyp}p.tt.
wtpsprwt;.s.tt;.q}x}v.t}px~};.
sxx}t;.}|q}t.p}s.r~vw=.at~.
~u.wxvw.p}s.tw~xp.~rrts.~}.~}{.
wtt.~rrpx~};.~}rt.x}.tprw.s~t.~u.
cWR=.cwtt.tt.}~.xv}xuxrp}.rwp}vt.
x}.
xp{.xv};.wtp<pt.
pxpqx{x;.~.
t}p{.u}rx~}=.^}t.qytr.xwst.
u~|.wt.s.st.~.x}rtpts.px}.
sx}v.ps|x}xpx~}.~u.E3.cWR.
|pxyp}p=.
cwt.pw~.r~}r{st.wp.|~zts.
|pxyp}p.tsrt.}t~pwxr.px};.
x|~
t.|~~s.p}s.pxs.x}.{tt;.q.
wp.|~zx}v.|pxyp}p.x.}~.p.
tutpq{t.~t.~u.ps|x}xpx~}=.cwt.
{x|xpx~}.~u.wx.s.x}r{stI.cwt.
{prz.~u.x}u~|px~}.~}.x|x}v.~u.
pt|t}.sx}v.wt.~pxt}.
~x~}.~u.wt.s.p}s.|px}t}p}rt.~u.
qytr.~}.~wt.p}p{vtxr.|tsxrpx~}.
wx{t.qtx}v.tts.xw.|pxyp}p=.
cwtt.{x|xpx~}.|pzt.x.sxuuxr{.~.
r~}r{st.wp.|pxyp}p.wp.p}p{vtxr.
~txt.~}.x.~}.p}s.wp.wt.prp{.
PT.ttxt}rts.sx}v.wt.s.x}.
t~}t.~.|pxyp}p.pt.~{tpq{t=.
W~t
t;.wt.s.~srts.~xx
t.
t{.vvtx}v.wp.|pxyp}p.w~{s.
qt.sxts.uwt.p.p}.psy}r.
tp|t}.u~.}r~}~{{ts.}t~pwxr.
•px}=.
fx{t.t.p{[email protected]~}srts.wt.
|~.trt}.s.t}x{ts;.[~<S~t.
ep~xts.Rp}}pqx.bxv}xuxrp}{.
X|~
t.]t~pwxr._px} =.cwx.s.
x.wt.~}{.~}t.x}.wx.t
xt.wp.
x{xts.
p~xpx~}.p.p.|tw~s.~u.
|pxyp}p.ps|x}xpx~}=.cwt.qytr.
tt.BE.pxt}.xw.p.}t~pwxr.
px}.sx~st.6Ra_b;.wp{p|xr.px};.
x}p{.r~s.x}y;.txwtp{.
}t~pw;.psxr{~pw;.~.}t
t.
x}y7.w~.tt.|px}px}ts.~}.wtx.
rt}.|tsxrpx~}.6~x~xs;.
p}xr~}
{p};.p}xsttp};.p}s.
]bPXS7=.P{w~vw.qytr.tt.
RQ!11111
Hto!11144
Hov!5812
Uhov!5813
6482;!
txts.~.wp
t.p.wx~.~u.|pxyp}p.
t;.wt.tupx}ts.u~|.t.~u.
rp}}pqx}~xs.u~.B?.sp.qtu~t.s.
tx~}=.
bqytr.pxrxpts.x}.wtt.tx~}.
x}.wxrw.wt.trtx
ts.@=AH3.~.B=DB3.
cWR.|pxyp}p.~.{prtq~.|pxyp}p=.
cwt.|pxyp}p.p.
p~xts.x}v.wt.
e~{rp}~.
p~xt.p}s.p.p}spsxts.
rts<uu.~rtstI.6@7.w~{s.wt.
p~xt.qpv.xw.~}t.wp}s.p}s..wt.
p~xt.|~wxtrt.x}.wtx.|~w .
6B?.tr~}s7;.6A7.vt.tps .6D.
tr~}s7;.6B7.x}wp{t .6D.tr~}s7;.6C7.
w~{s.
p~.x}.{}v .6@?.tr~}s7;.6D7.
twp{t.p}s.px .qtu~t.ttpx}v.uu.
rr{t.6C?.tr~}s7=.bqytr.x}wp{ts.C.
uu.p.E?.|x}t=.P.@G?.|x}t;.wt.
p~xt.p.tux{{ts.xw.|pxyp}p.
p~.p}s.qytr.tt.p{{~ts.~.
x}wp{t.C.~.G.uu.x}v.wt.rts.
~rtst=.cw;.r|{px
t.s~x}v.
p{{~ts.u~.p.p}vt.~u.G.~@A.uu.x}.
~p{.u~.tprw.tx~};.stt}sx}v.~}.wt.
qytr.stxts.t~}t.p}s.~{tp}rt=.
cwt.pw~.x|t.qttt}.tprw.tx~}.
p}vts.u~|.B¤@C.sp=.
cwt.x|p.~r~|t.
pxpq{t.p.
~}p}t~.px}.t{xtu;.p.ptts.
x}v.p.?¤@??.~x}.ePb.u~.rt}.
px}=.btr~}sp.|tpt.x}r{sts.wt.
_pxt}.V{~qp{.X|tx~}.~u.Rwp}vt.
6_VXR7;.wt.]t~pwxr._px}.brp{t.
6]_b7;.p.?¤@??.~x}.ePb.u~.p{{~s}xp=.
Prt.px}.wtw~{s.p.|tpts.
xw.p.wt|p{.px}.|~st{=.bqytrx
t.
|tpt.x}r{sts.?¤@??.~x}.}x~{p.
ePb.u~.p}.sv.tuutr;.v~~s.sv.
tuutr;.qps.sv.tuutr;.wxvw;.s}z;.
x|pxts;.~}ts;.sv.{xzx}v;.tspts;.
r~}uts;.}ptpts;.stxt.|~t.sv;.
p}x~;.s~}.p}s.w}v=.Qx~{p.?¤.
@??.~x}.ePb.x}r{sts.ps>wp;.
p}x~>t{pts;.yxt>rp{|;.qps>v~~s;.
pp}~xs>t{u<pts;.p}s.utpu{>
}pupxs=.]t~r~v}xx
t.pt|t}.
ptts.pt}x~}.p}s.r~}rt}px~};.
{tp}x}v.p}s.|t|~;.p}s.ux}t.|~~.
tts=.
P.B?3.tsrx~}.x}.px}.p.
prwxt
ts.x}.E@3.~u.qytr.w~.
trtx
ts.wt.B=DB3.cWR.|pxyp}p;.x}.
DF3.~u.qytr.w~.trtx
ts.wt.@=AH3.
cWR.|pxyp}p.p}s.x}.AE3.~u.qytr.
w~.trtx
ts.wt.{prtq~.|pxyp}p.6r!
L.?=??A.u~.{prtq~.
=.B=DB3.cWR;.r!L.
?=??F.u~.{prtq~.
.@=AH3.cWRJ.r!M.
?=?D.@=AH3.cWR.
=.B=DB3.cWR7=.Q~w.
t}vw.~u.|pxyp}p.xv}xuxrp}{.
strtpts.px}.x}t}x;.
}{tpp}}t;.wp}t;.p}s.
stt}t.~}.wt.]_b;.p.t{{.p.px}.
px}v.~}.wt._VXR;.r~|pts.~.
{prtq~=.cwtt.tuutr.~}.px}.tt.
|px|p{.xw.r|{px
t.s~x}v.~
t.
wt.r~t.~u.wt.s.tx~};.xw.
|px|p{.tuutr.p.@G?.|x}t=.cwtt.
tt.}~.tuutr.~u.|pxyp}p.r~|pts.
~.{prtq~.~}.|tpt.~u.p{{~s}xp.~.
G<^HT^HO^23CWR3/UIO
23CWR3
64831!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
wt|p{.px}=.bqytr.r~tr{.
xst}xuxts.wt.s.tp|t}.EB3.~u.
wt.x|t.u~.{prtq~;.E@3.~u.wt.x|t.u~.
@=AH3.cWR;.p}s.GH3.~u.wt.x|t.u~.
B=DB3.cWR=.
^}.qytrx
t.|tpt;.|pxyp}p.
~srts.s~t<stt}st}.x}rtpt.
r~|pts.~.{prtq~.~}.px}v.u~I.p}.
sv.tuutr;.v~~s.sv.tuutr;.sv.
{xzx}v;.wxvw;.~}ts;.tspts;.r~}uts;.
p}s.w}v=.Q~w.t}vw.~u.|pxyp}p.
~srts.x|x{p.x}rtpt.x}.s}z.~.
x|pxts.r~|pts.~.{prtq~=.X}.
r~}p;.stxt.u~.sv.p.pts.p.
wxvwt.u~.wt.@=AH3.cWR.|pxyp}p.
r~|pts.~.wt.B=DB3.cWR.|pxyp}p=.
cwtt.tt.}~.rwp}vt.r~|pts.~.
{prtq~.u~.qps.tuutr;.}pt~;.
p}xt;.utt{x}v.s~}.~.p}.~u.wt.
qx~{p.|~~s.pt|t}=.cwtt.p.
s~t<stt}st}.x|px|t}.~}.{tp}x}v.
p}s.|t|~.u~|.|pxyp}p.r~|pts.
~.{prtq~;.q.x|x{p.tuutr.qttt}.
wt.~.t}vw.~u.|pxyp}p.~}.
pt}x~}=.
cwt.pw~.r~}r{st.wp.
p~xpx~}.~u.t{px
t{.{~.s~t.~u.
|pxyp}p.rp}.~srt.x|~
t|t}.x}.
p}p{vtxp.x}.}t~pwxr.px}.pxt};.
ttrxp{{.wt}.pxt}.pt.p{{~ts.~.
xpt.wtx.t~t=.W~t
t;.wx.
x}sx
xsp{xpx~}.~u.s~t.|p.prr~}.
u~.wt.vt}tp{.{prz.~u.sxuutt}rt.
qttt}.wt.~.t}vw.~u.|pxyp}p=.
]~.spp.tt.tt}ts.tvpsx}v.wt.
~p{.p|~}.~u.cWR.r~}|ts.q.tprw.
qytr;.~.x.x.sxuuxr{.~.stt|x}t.p.
~t.s~t<t~}t.t
p{px~}=.
Pssxx~}p{.{x|xpx~}.~u.wx.s.pt.
wt.x}r{x~}.~u.qytr.xw.|p}.
u~|.~u.}t~pwxr.px}.p}s.
|px}t}p}rt.~u.qytr.~}.~wt.
p}p{vtxr.|tsxrpx~}.wx{t.qtx}v.tts.
xw.|pxyp}p=.cwtt.{x|xpx~}.|pzt.
x.sxuuxr{.~.r~}r{st.wp.|pxyp}p.
wp.p}p{vtxr.~txt.~}.x.~}=.X.x.
p{~.sxuuxr{.~.stt|x}t.xu.p}.
pxr{p.qt.~u.}t~pwxr.px}.
r~}sxx~}.~{s.qt}tux.trxuxrp{{.
u~|.|pxyp}p.ps|x}xpx~}=.
W~t
t;.wt.s.~srts.~xx
t.
t{.vvtx}v.wp.|pxyp}p.w~{s.
qt.sxts.uwt.p.p}.psy}r.
tp|t}.u~.}r~}~{{ts.}t~pwxr.
•px}=.
4/3 Crrgvkvg!Uvkowncvkqp!kp!JKX!
c~.p}s~|xts;.s~q{t<q{x}s;.
{prtq~<r~}~{{ts._wpt.A.sxt.
tp|x}ts.wt.tuutr.~u.|~zts.
|pxyp}p.~}.ptxt.x}.WXe<~xx
t.
qytr.6Wp}t.t.p{=;.A??DJ.Wp}t.t.
p{=;.A??F7=.cpq{t.A.~u.wt.Pt}sx.
||pxt.q~w.sxt=.
cwt.ux.s;.r~}srts.q.Wp}t.
t.p{=.6A??D7.x.t}x{ts;.S~}pqx}~{.p}s.
|pxyp}p.x}.WXe9.|pxyp}p.|~ztI.
prt.tuutr.~}.rp{~xr.x}pzt.p}s.
|~~s =.cwt.qytr.tt.B?.WXe<.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
~xx
t.pxt}.w~.tt.|px}px}ts.
~}.~.p}xt~
xp{.|tsxrpx~}.p}s.
txwt.wps.r{x}xrp{{.xv}xuxrp}.
strtpt.x}.{tp}.|r{t.|p 6{~<.
QXP.v~;.}.L.@D7.~.}~|p{.{tp}.
|r{t.|p.6}~|p{<QXP.v~;.}.L.
@D7=.P{{.qytr.wps.p.wx~.~u.
|~zx}v.|pxyp}p.p.{tp.xrt.ttz{.
u~.C.ttz.x~.~.t}.x}~.wt.s=.
^}.p
tpvt;.x}sx
xsp{.wps.|~zts.B.
|pxyp}p.rxvptt.t.sp;.D¤E.x|t.
t.ttz.u~.@?¤@A.tp=.
bqytr.pxrxpts.x}.G.tx~}.
wp.tts.wt.prt.tuutr.~u.?;.@?;.A?;.
p}s.B?.|v.s~}pqx}~{.~p{.rp{t.p}s.
|pxyp}p.rxvptt.xw.?3;.@=G3;.
A=G3;.p}s.B=H3.cWR.r~}rt}px~}.q.
txvw;.x}v.p.s~q{t<s||.stxv}.
6xw.~}{.~}t.prx
t.sv.t.tx~}7=.
cwt.s~t.~u.s~}pqx}~{.pt.wxvwt.wp}.
w~t.s~t.xrp{{.trxqts.u~.
ptxt.x|{px~}.x}.~st.~.wt{.
tt
t.wt.q{x}sx}v=.cwtt.p.p.~}t<.
sp.pw~.tx~s.qttt}.t.
tx~}=.
\pxyp}p.p.ps|x}xtts.x}v.p.
p}spsxts.rts.~rtstI.6@7.{xvw.
wt.rxvptt .6B?.tr~}s7;.6A7.
tpt .6D.tr~}s7;.6B7.x}wp{t .6D.
tr~}s7;.6C7.w~{s.|~zt.x}.{}v .6@?.
tr~}s7;.p}s.6D7.twp{t= .Tprw.qytr.
|~zts.wtt.uu.x}.wx.|p}}t;.xw.
p.C?<tr~}s.x}t
p{.qttt}.tprw.uu=.
Rp{~xr.x}pzt.p.ts.p.p.~vpt.
|tpt.u~.txvw.vpx}=.bqytr.
trtx
ts.p.q~.r~}px}x}v.p.
pxt.~u.
u~~s.p}s.qt
tpvt.xt|.p}s.tt.~{s.
~.tr~s.r~}|x~}.~u.wtt.xt|.
u~{{~x}v.wp.sp .ps|x}xpx~}.~u.
wt.t.sv=.bqytrx
t.|tpt.
x}r{sts.?¤@??.~x}.ePb.u~.utt{.sv.
tuutr;.v~~s.tuutr;.qps.tuutr;.pzt.sv.
pvpx};.sv.{xzx}v;.w}v;.u{{;.
}ptpts;.wx;.stxt.~.tp=.
]t~r~v}xx
t.|tpt.p}s.
xp{.xv}.
tt.|~}x~ts=.
cwt.{~.QXP.v~.r~}|ts.
xv}xuxrp}{.|~t.rp{~xt.x}.wt.@=G3.
p}s.B=H3.cWR.|pxyp}p.r~}sxx~}.6r!
[email protected]}s.wt.@?;.A?;.p}s.B?.|v.
s~}pqx}~{.r~}sxx~}.6r!K.?=?@7.
r~|pts.xw.wt.{prtq~.r~}sxx~}=.
X}.r~}p;.x}.wt.}~|p{.QXP.v~;.
}txwt.|pxyp}p.}~.s~}pqx}~{.
xv}xuxrp}{.puutrts.rp{~xr.x}pzt=.
cwx.{prz.~u.tuutr.|p.qt.prr~}pq{t;.
w~t
t;.q.wt.upr.wp.wx.v~.
r~}|ts.p~x|pt{.A??.rp{~xt.
|~t.wp}.wt.{~.QXP.v~.}st.
qpt{x}t.r~}sxx~}=.
apx}v.~u.wxvw.p}s.v~~s.sv.tuutr.
tt.x}rtpts.q.p{{.sv.tp|t}.x}.
q~w.wt.{~<QXP.p}s.}~|p{<QXP.
v~;.trt.x}.t~}t.~.wt.@?.|v.
[tp}.|r{t.|p.p.ptts.x}v.
qx~t{trxrp{.x|tsp}rt.p}p{x.6QXP7=.cwt.{~<.
QXP.v~.p.r{pxuxts.xw.wp
x}v.KH?3.QXP;.
p}s.wt.}~|p{<QXP.v~.p.r{pxuxts.xw.
wp
x}v.MH?3.QXP=.
RQ!11111
Hto!11145
Hov!5812
Uhov!5813
s~t.~u.s~}pqx}~{=.cwt.B=H3.cWR.
|pxyp}p.x}rtpts.px}v.~u.v~~s.
sv.tuutr;.sv.{xzx}v.p}s.stxt.~.
|~zt.pvpx}.r~|pts.xw.{prtq~=.
apx}v.~u.tspx~}.tt.x}rtpts.x}.
q~w.v~.q.@?.p}s.B?.|v.
s~}pqx}~{;.p}s.x}.wt.}~|p{.QXP.
v~.q.wt.A=G3.cWR.|pxyp}p=.
apx}v.~u.x|{px~}.tt.x}rtpts.x}.
wt.}~|p{.QXP.v~.q.A=G3.p}s.
B=H3.cWR.|pxyp}p.p}s.q.A?.|v.
s~}pqx}~{=.X}rtpt.x}.px}v.~u.
u~vtu{}t;.xwsp};.stp|x}v;.
r{|;.wtp
.{x|q;.wtp.~}sx}v;.
yxt;.p}s.strtpt.x}.px}v.~u.
t}tvtxr;.~rxp{;.p}s.p{zpx
t.tt.
t~ts.x}.wt.}~|p{.QXP.v~.xw.
B?.|v.s~}pqx}~{=.cwtt.tt.}~.
xv}xuxrp}.rwp}vt.x}.
xp{.xv}.~.
tu~|p}rt.~}.}t~r~v}xx
t.
|tpt.x}.t~}t.~.|pxyp}p=.
]~pq{;.wt.x|t.r~t.~u.qytrx
t.
tuutr.tpzts.xrz{.p}s.str{x}ts.
wttput.u~.|~zts.|pxyp}p;.wx{t.
~p{.s~}pqx}~{.t~}t.~~z.{~}vt.~.
tpz.p}s.txts.{~}vt=.Pssxx~}p{{;.
|pxyp}p.q.}~.s~}pqx}~{.~srts.
s.|~w.p}s.wx=.
X}.vt}tp{;.PT.t~ts.x}.wx.s.
tt.{~.x}.q~w.sv.r~}sxx~}.u~.
q~w.qytr.v~=.X}.wt.{~.QXP.
v~;.}ptp.p.t~ts.q.~}t.
qytr.x}.q~w.wt.@?.p}s.A?.|v.
s~}pqx}~{.r~}sxx~};.wx{t.p}.
}r~|u~pq{t.{t
t{.~u.x}~xrpx~}.p.
~srts.q.wt.B?.|v.s~t.x}.~.
qytr=.cwtt.tt.}~.PT.t~ts.x}.
wx.v~.u~{{~x}v.|pxyp}p.p.p}.
s~t=.X}.wt.}~|p{.QXP.v~;.wt.B?.
|v.s~t.~u.s~}pqx}~{.~srts.p}.
}r~|u~pq{t.{t
t{.~u.x}~xrpx~}.x}.
wtt.qytr.p}s.wtpsprwt.x}.~}t.
qytr;.wx{t.wt.B=H3.|pxyp}p.
~srts.sxpwtp.x}.~}t.qytr=.
cwt.pw~.r~}r{st.wp.|~zts.
|pxyp}p.rp}.prt{.x}rtpt.rp{~xr.
x}pzt.x}.{~.QXP.qytr.xw~.
xv}xuxrp}.r~v}xx
t.x|px|t}=.
W~t
t;.x.x.~xq{t.wp.wt.{~.
stvtt.~u.r~v}xx
t.x|px|t}.t~ts.
x}.wx.s.|p.tu{tr.wt.
st
t{~|t}.~u.~{tp}rt.~.
rp}}pqx}~xs.x}.wx.pxt}.~{px~};.
x}rt.p{{.x}sx
xsp{.wps.rt}.
wx~xt.~u.rw~}xr.|pxyp}p.t=.
Pssxx~}p{.{x|xpx~}.x}.wx.s.
x}r{st.}~.x{xx}v.prp{.txvw.vpx}.
p.p.x|p.|tpt=.W~t
t;.wt.
s.~srts.~xx
t.t{.
vvtx}v.wp.|pxyp}p.w~{s.qt.
sxts.uwt.p.p.tp|t}.u~.
ptxt.x|{px~}.x}.WXe.pxt}=.
P.tr~}s.s.r~}srts.q.Wp}t.
t.p{=.6A??F7.x.t}x{ts;.S~}pqx}~{.p}s.
|pxyp}p.x}.WXe<~xx
t.|pxyp}p.
|~ztI.Rp{~xr.x}pzt;.|~~s;.p}s.
{tt =.cwt.stxv}.~u.wx.s.p.
}tp{.xst}xrp{.~.wt.~}t.r~}srts.q.
wx.{pq~p~.x}.A??D.6tt.pq~
t7;.q.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
wtt.p.}~.pxuxrpx~}.~u.qytr.q.
QXP=.cwt.qytr.tt.@?.WXe<~xx
t.
pxt}.w~.tt.|px}px}ts.~}.~.
p}xt~
xp{.|tsxrpx~}.p}s.wps.p.
wx~.~u.|~zx}v.|pxyp}p.p.{tp.
xrt.ttz{.u~.C.ttz.x~.~.t}.
x}~.wt.s=.^}.p
tpvt;.x}sx
xsp{.
wps.|~zts.B.|pxyp}p.rxvptt.t.
sp;.D.x|t.t.ttz.u~.@H.tp=.
bqytr.pxrxpts.x}.G.tx~}.
wp.tts.wt.prt.tuutr.~u.?;.D.p}s.
@?.|v.s~}pqx}~{.~p{.rp{t.p}s.
|pxyp}p.rxvptt.xw.?;.A=?3.p}s.
B=H3.cWR.r~}rt}px~}.q.txvw;.
x}v.p.s~q{t<s||.stxv}.6xw.C.
tx~}.x}
~{
x}v.~}{.~}t.prx
t.sv.
p}s.C.x}ttts.{prtq~.tx~}7=.
Q~w.sv.p}s.{prtq~.tx~}.{pts.
u~.C.sp.tprw;.xw.prx
t.sv.
ps|x}xpx~}.~rrx}v.C.x|t.t.
sp.6t
t.C.w~7=.ctx}v.~rrts.x}.
~.@E<sp.x}pxt}.p=.X}.wt.
x}t
t}x}v.~pxt}.tx~s;.qytr.
tt.p{{~ts.~.|~zt.|pxyp}p.x~.
~.t<t}.~.wt.s.}x.u~.wt.
tr~}s.x}pxt}.p=.
\pxyp}p.p.ps|x}xtts.x}v.p.
p}spsxts.rts.~rtstI.6@7.{xvw.
wt.rxvptt .6B?.tr~}s7;.6A7.
tpt .6D.tr~}s7;.6B7.x}wp{t .6D.
tr~}s7;.6C7.w~{s.|~zt.x}.{}v .6@?.
tr~}s7;.p}s.6D7.twp{t= .Tprw.qytr.
|~zts.wtt.uu.x}.wx.|p}}t;.xw.
p.C?<tr~}s.x}t
p{.qttt}.tprw.uu=.
Rp{~xr.x}pzt.p.ts.p.p.~vpt.
|tpt.u~.txvw.vpx};.q.qytr.
tt.p{~.txvwts.w~vw~.wt.s.
6p.|tpt.wxrw.p.}~.r~{{trts.x}.
wt.A??D.s.q.wx.v~7=.bqytr.
trtx
ts.p.q~.r~}px}x}v.p.
pxt.~u.
u~~s.p}s.qt
tpvt.xt|.p}s.tt.~{s.
~.tr~s.r~}|x~}.~u.wtt.xt|.
u~{{~x}v.wp.sp .ps|x}xpx~}.~u.
wt.t.sv=.bqytrx
t.|tpt.
x}r{sts.?¤@??.~x}.ePb.u~.sv.
tuutr;.v~~s.tuutr;.qps.tuutr;.pzt.sv.
pvpx};.sv.{xzx}v;.w}v;.u{{;.
}ptpts;.wx;.stxt.~.tp=.
]t~r~v}xx
t.|tpt.p}s.
xp{.xv}.
tt.|~}x~ts=.b{tt.p.ptts.
x}v.q~w.wt.]xvwrp.{tt.
|~}x~x}v.t|.p}s.t{trts.ePb.
|tpt.t{pts.~.{tt=.
Q~w.D.p}s.@?.|v.s~}pqx}~{.6r!K.
?=??G7.p}s.A=?3.p}s.B=H3.cWR.
|pxyp}[email protected]~t<stt}st}{.
x}rtpts.rp{~xr.x}pzt.r~|pts.xw.
{prtq~=.cwx.x}rtpt.p.vt}tp{{.
prr~|{xwts.w~vw.x}rtpt.x}.
x}rxst}.~u.tpx}v;.pwt.wp}.p}.
x}rtpt.x}.wt.rp{~xt.r~}|ts.x}.
tprw.x}rxst}=.bqytr.p{~.vpx}ts.
x|x{p.p|~}.~u.txvw.put.wt.
wxvwt.s~t.~u.tprw.rp}}pqx}~xs.
tp|t}I.@=A.zv.6A=E.{q7.put.C.sp.
~u.@?.|v.s~}pqx}~{;.p}s.@[email protected]=C.{q7.
put.C.sp.~u.B=H3.cWR.|pxyp}p=.
cwt.B=H3.cWR.|pxyp}p.s~t.p{~.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
x}rtpts.wt.stxt.~.tp.p}s.px}v.~u.
w}vt=.
apx}v.~u.v~~s.sv.tuutr;.wxvw;.sv.
{xzx}v;.p}s.stxt.~.|~zt.pvpx}.tt.
xv}xuxrp}{.x}rtpts.q.@?.|v.
s•~}pqx}~{.p}s.A=?3.p}s.B=H3.cWR.
|pxyp}p.s~t.r~|pts.~.{prtq~=.
Q~w.|pxyp}p.s~t.x}rtpts.px}v.
~u.x|{pts;.uxt}s{;.p}s.t{u<.
r~}uxst}=.cwt.@?.|v.s~t.~u.s~}pqx}~{.
x}rtpts.px}v.~u.r~}rt}px~}.
x|px|t};.p}s.wt.A=?3.cWR.
|pxyp}p.s~t.x}rtpts.px}v.~u.
p}x~=.S.|~w.p.x}srts.q.@?.
|v.s•~}pqx}~{.6@?.|v7.p}s.A=?3.cWR.
|pxyp}p=.cwtt.tt.}~.rwp}vt.x}.
}t~r~v}xx
t.tu~|p}rt.~.~qytrx
t.
{tt.|tpt.u~|.ps|x}xpx~}.~u.
txwt.rp}}pqx}~xs=.W~t
t;.B=H3.cWR.
|pxyp}p.x}rtpts.qytrx
t.px}v.
~u.{tt=.
cwt.pw~.r~}r{st.wp.q~w.
s~}pqx}~{.p}s.|~zts.|pxyp}p.
x}rtpt.rp{~xr.x}pzt.p}s.~srt.
txvw.vpx}.x}.WXe<~xx
t.pxt}=.
W~t
t;.x.x.~xq{t.wp.wt.{~.
stvtt.~u.r~v}xx
t.x|px|t}.t~ts.
x}.wx.s.|p.tu{tr.wt.
st
t{~|t}.~u.~{tp}rt.~.
rp}}pqx}~xs.x}.wx.qytr.~{px~};.
x}rt.p{{.x}sx
xsp{.wps.rt}.
wx~xt.~u.rw~}xr.|pxyp}p.t=.cwx.
s.~srts.~xx
t.t{.
vvtx}v.wp.|pxyp}p.w~{s.qt.
sxts.uwt.p.p.tp|t}.u~.
ptxt.x|{px~}.x}.WXe.pxt}=.
4/4 Urcuvkekv{!kp!Ownvkrng!Uengtquku!
^}{.~}t.p}s~|xts;.s~q{t<q{x}s;.
{prtq~<r~}~{{ts._wpt.A.s.
tp|x}ts.wt.tuutr.~u.|~zts.
|pxyp}p.~}.pxrx.x}.\b=.
cwx.s.p.r~}srts.q.R~t<.
Q{~~|.t.p{[email protected]}s.x.t}x{ts;.
b|~zts.rp}}pqx.u~.pxrx.x}.
|{x{t.r{t~xI.P.p}s~|xts;.
{prtq~<r~}~{{ts.xp{ =.cwt.qytr.
tt.B?.pxt}.xw.\b<p~rxpts.
pxrx.p}s.wps.|~stpt.x}rtpt.x}.
~}t.6r~t."%B.~x}.~}.wt.|~sxuxts.
Pw~w.rp{t7=._pxrxp}.tt.
p{{~ts.~.r~}x}t.~wt.\b.
|tsxrpx~};.xw.wt.trtx~}.~u.
qt}~sxptx}t=.Txvw.trt}.~u.
qytr.wps.p.wx~.~u.|pxyp}p.t.
p}s.BB3.wps.ts.|pxyp}p.xwx}.wt.
t
x~.tp=.
bqytr.pxrxpts.x}.~.B<sp.t.
tx~};.xw.p}.@@.sp.pw~.
tx~s=.Sx}v.tprw.t.tx~}.wt.
|~zts.p.C=?3.cWR.|pxyp}p.rxvptt.
~}rt.t.sp.~.p.{prtq~.rxvptt.~}rt.
t.sp=.b|~zx}v.~rrts.w~vw.p.
p}spsxts.rts<uu.~rtstI.6@7.
X}wp{px~}.u~.D.tr~}s;.6A7.qtpw<w~{s.
p}s.twp{px~}.u~.@?.tr~}s;.6B7.pt.
qttt}.uu.u~.CD.tr~}s=.bqytr.
r~|{tts.p}.p
tpvt.~u.u~.uu.t.
rxvptt=.
RQ!11111
Hto!11146
Hov!5812
Uhov!5813
64832!
cwt.x|p.~r~|t.|tpt.p.
rwp}vt.x}.pxrx.~}.wt.|~sxuxts.
Pw~w.rp{t=.Pssxx~}p{{;.qytr.
tt.ptts.x}v.p.ePb.u~.px};.p.
x|ts.p{z;.p}s.r~v}xx
t.t.6_prts.
Psx~.btxp{.Pssxx~}.ct7.p}s.PT=.
ctp|t}.xw.C=?3.cWR.|pxyp}p.
tsrts.qytr.r~t.~}.wt.|~sxuxts.
Pw~w.rp{t.q.p}.p
tpvt.~u.A=FC.
~x}.|~t.wp}.{prtq~.6r!K.?=???@7.
p}s.tsrts.ePb.px}.r~t.r~|pts.
~.{prtq~.6r!L.?=??G7=.br~t.~}.wt.
r~v}xx
t.|tpt.strtpts.q.G=F.
~x}.|~t.wp}.{prtq~.6r!L.?=??B7=.
W~t
t;.|pxyp}p.sxs.}~.puutr.
r~t.u~.wt.x|ts.p{z.r~|pts.~.
{prtq~=.\pxyp}p.x}rtpts.px}v.~u.
utt{x}v.wxvw.r~|pts.~.{prtq~=.
F.qytr.sxs.}~.r~|{tt.wt.s.
st.~.ps
tt.t
t}.6~.qytr.ut{.
}r~|u~pq{.wxvw ;.~.wps.
sxx}t.p}s.~}t.wps.upxvt7=.^u.w~t.
F.qytr.w~.xwst;.D.wps.{x{t.~.
}~.t
x~.ttxt}rt.xw.|pxyp}p=.
fwt}.wt.spp.tt.t<p}p{ts.~.
x}r{st.wtt.s~<~.qytr;.xw.
wt.t|x~}.wt.sxs.}~.wp
t.p.
~xx
t.t~}t.~.tp|t};.wt.tuutr.
~u.|pxyp}p.p.x{{.xv}xuxrp}.~}.
pxrx=.
cwt.pw~.r~}r{st.wp.|~zts.
|pxyp}p.wps.tu{}t.x}.tsrx}v.
px}.p}s.pxrx.p~rxpts.xw.\b=.
X.x.r~}rt}x}v.wp.|pxyp}p<}p'¶
t.
qytr.s~ts.~.~u.wt.s.
qtrpt.wt.tt.}pq{t.~.~{tpt.wt.
rwxpxr.PT.x}srts.q.|pxyp}p=.
cwt.pw~.vvt.wp.ut.sxt.
w~{s.tp|x}t.wtwt.sxuutt}.s~t.
rp}.t{.x}.x|x{p.qt}tuxrxp{.tuutr.
xw.{t.r~v}xx
t.x|pr=.W~t
t;.
wt.rt}.s.~srts.~xx
t.
t{.vvtx}v.wp.|pxyp}p.w~{s.
qt.sxts.uwt.p.p}.psy}r.
tp|t}.u~.pxrx.x}.\b.pxt}=.
4/5 Cuvjoc!
cpwzx}.t.p{[email protected]p|x}ts.
q~}rw~sx{px~}.x}.@?.qytr.xw.
q~}rwxp{.pw|p.x}.wt.s.t}x{ts;.
Prt.Tuutr.~u.b|~zts.\pxyp}p.
p}s.^•p{.% <ctpws~rp}}pqx}~{.~}.
btrxuxr.Pxp.R~}srp}rt.x}.
Pw|pxr.bqytr =.cwt.s.p.p.
s~q{t<q{x}s;.{prtq~<r~}~{{ts;.
r~~
t.stxv}=.P{{.qytr.tt.
r{x}xrp{{.pq{t.p.wt.x|t.~u.wt.sJ.
u~.qytr.tt.|~|.utt;.p}s.
x.qytr.wps.rw~}xr.|~|.~u.
|x{s.~.|~stpt.t
tx=.bqytr.tt.
tts.xw.?=AD|{.~u.x~~tt}~{.WR{.
x~.~.wt.s.~.t}t.wt.
t~}sts.~.q~}rw~sx{p~.
|tsxrpx~}=.bqytr.tt.}~.p{{~ts.
~.pzt.q~}rw~sx{p~.|tsxrpx~}.
xwx}.G.w~.x~.~.wt.s=.
_t
x~.ttxt}rt.xw.|pxyp}p.p.
}~.txts.u~.pxrxpx~}.x}.wt.
s;.q.F.~u.wt.@?.qytr.t~ts.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64833!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
t
x~.t.~u.|pxyp}p.p.p.pt.~u.
{t.wp}.@.|pxyp}p.rxvptt.t.
|~}w=.]~.qytr.t~ts.|pxyp}p.
t.xwx}.F.sp.~u.wt.s=.
cwt.s.r~}xts.~u.u~.t.
tx~}.xw.p}.x}t
p{.~u.p.{tp.CG.
w~.qttt}.tx~}=.^}.~.t.
tx~}.qytr.|~zts.F.|v>zv.~u.
q~s.txvw.~u.txwt.|pxyp}p;.xw.
A3.cWR.r~}rt}px~}.q.txvw;.~.
{prtq~.|pxyp}p=.Sx}v.wt.~wt.~.
t.tx~};.qytr.x}vtts.rp{t.
xw.txwt.@D|v.~u.}wtxr.cWR.~.
{prtq~=.\pxyp}p.p.ps|x}xtts.
x}v.p.}xu~|.|~zx}v.trw}xtI.
qytr.x}wp{ts.stt{.u~.A¤C.tr~}s;.
wt{s.|~zt.x}.{}v.u~.@D.tr~}s;.p}s.
t|ts.}~|p{.qtpwx}v.u~.
p~x|pt{.D.tr~}s=.cwt.pw~.
sxs.}~.~
xst.p.strxx~}.~u.wt.
}|qt.~u.uu.pzt}.p.p}.|~zx}v.
tx~}=.cwt.pw~.pt.wp.wt.
|~zx}v.~rtst.p.ttpts.}x{.
wt.rxvptt.p.r~}|ts;.wxrw.~~z.
p~x|pt{.@?.|x}t=.
cwt.~r~|t.|tpt.ts.p.
trxuxr.pxp.r~}srp}rt.6bVp7;.p.
rp{r{pts.x}v.|tpt|t}.~u.
w~prxr.vp.
~{|t.6cVe7.p}s.pxp.
txp}rt.6ap7.x}v.p.
pxpq{t<.
tt.q~s.{tw|~vpw=.
Pssxx~}p{{;.p}.pt|t}.~u.stvtt.~u.
x}~xrpx~}.p.ps|x}xtts.~}{.~.
w~t.qytr.t~x}v.t
x~.
|pxyp}p.t=.cwx.pt|t}.
r~}xts.~u.qytr.px}v.w~.wxvw .
wt.ut{ .~}.p.rp{t.~u.?¤F;.F.
ttt}x}v.wt.wxvwt .wt.wps.t
t.
ut{.put.|~zx}v.|pxyp}p =.
\pxyp}p.~srts.p.xv}xuxrp}.
x}rtpt.~u.BB¤CG3.x}.p
tpvt.bVp.
r~|pts.~.q~w.qpt{x}t.p}s.{prtq~.
6_.K.?=?D7=.cwx.xv}xuxrp}.x}rtpt.x}.
bVp.{pts.u~.p.{tp.A.w~.put.
ps|x}xpx~}=.cwt.p
tpvt.cVe.
xv}xuxrp}{.strtpts.q.C¤@B3.
r~|pts.~.qpt{x}t.p}s.{prtq~.6_.K.
?=?D7=.cwt.pw~.pts.wp.p{{.qytr.
t~ts.utt{x}v.~u.x}~xrpx~}.put.
|pxyp}p.ps|x}xpx~}=.
cwt.pw~.r~}r{st.wp.|pxyp}p.
~srts.q~}rw~sx{px~}.x}.r{x}xrp{{.
pq{t.pw|pxr.qytr.xw.|x}x|p{.
~.|~stpt.q~}rw~p|=.bs.
{x|xpx~}.x}r{stI.x}r{x~}.~u.
qytr.xw.
px}v.t
tx.~u.
pw|pxr.|~|;.t.~u.bVp.~.
|tpt.{}v.t~}t.~.|pxyp}p.
ps|x}xpx~};.p}s.ps|x}xpx~}.~u.
|~zt.~.pw|pxr.qytr=.b|~zt.
st{x
t.p.}|qt.~u.wp|u{.qp}rt.
p}s.x.}~.p}.~x|p{.st{x
t.|~|;.
ttrxp{{.u~.pw|pxr.pxt}=.UTe@.
xp.x~|t.x.wt.v~{s.p}sps.~.
pt.rwp}vt.x}.{}v.u}rx~};.t.p}s.
~.pw|p.tp|t};.q.
wp|pr~wtp=.bVp.wp.qtt}.
w~}.~.qt.p.
p{xs.~~{.x}.
q~}rw~r~}xrx~}.{}v.pt|t}J.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
}st.st
t{~|t};.wt.tuutr.~ut}.
t{pt.~.p.w~<t|.r{x}xrp{.~r~|t.
qtx}v.x}
txvpts=._~~u<~u<r~}rt.
sxt.t
t.p.wt.{x}z.qttt}.
tr{x}xrp{.sxt.p}s.s~t.p}vx}v.
4/6 Incweqoc!
r{x}xrp{.sxt=.cwttu~t;.~~u<~u<.
c~.p}s~|xts;.s~q{t<q{x}s;.
r~}rt.sxt.pt.}~.uuxrxt}.~.
{prtq~<r~}~{{ts._wpt.A.r{x}xrp{.
st|~}pt.tuuxrpr.~u.p.sv.qtrpt.
sxt.tp|x}ts.|~zts.|pxyp}p.x}. wt.~
xst.~}{.t{x|x}p.
v{pr~|p.6Rpu~s.p}s.\tx;.@HFHJ.
x}u~|px~}.pq~.wt.tuutr.~u.p.sv=.
\tx.t.p{=;.@HG?7=.X}.q~w.sxt;.
P{w~vw.wtt.sxt.s~.}~.~
xst.
x}p~r{p.tt.6X^_7.p.
t
xst}rt.wp.|pxyp}p.x.tuutrx
t.x}.
xv}xuxrp}{.tsrts.B?.|x}t.put.
tpx}v.p.trxuxr;.tr~v}xts.sx~st;.
|~zx}v.|pxyp}p=.\px|p{.tuutr.
wtt.sxt.s~.~.ut.{pvt.
~rrts.E?¤H?.|x}t.put.|~zx}v;.
t{{<r~}~{{ts.sxt.~.pt.wt.
xw.X^_.t}x}v.~.qpt{x}t.xwx}.B¤.
put.p}s.tuuxrpr.~u.|pxyp}p.u~.p.
C.w~=.cwtt.~.sxt.tt.
trxuxr.|tsxrp{.x}sxrpx~}=.^
tp{{;.wt.
x}r{sts.x}.[email protected]^\.t~.~}.wt.
r~}r{x~}.qt{~.pt.t{x|x}p;.
|tsxrp{.t.~u.|pxyp}p=.Qtrpt.~.
qpts.~}.
t.{x|xts.t
xst}rt=.
x}stt}st}.p}p{x.~u.wtt.sxt.
r~}rts.xw.wt.r~}r{x~}.u~|.
B=E=@ R~}r{x~}.u~.Rw~}xr.
[email protected]^\.t~;.wtt.sxt.x{{. ]t~pwxr._px}.
}~.qt.sxrts.x}.uwt.stpx{.x}.wx.
X}.qytr.xw.rw~}xr.}t~pwxr.
t
xt=.]~.trt}.sxt.wp
t.qtt}.
px}.w~.pt.tupr~.~.~wt.px}.
r~}srts.tp|x}x}v.wt.tuutr.~u.
tp|t};.ux
t.~~u<~u<r~}rt.sxt.
x}wp{ts.|pxyp}p.~}.X^_.x}.v{pr~|p.
pxt}=.cwx.{prz.~u.trt}.sxt.|p. ~srts.~xx
t.t{.tvpsx}v.wt.
qt.pxqts.~.wt.r~}r{x~}.|pst.x}. t.~u.|~zts.|pxyp}p.u~.p}p{vtxp=.
W~t
t;.wt.qytr.x}.wtt.sxt.
[email protected]^\.t~.wp.wx{t.
r~}x}ts.~.t.wtx.rt}.p}p{vtxr.
rp}}pqx}~xs.rp}.tsrt.x}p~r{p.
tt.6X^_7;.wt.wtptxr.tuutr.
sv.tvx|t;.p}s.w.}~.r~}r{x~}.
txt.wxvw.s~t.wp.~srt.w~<.
rp}.qt.|pst.tvpsx}v.wt.~t}xp{.
{px}v.t~}t;.xw.p.wxvw.stvtt.~u. tuuxrpr.~u.|pxyp}p.u~.}t~pwxr.
PT=.cwx.wxvw.stvtt.~u.PT.|tp}.wp. px}.x}.pxt}.}~.pzx}v.~wt.
wt.~t}xp{.wp|u{.tuutr.~u.rw~}xr. p}p{vtxr.sv=.bqytr.p{~.wps.
|pxyp}p.|~zx}v.|p.~txvw.x.
}|t~.u~|.~u.}t~pwxr.px};.
|~st.qt}tux.x}.wt.tp|t}.~u.
|pzx}v.x.sxuuxr{.~.xst}xu.wtwt.p.
v{pr~|p=.
trxuxr.t.~u.|~|.|xvw.qt.|~t.
t~}x
t.~.wt.tuutr.~u.|pxyp}p=.X.
4/7 Eqpenwukqpu!
x.ttrxp{{.r~}rt}x}v.wp.~|t.
^u.wt.t{t
t}.p}s~|xts;.s~q{t<.
|pxyp}p<}p'¶
t.qytr.wps.x}~{tpq{t.
q{x}s;.{prtq~<r~}~{{ts._wpt.A.
rwxpxr.t~}t.~.|pxyp}p.
r{x}xrp{.sxt.wp.|t.wt.rxtxp.u~.
t~t.p.p}p{vtxr.s~t=.
t
xt.6tt.btrx~}.A=A.p}s.A=B7;.t}.
sxt.ps|x}xtts.|pxyp}p.w~vw. B=E=A R~}r{x~}.u~.Ptxt.
bx|{px~}.x}.WXe.
|~zx}v;.wx{t.~}t.s.x{xts.
|pxyp}p.
p~xpx~}=.X}.wtt.t{t
t}.
X}.qytr.w~.tt.WXe<~xx
t;.
sxt;.wtt.tt.ux
t.sxuutt}.
wtptxr.x}sxrpx~}I.Ux
t.tp|x}ts. ~.~~u<~u<r~}rt.sxt.~srts.
~xx
t.t{.xw.wt.t.~u.q~w.
rw~}xr.}t~pwxr.px};.~.
s~}pqx}~{.p}s.|~zts.|pxyp}p.~.
tp|x}ts.ptxt.x|{px~}.x}.WXe.
x}rtpt.rp{~xr.x}pzt.p}s.~srt.
pxt};.~.tp|x}ts.v{pr~|p;.~}t.
txvw.vpx}.x}.WXe<~xx
t.pxt}=.
tp|x}ts.pxrx.x}.\b;.p}s.~}t.
W~t
t;.wt.p|~}.~u.cWR.x}.wt.
tp|x}ts.pw|p=.
cwtt.pt.{x|xts.r~}r{x~}.wp.rp}. |pxyp}p.tts.x}.wtt.sxt.x.u~.
x|t.vtpt.wp}.wt.s~t.~u.
qt.sp}.u~|.wt.spp.x}.wtt.
q{xwts.sxt.t
p{px}v.|pxyp}p. s~}pqx}~{.xrp{{.tts.u~.ptxt.
u~.wt.tp|t}.~u.sxuutt}.wtptxr. x|{px~}.6@?.|v.
=.A=D.|vJ.Wp}t.t.
p{=;.A??D7=.cw;.x.x.~xq{t.wp.wt.
x}sxrpx~}=.cwt.p}p{x.t{xts.~}.
{~.stvtt.~u.PT.t~ts.x}.wx.s.
q{xwts.sxt;.w.x}u~|px~}.
|p.tu{tr.wt.st
t{~|t}.~u.
p
px{pq{t.pq~.~~r~{;.~rtst;.
~{tp}rt.~.rp}}pqx}~xs.x}.wx.pxt}.
p}s.t{.tt.{x|xts.~.s~r|t}.
~{px~};.x}rt.p{{.x}sx
xsp{.wps.
q{xwts.p}s.xst{.p
px{pq{t.x}.wt.
rt}.wx~xt.~u.rw~}xr.|pxyp}p.
q{xr.s~|px}=.cwt.q{xwts.sxt.
t=.cw;.x}sx
xsp{.xw.{x{t.x~.
~}.|tsxrp{.|pxyp}p.pt.tuutrx
t{.
t~t.~.|pxyp}p.|p.}~.t~}s.
~~u<~u<r~}rt.sxt=._~~u<~u<.
x|x{p{.p}s.|p.}~.qt.pq{t.~.~{tpt.
r~}rt.sxt.~
xst.t{x|x}p.
uuxrxt}.|pxyp}p.~.~srt.ptxt.
t
xst}rt.~}.p.~~ts.w~wtx.
x|{px~}=.
tvpsx}v.p.sv .tuutr=.U~.sv.
w~t
t;.x}rt.wt.UTe@.|tw~s.p.
}~.x{xts;.x.x.}r{tp.wtwt.wtt.
t{.~{s.r~t{pt.xu.wt.UTe@.
|tw~s.wps.qtt}.t|{~ts=.
RQ!11111
Hto!11147
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
B=E=B
R~}r{x~}.u~.bpxrx.x}.\b.
X}.qytr.xw.\b;.p.~~u.~u.
r~}rt.s.~srts.~xx
t.t{.
x}v.|~zts.|pxyp}p.p.p.tp|t}.
u~.px}.p}s.|~|.p~rxpts.xw.
tp|t}<txp}.pxrx=.cwt.
qytr.x}.wx.s.r~}x}ts.~.pzt.
wtx.rt}.|tsxrpx~}.tvx|t};.p}s.
w.}~.r~}r{x~}.rp}.qt.|pst.
tvpsx}v.wt.~t}xp{.tuuxrpr.~u.
|pxyp}p.wt}.pzt}.~}.x.~}=.X.x.
p{~.r~}rt}x}v.wp.|pxyp}p<}p'¶
t.
qytr.s~ts.~.~u.wt.s.
qtrpt.wt.tt.}pq{t.~.~{tpt.wt.
rwxpxr.PT.x}srts.q.|pxyp}p=.
cwt.pw~.vvt.wp.ut.sxt.
w~{s.tp|x}t.wtwt.sxuutt}.s~t.
rp}.t{.x}.x|x{p.qt}tuxrxp{.tuutr.
xw.{t.r~v}xx
t.x|pr=.
B=E=C
R~}r{x~}.u~.Pw|p.
X}.qytr.xw.r{x}xrp{{.pq{t.
pw|p;.p.~~u.~u.r~}rt.s.
~srts.~xx
t.t{.~u.|~zts.
|pxyp}p.~srx}v.q~}rw~sx{px~}=.
W~t
t;.x}.wx.s.|pxyp}p.p.
ps|x}xtts.p.t.p}s.}~.wx{t.
ttxt}rx}v.q~}rw~p|=.
Pssxx~}p{{;.wt.ps|x}xpx~}.~u.
|pxyp}p.w~vw.|~zx}v.x}~srt.
wp|u{.p}s.xxpx}v.qp}rt.~.wt.
qytr;.wxrw.x.}stxpq{t.ttrxp{{.
x}.pw|pxr.pxt}=.cw.wt.t{.
vvt.|pxyp}p.|p.wp
t.
q~}rw~sx{p~.tuutr;.q.x.|p.p{~.
wp
t.}stxpq{t.ps
tt.tuutr.x}.
qytr.xw.pw|p=.
B=E=D
R~}r{x~}.u~.V{pr~|p.
P.}~ts.x}.btrx~}.B=D;.wt.~.
sxt.wp.t
p{pts.|~zts.
|pxyp}p.u~.v{pr~|p.tt.r~}srts.
strpst.pv~;.p}s.wt.wp
t.qtt}.
w~~vw{.t
p{pts.x}.[email protected]^\.
t~[email protected]^\.t~.r~}r{st.
wp.wx{t.wt.sxt.xw.|pxyp}p.
w~ts.~xx
t.t{.u~.tsrx~}.x}.
X^_;.wt.tuutr.x.w~<{px}v;.txt.
p.wxvw.s~t;.p}s.x.p~rxpts.xw.
|p}.PT=.cw;.wt.~t}xp{.wp|u{.
tuutr.|p.~txvw.p}.|~st.
qt}tux.~u.|pxyp}p.u~.wx.r~}sxx~}=.
ft.pvtt.xw.wt.r~}r{x~}.sp}.x}.
[email protected]^\.t~=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
4/8 Fgukip!Ejcnngpigu!hqt!Hwvwtg!
Uvwfkgu!
cwt.~xx
t.t{.t~ts.q.wt.
sxt.sxrts.x}.wx.t
xt.~.
wt.r~}sr.~u.|~t.xv~~.sxt.x}.
wt.ut=.cwx.trx~}.sxrt.
|tw~s~{~vxrp{.rwp{{t}vt.wp.wp
t.
~rrts.x}.r{x}xrp{.sxt.xw.
|~zts.|pxyp}p=.cwtt.stxv}.xt.
w~{s.qt.psstts.wt}.{pvt<rp{t.
r{x}xrp{.sxt.pt.r~}srts.~.t}t.
wp.
p{xs.rxt}xuxr.spp.pt.vt}tpts.
x}.sxt.t
p{px}v.|pxyp}p .put.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
p}s.tuuxrpr.u~.p.pxr{p.wtptxr.
t=.
B=F=@ bp|{t.bxt.
cwt.pqx{x.u~.t{.u~|.p.r{x}xrp{.
s.~.qt.vt}tp{xts.~.p.q~pst.
~{px~}.x.t{xp}.~}.wp
x}v.p.
uuxrxt}{.{pvt.s.p|{t.xt=.
W~t
t;.p.}~ts.pq~
t;.p{{.~u.wt.@@.
sxt.t
xtts.x}.wx.s~r|t}.tt.
tp{._wpt.A.~~u.~u.r~}rt.sxt.
u~.tuuxrpr.p}s.put=.cw;.wt.p|{t.
xt.ts.x}.wtt.sxt.tt.
x}wtt}{.|p{{;.p}vx}v.u~|.@?.
qytr.t.tp|t}.v~.6cpwzx}.t.
p{=;[email protected]}t.t.p{=;.A??F7.~.AD.
qytr.t.tp|t}.v~.6Pqp|.t.
p{=;.A??F7=.cwtt.p|{t.xt.pt.
pxxrp{{.x}pstpt.~.~.p.
w~x}v.~u.put.~.tuuxrpr=.USP .
tr~||t}spx~}.pq~.p|{t.xt.u~.
r{x}xrp{.xp{.rp}.qt.u~}s.x}.wt.
Iwkfcpeg!hqt!Kpfwuvt{<!G;!Uvcvkuvkecn!
Rtkpekrngu!hqt!Enkpkecn!Vtkcnu!)2;;9*/ .
U~.tp|{t;.wt.}|qt.~u.qytr.x}.
p.r{x}xrp{.xp{.w~{s.p{p.qt.{pvt.
t}~vw.~.~
xst.p.t{xpq{t.p}t.~.
wt.tx~}.psstts=.cwx.}|qt.x.
p{{.stt|x}ts.q.wt.x|p.
~qytrx
t.~u.wt.xp{=.cwt.|tw~s.q.
wxrw.wt.p|{t.xt.x.rp{r{pts.
w~{s.qt.vx
t}.x}.wt.~~r~{;.~vtwt.
xw.wt.tx|pt.~u.p}.p}xxt.
ts.x}.wt.rp{r{px~}.6rw.p.
pxp}rt;.|tp}.
p{t;.t~}t.pt;.
t
t}.pt;.sxuutt}rt.~.qt.sttrts7= .
6v=.A@7=.^wt.r{x}xrp{.USP.Iwkfcpeg!
hqt!Kpfwuvt{ |p.p{~.r~}px}.
tr~||t}spx~}.tvpsx}v.wt.
p~xpt.}|qt.~u.qytr.wp.
w~{s.qt.x}
txvpts.u~.p.trxuxr.
|tsxrp{.x}sxrpx~}=.
B=F=A \pxyp}p.S~t.bp}spsxpx~}.
S~t.p}spsxpx~}.x.rxxrp{.u~.
p}.r{x}xrp{.s.x}.~st.~.t}t.
wp.tprw.qytr.trtx
t.p.r~}xt}.
t~t.~.wt.t.sv=.cwt.Iwkfcpeg!
hqt!Kpfwuvt{<!Dqvcpkecn!Ftwi!Rtqfwevu!
6A??C7 ~
xst.trxuxr.x}u~|px~}.
~}.wt.st
t{~|t}.~u.q~p}xrp{.sv.
~sr=.btrxuxrp{{;.wx.vxsp}rt.
x}r{st.x}u~|px~}.pq~.wt.}tts.u~.
t{{<rwpprtxts.p}s.r~}xt}.
rwt|x.u~.wt.q~p}xrp{.{p}.
~sr.p}s.u~.r~}xt}.p}s.t{xpq{t.
cwt.Iwkfcpeg!hqt!Kpfwuvt{<!G;!Uvcvkuvkecn!
Rtkpekrngu!hqt!Enkpkecn!Vtkcnu!rp}.qt.u~}s.pI.
yyy/hfc/iqx0fqypnqcfu0Ftwiu0
IwkfcpegEqornkcpegTgiwncvqt{Kphqtocvkqp0
Iwkfcpegu0weo184248/rfh/!
^wt.Vxsp}rt.u~.X}s.rp}.qt.u~}s.pI.
yyy/hfc/iqx0Ftwiu0
IwkfcpegEqornkcpegTgiwncvqt{Kphqtocvkqp0
Iwkfcpegu0weo175;92/jvo/!
cwt.Iwkfcpeg!hqt!Kpfwuvt{<!Dqvcpkecn!Ftwi!
Rtqfwevu!rp}.qt.u~}s.pI.jvvr<00yyy/hfc/iqx0
fqypnqcfu0Ftwiu0
IwkfcpegEqornkcpegTgiwncvqt{Kphqtocvkqp0
Iwkfcpegu0weo1815;2/rfh=.
RQ!11111
Hto!11148
Hov!5812
Uhov!5813
64834!
s~x}v=.btrxuxrp{{.u~.|pxyp}p.
sxt;.s~t.p}spsxpx~}.x.
x|~p}.qtrpt.xu.|pxyp}p.{tps.~.
{p|p.{t
t{.~u.rp}}pqx}~xs.wp.pt.
xv}xuxrp}{.sxuutt}.qttt}.qytr;.
wx.
pxpx~}.|p.{tps.~.sxuutt}rt.x}.
wtptxr.t~}x
x.~.x}.wt.
t
p{t}rt.~u.rwxpxr.PT=.
X}.|~.|pxyp}p.sxt.sxrts.
x}.wx.t
xt;.x}
txvp~.t.p.
p}spsxts.rts.|~zx}v.~rtst=.
X}.wx.~rtst;.p.qytr.x.x}rts.
~.x}wp{t.|pxyp}p.|~zt.u~.D.
tr~}s;.w~{s.wt.|~zt.x}.wt.{}v.u~.
@?.tr~}s;.twp{t.p}s.qtpwt.
}~|p{{.u~.C?.tr~}s=.cwx.~rt.x.
ttpts.~.~qpx}.wt.stxts.s~t.~u.
wt.sv=.W~t
t;.wx.~rtst.|p.
}~.{tps.~.tx
p{t}.t~t.~.
|pxyp}p.p}s.x.r~}xt}.
rp}}pqx}~xs;.qpts.~}.t
tp{.upr~I.
X}t}x~}p{.~.}x}t}x~}p{.
sxuutt}rt.x}.wt.stw.~u.x}wp{px~}.
|p.rwp}vt.wt.p|~}.~u.|~zt.x}.wt.
qytr .{}v=.
b|~zx}v.t{.x}.{~.u~|.xst.
tp|.|~zt;.rw.wp.wt.t}xt.s~t.
x.}~.st{x
tts.~.wt.qytr=.
cwtt.|p.qt.sxuutt}rt.x}.cWR.
r~}rt}px~}.p{~}v.wt.{t}vw.~u.p.
|pxyp}p.rxvptt=.Prr~sx}v.~.
cpwzx}.t.p{=.6@HH@7;.wt.ptp.~u.wt.
rxvptt.r{~t.~.wt.|~w.t}s.~.
prr|{pt.p.wxvwt.r~}rt}px~}.~u.
cWR;.q.wx.trx~}.~u.wt.rxvptt.x.
}~.|~zts.sx}v.p.s=.
U~.tp|{t;.fx{t.t.p{=.6A??G7.ts.
wx.p}spsxts.|~zx}v.~rtst=.
cwt.t~ts.|tp}.6p}vt7.~u.|pxyp}p.
rxvptt.r~}|ts.p.DD?.|v.6A??¤.
GB?|v7.u~.wt.{~.t}vw.|pxyp}p.
6B=D3.cWR7.p}s.CH?.|v.6AF?¤GF?|v7.
u~.wt.wxvw.t}vw.|pxyp}p.6F3.
cWR7=.cwx.xst.p}vt.~u.p|~}.~u.
|pxyp}p.rxvptt.|~zts.q.wt.
x}sx
xsp{.qytr;.t
t}.xw.
p}spsxts.|~zx}v.~rtst.p}s.
r~}~{{ts.}|qt.~u.uu;.~.wt.
xt.xw.st{x
tx}v.r~}xt}.s~t.
xw.|~zt.|pxyp}p=.
X}.~wt.|pxyp}p.sxt.wp.s~.}~.
t.p.rts.|~zx}v.~rtst;.qytr.
pt.x|{.~{s.~.|~zt.wt.|pxyp}p.
rxvptt.~
t.p.trxuxr.p|~}.~u.x|t.
6p{{.@?.|x}t7.xw~.uwt.
x}rx~}.6Rpu~s.p}s.\tx;.@HFHJ.
\tx.t.p{=;.@HG?J.T{{x.t.p{=;.A??H7=.
cwt.t.~u.p.}~}p}spsxts.~rtst.
|p.{tps.~.}~}<tx
p{t}.t~t.~.
|pxyp}p.p}s.x.r~}xt}.
rp}}pqx}~xs.qttt}.qytr.qtrpt.
~u.pssxx~}p{.upr~.wp.pt.}~.{xts.
pq~
t;.rw.pI.
Sxuutt}rt.x}.pq~x~}.p}s.sv.
t~}t.xu.qytr.6ttrxp{{.
|pxyp}p<}p'¶
t.~}t7.pt.}~.x}rts.
~.w~{s.|pxyp}p.|~zt.x}.wtx.{}v.
u~.p.rtpx}.tx~s.~u.x|t=.
G<^HT^HO^23CWR3/UIO
23CWR3
64835!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
_~{~}vts.tx~s.qttt}.uu.
|p.x}rtpt.{~.~.xst.tp|.|~zt=.
bqytr.|p.pt|.~.|~zt.wt.
|pxyp}p.rxvptt.x}.wt.p.wt.
~{s.|~zt.p.~qprr~.rxvptt;.wxrw.
t{xt.x|px{.~}.w~;.wp{{~.uu=.
X}.q~w.p}spsxts.p}s.}~}<.
p}spsxts.|~zx}v.~rtst;.
qytr.|p.ttz.~.r~}~{.wt.s~t.~u.
cWR.w~vw.t{u<xpx~}.6Rpu~s.
p}s.\tx;.@HFHJ.\tx.t.p{=;.@HG?J.
cpwzx}.t.p{=;[email protected]p|.t.p{=;.A??FJ.
T{{x.t.p{=;.A??H7=.bt{u<xpx~}.x}
~{
t.
p}.x}sx
xsp{.|~stpx}v.wt.p|~}.~u.
|pxyp}p.|~zt.x}wp{ts.~
t.x|t.x}.
~st.~.~qpx}.p.tutts.{t
t{.~u.
rw~prx
t.~.r{x}xrp{.t~}t=.cwt.
pqx{x.~u.p}.x}sx
xsp{.~.t{u<xpt.q.
|~zx}v.x.~}t.tp~}.vx
t}.q.
ps
~rpt.~u.|tsxrp{.|pxyp}p .x}.
~.~u.|~zx}v.~u.|pxyp}p.pwt.
wp}.w~vw.x.x}vtx~}.
xp.tsxq{t=.
W~t
t;.u~.ttprw.~t;.t{u<.
xpx~}.x}tutt.xw.wt.pqx{x.~.
|px}px}.r~}xt}.s~x}v.{t
t{.
qttt}.qytr;.p}s.w;.
p{xs.
r~|px~}.qttt}.s.v~=.
P{{.~u.wtt.upr~.rp}.|pzt.wt.tpr.
s~t.~u.rp}}pqx}~xs.trtx
ts.q.p.
qytr.x}.p.|pxyp}p.s.sxuuxr{.~.
stt|x}t.xw.prrpr=.ctx}v.
wtwt.{p|p.{t
t{.~u.cWR.~.~wt.
rp}}pqx}~xs.pt.x|x{p.qttt}.
qytr.u~{{~x}v.wt.|~zx}v.
~rtst.~{s.tpq{xw.wtwt.wt.
~rtst.x.~srx}v.p~xpt.
t{=.Pssxx~}p{{;.sxt.r~{s.qt.
r~}srts.~.stt|x}t.xu.
p~xpx~}.
rp}.qt.ts.~.st{x
t.r~}xt}.s~t.
~u.rp}}pqx}~xs.u~|.|pxyp}p.{p}.
|ptxp{=.btrxuxrp{{;.
p~xpx~}.
st
xrt.wp.x}
~{
t.wt.r~{{trx~}.~u.
p~.x}.p}.t}r{~ts.qpv.~.rwp|qt.
|p.wt{.xw.st{x
t.~u.r~}xt}.
s~t.~u.|pxyp}p=.cw;.|~t.
x}u~|px~}.r~{s.qt.r~{{trts.~}.
wtwt.
p~xpx~}.x.r~|ppq{t.~.
~.sxuutt}.wp}.|~zx}v.x}.t|.~u.
~srx}v.x|x{p.{p|p.{t
t{.~u.cWR.
x}.qytr.x}v.xst}xrp{.|pxyp}p.
{p}.|ptxp{=.
wxrp{.stt}st}rt;.~.trxuxr.
ps
tt.t
t}.st
t{~.~
t.wt.r~t.
~u.x|t.xw.r~}x}x}v.t.~u.
wtptxr.|pxyp}p=.
B=F=C b|~zx}v.p.p.a~t.~u.
Ps|x}xpx~}.
P.wp.qtt}.~x}ts.~.q.wt.X^\.
p}s.~wt.v~;.|~zx}v.x.}~.p}.
~x||.~t.~u.ps|x}xpx~}.u~.
|pxyp}p<stx
ts.wtptxr.sv.
~sr;.x|px{.qtrpt.x}~srx}v.
wt.|~zt.u~|.p.q}.q~p}xrp{.
qp}rt.x}~.wt.{}v.~u.x}sx
xsp{.
xw.p.sxtpt.pt.x.}~.tr~||t}sts.
wt}.wtx.q~sxt.|p.qt.wxrp{{.
r~|~|x[email protected]^\.t~.~}.
|tsxrx}p{.t.~u.|pxyp}p.}~ts.wp.
p{t}px
t.st{x
t.|tw~s.~uutx}v.
wt.p|t.pqx{x.~u.s~t.xpx~}.p.
|~zx}v.|pxyp}p.x{{.qt.qt}tuxrxp{.
p}s.|p.{x|x.~|t.~u.wt.~xq{t.{~}v<.
t|.wtp{w.r~}tt}rt.~u.|~zx}v.
|pxyp}p=.cwt.x|p.p{t}px
t.~.
|~zts.|pxyp}p.x.
p~xpx~};.
wxrw.rp}.tsrt.t~t.~.
r~|qts.{p}.|ptxp{.r~}px}x}v.
rp}}pqx}~xs=.cwt.~}{.s.~.t.
p~xpx~}.p.wt.st{x
t.|tw~s.p.
fx{t.t.p{=.6A?@B7=.cwt.t{.u~|.
fx{t.t.p{=.6A?@B7.w~ts.p.x|x{p.
tuutr.~u.strtpts.px}.p.tt}.x}.wt.
~wt.sxt.x}v.|~zx}v.p.wt.
st{x
t.|tw~s.6fpt.t.p{=;.A?@?J.
fx{t.t.p{=;.A??G7=.cwx.x|x{p.tuutr.
~u.strtpt.px}.~.
p~xpx~}.
p.p.~xq{.
xpq{t.~t.~.ps|x}xt.
|pxyp}p.x}.ttprw;.wx{t.~t}xp{{.
{x|xx}v.wt.xz.p~rxpts.xw.
|~zx}v=.
B=F=D Sxuuxr{.x}.Q{x}sx}v.~u.Sv.
R~}sxx~}.
P}.pstpt.p}s.t{{<r~}~{{ts.
r{x}xrp{.s.x}
~{
t.s~q{t<q{x}sx}v;.
wtt.q~w.wt.qytr.p}s.wt.
x}
txvp~.pt.}pq{t.~.t{{.wt.
sxuutt}rt.qttt}.wt.t.tp|t}.
6xrp{{.r~}xx}v.~u.p.{tp.p.t.
sv.p}s.{prtq~7.wt}.wt.pt.
ps|x}xtts=.P{{.~u.wt.sxt.
t
xtts.x}.wx.s~r|t}.ps|x}xtts.
B=F=B Prt.
=.Rw~}xr.cwtptxr.
s.tp|t}.}st.s~q{t<q{x}s.
\pxyp}p.dt.
r~}sxx~}.p}s.w.tt.r~}xstts.~.
cwt.sxt.wp.tt.t
xtts.
wp
t.p}.p~xpt.s.stxv}=.
ps|x}xtts.wt.sv.u~.w~.
W~t
t;.t
t}.}st.wt.|~.
spx~}.{px}v.}~.{~}vt.wp}.D.sp.
xv~~.ttx|t}p{.r~}sxx~};.
6Pqp|.t.p{=;.A??FJ.T{{x.t.p{=;.A??HJ.
q{x}sx}v.rp}.qt.sxuuxr{.x}.sxt.xw.
fpt.t.p{=;.A?@?7=.cw.p{{.sxt.
|~zts.|pxyp}p.qtrpt.wt.pxs.
tp|x}ts.wt.w~<t|.tuutr.~u.
~}t.~u.rw~prx
t.tuutr.tpsx{.
|pxyp}p.ps|x}xpx~}.u~.
sxx}vxwt.prx
t.u~|.{prtq~.
wtptxr.~t=.W~t
t;.|p}.~u. |pxyp}p=.cwt.tt}rt.~u.
wt.|tsxrp{.r~}sxx~}.wp.wp
t.qtt}.
rw~prx
t.tuutr.p{~.~rr.xw.
sxts.pt.txt}.~.ttrts.~.{p. ~wt.sv=.W~t
t;.|~.~wt.sv.
wt.t.~u.p.pxt} .{xut=.cwttu~t;.
wp
t.p.x|x{p.rw~prx
t.tuutr.xw.
spp.~}.rw~}xr.t~t.~.|~zts.
qp}rt.xw.x|x{p.|trwp}x|.~u.
|pxyp}p.x}.r{x}xrp{.sxt.x.}ttsts=. prx~}=.cwtt.qp}rt.rp}.qt.ts.p.
X}.wx.p;.|~t.x}u~|px~}.x{{.qt.
~xx
t.r~}~{.~.wt{.|px}px}.
p
px{pq{t.tvpsx}v.wtwt.~{tp}rt;.
q{x}sx}v.~.wt.prx
t.sv.qtx}v.tts=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11149
Hov!5812
Uhov!5813
\pxyp}p.~}.wt.~wt.wp}s;.wp.p.
}xt.t.~u.rw~prx
t.tuutr.wxrw.
|pzt.wt.t.~u.p~xpt.~xx
t.
r~}~{.sxuuxr{.6Qpt.t.p{=;.@HHD7=.
W~t
t;.~.sxt.sxs.t.
S~}pqx}~{.p.p.~xx
t.r~}~{.sv.~.
wt{.|px}px}.q{x}sx}v.6Wp}t.t.p{=;.
A??DJ.Wp}t.t.p{=;.A??F7=.
fwt}.q{x}sx}v.x.s~}t.x}v.~}{.
{prtq~.|pxyp}p;.wt.pqx{x.~.
sxx}vxw.prx
t.u~|.{prtq~.
|pxyp}p.|p.{tps.~.ttrpx~}.qxp.
p}s.p}.p{tpx~}.x}.trtx
ts.
t~}x
x.~.wt.wtptxr.~r~|t.
|tpt=.fxw.|pxyp}p<ttxt}rts.
qytr;.u~.tp|{t;.wtt.|p.qt.p}.
tp{.tr~v}xx~}.~u.wt.|~t.q{t.
rp}}pqx}~xs.tuutr.wp.rp}.t
t.p.p.
wpqx}vt.~u.~}vt.tuutr;.wxrw.x.
{t.{xzt{.~.~rr.xw.|pxyp}p<.
}p'¶
t.qytr=.c~.tsrt.wx.
~xqx{x;.x}
txvp~.wp
t.tts.
s~t.~u.|pxyp}p.~wt.wp}.wt.~}t.
wt.tt.x}ttts.x}.ttx|t}p{{.
~.|px}px}.wt.q{x}s.6fpt.t.p{=;.A?@?7=.
Q{x}sx}v.rp}.p{~.qt.r~|~|xts.q.
sxuutt}rt.x}.wt.ptpp}rt.~u.
|pxyp}p.{p}.|ptxp{.qpts.~}.cWR.
r~}rt}px~}=.\pxyp}p.xw.wxvwt.
r~}rt}px~}.~u.cWR.t}s.~.qt.
wtp
xt.p}s.tt|x}v{.spzt;.xw.
|~t.p<{xzt .qp}rt=.bqytr.w~.
wp
t.ttxt}rt.xw.|pxyp}p.wp
t.
t~ts.qtx}v.pq{t.~.xst}xu.
|pxyp}p.u~|.{prtq~.rxvptt.q.
xvw.p{~}t.wt}.wt.{p}.|ptxp{.x}.
p.rxvptt.p.
xxq{t.6cpwzx}.t.p{=;.
@HFCJ.fpt.t.p{=;.A?@?7=.cw;.~.
|px}px}.p.s~q{t<q{x}s.stxv};.|p}.
sxt.~qrt.wt.ptpp}rt.~u.{p}.
|ptxp{.q.r{~x}v.q~w.t}s.~u.wt.
|pxyp}p.rxvptt.p}s.{prx}v.x.x}.x}.
p}.~pt.{pxr.qt=.
fwx{t.}~}t.~u.wtt.|tw~s.~.
trt.q{x}sx}v.|p.qt.r~|{tt{.
tuutrx
t;.x.x.x|~p}.~.tsrt.qxp.
p.|rw.p.~xq{t.~.~srt.
r~}xt}.t{.qttt}.qytr.
}st.wt.p|t.ttx|t}p{.
r~}sxx~}=.
B=F=E _x~.\pxyp}p.Ttxt}rt.
\pxyp}p.t.wx~xt.x}.t.
qytr.|p.x}u{t}rt.~r~|t;.
t{pts.~.q~w.wtptxr.t~}x
x.
p}s.rwxpxr.PT=.\pxyp}p<}p'¶
t.
qytr.|p.p{~.ttxt}rt.p.
|pxyp}p.sv.~sr.p.~.p
tx
t.
wp.wt.~{s.}~.p}.~.t.wt.
sv.~sr=.cw;.qytr .x~.
ttxt}rt.xw.|pxyp}p.|p.puutr.
wt.r~}sr.p}s.t{.~u.sxt=.
\~.~u.wt.sxt.t
xtts.x}.wx.
s~r|t}.txts.wp.qytr.wp
t.p.
wx~.~u.|pxyp}p.t.6tt.pq{t.x}.
Pt}sx.wp.strxqt.trxuxr.
txt|t}.u~.tprw.s7=.W~t
t;.
x}.sxt.q{xwts.x}.wt.rxt}xuxr.
{xtpt;.wt.u{{.x}r{x~}.rxtxp.xw.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
tvps.~.trxuxr.p|~}.~u.ttxt}rt.
xw.|pxyp}p.|p.}~.qt.~
xsts=.
U~.w~t.sxt.wp.s~.~
xst.
x}r{x~}.rxtxp;.prrtpq{t.ttxt}rt.
xw.|pxyp}p.rp}.p}vt.u~|.~}rt.x}.
p.{xutx|t.~.t.|{x{t.x|t.p.sp=.
cwt.
px}v.wx~xt.~u.t.|xvw.
puutr.t
twx}v.u~|.r~t.~}.ps
tt.
t
t}.|tpt;.put.|tpt;.~.
tuuxrpr.|tpt=.Pssxx~}p{{;.
px}v.
p|~}.~u.ttxt}rt.rp}.x|pr.
r~v}xx
t.tuutr.|tpt.ptts.
sx}v.prt.ps|x}xpx~}.sxt=.U~.
x}p}rt;.brwtx}t.p}s.S}}.6A?@A7.
r~}t}s.r~v}xx
t.stuxrx.x}.wtp
.
|pxyp}p.t.r~}x}t.u~.
p~x|pt{.AG.sp.put.rtpx~}.~u.
|~zx}v=.bsxt.txx}v.{t.wp}.p.
|~}w.~u.pqx}t}rt.x~.~.wt.s.
|p.x{{.tt.txsp{.tuutr.~u.wtp
.
t.p.qpt{x}t.p}s.put.{prtq~.
|pxyp}p.ps|x}xpx~};.w.w~x}v.
}~.xv}xuxrp}.tuutr.~}.r~v}xx
t.
|tpt=.W~t
t;.wtt.p|t.
|tpt|t}.x}.~rrpx~}p{.~.}p'¶
t.
|pxyp}p.t.|p.st|~}pt.p.
xv}xuxrp}.tuutr.put.prt.|pxyp}p.
ps|x}xpx~}=.cwttu~t;.wt.p|~}.
~u.ttxt}rt.p}s.wt.spx~}.~u.
pqx}t}rt.~u.|pxyp}p.t.pt.
x|~p}.~.ztt.x}.|x}s.wt}.
p}p{x}v.t{.u~.r~v}xx
t.p}s.~wt.
ps
tt.t
t}.|tpt=.[p{;.p.s.
~{px~}.xw.t
x~.ttxt}rt.
xw.|pxyp}p.|p.}stt~.wt.
x}rxst}rt.p}s.t
tx.~u.ps
tt.
t
t}=.Qtrpt.|~.sxt.ts.
qytr.xw.x~.|pxyp}p.
ttxt}rt;.t.pt.{x|xts.x}.~.pqx{x.
~.vt}tp{xt.wt.t{;.ttrxp{{.u~.
put.|tpt;.~.|pxyp}p.}p'¶
t.
~{px~}=.
Ux
t.~u.@@.sxt.t
xtts.x}.wx.
s~r|t}.x}r{sts.q~w.|pxyp}p<.
}p'¶
t.p}s.|pxyp}p<ttxt}rts.
qytr.6R~t<Q{~~|.t.p{=;[email protected]{{x.
t.p{=;.A??HJ.fpt.t.p{=;.A?@?J.\tx.t.
p{=;.@HG?J.cpwzx}.t.p{=;.@HFC7=.bx}rt.wt.
}|qt.~u.|pxyp}p<}p'¶
t.qytr.x}.
wtt.sxt.p.{~;.x.p.}~.
~xq{t.~.r~}sr.p.tppt.p}p{x.
r~|pts.~.ttxt}rts.t=.
W~t
t;.t|pxrp{{.t
p{px}v.wt.
tuutr.~u.|pxyp}p.ttxt}rt.~}.s.
~r~|t.x.x|~p};.x}rt.|p}.
pxt}.w~.|xvw.t.p.|pxyp}p.
~sr.u~.p.wtptxr.t.x{{.qt.
|pxyp}p<}p'¶
t=.
attprw.w~.wp.|pxyp}p<.
ttxt}rts.qytr.wp
t.p.wxvwt.
pqx{x.~.~{tpt.~}vt.s~t.~u.~p{.
s~}pqx}~{.wp}.|pxyp}p<}p'¶
t.
qytr.6Wp}t.t.p{=;.A??D7=._~xq{;.
wx.x}rtpts.~{tp}rt.x.p{~.wt.rpt.
wt}.qytr.|~zt.~.
p~xt.
|pxyp}p=.cw;.sxt.r~{s.qt.
r~}srts.wp.x}
txvpt.wt.~{t.~u.
|pxyp}p.ttxt}rt.x}.stt|x}x}v.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
~{tpqx{x.~u.p}s.t~}t.~.p.
pxt.
~u.cWR.r~}rt}px~}.x}.|pxyp}p=.
64836!
5/!Tghgtgpegu!
@HHH=.Octklwcpc!cpf!Ogfkekpg<!Cuuguukpi!vjg!
Uekgpeg!Dcug/!fpwx}v~};.SRI.]px~}p{.
Prpst|._t=.
Pqp|.SX;.Wx{~}.YU;.[txt.aY;.bwpst.bQ;.
T{qtxz.cP;.Pttzp.Uc;.Qt}~x.][;.
Qts.Q\;.Z~t{.Q;.Pqtv.YP;.Sttz.bV;.
\xrwt{{.cU;.\{{xvp}.Z;.Qprrwtx._;.
\rR}t.Y\;.p}s.brwp|qt{p}.\=.A??B=.
bw~<t|.tuutr.~u.rp}}pqx}~xs.x}.
pxt}.xw.WXe¤@.x}utrx~}I.p.
p}s~|xts;.{prtq~<r~}~{{ts.r{x}xrp{.
xp{=.Cppcnu!qh!Kpvgtpcn!Ogfkekpg!@BH.
6C7I.ADG¤AEE=.
Pqp|.SX;.Yp.RP;.bwpst.bQ;.ex~~.W;.
atsp.W;._t.b;.Zt{{.\T;.a~q~wp|.
\R;.p}s._ttt}.Z[=.A??F=.Rp}}pqx.x}.
px}u{.WXe<p~rxpts.t}~.
}t~pwI.p.p}s~|xts.{prtq~<.
r~}~{{ts.xp{=.Pgwtqnqi{!EG.6F7I.D@D¤.
DA@=.
Pt}sx}~.V;.Rwxp}tt.V;.cpv{xp{pt{p<.
brpupx.^=.A?@@=.Rp}}pqx}~xsI.
~rrt}rt.p}s.|tsxrx}p{.rwt|x=.
Ewtt!Ogf!Ejgo/!@G6F7I@?GD¤HH=.
Qpt.a[;.fx{t.Y[;.Qp{t.a[;.p}s.\px}.
Qa=.@HHD=._wp|pr~{~vxrp{.trxuxrx.~u.
% <tpws~rp}}pqx}~{.sxrx|x}px~}.
x}.p=.Ru{ejqrjctoceqnqi{!@@G6C7I.
C@H¤CAC=.
Rwpx.[S;.p}s._xtx.Y=.@HGH=.b~|t.wxrp{.
rwpprtxxr.~u.]XSP.|pxyp}p.
rxvptt[email protected]@7I.
E@¤EF=.
Rwp}v.PT;.bwx{x}v.SY;.bx{{|p}.aR;.V~s{qtv.
]W;.btx.RP;.Qp~uz.X;.bx|~}.a\;.
p}s.a~t}qtv.bP=.@HFH=.St{p<H<.
tpws~rp}}pqx}~{.p.p}.p}xt|txr.x}.
rp}rt.pxt}.trtx
x}v.wxvw<s~t.
|tw~tpt=.Cppcnu!qh!Kpvgtpcn!
[email protected]@H¤GAC=.
R~t<Q{~~|.Y;.f~{u~}.c;.Vp|.P;.Yx}.b;.
\pr~t.cS;.Qt}{t.W;.p}s.V~p.Q=.
A?@A=.b|~zts.rp}}pqx.u~.pxrx.x}.
|{x{t.r{t~xI.p.p}s~|xts;.
{prtq~<r~}~{{ts.xp{=.Ecpcfkcp!
[email protected]@?7I.
B=F=G ]|qt.~u.Ut|p{t.bqytr.
@@CB¤@@D?=.
Rp}t.]P;.brwt.a\;.Up<_~{x._;.p}s.
P.r~||~}.••~q{t|.x}.r{x}xrp{.
V~}p{t.a=.A?@A=.Tuutr.~u.Rp}}pqx.~}.
ttprw.x.wt.{x|xts.}|qt.~u.
]t~r~v}xx
t.U}rx~}x}vI.atrt}.
ut|p{t.w~.pxrxpt.x}.wt.sxt=.
Ps
p}rt;.]t~st
t{~|t}p{.
cwx.~q{t|.x.tt}.x}.wt.@@.
X}u{t}rt;.p}s.bt.Sxuutt}rt=.
sxt.t
xtts.x}.wx.s~r|t};.x}.
Pgwtqru{ejqnqi{!Tgxkgy/!
wxrw.~}t.s.sxs.}~.x}r{st.p}.
Rpu~s.fY;.p}s.\tx.YR=.@HFH=.Tuutr.~u.
ut|p{t.qytr.6T{{x.t.p{=;.A??H7;.p}s.
tpws~rp}}pqx}~{.~}.ptxp{.p}s.
wtt.sxt.wps.p.{~.trt}pvt.~u.
x}p~r{p.wtt}x~}=.Kpvgtpcvkqpcn!
Lqwtpcn!qh!Enkpkecn!Rjctoceqnqi{!cpf!
ut|p{t.qytr.6Pqp|.t.p{=;.A??FJ.
dkqrjctoce{[email protected].@H@¤@HE=.
Wp}t.t.p{=;.A??DJ.Wp}t.t.p{=;.A??F7=.
T{{x.aY;.c~t~uu.f;.epxsp.U;.ep}.St}.
W~t
t;.tprw.~u.wtt.u~.sxt.
Qp}st.V;.V~}p{t.Y;.V~p.Q;.Qt}{t.
x}
txvpts.p}.WXe<~xx
t.pxt}.
W;.p}s.Pzx}~}.YW=.A??H=.b|~zts.
~{px~};.wtt.wtt.|p.wp
t.qtt}.
|tsxrx}p{.rp}}pqx.u~.}t~pwxr.px}.
p.{pvt.|p{t.~{px~}.~~{.u~|.
x}.WXeI.p.p}s~|xts;.r~~
t.r{x}xrp{.
wxrw.~.trx.r~|pts.~.ut|p{t=.
xp{=.Pgwtqru{ejqrjctoceqnqi{!BC.6B7I.
bx}rt.wtt.x.~|t.t
xst}rt.wp.wt.
EFA¤EG?=.
U{~|.\R;.Psp|.PY;.p}s.Y~}t.ac=.@HFD=.
st}x.~u.RQ@.trt~.x}.wt.qpx}.
\pxyp}p.|~zx}v.p}s.tsrts.tt.
|p.
p.qttt}.|p{t.p}s.ut|p{t.
x}.w|p}.ttI.sv.prx~}.~.
6Rp}t.t.p{=;.A?@A7;.wtt.|p.qt.
t•x•wt}~|t}~}N.Kpxguvkicvkxg!
sxuutx}v.wtptxr.~.qytrx
t.
Qrvjcnoqnqi{!@[email protected]¤DD=.
t~}x
x.~.|pxyp}p=.bsxt.x}v. U~{x}.af;.Qps.Ye;.p}s.Uxrwp|}.\f=.
p.s.~{px~}.wp.x.tp{.p.
@HGE=.Qtwp
x~p{.p}p{x.~u.|pxyp}p.
|p{t.p}s.ut|p{t.|p.w~.wtwt.p}s.
tuutr.~}.u~~s.x}pzt.x}.w|p}=.
w~.wt.tuutr.~u.|pxyp}p.sxuut.
Rjctoceqnqi{!Dkqejgokuvt{!cpf!
qttt}.|p{t.p}s.ut|p{t.qytr=.
Dgjcxkqt!ADI.DFF¤DGA=.
B=F=F X}r{x~}.p}s.Tr{x~}.Rxtxp.
U~.put.tp~};.p{{.r{x}xrp{.sxt.
wp
t.x}r{x~}.p}s.tr{x~}.rxtxp.
wp.txr.wt.pxrxpx~}.~u.
x}sx
xsp{.xw.rtpx}.|tsxrp{.
r~}sxx~}=.U~.sxt.wp.t.
|pxyp}p;.wtt.rxtxp.|p.qt.qpts.
~}.xz.p~rxpts.xw.t~t.~.
|~zts.|ptxp{.p}s.wt.tuutr.~u.cWR=.
cw;.|~.sxt.x}
txvpx}v.
|pxyp}p.txt.wp.qytr.p{xu.
u~.wt.s.qpts.~}.txrx
t.
|~|.rxtxp.rw.wp.x}sx
xsp{.
s~.}~.wp
t.~wt.|~|.wp.|p.qt.
z}~}.~.x}tpr.~~{.xw.
rp}}pqx}~xs=.
bx|x{p{;.r{x}xrp{.sxt.xw.
|pxyp}p.xrp{{.tr{st.x}sx
xsp{.
xw.rpsxpr.~.{|~}p.~q{t|;.p.
t{{.p.rwxpxr.sx~st=.cwtt.
tr{x~}.rxtxp.pt.qpts.~}.wt.t{{<.
z}~}.tuutr.~u.|pxyp}p.|~zt.~.
~srt.x}rtpt.x}.wtp.pt.p}s.
q{~~s.tt;.{}v.xxpx~};.p}s.wt.
tprtqpx~}.~u.rwxpxr.sxqp}rt.
x}.
{}tpq{t.x}sx
xsp{=.P{w~vw.
wtt.rxtxp.pt.|tsxrp{{.tp~}pq{t.
u~.ttprw.~~r~{;.x.x.{xzt{.wp.
ut.|pxyp}p.~sr.x{{.qt.ts.
x}.pxt}.w~.wp
t.rpsxpr;.
{|~}p.~.rwxpxr.r~}sxx~}=.
cw;.x}sx
xsp{.xw.wtt.r~}sxx~}.
w~{s.qt.t
p{pts;.wt}t
t.~xq{t=.
Pssxx~}p{{;.p{{.sxt.t
xtts.x}.
wx.s~r|t}.p{{~ts.wt.qytr.~.
r~}x}t.pzx}v.wtx.rt}.tvx|t}.~u.
|tsxrpx~}=.cw.p{{.t{.t
p{pts.
|pxyp}p.p.p}.psy}r.tp|t}.u~.
tprw.wtptxr.x}sxrpx~}=.
RQ!11111
Hto!1114;
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
64837!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
U~{x}.af;.Uxrw|p}.\f;.p}s.Q}t.\U=.
@HGG=.Tuutr.~u.|~zts.|pxyp}p.~}.
u~~s.x}pzt.p}s.q~s.txvw.~u.w|p}.
{x
x}v.x}.p.txst}xp{.{pq~p~=.
Crrgvkvg!@@I.@¤@C=.
Vtt}qtv.Wb;.ft}t.bP;._vw.YT;.P}s.
a^;.P}st~}.SY;.p}s.S~|x}~.TU=.@HHC=.
bw~<t|.tuutr.~u.|~zx}v.|pxyp}p.
~}.qp{p}rt.x}.pxt}.xw.|{x{t.
r{t~x.p}s.}~|p{.
~{}tt=.Enkpkecn!
Rjctoceqnqi{!cpf!Vjgtcrgwvkeu!DD.6B7I.
BAC¤BAG=.
Vtt}p{s.\Z.p}s.bxt.\[=.A???=.
P}x}~rxrtx
t;.qytrx
t;.p}s.
qtwp
x~p{.tuutr.~u.|~zts.|pxyp}p.
x}.w|p}=.Ftwi!cpf!Cneqjqn!
Fgrgpfgpeg!DHI.AE@¤AFD=.
Wp}t.\;.V}st~}.Tf;.apqzx}.Y;.Wp.R[;.
e~qv.bZ;.R~|t.bS;.p}s.U~{x}.af=.
A??F=.S~}pqx}~{.p}s.|pxyp}p.x}.WXe<.
~xx
t.|pxyp}p.|~zt=.Rp{~xr.
x}pzt;.|~~s;.p}s.{tt=.Lqwtpcn!qh!
Ceswktgf!Koowpg!Fghkekgpe{!U{pftqogu!
)2;;;*!CD.6D7I.DCD¤DDC=.
Wp}t.\;.apqzx}.Y;.V}st~}.T;.p}s.U~{x}.
af=.A??D=.S~}pqx}~{.p}s.|pxyp}p.x}.
WXe697.|pxyp}p.|~ztI.prt.tuutr.
~}.rp{~xr.x}pzt.p}s.|~~s=.
Ru{ejqrjctoceqnqi{!@[email protected]@7I.@F?¤@FG=.
Wx{{.bh;.brwx}.a;.V~~sx}.Sf;.p}s._~t{{.
QY=.@HFC=.\pxwp}p.p}s.px}=.Lqwtpcn!qh!
Rjctoceqnqi{!cpf!Gzrgtkogpvcn!
[email protected]@D¤C@G=.
Yp|t{.W=.A?@?=.P|txrp}.v{pr~|p.~rxt.
~xx~}.pt|t}I.|pxyp}p.p}s.wt.
tp|t}.~u.v{pr~|p=.Lqwtpcn!qh!
[email protected]¤FE=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
\tx.YR;.Rpu~s.fY;.P{tp}st._R;.
P}st.P[;.p}s.Vt{qp.bb=.@HG?=.Tuutr.
~u.|pxwp}p.~}.x}p~r{p.p}s.q{~~s.
tt.x}.v{pr~|p=.Qrjvjcnoqnqi{!GF.
6B7I.AAA¤AAG=.
\x{tx}.b[;.\prRp}}t{{.Z[;.Zp.Vf;.p}s.
R{pz.b=.@HFC=.\pxyp}p.~srts.
rwp}vt.x}.rp}t~.t}xx
x.p}s.
puutrI.t.p}s.}~}<t=.
Rjctoceqnqi{!Dkqejgokuvt{!cpf!
Dgjcxkqt!AIBEF¤BFC=.
\x{tx}.b[;.\prRp}}t{{.Z;.Zp.V;.p}s.R{pz.
b=.@HFD=.\pxyp}p<~srts.rwp}vt.x}.
px}.~{tp}rtI.Ttxt}rt.p}s.}~}<.
ttxt}rts.qytr=.Kpv/!
Rjctoceqru{ejkcv!@?I.@FF¤@GA=.
]pup{x.c;.brw{txst.[Q;.S~p}.X;.[p}z.T_;.
Qt}yp|x}~
.Ub;.p}s.Z~}xz~uu.U\=.A?@B=.
Rp}}pqx.x}srt.p.r{x}xrp{.t~}t.x}.
pxt}.xw.R~w} .sxtptI.P.
~trx
t.{prtq~<r~}~{{ts.s=.
Enkpkecn!Icuvtqgpvgtqnqi{!cpf!
Jgrcvqnqi{!@@I.@AFE¤@AG?=.
a~.T;.\pwt.\[;.Q}t.P;.et{x}.a;.Qprw.
_Y;.bp}rwt<ap|~.Y;.p}s.Zx{x}.ZP=.
A??A=.Rw~}xr.Rp}}pqx.dt.x}.wt.
R~|px~}pt.X}
txvpx~}p{.]t.Sv.
_~vp|I.P}.Tp|x}px~}.~u.Qt}tux.
p}s.Ps
tt.Tuutr.~u.[tvp{.R{x}xrp{.
Rp}}pqx=.Lqwtpcn!qh!Ecppcdku!
[email protected]¤DF=.
b~stp{|.PWe;.brwt.P;.p}s.st.fx.W=.
A??@=.P}xt|txr.tuuxrpr.~u.|~zts.
|pxyp}p.qytrx
t.p}s.qtwp
x~p{.
tuutr.~}.}ptp.x}srts.q..~u.
x•trpr=.Rjctoceqnqi{!Dkqejgokuvt{!cpf!
Dgjcxkqt!EHI.BCB¤BD?=.
RQ!11111
Hto!11151
Hov!5812
Uhov!5813
cpwzx}.S_;.V{xtstt.U;.a~t.Y;.Rwp}v._;.Wx.
ZZ;.h.Y[;.p}s.f.cR=.@HH@=.cp;.R^.
p}s.st{p.HcWR.st{x
t.u~|.wt.@.
p}s.A}s.wp{
t.~u.p.|pxyp}p.rxvptt=.
Rjctoceqnqi{!Dkqejgokuvt{!cpf!
Dgjcxkqt!C?.6B7I.EDF¤EE@=.
cpwzx}.S_;.bwpx~.QY;.[tt.hT;.Wpt.RT=.
@HFD=.Tuutr.~u.|~zts.|pxyp}p.x}.
ttx|t}p{{.x}srts.pw|p=.
Cogtkecp!Tgxkgy!qh!Tgurktcvqt{!Fkugcug!
@@AI.BFF¤BGE=.
fp{{prt.\;.brw{tx.V;.Pzx}~}.YW;.
f~{u~}.c;.[ppt~.S;.Qt}{t.W;.
V~p.Q;.p}s.Pqp|~}.X=.A??F=.S~t<.
stt}st}.tuutr.~u.|~zts.rp}}pqx.~}.
rppxrx}<x}srts.px}.p}s.wtp{vtxp.
x}.wtp{w.
~{}tt=.Cpguvjgukqnqi{!
@?F.6D7I.FGD¤FHE=.
fpt.\P;.fp}v.c;.bwpx~.b;.a~qx}~}.P;.
Srt.c;.W}w.c;.Vp|p.P;.Qt}}t.
VY;.p}s.R~{{t.Y_=.A?@?=.b|~zts.rp}}pqx.
u~.rw~}xr.}t~pwxr.px}I.p.
p}s~|xts.r~}~{{ts.xp{=.Ecpcfkcp!
[email protected]@C7I.
TEHC¤TF?@=.
fx{t.Q;.\pr~t.c;.c~sxz~
.P;.\x{{|p}.
Y;.Qt}{t.W;.V~p.Q;.p}s.Uxw|p}.b=.
A??G=.P.p}s~|xts;.{prtq~<r~}~{{ts;.
r~~
t.xp{.~u.rp}}pqx.rxvptt.x}.
}t~pwxr.px}=.L/!Rckp!H.6E7I.D?E¤DA@=.
fx{t.Q;.\pr~t.c;.Strw.a;.V~p.Q;.
bpzpx.b;.p}s.S~}pvwt.W=.A?@B=.[~<s~t.
p~xts.rp}}pqx.xv}xuxrp}{.
x|~
t.}t~pwxr.px}=.L/!Rckp!
@C6A7I@BE¤CG=.
G<^HT^HO^23CWR3/UIO
23CWR3
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11152
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
64838!
GR23CW27/119>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11153
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/11;>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64839!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11154
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
6483;!
GR23CW27/121>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11155
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/122>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64841!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11156
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
64842!
GR23CW27/123>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11157
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/124>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64843!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11158
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
64844!
GR23CW27/125>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11159
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/126>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64845!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!1115;
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
64846!
GR23CW27/127>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11161
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/128>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64847!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11162
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
64848!
GR23CW27/129>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
XgtFcvg!Ugr>22@3125!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11163
Hov!5812
Uhov!5836
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/12;>0IRJ@
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64849!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
W/U/!Fgrctvogpv!qh!LwuvkegFtwi!
Gphqtegogpv!Cfokpkuvtcvkqp!
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
STP.q|xts.wt.txx~}.p}s.
}trtp.spp.~.wt.Stp|t}.~u.
Wtp{w.p}s.W|p}.bt
xrt.6WWb7.~}.
Uejgfwng!qh!Eqpvtqnngf!Uwduvcpegu<!
Ockpvckpkpi!Octklwcpc!kp!Uejgfwng!K!qh! Y}t.@@;.A?@B;.p}s.ttts.wp.WWb.
~
xst.p.rxt}xuxr.p}s.|tsxrp{.
vjg!Eqpvtqnngf!Uwduvcpegu!Cev!
t
p{px~}.p}s.rwts{x}v.
Dcemitqwpf-!Fcvc-!cpf!Cpcn{uku<!Gkijv! tr~||t}spx~}.u~.|pxyp}p=.X}.
Hcevqtu!Fgvgtokpcvkxg!qh!Eqpvtqn!cpf!
s~r|t}.spts.Y}t.B.p}s.Y}t.AD;.
Hkpfkpiu!Rwtuwcpv!vq!32!W/U/E/!923)d*!
A?@D;.wt.prx}v.Pxp}.btrtp.u~.
Wtp{w.~u.wt.WWb tr~||t}sts.~.
Rtgrctgf!d{<!Qhhkeg!qh!Fkxgtukqp!
Eqpvtqn-!Ftwi!cpf!Ejgokecn!Gxcnwcvkqp! wt.STP.wp.|pxyp}p.r~}x}t.~.qt.
r~}~{{ts.x}.brwts{t.X.~u.wt.RbP;.p}s.
Ugevkqp-!Ycujkpivqp-!FE!31648!
~
xsts.~.wt.STP.x.rxt}xuxr.p}s.
Lwn{!3127!
|tsxrp{.t
p{px~}.x{ts.Qpx.u~.wt.
atr~||t}spx~}.u~.\px}px}x}v.
Dcemitqwpf!
\pxyp}p.x}.brwts{t.X.~u.wt.
^}.]~
t|qt.B?;.A?@@;.V~
t}~.
R~}~{{ts.bqp}rt.Pr= .cwt.WWb .
[x}r~{}.S=.Rwputt.~u.aw~st.X{p}s.p}s.
tr~||t}spx~}.pt.qx}sx}v.~}.wt.
Rwxx}t.^=.Vtv~xt.~u.fpwx}v~}.
STP.p.~.rxt}xuxr.p}s.|tsxrp{.
q|xts.p.txx~}.~.wt.Sv.
|pt[email protected]=b=R=.G@@6q7=.
T}u~rt|t}.Ps|x}xpx~}.6STP7.~.
Qtu~t.x}xxpx}v.~rttsx}v.~.
x}xxpt.~rttsx}v.u~.p.ttp{.~u.wt.
trwts{t.p.qp}rt;.wt.RbP.
{t.~.tv{px~}.wp.{prt.
txt.wt.Prx}v.Ps|x}xp~.~.
|pxyp}p x}.rwts{t.X.~u.wt.
stt|x}t.wtwt.wt.WWb.rwts{x}v.
R~}~{{ts.bqp}rt.Pr.6RbP7=.cwt.
tr~||t}spx~};.rxt}xuxr.p}s.|tsxrp{.
txx~}.tt.wp.|pxyp}p p}s.
t
p{px~};.p}s.p{{.~wt.t{t
p}.spp .
t{pts.xt| .qt.trwts{ts.x}.
r~}xt.qp}xp{.t
xst}rt.wp.wt.
rwts{t.XX.~u.wt.RbP=.cwt.txx~}t.
sv.w~{s.qt.trwts{ts.p.
r{px|.wpI.
~~[email protected]=b=R=.G@@6q7=.cwt.Prx}v.
@=.Rp}}pqx.wp.prrtts.|tsxrp{.t.
Ps|x}xp~.|.stt|x}t.wtwt.
x}.wt.d}xts.bptJ.
wtt.x.qp}xp{.t
xst}rt.~.
A=.Rp}}pqx.x.put.u~.t.}st.
r~}r{st.wp.wt.sv.|tt.wt.rxtxp.
|tsxrp{.t
xx~}J.
B=.Rp}}pqx.u~.|tsxrp{.~t.wp. u~.{prt|t}.x}.p}~wt.rwts{t.qpts.
~}.wt.rxtxp.t.u~w.x}[email protected]=b=R=.
p.t{px
t{.{~.~t}xp{.u~.pqt;.
G@A6q7=.cwt.RbP.txt.wp.q~w.wt.
ttrxp{{.x}.r~|px~}.xw.~wt.
STP.p}s.wt.WWb.r~}xst.wt.txvw.
rwts{t.XX.sv=.
upr~.trxuxts.q.R~}vt.x}.A@.
cwt.STP.prrtts.wx.txx~}.u~.
d=b=R=.G@@6r7=.cwx.s~r|t}.{p.~.
ux{x}v.~}.Yp}p.B?;.A?@A=.
w~t.r~}xstpx~}.p}s.x.~vp}xts.
cwt.P~}t.Vt}tp{.|p.q.{t.
prr~sx}v.~.wt.txvw.upr~=.P.STP.
p}ut.p.sv.~.~wt.qp}rt.
t.u~w.x}.stpx{.qt{~;.wt.t
xst}rt.
qttt}.rwts{t.~u.wt.RbP.xu.wt.
w~I.
ux}s.wp.rw.sv.~.~wt.qp}rt.
@=.Cevwcn!qt!tgncvkxg!rqvgpvkcn!hqt!
wp.p.~t}xp{.u~.pqt;.p}s.|pzt.wt.
ux}sx}v.trxqts.q[email protected]=b=R=.G@A6q7. cdwug/!\pxyp}p.wp.p.wxvw.~t}xp{.
u~.pqt=._tr{x}xrp{.p}s.r{x}xrp{.spp.
u~.wt.rwts{t.x}.wxrw.rw.sv.x.
w~.wp.x.wp.tx}u~rx}v.tuutr.
~.qt.{[email protected]=b=R=.G@@6p76@7=.cwt.
rwpprtxxr.~u.sv.~u.pqt=.
P~}t.Vt}tp{.wp.st{tvpts.wx.
]px~}p{.sppqpt.~}.prp{.pqt.
t~}xqx{x.~.wt.Prx}v.
w~.|pxyp}p.x.wt.|~.xst{.
Ps|x}xp~.~u.wt.STP=.AG.RUa.
pqts.sv;.x}r{sx}v.xv}xuxrp}.
?=@??6q7=.
}|qt.~u.qp}rt.pqt.tp|t}.
X}.prr~sp}rt.x[email protected]=b=R=.G@@6q7;.
ps|xx~}=.Spp.~}.|pxyp}p.txt.
put.vpwtx}v.wt.}trtp.spp;.wt.
w~.xsttps.p
px{pqx{x.p}s.
puuxrzx}v=.
cwt.R~}~{{ts.bqp}rt.Pr.6RbP7.stux}t.
A=.Uekgpvkhke!gxkfgpeg!qh!kvu!
|pxyp}p.pI.P{{.p.~u.wt.{p}.Ecppcdku!
ucvkxc!N/-!wtwt.v~x}v.~.}~J.wt.tts.wtt~uJ. rjctoceqnqikecn!ghhgev/!cwt.rxt}xuxr.
wt.tx}.tprts.u~|.p}.p.~u.rw.{p}J.p}s. }stp}sx}v.~u.|pxyp}p;.
t
t.r~|~}s;.|p}uprt;.p{;.stx
px
t;.
rp}}pqx}~xs.trt~;.p}s.wt.
|xt;.~.tppx~}.~u.rw.{p};.x.tts.~.
tx}=.brw.t|.s~t.}~.x}r{st.wt.|pt.p{z. t}s~rp}}pqx}~xs.t|.r~}x}t.~.
~u.rw.{p};.uxqt.~srts.u~|.rw.p{z;.~x{.
~.rpzt.|pst.u~|.wt.tts.~u.rw.{p};.p}.
~wt.r~|~}s;.|p}uprt;.p{;.stx
px
t;.
|xt;.~.tppx~}.~u.rw.|pt.p{z.
6trt.wt.tx}.tprts.wtt.u~|7;.uxqt;.~x{;.~.
rpzt;.~.wt.tx{xts.tts.~u.rw.{p}.wxrw.x.
x}rppq{t.~u.vt|x}px~}= [email protected]=b=R=.G?A6@E7=.]~t.
wp.|pxwp}p .x.wt.t{{x}v.ts.x}.wt.RbP=.
cwx.s~r|t}.t.wt.t{{x}v.wp.x.|~t.
r~||~}.x}.rt}.pvt;.|pxyp}p= .
_txx~}t.stux}ts.|pxyp}p.p.p{{.r{x
pts.
px}.~u.rp}}pqx=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
P.t.u~w.x}.p.|t|~p}s|.~u.
}stp}sx}v.t}tts.x}~.q.wt.WWb;.wt.U~~s.
p}s.Sv.Ps|x}xpx~}.6USP7;.p}s.wt.]px~}p{.
X}xt.~}.Sv.Pqt.6]XSP7;.wt.USP.pr.p.wt.
{tps.pvt}r.xwx}.wt.WWb.x}.rpx}v.~.wt.
btrtp .rwts{x}v.t~}xqx{xxt.}st.wt.
RbP;.xw.wt.r~}rt}rt.~u.wt.]XSP=.D?.Ua.
HD@G;.\p=.G;.@HGD=.cwt.btrtp.~u.wt.WWb.wp.
st{tvpts.~.wt.Pxp}.btrtp.u~.Wtp{w.~u.
wt.WWb.wt.pw~x.~.|pzt.s~|txr.sv.
rwts{x}v.tr~||t}spx~}=.
RQ!11111
Hto!11164
Hov!5812
Uhov!5813
6484;!
qt.sxts.p}s.t{rxspts=.\pxyp}p.
~srt.
px~.wp|pr~{~vxrp{.
tuutr;.x}r{sx}v.qytrx
t.6g/i/-!
tw~xp;.sxx}t;.sxx}wxqxx~}7;.
rpsx~
pr{p;.prt.p}s.rw~}xr.
txp~;.x||}t.t|;.p}s.
t}pp{.t~t.tuutr;.p.t{{.p.
qtwp
x~p{.p}s.r~v}xx
t.x|px|t}=.
B=.Ewttgpv!uekgpvkhke!mpqyngfig/!cwt•t.
x.}~.rt}{.prrtts.|tsxrp{.t.u~.
|pxyp}p.x}.wt.d}xts.bpt=.
\pxyp}p.~rt.pt.stx
ts.u~|.
}|t~.r{x
pts.px}.p}s.|p.
wp
t.sxuutt}.{t
t{.~u.% <cWR.p}s.
~wt.rp}}pqx}~xs=.d}st.wt.ux
t<.
t{t|t}.t.u~.rt}{.prrtts.
|tsxrp{.t.sxrts.x}.|~t.stpx{.
qt{~.p}s.wt{s.q.wt.R~.~u.
Ptp{.u~.wt.Sxxr.~u.R~{|qxp.x}.
Cnnkcpeg!hqt!Ecppcdku!Vjgtcrgwvkeu!
=.
[email protected]=Bs.@@B@;.@@BD.6S=R=.Rx•=.
@HHC7.6wttx}put.CEV 7;.wtt.x.}~.
r~|{tt.rxt}xuxr.p}p{x.~u.
|pxyp}p .rwt|xrp{.r~|~}t}J.wtt.
pt.}~.pstpt.put.sxtJ.wtt.pt.
}~.pstpt.p}s.t{{<r~}~{{ts.
tuuxrpr.sxtJ.wtt.x.}~.p.r~}t}.
~u.|tsxrp{.~•x}x~}.r~}rt•}x}v.|tsxrp{.
p{xrpx~}.~u.|pxyp}pJ.p}s.wt.
rxt}xuxr.t
xst}rt.tvpsx}v.
|pxyp}p .put.p}s.tuuxrpr.x.}~.
xst{.p
px{pq{t=.c~.spt;.rxt}xuxr.p}s.
|tsxrp{.ttprw.wp.}~.~vtts.~.
wt.~x}.wp.|pxyp}p.wp.p.rt}{.
prrtts.|tsxrp{.t;.t
t}.}st.
r~}sxx~}.wtt.x.t.x.t
tt{.
txrts=.
C=.Jkuvqt{!cpf!ewttgpv!rcvvgtp!qh!
cdwug/!\pxyp}p.r~}x}t.~.qt.wt.
|~.xst{.ts.x{{xrx.sv=.X}.A?@C;.
wtt.tt.AA=A.|x{{x~}.rt}.t=.
cwtt.tt.p{~.A=E.|x{{x~}.}t.t;.
|~.~u.w~|.tt.{t.wp}.@G.tp.
~u.pvt=.Sx}v.wt.p|t.tx~s;.
|pxyp}p.p.wt.|~.utt}{.
xst}xuxts.sv.twxqx.x}.utstp{;.pt;.
p}s.{~rp{.u~t}xr.{pq~p~xt=.
D=.Ueqrg-!fwtcvkqp-!cpf!ukipkhkecpeg!
qh!cdwug/!Pqt.~u.|pxyp}p.x.
xsttps.p}s.xv}xuxrp}=.X}.A?@C;.u~.
tp|{t;.p}.tx|pts.E=D.|x{{x~}.
t~{t.pvts.@A.~.~{st.ts.|pxyp}p.
~}.p.spx{.~.p{|~.spx{.qpx.~
t.p.
@A<|~}w.tx~s=.X}.pssxx~};.p.
xv}xuxrp}.~~x~}.~u.p{{.ps|xx~}.
u~.qp}rt.pqt.tp|t}.pt.u~.
|pxyp}p>wpwxw.p.wtx.x|p.sv.
~u.pqt=.X}.A?@B;.@E=G3.~u.p{{.rw.
ps|xx~}¥AG@;HH@.~
t.wt.r~t.~u.
wt.tp¥tt.u~.x|p.|pxyp}p>
wpwxw.pqt=.
E=.Tkum-!kh!cp{-!vq!rwdnke!jgcnvj/!
c~vtwt.xw.wt.wtp{w.xz.~{x}ts.
x}.t|.~u.wp|pr~{~vxrp{.tuutr.
pq~
t;.q{xr.wtp{w.xz.u~|.prt.
t.~u.|pxyp}p.x}r{st.x|pxts.
rw~|~~.tu~|p}rt;.x|pxts.
sx
x}v;.p}s.x|pxts.tu~|p}rt.~}.
t.~u.{tp}x}v.p}s.p~rxpx
t.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64851!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
~rtt=.Rw~}xr.t.~u.|pxyp}p.
~t.p.}|qt.~u.~wt.xz.~.wt.
q{xr.wtp{w.x}r{sx}v.wxrp{.p.t{{.
p.rw~{~vxrp{.stt}st}rt=.
F=.Ru{ejke!qt!rj{ukqnqikecn!
fgrgpfgpeg!nkcdknkv{/![~}v<t|;.wtp
.
t.~u.|pxyp}p.rp}.{tps.~.wxrp{.
stt}st}rt.p}s.xwspp{.u~{{~x}v.
sxr~}x}px~};.p.t{{.p.rwxr.~.
rw~{~vxrp{.stt}st}rt=.X}.pssxx~};.
p.xv}xuxrp}.~~x~}.~u.p{{.
ps|xx~}.u~.tp|t}.u~.qp}rt.
pqt.pt.u~.x|p.|pxyp}p.pqtJ.
x}.A?@B;.@E=G3.~u.p{{.ps|xx~}.tt.
u~.x|p.|pxyp}p>wpwxw.pqt;.
ttt}x}v.AG@;[email protected]}sx
xsp{=.
G=.Koogfkcvg!rtgewtuqt/!\pxyp}p.x.
}~.p}.x||tsxpt.tr~.~u.p}.
r~}~{{ts.qp}rt=.
P.trxuxts.x}[email protected]=b=R=.G@A6q76@7;.x}.
~st.u~.p.qp}rt.~.qt.{prts.x}.
rwts{t.X;.wt.Prx}v.Ps|x}xp~.
|.ux}s.wpI.
P=.cwt.sv.~.~wt.qp}rt.wp.p.
wxvw.~t}xp{.u~.pqt=.
Q=.cwt.sv.~.~wt.qp}rt.wp.}~.
rt}{.prrtts.|tsxrp{.t.x}.
tp|t}.x}.wt.d}xts.bpt=.
R=.cwtt.x.p.{prz.~u.prrtts.put.
u~.t.~u.wt.sv.~.~wt.qp}rt.
}st.|tsxrp{.t
xx~}=.
c~.qt.r{pxuxts.x}.p}~wt.rwts{t.
}st.wt.RbP.6g/i/-!XX;.XXX;.Xe;.~•.e7;.p.
qp}rt.|.wp
t.p.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bpt= [email protected]=b=R=.G@A6q76A7¤6D7=.
P.qp}rt.p{~.|p.qt.{prts.x}.
rwts{t.XX.xu.x.x.u~}s.~.wp
t.p.
rt}{.prrtts.|tsxrp{.t.xw.
t
tt.txrx~}= [email protected]=b=R=.G@A6q76A7=.
Xu.p.r~}~{{ts.qp}rt.wp.}~.rw.
rt}{.prrtts.|tsxrp{.t;.x.|.
qt.{prts.x}.rwts{t.X=.Ugg!]~xrt.~u.
St}xp{.~u._txx~};.EE.Ua.A??BG.6P=.@G;.
[email protected]R~}vt.tpq{xwts.~}{.~}t.
rwts{t¥rwts{t.X¥u~.sv.~u.
pqt.xw.}~.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt .p}s.{prz.~u.prrtts.put.u~.
t.= = =.}st.|tsxrp{.t
xx~}= 7=.
P.sv.wp.x.wt.qytr.~u.p}.
p~
ts.}t.sv.p{xrpx~}.6]SP7.
~.pqqt
xpts.}t.sv.p{xrpx~}.
6P]SP7.}st.Utstp{.U~~s;.Sv;.p}s.
R~|txr.Pr[email protected]=b=R=.BDD7;.x.
r~}xstts.~.wp
t.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt.u~.~t.~u.wt.RbP=.cwt.
WWb.pts.x}.x.t
xt;.w~t
t;.wp.
USP.wp.}~.p~
ts.p}.]SP.u~.
|pxyp}p.u~.p}.x}sxrpx~}=.
X}.wt.pqt}rt.~u.]SP.~.P]SP.
p~
p{;.STP.wp.tpq{xwts.p.ux
t<.
t{t|t}.t.u~.stt|x}x}v.wtwt.
wt.sv.wp.p.rt}{.prrtts.
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt=.d}st.wx.t;.p.sv.x{{.qt.
r~}xstts.~.wp
t.p.rt}{.prrtts.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
|tsxrp{.t.~}{.xu.wt.u~{{~x}v.ux
t.
t{t|t}.pt.pxuxtsI.
@=.cwt.sv .rwt|x.x.z}~}.p}s.
t~srxq{tJ.
A=.cwtt.pt.pstpt.put.sxtJ.
B=.cwtt.pt.pstpt.p}s.t{{<.
r~}~{{ts.sxt.~
x}v.tuuxrprJ.
C=.cwt.sv.x.prrtts.q.p{xuxts.
ttJ.p}s.
D=.cwt.rxt}xuxr.t
xst}rt.x.xst{.
p
px{pq{t=.
6DF.Ua.@?CHH;.@?D?E.6\p•rw.AE;.@HHA77=.
[email protected]=Bs.p.@@BD=.
P.sxrts.x}.Upr~.B;.qt{~;.WWb.
r~}r{sts;.p}s.STP.pvtt;.wp.wt.
rxt}xuxr.t
xst}rt.x.x}uuxrxt}.~.
st|~}pt.wp.|pxyp}p.wp.p.
rt}{.prrtts.|tsxrp{.t.}st.
wt.ux
t<t{t|t}.t=.cwt.t
xst}rt.p.
x}uuxrxt}.x}.wx.tvps.p{~.wt}.wt.
STP.r~}xstts.txx~}.~.trwts{t.
|pxyp}p.x}[email protected].@?CHH7; x}.
[email protected]??BG7;.p}s.x}.A?@@.6FE.Ua.
C?DDA7= [x{t.wp.rwp}vts.x}rt.A?@@.
xw.ttr.~.wt.{prz.~u.r{x}xrp{.
t
xst}rt.}trtp.~.tpq{xw.wp.
|pxyp}p.wp.p.rt}{.prrtts.
|tsxrp{.t=.]~.sxt.wp
t.
rxt}xuxrp{{.ptts.wt.tuuxrpr.p}s.
u{{.put.~ux{t.~u.|pxyp}p.u~.p}.
trxuxr.|tsxrp{.r~}sxx~}=.
cwt.{x|xts.txx}v.r{x}xrp{.t
xst}rt.
x.}~.pstpt.~.pp}.trwts{x}v.
~u.|pxyp}p.}st.wt.RbP=.c~.wt.
r~}p;.wt.spp.x}.wx.rwts{x}v.
t
xt.s~r|t}.w~.wp.|pxyp}p.
r~}x}t.~.|tt.wt.rxtxp.u~.
rwts{t.X.r~}~{.}st.wt.RbP.u~.wt.
u~{{~x}v.tp~}I.
@=.\pxyp}p.wp.p.wxvw.~t}xp{.u~.
pqt=.
A=.\pxyp}p.wp.}~.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bpt=.
B=.\pxyp}p.{prz.prrtts.put.u~.
t.}st.|tsxrp{.t
xx~}=.
wp.|pxyp}p.~srt.tx}u~rx}v.
tuutr.x}.w|p}=.brw.tx}u~rx}v.
tuutr.rp}.prr~}.u~.wt.ttpts.
pqt.~u.|pxyp}p=.
C/!Kpfkecvqtu!qh!Cdwug!Rqvgpvkcn!
cwt.WWb.wp.r~}r{sts.x}.x.
s~r|t};.Qpx.u~.wt.
atr~||t}spx~}.u~.\px}px}x}v.
\pxyp}p.x}.brwts{t.X.~u.wt.
R~}~{{ts.bqp}rt.Pr; .wp.
|pxyp}p.wp.p.wxvw.~t}xp{.u~.pqt=.
cwt.ux}sx}v.~u.pqt.~t}xp{ .x.
rxxrp{.u~.r~}~{.}st.wt.R~}~{{ts.
bqp}rt.Pr.6RbP7=.P{w~vw.wt.
t|.x.}~.stux}ts.x}.wt.RbP;.
vxsp}rt.x}.stt|x}x}v.pqt.~t}xp{.
x.~
xsts.x}.wt.{tvx{px
t.wx~.~u.
wt.Pr.6R~|twt}x
t.Sv.Pqt.
_t
t}x~}.p}s.R~}~{.Pr.~u.@HF?;.
W=a=.at=.]~=.H@¤@CCC;.H@.R~}v=;.bt=.
A.6@HF?7;.tx}ts.x}.@HF?.d=b=R=R=P=]=.
CDEE;.CE?B7=.Prr~sx}v{;.wt.u~{{~x}v.
xt|.pt.x}sxrp~.wp.p.sv.~.~wt.
qp}rt.wp.~t}xp{.u~.pqtI.
Vjgtg!ku!gxkfgpeg!vjcv!kpfkxkfwcnu!
ctg!vcmkpi!vjg!ftwi!qt!ftwiu!eqpvckpkpi!
uwej!c!uwduvcpeg!kp!coqwpvu!uwhhkekgpv!
vq!etgcvg!c!jc|ctf!vq!vjgkt!jgcnvj!qt!vq!
vjg!uchgv{!qh!qvjgt!kpfkxkfwcnu!qt!qh!vjg!
eqoowpkv{=!qt!
Vjgtg!ku!ukipkhkecpv!fkxgtukqp!qh!vjg!
ftwi!qt!ftwiu!eqpvckpkpi!uwej!c!
uwduvcpeg!htqo!ngikvkocvg!ftwi!
ejcppgnu=!qt!
Kpfkxkfwcnu!ctg!vcmkpi!vjg!ftwi!qt!
ftwiu!eqpvckpkpi!uwej!c!uwduvcpeg!qp!
vjgkt!qyp!kpkvkcvkxg!tcvjgt!vjcp!qp!vjg!
dcuku!qh!ogfkecn!cfxkeg!htqo!c!
rtcevkvkqpgt!nkegpugf!d{!ncy!vq!
cfokpkuvgt!uwej!ftwiu!kp!vjg!eqwtug!qh!
jku!rtqhguukqpcn!rtcevkeg=!qt!
Vjg!ftwi!qt!ftwiu!eqpvckpkpi!uwej!
c!uwduvcpeg!ctg!pgy!ftwiu!uq!tgncvgf!kp!
vjgkt!cevkqp!vq!c!ftwi!qt!ftwiu!cntgcf{!
nkuvgf!cu!jcxkpi!c!rqvgpvkcn!hqt!cdwug!vq!
ocmg!kv!nkmgn{!vjcv!vjg!ftwi!yknn!jcxg!vjg!
Hcevqt!2<!Vjg!Ftwiu!Cevwcn!qt!Tgncvkxg! ucog!rqvgpvkcnkv{!hqt!cdwug!cu!uwej!
ftwiu-!vjwu!ocmkpi!kv!tgcuqpcdng!vq!
Rqvgpvkcn!hqt!Cdwug!
cuuwog!vjcv!vjgtg!oc{!dg!ukipkhkecpv!
\pxyp}p.x.wt.|~.r~||~}{.
fkxgtukqpu!htqo!ngikvkocvg!ejcppgnu-!
pqts.x{{tvp{.sv.x}.wt.d}xts.bpt=. ukipkhkecpv!wug!eqpvtct{!vq!qt!ykvjqwv!
X.x.p{~.wt.|~.r~||~}{.ts.x{{xrx. ogfkecn!cfxkeg-!qt!vjcv!kv!jcu!c!
sv.q.wxvw.rw~~{.st}.x}.wt.
uwduvcpvkcn!ecrcdknkv{!qh!etgcvkpi!
d}xts.bpt=.Uwt;.|pxyp}p.x.wt. jc|ctfu!vq!vjg!jgcnvj!qh!vjg!wugt!qt!vq!
|~.utt}{.xst}xuxts.sv.q.pt;. vjg!uchgv{!qh!vjg!eqoowpkv{/!
{~rp{.p}s.utstp{.u~t}xr.{pq~p~xt=.
^u.r~t;.t
xst}rt.~u.prp{.pqt.~u.
\pxyp}p .|px}.rw~prx
t.
p.qp}rt.x.x}sxrpx
t.wp.p.sv.wp.
x}vtsxt};.% <tpws~rp}}pqx}~{.
p.~t}xp{.u~.pqt=.
6% <cWR7; x.p}.tuutrx
t.tx}u~rt.x}.
X}.x.tr~||t}spx~};.wt.WWb.
{pq~p~.p}x|p{;.x}r{sx}v.x|pt.
p}p{ts.p}s.t
p{pts.spp.~}.
p}s.~st}=.cwtt.p}x|p{.sxt.q~w. |pxyp}p.p.p{xts.~.tprw.~u.wt.
tsxr.p}s.~.wt.~qt
px~}.
pq~
t.u~.rxtxp=.cwt.p}p{x.
tt}ts.x}.wt.tr~||t}spx~}.6WWb;.
Ugg!Cnnkcpeg!hqt!Ecppcdku!Vjgtcrgwvkeu!
=.FGC-! [email protected].sxrts.qt{~I.
@D.U=Bs.@@[email protected]=R=.Rx•=.@HHC7=.
2/!Vjgtg!ku!gxkfgpeg!vjcv!kpfkxkfwcnu!
Ugg!P|txrp}.u~.bput.Prrt.
=.STP;.F?E.
ctg!vcmkpi!vjg!ftwi!qt!ftwiu!eqpvckpkpi!
U=Bs.CBG.6S=R=.Rx•[email protected]•wv.st}=.A?@B7=.
uwej!c!uwduvcpeg!kp!coqwpvu!uwhhkekgpv!
cwt.t|.% <cWR.p}s.cWR.pt.ts.
x}trwp}vtpq{.w~vw.~.wx.s~r|t}=.
vq!etgcvg!c!jc|ctf!vq!vjgkt!jgcnvj!qt!vq!
RQ!11111
Hto!11165
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
vjg!uchgv{!qh!qvjgt!kpfkxkfwcnu!qt!qh!vjg!
eqoowpkv{/!
cwt.WWb.pts.wp.~|t.x}sx
xsp{.
pt.pzx}v.|pxyp}p.x}.p|~}.
uuxrxt}.~.rtpt.p.wpps.~.wtx.
wtp{w.p}s.~.wt.put.~u.~wt.
x}sx
xsp{.p}s.wt.r~||}x=.Spp.
u~|.}px~}p{.sppqpt.~}.prp{.pqt.
~u.|pxyp}p.~.wt.xstp.wp.p.
{pvt.}|qt.~u.x}sx
xsp{.t.
|pxyp}p=.X}.x.tr~||t}spx~}.6WWb;.
A?@D7;.wt.WWb.tt}ts.spp.u~|.wt.
]px~}p{.b
t.~}.Sv.p}s.Wtp{w.
6]bSdW7.~u.wt.bqp}rt.Pqt.p}s.
\t}p{.Wtp{w.bt
xrt.Ps|x}xpx~}.
6bP\WbP7.p}s.wt.\~}x~x}v.wt.
Ut.6\cU7.
t.~u.wt.]px~}p{.
X}xt.~}.Sv.Pqt.6]XSP7;.p}s.wt.
STP.wp.x}rt.spts.wx.x}u~|px~}=.
cwt.|~.trt}.spp.u~|.bP\WbP .
]bSdW.x}.A?@C.t~ts.wp.|pxyp}p.
p.wt.|~.ts.x{{xrx.sv=.P|~}v.
P|txrp}.pvts.@A.tp.p}s.~{st;.p}.
tx|pts.AA=A.|x{{x~}.P|txrp}.ts.
|pxyp}p.xwx}.wt.p.|~}w.
prr~sx}v.~.wt.A?@C.]bSdW=.X}.A??C;.
p}.tx|pts.@C=E.|x{{x~}.x}sx
xsp{.
t~ts.x}v.|pxyp}p.xwx}.wt.
|~}w.x~.~.wt.s=.cwt.tx|pts.
pt.x}.A?@C.w.tu{tr.p}.x}rtpt.~u.
p~x|pt{.F=E.|x{{x~}.x}sx
xsp{.
~
t.p.@?<tp.tx~s=.Prr~sx}v.~.wt.
A?@B.]bSdW.t~;.p}.tx|pts.@H=G.
|x{{x~}.x}sx
xsp{.t~ts.x}v.
|pxyp}p=.cw;.~
t.p.tx~s.~u.~}t.
tp.6A?@B.]bSdW¤A?@C.]bSdW7;.wtt.
p.p}.tx|pts.x}rtpt.~u.A=C.|x{{x~}.
x}sx
xsp{.x}.wt.d}xts.bpt.x}v.
|pxyp}p=.
cwt.t{.u~|.wt.A?@D.\~}x~x}v.
wt.Ut.
t.~u.Gw;.@?w;.p}s.@Aw.
vpst.st}.x}sxrpt.wp.|pxyp}p.
p.wt.|~.xst{.ts.x{{xrx.sv.x}.
wtt.pvt.v~=.Rt}.|~}w{.t.
p.E=D3.~u.Gw.vpst;.@C=G3.~u.@?w.
vpst;.p}s.A@=B3.~u.@Aw.vpst=.cwt.
ctp|t}.Tx~st.Spp.bt.6cTSb7.x}.
A?@B.t~ts.wp.|pxyp}p.pqt.p.
wt.x|p.upr~.x}.@E=G.trt}.~u.
}~}<x
pt.qp}rt<pqt.tp|t}.
uprx{x.ps|xx~}=.X}.A?@@;.bP\WbP .
Sv.Pqt.fp}x}v.]t~z.6SPf]7.
t~ts.wp.|pxyp}p.p.|t}x~}ts.
x}.BE=C3.6CDD;EEG.~.~u.p~x|pt{.
@=AD.|x{{x~}7.~u.x{{xrx.sv<t{pts.
T|tvt}r.Stp|t}.6TS7.
xx=.
Spp.~}.wt.tt}.p}s.r~t.~u.
|pxyp}p.pqt.pt.tt}ts.}st.
Upr~.C.p}s.D.~u.wx.p}p{x=.
Sxrx~}.~u.wt.wtp{w.tuutr.~u.
|pxyp}p.x.tt}ts.}st.Upr~.A;.
p}s.wt.pt|t}.~u.xz.~.wt.q{xr.
wtp{w.~ts.q.prt.p}s.rw~}xr.
|pxyp}p.pqt.x.tt}ts.}st.
Upr~.E.~u.wx.p}p{x=.
3/!Vjgtg!ku!ukipkhkecpv!fkxgtukqp!qh!vjg!
ftwi!qt!ftwiu!eqpvckpkpi!uwej!c!
uwduvcpeg!htqo!ngikvkocvg!ftwi!
ejcppgnu/!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
X}.prr~sp}rt.xw.wt.RbP;.wt.~}{.
{pu{.~rt.~u.|pxyp}p.x}.wt.d}xts.
bpt.x.wp.~srts.p}s.sxxqts.
u~.ttprw.~t.}st.wt.
~
txvw.~u.]XSP.p}s.x}.r~}u~|x.
xw.d}xts.bpt.~q{xvpx~}.}st.wt.
bx}v{t.R~}
t}x~}.~}.]pr~xr.Sv= .
cwt.WWb.pts.wp.wtt.x.p.{prz.~u.
xv}xuxrp}.sx
tx~}.u~|.{tvxx|pt.
sv.~rt;.q.wp.wx.x.{xzt{.st.
~.wxvw.p
px{pqx{x.~u.|pxyp}p.u~|.
x{{xrx.~rt=.\pxyp}p.x.}~.p}.USP<.
p~
ts.sv.~sr=.]txwt.p.]t.
Sv.P{xrpx~}.6]SP7.}~.p.Qx~{~vxr.
[xrt}t.P{xrpx~}.6Q[P7.wp.qtt}.
p~
ts.u~.|pztx}v.x}.wt.d}xts.
bpt=.W~t
t;.wt.|pxyp}p.ts.u~.
}~}r{x}xrp{.p}s.r{x}xrp{.ttprw.
ttt}.p.
t.|p{{.p|~}.~u.wt.
~p{.p|~}.~u.|pxyp}p.p
px{pq{t.x}.
wt.d}xts.bpt.p}s.wttu~t.
x}u~|px~}.pq~.|pxyp}p.sx
tx~}.
u~|.{tvxx|pt.~rt.x.{x|xts.~.}~.
p
px{pq{t=.
cwt.STP.}~t.wp.wt.|pv}xst.~u.
wt.st|p}s.u~.x{{xrx.|pxyp}p.x.
t
xst}rts.q.x}u~|px~}.u~|.p.
}|qt.~u.sppqpt.tt}ts.}st.
Upr~.C=.Qxtu{;.|pxyp}p.x.wt.|~.
r~||~}{.ts.x{{tvp{.sv.x}.wt.
d}xts.bpt=.X.x.p{~.wt.|~.
r~||~}{.ts.x{{xrx.sv.q.P|txrp}.
wxvw.rw~~{t=.\pxyp}p.x.wt.|~.
utt}{.xst}xuxts.sv.x}.pt;.{~rp{;.
p}s.utstp{.u~t}xr.{pq~p~xt;.xw.
x}rtpx}v.p|~}.~u.q~w.s~|txrp{{.
v~}.p}s.~u.x{{xrx{.|vv{ts.
|pxyp}p=.
Vx
t}.wp.|pxyp}p.wp.{~}v.qtt}.
wt.|~.xst{.puuxrzts.p}s.pqts.
r~}~{{ts.qp}rt.x}.wt.d}xts.
bpt;.p}s.wp.p{{.ptr.~u.rw.x{{xrx.
prx
x.pt.t}xt{.~xst.~u.wt.
r{~ts.t|.~u.sxxqx~}.|p}spts.
q.wt.RbP;.x.|p.t{{.qt.wt.rpt.wp.
wtt.x.{x{t.w~vw.vx
t}.~.sx
tx}v.
|pxyp}p.u~|.wt.|p{{.{xt.
~srts.u~.{tvxx|pt.ttprw.
~t=.cw;.wt.{prz.~u.spp.
x}sxrpx}v.sx
tx~}.~u.|pxyp}p.u~|.
{tvxx|pt.rwp}}t{.~.wt.x{{xrx.|pzt.
x.}~.x}sxrpx
t.~u.p.{prz.~u.~t}xp{.u~.
pqt.~u.wt.sv=.
4/!Kpfkxkfwcnu!ctg!vcmkpi!vjg!ftwi!qt!
ftwiu!eqpvckpkpi!uwej!c!uwduvcpeg!qp!
vjgkt!qyp!kpkvkcvkxg!tcvjgt!vjcp!qp!vjg!
dcuku!qh!ogfkecn!cfxkeg!htqo!c!
rtcevkvkqpgt!nkegpugf!d{!ncy!vq!
cfokpkuvgt!uwej!ftwiu!kp!vjg!eqwtug!qh!
jku!rtqhguukqpcn!rtcevkeg/!
cwt.WWb.pts.wp.wt.USP.wp.}~.
t
p{pts.~.p~
ts.p}.]SP.~.Q[P.
u~.|pxyp}p.u~.p}.wtptxr.
x}sxrpx~}=.R~}xt}.xw.utstp{.{p;.
wttu~t;.p}.x}sx
xsp{.{tvxx|pt{.rp}.
Ugg!FE.Ua.D@C?B;.D@C?H¤D@C@?.6A?@@7.
6sxrx}v.rp}}pqx.r~}~{.txts.}st.wt.
bx}v{t.R~}
t}x~}7=.
RQ!11111
Hto!11166
Hov!5812
Uhov!5813
64852!
pzt.|pxyp}p.qpts.~}.|tsxrp{.ps
xrt.
u~|.p.prxx~}t.~}{.q.pxrxpx}v.
x}.ttprw.wp.x.qtx}v.r~}srts.
}st.p}.X}
txvpx~}p{.]t.Sv.
6X]S7.p{xrpx~}=.cwt.WWb.}~ts.wp.
wtt.pt.t
tp{.pt.p.t{{.p.wt.
Sxxr.~u.R~{|qxp.wxrw.wp
t.pts.
{p.p{{~x}v.u~.x}sx
xsp{.~.t.
|pxyp}p.u~.~ts.|tsxrp{ .t.
}st.rtpx}.rxr|p}rt;.q.spp.
pt.}~.p
px{pq{t.t.~.stt|x}t.wt.
}|qt.~u.x}sx
xsp{.x}v.|pxyp}p.
}st.wtt.pt.{p=.]~}twt{t;.
prr~sx}v.~.A?@C.]bSdW.spp;.AA=A.
|x{{x~}.P|txrp}.ps{.rt}{.t.
|pxyp}p.6bP\WbP;.A?@Dp7=.Qpts.~}.
wt.{pvt.}|qt.~u.x}sx
xsp{.w~.t.
|pxyp}p.p}s.wt.{prz.~u.p}.USP<.
p~
ts.sv.~sr;.wt.WWb.
r~}r{sts.wp.wt.|py~x.~u.
x}sx
xsp{.x}v.|pxyp}p.s~.~.~}.
wtx.~}.x}xxpx
t.pwt.wp}.q.
u~{{~x}v.|tsxrp{.ps
xrt.u~|.p.
{xrt}ts.prxx~}t=.
5/!Vjg!ftwi!qt!ftwiu!eqpvckpkpi!uwej!
c!uwduvcpeg!ctg!pgy!ftwiu!uq!tgncvgf!kp!
vjgkt!cevkqp!vq!c!ftwi!qt!ftwiu!cntgcf{!
nkuvgf!cu!jcxkpi!c!rqvgpvkcn!hqt!cdwug!vq!
ocmg!kv!nkmgn{!vjcv!vjg!ftwi!yknn!jcxg!vjg!
ucog!rqvgpvkcnkv{!hqt!cdwug!cu!uwej!
ftwiu-!vjwu!ocmkpi!kv!tgcuqpcdng!vq!
cuuwog!vjcv!vjgtg!oc{!dg!ukipkhkecpv!
fkxgtukqpu!htqo!ngikvkocvg!ejcppgnu-!
ukipkhkecpv!wug!eqpvtct{!vq!qt!ykvjqwv!
ogfkecn!cfxkeg-!qt!vjcv!kv!jcu!c!
uwduvcpvkcn!ecrcdknkv{!qh!etgcvkpi!
jc|ctfu!vq!vjg!jgcnvj!qh!vjg!wugt!qt!vq!
vjg!uchgv{!qh!vjg!eqoowpkv{/!
\pxyp}p.p}s.x.x|p.
rw~prx
t.x}vtsxt};.% <cWR;.p•t.
r~}~{{ts.qp}rt.x}.rwts{t.X.
}st.wt.RbP=.
cwt.WWb.pts.wp.~}t.p~
ts;.
|pztts.sv.~sr.r~}px}.
}wtxr.% <cWR;.p{~.z}~}.p.
s~}pqx}~{;.p}s.p}~wt.p~
ts;.
|pztts.sv.~sr.r~}px}.p.
rp}}pqx}~xs<{xzt.}wtxr.r~|~}s.
wp.x.rp{{.t{pts.~.% <cWR;.
wt.|px}.prx
t.r~|~}t}.x}.
|pxyp}p=.Q~w.~sr.pt.r~}~{{ts.
}st.wt.RbP=.
\px}~{.x.p.rwts{t.XXX.sv.~sr.
r~}px}x}v.}wtxr.% <cWR.
6s~}pqx}~{7.u~|{pts.x}.tp|t.~x{.x}.
~u.vt{px}.rp{t=.\px}~{.p.
p~
ts.q.wt.USP.x}[email protected]~.wt.
tp|t}.~u.}ptp.p}s.
~|xx}v.
p~rxpts.xw.rp}rt.rwt|~wtp.x}.
pxt}.w~.sxs.}~.t~}s.~.
r~}
t}x~}p{.p}x<t|txr.tp|t}=.X}.
@HHA;.USP.p~
ts.\px}~{.u~.wt.
tp|t}.~u.p}~txp.p~rxpts.xw.
txvw.{~.x}.pxt}.xw.prxts.
x||}~stuxrxt}r.}s~|t.6PXSb7=.
\px}~{.p.~xvx}p{{.{prts.x}~.
rwts{t.XX.p}s.{pt.trwts{ts.~.
rwts{t.XXX.}st.wt.RbP.st.~.wt.
G<^HT^HO^23CWR3/UIO
23CWR3
64853!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
{~.t~.~u.pqt.t{px
t.~.
|pxyp}p=.
Rtp|t.x.p.sv.~sr.r~}px}x}v.
wt.rwts{t.XX.qp}rt.}pqx{~}t;.p.
}wtxr.qp}rt.rp{{.t{pts.
~.% <cWR=.Rtp|t.p.p~
ts.u~.
|pztx}v.q.wt.USP.x}[email protected]~.wt.
tp|t}.~u.}ptp.p}s.
~|xx}v.
p~rxpts.xw.rp}rt.rwt|~wtp=.
P{{.~wt.}pp{{.~rrx}v.
rp}}pqx}~xs.x}.|pxyp}p.p}s.wtx.
}wtxr.tx
p{t}.xw.x|x{p.
rwt|xrp{.rt.p}s.wp|pr~{~vxrp{.
prx
x.pt.p{tps.x}r{sts.p.
rwts{t.X.sv.}st.wt.RbP=.
p}s.p.p.pt.~u.B?.x}ytrx~}.t.~}t.
w~.tx~}=.cp}sp.t.p{=.6A???7.ts.p.
cwt.sv.sxrx|x}px~}.ppsxv|.x.
{~t.s~t.~u.% <cWR.wp.p.pxs{.
ts.p.p}.p}x|p{.|~st{.~u.w|p}.
st{x
tts.6?=A.|{.x}ytrx~}.~
t.A??.|7.
qytrx
t.tuutr.6b~{x}p.t.p{=;.A??E7.
p}s.x.p.|tw~s.wtt.p}x|p{.pt.pq{t. wp}.x}.t
x~.t{u<ps|x}xpx~}.
sxt.rw.wp.p}p{vtxr.prx
x.~u.
~.x}sxrpt.wtwt.p.t.sv.x.pq{t.
% <cWR.p.}~.p.r~}u~}sx}v.upr~=.
~.~srt.wxrp{.~.rw~{~vxrp{.
cwt.pw~.p{~.pts.wp.wt.s~t.
rwp}vt.x|x{p.~.p.z}~}.sv.~u.
tt.r~|ppq{t.~.w~t.s~t.ts.q.
pqt=.P}x|p{.pt.px}ts.~.t.~}t.
w|p}.w~.|~zt.|pxyp}p=.P.RQ@.
qp.6x}.p}.~tp}.rwp|qt7.wt}.wt.
trt~.p}pv~}x.6ba@C@[email protected]{~rzts.
trtx
t.p.z}~}.sv.~u.pqt.p}s.
wx.tpsx}v.tuutr.~u.cWR=.
p}~wt.qp.wt}.wt.trtx
t.p.
Yx}~
p.t.p{=.6A??B7.tt.pq{t.~.
{prtq~=.fwt}.p.px}ts.p}x|p{.trtx
t. st|~}pt.t{u<ps|x}xpx~}.~u.% <.
p.t.sv;.xu.wt.sv.x.x|x{p.~.wt.
cWR.x}.sv<}p'¶
t.xt{.|~}zt.6}~.
z}~}.sv.~u.pqt;.x.x{{.t.wt.
t
x~.t~t.~.~wt.sv7=.cwt.
qp.p~rxpts.xw.wt.sv=.
D/!Cdwug!Nkcdknkv{!Uvwfkgu!
pw~.tts.wt.|~}zt.xw.t
tp{.
Sxrx|x}px
t.x|{.tuutr.~u.% <. s~t.~u.% <cWR.6@;.A;.C;.G;.p}[email protected]>.
X}.pssxx~}.~.wt.x}sxrp~.vvtts. cWR.wp
t.trxuxrx.u~.wt.
q.wt.RbP .{tvx{px
t.wx~;.spp.p. wp|pr~{~vxrp{.tuutr.~u.rp}}pqx}~xs. zv;.x=
=7.p}s.u~}s.wp.wt.|px|p{.
pt.~u.t{u<ps|x}xpx~}.tt.
~.tr{x}xrp{.p}s.r{x}xrp{.pqt.{xpqx{x. u~}s.x}.|pxyp}p.6Qp{t.p}s.
~qt
ts.xw.wt.C.Iv>zv>x}ux~}=.
sxt;.p.t{{.p.prp{.pqt;.
_tr~;[email protected]~}t.p}s.ftx|p};.
bqtt}{;.Qpxsp.t.p{=.6A??C7.
x}r{sx}v.r{p}stx}t.|p}uprt;.
@[email protected]{t.t.p{=;[email protected]{t.t.p{=;.
t~ts.wp.p.x{{.t{u<ps|x}xt.
puuxrzx}v;.p}s.sx
tx~}.u~|.
@HHD7=.P.|t}x~}ts.q.wt.WWb;.wt.
% <cWR.wt}.st{x
tts.
{tvxx|pt.~rt;.pt.r~}xstts.x}.
sxrx|x}px
t.x|{.tuutr.~u.
x}prttq~
t}xr{p{.6x=r=
=7;.q.
wx.upr~=.
rp}}pqx}~xs.ptp.~.qt.}xt.
~}{.p.wt.{~t.s~t.tts.6?=?@¤.
Pqt.{xpqx{x.t
p{px~}.pt.
qtrpt.pqts.sv.~u.~wt.r{pt.
?=?A.Iv>x}ux~};.x=r=
=7=.
~qpx}ts.u~|.sxt.x}.wt.rxt}xuxr.
x}r{sx}v.x|{p};.wp{{rx}~vt};.
bt{u<ps|x}xpx~}.qtwp
x~.xw.% <.
p}s.|tsxrp{.{xtpt=.cwtt.pt.|p}.
~x~xs;.qt}~sxptx}t;.qpqxpt;. cWR.p.u~}s.~.qt.p}pv~}xts.x}.p.
tr{x}xrp{.|tpt.~u.p.sv .tuutr.
]\SP.p}pv~}x;.p}s.p}xrw~xr.
p}s.xt{.|~}zt.q.x|~}pqp}.
wp.wt}.pzt}.~vtwt.~
xst.p}.
s~.}~.u{{.qxt.u~.% <cWR=.
6ba@C@F@EP;[email protected]}pv~}x7.p}s.wt.
prrpt.tsxrx~}.~u.wt.w|p}.pqt.
[pq~p~.p}x|p{.x}r{sx}v.
~x~xs.p}pv~}x.6}p{~~}t.p}s.
{xpqx{x=.R{x}xrp{.sxt.~u.wt.
|~}zt.6\r\pw~}.t.p{=;.A??H7;.|xrt.
}p{t~}t7.6cp}sp.t.p{=;.A???J.Qpxsp.t.
qytrx
t.p}s.tx}u~rx}v.tuutr.x}.
6\r\pw~}.t.p{=;.A??G7;.p}s.p.6V~{s.
p{=;.A??CJ.Yx}~
p.t.p{=;.A??C7=.
w|p}.p}s.txst|x~{~vxrp{.sxt.
t.p{=;[email protected]t.pq{t.~.sxrx|x}pt.
~
xst.p}xpx
t.spp.~}.pqt.
r=.R~}sxx~}ts._{prt._tutt}rt.bsxt.
rp}}pqx}~xs.u~|.~wt.sv.p}s.
{xpqx{x.x}.w|p}.p}s.~|t.x}sxrpx~}. {prtq~=.cwt.|py~.prx
t.|tpq~{xt.~u.
R~}sxx~}ts.{prt.tutt}rt.6R__7.x.
~u.prp{.pqt.t}s=.Q~w.tr{x}xrp{. % <cWR;.@@<ws~<% <cWR;.
p.qtwp
x~p{.pp.wtt.p}x|p{.pt.
p}s.r{x}xrp{.sxt.wp
t.r{tp{.
vt}tp{xt.~.% <cWR.6Q~}t.p}s.
vx
t}.wt.~~}x.~.t}s.x|t.x}.
st|~}pts.wp.|pxyp}p.p}s.% <.
ftx|p};.@HG@7=.X}.pssxx~};.prr~sx}v. ~.sxx}r.t}
x~}|t}I.~}t.wtt.
cWR.~t.wt.pxqt.p~rxpts.
~.wt.WWb;.t}<~.~wt.
wt.t
x~{.trtx
ts.p.sv.p}s.~}t.
xw.sv.~u.pqtI.cwt.u}rx~}.p.p. rp}}pqx}~xs.u~}s.x}.|pxyp}p.p{~.
wtt.wt.trtx
ts.p.{prtq~=.Xu.wt.
~xx
t.tx}u~rt.~.|px}px}.sv<.
qxt.u~.% <cWR=.P.{tp.~}t.
sv.x.tx}u~rx}v;.p}x|p{.x}.p.sv<.
ttzx}v.qtwp
x~;.wt.u}rx~}.p.p.
rp}}pqx}~xs;.RQS;.s~t.}~.qxt.
utt.pt.x{{.rw~~t.~.t}s.|~t.
sxrx|x}px
t.x|{;.p}s.wt.wp
t. u~•.% <cWR.x}.p.6ep}}.t.p{=;.A??G7=.
x|t.x}.wt.t}
x~}|t}.pxts.xw.wt.
stt}st}rt.~t}xp{=.
sv.wt}.q~w.t}
x~}|t}.pt.
q=.bt{u<Ps|x}xpx~}.bsxt.
_tr{x}xrp{.p}s.|~.r{x}xrp{.pqt.
tt}ts.x|{p}t~{=.
P}x|p{.t{u<ps|x}xpx~}.qtwp
x~.
{xpqx{x.sxt.wp
t.qtt}.r~}srts.
R__.wp.qtt}.st|~}pts.xw.% <.
p~rxpts.xw.p.sv.x.p.r~||~}{.
xw.wt.rw~prx
t.r~}xt}.~u.
cWR.x}.p.q.~}{.p.{~.s~t.
ts.|tw~s.u~.t
p{px}v.xu.wt.sv.
|pxyp}p;.x|px{.% <cWR.p}s.x.
6?=?FD¤@=?.|v>zv;.x==J.Qpxsp.t.p{=;.
~srt.tpsx}v.tuutr.p}s.u~.
|tpq~{xt;.@@<ws~<% <cWR=.% <.
A??C7=.ax|~}pqp}.6?=AD¤@=?.|v>zv;.x==7.
cWR .qytrx
t.tuutr.pt.r~}xstts. tsxrx}v.pqt.~t}xp{.6Qp{t;.
p}s.}p{~~}t.6?=D¤A=?.|v>zv;.x==7.
@[email protected]{t.p}s.Qxvt{~;.A??B7=.Sv. p}pv~}xts.% <cWR<|tsxpts.R__.
~.qt.wt.qpx.u~.|pxyp}p .pqt.
{xpqx{x=.cwt.u~{{~x}v.sxt.~
xst. wp.pt.t{u<ps|x}xtts.q.p}x|p{.pt. 6Qpxsp.t.p{=;.A??C7=.W~t
t;.x}.
{xzt{.~.~srt.tpsx}v.tuutr.x}.
p.||p.~u.wp.spp=.
p}~wt.s.xw.p;.x|~}pqp}.p.
w|p}=.P.|t}x~}ts.x}.wt.WWb.
st|~}pts.~.x}srt.R__.p.s~t.
2/!Rtgenkpkecn!Uvwfkgu!
t
xt.s~r|t};.tp{xt.pt|.~.
p}vx}v.u~|.?=AD¤B=?.|v>zv.6Rwtt.t.
% <cWR;.wt.x|p.rw~prx
t.
st|~}pt.t{u<ps|x}xpx~}.~u.% <.
p{=;.A???7=.\xrt.xw~.I<~•x~xs.
r~|~}t}.x}.|pxyp}p;.x.p}.tuutrx
t. cWR.tt.}rrtu{.p}s.r~}u~}sts. trt~.sxs.}~.twxqx.R__.~.% <cWR.
tx}u~rt.x}.{pq~p~.p}x|p{;.
q.sxt.txrx~};.p}x|p{.tpx};.
6pxts.xw.@.|v>zv.% <cWR;.x=•=7.
x}r{sx}v.x|pt.p}s.~st};.p.wtt. p}s.z}~}.p}p{vtxr.prx
x.~u.% <cWR. 6Vw~{p}s.t.p{=;.A??A7=.
p}x|p{.x{{.t{u<ps|x}xt.% <cWR=.
p.tx}v.s~t.6cp}sp.p}s.V~{sqtv;.
3/!Enkpkecn!Uvwfkgu!
cwtt.p}x|p{.sxt.q~w.tsxr.p}s.
A??BJ.Yx}~
p.t.p{=;.A??B7=.bt{u<.
~.wt.~qt
px~}.wp.% <cWR;.
ps|x}xpx~}.~u.% <cWR.p.ux.
X}.x.rxt}xuxr.t
xt.6WWb;.A?@D7;.
wtwt.|~zts.p.|pxyp}p.~.
st|~}pts.q.cp}sp.t.p{=.6A???7=.
wt.WWb.~
xsts.p.{x.~u.r~||~}.
ps|x}xtts.q.~wt.~t;.~srt. cp}sp.t.p{=.6A???7.w~ts.wp.xt{. qytrx
t.rw~prx
t.t~}t.~.
tx}u~rx}v.tuutr.x}.w|p}=.brw.
|~}zt.wp.tt.x}xxp{{.px}ts.~.
rp}}pqx}~xs.qpts.~}.x}u~|px~}.u~|.
tx}u~rx}v.tuutr.rp}.prr~}.u~.wt.
t{u<ps|x}xt.r~rpx}t.6B?.Iv>zv;.x=
=7.
t
tp{.tutt}rt.6Psp|.p}s.\px};.
ttpts.pqt.~u.|pxyp}p=.
t{u<ps|x}xtts.A.Iv>zv.% <cWR.6x=
=7.
@HHEJ.V~}p{t;.A??FJ.W~{{xt;.@HGEJ.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
p=.Sv.Sxrx|x}px~}.bsxt.
RQ!11111
Hto!11167
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
W~{{xt;[email protected]}xt.~u.\tsxrx}t;.
@HGA7=.Uwt|~t;.\p{s~}ps~.6A??A7.
rwpprtxts.wtt.qytrx
t.t~}t.
p.{tppq{t.~.|~.w|p}.p}s.pt.
vt}tp{{.p~rxpts.xw.sv<ttzx}v.
p}s>~.sv<pzx}v=.[pt.sxt.
6brwtt.t.p{=;.A??HJ.itxvt.t.p{=;.A?@?7.
t~ts.wp.wxvw.{t
t{.~u.~xx
t.
rw~prx
t.tuutr.r~t{pt.xw.
x}rtpts.|pxyp}p.t;.pqt;.p}s.
stt}st}rt=.cwt.{x.~u.wt.r~||~}.
qytrx
t.rw~prx
t.tuutr.~
xsts.
q.wt.WWb.6WWb;[email protected].tt}ts.
qt{~I.
[email protected]!tgnczcvkqp-!
kpetgcugf!uqekcdknkv{-!cpf!vcnmcvkxgpguu/!
6A7.Kpetgcugf!ogttkogpv!cpf!crrgvkvg-!
cpf!gxgp!gzjknctcvkqp!cv!jkij!fqugu/!
6B7.Gpjcpegf!ugpuqt{!rgtegrvkqp-!
yjkej!ecp!igpgtcvg!cp!kpetgcugf!
crrtgekcvkqp!qh!owuke-!ctv-!cpf!vqwej/!
6C7.Jgkijvgpgf!kocikpcvkqp-!yjkej!
ecp!ngcf!vq!c!uwdlgevkxg!ugpug!qh!
kpetgcugf!etgcvkxkv{/!
6D7.Kpkvkcn!fk||kpguu-!pcwugc-!
vcej{ectfkc-!hcekcn!hnwujkpi-!ft{!oqwvj-!
cpf!vtgoqt/!
6E7.Fkuqticpk|gf!vjkpmkpi-!kpcdknkv{!vq!
eqpxgtug!nqikecnn{-!vkog!fkuvqtvkqpu-!cpf!
ujqtv.vgto!ogoqt{!korcktogpv/!
6F7.Cvczkc!cpf!korcktgf!lwfiogpv-!
yjkej!ecp!korgfg!ftkxkpi!cdknkv{!qt!ngcf!
vq!cp!kpetgcug!kp!tkum.vcmkpi!dgjcxkqt/!
6G7.Knnwukqpu-!fgnwukqpu-!cpf!
jcnnwekpcvkqpu!vjcv!kpvgpukh{!ykvj!jkijgt!
fqugu/!
6H7.Goqvkqpcn!ncdknkv{-!kpeqpitwkv{!qh!
chhgev-!f{urjqtkc-!cikvcvkqp-!rctcpqkc-!
eqphwukqp-!ftqyukpguu-!cpf!rcpke!
cvvcemu-!yjkej!ctg!oqtg!eqooqp!kp!
kpgzrgtkgpegf!qt!jkij.fqugf!wugtu/!
cwt.WWb.|t}x~}ts.wp.|pxyp}p.
t.tut.wxvwt.r~}rt}px~}.~u.wt.
x}rxp{.rw~prx
t.r~|~}t}.6% <.
cWR7.~
t.{~t.r~}rt}px~}=.X}.p.
r{x}xrp{.s.xw.|pxyp}p.t.6}.L.
@A;.pvt.p}vts.u~|.~}rt.p.|~}w.~.
C.x|t.p.ttz7;.qytr.tt.vx
t}.p.
rw~xrt.~u.@=HD3.% <cWR.|pxyp}p.~.
?=EB3.% <cWR.|pxyp}p.put.p|{x}v.
q~w.|pxyp}p.rxvptt.x}.~.rw~xrt.
tx~}=.cwt.|pxyp}p.rxvptt.xw.
wxvw.cWR.p.rw~t}.x}.A@.~.~u.AC.
rw~xrt.tx~}.~.GF=D3.~u.wt.x|t.
6Rwpx.p}s.Qzt;.@HHC7=.Uwt|~t;.
x}.p.s~q{t<q{x}s.s;.utt}.
|pxyp}p.t.6}.L.@@;.pvt.p.{tp.
A.x|t.t.|~}w.xw.p.{tp.@??.
~rrpx~}7.wt}.vx
t}.p.{~<s~t.~u.
~•p{.% <cWR.6F=D.|v7.tt.pq{t.~.
sxx}vxw.wt.rw~prx
t.tuutr.
qtt.wp}.~rrpx~}p{.t.6}.L.@?;.}~.
t.xwx}.wt.p.C.tp.xw.@?.~.
utt.{xutx|t.t7.p}s.p{~.
ttxt}rts.utt.tspx
t.tuutr.6Zxz.
p}s.st.fx;.@HHH7=.
\pxyp}p.wp.p{~.qtt}.tr~v}xts.
q.rxt}xuxr.tt.~.wp
t.xwspp{.
|~|.6}tvpx
t.tx}u~rt|t}7.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
u~{{~x}v.|~stpt.p}s.wtp
.t=.P.
sxrts.uwt.x}.Upr~.F;.wt.STP.
}~t.wp.wt.P|txrp}._rwxpxr.
P~rxpx~} .6P_P7.Sxpv}~xr.p}s.
bpxxrp{.\p}p{.~u.\t}p{.Sx~st;.
Uxuw.Tsxx~}.6Sb\¤D7.x}r{sts.p.{x.~u.
xwspp{.|~|.u~{{~x}v.
|pxyp}p.jrp}}pqxl.t.6Sb\¤D;.
A?@B7=.
E/!Cevwcn!Cdwug!qh!Octklwcpc!
Pcvkqpcn!Fcvcdcugu!Tgncvgf!vq!
Octklwcpc!Cdwug!cpf!Vtchhkemkpi!
\pxyp}p.r~}x}t.~.qt.wt.|~.
xst{.ts.x{{xrx.sv=.T
xst}rt.~u.
prp{.pqt.rp}.qt.stux}ts.q.
tx~st>|t}x~}.x}.sppqpt.
x}sxrpx
t.~u.pqt>stt}st}rt=.cwt.
WWb.~
xsts.x}.x.tr~||t}spx~}.
6WWb;[email protected]}u~|px~}.t{t
p}.~.
prp{.pqt.~u.|pxyp}p.x}r{sx}v.spp.
t{.u~|.wt.]px~}p{.b
t.~}.
Sv.dt.p}s.Wtp{w.6]bSdW7;.p.
\~}x~x}v.wt.Ut.6\cU7.
t;.
wt.Sv.Pqt.fp}x}v.]t~z.
6SPf]7;.p}s.wt.ctp|t}.Tx~st.
Spp.bt.6cTSb7=.cwtt.spp.~rt.
~
xst.p}xpx
t.x}u~|px~}.~}.
|p}.upr~.t{pts.~.pqt.~u.p.
pxr{p.qp}rt;.x}r{sx}v.
x}rxst}rt.p}s.pt}.~u.t;.p}s.
~ux{t.~u.wt.pqt.~u.trxuxr.
qp}rt=.cwt.STP.x.~
xsx}v.
spts.x}u~|px~}.u~|.wtt.
sppqpt.x}.wx.sxrx~}=.cwt.STP.
p{~.x}r{st.spp.~}.puuxrzx}v.p}s.
x{{xrx.p
px{pqx{x.~u.|pxyp}p.u~|.
STP.sppqpt.x}r{sx}v.wt.]px~}p{.
U~t}xr.[pq~p~.X}u~|px~}.bt|.
6]U[Xb7.p}s.wt.]px~}p{.btxt.
bt|.6]bb7;.u~|t{.wt.Utstp{<.
xst.Sv.btxt.bt|.6USbb7;.p.
t{{.p.~wt.~rt.~u.spp.trxuxr.~.
|pxyp}p;.x}r{sx}v.wt._~t}r.
\~}x~x}v._~ytr.p}s.wt.S~|txr.
Rp}}pqx.Tpsxrpx~}.p}s.btx~}.
_•~v•p|.6SRT>b_7=.
2/!Pcvkqpcn!Uwtxg{!qp!Ftwi!Wug!cpf!
Jgcnvj!)PUFWJ*!
cwt.]px~}p{.b
t.~}.Sv.dt.p}s.
Wtp{w.6]bSdW7.x.r~}srts.p}}p{{.
q.wt.Stp|t}.~u.Wtp{w.p}s.W|p}.
bt
xrt .bqp}rt.Pqt.p}s.\t}p{.
Wtp{w.bt
xrt.Ps|x}xpx~}.
6bP\WbP7=.bP\WbP.x.wt.x|p.
~rt.~u.tx|pt.~u.wt.t
p{t}rt.
p}s.x}rxst}rt.~u.wp|prtxrp{.sv;.
x{{xrx.sv;.p{r~w~{;.p}s.~qprr~.t.x}.
wt.d}xts.bpt=.cwt.
t.x.qpts.
~}.p.}px~}p{{.ttt}px
t.p|{t.~u.
wt.rx
x{xp};.}~}<x}xx~}p{xts.
~{px~}.@A.tp.~u.pvt.p}s.~{st=.
cwt.
t.tr{st.w~|t{t.t~{t.
w~.s~.}~.t.wt{t;.prx
t.|x{xp.
t~}}t{;.p}s.txst}.~u.x}xx~}p{.
v~.pt.rw.p.ypx{.p}s.
w~xp{=.
RQ!11111
Hto!11168
Hov!5812
Uhov!5813
64854!
Prr~sx}v.~.wt.A?@C.]bSdW.t~;.
|pxyp}p.p.wt.|~.r~||~}{.ts.
p}s.pqts.x{{xrx.sv=.cwp.spp.
w~ts.wp.wtt.tt.AA=A.|x{{x~}.
t~{t.w~.tt.p.|~}w.t.
6G=C37.p|~}v.w~[email protected]}s.~{st.
x}.wt.d}xts.bpt=.6]~tI.]bSdW.
uxvt.~}.|pxyp}p.t.x}r{st.
wpwxw.tJ.wt.t{px
t.~~x~}.~u.
wpwxw.t.~.|pxyp}p.t.x.
t.
{~=7.\pxyp}p.wps.wt.wxvwt.pt.~u.
p<tp.stt}st}rt.~.pqt.x}.A?@C=.
cwt.]bSdW.t~.tx|pt.wp.B=?.
|x{{x~}.t~{t.pvts.@A.~.~{st.ts.p}.
x{{xrx.sv.u~.wt.ux.x|t.x}[email protected].
|py~x.6F?=B37.~u.wtt.p.tp.
x}xxpt.t~ts.wp.wtx.ux.sv.
ts.p.|pxyp}p=.P|~}v.w~t.w~.
qtvp}.x}v.x{{xrx.sv.x}.wt.p.tp;.
ED=E3;.F?=B3;.p}s.EF=E3.t~ts.
|pxyp}p.p.wt.ux.x{{xrx.sv.
x}xxpts.x}.A?@A;.A?@B;.p}s.A?@C.
ttrx
t{=.X}.A?@C;.wt.p
tpvt.pvt.~u.
|pxyp}p.x}xxpt.p|~}v.@A<.~.CH<.
tp<~{s.p.@G=D.tp=.cwtt.pvt.
pt.p}s.st|~vpwxr.pt.t{t
p}.x}.
{xvw.~u.wt.xz.tt}ts=.
\pxyp}p.wps.wt.wxvwt.pt.~u.p.
tp.stt}st}rt.~.pqt.~u.p}.x{{xrx.
sv.x}.A?@C=.cwt.A?@C.]bSdW.t~.
pts.wp.C=A.|x{{x~}.t~}.tt.
r{pxuxts.xw.qp}rt.stt}st}rt.~.
pqt.~u.|pxyp}p.x}.wt.p.tp.
6ttt}x}v.@=E3.~u.wt.~p{.
~{px~}.pvts.@A.~.~{st;.p}s.DH=?3.
~u.w~t.r{pxuxts.xw.x{{xrx.sv.
stt}st}rt.~.pqt7.qpts.~}.rxtxp.
trxuxts.x}.wt.Sxpv}~xr.p}s.
bpxxrp{.\p}p{.~u.\t}p{.Sx~st;.
Cw.tsxx~}.6Sb\¤Xe7=.
P|~}v.p.tp.|pxyp}p.t.pvt.
@A.~.~{st;.@G=D3.ts.|pxyp}p.~}.
B??.~.|~t.sp.xwx}.wt.t
x~.@A.
|~}w.x}.A?@C=.cwx.p}{pt.x}~.E=D.
|x{{x~}.t~{t.x}v.|pxyp}p.~}.p.
spx{.~.p{|~.spx{.qpx.~
t.p.@A<.
|~}w.tx~s;.xv}xuxrp}{.|~t.wp}.
wt.tx|pts.D=F.|x{{x~}.spx{.~.p{|~.
spx{.t.x}.y.wt.tp.qtu~t=.
P|~}v.p.|~}w.|pxyp}p.t;.
C@=E3.6H=A.|x{{x~}7.ts.wt.sv.~}.A?.
~.|~t.sp.x}.wt.p.|~}w;.p.
xv}xuxrp}.x}rtpt.u~|.wt.G=@.|x{{x~}.
w~.ts.|pxyp}p.A?.sp.~.|~t.x}.
A?@B=.
3/!Oqpkvqtkpi!vjg!Hwvwtg!)OVH*!
\~}x~x}v.wt.Ut.6\cU7.x.p}.
~}v~x}v.s.wxrw.x.u}sts.}st.p.
txt.~u.x}
txvp~<x}xxpts.
r~|tx}v.ttprw.vp}.u~|.wt.
]px~}p{.X}xt.~}.Sv.Pqt.
6]XSP7=.\cU.prz.sv.t.t}s.
p|~}v.P|txrp}.ps~{trt}.x}.wt.Gw;.
@?w;.p}s.@Aw.vpst=.Prr~sx}v.~.x.
A?@D.
t.t{;.|pxyp}p.p.wt.
|~.r~||~}{.ts.x{{xrx.sv;.p.p.
wt.rpt.x}.t
x~.tp=.
P~x|pt{.E=D3.~u.Gw.vpst;.
G<^HT^HO^23CWR3/UIO
23CWR3
64855!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
TS.
xx.x}
~{
x}v.p{{.pqt.~.|xt.
x}.wt.d}xts.bpt.p}s.~.~u.@=AD.
|x{{x~}.
xx.x}
~{
x}v.pqt.~.|xt.
~u.x{{xrx.sv.6tr{sx}v.p{r~w~{<.
t{pts.
xx7;.p.tx|pts.q.SPf]=.
cwx.x.{~t.wp}.wt.}|qt.~u.TS.
xx.x}
~{
x}v.r~rpx}t.6D?D;AAC7.p}s.
wxvwt.wp}.wt.}|qt.~u.TS.
xx.
x}
~{
x}v.wt~x}.6ADG;CGA7.p}s.
4/!Ftwi!Cdwug!Yctpkpi!Pgvyqtm!
x|{p}.6g/i/-!p|wtp|x}t;.
)FCYP*-!Gogtigpe{!Fgrctvogpv!)GF*!
|twp|wtp|x}t7.6@DH;GC?7=.exx.
Xkukvu!
x}
~{
x}v.wt.~wt.|py~.x{{xrx.sv;.
cwt.Sv.Pqt.fp}x}v.]t~z.
rw.p.\S\P;.VWQ;.[bS.p}s.~wt.
6SPf]7.x.p.q{xr.wtp{w.
tx{{p}rt. wp{{rx}~vt};._R_;.p}s.x}wp{p};.tt.
t|.wp.|~}x~.sv<t{pts.
|rw.{t.utt};.r~|ppx
t{=.
w~xp{.t|tvt}r.stp|t}.6TS7.
X}.~}v.pxt};.|pxyp}p.x.wt.
xx.~.prz.wt.x|pr.~u.sv.t;.
x{{xrx.sv.|~.utt}{.x}
~{
ts.x}.
|xt;.p}s.pqt.x}.wt.d}xts.bpt=. TS.
xx;.prr~sx}v.~.SPf].tx|pt;.
U~.wt.~t.~u.SPf];.wt.t|.
xw.AC?=A.|pxyp}p<t{pts.TS.
xx.
sv.pqt .p{xt.xu.wt.u~{{~x}v.
t.@??;???.~{px~}.pvt.@A.~.@F;.
r~}sxx~}.pt.|t[email protected]t.
CCB=G.t.@??;???.~{px~}.pvt.@G.~.
x}
~{
ts.p.{tp.~}t.~u.wt.u~{{~x}vI.
A?;.p}s.CCE=H.t.@??;???.~{px~}.
t.~u.p}.x{{tvp{.sv;.t.~u.p.{tvp{.sv. pvt.A@.~.AC=.
r~}p.~.sxtrx~};.~.x}wp{px~}.~u.p.
5/!Vtgcvogpv!Grkuqfg!Fcvc!Ugv!)VGFU*!
}~}<wp|prtxrp{.qp}rtJ.p}s.6A7.
U{uvgo!
wt.qp}rt.p.ts.u~.~}t.~u.wt.
cwt.ctp|t}.Tx~st.Spp.bt.
u~{{~x}v.tp~}I.Qtrpt.~u.sv.
6cTSb7.t|.x.p.~u.wt.bP\WbP.
stt}st}rt;.~.r~||x.xrxst.6~.
Sv.p}s.P{r~w~{.bt
xrt.X}u~|px~}.
pt|.~.r~||x.xrxst7;.u~.
bt|.p}s.x.p.}px~}p{.rt}.~u.
trtpx~}p{.~t;.~.~.prwxt
t.
~wt.rwxr.tuutr=.X|~p}{;.|p}. p}}p{.ps|xx~}.~.pt.{xrt}ts.~.
upr~.rp}.x}u{t}rt.wt.tx|pt.~u.TS. rtxuxts;.~.ps|x}xpx
t{.przts;.
xx;.x}r{sx}v.t}s.x}.~
tp{{.t.~u. qp}rt.pqt.tp|t}.uprx{xxt=.cwt.
cTSb.t|.r~}px}.x}u~|px~}.~}.
p.qp}rt.p.t{{.p.t}s.x}.wt.
tp~}.u~.TS.pvt=.U~.x}p}rt;.~|t. pxt}.st|~vpwxr.p}s.qp}rt.
pqt.~q{t|.~u.ps|xx~}.~.
sv.t.|p.
xx.TS.u~.{xut<.
tp|t}.u~.pqt.~u.p{r~w~{.p}s>~.
wtpt}x}v.xt.wx{t.~wt.|p.
sv.x}.uprx{xxt.wp.t~.~.pt.
xx.~.ttz.rpt.u~.st~xuxrpx~}.
qtrpt.wt.}ttsts.rtxuxrpx~}.qtu~t. ps|x}xpx
t.spp.t|=.U~.wx.
sppqpt;.wt.x|p.qp}rt.~u.
t}tx}v.tp|t}=.Pssxx~}p{{;.
SPf].spp.s~.}~.sxx}vxw.wt.sv. pqt.x.stux}ts.p.wt.|px}.qp}rt.
~u.pqt.t~ts.p.wt.x|t.~u.
t~}xq{t.u~.wt.TS.
xx.u~|.~wt.
sv.wp.|p.wp
t.qtt}.ts.
ps|xx~}=.cTSb.p{~.p{{~.u~.wt.
r~}r~|xp}{=.P.pts.x}.p.SPf].
tr~sx}v.~u.~.~wt.qp}rt.~u.
t~;.bx}rt.|pxyp}p>wpwxw.x.
pqt.6tr~}sp.p}s.txp7=.
X}.A?@@;.wt.cTSb.t|.x}r{sts.
utt}{.tt}.x}.r~|qx}px~}.xw.
@;HAG;FHA.ps|xx~}.~.qp}rt.
~wt.sv;.wt.tp~}.u~.wt.TS.
xx.
pqt.tp|t}J.x}.A?@A.wtt.tt.
|p.qt.|~t.t{t
p}.~.wt.~wt.
@;G?@;BGD.ps|xx~}J.p}s.x}.A?@B.wtt.
sv67.x}
~{
ts.x}.wt.tx~st= .
tt.@;EGB;[email protected]|xx~}=.\pxyp}p>
X}.A?@@;.|pxyp}p.p.x}
~{
ts.x}.
CDD;EEG.TS.
xx.~.~u.A;CEA;HCG.~p{. wpwxw.p.wt.x|p.qp}rt.~u.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
@C=G3.~u.@?w.vpst;.p}s.A@=B3.~u.
@Aw.vpst.
tts.x}.A?@D.t~ts.
|pxyp}p.t.sx}v.wt.p.|~}w.
x~.~.wt.
t=.P.}|qt.~u.wxvw.
rw~~{.st}.x}[email protected]{~.t~ts.
spx{.t.x}.wt.p.|~}w;.x}r{sx}v.
@=@3;.B=?3;.p}s.E=?3.~u.Gw;.@?w;.p}s.
@Aw.vpst;.ttrx
t{=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11169
Hov!5812
Uhov!5813
pqt.u~.@G=B3.6BDA;BHF7.~u.ps|xx~}.
x}.A?@@J.@F=D3.6B@D;A??7.x}[email protected]}s.
@E=G3.6AG@;[email protected]}.A?@B=.^u.wt.AG@;HH@.
ps|xx~}.u~.|pxyp}p>wpwxw.
tp|t}.x}.A?@B;.AC=B3.ts.
|pxyp}p>wpwxw.spx{=.P|~}v.w~t.
tpts.u~.|pxyp}p>wpwxw.p.wt.
x|p.qp}rt.x}.A?@B;.AF=C3.tt.
pvt.@A.~.@F.tp.p}s.AH=F3.tt.pvt.
@G.~.AC.tp=.cw~t.ps|xts.u~.
|pxyp}p>wpwxw.tt.|~{.|p{t.
6FA=E37.p}s.}~}<Wxp}xr.6GA=A37=.
]~}<wxp}xr.wxt.6CB=A37.
ttt}ts.wt.{pvt.tw}xr.v~.~u.
|pxyp}p.ps|xx~}=.
6/!Hqtgpuke!Ncdqtcvqt{!Fcvc!
Spp.~}.|pxyp}p.txt.u~|.
utstp{;.pt;.p}s.{~rp{.u~t}xr.
{pq~p~xt.wp
t.x}sxrpts.wp.wtt.x.
xv}xuxrp}.puuxrzx}v.~u.|pxyp}p=.cwt.
]px~}p{.U~t}xr.[pq~p~.bt|.
6]U[Xb7.x.p.~vp|.~}~ts.q.wt.
Sv.T}u~rt|t}.Ps|x}xpx~} .
^uuxrt.~u.Sx
tx~}.R~}~{=.]U[Xb.
t|pxrp{{.r~{{tr.sv.
xst}xuxrpx~}.t{.p}s.p~rxpts.
x}u~|px~}.u~|.sv.twxqx.
t}r~}tts.q.{p.t}u~rt|t}.p}s.
p}p{ts.x}.utstp{;.pt;.p}s.{~rp{.
u~t}xr.{pq~p~xt=.]U[Xb.x.p.
r~|twt}x
t.x}u~|px~}.t|.wp.
x}r{st.spp.u~|.AFG.x}sx
xsp{.
u~t}xr.{pq~p~xt.wp.t~.|~t.
wp}.H@3.~u.wt.sv.rpt{~ps.x}.wt.
d=b=.]U[Xb.rpt.spp.u~.p{{.sv.
p}s.rwt|xrp{.xst}xuxts.p}s.t~ts.
q.u~t}xr.{pq~p~xt=.\~t.wp}.
@;F??.}xt.qp}rt.pt.ttt}ts.
x}.wt.]U[Xb.sppqpt=.
Spp.u~|.]U[Xb.w~ts.wp.
|pxyp}p.p.wt.|~.utt}{.
xst}xuxts.sv.x}.utstp{;.pt;.p}s.
{~rp{.{pq~p~xt.u~|.Yp}p.A??C.
w~vw.Strt|qt.A?@C=.\pxyp}p.
prr~}ts.u~.qttt}.AH=CF3.p}s.
BC=GC3.~u.p{{.sv.twxqx.p}p{ts.
p}}p{{.sx}v.wp.x|t.up|t.
6cpq{t.@7=.
G<^HT^HO^23CWR3/UIO
23CWR3
7/!Hgfgtcn.Ykfg!Ftwi!Ugk|wtg!U{uvgo!
cwt.Utstp{<xst.Sv.btxt.
bt|.6USbb7.r~}px}.x}u~|px~}.
pq~.sv.txt.|pst.xwx}.wt.
yxsxrx~}.~u.wt.d}xts.bpt.q.wt.
Sv.T}u~rt|t}.Ps|x}xpx~};.wt.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
8/!Rqvgpe{!Oqpkvqtkpi!Rtqlgev!
cwt.d}x
tx.~u.\xxxx .
_~t}r.\~}x~x}v._~ytr.6_\_7;.
w~vw.p.r~}pr.xw.wt.]px~}p{.
X}xt.~}.Sv.Pqt.6]XSP7;.
p}p{t.p}s.r~|x{t.spp.~}.wt.% <.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
Utstp{.Qtp.~u.X}
txvpx~};.d}xts.
bpt.R~|.p}s.Q~st._~trx~};.
p}s.d}xts.bpt.X||xvpx~}.p}s.
R~|.T}u~rt|t}=.X.p{~.tr~s.
|pxx|t.txt.|pst.q.wt.d}xts.
bpt.R~p.Vps=.Sv.txt.|pst.
q.~wt.Utstp{.pvt}rxt.pt.x}r{sts.
x}.wt.USbb.sppqpt.wt}.sv.
t
xst}rt.r~s.x.p}utts.~.~}t.~u.
wt.pvt}rxt.xst}xuxts.pq~
t=.USbb.x.
}~.x}r~~pts.x}~.wt.]px~}p{.
btxt.bt|.6]bb7;.wxrw.x.p.
t~x~.u~.x}u~|px~}.~}.
r{p}stx}t.{pq~p~.p}s.r~}pqp}s.
6rwt|xrp{.p}s.tr~;.rt}r;.
sv;.tx|t}.p}s.tp~}7=.USbb.
t~.~p{.utstp{.sv.txt.jx}.
zx{~vp|.6zv7l.~u.qp}rt.rw.p.
r~rpx}t;.wt•~x};.\S\P;.
|twp|wtp|x}t;.p}s.rp}}pqx.
6|pxyp}p.p}s.wpwxw7=.cwt.tp{.
~{|t.~u.rp}}pqx.txts.6cpq{t.A7;.
r~}xt}{.trttsx}v.p.w~p}s.
|txr.~}.t.tp;.w~.wp.
rp}}pqx.x.
t.xst{.puuxrzts.x}.wt.
d}xts.bpt=.
cWR.r~}rt}px~}.~u.|pxyp}p;.
wpwxw.p}s.wpw.~x{.p|{t.~
xsts.
q.STP.tvx~}p{.{pq~p~xt.p}s.q.
pt.p}s.{~rp{.~{xrt.pvt}rxt=.Put.
A?@?;._\_.wp.p}p{ts.~}{.|pxyp}p.
p|{t.~
xsts.q.STP.tvx~}p{.
{pq~p~xt=.P.x}sxrpts.x}.Uxvt.@;.
wt.trt}pvt.~u.% <cWR.x}rtpts.
u~|.@HHD.~.A?@?.xw.p}.p
tpvt.cWR.
r~}t}.~u.B=FD3.x}[email protected]}s.H=DB3.x}.
A?@?=.X}.tp|x}x}v.|pxyp}p.p|{t.
~}{.~
xsts.q.STP.{pq~p~xt;.wt.
p
tpvt.% <cWR.r~}t}.p.B=HE3.x}.
@HHD.x}.r~|px~}.~.@@[email protected]}.A?@D=.
RQ!11111
Hto!1116;
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/132>0IRJ@
bx}rt.A??C;.wt.~p{.}|qt.~u.
t~.~u.|pxyp}p.p}s.wt.p|~}.~u.
|pxyp}p.t}r~}tts.utstp{{.wp.
t|px}ts.wxvw.6tt.spp.u~|.Utstp{<.
xst.Sv.btxt.bt|.p}s.S~|txr.
Rp}}pqx.Tpsxrpx~}.p}s.btx~}.
_~vp|.qt{~7=.
64856!
GR23CW27/131>0IRJ@
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
9/!Vjg!Fqoguvke!Ecppcdku!Gtcfkecvkqp!
cpf!Uwrrtguukqp!Rtqitco!
cwt.S~|txr.Rp}}pqx.Tpsxrpx~}.
p}s.btx~}._~vp|.6SRT>b_7.p.
tpq{xwts.x}.@HFH.~.tsrt.wt.{.
~u.s~|txrp{{.r{x
pts.|pxyp}p.x}.
wt.d}xts.bpt=.cwt.~vp|.p.
stxv}ts.~.t
t.p.p.p}twx.
qttt}.utstp{;.pt;.p}s.{~rp{.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
pvt}rxt=.^}{.Rp{xu~}xp.p}s.Wppxx.
tt.prx
t.pxrxp}.x}.wt.~vp|.
p.x.x}rtx~}=.W~t
t;.q.@HGA.wt.
~vp|.wps.tp}sts.~.AD.pt.p}s.
q[email protected]{{.D?.pt.tt.pxrxp}=.
Rp}}pqx.x.r{x
pts.x}.t|~t.
{~rpx~}.p}s.utt}{.~}.q{xr.{p}s.
p}s.x{{xrx{.v~}.x}.p{{.pt=.Spp.
~
xsts.q.wt.SRT>b_.6cpq{t.B7.w~.
RQ!11111
Hto!11171
Hov!5812
Uhov!5813
wp.x}.wt.d}xts.bpt.x}.A?@C;.wtt.
tt.B;H?C;A@B.{p}.tpsxrpts.x}.
~s~~.rp}}pqx.r{x
px~}.ptp.
r~|pts.~.A;DHF;FHG.{p}.x}.A???=.
bxv}xuxrp}.p}xxt.~u.|pxyp}p.tt.
p{~.tpsxrpts.u~|.x}s~~.r{x
px~}.
~tpx~}=.cwtt.tt.BHE;EA?.x}s~~.
{p}.tpsxrpts.x}[email protected]~|pts.~.
A@F;@?D.tpsxrpts.x}.A???=.
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/133>0IRJ@
64857!
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
cwt.trt}.pxxr.u~|.wtt.
px~.
t.p}s.sppqpt.w~.wp.
|pxyp}p.r~}x}t.~.qt.wt.|~.
r~||~}{.ts.x{{xrx.sv;.xw.
r~}xstpq{t.pt.~u.wtp
.pqt.p}s.
stt}st}rt=.cwt.p{~.w~.wp.
|pxyp}p.x.wt.|~.tpsx{.p
px{pq{t.
x{{xrx.sv.x}.wt.d}xts.bpt=.
Uwt|~t;.p.}~ts.q.wt.WWb;.
|p}.qp}rt.rwts{ts.}st.wt.
RbP.pt.t
p{pts.xwx}.wt.r~}t.~u.
sv.st
t{~|t}.x}v.spp.q|xts.
}st.p.]t.Sv.P{xrpx~}.6]SP7=.
W~t
t;.wt.txx~}t.wp
t.}~.
xst}xuxts.p.trxuxr.x}sxrpx~}.u~.t.~u.
|pxyp}p.p}s.wttu~t.wt.WWb.}~t.
_txx~}t .\py~.R~||t}.x}.at{px~}. wp.p}.p~xpt.r~|pp~.qpts.~}.
x}sxrpx~}.rp}}~.qt.xst}xuxts=.
~.Upr~[email protected]}s.wt.V~
t}|t} .
at~}t.
6A7.cwt.txx~}t.x}sxrpts.wp.wt.
prp{.~.t{px
t.~t}xp{.~u.pqt.~u.
[email protected]}.Twxqx.Q;.wt.txx~}t.
|pxyp}p.x.{~=.cwt.txx~}t.pt;.
r~|pts.wt.tuutr.~u.|pxyp}p.~.
Uqog!tgugctejgtu!encko!vjcv!ecppcdku!
rt}{.r~}~{{ts.rwts{t.XX.
ku!pqv!rctvkewnctn{!cffkevkxg/!Gzrgtvu!
qp}rt.p}s.|pst.ttpts.r{px|.
cuugtv!vjcv!ecppcdkuu!cffkevkxg!
pq~.wt.r~|ppx
t.tuutr=.
rqvgpvkcn!rctcnngnu!echhgkpgu/ .6Twxqx.
cwt.WWb.}~ts.wp.r~|px~}.
Q;.pvt.@H;.{x}t.A?¤A@7=.Uwt|~t;.
qttt}.|pxyp}p.p}s.rwts{t.XX.
txx~}t.pts.wp;.Ecppcdku!wug!
qp}rt.pt.sxuuxr{.qtrpt.~u.
kpfkecvgu!c!nqygt!nkmgnkjqqf!qh!cffkevkqp!
sxuutt}rt.x}.wt.prx~}.~u.sxuutt}.
cpf!cdwug!rqvgpvkcn!cu!eqorctgf!vq!
wp|pr~{~vxrp{.r{pt.~u.rwts{t.XX.
qvjgt!uwduvcpegu/ .6Twxqx.Q;.pvt.AA;.
sv.x}.wt.RbP=.cwt.WWb.}~t.wp.
{x}t.@A¤@B7=.
rwts{t.XX.qp}rt.x}r{st.
d}st.wt.RbP;.u~.p.qp}rt.~.qt.
x|{p}<{xzt.sv.6g/i/-!r~rpx}t;.
{prts.x}.rwts{t.XX;.XXX;.Xe;.~.e;.x.
p|wtp|x}t7;.~x~xs.6g/i/-!ut}p}{;.
|.wp
t.p.rt}{.prrtts.|tsxrp{.
~r~s~}t7;.sttp}.sv.6g/i/-!
t.x}.tp|t}.x}.wt.d}xts.bpt= .
t}~qpqxp{7;.sx~rxpx
t.p}twtxr.
P.STP.wp.t
x~{.pts;.R~}vt.
6g/i/-!wt}rr{xsx}t7;.p}s.}pp{{.
tpq{xwts.~}{.~}t.rwts{t;.rwts{t.
~rrx}v.{p}.r~|~}t}.6g/i/-!r~rp.
X;.u~.sv.~u.pqt.xw.}~.rt}{.
{tp
t.p}s.~.p7=.cwt.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
|trwp}x|.~u.prx~}.~u.% <cWR.p}s.
|pxyp}p;.wxrw.pr.x|px{.w~vw. d}xts.bpt= .FE.Ua.C?DDA.6A?@@7=.
cw;.p}.pt|.~.r~|pt.wt.
wt.rp}}pqx}~xs.trt~.6sxrts.
t{px
t.pqt.~t}xp{.~u.rwts{t.X.
uwt.x}.Upr~.A7.pt.r~|{tt{.
qp}rt.~.wp.~u.p.qp}rt.x}.
sxuutt}.u~|.wt.pq~
t<|t}x~}ts.
p}~wt.rwts{t.x.x}r~}tt}xp{.
r{pt.~u.rwts{t.XX.qp}rt=.cwt.
x}rt.p.rwts{t.X.qp}rt.|.
WWb.r~}r{st.wp.wt.sxuutt}rt.x}.
wt.|trwp}x|.~u.prx~}.x}.wt.
px~. t|px}.x}.rwts{t.X.}x{.x.wp.qtt}.
u~}s.~.wp
t.p.rt}{.prrtts.
r{pt.~u.rwts{t.XX.qp}rt.|pzt.
x.x}p~xpt.~.r~|pt.wt.p}vt.~u.
w~t.qp}rt.xw.|pxyp}p=.
Ugg!P|txrp}.u~.bput.Prrt;.F?E.U=Bs.p.CC?=.
XgtFcvg!Ugr>22@3125!
32<55!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11172
Hov!5812
Uhov!5813
64858!
|tsxrp{.t.x}.tp|t}.x}.wt.d}xts.
bpt=.
\~t~
t;.wt.txx~}t.upx{ts.~.
t
xt.wt.x}sxrp~.~u.pqt.~t}xp{;.
p.sxrts.x}.wt.{tvx{px
t.wx~.~u.
wt.RbP=.cwt.txx~}t.sxs.}~.t.
spp.~}.|pxyp}p.pvt;.sx
tx~};.
rw~prx
t.~txt;.p}s.
stt}st}rt.x}.wtx.t
p{px~}.~u.
|pxyp}p.pqt.~t}xp{=.cwt.WWb.p}s.
wt.STP.sxr.w~t.x}sxrp~.pq~
t.
x}.wx.upr~=.WWb .t
p{px~}.~u.wt.
u{{.p}vt.~u.spp.{ts.WWb.p}s.STP.~.
r~}r{st.wp.|pxyp}p.wp.p.wxvw.
~t}xp{.u~.pqt=.
cwt.txx~}t;.qpts.~}.wtx.t
xt.
~u.p.
t.q.V~t.p}s.Tp{tx}t.
6A??F7;.r~}r{sts.wp.|pxyp}p.wp.p.
{~.pqt.~t}xp{=.V~t.p}s.
Tp{tx}t.
tts.FCE.|t}p{.wtp{w.
~utx~}p{.p}s.pzts.wt|.~.pt.wt.
pssxrx
t}t.6qpts.~}.p.t
t}<~x}.
rp{t7.~u.t
tp{.sv.6wt~x};.}xr~x}t;.
r~rpx}t>rprz;.~r~s~}t;.
|twp|wtp|x}t;.p|wtp|x}t;.
rpuutx}t;.p{r~w~{;.p}s.|pxyp}p7=.cwt.
txx~}t.pts.wp.wt.wtp{w.
~utx~}p{.pts.|pxyp}p.p.{tp.
pssxrx
t.~u.wt.sv.
tts=.cwt.
STP.}~t.wp.wt.
t.rxts.q.wt.
txx~}t.x.qpts.~}.qytrx
t.
~x}x~}.u~|.wtp{w.~utx~}p{=.
6B7.cwt.txx~}t.|t}x~}ts.wp.
|p}.~u.wt.rp}}pqx}~xs.x}.|pxyp}p.
strtpt.wt.rw~prx
t.tuutr.~u.% <.
cWR;.p}s.wttu~t.|pxyp}p.{prz.
uuxrxt}.pqt.~t}xp{.u~.{prt|t}.
x}~.rwts{t.X=.Uwt;.wt.txx~}t.
|t}x~}ts.~}.pvt.C.x}.Twxqx.Q.6{x}t.
@@¤@D7;.Yjkng!vjg!FGC!eqpukfgtu!
ecppcdku!c!uejgfwng!K!ftwi-!kv!encuukhkgu!
ftqpcdkpqn!)Octkpqn*!cu!uejgfwng!KKK/!
Ftqpcdkpqn!ku!211!rgtegpv!VJE!cpf!ku!
G<^HT^HO^23CWR3/UIO
23CWR3
GR23CW27/134>0IRJ@
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64859!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
rqvgpvkcnn{!xgt{!ru{ejqcevkxg/!Pcvwtcn!
ecppcdku!v{rkecnn{!yqwnf!dg!pq!oqtg!
vjcp!26!rgtegpv!VJE!d{!ygkijv/!Vjwu!kv!
ku!kpeqpukuvgpv!vjcv!ecppcdku-!ykvj!26!
rgtegpv!ygkijv!VJE-!tgockpu!c!
]u_ejgfwng!K!ftwi-!yjkng!ftqpcdkpqn-!cv!
211!rgtegpv!VJE-!ku!uejgfwng!KKK/ .
cwt.WWb.psstts.wx.xt.q.
x}sxrpx}v.wp.wt.|~s{px}v.tuutr.~u.
wt.~wt.rp}}pqx}~xs.x}.|pxyp}p.~}.
% <cWR.wp
t.}~.qtt}.st|~}pts.x}.
r~}~{{ts.sxt=.cwt.WWb.p}s.wt.
STP.p{~.}~t.wp.wt.stt|x}px~}.~u.
wt.pqt.~t}xp{.~u.p.qp}rt.
r~}xst.}~.~}{.rw~prx
t.tuutr.
q.p{~.rwt|x;.wp|pr~{~v;.
wp|pr~zx}txr;.pvt.pt};.p}s.
sx
tx~}.wx~.p|~}v.~wt.|tpt=.
\px}~{.6s~}pqx}~{.x}.tp|t.~x{7.
p.trwts{ts.u~|.rwts{t.XX.~.
rwts{t.XXX.~}.Y{.A;[email protected].
BDHAG;.STP.@HHH7=.X}.ptx}v.\px}~{;.
WWb.r~|pts.\px}~{.~.|pxyp}p.~}.
t
tp{.ptr.~u.pqt.~t}xp{.p}s.
u~}s.wp.|py~.sxuutt}rt.qttt}.
wt.~;.rw.p.u~|{px~};.
p
px{pqx{x;.p}s.pvt;.r~}xqt.~.
sxuutt}rt.x}.pqt.~t}xp{=.cwt.
rw~prx
t.tuutr.u~|.|~zx}v.pt.
vt}tp{{.|~t.pxs.p}s.x}t}t.wp}.
w~t.wp.~rr.w~vw.~p{.
ps|x}xpx~}.6WWb;[email protected]~}.p}s.
fpwq};.@HH?J.W~{{xt.p}s.
Vx{{txt;.@HFB7=.cwttu~t;.p.
r~}r{sts.q.q~w.wt.WWb.p}s.wt.
STP;.wt.st{pts.~}t.~u.prx~}.p}s.
{~}vt.spx~}.~u.prx~}.u~|.p}.~p{.
s~t.~u.\px}~{.|p.r~}xqt.x}.
{x|xx}v.wt.pqt.~t}xp{.~u.\px}~{.
t{px
t.~.|pxyp}p;.wxrw.x.|~.
~ut}.|~zts=.cwt.WWb.p{~.pts.wp.
wt.tprx~}.p}s.xuxrpx~}.~u.
s~}pqx}~{.u~|.wt.t}rp{pts.
tp|t.~x{.|xt.~u.\px}~{.x.wxvw{.
r~|{t.p}s.sxuuxr{;.p}s.wp.wt.
tt}rt.~u.tp|t.~x{.|xt.|p.
tr{st.wt.|~zx}v.~u.\px}~{<{prts.
rxvptt=.
Uwt|~t;.|pxyp}p.p}s.\px}~{.
w~.xv}xuxrp}.sxuutt}rt.x}.prp{.
pqt.p}s.x{{xrx.puuxrzx}v=.cwtt.wp
t.
qtt}.}~.t~.~u.pqt;.sx
tx~};.~.
q{xr.wtp{w.xz.st.~.\px}~{=.X}.
r~}p;.AA=A.|x{{x~}.P|txrp}.ps{.
t~.rt}{.x}v.|pxyp}p.
6bP\WbP;.A?@Dp7=.cwt.STP.sppqpt;.
]U[Xb;.w~ts.wp.|pxyp}p.p.wt.
|~.utt}{.xst}xuxts.sv.x}.pt.
p}s.{~rp{.u~t}xr.{pq~p~xt.u~|.
Yp}p.A??@.~.Strt|qt[email protected]}s.
x}sxrpt.wt.wxvw.p
px{pqx{x.~u.
|pxyp}p=.cwt.sxuutt}rt.x}.
r~|~xx~};.prp{.pqt;.p}s.
sx
tx~}.r~}xqt.~.wt.sxuutt}rt.
x}.rwts{x}v.qttt}.|pxyp}p.p}s.
\p•x}~{=.
Pssxx~}p{{;.wt.USP.p~
ts.p.
]t.Sv.P{xrpx~}.6]SP7.u~.
\px}~{;.x}sxrpx}v.p.{tvxx|pt.|tsxrp{.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
t.u~.\px}~{.x}.wt.d}xts.bpt.p}s.
p{{~x}v.u~.\px}~{.~.qt.trwts{ts.
x}~.rwts{t.XX.p}s.qtt}{.x}~.
rwts{t.XXX.~u.wt.RbP=.cwt.WWb.
|t}x~}ts.wp.|pxyp}p.p}s.\px}~{.
sxuut.~}.p.xst.
pxt.~u.upr~.p}s.
wtt.sxuutt}rt.pt.|py~.tp~}.u~.
sxuutt}xp{.rwts{x}v.~u.|pxyp}p.p}s.
\px}~{=.\pxyp}p;.p.sxrts.|~t.
u{{.x}.Upr~.B.p}s.E;.s~t.}~.wp
t.
p.rt}{.prrtts.|tsxrp{.t.x}.wt.
d}xts.bpt;.x.wxvw{.pqts;.p}s.wp.
p.{prz.~u.prrtts.put=.
Hcevqt!3<!Uekgpvkhke!Gxkfgpeg!qh!vjg!
Ftwi!Rjctoeqnqikecn!Ghhgevu-!kh!Mpqyp!
cwt.WWb.pts.wp.wtt.pt.{pvt.
p|~}.~u.rxt}xuxr.spp.~}.wt.
}t~rwt|x;.|trwp}xxr.tuutr;.
~xr~{~v;.p}s.wp|pr~{~v.~u.
|pxyp}p=.P.rxt}xuxr.t
p{px~};.p.
r~}srts.q.wt.WWb.p}s.wt.STP;.~u.
|pxyp}p .}t~rwt|x;.w|p}.p}s.
p}x|p{.qtwp
x~p{.wp|pr~{~v;.
rt}p{.}t
~.t|.tuutr;.p}s.
~wt.wp|pr~{~vxrp{.tuutr.6g/i/-!
rpsx~
pr{p;.x||}~{~vxrp{.tuutr7.
x.tt}ts.qt{~=.
Pgwtqejgokuvt{!
\pxyp}p.r~}px}.}|t~.
r~}xt}.rw.p.rp}}pqx}~xs.wp.
wp
t.p.
pxt.~u.wp|pr~{~vxrp{.
prx~}=.cwt.txx~}.stux}ts.|pxyp}p.
p.x}r{sx}v.p{{.rp}}pqx.r{x
pts.
px}=.cwt.WWb.pts.wp.sxuutt}.
|pxyp}p.p|{t.stx
ts.u~|.
px~.
r{x
pts.px}.|p.sxuut.x}.wtx.
rwt|xrp{.r~}xt}.x}r{sx}v.% <cWR.
p}s.~wt.rp}}pqx}~xs=.cwttu~t.
|pxyp}p.~sr.u~|.sxuutt}.
px}.x{{.wp
t.sxuutt}.qx~{~vxrp{.p}s.
wp|pr~{~vxrp{.tuutr=.cwt.rwt|xrp{.
r~}xt}.~u.|pxyp}p.pt.sxrts.
uwt.x}.Upr~.B=.
cwt.x|p.xt.~u.prx~}.u~.
rp}}pqx}~xs.rw.p.% <cWR.x.p.wt.
rp}}pqx}~xs.trt~=.c~.rp}}pqx}~xs.
trt~;[email protected]}s.RQA;.wp
t.qtt}.
xst}xuxts.p}s.rwpprtxts.6Qpxp.t.
p{=;.A?@AJ._x~|t{{x;.A??D7.p}s.p•t.V<.
~tx}<r~{ts.trt~=.Prx
px~}.~u.
wtt.x}wxqx~.V<~tx}<r~{ts.
trt~.x}wxqx.pst}{pt.rr{pt.
prx
x;.wxrw.t
t}.r~}
tx~}.~u.
Pc_.~.rr{xr.P\_=.Rp}}pqx}~xs.
trt~.prx
px~}.p{~.t{.x}.
x}wxqxx~}.~u.]<.p}s._>`<t.rp{rx|.
rwp}}t{.p}s.prx
pt.x}ps{.
trxux}v.~px|.rwp}}t{.6\przxt.
t.p{=;[email protected]xrwt{{.t.p{=;.@HHF7=.cwt.
WWb.|t}x~}ts.wp.x}wxqxx~}.~u.]<.
t.rp{rx|.rwp}}t{.strtpt.
}t~p}|xt.t{tpt.p}s.wx.|p.
qt.wt.}st{x}v.|trwp}x|.x}.wt.
pqx{x.~u.rp}}pqx}~xs.~.x}wxqx.
prt{rw~{x}t;.}~tx}twx}t.p}s.
v{p|pt.u~|.trxuxr.ptp.~u.wt.
qpx}=.cwtt.rt{{{p.prx~}.|p.
RQ!11111
Hto!11173
Hov!5812
Uhov!5813
}st{xt.wt.p}x}~rxrtx
t.p}s.
rw~prx
t.tuutr.~u.rp}}pqx}~xs=.
% <cWR.pr.p.p}.pv~}x.p.
rp}}pqx}~xs.trt~=.
RQ@.trt~.pt.x|px{.u~}s.x}.
wt.rt}p{.}t
~.t|.p}s.pt.
{~rpts.|px}{.x}.wt.qpp{.vp}v{xp;.
wx~rp|.p}s.rttqt{{|.~u.wt.
qpx}.6W~{t.t.p{=;.A??C7=.RQ@.
trt~.pt.p{~.{~rpts.x}.txwtp{.
xt.rw.p.wt.x||}t.t|.6St.
_t~rt{{x.p}s.Sx.\p~;.A??H7;.q.wt.
r~}rt}px~}.~u.RQ@.trt~.wtt.x.
r~}xstpq{.{~t.wp}.x}.wt.rt}p{.
}t
~.t|.6Wtzt}wp|.t.p{=;.@HH?J.
@HHA7=.RQA.trt~.pt.u~}s.
x|px{.x}.wt.x||}t.t|.p}s.
ts~|x}p}{.x}.Q.{|w~rt.p}s.
}pp{.zx{{t.rt{{.6Q~pq~{p.t.p{=;.
@HHB7=.RQA.trt~.pt.p{~.u~}s.x}.
wt.rt}p{.}t
~.t|;.x|px{.x}.
wt.rttqt{{|.p}s.wx~rp|.6V~}v.
t.p{=;.A??E7=.
c~.t}s~vt}~.{xvp}s.~.wt.
rp}}pqx}~xs.trt~;.p}p}sp|xst.p}s.
pprwxs~}{.v{rt~{.6A¤PV7;.tt.
xst}xuxts.x}[email protected]
p}t.t.p{=;.@HHA7.
p}[email protected]\trw~{p|.t.p{=;.@HHD7;.
ttrx
t{=.P}p}sp|xst.x.p.{~<.
tuuxrpr.pv~}x.6Qxt
~vt{.p}s.Rwx{st;.
A???7.p}s.A¤PV.x.p.wxvw.tuuxrpr.
pv~}x.6V~}x~tz.t.p{=;.A???7.~.wt.
rp}}pqx}~xs.trt~=.cwtt.
t}s~vt}~.{xvp}s.pt.tt}.x}.q~w.
wt.rt}p{.}t
~.t|.p}s.x}.wt.
txwt.6WWb;.A?@D7=.
% <cWR.p}s.rp}}pqxsx~{.6RQS7.p•t.
~.~u.wt.|py~.rp}}pqx}~xs.x}.
|pxyp}p=.% <cWR.x.wt.|py~.
rw~prx
t.rp}}pqx}~xs.6fprwt{.t.
p{=;.A??A7=.% <cWR.wp.x|x{p.puux}x.
u~[email protected]}s.RQA.trt~.p}s.pr.p.
p.tpz.pv~}x.p.RQA.trt~=.cwt.
WWb.x}sxrpts.wp.prx
px~}.~u.RQ@.
trt~.|tsxpt.rw~~xr.tuutr.
~u.rp}}pqx}~xs=.RQS.wp.{~.puux}x.
u~.q~[email protected]}s.RQA.trt~=.RQS.
wp.p}pv~}xxr.tuutr.p.RQ@.trt~;.
p}s.~|t.x}
tt.pv~}xxr.~txt.p.
RQA.trt~=.
Cpkocn!Dgjcxkqtcn!Ghhgevu!
P}x|p{.pqt.~t}xp{.sxt.6sv.
sxrx|x}px~};.t{u<ps|x}xpx~};.
r~}sxx~}ts.{prt.tutt}rt7.pt.
sxrts.|~t.u{{.x}.Upr~.@=.
Qxtu{;.x.p.r~}xt}{.
st|~}pts.wp.% <cWR;.wt.x|p.
rw~prx
t.r~|~}t}.x}.|pxyp}p;.
p}s.~wt.rp}}pqx}~xs.x}.|pxyp}p.
wp
t.p.sxx}r.sv.sxrx|x}px
t.
~ux{t=.X}.pssxx~};.p}x|p{.t{u<.
ps|x}xt.% <cWR;.p}s.% <cWR.x}.{~.
s~t.~srt.r~}sxx~}ts.{prt.
••tut•t}rt=.
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Egpvtcn!Pgtxqwu!U{uvgo!Ghhgevu!
_rw~prx
t.Tuutr.
cwt.r{x}xrp{.rw~prx
t.tuutr.~u.
|pxyp}p.pt.sxrts.|~t.u{{.x}.
Upr~.@=.Qxtu{;.wt.rw~prx
t.
tuutr.u~|.|pxyp}p.t.pt.
r~}xstts.{tppq{t.p}s.p~rxpts.
xw.sv<ttzx}v.~.sv<pzx}v.6WWb;.
A?@DJ.\p{s~}ps~;.A??A7=.Uwt;.x.p.
}~ts.q.WWb.wp.|pxyp}p.t.
tut.wxvwt.r~}rt}px~}.~u.wt.
x}rxp{.rw~prx
t.r~|~}t}.6% <.
cWR7.~
t.{~t.r~}rt}px~}.6WWb;.
A?@D7=.
bsxt.wp
t.t
p{pts.rw~prx
t.
tuutr.~u.cWR.x}.wt.tt}rt.~u.wxvw.
RQS;.RQR;.~.RQ].px~=.T
t}.w~vw.
~|t.sxt.vvt.wp.RQS.|p.
strtpt.~|t.~u.% <cWR .rw~prx
t.
tuutr;.wt.WWb.u~}s.wp.wt.px~.~u.
RQS.~.% <cWR.ps|x}xtts.x}.wt.
sxt.tt.}~.r~|ppq{t.~.wt.
p|~}.u~}s.x}.|pxyp}p.ts.q.
|~.t~{t.6Sp{~}.t.p{=;[email protected]}x~{.
t.p{=;[email protected]psx.t.p{=;.@HGA7=.X}.upr;.
wt.RQS.px~.x}.wtt.sxt.pt.
xv}xuxrp}{.wxvwt.wp}.wt.RQS.u~}s.
x}.|~.|pxyp}p.rt}{.u~}s.~}.
wt.tt.6\tw|tsxr.t.p{=;.A?@?7=.WWb.
x}sxrpts.wp.|~.~u.wt.|pxyp}p.
p
px{pq{t.~}.wt.tt.wp.p.wxvw.cWR.
p}s.{~.RQS.r~}t}.p}s.wttu~t.p}.
{tt}x}v.~u.cWR .rw~prx
t.tuutr.
q.RQS.x{{.}~.~rr.u~.|~.
|pxyp}p.t.6WWb;.A?@D7=.Sp{~}.t.
p{=.6@HFE7.t~ts.wp.wt}.
~{}tt.
|~zts.rxvptt.xw.p.px~.~u.F.RQS.
~.@.% <cWR.6?=@D.|v>zv.RQS.p}s.?=?AD.
|v>zv.% <cWR7;.wtt.p.p.xv}xuxrp}.
strtpt.x}.px}v.~u.prt.qytrx
t.
tuutr.p}s.prwxt
x}v.p.wxvw .x}.
r~|px~}.~.|~zx}v.% <cWR.p{~}t=.
X}.~p{.ps|x}xpx~}.sxt;.wt.
qytrx
t.tuutr.p}s.p}xt.~srts.
q.r~|qx}px~}.~u.RQS.p}s.cWR.x}.p.
px~.~u.p.{tp[email protected].~.% <cWR.6@D;.
B?;.E?.|v.RQS.~.B?.|v.% <cWRJ.
Zp}x~{.t.p{=;.@HFC7.~.p.p[email protected].
~.% <cWR.6@.|v>zv.RQS.~.?=D.|v>zv.
% <cWRJ.ipsx.t.p{=;[email protected]t.{t.
wp}.w~t.~srts.q.% <cWR.
ps|x}xtts.p{~}t=.
X}.~}t.s.6X{p}.t.p{=;.A??D7;.wt.
pw~.rp{r{pts.wt.}pp{{.
~rrx}v.r~}rt}px~}.~u.RQR.p}s.
RQS.x}.|pxyp}p.rxvptt.xw.txwt.
@=G.~•.B=E3.% <cWR.q.txvw=.cwt.
pw~.
pxts.wt.r~}rt}px~}.~u.
RQR.p}s.RQS.u~.tprw.r~}rt}px~}.~u.
% <cWR.x}.wt.|pxyp}p.rxvptt=.
Ps|x}xpx~}.x}.wtp{w.|pxyp}p.
t.6}LAB7.r~}xts.~u.txwtI.6@7.[~.
[email protected].txvw7.p}s.{~.RQS.
6?=A3.q.txvw7J.6A7.wxvw.RQR.6?=D3.q.
txvw7.p}s.{~.RQSJ.6B7.{~.RQR.p}s.
wxvw.RQS.6@=?3.q.txvw7J.~.6C7.wxvw.
RQR.p}s.wxvw.RQS.p}s.wt.t.tt.
sx
xsts.x}~.{~.% <cWR.6@=G3.q.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
txvw7.p}s.wxvw.% <cWR.6B=E3.q.
txvw7.v~=.bqytrx
t.rw~prx
t.
tuutr.tt.xv}xuxrp}{.vtpt.u~.p{{.
v~.x}.r~|px~}.~.{prtq~.p}s.
wtt.tt.}~.xv}xuxrp}.sxuutt}rt.x}.
tuutr.p|~}v.wt.tp|t}.6X{p}.t.p{=;.
A??D7=.
cwt.WWb.p{~.tutts.~.p.s.xw.
% <cWR.p}s.rp}}pqx}~{.6RQ]7.6Zp•}x~{.
t.p{=;.@HFD7=.X}.wx.s;.~p{.
ps|x}xpx~}.~u.txwt.@A=D;.AD;.~.D?.
|v.RQ].r~|qx}ts.xw.AD.|v.% <cWR.
6px~.~u.p.{tp[email protected]].~.% <cWR7.
xv}xuxrp}{.x}rtpts.qytrx
t.
rw~prx
t.px}v.~u.% <cWR.
r~|pts.~.% <cWR.p{~}t.6Zp}x~{.t.
p{=;.@HFD7=.
Qtwp
x~p{.X|px|t}.
bt
tp{.upr~.|p.x}u{t}rt.
|pxyp}p .qtwp
x~p{.tuutr.x}r{sx}v.
wt.spx~}.6rw~}xr.~.w~.t|7;.
utt}r.6spx{;.ttz{;.~.
~rrpx~}p{{7;.p}s.p|~}.~u.t.6wtp
.
~.|~stpt7=.attprwt.wp
t.
tp|x}ts.w~.{~}v.qtwp
x~p{.
x|px|t}.tx.u~{{~x}v.rw~}xr.
|pxyp}p.t=.cwtt.sxt.ts.t{u<.
t~ts.wx~xt.~u.t~t.spx~};.
utt}r;.p}s.p|~}.~u.|pxyp}p.
t;.p}s.ps|x}xtts.t
tp{.
tu~|p}rt.p}s.r~v}xx
t.t.p.
sxuutt}.x|t.~x}.u~{{~x}v.
|pxyp}p.pqx}t}rt=.Prr~sx}v.~.
WWb;.qtwp
x~p{.x|px|t}.|p.
tx.u~..~.AG.sp.~u.pqx}t}rt.
x}.rw~}xr.|pxyp}p.t=.
_rw~prx
t.tuutr.~u.|pxyp}p.rp}.
{tps.~.qtwp
x~p{.x|px|t}.x}r{sx}v.
r~v}xx
t.strt|t}.p}s.strtpts.
pqx{x.~.~tpt.|~~.
twxr{t.6WWb;.
A?@D7=.Q{~rz.t.p{[email protected]
p{pts.
r~v}xx
t.|tpt.x}.CG.wtp{w.|p{t.
qytr.u~{{~x}v.|~zx}v.p.|pxyp}p.
rxvptt.wp.r~}px}ts.A=DF3.~.@H.|v.
% <cWR.q.txvw.~.{prtq~=.Tprw.
qytr.pxrxpts.x}.txvw.tx~}.
6u~.tx~}.xw.|pxyp}pJ.u~.
tx~}.xw.{prtq~7.p}s.t
tp{.
r~v}xx
t.p}s.rw~|~~.t.tt.
ps|x}xtts.6g/i/!
tqp{.trp{{;.uprxp{.
tr~v}xx~};.t.{tp}x}v;.tprx~}.
x|t7=.\pxyp}p.xv}xuxrp}{.x|pxts.
tu~|p}rt.x}.|~.~u.wtt.r~v}xx
t.
p}s.rw~|~~.t.6Q{~rz.t.p{=;.
@HHA7=.
ap|ptzt.t.p{=.6A??E7.t~ts.wp.
x}.A?.trtpx~}p{.t.~u.|pxyp}p;.
prt.ps|x}xpx~}.~u.AD?.Iv>zv.p}s.
D??.Iv>zv.% <cWR.x}.|~zts.|pxyp}p.
t{ts.x}.s~t<stt}st}.x|px|t}.
x}.r~v}xx~};.|~~.x|{x
x;.|~~.
r~}~{.6przx}v.x|px|t}7;.p}s.xz.
pzx}v=.X}.p}~wt.s.6Zwp{t.t.
p{=;.@HHH7;.wt}.AH?.Iv>zv.% <cWR.p.
ps|x}xtts.
xp.p.|~zts.|pxyp}p.
rxvptt.x}.B?.wtp{w.
~{}tt.xw.
}~.wx~.~u.qp}rt.pqt.wtt.tt.
xv}xuxrp}.x|px|t}.~u.|~~.tts.
RQ!11111
Hto!11174
Hov!5812
Uhov!5813
6485;!
p}s.prrpr=.Uwt|~t;.
ps|x}xpx~}.~u.B=HD3.% <cWR.x}.p.
|~zts.|pxyp}p.rxvptt.x}rtpts.
wt.{pt}r.x}.p.pz.~u.x|{pts.
qpzx}v.x}.p.
twxr{t.6[xv~x.t.p{=;.
@HHG7=.cwt.WWb.}~ts.wp.wt.|~~.
x|px|t}.t~ts.x}.wtt.sxt.
6Zwp{t.t.p{=;.@HHHJ.[xv~x.t.p{=;.
@HHG7.pt.rxxrp{.zx{{.}ttsts.u~.
~tpx}v.p.
twxr{t=.
P.|t}x~}ts.x}.wt.WWb.s~r|t};.
~|t.sxt.tp|x}ts.wt.txt}rt.
~u.wt.qtwp
x~p{.x|px|t}.
x||tsxpt{.put.|pxyp}p.
ps|x}xpx~}=.b~|t.~u.|pxyp}p .
prt.tuutr.|p.x{{.qt.tt}.u~.p.
{tp.AC.w~.put.wt.prt.
rw~prx
t.tuutr.wp
t.qxsts=.X}.p.
qxtu.r~||}xrpx~};.Wtxw|p}}.t.p{=.
6@HH?7.t~ts.wp.wtt.tt.
r~v}xx
t.x|px|t}.6sxvx.trp{{.p}s.
pxw|txr.pz7.x}.~.~.~u.wtt.
ttxt}rts.|pxyp}p.|~zt.u~.AC.
w~.put.|~zx}v.|pxyp}p.rxvptt.
r~}px}x}v.A=DF3.% <cWR=.W~t
t;.
Up}.t.p{[email protected]
p{pts.qytrx
t.
tuutr.p}s.tu~|p}rt.|tpt.u~..
~.AD.w~.x}.@?.wtp{w.|p{t.put.
t~t.~.txwt.@=G3.~.B=E3.% <cWR.
x}.|pxyp}p.rxvptt=._tpz.
strt|t}.x}.qytrx
t.p}s.
tu~|p}rt.|tpt.tt.}~ts.
xwx}.A.w~.~u.|pxyp}p.t~t.
q.wtt.tt.|x}x|p{.txsp{.
p{tpx~}.x}.qytrx
t.~.tu~|p}rt.
|tpt.p.AB¤AD.w~.put.t~t=.
_txt}rt.~u.qtwp
x~p{.x|px|t}.
u~{{~x}v.ttpts.p}s.rw~}xr.t.~u.
|pxyp}p.wp.p{~.qtt}.x}
txvpts.
p}s.p.t
xtts.x}.wt.WWb.s~r|t}.
6WWb;.A?@D7=.X}.pxr{p;.ttprwt.
tp|x}ts.w~.{~}v.qtwp
x~p{.
x|px|t}.{p.u~{{~x}v.rw~}xr.
|pxyp}p.t=.X}.sxt.tp|x}x}v.
txt}rt.~u.tuutr.x}.rw~}xr.p}s.
wtp
.|pxyp}p.t;.wtt.tt.
xv}xuxrp}.strt|t}.x}.r~v}xx
t.p}s.
|~~.u}rx~}.pz.x}.p{{.sxt.~u..
~.AF.sp;.p}s.x}.|~.sxt.p.AG.
sp.6b~{~xy.t.p{=;.A??AJ.\tx}x.t.
p{=;.A??EJ.[xspw{.p}s._xrt;.A?@AJ._~t.
t.p{=;.A??AJ.Q~{{p.t.p{=;.A??AJ.Q~{{p.t.
p{=;.A??D7=.X}.sxt.wp.u~{{~ts.wtp
.
|pxyp}p.t.u~.{~}vt.wp}.AG.sp.
p}s..~.A?.tp.~u.|pxyp}p.
pqx}t}rt;.r~v}xx
t.p}s.rw~|~~.
x|px|t}.tt.}~.{~}vt.sttrts.
6Uxts.t.p{=;.A??DJ.[~}.t.p{=;.A??CJ.
cpx.t.p{=;.A?@@7=.U~.tp|{t;.Uxts.t.
p{=.6A??D7.t~ts.wp.put.B.|~}w.
~u.pqx}t}rt.u~|.|pxyp}p;.p}.
stuxrx.x}.x}t{{xvt}rt.6X`7;.|t|~;.
p}s.~rtx}v.tts.u~{{~x}v.wtp
.
|pxyp}p.t.tt.}~.{~}vt.~qt
ts.
6Uxts.t.p{=;.A??D7=.X}.p.|tp<p}p{x.
wp.tp|x}ts.}~}<prt.p}s.{~}v<.
{px}v.tuutr.~u.|pxyp}p;.p}.stuxrx.
x}.}t~r~v}xx
t.tu~|p}rt.wp.
tt.~qt
ts.xwx}.wt.ux.|~}w.
G<^HT^HO^23CWR3/UIO
23CWR3
64861!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
tt.}~.{~}vt.ppt}.put.
p~x|pt{.~}t.|~}w.~u.pqx}t}rt.
6brwtx}t.p}s.S}};.A?@A7=.WWb.
uwt.}~t.wp.x}.|~stpt.|pxyp}p.
t.stuxrx.x}.strxx~}<|pzx}v.zx{{.
tt.}~.~qt
ts.put.AD.sp.~u.
pqx}t}rt.p}s.pssxx~}p{{.X`;.
x||tsxpt.|t|~.p}s.st{pts.
|t|~.zx{{.tt.}~.xv}xuxrp}{.
x|prts.p.~qt
ts.xw.wtp
.p}s.
rw~}xr.|pxyp}p.t.6Uxts.t.p{=;.
A??DJ.WWb;.A?@D7=.
P.|t}x~}ts.x}.wt.WWb.s~r|t}.
6WWb;.A?@D7;.wt.x}t}x.p}s.
txt}rt.~u.}t~{~vxrp{.x|px|t}.
u~|.rw~}xr.|pxyp}p.t.p{~.|p.qt.
stt}st}.~}.wt.pvt.~u.ux.t=.X}.~.
tppt.|p{{t.rp{t.sxt.6{t.wp}.
@??.pxrxp}.t.t~t.v~7;.
U~}t.t.p{=.6A?@@7.p}s.Vqt.t.p{=.
[email protected]~|pts.}t~{~vxrp{.u}rx~}.
x}.tp{.~}t.6rw~}xr.|pxyp}p.t.
x~.~.pvt.@D.~[email protected]}s.{pt.~}t.
6rw~}xr.|pxyp}p.t.put.pvt.@D.~.
@E7.wtp
.|pxyp}p.t.p}s.u~}s.
wp.wtt.tt.xv}xuxrp}.stuxrx.x}.
ttrx
t.}t~{~vxrp{.u}rx~}.x}.tp{.
~}t.t.wxrw.tt.}~.~qt
ts.~.
tt.{t.ppt}.x}.{pt.~}t.t=.
X}.p.~trx
t.{~}vxsx}p{.qxw.
r~w~.s.u~{{~x}v.@;?BF.
x}sx
xsp{.6\txt.t.p{=;.A?@A7;.p.
xv}xuxrp}.strtpt.x}.X`.p}s.
}t~rw~{~vxrp{.tu~|p}rt.p.
~qt
ts.x}.ps~{trt}<~}t.t.p}s.
txts.t
t}.put.pqx}t}rt.u~|.
|pxyp}p.u~.p.{tp.~}t.tp=.
W~t
t;.\txt.t.p{=.6A?@A7.t~ts.x}.
wtt.p.}~.xv}xuxrp}.rwp}vt.x}.X`.x}.
ps{<~}t.t=.
cwt.WWb.}~ts.wp.wtt.x.~|t.
t
xst}rt.wp.wt.t
tx.~u.wt.
txt}.}t~{~vxrp{.x|px|t}.
|p.p{~.qt.st.x}.p.~.wt.p|~}.
~u.|pxyp}p.pvt=.X}.wt.s.
|t}x~}ts.pq~
t;.Vqt.t.p{=.6A?@A7.
u~}s.wp.wt.tp{.~}t.t.
r~}|ts.wtt.x|t.p.|rw.
|pxyp}p.t.ttz.p}s.ts.x.xrt.p.
~ut}.p.{pt.~}t.t=.\txt.t.p{=.
6A?@A7.t~ts.x}.wtx.s;.
|t}x~}ts.pq~
t;.wp.wtt.p.p.
r~t{px~}.qttt}.X`.stuxrx.x}.
ps~{trt}.~}t.t.p}s.wt.x}rtpts.
p|~}.~u.|pxyp}p.ts=.
Qtwp
x~p{.Tuutr.~u._t}pp{.T~t.
X}.sxt.wp.tp|x}ts.tuutr.~u.
t}pp{.|pxyp}p.t~t;.|p}.~u.
wt.tv}p}.~|t}.p{~.ts.p{r~w~{.
p}s.~qprr~.x}.pssxx~}.~.|pxyp}p=.
T
t}.w~vw.~wt.sv.tt.ts.x}.
r~}y}rx~}.xw.|pxyp}p;.wtt.x.
t
xst}rt.~u.p}.p~rxpx~}.qttt}.
wtp
.t}pp{.|pxyp}p.t~t.p}s.
stuxrx.x}.~|t.r~v}xx
t.u}rx~}=.
cwtt.wp
t.qtt}.~.~trx
t.
{~}vxsx}p{.qxw.r~w~.sxt.
u~{{~x}v.x}sx
xsp{.t}pp{{.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
R{pxuxrpx~}.~u.Sxtpt.6XRS¤@?7.p}s.
~{s.}~.qt.p~xpt.u~.t.x}.
t
p{px}v.wt.p~rxpx~}.qttt}.
|pxyp}p.t.p}s.rw~x=.U~.
tp|{t;.ttprwt.rwpprtxts.
rw~x.p.rwx~wt}xr.r{t .
6\pt||p}x.t.p{=;.A??C7;.qr{x}xrp{.
rw~xr.|~| .6
p}.Vpt{.t.p{=;.
A?@A7;.t<rw~xr.r{x}xrp{.wxvw.xz .
6
p}.st.\tt.t.p{=;.A?@A7;.p}s.
|~|.t{pts.~.rw~x.
{}tpqx{x .6Vxuuxw<[t}stx}v.t.p{=;.
A?@A7=.
cwt.WWb.sxrts.p}.tp{.
txst|x~{~vxrp{.s.r~}srts.q.
P}stp~}.t.p{=.6@HGF7;.wxrw.
tp|x}ts.wt.{x}z.qttt}.rw~x.
p}s.|pxyp}p.t=.X}.wx.s;.pq~.
CD;???.@G<.p}s.@H<tp<~{s.|p{t.
btsxw.qytr.~
xsts.stpx{ts.
x}u~|px~}.~}.wtx.sv<pzx}v.wx~.
p}s.AFC.~u.wtt.qytr.tt.
sxpv}~ts.xw.rwx~wt}xp.~
t.p.@C<.
tp.tx~s.6@HEH¤@HGB7=.^.~u.wt.AFC.
qytr.sxpv}~ts.xw.rw~x;.A@.
x}sx
xsp{.6F=F37.wps.ts.|pxyp}p.
|~t.wp}.D?.x|t;.wx{t.@HF.
x}sx
xsp{.6FA37.}t
t.ts.
|pxyp}p=.P.tt}ts.q.wt.pw~.
6P}stp~}.t.p{=;.@HGF7;.x}sx
xsp{.
w~.r{px|ts.~.pzt.|pxyp}p.~}.|~t.
wp}.D?.~rrpx~}.tt.E.x|t.|~t.
P~rxpx~}.~u.\pxyp}p.dt.fxw.
{xzt{.~.qt.sxpv}~ts.xw.
_rw~x.
rwx~wt}xp.wp}.w~t.w~.wps.}t
t.
cwtt.wp.qtt}.tt}x
t.ttprw.~.
r~}|ts.wt.sv=.cwt.pw~.
stt|x}t.wtwt.|pxyp}p.pvt.x.
r~}r{sts.wp.|pxyp}p.t.w~.pt.
p~rxpts.xw.st
t{~|t}.~u.
{}tpq{t.~.st
t{~x}v.rw~t.pt.
rwx~wt}xp.~.~wt.rw~t;.p}s.
p.wt.vtpt.xz.u~.rwx~wt}xp=.X}.
wt.WWb.x}sxrpts.wp.wt.p
px{pq{t.
p.BD.tp.u~{{~..~.wt.qytr.
spp.s~.}~.vvt.p.rppx
t.{x}z.
t
p{pts.x}.P}stp~}.t.p{=.6@HGF7;.
qttt}.|pxyp}p.p}s.wt.
\p}xt<Vprxp.t.p{=.6A?@A7.t~ts.
st
t{~|t}.~u.rw~x.6WWb;.A?@DJ.
x|x{p.ux}sx}v=.X}.wt.u~{{~..s;.
\x}~x.t.p{=;.A?@?7=.P.|t}x~}ts.x}.
wt.WWb.t
xt.6WWb;.A?@D7;.}|t~. BDC.x}sx
xsp{.st
t{~ts.
rwx~wt}xp=.^u.w~t;.BA.x}sx
xsp{.
{pvt.rp{t.{~}vxsx}p{.sxt.
6H37.wps.ts.|pxyp}p.|~t.wp}.D?.
st|~}pts.wp.qytr.w~.ts.
x|t.p}s.tt.E=B.x|t.|~t.{xzt{.~.
|pxyp}p.s~.}~.wp
t.p.vtpt.
st
t{~.rwx~wt}xp=.ADD.~u.wt.BDC.
x}rxst}rt.~u.rw~xr.sxpv}~t.
x}sx
xsp{.6FA37.}t
t.ts.
r~|pts.~.}~}<|pxyp}p.t.6
p}.
|pxyp}p=.
^.t.p{=;.A??AJ.Utv~}.t.p{=;.A??DJ.
cwt.WWb.p{~.}~ts.wp.|p}.sxt.
Ztt.t.p{=;.A?@@7=.Uwt;.wt.WWb.
~.wt.ptx~}.wp.rw~x.
r~||t}ts.wp.wt}.p}p{x}v.wt.
p
px{pq{t.spp.tp|x}x}v.wt.p~rxpx~}. u~|.|pxyp}p.pvt.|p.|p}xut.~}{.
x}.x}sx
xsp{.p{tps.tsx~ts.~.
qttt}.|pxyp}p.p}s.rw~x;.x.x.
st
t{~|t}.~u.rw~xr.sx~st=.
rxxrp{.~.sxuutt}xpt.wtwt.wt.
\pxyp}p.t.|p.trtst.sxpv}~x.~u.
pxt}.x}.p.s.pt.p{tps.
rw~x.6brwx||t{|p}}.t.p{=;.A?@@7;.
sxpv}~ts.xw.rw~x.~.xu.wt.
q.|~.t~.x}sxrpt.wp.~s~|p{.
x}sx
xsp{.wp
t.p.{x|xts.}|qt.~u.
|~|.~u.rwx~wt}xp.pt.
|~|.p~rxpts.xw.rw~x.
xw~.p{xux}v.u~.p.sxpv}~x.~u.wt. ~qt
ts.x~.~.|pxyp}p.t.
6brwxuu|p}.t.p{=;.A??D7=.X}.p.t
xt.
sx~st=.
tp|x}x}v.vt}t<t}
x~}|t}p{.
P.|t}x~}ts.q.wt.WWb;.~|t.~u.
x}tprx~}.qttt}.|pxyp}p.t~t.
wt.sxt.tp|x}x}v.wt.p~rxpx~}.
qttt}.|pxyp}p.p}s.rw~x.
p}s.wt.st
t{~|t}.~u.rw~x;.x.
p.r~}r{sts.wp.wtt.x.~|t.
x{xts.}~}<p}sps.|tw~s.~.
rptv~xt.rw~x.p}s.wtt.|tw~s. t
xst}rt.~.~.wp.|pxyp}p.t.
sxs.}~.r~}u~|.~.wt.rxtxp.x}.wt.
|p.x}u{t}rt.wt.st
t{~|t}.~u.
rw~x.q.~}{.u~.rtxq{t.
Sxpv}~xr.p}s.bpxxrp{.\p}p{.
x}sx
xsp{.6_t{p~<ctp}.t.p{=;.A?@A7=.
6Sb\¤D7.~.wt.X}t}px~}p{.
t~ts.~.|pxyp}p.u~|.qxw.}x{.
ps{w~~sI.cwt.^pp._t}pp{.
_~trx
t.bs.6^__bJ.Uxts.t.p{=;.
@HG?7;.p}s.wt.\pt}p{.Wtp{w._prxrt.
p}s.Rwx{s.St
t{~|t}._~ytr.
6\W_RSJ.Sp.t.p{=;.@HGD7=.Q~w.
{~}vxsx}p{.sxt.t~.wp.wtp
.
t}pp{.|pxyp}p.t.x.p~rxpts.
xw.strtpts.tu~|p}rt.~}.pz.
ptx}v.|t|~;.
tqp{.p}s.
p}xpx
t.tp~}x}v.x}.C<tp<~{s.
6Uxts.p}s.fpzx}~};.@HH?7.p}s.x}.E.
tp.~{s.6V~{srw|xs.t.p{=;.A??G7=.X}.
qtt}.sxt.xw.wt.^__b.
r~w~;.stuxrx.x}.px}ts.pt}x~}.
tt.t~ts.x}.rwx{st}.pvt.E.p}s.
@B¤@E.tp.6Uxts.t.p{=;[email protected]xts;.
A??A7.p}s.stuxrx.x}.ttrx
t.
}t~{~vxrp{.u}rx~}.tt.~qt
ts.x}.
H<.p}s.@A<tp<~{s.rwx{st}.6Uxts.t.p{=;.
@HHG7=.STP.uwt.}~t.wp.xw.wt.
\W_RS.r~w~;.u~{{~<.sxt.
t~ts.p}.x}rtpts.pt.~u.st{x}t}.
qtwp
x~.6Sp.t.p{=;.A?@@7.p}s.
strtpts.prwxt
t|t}.t.r~t.
6V~{srw|xs.t.p{=;[email protected].pvt.@C=.
fwt}.wt.\W_RS.r~w~.p.u~{{~ts.
~.pvt.AA;.wtt.p.p.|pvx}p{.6r!L.
?=?E7.x}rtpt.x}.rw~x.xw.
t}pp{.|pxyp}p.t~t.p}s.tp{.
~}t.~u.|pxyp}p.t.6Sp.t.p{=;.A?@D7=.
RQ!11111
Hto!11175
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
Stvt}wps.t.p{=.6A??B7.|~st{ts.wt.
t
p{t}rt ~u.rwx~wt}xp.pvpx}.
|pxyp}p.t.pr~.txvw.qxw.r~w~.
x}.x}sx
xsp{.q~}.sx}v.@HC?.~.@HFH.
x}.Pp{xp=.T
t}.w~vw.wtt.p.p}.
x}rtpt.x} |pxyp}p.t.x}.wt.ps{.
qytr ~
t.wx.x|t.tx~s;.wtt.p.
}~.p}.x}rtpt.x}.sxpv}~t.~u.
rw~x.u~.wtt.p|t.qytr=.cwt.
pw~.r~}r{sts.wp.t.~u.|pxyp}p.
|p.x}rtpt rwx~wt}xp.~}{.x}.
t~}.
{}tpq{t.~.st
t{~x}v.
rw~x=.
cwt.WWb.pts.wp.prt.t.~u.
|pxyp}p.rpt.p}.x}rtpt.x}.wtp.
pt.6prwrpsxp7 p}s.|p.x}rtpt.
q{~~s.tt.6Rpx~x.t.p{=;.@HGGJ.
Qt}~x p}s Y~}t;.@HFD7=.cwtt.x.
~|t.t
xst}rt.wp.p~rxpt.wt.
x}rtpts.wtp pt.u~|.
t~t.xw.trxpx~}.~u.wt.
|pwtxr.p}s.sttx~}.~u.wt.
pp|pwtxr.}t
~.t|.
6\p{x}~zp.t.p{=;.A?@A7=.c~{tp}rt.~.
prwrpsxp st
t{~.xw.rw~}xr.
t~t.~ |pxyp}p.6Y~}t;.A??AJ.
bxs}t;.A??A7=.
_~{~}vts t~t.~. <cWR.t{.
x}.p.strtpt.x}.wtp.pt.6qpsrpsxp7.
p}s.w~t}x~}.6Qt}~x.p}s.Y~}t;.
@HFD7=.cwtt.tuutr.pt.w~vw.~.qt.
|tsxpts.w~vw.txwtp{{.{~rpts;.
t}pxr.RQ@.trt~.x}wxqxx~}.~u.
}~tx}twx}t.t{tpt.xw.~xq{t.
sxtr prx
px~}.~u.
pr{p.
rp}}pqx}~xs.trt~.6fpv}t.t.p{=;.
@HHGJ._prwt.t p{=;.A??E7=.
P.pts.x}.wt.WWb.tr~||t}spx~}.
6WWb;.A?@D7;.|pxyp}p.t~t.rpt.
~w~pxr.w~t}x~}.6upx}x}v<{xzt.
utt{x}vJ.sst}.s~.x}.q{~~s.tt.
~}.p}sx}v.7.p}s.~{tp}rt.rp}.
st
t{~ ~.wx.tuutr.~}.ttpts;.
rw~}xr.t~t.6Y~}t;.A??A7=.
c~{tp}rt.~.~w~pxr.w~t}x~}.x.
~t}xp{{ t{pts.~.{p|p.
~{|t.
tp}x~}; q.~{tp}rt.s~t.}~.
st
t{~ ~.x}t.w~t}x
t.tuutr.
6Qt}~x p}s Y~}t;.@HFD7=.
\pxyp}p |~zx}v;.pxr{p{.q.
w~t.xw.~|t.stvtt.~u.r~~}p.
pt ~.rttq~
pr{p.sxtpt;.~t.
xz.rw.p.x}rtpts.rpsxpr.~z;.
x}rtpts.rptrw~{p|x}t.p}s.
rpq~wt|~v{~qx}; |~rpsxp{.
x}uprx~}.p}s.~p{.w~t}x~}.
6Qt}~x p}s Y~}t;.@[email protected]~{{xt;.
@HGGJ.\x{t|p}.t.p{=;.A??@J.
\p{x}~zp.t.p{=;.A?@A7=.W~t
t;.
t{tr~rpsx~vpwxr.rwp}vt.tt.
|x}x|p{.put.ps|x}xpx~}.~u.{pvt.
r|{px
t.s~t ~u. <cWR.6Qt}~x.
p}s.Y~}t;.@HFD7=.
cwt.STP.}~t.~.trt}.t~.wp.
t
xtts.t
tp{.rpt.sxt.~}.
|pxyp}p p}s rpsx~
pr{p
r~|{xrpx~}.6_p}px~xst;.A?@DJ.
Wprzp|; A?@D7=._p}px~xst.6A?@D7.
t~ts.wp.p~x|pt{.AD=E3.~u.
wt.rpsx~
pr{p.rpt.u~|.|pxyp}p.
t.t{ts.x}.stpw.u~|.spp.
~
xsts.q.wt.Ut}rw.
Pssxr~
xvx{p}rt.]t~z.sx}v.wt.
tx~s ~u.A??E¤A?@?=.bt
tp{.rpt.
sxt.~}.|pxyp}p.pvt.p}s.
rpsx~
pr{p.t
t}.tt.sxrts.
p}s.x p.r~}r{sts.wp.p{w~vw.p.
rpp{.{x}z rp}}~.qt.tpq{xwts.st.~.
}~.z}~x}v.tpr.p|~}.~u.
|pxyp}p.ts.x}.wt.rpt.p}s.
r~}u~}sx}v.
pxpq{t;.wt.p
px{pq{t.
t
xst}rt.~.p.{x}z.qttt}.
|pxyp}p.p}s.rpsx~~xrx=.Wprzwp|.
6A?@D7.t
xtts.BC.rpt.t~.~.rpt.
txt.t~.~u.|pxyp}p.p}s.~zt>.
xrwt|xp.x} EC.~zt.pxt}.p}s.
t~ts.wp.x}.G@3.~u.wt.rpt.wtt.
p.p.t|~p{.t{px~}wx.qttt}.
|pxyp}p.pvt.p}s.~zt.~.xrwt|xr.
t
t}=.cwt.pw~.r~}r{sts.wp.
r~{{trx
t.p}p{x.~u.wt.rpt.t~.
~ p.rpp{.{x}z.qttt}.
|pxyp}p.t.p}s.~zt=.
cwt.WWb.pts.wp.p}xt}.
q~}rw~sx{px~}.x.wt.|~.xrp{.
txp~ tuutr.~u.prt.t~t.~.
|pxyp}p.6V~}v.t.p{=;.@HGC7=.X}.p.trt}.
{~}vxsx}p{ s;.x}u~|px~}.~}.
|pxyp}p.t.p}s.{|~}p.spp.
u}rx~}.tt.r~{{trts.u~|.D;@@D.
x}sx
xsp{.~
t.A?.tp.u~|.C.
r~||}xxt x}.wt.d}xts.bpt.
6^pz{p}s;.RPJ.Rwxrpv~;.X[J.\x}}tp~{x;.
\]J Qx|x}vwp|;.P[7.6_{trwt.t.p{=;.
A?@A7=.^u.wt.D;@@D.x}sx
xsp{;.FHD.
x}sx
xsp{.t~ts.t.~u.~}{.
|pxyp}p.6xw~.~qprr~7=.cwt.
pw~.t~ts.wp.~rrpx~}p{.t.~u.
|pxyp}p.6F.y~x}<tp.u~.{xutx|t.~.@.
y~x}>sp.u~.F.tp.~[email protected]~x}>ttz.u~.
CH.tp7.s~t.}~.ps
tt{.puutr.
{|~}p.u}rx~}=._{trwt.t.p{=.
[email protected]wt r~}r{sts.wp.wtt.x.
~|t.t{x|x}p.t
xst}rt.vvtx}v.
wp.wtp
.|pxyp}p.t.|p.wp
t.p.
stx|t}p{ tuutr.~}.{|~}p.
u}rx~};.q.wt.p|{t.xt.~u.wtp
.
|pxyp}p.t.x}.wt.s.p.~~.
|p{{=.Uwt;.p.|t}x~}ts.x}.wt.WWb.
tr~||t}spx~}.s~r|t} 6WWb;.
A?@D7;.{~}v<t|.t.~u.|pxyp}p.|p.
{tps.~.rw~}xr.r~vw;.x}rtpts.
|;.p t{{.p.x}rtpts.utt}r.
~u.rw~}xr q~}rwxx.p}s.wp}vxx.
6Psp|.p}s.\px};[email protected]~{{xt;.
cwt WWb pts wp wt t
xst}rt wp
|pxyp}p.|p.{tps.~.rp}rt.~u.wt.
txp~ t|.x.x}r~}xt};.xw.
~|t sxt vvtx}v p ~xx
t
r~t{px~} wx{t.~wt.s~.}~.6[tt.p}s.
Wp}r~;.A?@@J.cpwzx};.A??D7=.cwt.WWb.
}~ts p rpt txt wp t~ts {}v
rp}rt ~rrt}rt.x}.wtt.|pxyp}p.
|~zt.6pvt.p}vt.B@¤BF.tp7.xw.}~.
wx~.~u.~qprr~.|~zx}v.6U}v.t.p{=;.
@HHH7= Uwt|~t; x} p rpt<r~}~{
s.6}[email protected]}sx
xsp{.xw.
p|~.rt{{.rprx}~|p.~u.wt.wtps.
p}s }trzJ } L @FE r~}~{J iwp}v t p{=;
@HHH7;.t
p{t}rt.~u.|pxyp}p.t.p.
H=F3.x}.r~}~{.p}s.@B=H3.x}.rpt.
p}s wt pw~ t~ts wp |pxyp}p
t.|p.s~t<stt}st}{.x}tpr.xw.
|pvt}xr.t}xx
x;.rxvptt.
|~zx}v; p}s p{r~w~{ t ~ x}rtpt
xz.p~rxpts.xw.wtps.p}s.}trz.
rp}rt.6iwp}v.t.p{=;.@HHH7=.W~t
t;.
x} p {pvt r{x}xrp{ s xw @;ED?
qytr;.}~.~xx
t.r~t{px~}.p.
u~}s.qttt}.|pxyp}p.t.p}s.{}v.
rp}rt 6cpwzx} t p{=; A??E7= cwx
ux}sx}v wt{s t.tvps{t.~u.wt.tt}.
~u.|pxyp}p.t.wt}.q~w.~qprr~.t.
p}s ~wt ~t}xp{ r~}u~}sx}v upr~
tt.r~}~{{ts=.cwt.WWb.r~}r{sts.
wp.}t t
xst}rt.vvt.wp.wt.
tuutr ~u |~zx}v |pxyp}p ~}
txp~ u}rx~}.p}s.rp}rt.pt.
sxuutt}.u~|.wt.tuutr.~u.|~zx}v.
~qprr~ 6[tt p}s Wp}r~; A?@@7=
cwt STP uwt }~t wt q{xrpx~}
~u.trt} t
xt.pxr{t.rxxrp{{.
t
p{px}v.wt.p~rxpx~}.qttt}.
|pxyp}p.p}s.{}v.rp}rt=.\~.~u.wt.
t
xt.pvtt.wp.wt.p~rxpx~}.x.
tpz.~.x}r~}xt}.6Wp}v.t.p{=;.A?@DJ.
iwp}v.t p{=;[email protected]t.t.p{=;.A?@CJ.
Wp{{.p}s.Stvt}wps;.A?@C7=.Wp}v.t.p{=.
[email protected]}xuxts p}s.t
xtts.x.
sxt.t
p{px}v wt.p~rxpx~}.
qttt}.|pxyp}p.t.p}s.{}v.rp}rt.
p}s.wt.pw~.r~}r{sts.wp.p}.
p~rxpx~}.x.}~.~ts.|~.{xzt{.
st.~.wt.|p{{.p|~}.~u.|pxyp}p.
|~zts.x}.r~|px~}.~.~qprr~=.
iwp}v.t p{[email protected]p|x}ts.x.rpt.
r~}~{.sxt.u~|.wt.db;.dZ;.]t.
itp{p}s; p}s Rp}psp.xwx}.wt.
X}t}px~}p{.[}v.Rp}rt.R~}~x|.
p}s.u~}s.wp.wtt.p.p.tpz.
p~rxpx~}.qttt}.|~zx}v.|pxyp}p.
p}s.{}v.rp}rt.x}.x}sx
xsp{.w~.
}t
t.|~zts.~qprr~;.q.trxx~}.~u.
wt.p~rxpx~}.p.{~.p.wxvw.
|pxyp}p.t~t.{t
t{=.Wp{{.p}s.
Stvt}wps.6A?@C7.}~ts.wp.t
t}.
w~vw.|pxyp}p.|~zt.r~}px}.
t
tp{.~u.wt.p|t.rprx}~vt}.p}s.r~<.
rprx}~vt}.p ~qprr~.|~zt.6a~w.t.
p{=;[email protected]}s.wp.qtt}.u~}s.~.qt.
|pvt}xr.p}s.rprx}~vt}xr.x}.wt.
|~t zx}.t;.txst|x~{~vxrp{.sxt.
wp
t.qtt}.x}r~}xt};.q.|~t.
r~}xt} ~xx
t.p~rxpx~}.wp
t.
qtt}.t~ts x}.rpt.r~}~{.sxt=.
Ux}p{{.Vpt.t.p{=.6A?@C7;.t
xtts.wt.
sxt.t
p{px}v |pxyp}p.t.p}s.
{}v.rp}rt p}s.r~}r{sts.wp.wtt.x.
t
xst}rt.wp.|pxyp}p.~srt.
rwp}vt x}.wt.txp~.t|.
6tr~.~ rp}rt7.wp.r~{s.{tps.~.
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
{}v.rp}rt;.q.~
tp{{.p~rxpx~}.x.
tpz.qttt}.|pxyp}p.t.p}s.{}v.
rp}rt ttrxp{{.wt}.r~}~{{x}v.u~.
~qprr~.t=.
cwt.WWb.pts.wp.trt}.sxt.
~.p ~xq{t.p~rxpx~}.qttt}.
utt};.{~}v<t|.|pxyp}p.t.p}s.
x}rtpts.xz.~u.txr{p.vt|.rt{{.
|~=.X}.p.w~xp{<qpts.rpt<r~}~{.
s;.wt.utt}r.~u.|pxyp}p.t.
p.r~|pts.qttt}.txr{p.vt|.
rt{{ |~.6cVRc7.pxt}.6}.L.@GF7.
p}s.r~}~{.6}[email protected]pqt.t.p{=;.
cVRc.pxt} tt.|~t.{xzt{.~.qt.
utt}.|pxyp}p.t.wp}.r~}~{.
xw.p}.~ss.px~.6^a7.~u.A=A.6HD3.
r~}uxst}rt.{x|x ~u.@=?¤[email protected]}s.tt.
{t.{xzt{.~.qt.x}utt}.~.w~<t|.
t.xw.~ss.px~.~u.?=D.p}s.?=E;.
ttrx
t{.x}.r~|px~}.~.r~}~{.
6cpqt t.p{=;.A?@@7=.cwt.STP.uwt.
}~t.wp.x}.~.~{px~}<qpts.rpt<.
r~}~{.sxt.6Sp{x}v.t.p{=;.A??HJ.
[pr~}.t p{=;.A?@A7;.|pxyp}p.t.p.
r~|pts.qttt} pxt}.sxpv}~ts.
xw.cVRc.p}s.|prwts.r~}~{.x}.
fpwx}v~}.bpt.~.[~.P}vt{t.
R~}=.X}.q~w.sxt;.x.p.t~ts.
wp.cRVc.pxt}.tt.xrt.p.{xzt{.
p.r~}~{.~.t.|pxyp}p=.Pw~.~u.
q~w.sxt.r~}r{sts.wp.|pxyp}p.
t.x.p~rxpts.xw.p}.t{t
pts.xz.
~u.cVRc.6Sp{x}v.t.p{=;.A??HJ.[pr~}.t.
cwt.WWb.rxts.p.s.6bpupp.t.p{=;.
A??D7.st|~}px}v.wp.fX].DD;A@A¤A.
6p.|xts RQ@>RQA.pv~}x7.x}srt.
p~~x.6~}t u~|.~u.rt{{.stpw7.x}.
~pt rp}rt.rt{{.p}s.strtpt.
ttx~}.~u.p}s~vt}.trt~.p}s.
~pt trxuxr.p}xvt};.vvtx}v.p.
~t}xp{.wtptxr.
p{t.u~.
rp}}pqx}~xs.pv~}x.x}.wt.tp|t}.~u.
~pt rp}rt;.p}.p}s~vt}<x|{pts.
t ~u.rprx}~|p=.
|p.x}tpr.xw.wt.v{r~r~xr~xs.
trt~.t|=.
X||}t.bt|.
cwt.WWb.pts.wp.rp}}pqx}~xs.
p{t.x||}t.u}rx~}.q.wp.wtt.rp}.
qt.sxuutt}rt.qttt}.wt.tuutr.~u.
at~srx
t.W~|~}t.
}wtxr;.}pp{; p}s.t}s~vt}~.
rp}}pqx}~xs.6R~u~s.p}s.hp|p|p;.
cwt WWb pts wp ps|x}xpx~} ~u
A??DJ.cp}ptr.p}s.R~}p}x}tr;.
|pxyp}p.~.w|p}.s~t.}~.
r~}xt}{.p{t.wt.t}s~rx}t.t|=.
cwt.WWb.}~ts.wp.wtt.pt.
X}.p.r~}~{{ts.w|p}.t~t.s.
r~}u{xrx}v.t{.x}.p}x|p{.p}s.w|p}.
6}.L.C.|p{t7;.qytr.tt.prt{.
sxt.xw.ttr.~.rp}}pqx}~xs.
ps|x}xtts.|~zts.|pxyp}p.
tuutr.~}.x||}t.u}rx~}x}v.x}.
r~}px}x}v.A=G3.
qytr xw.r~|~|xts.x||}t.
p}.x||tsxpt.xv}xuxrp}.strtpt.x}.
t|=.Pqp|.t.p{=.6A??B7.tp|x}ts.
{tx}xx}v w~|~}t.p}s.p}.x}rtpt.x}.
wt.tuutr ~u.|pxyp}p.p}s.
r~x~{.p.t~ts.x}.wt.qytr.wp.
EA.WXe¤@<x}utrts.pxt}=.bqytr.
|~zts.|pxyp}p.6R~}t.t.p{=;.@HGE7=.
trtx
ts.~}t.~u.wtt.tp|t};.wtt.
Uwt|~t;.p.rxts.q.wt.WWb;.~.
x|t.p.spI.|~zts.|pxyp}p.rxvptt.
{pt.sxt.6Sp.t.p{=;[email protected]{~rz.t.
r~}px}x}v.B=HD3. <cWR;.~p{.pq{t.
p{=;.@HH@7.t~ts.}~.rwp}vt.x}.
r~}px}x}v.
w~|~}t.{t
t{= Sp.t.p{=.6@HGH7.
s~}pqx}~{7; ~.~p{.{prtq~=.cwtt.tt.
trxts |p{t
~{}tt 6} L @F7 wp
}~.rwp}vt.x}.RSC9.p}s.RSG9.rt{{.
tt.~rrpx~}p{ ~.wtp
.t.~u.
r~}; WXe.a]P.{t
t{;.~.~tpt.
|pxyp}p=.U~{{~x}v.t~t.~.
x}wxqx~.{t
t{ x}.p}.~u.wt.tp|t}.
|~zts. <cWR.6@G.|v>rxvptt7.~.~p{.
v~.6Pqp|.t.p{=;.A??B7=.cwttu~t;.
<cWR.6@?.|v.wtt.x|t.t.sp.u~.
t.~u.rp}}pqx}~xs.w~ts.}~.w~<.
wtt sp.p}s.~}.wt.|~}x}v.~u.wt.
t|.ps
tt.
x~{~vxr.tuutr.x}.
u~w.sp7;.wt.qytr.x}.wp.s.
x}sx
xsp{.xw.r~|~|xts.x||}t.
w~ts.}~ rwp}vt.x}.{p|p.
t|=.R~}
tt{;.a~w.t.p{=.6A??D7.
pst}~r~xr~~xr.w~|~}t.6PRcW7;.
t~ts.wp.x}.x||}~stuxrxt}.|xrt.
r~x~{;.~{prx};.{tx}xx}v.w~|~}t;.
x|{p}ts.xw.w|p}.q{~~s.rt{{.
~.t~t~}t.{t
t{=.Pssxx~}p{{;.
x}utrts.xw.WXe;.t~t.~.
Q{~rz.t.p{=.6@[email protected]~|pts.{p|p.
tt.x||}t.u}rx~};.
w~|~}t.{t
t{.p|~}v.}~}<t.p.
x}rtpt.WXe r~<trt~.ttx~};.
t{{ p.x}utt};.|~stpt;.p}s.
p}s.pr.p.p.r~upr~.~.t}wp}rt.WXe.
t{xrpx~}=.
utt}.t.~u.|pxyp}p.6}.L.HB.|t}.
p}s.DE.~|t}7.p}s.u~}s.wp.rw~}xr.
cwt.STP.}~t.~.trt}.r{x}xrp{.
t.~u.|pxyp}p.6x}utt};.|~stpt;.
sxt.t~x}v.p.strtpt.x}.r~zx}t.
p}s.x}t{tzx}.{t
t{.u~{{~x}v.
p}s.utt}.t7.sxs.}~.xv}xuxrp}{.
|pxyp}p.t=.Ztt}.t.p{=.6A?@C7.
p{t.r~}rt}px~}.~u.t~t~}t;.
r~|pts.wt.sxuutt}rt.x}.wt.{t
t{.~u.
{tx}xx}v w~|~}t; u~{{xr{t.x|{px}v.
w~|~}t;.~{prx};.~.r~x~{=.
^wt.W~|~}t.6cw~xs;.Ptxt7. X[¤E.6x}t{tzx}<E7;.p.~x}u{p||p~.
r~zx}t;.p|~}v }~}<sv.t.6}.L.
cwt.WWb.}~ts.wp.wtt.x.p.
X}.|~t.trt}.sxt;.p.rxts.q.wt. FG7;.|pxyp}p.~}{.t.6}.L.CE7.p}s.
sxrtp}r.x}.wt.tuutr.~u.|pxyp}p.
WWb;.rw~}xr.|pxyp}p.t.q.qytr. |pxyp}p.{.~wt.sv.t.6}.L.CD7.
~}.ut|p{t.t~srx
t.t|.
6}.L.BH7.rwpprtxts.p.stt}st}.~}.
x}.p.r~||}x<qpts.p|{t.~u.
u}rx~}p{x.qttt}.p}x|p{.p}s.
|pxyp}p.prr~sx}v.~.wt.XRS¤@?.
|xss{t<pvts.Puxrp}.P|txrp}.6Ztt}.
w|p} 6WWb;.A?@D7=.Ut|p{t.wt.
rxtxp.sxs.}~.puutr.t|.{t
t{.~u.
t.p{=;.A?@C7=.Put.psyx}v.u~.
|~}zt.wp.tt.ps|x}xtts.A=D.
w~xs.w~|~}tI.cbW.6w~~x}7;.
r~}u~}st; p}p{t.t
tp{ts.wp.
<cWR;.x=|=;.sx}v.sp.@¤@G.~u. cC.6w~x}t7;.p}s.cB.
{xutx|t.|pxyp}p.~}{.t.wps.
wt.|t}p{ rr{t.wps.tsrts.
6xx~s~w~}x}t7.6Q~}}t;.A?@B7=.fxw. xv}xuxrp}{.{~t.X[¤E.{t
t{.wp}.wt.
~vtt~}t.{t
t{.p}s.~
{px~}.p.
ttr.~.ptxt.w~|~}t;.x}.p.x{~. }~}t.v~=.Uwt;.bt~}.t.p{=.
tts.6Prw.t.p{=;.@HG@7=.W~t
t;. s.xw.WXe<~xx
t.|p{t;.|~zx}v. [email protected]~|pts.t
tp{.x||}t.
~|t}.w~ |~zts.|pxyp}[email protected]p|. |pxyp}p.s~t<stt}st}{.x}rtpts.
pp|tt.x}.wtp{w.x}sx
xsp{.p}s.
|pxyp}p rxvptt xw @=G3 {p|p.{t
t{.~u.vwt{x}.p}s.{tx}.p}s. qytr xw.|{x{t.r{t~x.6\b7.
sx}v.wt.tx~
{p~.tx~s.6AC¤BE. strtpts.{p|p.{t
t{.~u.txst.hh.
p}s.u~}s.wp.wt.rw~}xr.t.~u.
w~.x~.~.~
{px~}7.sxs.}~.twxqx. 6axvv.t.p{=;.A?@A7=.
|pxyp}p.t{ts.x}.tsrts.|~}~rt.
rwp}vt x}.t~srx
t.w~|~}t.{t
t{.
|xvpx~};.p}s.strtpts.{t
t{.~u.RR[A.
cwt.WWb.pts.wp. <cWR tsrt
~.wtx.|t}p{.rr{t.6\t}st{~}.
p}s.X[¤@F.x}.q~w.wtp{w.p}s.\b.
qx}sx}v.~u.wt.r~xr~t~xs.
p}s.\t{{~;.@HGC7=.X}.p.t
xt.pxr{t.q. stp|twp~}t.x} wx~rp|p{.xt.
v~=.
Q~}.p}s.S~q.6A??A7;.wt.pw~.
u~| pst}p{tr~|xts p p}s prt
cwt STP p{~ }~t p t
xt
pt.wp.t}s~rx}t.rwp}vt.~qt
ts.
<cWR.t{tpt.r~xr~t~}t;.xw.
vvtx}v.wp. <cWR.tt.wt.
xw.|pxyp}p pt.}~.{~}vt.~qt
ts.
~{tp}rt.st
t{~x}v.~.wx.tuutr.xw. x||}t.t~}t.x}.ttx|t}p{.
xw.rw~}xr.ps|x}xpx~}.p}s.wx.
rw~}xr ps|x}xpx~} 6T{sxsvt t p{=;
p}x|p{ |~st{ p}s p}s wp
@HH@7=.cwtt.spp.vvt.wp.
wtt.rwp}vt |p.qt.x|px{.
|p.qt.st.~.sv.~{tp}rt=.
at~srx
t.Rp}rt.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
|tsxpts.w~vw.wt.RQA.rp}}pqx}~xs. stpw.pxqts.~.p.sv.x.}~.
trt~.6Txt}tx}.p}s.\tx{t;.A?@D7=. t
xst}rt.wp.wt.sv.x.put.u~.
|tsxrp{.t=.
Rgvkvkqpgtu!Oclqt!Eqoogpvu!kp!Tgncvkqp!
vq!Hcevqt!3!cpf!vjg!Iqxgtpogpvu!
Hcevqt!4<!Vjg!Uvcvg!qh!vjg!Ewttgpv!
Tgurqpugu!
Uekgpvkhke!Mpqyngfig!Tgictfkpi!vjg!
Ftwi!qt!Uwduvcpeg!
[email protected].txx~}t.pt.wp.
]o_gfkecn!wug!qh!ecppcdku!ku!eqpukfgtgf! Ejgokuvt{!
uchg/!6Twxqx.Q;.pvt.F7J.p}s.wp.
cwt.WWb.pts.wp.|pxyp}p;.p{~.
]v_jgtg!ctg!cfgswcvg!cpf!ygnn.!
z}~}.p.Ecppcdku!ucvkxc!N/-!x.p.~u.
eqpvtqnngf!uvwfkgu!rtqxkpi!vjg!ogfkecn!
wt.Rp}}pqprtpt.{p}.up|x{.p}s.x.~}t.
ghhkece{!qh!ecppcdku/!6Twxqx.Q;.pvt.
~u.wt.~{st.r{x
pts.r~=.cwt.t|.
@?7=.cwt.txx~}t.p{~.p{{tvt.wp.
|pxyp}p .x.vt}tp{{.ts.~.tut.~.
Ecppcdku!ku!uchgt!vjcp!ewttgpv-!ngicn!
p.|xt.~u.wt.sxts.u{~tx}v.~.
Uejgfwng!KK!qrkcvg!ftwiu!p}s.wp.x.
p}s.{tp
t.u~|.Ecppcdku/!\pxyp}p.
tt}.|x{st.xst.tuutr.6Twxqx.Q;.
t.x|px{.|~zt.wt.|pxyp}p.
•pvt.H¤@?7=.
P.stpx{ts.x}.wt.WWb.t
xt.p}s.p. {tp
t;.q.x}sx
xsp{.p{~.x}vt.
|pxyp}p.w~vw.u~~s.x}uts.xw.
sxrts.{pt.x}.wx.s~r|t}.6tt.
|pxyp}p.p}s.x.tpr=.Ecppcdku!
Upr~.B7;.wtt.pt.}txwt.pstpt.
ucvkxc!x.wt.x|p.trxt.~u.
put.sxt.}~.pstpt;.t{{<.
Ecppcdku!wp.x.x{{tvp{{.|pztts.x}.
r~}~{{ts.sxt.~
x}v.|pxyp}p .
tuuxrpr=.cwt.STP.}~t.wp.}txwt.wt. wt.d}xts.bpt=.\pxyp}p.x.~}t.~u.
RbP.}~.tpq{xwts.rwts{x}v.rxtxp. wtt.|py~.stx
px
t.~{s.p.tppt.
x{{xrx.~sr;.wt.~wt.~.qtx}v.
vvt.wp.wt.WWb.p}s.STP.w~{s.
wpwxw.p}s.wpw.~x{=.Wpwxw.x.
r~}xst.wt.t{px
t.put.~ux{t.~u.
r~|~ts.~u.wt.sxts.p}s.r~|tts.
sv.wt}.stt|x}x}v.wt.~t.
rp}}pqx}~xs<xrw.tx}~.|ptxp{.~u.
rwts{t=.c~.wt.tt}.wp.wt.
Ecppcdku!p}s.x.u~}s.p.qp{{.p}s.
txx~}t.tt.tutx}v.~.pqt.p}s.
rpzt.p.t{{.p.~wt.u~|=.X}sx
xsp{.
stt}st}rt.{xpqx{x;.wx.s~r|t}.
|p.qtpz.~uu.xtrt.p}s.{prt.wt|.
sxrt.w~t.tuutr.x}.upr~.@;.C;.
x}~.p.xt.~.|~zt=.Wpw.~x{;.p.
xr~.
p}s.F=.
q~}.~.p|qt.r~{~ts.{xxs;.x.
6A7.cwt.txx~}t.pt.wp.
uekgpvkhke!gxkfgpeg!tgictfkpi!vjg!uchgv{! ~srts.q.~{
t}.tprx~}.~u.
rp}}pqx}~xs.u~|.Ecppcdku!p}s.
cpf!ghhkece{!qh!ecppcdku!ku!tgcfkn{!
r~}px}.p~x|pt{.D?3.
cxckncdng!fktgevn{!htqo!vjg!Pcvkqpcn!
rp}}pqx}~xs=.^}t.~.~.s~.~u.wpw.
Nkdtct{!qh!Ogfkekpg/!6Twxqx.Q;.pvt.
~x{.~}.p.rxvptt.wp.qtt}.t~ts.~.
@C7=.
~srt.wt.tx
p{t}.~u.p.x}v{t.
cwt.v~
t}|t}.pvtt.wp.|p}.
|pxyp}p.rxvptt.6STP;.A?@D7=.
pxr{t.fkuewuu!|pxyp}p.p}s.x.
cwt.WWb.x}sxrpts.x}.x.t
p{px~}.
r~}xt}=.ht;.wtt.pxr{t.x}.}~.
wp.wt.txx~}t.stux}ts.|pxyp}p.p.
p.st|~}pt.wp.|pxyp}p.x.put.
x}r{sx}v.p{{.Ecppcdku!r{x
pts.
p}s.tuutrx
t.u~.wt.tp|t}.~u.p}.
px}=.W~t
t;.sxuutt}.|pxyp}p.
sxtpt.~.r~}sxx~}=.P.|t}x~}ts.x}.
p|{t.pt.stx
ts.u~|.}|t~.
wt.WWb.t
xt.p}s.p.sxrts.{pt.
r{x
pts.px}.p}s.|p.wp
t.
x}.wx.s~r|t}.6tt.Upr~.B7;.wt.
sxuutt}.rwt|xrp{.r~|~xx~}.
rt}.ttprw.s~t.}~.~
xst.
x}r{sx}v.{t
t{.~u.% <cWR.p}s.~wt.
pstpt.stpx{ts.rxt}xuxr.t
xst}rt.
rp}}pqx}~xs.6Pt}sx}~.t.p{=;.A?@@7=.
tvpsx}v.rwt|x;.wp|pr~{~v;.
P.r~}tt}rt.~u.wp
x}v.sxuutt}.
~xr~{~v;.p}s.tuutrx
t}t.stx
ts.
rwt|xrp{.r~|~xx~}.x}.wt.
px~.
u~|.t{{<r~}~{{ts.r{x}xrp{.
|pxyp}p.p|{t.x.wp.wtt.x{{.qt.
x}
txvpx~}.~.t|x.p.r~}r{x~}.
xv}xuxrp}.sxuutt}rt.x}.put;.
wp.|pxyp}p.x.put.p}s.tuutrx
t.u~.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=. qx~{~vxrp{;.•wp•|pr~{~vxrp{;.p}s.
6B7.cwt.txx~}t.|t}x~}ts.~}.pvt. ~xr~{~vxrp{.~ux{t.p}s.wttu~t;.
prr~sx}v.~.wt.WWb;.p{{.Ecppcdku!
H.~u.twxqx.Q.wp.]v_jgtg!jcu!pgxgt!
px}.rp}}~.qt.r~}xstts.
dggp!c!ngvjcn!qxgtfqug!qh!octklwcpc!
tgrqtvgf!kp!jwocpu!p}s.wp.]v_jgtg!ku! r~{{trx
t{.qtrpt.~u.wt.
pxpx~}.x}.
rwt|xrp{.r~|~xx~}=.Uwt|~t;.wt.
pq!mpqyp!NF61!hqt!cp{!hqto!qh!
r~}rt}px~}.~u.% <cWR.p}s.~wt.
ecppcdku/!
P.|~t.u{{.sxrts.x}.Upr~.B.
rp}}pqx}~xs.tt}.x}.|pxyp}p.|p.
qt{~;.wt.WWb.p}s.STP.r~}r{st.wp.
p.st.~.v~x}v.r~}sxx~}.p}s.
~rtx}v.~u.wt.{p}.put.wp
tx}v=.
wtt.pt.}~.pstpt.sxt.~.
stt|x}t.wt.put.~u.|pxyp}p=.P.
U~.tp|{t;.wt.{p}.p.r~{{trts.
rw.p.u{~t;.{tp
t.p}s.t|.rp}.
sxrts.x}.wt.WWb.s~r|t}.p}s.
x}u{t}rt.|pxyp}p .~t}r;.p{x;.
qt{~;.wt.stt|x}px~}.~u.put.x.
|~t.r~|{t.wp}.p.|tt.
p}s.x.6Psp|.p}s.\px};.@HHEJ.
Pvt{{.t.p{=;.@HGCJ.\trw~{p|;.@HFB7=.
stt|x}px~}.~u.wt.pt.~.{xzt{xw~~s.
~u.stpw=.\~t~
t;.wt.{prz.~u.~
ts~t. epxpx~}.x}.|pxyp}p.wp
tx}v.wp
t.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11178
Hov!5812
Uhov!5813
64864!
t{ts.x}.~t}rxt.p}vx}v.u~|.p.{~.
~u.@.~.A3..[email protected].
x}sxrpts.q.rp}}pqx}~xs.r~}t}=.cwt.
r~}rt}px~}.~u.% <cWR.p
tpvt.
p~x|pt{[email protected].txvw.x}.p.
xrp{.|pxyp}p.|xt.~u.{tp
t.p}s.
t|=.W~t
t;.~|t.trxuxrp{{.
v~}.p}s.t{trts.|pxyp}p.p|{t.
rp}.r~}px}.@D3.~.vtpt.% <cWR.
6Pt}sx}~.t.p{=;.A?@@7=.P.p.t{;.wt.
% <cWR.r~}t}.x}[email protected]p|.|pxyp}p.
rxvptt.rp}.p}vt.u~|.p.{x{t.p.B.
|x{{xvp|.~.@D?.|x{{xvp|.~.|~t=.X}.
p.t|pxr.t
xt.r~}srts.q.
Rprx}x.t.p{=.6A?@A7;.x.p.t~ts.wp.
|pxyp}p .% <cWR.r~}t}.wp.
x}rtpts.xv}xuxrp}{.u~|.@HFH¤A??H=.
bx}rt.wtt.x.r~}xstpq{t.
pxpqx{x.
x}.wt.rp}}pqx}~xs.r~}rt}px~}.p}s.
rwt|xrp{.r~}xt}r.p|~}v.|pxyp}p.
p|{t;.wt.x}ttpx~}.~u.r{x}xrp{.
spp.xw.|pxyp}p.x.r~|{xrpts=.P.
x|p.xt.x.wt.{prz.~u.r~}xt}.
r~}rt}px~}.~u.% <cWR.p}s.~wt.
qp}rt.x}.|pxyp}p.wxrw.
r~|{xrpt.wt.x}ttpx~}.~u.wt.
tuutr.~u.sxuutt}.|pxyp}p.
r~}xt}=.P}.pssts.xt.x.wp.wt.
}~}<rp}}pqx}~xs.r~|~}t}.x}.
|pxyp}p.|p.~t}xp{{.|~sxu.wt.
~
tp{{.wp|pr~{~vxrp{.p}s.
~xr~{~vxrp{.~txt.~u.
px~.
|pxyp}p.px}.p}s.~sr=.
epx~.Ecppcdku!px}.r~}px}.
|~t.wp}.DAD.xst}xuxts.}pp{.
r~}xt}.x}r{sx}v.rp}}pqx}~xs;.A@.
6~.AA7.rpq~}.tt}~xs.u~}s.x}.wt.
{p};.p.t{{.p.wtx.rpq~{xr.prxs;.
p}p{~vt;.p}s.p}u~|px~}.~sr.
6Pvt{{.t.p{=;.@HGCJ.@HGEJ.\trw~{p|;.
@HFBJ.Pt}sx}~.t.p{=;.A?@@7=.c~.spt;.
|~t.wp}.@??.rp}}pqx}~xs.wp
t.qtt}.
rwpprtxts.6T{b~w{.p}s.b{pst;.A??DJ.
apsp}.t.p{=;.A??HJ.Pt}sx}~.t.p{=;.
A?@@7;.p}s.|~.|py~.rp}}pqx}~xs.
r~|~}s.~rrx}v.}pp{{.wp
t.
qtt}.xst}xuxts=.cwtt.pt.x{{.}t.p}s.
r~|ppq{.|~t.|x}~.rp}}pqx}~xs.
qtx}v.rwpprtxts.6_~{{p~.t.p{=;.
A?@@7=.cwt.|py~x.~u.wt.rp}}pqx}~xs.
pt.u~}s.x}.Ecppcdku/!^}t.s.
t~ts.prr|{px~}.~u.~.
rp}}pqx}~xs;.rp}}pqxvt~{.p}s.x.
r~t~}sx}v.prxs;.x}.Jgnkejt{uwo!6J/!
wodtcewnkigtwo7.wxrw.x.p.}~}<.
Ecppcdku!~rt.6Pt}sx}~.t.p{=;.
A?@@7=.
^u.wt.rp}}pqx}~xs.u~}s.x}.
|pxyp}p;.% <cWR.6t
x~{.z}~}.
p.% <cWR7.p}s.st{p<G<.
tpws~rp}}pqx}~{.6% <cWR;.% <cWR7.
wp
t.qtt}.st|~}pts.~.~srt.
|pxyp}p .rw~prx
t.tuutr=.
_rw~prx
t.tuutr.u~|.|pxyp}p.
pvt.wp
t.qtt}.|px}{.pxqts.~.% <.
cWR.qtrpt.% <cWR.x.tt}.x}.
xv}xuxrp}{.|~t.p}xxt.wp}.% <.
cWR.x}.|~.|pxyp}p.
pxtxt=.cwtt.
pt.~}{.p.ut.|pxyp}p.px}.wp.
G<^HT^HO^23CWR3/UIO
23CWR3
64865!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
r~}px}.% <cWR.x}.xv}xuxrp}.p|~}.
6Wx
t{.t.p{=;.@HEE7=.% <cWR.x.p}.
~xrp{{.prx
t.tx}~.qp}rt.wp.
x.tt|t{.{x~wx{xr=.cwt.rwt|xrp{.
}p|t.u~•.% <cWR.x.6EpT<p}7<.
Ep;F;G;@?p<tpws~<E;E;H<x|tw{<B<.
t}{<EJ<sxqt}~<jq;slp}<@<~{;.~.
6¤7<st{pH<6p}7<tpws~rp}}pqx}~{=.
cwt.6¤7<p}.% <cWR.x~|t.x.
wp|pr~{~vxrp{{.E.~.@??.x|t.|~t.
~t}.wp}.wt.697<p}.x~|t.6Stt.
t.p{=;.@HGC7=.
^wt.t{px
t{.t{{<rwpprtxts.
rp}}pqx}~xs.tt}.x}.|pxyp}p.
x}r{st.rp}}pqxsx~{.6RQS7;.
rp}}pqxrw•~|t}t.6RQR7;.p}s.
rp}}pqx}~{.6RQ]7=.RQS.p}s.RQR.p•t.
|py~.rp}}pqx}~xs.x}.|pxyp}p.p}s.
pt.q~w.{x~wx{xr=.cwt.rwt|xrp{.}p|t.
u~.RQS.x.A<j6@T-ET7<B<|tw{<E<~<@<.
t}<A<{rr{~wt<A<t}<@<{l<D<.
t}{qt}t}t<@;B<sx~{.p}s.wt.
rwt|xrp{.}p|t.u~.RQR.x.A<|tw{<A<6C<.
|tw{t}<B<t}{7<F<t}{<D<.
rw~|t}~{=.RQ].x.p.|x}~.}pp{{<.
~rrx}v.rp}}pqx}~xs.xw.tpz.
rw~prx
x.p}s.x.p{~.p.|py~.
|tpq~{xt.~u.% <cWR=.cwt.rwt|xrp{.
}p|t.u~.RQ].x.E;E;H<x|tw{<B<.
t}{<qt}~jrlrw~|t}<@<~{=.
X}.||p;.|pxyp}p.wp.t
tp{.
px}.xw.wxvw.
pxpqx{x.x}.wt.
r~}rt}px~}.~u.% <cWR;.wt.|px}.
rw~prx
t.r~|~}t};.p.t{{.p.
~wt.rp}}pqx}~xs.p}s.r~|~}s=.
\pxyp}p.x.}~.p.x}v{t.rwt|xrp{.p}s.
s~t.}~.wp
t.p.r~}xt}.p}s.
t~srxq{t.rwt|xrp{.~ux{t.xw.
tsxrpq{t.~.r~}xt}.r{x}xrp{.tuutr=.
X}.wt.WWb.tr~||t}spx~}.u~.
|pxyp}p.rwts{x}v.6WWb;.A?@D7;.x.
p.tr~||t}sts.wp.x}
txvp~.
r~}{.p.vxsp}rt.u~.x}s.t}x{ts;.
Dqvcpkecn!Ftwi!Rtqfwevu- wxrw.
~
xst.x}u~|px~}.~}.wt.p~
p{.~u.
q~p}xrp{.sv.~sr=.btrxuxrp{{;.x}.
~st.~.x}
txvpt.|pxyp}p.x}.
~.~u.p.]t.Sv.P{xrpx~}.
6]SP7;.r{x}xrp{.sxt.}st.p}.
X}
txvpx~}p{.]t.Sv.6X]S7.
p{xrpx~}.w~{s.x}r{st.r~}xt}.
qprwt.~u.p.pxr{p.|pxyp}p.
~sr.u~.jpl.pxr{p.sxtpt= .
6WWb;.A?@D7=.Uwt|~t;.wt.WWb.
}~ts.wp.x}
txvp~.|.~
xst.
spp.|ttx}v.wt.txt|t}.u~.}t.
sv.p~
p{.p.x{pts.x}[email protected].
B@C=D?.6WWb;.A?@D7=.
6txvwx}v.qttt}.?=D.p}[email protected]p|7.~.x}.
p.xt=.\~t.trt}{;.
p~xt.wp
t.
qtt}.ts.p.p}~wt.|tp}.u~.
x}sx
xsp{.~.x}wp{t.|pxyp}p=.
\pxyp}p.|p.p{~.qt.x}vtts.~p{{.x}.
u~~s.~.p.p}.tpr.x}.twp}~{.~.~wt.
~{
t}=._wp|pr~zx}txr.sxt.xw.
|pxyp}p.u~rts.~}.t
p{px}v.wt.
pq~x~};.|tpq~{x|;.p}s.t{x|x}px~}.
••~ux{t.~u.% <cWR.p}s.~wt.
rp}}pqx}~xs.6Psp|.p}s.\px};.@HHEJ.
Pvt{{.t.p{=;[email protected]t{{.t.p{=;.@HGE7=.
Pq~x~}.p}s.Sxxqx~}.~u.^p{{.
Ps|x}xtts.\pxyp}p.
U~{{~x}v.~p{.ps|x}xpx~}.~u.% <.
cWR.~.|pxyp}p;.~}t.~u.tuutr.p.
xwx}.B?.~.H?.|x}t;.tpz.put.A.~.
B.w~.p}s.tuutr.t|px}.u~.C.~.@A.
w~.6V~t}wt|t};.A??BJ.Psp|.p}s.
\px};[email protected]t{{.t.p{=;.@HGCJ.
Pvt{{.t.p{=;.@HGE7=.S~t.xpx~}.~u.% <.
cWR.u~|.~p{{.x}vtts.|pxyp}p.x.
sxuuxr{.u~.t.x}.r~|px~}.~.
|~zts.~.x}wp{ts.|pxyp}p.st.~.wt.
st{p.x}.wt.~}t.~u.tuutr=.^p{.
Pq~x~}.p}s.Sxxqx~}.~u.X}wp{ts.
qx~p
px{pqx{x.~u.% <cWR;.txwt.x}.x.
\pxyp}p.b|~zt.
t.u~|.~.x}.|pxyp}p;.x.{~.p}s.
pxpq{t.xw.p.p}vt.u~|.D3.~.A?3.
cwtt.x.wxvw.
pxpqx{x.x}.wt.
6Pvt{{.t.p{=;[email protected]t{{.t.p{=;.
wp|pr~zx}txr.~u.% <cWR.p}s.~wt.
@HGE7=.cwtt.x.p{~.x}t<.p}s.x}p<.
rp}}pqx}~xs.u~|.|~zts.|pxyp}p.
qytr.
pxpqx{x.~u.~p{{.ps|x}xtts.
st.~.sxuutt}rt.x}.x}sx
xsp{.
|~zx}v.qtwp
x~.t
t}.}st.r~}~{{ts. % <cWR.}st.ttx|t}p{.r~}sxx~}.
ttx|t}p{.r~}sxx~}.6Pvt{{.t.p{=;. p}s.t
t}.}st.ttpts.s~x}v.
ttx|t}.6WWb;.A?@D7=.cwt.WWb.
@HGEJ.Wt}x}v.t.p{=;[email protected]tx.t.p{=;.
}~ts.wp.x}.qx~p
px{pqx{x.sxt.
@HHAp7=.Ttxt}rts.|pxyp}p.t.
x}v.psx~{pqt{ts.% <cWR;.% <cWR.
rp}.xpt.p}s.tv{pt.wt.s~t.q.
{p|p.{t
t{.u~{{~x}v.~p{.
w~{sx}v.|pxyp}p.|~zt.x}.wtx.{}v. ps|x}xpx~}.~u.% <cWR.tt.{~.
u~.p}.tt}sts.tx~s.~u.x|t.t{x}v. t{px
t.~.{p|p.{t
t{.put.x}wp{ts.~.
x}.x}rtpts.rw~prx
t.tuutr.q.
x}p
t}~{.ps|x}xtts.% <cWR=.
~{~}vx}v.pq~x~}.~u.wt.|~zt=.
cwt.{~.p}s.
pxpq{t.qx~p
px{pqx{x.~u.
cwx.~t.|p.p{~.wt{.t{px}.
~p{{.ps|x}xtts.% <cWR.x.st.~.
w.wtt.x.p.~~.r~t{px~}.qttt}. ux.p.wtpxr.t{x|x}px~}.u~|.
t}~.{t
t{.~u.% <cWR.p}s.wt.
q{~~s.p}s.tpxr.pq~x~}.u~|.
x}t}x.~u.tuutr.p}s.x}~xrpx~}.
~|prw.p}s.q~t{.6WWb;.A?@D7=.
6Pvt{{.t.p{=;[email protected]}t.t.p{=;.@HGDJ.
\tpq~{x|.p}s.Trtx~}.~u.
Wtx.t.p{=;.@HHAp7=.cwt.WWb.
Rp}}pqx}~xs.U~|.\pxyp}p.
tr~||t}sts.wp.uu.p}s.x}wp{px~}.
bsxt.t
p{px}v.rp}}pqx}~xs.
~{|t.w~{s.qt.przts.x}.
|tpq~{x|.p}s.trtx~}.u~rts.~}.
ttx|t}p{.sxt.qtrpt.wt.
r~}rt}px~}.~u.rp}}pqx}~xs.rp}.
p. % <cWR.qtrpt.x.x.wt.x|p.
rw~prx
t.r~|~}t}.x}.|pxyp}p=.
p.sxuutt}.pvt.~u.|~zx}v=.
% <cWR.x.|tpq~{xts.
xp.|xr~~|p{.
% <cWR.u~|.|~zts.|pxyp}p.x.
ws~{px~}.p}s.~xspx~}.~.q~w.
pxs{.pq~qts.xwx}.tr~}s=.
prx
t.p}s.x}prx
t.|tpq~{xt.
_rw~prx
t.tuutr.pt.~qt
ts.
6[t|qtvt.t.p{=;.@HF?J.[t|qtvt.t.p{=;.
x||tsxpt{.u~{{~x}v.pq~x~}.xw.
@HFApJ.[t|qtvt.t.p{=;[email protected]t{{.
|tppq{t.}t~{~vxrp{.p}s.qtwp
x~p{.
t.p{=;[email protected]~{{xt;.@HGG7=.\tpq~{x|.
rwp}vt.u~..~.E.w~.
~u.% <cWR.x.r~}xt}.p|~}v.utt}.
6V~t}wt|t};.A??BJ.W~{{xt;.@HGEJ.
p}s.x}utt}.|pxyp}p.t.6Pvt{{.
W~{{xt;.@HGG7=.% <cWR.x.sxxqts.
t.p{=;.@HGE7=.cwt.x|p.prx
t.
~.wt.qpx}.x}.p.pxs.p}s.tuuxrxt}.
|tpq~{xt.~u.% <cWR.u~{{~x}v.~p{.
|p}}t=.Qx~p
px{pqx{x.~u.% <cWR.u•~|. x}vtx~}.x.@@<ws~<% <cWR.wxrw.
|pxyp}p.6u~|.p.rxvptt.~.xt7.
x.tx~t}.~.% <cWR.x}.~srx}v.
p}vt.u~|.@.~.AC3.xw.wt.uprx~}.
|pxyp}p<{xzt.qytrx
t.tuutr.
pq~qts.pt{.trttsx}v.@?.~.A?3.
6Pvt{{.t.p{=;.@HGEJ.[t|qtvt.p}s.
6Pvt{{.t.p{=;[email protected]~{{xt;.@HGG7=.
aqx};.@HFD7=.\tpq~{xt.{t
t{.
cwt.{~.p}s.
pxpq{t.qx~p
px{pqx{x.~u.
u~{{~x}v.~p{.ps|x}xpx~}.|p.qt.
% <cWR.x.st.~.{~.x}.xst<tp|.
vtpt.wp}.wp.~u.% <cWR.p}s.|p.
|~zt;.
pxpx~}.x}.x}sx
xsp{.|~zx}v. r~}xqt.vtp{.~.wt.
qtwp
x~.p}s.ttxt}rt;.x}r~|{tt.
Jwocp!Rjctoceqmkpgvkeu!
wp|pr~{~vxrp{.tuutr.~u.~p{.% <cWR.
pq~x~}.~u.x}wp{ts.|~zt;.p}s.
~.|pxyp}p=.
_wp|pr~zx}txr.~u.|pxyp}p.x}.
|tpq~{x|.x}.{}v.6Wt}x}v.t.p{=;.
_{p|p.r{tpp}rt.~u.% <cWR.
w|p}.x.stt}st}.~}.wt.~t.~u.
p~x|pt.wtpxr.q{~~s.u{~.p.p.
ps|x}xpx~}.p}s.u~|{px~}.6Psp|. @HGEJ.Y~wp}~}.t.p{=;.@HGH7=.Put.
rtpx~}.~u.|~zx}v;.% <cWR.
t}~.
pt.~u.p~x|pt{.HD?.|{>|x}.~.
p}s.\px};[email protected]t{{.t.p{=;.@HGCJ.
{t
t{.str{x}t.xwx}.|x}t.p}s.
vtpt=.apxs.r{tpp}rt.~u.% <cWR.u•~|.
Pvt{{.t.p{=;.@HGE7=.X}sx
xsp{.
x|px{.|~zt.|pxyp}p.p.p.rxvptt. r~}x}t.~.str{x}t.~.pq~.D3.~.@?3. q{~~s.x.x|px{.st.~.tsxxqx~}.
~u.wt.tpz.{t
t{.xwx}.p}.w~.
~.~wt.xt.x}.wt.q~s.pwt.wp}.
6Pvt{{.t.p{=;[email protected]tx.t.p{=;.
~.|tpq~{x|.6Pvt{{.t.p{=;.@HGCJ.
P
px{pq{t.p.jvvr<00yyy/hfc/iqx0Ftwiu0
fghcwnv/jvo!}st.Vxsp}rt.6Sv7=.
@HHApJ.Wtx.t.p{=;.@HHAq7=.
Pvt{{.t.p{=;.@HGE7=.^xst.~u.wt.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11179
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
{x
t;.|tpq~{x|.x}.|~.xt.x.
r~}xstpq{.{~.~.s~t.}~.~rr=.
cwt.t{x|x}px~}.wp{u<{xut.~u.% <cWR.
p}vt.u~|.A?.w~.~.qttt}.@?.p}s.
@B.sp.6W}.p}s.Y~}t;.@HG?7=.
[t|qtvt.t.p{=.6@HF?7.t~ts.wp.wt.
wp{u<{xut.~u.% <cWR.•p}vts.u•~|.AB¤AG.
w~.x}.wtp
.|pxyp}p.t.p}s..
~.E?.~.F?.w~.x}.}p'¶
t.t=.cwt.
{~}v.t{x|x}px~}.wp{u<{xut.~u.% <cWR.x.
st.~.{~.t{tpt.~u.% <cWR.p}s.~wt.
rp}}pqx}~xs.u~|.xt.p}s.
qtt}.|tpq~{x|=.X}prx
t.
rpq~.|tpq~{xt.~u.% <cWR.wp
t.
t|x}p{.wp{u<{x
t.~u.D?.w~.~.E.sp.
~.|~t.p}s.t
t.p.{~}v<t|.|pzt.
x}.x}t.t.u~.|pxyp}p.t=.
\~.~u.wt.pq~qts.% <cWR.s~t.x.
t{x|x}pts.x}.wt.utrt.p}s.pq~.BB3.
x}.x}t=.cwt.v{r~}xst.|tpq~{xt.~u.
% <cWR.x.trtts.p.wt.|py~.x}t.
|tpq~{xt.p{~}v.xw.@G.}~}<.
r~}yvpts.|tpq~{xt.6Pvt{{.t.p{=;.
@HGE7=.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Tgugctej!Uvcvwu!cpf!Vguv!qh!Ewttgpvn{!
Ceegrvgf!Ogfkecn!Wug!hqt!Octklwcpc!
Prr~sx}v.~.wt.WWb;.wtt.pt.
}|t~.w|p}.r{x}xrp{.sxt.xw.
|pxyp}p.x}.wt.d}xts.bpt.}st.
USP<tv{pts.X]S.p{xrpx~}=.at{.
~u.|p{{.r{x}xrp{.t{~p~.sxt.
wp
t.qtt}.q{xwts.x}.wt.|tsxrp{.
{xtpt=.P~
p{.~u.p.w|p}.sv.u~.
|pztx}v;.w~t
t;.x.r~}x}vt}.~}.
USP.p~
p{.~u.p.]t.Sv.
P{xrpx~}.6]SP7.~.p.Qx~{~vxr.
[xrt}t.P{xrpx~}.6Q[P7=.Prr~sx}v.
~.wt.WWb;.wt.USP.wp.}~.p~
ts.
p}.sv.~sr.r~}px}x}v.|pxyp}p.
u~.|pztx}v=.
cwt.WWb.}~ts.wp.p.sv.|p.qt.
u~}s.~.wp
t.p.|tsxrp{.t.x}.
tp|t}.x}.wt.d}xts.bpt.u~.
~t.~u.wt.RbP.xu.wt.sv.|tt.
wt.ux
t.t{t|t}.strxqts.q.wt.STP.
x}.@HHA=.cw~t.ux
t.t{t|t}.pt.q~w.
}trtp.p}s.uuxrxt}.~.tpq{xw.p.
x|p.uprxt.rpt.~u.rt}{.prrtts.
|tsxrp{.t .x}.tp|t}.x}.wt.d}xts.
bpt= .6DF.Ua.@?CHH;.@?D?C.6\prw.AE;.
@HHA77=.cwx.ux
t<t{t|t}.t;.wxrw.wt.
WWb.p}s.STP.wp
t.x{xts.x}.p{{.rw.
p}p{t.u~.|~t.wp}.~.strpst;.wp.
qtt}.wt{s.q.wt.R~.~u.Ptp{=.
[email protected]=Bs.p.@@BD=.cwt.ux
t.t{t|t}.
wp.rwpprtxt.rt}{.prrtts.
|tsxrp{.t .u~.p.sv.pt.||pxts.
wtt.p}s.tp}sts.~}.x}.wt.
sxrx~}.qt{~I.
2/!Vjg!ftwiu!ejgokuvt{!owuv!dg!
mpqyp!cpf!tgrtqfwekdng=!
3/!Vjgtg!owuv!dg!cfgswcvg!uchgv{!
uvwfkgu=!
4/!Vjgtg!owuv!dg!cfgswcvg!cpf!ygnn.!
eqpvtqnngf!uvwfkgu!rtqxkpi!ghhkece{=!
5/!Vjg!ftwi!owuv!dg!ceegrvgf!d{!
swcnkhkgf!gzrgtvu=!cpf!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
6/!Uekgpvkhke!gxkfgpeg!owuv!dg!ykfgn{!
cxckncdng/!
X}.x.t
xt.6WWb;.A?@D7;.wt.WWb.
t
p{pts.wt.ux
t.t{t|t}.xw.ttr.
~.wt.rt}{.p
px{pq{t.ttprw.u~.
|pxyp}p=.cwt.WWb.r~}r{sts.wp.
|pxyp}p.s~t.}~.|tt.p}.~u.wt.ux
t.
t{t|t}¥p{{.~u.wxrw.|.qt.
st|~}pts.~.ux}s.wp.p.sv.wp.p.
rt}{.prrtts.|tsxrp{.t= .P.
qxtu.||p.~u.wt.WWb .t
p{px~}.
x.~
xsts.qt{~=.
Gngogpv!$2<!Vjg!ftwiu!ejgokuvt{!
owuv!dg!mpqyp!cpf!tgrtqfwekdng/!
Vjg!uwduvcpegu!ejgokuvt{!owuv!dg!
uekgpvkhkecnn{!guvcdnkujgf!vq!rgtokv!kv!vq!
dg!tgrtqfwegf!kpvq!fqucigu!yjkej!ecp!
dg!uvcpfctfk|gf/!Vjg!nkuvkpi!qh!vjg!
uwduvcpeg!kp!c!ewttgpv!gfkvkqp!qh!qpg!qh!
vjg!qhhkekcn!eqorgpfkc-!cu!fghkpgf!d{!
ugevkqp!312)l*!qh!vjg!Hqqf-!Ftwi!cpf!
Equogvke!Cev-!32!W/U/E/!432)l*-!ku!
uwhhkekgpv!igpgtcnn{!vq!oggv!vjku!
tgswktgogpv/!68!HT!215;;-!21617!
)Octej!37-!2;;3*/!
\pxyp}p;.p.stux}ts.x}.wt.txx~};.
x}r{st.p{{.Ecppcdku!px}=.6U~.
~t.~u.wt.RbP;.|pxyp}p.
x}r{st.p{{.trxt.~u.wt.vt}.
Ecppcdku-!x}r{sx}v.p{{.px}.
wttx} 7=.Qpts.~}.wt.stux}xx~}.~u.
|pxyp}p.x}.wt.txx~};.wt.rwt|x.
~u.|pxyp}p.x.}~.t~srxq{t.rw.
wp.p.p}spsxts.s~t.rp}.qt.rtpts=.
Rwt|xrp{.r~}xt}.x}r{sx}v.% <.
cWR.p}s.~wt.rp}}pqx}~xs.
p.
xv}xuxrp}{.x}.|pxyp}p.p|{t.
stx
ts.u~|.sxuutt}.px}.
6Pt}sx}~.t.p{=;.A?@@7=.P.p.t{;.
wtt.x{{.qt.xv}xuxrp}.sxuutt}rt.x}.
put;.qx~{~vxrp{;.wp|pr~{~vxrp{;.p}s.
~xr~{~vxrp{.pp|tt.p|~}v.wt.
px~.|pxyp}p.p|{t=.St.~.wt.
pxpx~}.~u.wt.rwt|xrp{.r~|~xx~}.x}.
|pxyp}p.p|{t;.x.x.}~.~xq{t.~.
t~srt.p.p}spsxts.s~t.wt}.
r~}xstx}v.p{{.px}.~vtwt=.cwt.
WWb.s~t.ps
xt.wp.xu.p.trxuxr.
Ecppcdku!px}.x.r{x
pts.p}s.
~rtts.}st.r~}~{{ts.r~}sxx~};.
wt.{p}.rwt|x.|p.qt.r~}xt}.
P{w~vw.wt.RbP.stux}xx~}.~u.|pxyp}p.
tut.~}{.~.wt.trxt.Ecppcdku!ucvkxc!N/-!
utstp{.r~.wp
t.r~}xt}{.{ts.wp.p{{.
trxt.~u.wt.vt}.rp}}pqx.pt.x}r{sts.x}.wx.
stux}xx~}=.Ugg!Wpkvgf!Uvcvgu!
=.Mgnn{-!DAF.U=As.HE@;.
HEB¤HEC.6Hw.Rx[email protected]~{{trx}v.p}s.tp|x}x}v.
rpt7=.cwt.bx}v{t.R~}
t}x~}.6pxr{t.@;.p=.@6r77.
{xztxt.stux}t.wt.rp}}pqx.{p} .~.|tp}.p}.
{p}.~u.wt.vt}.Rp}}pqx= .P.t{px}ts.pq~
t.
x}.wt.pprw|t}.x{ts._t{x|x}p.]~t.
atvpsx}v.ctp.R~}xstpx~}; [email protected]=b=R=.
G@@6s76@7.~
xst.wp;.wtt.p.sv.x.qytr.~.
r~}~{.}st.wt.bx}v{t.R~}
t}x~};.wt.STP.
Ps|x}xp~.|.r~}~{.wt.sv.}st.wt.
rwts{t.wt.stt|.|~.p~xpt.~.rp.~.
rw.tp.~q{xvpx~};.xw~.tvps.~.wt.
ux}sx}v.txts.q[email protected]=b=R=.G@@6p7.~[email protected]}s.
xw~.tvps.~.wt.~rtst.trxqts.q.A@.
d=b=R=.G@@6p7.p}s.6q7=.
RQ!11111
Hto!1117;
Hov!5812
Uhov!5813
64866!
t}~vw.~.stx
t.t~srxq{t.p}s.
p}spsxts.s~t=.
Gngogpv!$3<!Vjgtg!owuv!dg!cfgswcvg!
uchgv{!uvwfkgu/!
Vjgtg!owuv!dg!cfgswcvg!
rjctoceqnqikecn!cpf!vqzkeqnqikecn!
uvwfkgu-!fqpg!d{!cnn!ogvjqfu!tgcuqpcdn{!
crrnkecdng-!qp!vjg!dcuku!qh!yjkej!kv!
eqwnf!hcktn{!cpf!tgurqpukdn{!dg!
eqpenwfgf-!d{!gzrgtvu!swcnkhkgf!d{!
uekgpvkhke!vtckpkpi!cpf!gzrgtkgpeg!vq!
gxcnwcvg!vjg!uchgv{!cpf!ghhgevkxgpguu!qh!
ftwiu-!vjcv!vjg!uwduvcpeg!ku!uchg!hqt!
vtgcvkpi!c!urgekhke-!tgeqipk|gf!fkuqtfgt/!
68!HT!215;;-!21617!)Octej!37-!2;;3*/!
cwt.WWb.pts.wp.wtt.pt.}~.
pstpt.put.sxt.~}.|pxyp}p=.
P.x}sxrpts.x}.wtx.t
p{px~}.~u.
T{t|t}.1@;.wt.r~}xstpq{t.
pxpx~}.
x}.wt.rwt|x.~u.|pxyp}p.
r~|{xrpt.wt.put.t
p{px~}=.cwt.
WWb.r~}r{sts.wp.|pxyp}p.s~t.}~.
pxu.T{t|t}.1A.u~.wp
x}v.pstpt.
put.sxt.rw.wp.|tsxrp{.p}s.
rxt}xuxr.tt.|p.r~}r{st.wp.x.x.
put.u~.tpx}v.p.trxuxr.px{|t}=.
Gngogpv!$4<!Vjgtg!owuv!dg!cfgswcvg!
cpf!ygnn.eqpvtqnngf!uvwfkgu!qh!ghhkece{/!
Vjgtg!owuv!dg!cfgswcvg-!ygnn.!
eqpvtqnngf-!ygnn.fgukipgf-!ygnn.!
eqpfwevgf!cpf!ygnn.fqewogpvgf!
uvwfkgu-!kpenwfkpi!enkpkecn!
kpxguvkicvkqpu-!d{!gzrgtvu!swcnkhkgf!d{!
uekgpvkhke!vtckpkpi!cpf!gzrgtkgpeg!vq!
gxcnwcvg!vjg!uchgv{!cpf!ghhgevkxgpguu!qh!
ftwiu-!qp!vjg!dcuku!qh!yjkej!kv!eqwnf!dg!
hcktn{!cpf!tgurqpukdn{!eqpenwfgf!d{!
uwej!gzrgtvu!vjcv!vjg!uwduvcpeg!yknn!
jcxg!vjg!kpvgpfgf!ghhgev!kp!vtgcvkpi!c!
urgekhke-!tgeqipk|gf!fkuqtfgt/!68!HT!
215;;-!21617!)Octej!37-!2;;3*/!
P.x}sxrpts.x}.wt.WWb .t
xt.~u.
|pxyp}p.6WWb;.A?@D7;.wtt.pt.}~.
pstpt.~.t{{<r~}~{{ts.sxt.wp.
~
t.|pxyp}p .tuuxrpr=.cwt.USP.
x}stt}st}{.t
xtts.6USP;.A?@D7.
q{xr{.p
px{pq{t.r{x}xrp{.sxt.~}.
|pxyp}p.q{xwts.x~.~.Utqp.
A?@B.~.stt|x}t.xu.wtt.tt.
p~xpt.sxt.~.stt|x}t.
|pxyp}p .tuuxrpr.6{tpt.tut.~.
USP;[email protected]}s.WWb;[email protected]~•.|~•t.
stpx{7=.Put.t
xt;.wt.USP.
stt|x}ts.wp.~.~u.wt.xst}xuxts.
pxr{t;.x}r{sx}v.w~t.xst}xuxts.
w~vw.p.tprw.~u.qxq{x~vpwxr.
tutt}rt.p}s.DEE.pqpr.{~rpts.~}.
_q\ts;.@@.sxt.|t.wt.c!rtkqtk!
t{trx~}.rxtxp;.x}r{sx}v.{prtq~.
r~}~{.p}s.s~q{t<q{x}sx}v=.USP.p}s.
WWb.rxxrp{{.t
xtts.tprw.~u.wt.@@.
sxt.~.stt|x}t.xu.wt.sxt.|t.
prrtts.rxt}xuxr.p}sps=.USP.p}s.
WWb.r~}r{sts.wp.wtt.sxt.s~.
}~.rt}{.~
t.tuuxrpr.~u.
|pxyp}p .u~.p}.wtptxr.
x}sxrpx~}.st.~.{x|xpx~}.x}.wt.
s.stxv}=.cwt.WWb.x}sxrpts.wp.
wtt.sxt.r~{s.qt.ts.p.~~u.~u.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64867!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
r~}rt.sxt;.~
xsx}v.t{x|x}p.
t
xst}rt.~}.p.~~ts.w~wtx.
x}
~{
x}v.p.sv .tuutr=.
Gngogpv!$5<!Vjg!ftwi!owuv!dg!
ceegrvgf!d{!swcnkhkgf!gzrgtvu/!
]C_!eqpugpuwu!qh!vjg!pcvkqpcn!
eqoowpkv{!qh!gzrgtvu-!swcnkhkgf!d{!
uekgpvkhke!vtckpkpi!cpf!gzrgtkgpeg!vq!
gxcnwcvg!vjg!uchgv{!cpf!ghhgevkxgpguu!qh!
ftwiu-!ceegrvu!vjg!uchgv{!cpf!
ghhgevkxgpguu!qh!vjg!uwduvcpeg!hqt!wug!kp!
vtgcvkpi!c!urgekhke-!tgeqipk|gf!fkuqtfgt/!
C!ocvgtkcn!eqphnkev!qh!qrkpkqp!coqpi!
gzrgtvu!rtgenwfgu!c!hkpfkpi!qh!
eqpugpuwu/!68!HT!215;;-!21617!)Octej!
37-!2;;3*/!
cwt.WWb.r~}r{sts.wp.wtt.x.
rt}{.}~.t
xst}rt.~u.p.r~}t}.
p|~}v.p{xuxts.tt.wp.|pxyp}p.
x.put.p}s.tuutrx
t.x}.tpx}v.p.trxuxr.
p}s.tr~v}xts.sx~st=.cwt.WWb.
x}sxrpts.wp.|tsxrp{.prxx~}t.
w~.pt.}~.tt.x}.t
p{px}v.sv.
rp}}~.qt.r~}xstts.p{xuxts.tt.
6WWb;[email protected].@?CHH;.@?D?D7=.
Uwt;.wt.WWb.}~ts.wp.wt.A??H.
P|txrp}.\tsxrp{.P~rxpx~}.6P\P7.
t~.t}x{ts;.dt.~u.Rp}}pqx.u~.
\tsxrx}p{._~t .s~t.}~.r~}r{st.
wp.wtt.x.p.rt}{.prrtts.
|tsxrp{.t.u~.|pxyp}p=.WWb.p{~.
~x}ts.~.wp.pt<{t
t{.|tsxrp{.
|pxyp}p .{p.s~.}~.~
xst.
t
xst}rt.~u.rw.p.r~}t}.p|~}v.
p{xuxts.tt=.
Gngogpv!$6<!Vjg!uekgpvkhke!gxkfgpeg!
owuv!dg!ykfgn{!cxckncdng/!
Kp!vjg!cdugpeg!qh!PFC!crrtqxcn-!
kphqtocvkqp!eqpegtpkpi!vjg!ejgokuvt{-!
rjctoceqnqi{-!vqzkeqnqi{-!cpf!
ghhgevkxgpguu!qh!vjg!uwduvcpeg!owuv!dg!
tgrqtvgf-!rwdnkujgf-!qt!qvjgtykug!ykfgn{!
cxckncdng-!kp!uwhhkekgpv!fgvckn!vq!rgtokv!
gzrgtvu-!swcnkhkgf!d{!uekgpvkhke!vtckpkpi!
cpf!gzrgtkgpeg!vq!gxcnwcvg!vjg!uchgv{!
cpf!ghhgevkxgpguu!qh!ftwiu-!vq!hcktn{!cpf!
tgurqpukdn{!eqpenwfg!vjg!uwduvcpeg!ku!
uchg!cpf!ghhgevkxg!hqt!wug!kp!vtgcvkpi!c!
urgekhke-!tgeqipk|gf!fkuqtfgt/!68!HT!
215;;-!21617!)Octej!37-!2;;3*/!
cwt.WWb.r~}r{sts.wp.wt.rt}{.
p
px{pq{t.spp.p}s.x}u~|px~}.~}.
|pxyp}p.x.}~.uuxrxt}.~.p{{~.
rxt}xuxr.rx}.~u.wt.rwt|x;.
wp|pr~{~v;.~xr~{~v;.p}s.
tuutrx
t}t=.X}.pxr{p;.rxt}xuxr.
t
xst}rt.st|~}px}v.wt.rwt|x.
~u.p.trxuxr.Ecppcdku!px}.wp.r~{s.
~
xst.p}spsxts.p}s.t~srxq{t.
s~t.x.}~.p
px{pq{t=.
ewttgpv-!oqfgtp!ceegrvgf!uvcpfctfu!hqt!
yjcv!eqpuvkvwvgu!ogfkekpg/ .6Twxqx.Q;.
pvt.@B7=.^}.pvt.B.~u.wt.r~
t.{tt.
~u.wt.txx~};.wt.txx~}t.pts;.
Vjg!Cogtkecp!ogfkecn!eqoowpkv{!
uwrrqtvu!tguejgfwnkpi-!cpf!vjgtg!ctg!
uchg!rjctoce{.dcugf!ogvjqfu!vq!
fkurgpug!ogfkecn!ecppcdku/ .
Uwt|~t;.wt.pts.wp.]k_p!
311;-!vjg!Cogtkecp!Ogfkecn!Cuuqekcvkqp!
)COC*!tgxgtugf!kvu!gctnkgt!rqukvkqp!vjcv!
uwrrqtvgf!]u_ejgfwng!K!encuukhkecvkqp!qh!
ecppcdku/!Vjg!COC!pqy!uwrrqtvu!
kpxguvkicvkqp!cpf!enkpkecn!tgugctej!qh!
ecppcdku!hqt!ogfkekpcn!wug-!cpf!wtigf!
vjg!hgfgtcn!iqxgtpogpv!vq!tgcuuguu!vjg!
]u_ejgfwng!K!encuukhkecvkqp/!Vjg!
Cogtkecp!Eqnngig!qh!Rj{ukekcpu!]CER_!
tgegpvn{!gzrtguugf!ukoknct!uwrrqtv/ .X}.
pssxx~};.wt.}~t.wp.wt.X}xt.~u.
\tsxrx}t.6X^\7.p{~.s~r|t}ts.wt.
rxt}xuxr.qpx.p}s.wtptxr.tuutr.~u.
rp}}pqx.6Twxqx.Q;.pvt.@B7=.
cwt.STP.}~t.wp.wt.pt|t}.q.
wt.rxts.~vp}xpx~}.6P\P;.PR_;.
X^\7.~.|~t.ttprw.x}~.wt.
~t}xp{.|tsxrp{.~txt.p~rxpts.
xw.|pxyp}p=.cwt.WWb.sxs.}~.ux}s.
wp.wt.pt|t}.q.wtt.
~vp}xpx~}.~
xst.t
xst}rt.
~x}v.p.r~}r{x~}.wp.pstpt.
put.sxt.p}s.pstpt;.t{{<.
r~}~{{ts.tuuxrpr.sxt.st|~}pt.
wt.put.p}s.tuuxrpr.~u.|pxyp}p.
6WWb;.A?@D7=.cwt.P\P .~uuxrxp{.~{xr.
~}.|tsxrx}p{.t.~u.|pxyp}p.x.p.
u~{{~I.Qwt!COC!wtigu!vjcv!
octklwcpcu!uvcvwu!cu!c!hgfgtcn!
]u_ejgfwng!K!eqpvtqnngf!uwduvcpeg!dg!
tgxkgygf!ykvj!vjg!iqcn!qh!hceknkvcvkpi!vjg!
eqpfwev!qh!enkpkecn!tgugctej!cpf!
fgxgnqrogpv!qh!ecppcdkpqkf.dcugf!
ogfkekpgu-!cpf!cnvgtpcvkxg!fgnkxgt{!
ogvjqfu/!Vjku!ujqwnf!pqv!dg!xkgygf!cu!
cp!gpfqtugogpv!qh!uvcvg.dcugf!ogfkecn!
ecppcdku!rtqitcou-!vjg!ngicnk|cvkqp!qh!
octklwcpc-!qt!vjcv!uekgpvkhke!gxkfgpeg!qp!
vjg!vjgtcrgwvke!wug!qh!ecppcdku!oggvu!
vjg!ewttgpv!uvcpfctfu!hqt!c!rtguetkrvkqp!
ftwi!rtqfwev/ .6P\P;.A??H7=.
cwt.STP.uwt.}~t.wp.wt.A?@B.
P\P.W~t.~u.St{tvpt.t~.pt.
wp;.ecppcdku!ku!c!fcpigtqwu!ftwi!cpf!
cu!uwej!ku!c!rwdnke!jgcnvj!eqpegtp/ .
6P\P;.A?@B7=.X}.A??G;.wt.PR_.
x}sxrpts.wp.hwtvjgt!tgugctej!ku!
pggfgf!vq!eqorctg!ecppcdkpqkfu!
ghhkece{!cpf!uchgv{!ykvj!ewttgpv!
vtgcvogpvu/ .6PR_;.A??G7=.cwt.PR_.
pts.wp;.CER!wtigu!cp!gxkfgpeg.!
dcugf!tgxkgy!qh!octklwcpcu!uvcvwu!cu!c!
Rgvkvkqpgtu!Oclqt!Eqoogpvu!kp!Tgncvkqp! ]u_ejgfwng!K!eqpvtqnngf!uwduvcpeg!vq!
vq!Hcevqt!4!cpf!vjg!Iqxgtpogpvu!
fgvgtokpg!yjgvjgt!kv!ujqwnf!dg!
Tgurqpugu!
tgencuukhkgf!vq!c!fkhhgtgpv!uejgfwng/!Vjku!
[email protected].txx~}t.x}sxrpt.wp.wtt. tgxkgy!ujqwnf!eqpukfgt!vjg!uekgpvkhke!
x.|tsxrp{.~.p}s.prrtp}rt.u~.
hkpfkpiu!tgictfkpi!octklwcpcu!uchgv{!
wt.|tsxrp{.t.~u.|pxyp}p.p}s.pts. cpf!ghhkece{!kp!uqog!enkpkecn!eqpfkvkqpu!
wp.]e_cppcdku!jcu!dggp!ceegrvgf!d{!
cu!ygnn!cu!gxkfgpeg!qp!vjg!jgcnvj!tkumu!
vjg!ogfkecn!eqoowpkv{!cu!oggvkpi!vjg!
cuuqekcvgf!ykvj!octklwcpc!
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11181
Hov!5812
Uhov!5813
eqpuworvkqp-!rctvkewnctn{!kp!kvu!etwfg!
uoqmgf!hqto .6PR_;.A??G7=.cwt.X^\;.
r~}xt}.xw.~wt.x}.wt.|tsxrp{.
r~||}x;.t}s~t.uwt.sxt.
x}~.wt.~t}xp{.wtptxr.t.~u.
|pxyp}p;.q.sxs.}~.ps
~rpt.u~.
|tsxrx}p{.t.xw~.uwt.tx}v.
6X^\;.A??H7=.
P.stpx{ts.x}.wt.WWb.t
xt;.x}.
~st.u~.p.sv.~.qt.u~}s.~.wp
t.p.
rt}{.prrtts.|tsxrp{.t; .x.
|.qt.prrtts.q.p{xuxts.tt=.
cwtt.x.}~.t
xst}rt.wp.wtt.x.p.
r~}t}.p|~}v.p{xuxts.tt.wp.
|pxyp}p.x.put.p}s.tuutrx
t.u~.t.x}.
tpx}v.p.trxuxr;.tr~v}xts.sx~st=.
6A7.cwt.txx~}t.r{px|.wp;.Vjg!
ejgokuvt{!qh!ecppcdku!ku!mpqyp!cpf!
tgrtqfwekdng .6Twxqx.Q;.pvt.E7.p}s.
pgygt!ogfkekpcn!uvtckpu!qh!ecppcdku!
ctg!nqygt!kp!VJE!cpf!jkijgt!kp!vjg!pqp.!
ru{ejqcevkxg-!oqtg!vjgtcrgwvke!
ecppcdkpqkfu-!uwej!cu!EDF-!cpf!EDP/!
Vjgug!eqorqwpfu!hwtvjgt!kortqxgf!vjg!
ghhkece{!qh!ecppcdku/ .6Twxqx.Q;.pvt.
@?7=.
P.x}sxrpts.q.wt.WWb;.wt.
txx~}t.stux}ts.|pxyp}p.~.x}r{st.
p{{.Ecppcdku!px}=.P.rw;.wt.
rwt|x.~u.|pxyp}p.x.}~.
t~srxq{t.rw.wp.p.p}spsxts.
s~t.rp}.qt.rtpts=.Rwt|xrp{.
r~}xt}.x}r{sx}v.% <cWR.p}s.
~wt.rp}}pqx}~xs.
p.xv}xuxrp}{.x}.
sxuutt}.|pxyp}p.p|{t.6Pt}sx}~.
t.p{=;.A?@@7=.Uwt|~t;.wt.WWb.rxts.
p.q{xwts.t~.wp.x}sxrpt.wp.
}t.qp}rt.x}.|pxyp}p.pt.
r~}x}p{{.qtx}v.rwpprtxts.
6_~{{p~.t.p{=;.A?@@7=.Xu.wtt.x.
xv}xuxrp}.
pxp}rt.x}.wt.rwt|xrp{.
r~|~xx~}.~u.|pxyp}p.qttt}.
p|{t;.x.x.}~.~xq{t.u~.wt.
rwt|x.~.qt.t~srxq{t=.
Qtrpt.wt.txx~}.stux}t.
|pxyp}p.p.x}r{sx}v.p{{.r{x
pts.
px};.wt.STP.qt{xt
t.wp.wt.cWR.
p}s.RQS.{t
t{.~u.trxuxr.px}.x.}~.
t{t
p}.~.wx.r~}xstpx~}=.X}.upr;.
wt.p
tpvt.% <cWR.r~}t}.x}.
|pxyp}p.wp.tpsx{.xt}.u~|.@HHD.
[email protected].t~ts.q.wt.d}x
tx.~u.
\xxxx._~t}r.\~}x~x}v._~ytr;.
p.tt}ts.x}.Upr~.@=.X}.@HHD;.wt.% <.
cWR.r~}t}.p.C3.~}.p
tpvt.p}s.q.
A?@D;.wt.p
tpvt.r~}t}.~u.cWR.wps.
xt}.~.@@=A3.~
t.p.A?.tp.tx~s=.X}.
wt.p|t.x|t.tx~s;.RQS.p}s.RQ].
trt}pvt.wp
t.p}vts.u~|.?=@D3.~.
?=E?3.~}.p
tpvt=.
cwt.STP.p{~.}~t.pt|t}.x}.wt.
txx~}t .s~r|t}.wp.~.wt.
r~}r{x~}.tprwts.q.STP.p}s.WWb.
wp.wt.rwt|x.~u.|pxyp}p.p.
q~ps{.stux}ts.q.wt.txx~}t.x.}~.
t~srxq{t.~.t{{<stux}ts=.U~.
tp|{t;.wt.txx~}t.prz}~{tsvt.
wp.Rp}}pqx.x.p.r~|{t.{p};.xw.
t
tp{.qt.~u.rp}}pqx= .6Twxqx.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
Q;.pvt.E7=.cwt.txx~}t.p{~.
prz}~{tsvt.wp.wt.px~.~u.wt.
px~.rp}}pqx}~xs.sxuut.prr~sx}v.~.
wt.{p}.px};.p}s;.~.~|t.tt};.
w~.wt.{p}.x.v~}= .6Twxqx.Q;.
•pvt.@A7=.
6B7.cwt.txx~}t.pts.x}.Twxqx.Q;.
pvt.G;.wp.]q_xgtcnn-!vjg!44!eqorngvgf!
cpf!rwdnkujgf!Cogtkecp!eqpvtqnngf!
enkpkecn!vtkcnu!ykvj!ecppcdku!jcxg!
uvwfkgf!kvu!uchgv{-!tqwvgu!qh!
cfokpkuvtcvkqp-!cpf!wug!kp!eqorctkuqp!
ykvj!rncegdqu-!uvcpfctf!ftwiu-!cpf!kp!
uqog!ecugu!ftqpcdkpqn!/ / / - .p}s.
uwt.rxts.p.t|pxr.t
xt.q.
fp}v.t.p{=.6A??G7;.wp.t
p{pts.AB.
p}s~|xts.r~}~{{ts.xp{.p}s.G.
~qt
px~}p{.sxt;.px}v.wp;.]q_h!
cnn!vjg!cfxgtug!gxgpvu!tgrqtvgf-!;8!
rgtegpv!ygtg!eqpukfgtgf!pqv!ugtkqwu-!
ykvj!vjg!oquv!eqooqpn{!tgrqtvgf!
fk||kpguu/ .
cwt.txx~}t.p{~.rxts.x}.Twxqx.
Q;.pvt.G;.Vjgtg!jcu!dggp!c!nqpi.vgto-!
rtqurgevkxg-!hgfgtcnn{!hwpfgf!ecppcdku!
enkpkecn!uvwf{!lqkpvn{!cfokpkuvgtgf!d{!
Pcvkqpcn!Kpuvkvwvg!qp!Ftwi!Cdwug!
)PKFC*!cpf!HFC/!Vjku!uvwf{!jcu!dggp!
twppkpi!hqt!qxgt!41!{gctu!ykvjqwv!cp{!
fgoqpuvtcdng!cfxgtug!qwveqogu!tgncvgf!
vq!ejtqpke!ogfkekpcn!ecppcdku!wug/ .
P.rxts.x}.wt.WWb.tr~||t}spx~}.
s~r|t}.6WWb;.A?@D7;.wt.USP.
r~}srts.x.~}.t
p{px~}.~u.wt.
q{xwts.r{x}xrp{.sxt.~}.wt.
|tsxrp{.p{xrpx~}.~u.|pxyp}p.x~.
~.Utqp[email protected];.A?@D7=.Uwt.
stpx{.~}.wt.USP.t
xt.rp}.qt.u~}s.
x}.wt.q{xwts.t~.6USP;.A?@D7=.
Qpts.~}.wt.p}p{x;.@@.sxt.tt.
t
p{pts.uwt.p}s.wt.USP.
r~}r{sts.wp.}~}t.~u.wtt.sxt.
|tt.wt.rxtxp.txts.q.wt.USP.
~.stt|x}t.xu.|pxyp}p.x.put.p}s.
tuutrx
t.x}.trxuxr.wtptxr.ptp= .
6•pvt.EJ.USP;.A?@D7=.
cwt.STP.wp.t
xtts.wt.t|pxr.
t
xt.q.fp}v.t.p{=.6A??G7.p}s.}~t.
wp.|~.~u.wt.sxt.x}r{sts.x}.wt.
t
xt.tt.}wtxr.rp}}pqx}~xs.
|tsxrx}t.6g/i/!s•~}pqx}~{7.~•.
rp}}pqx}~xs.tpr.6g/i/!bpx
t 7J.
wtt.t.~u.sxt.tt.tr{sts.x}.
wt.USP.t
xt.p.wt.p}p{x.u~rts.
~{t{.~}.}pp{.u~|.~u.|pxyp}p.
6USP;.A?@D7=.fp}v.t.p{=.6A??G7.
r~}r{sts.wp.v~~s.put.p}s.
tuuxrpr.spp.~}.|~zts.rp}}pqx.pt.
vt}{.}ttsts= .
fxw.ttr.~.wt.B?<tp.s.
rxts.q.wt.txx~}t.6a~.t.p{=;.
A??@7.~}.pvt.G.~u.Twxqx.Q;.x.w~{s.
qt.r{pxuxts.wp.wt.tutt}rts.s.
p.}~.y~x}{.ps|x}xtts.q.]XSP.
p}s.wt.USP=.P.xw.~wt.r{x}xrp{.
sxt;.p}.X]S.p{xrpx~}.p.
p~
ts.q.wt.USP.p}s.|pxyp}p.
p.{xts.q.]XSP=.cwt.pw~.
t
p{pts.~}{.G.pxt}.~
t.wx.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
tx~s;.~u.wxrw.~}t.pxt}.sxts=.
fwx{t.wt.ux}sx}v.rxts.q.wt.
txx~}t.p}s.pw~.6g/i/!}~.ps
tt.
~r~|t.xw.{~}v.t|.|pxyp}p.t7.
pt.x}u~|px
t;.r~}r{x~}.~}.{~}v<.
t|.t.~u.|pxyp}p.rp}}~.qt.p{xts.
~.wt.vt}tp{.~{px~}=.
64868!
vpst;.ttrx
t{;.t~ts.x}v.
|pxyp}p.x}.wt.p.tp=.P.}|qt.~u.
wxvw.rw~~{.st}.t~ts.spx{.t.
x}.wt.p.|~}w;.x}r{sx}v.@=@3;.
B=?3;.p}s.E=?3.~u.Gw;.@?w;.p}s.@Aw.
vpst;.ttrx
t{=.
cwt.t
p{t}rt.~u.|pxyp}p.t.p}s.
pqt.x.p{~.x}sxrpts.q.rx|x}p{.
Hcevqt!5<!Kvu!Jkuvqt{!cpf!Ewttgpv!
x}
txvpx~}.u~.wxrw.sv.t
xst}rt.
Rcvvgtp!qh!Cdwug!
p.p}p{ts.x}.utstp{;.pt;.p}s.{~rp{.
\pxyp}p.r~}x}t.~.qt.wt.|~.
u~t}xr.{pq~p~xt;.p.sxrts.pq~
t.
xst{.ts.x{{xrx.sv=.X}.A?@B;.p}.
x}.Upr~.@=.cwt.]px~}p{.U~t}xr.
tx|pts.AC=E.|x{{x~}.P|txrp}.pvt.
[pq~p~.bt|.6]U[Xb7;.p.STP.
@A.~.~{st.tt.rt}.6p.|~}w7.
~vp|;.t|pxrp{{.r~{{tr.sv.
x{{xrx.sv.t=.^u.w~t;.@H=G.|x{{x~}. xst}xuxrpx~}.t{.p}s.p~rxpts.
tt.rt}.6p.|~}w7.|pxyp}p.
x}u~|px~}.u~|.sv.rpt.q|xts.
t=.P.~u.A?@B;.p}.tx|pts.@@C=F.
~.p}s.p}p{ts.q.utstp{;.pt;.p}s.
|x{{x~}.P|txrp}[email protected]}s.~{st.wps. {~rp{.u~t}xr.{pq~p~xt=.]U[Xb.spp.
ts.|pxyp}p.~.wpwxw.x}.wtx.
w~.wp.|pxyp}p.p.wt.|~.
{xutx|t.p}s.BB=?.|x{{x~}.wps.ts.x.x}. utt}{.xst}xuxts.sv.u~|.Yp}p.
wt.p.tp=.
A??@.w~vw.Strt|qt.A?@C=.X}.A?@C;.
Prr~sx}v.~.wt.]bSdW.tx|pt;.
|pxyp}p.prr~}ts.u~.AH=B3.
B=?.|x{{x~}.t~{t.pvt.@A.~.~{st.ts. 6CBA;HGH7.~u.p{{.sv.twxqx.x}.]U[Xb=.
p}.x{{xrx.sv.u~.wt.ux.x|t.x}.A?@C=.
cwt.wxvw.r~}|x~}.~u.|pxyp}p.x.
\pxyp}p.x}xxpt.~p{ts.A=E.|x{{x~}.x}. qtx}v.ut{ts.q.x}rtpx}v.p|~}.~u.
A?@C=.]tp{.wp{u.6CE=G37.~u.wt.A=E.
s~|txrp{{.v~}.|pxyp}p.p.t{{.p.
|x{{x~}.}t.t.tt.{t.wp}.@G.
x}rtpts.p|~}.~u.u~txv}.~rt.
tp.~u.pvt=.X}.A?@C;.|pxyp}p.p.
|pxyp}p.qtx}v.x{{xrx{.|vv{ts.x}~.
ts.q.GA=A3.~u.rt}.6p.|~}w7.
wt.d}xts.bpt=.X}.A?@C;.wt.S~|txr.
x{{xrx.sv.t=.X}.A?@C;.p|~}v.p.
Rp}}pqx.Tpsxrpx~}.p}s.btx~}.
tp.|pxyp}p.t.pvt.@A.~.~{st;.
_~vp|.6SRT>b_7.t~ts.wp.
@G=D3.ts.|pxyp}p.~}.B??.~.|~t.
B;H?C;A@B.{p}.tt.tpsxrpts.x}.
sp.xwx}.wt.t
x~.@A.|~}w=.
~s~~.rp}}pqx.r{x
px~}.ptp.
cwx.p}{pt.x}~.E=D.|x{{x~}.t~{t.
r~|pts.~.A;DHF;FHG.x}.A???;.p.
x}v.|pxyp}p.~}.p.spx{.~.p{|~.
w~}.pq~
t.x}.cpq{t.B=.bxv}xuxrp}.
spx{.qpx.~
t.p.@A<|~}w.tx~s;.p.
p}xxt.~u.|pxyp}p.tt.p{~.
xv}xuxrp}.x}rtpt.u~|.wt.B=@.|x{{x~}. tpsxrpts.u~|.x}s~~.r{x
px~}.
spx{.~.p{|~.spx{.t.x}.A??E.p}s.
~tpx~}=.cwtt.tt.BHE;EA?.x}s~~.
u~|.wt.D=F.|x{{x~}.x}.y.wt.t
x~. {p}.tpsxrpts.x}[email protected]~|pts.~.
tp=.X}.A?@C;.p|~}v.p.|~}w.
A@F;@?D.tpsxrpts.x}.A???=.P.w~}.
|pxyp}p.t;.C@=E3.6H=A.|x{{x~}.
x}.cpq{t.A.pq~
t;.x}.A?@C;.wt.]px~}p{.
t~{t7.ts.wt.sv.~}.A?.~.|~t.
btxt.bt|.6]bb7.t~ts.txt.
sp.x}.wt.p.|~}w;.p.xv}xuxrp}.
~u.@;FEF;[email protected].~u.|pxyp}p=.
x}rtpt.u~|.wt.G=@.|x{{x~}.x}.A?@B=.
\pxyp}p.x.p{~.wt.x{{xrx.sv.xw. Rgvkvkqpgtu!Oclqt!Eqoogpvu!kp!Tgncvkqp!
vq!Hcevqt!5!cpf!vjg!Iqxgtpogpvu!
wt.wxvwt.}|qt.~u.p.tp.
Tgurqpugu!
stt}st}rt.~.pqt.x}.wt.d=b=.
[email protected].txx~}t.x}sxrpts.wp.wt.
~{px~}=.Prr~sx}v.~.wt.A?@C.
]bSdW.t~;.~u.wt.F=@.|x{{x~}.
wx~.p}s.rt}.pt}.~u.pqt.x.
t~}.pvts.@A.~.~{st.w~.tt.
sxuuxr{.~.tx|pt.x}rt.c!nctig!
rgtegpvcig!qh!Wpkvgf!Uvcvgu!ekvk|gpu!
r{pxuxts.xw.x{{xrx.sv.stt}st}rt.
~.pqt;.C=A.|x{{x~}.~u.wt|.pqts.~. wp
t.ts.|pxyp}p.p.{tp.~}rt.x}.
tt.stt}st}.~}.|pxyp}p.
wtx.{xutx|t.p}s.~|t.tx|pt.wp
t.
x}sxrpts.wp.qxgt!51!rgtegpv!qh!vjg!
6ttt}x}v.DH=?3.~u.p{{.w~t.
r{pxuxts.xw.x{{xrx.sv.stt}st}rt.
pcvkqp!jcu!vtkgf!vjg!rncpv/!Uwt;.wt.
~.pqt.p}s.@=E3.~u.wt.~p{.d=b=.}~}<. txx~}t.pts.wp.vt{kpi!octklwcpc!
x}xx~}p{xts.~{px~}.pvts.@A.~. qpeg!ujqwnf!pqv!dg!eqphwugf!ykvj!c!
jgcnvj!rtqdngo-!ngv!cnqpg!c!fkcipquku!qh!
~{st•7=.
Prr~sx}v.~.wt.A?@D.\~}x~x}v.wt. fgrgpfgpeg!qt!cdwug/!6Twxqx.Q;.pvt.
Ut.6\cU7.
t;.|pxyp}p.x.ts. AE7=.
\pxyp}p.pvt.}|qt.|t}x~}ts.
q.p.{pvt.trt}pvt.~u.P|txrp}.
~w;.p}s.x.wt.|~.r~||~}{.ts. x}.q~w.wt.WWb.atr~||t}spx~}.p}s.
wx.STP.s~r|t}.x}r{st.
t.
x{{xrx.sv.p|~}v.P|txrp}.~w=.
u~|.]bSdW.p}s.\cU=.cwtt.
t.
P|~}v.st}.
tts.x}.A?@D;.
|tpt.tt}.~u.t.~u.|pxyp}p=.P.
@D=D3.~u.Gw.vpst;.B@=@3.~u.@?w.
|t}x~}ts.x}.wx.Upr~;.prr~sx}v.~.
vpst;.p}s.CC=F3.~u.@Aw.vpst.
wt.t{.~u.wt.A?@B.]bSdW.
t;.
t~ts.wp.wt.wps.ts.|pxyp}p.
@F=C3.~u.p.tp.|pxyp}p.t.pvt.
x}.wtx.{xutx|t=.X}.pssxx~};.@@=G3;.
AD=C3;.p}s.BC=H3.~u.Gw;.@?w;.p}s.@Aw. @A.~.~{st.ts.|pxyp}p.~}.B??.~.
RQ!11111
Hto!11182
Hov!5812
Uhov!5813
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64869!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
|~t.sp.xwx}.wt.t
x~.@A.
|~}w=.cwx.x}sxrpt.wp.D=F.|x{{x~}.
t~{t.ts.|pxyp}p.~}.p.spx{.~.
p{|~.spx{.qpx.~
t.wx.@A<|~}w.
tx~s;.wxrw.x.p.@=G<u~{s.x}rtpt.u~|.
wt.B=@.|x{{x~}.spx{.~.p{|~.spx{.
t.x}.A??E=.Uwt|~t;.E3.~u.p{{.
t{uw.vpst.x}.wt.d}xts.bpt.
t~ts.spx{.t.~u.|pxyp}p.x}.wt.
A?@D.\cU.
t=.cwtt.spp.~}v{.
x}sxrpt.wp.wtt.x.p.xv}xuxrp}.
~x~}.~u.wt.d=b=.~{px~}.x}v.
|pxyp}p.~}.p.spx{.qpx=.
6A7.P.pts.x}.Twxqx.Q.~}.pvt.AE;.
qp.P;.Tcvgu!qh!fgrgpfgpeg!qt!
cdwug!ctg!tgoctmcdn{!nqy!p}s.uwt.
vvt.wp.]k_pvgtxkgyu!hqt!vjg!
Pcvkqpcn!Nqpikvwfkpcn!Cneqjqn!
Grkfgokqnqikecn!Uwtxg{!)]PNCGU_!jxrl.
cpf!Pcvkqpcn!Grkfgokqnqikecn!Uwtxg{!
qp!Cneqjqn!cpf!Tgncvgf!Eqpfkvkqpu!
)]PGUCTE_!jxrl.gcej!eqphkto!vjcv!tcvgu!
qh!fgrgpfgpeg!qt!cdwug!qh!ecppcdku!
jcxg!pgxgt!gzeggf!)uke*!vyq!rgtegpv!kp!
c!ikxgp!{gct/!
cwt.pw~.~u.s.rxts.q.wt.
txx~}t.6R~|~}.t.p{=;.A??C7.
r~}r{sts.wp.p.wxvwt.trt}pvt.~u.
P|txrp}.ps{.wps.p.|pxyp}p.t.
sx~st.x}.A??@¤A??A.6@=D37.wp}.x}.
@HH@¤@HHA.6@=A37=.R~|~}.t.p{=.
6A??C7.}~ts.wp.wt.|pxyp}p.t.
sx~st.x}rtpt.~u.?=B3.~
t.wt.@?.
tp.tx~s.~{s.tpt.~.p}.x}rtpt.
u~|.A=A.|x{{x~}.t~{t.~.B.|x{{x~}.
t~{t.x}.wt.d}xts.bpt=.cwt.
txx~}t.upx{ts.~.t{px}.wt.x|pr.
~u.@=D3.6~.{t.wp}.A.trt}7.~u.wt.
d=b=.~{px~}.wp
x}v.p.|pxyp}p.t.
sx~st=.X}.~st.~..wtt.}|qt.
x}~.ttrx
t;.wt.STP.t
xtts.wt.
{xtpt.p}s.u~}s.wp.}~}<|tsxrp{.
trxx~}.sv.t.p}s.pqt.pt.
tt.tp|x}ts.x}.wt.p|t.][PTb.p}s.
]TbPaR.6@HH@¤@HHA.p}s.A??@¤A??A7.
~{px~}.6Q{p}r~.t.p{=;.A??F7=.Q{p}r~.
t.p{.6A??F7.tp|x}ts.}~}<|tsxrp{.
trxx~}.sv.t.p}s.pqt.pt.
u~|.wt.tx~s.~u.@HH@¤@HHA.p}s.
A??@¤A??A=.X}.@HH@.w~vw.@HHA;.wt.
t
p{t}rt.~u.}~}<|tsxrp{.trxx~}.
sv.6~x~xs;.x|{p};.p}s.
p}x{xt7.pqt.p}s.stt}st}rt.p.
?=@3=.]~}<|tsxrp{.trxx~}.sv.
6x|px{.~x~xs<qpts.sv7.pqt.
p}s.stt}st}rt.x}rtpts.~.?=B3.x}.
A??@.w~vw.A??A=.cwttu~t;.x}.wt.
p|t.A??@¤A??A.][PTb.p}s.]TbPaR.
~{px~};.wt.trt}pvt.~u.t~{t.
xw.p.|pxyp}p.t.sx~st.p.
p~x|pt{.ux
t<u~{s.wxvwt.6@=D3.
t.?=B37.wp}.w~t.xw.~x~xs.
pqt.p}s.stt}st}rt.t{x}v.u~|.
}~}<|tsxrp{.trxx~}.sv.t=.
Uwt;.e~{z~.t.p{=.6A?@C7.t~ts.
wp.x}.{~}v<t|.~.wtp
.|pxyp}p.
t;.H3.~u.t.qtr~|t.pssxrts.~.
|pxyp}p=.cwx.trt}pvt.x}rtpt.~.
@F3.wt}.|pxyp}p.t.p.x}.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
ps~{trt}rt.p}s.x.x}rtpt.~.AD.~.
D?3.~u.w~t.w~.pt.spx{.t=.
stt}st}rt.x}.tt}pvt.p}s.~}v.
ps{=.
Hcevqt!6<!Vjg!Ueqrg-!Fwtcvkqp-!cpf!
Ukipkhkecpeg!qh!Cdwug!
Pqt.~u.|pxyp}p.x.xsttps.p}s.
xv}xuxrp}=.P.t
x~{.}~ts;.
prr~sx}v.~.wt.]bSdW;.x}.A?@C;.p}.
tx|pts.@@F=A.|x{{x~}.P|txrp}.
6CC=A37.pvt.@A.~.~{st.wps.ts.
|pxyp}p.~.wpwxw.x}.wtx.{xutx|t;.
BD=@.|x{{x~}.6@B=A37.wps.ts.x.x}.wt.
p.tp;.p}s.AA=A.|x{{x~}.6G=C37.wps.
ts.x.x}.wt.p.|~}w=._p.tp.p}s.
p.|~}w.|pxyp}p.t.wp.x}rtpts.
xv}xuxrp}{.x}rt.A?@B=._p.|~}w.
|pxyp}p.t.x.wxvwt.p|~}v.@G¤A@.
tp.~{s.p}s.x.str{x}t.p|~}v.w~t.AA.
tp.~u.pvt.p}s.~{st=.X}.A?@C;.p}.
tx|pts.@G=D3.~u.p.tp.|pxyp}p.
t.pvt.@A.~.~{st.ts.|pxyp}p.~}.
B??.~.|~t.sp.xwx}.wt.p.@A.
|~}w=.cwx.p}{pt.x}~.E=D.|x{{x~}.
t~}.x}v.|pxyp}p.~}.p.spx{.~.
p{|~.spx{.qpx.~
t.p.@A<|~}w.
tx~s=.X}.A?@C;.p}.tx|pts.C@=E3.6H=A.
|x{{x~}7.~u.p.|~}w.|pxyp}p.t.
pvt.@A.~.~{st.ts.wt.sv.~}.A?.~.
|~t.sp.x}.wt.p.|~}w.6bP\WbP;.
]bSdW7=.Rw~}xr.t.~u.|pxyp}p.x.
p~rxpts.xw.p.}|qt.~u.wtp{w.xz.
6tt.Upr~.A.p}s.E7=.
Uwt|~t;.wt.p
tpvt.trt}pvt.
~u.% <cWR.x}.txts.|pxyp}p.wp.
x}rtpts.~
t.wt.p.~.strpst.
6cwt.d}x
tx.~u.\xxxx._~t}r.
\~}x~x}v._~ytr7=.Pssxx~}p{.sxt.
pt.}ttsts.~.r{pxu.wt.x|pr.~u.
vtpt.~t}r;.q.~}t.s.w~.
wp.wxvwt.{t
t{.~u.% <cWR.x}.wt.q~s.
pt.p~rxpts.xw.vtpt.rw~prx
t.
tuutr.6Wpst.p}s.axtq~rz;.@HHF7;.
wxrw.rp}.qt.r~t{pts.xw.wxvwt.
pqt.~t}xp{.6Rwpx.p}s.Qzt;.@HHC7=.
cTSb.spp.w~.wp.x}.A?@B;.
|pxyp}p>wpwxw.p.wt.x|p.
qp}rt.~u.pqt.x}.@E=G3.~u.p{{.
ps|xx~}.~.qp}rt.pqt.tp|t}.
p|~}v.pxt}[email protected]}s.~{st=.cTSb.
spp.p{~.w~.wp.|pxyp}p>wpwxw.
p.wt.x|p.qp}rt.~u.pqt.u~.
FF=?3.~u.p{{.@A<.~.@C<tp<~{s.
ps|xts.u~.sv.tp|t}.p}s.FD=D3.
~u.p{{.@D<.~.@F<tp<~{s.ps|xts.u~.
sv.tp|t}.x}.A?@B=.P|~}v.wt.
AG@;[email protected]|xx~}.~.sv.tp|t}.x}.
[email protected]}.wxrw.|pxyp}p>wpwxw.p.
wt.x|p.sv;.wt.p
tpvt.pvt.p.
ps|xx~}.p.AD.tp.p}s.wt.tpz.
pvt.r~w~.p.@D.~.@F.tp.6AA=D37=.
cwx<}x}t.trt}.~u.wt.AG@;HH@.
x|p.|pxyp}p>wpwxw.ps|xx~}.
6BD=H37.tt.}st.wt.pvt.~u.A?=.
X}.||p;.wt.trt}.pxxr.u~|.
wtt.
px~.
t.p}s.sppqpt.6tt.
Upr~[email protected]~.|~t.stpx{7.st|~}pt.
wp.|pxyp}p.r~}x}t.~.qt.wt.|~.
r~||~}{.ts.x{{xrx.sv;.xw.{pvt.
x}rxst}rt.~u.wtp
.t.p}s.
Rgvkvkqpgtu!Oclqt!Eqoogpv!kp!Tgncvkqp!
vq!Hcevqt!6!cpf!FGCu!Tgurqpug!
6@7._txx~}t .r~}t}s.wp;.Vjg!
rtgxcngpeg!cpf!ukipkhkecpeg!qh!rqvgpvkcn!
cdwug!ctg!nkokvgf!hqt!ecppcdku-!
gurgekcnn{!kp!tgncvkqp!vq!qvjgt!]u_ejgfwng!
KK!uwduvcpegu/!cwt.txx~}t.rxts.
t{.u~|.wt.@HH?.]XSP.W~tw~{s.
b
t.~}.Sv.Pqt.p}s.x}sxrpts.
wp;.oqtg!vjcp!hqwt!qwv!qh!hkxg!rgqrng!
yjq!jcf!wugf!ecppcdku!kp!vjg!rtgxkqwu!
{gct!tgrqtvgf!pq!rtqdngou!tgncvgf!vq!vjg!
ftwi/!6Twxqx.Q;.pvt.AG7=.
cwt.t
p{t}rt.~u.|pxyp}p.pvt.
p}s.|pxyp}p.stt}st}rt.x.xv}xuxrp}.
x}.wt.d}xts.bpt=.cwt.A?@C.]bSdW.
ux}sx}v.x}sxrpt.wp.wtt.pt.
p~x|pt{.E=D.|x{{x~}.P|txrp}.
x}v.|pxyp}p.~}.p.spx{.~.p{|~.
spx{.qpx=.Uwt;.e~{z~.t.p{=.6A?@C7.
t~ts.wp.x}.{~}v<t|.~.wtp
.
|pxyp}p.t;.H3.~u.t.qtr~|t.
pssxrts.~.|pxyp}p=.P|~}v.w~t.
w~.qtvp}.x}v.|pxyp}p.x}.
ps~{trt}rt;.|pxyp}p.stt}st}rt.
x}rtpt.~.@F3;.p}s.x.uwt.
x}rtpt.~.AD.~.D?3.~u.spx{.t.
wp.pts.x}v.|pxyp}p.sx}v.
ps~{trt}rt=.cwtt.r~{{trx
t.ux}sx}v.
x}sxrpt.wp.wtt.x.r~}xstpq{t.
xv}xuxrp}rt.p~rxpts.xw.|pxyp}p.
t.p}s.pqt.x}rt.H3.~u.t.qtr~|t.
pssxrts.~.|pxyp}p;.AD.~.D?3.~u.
spx{.|pxyp}p.t.pts.sx}v.
ps~{trt}rt;.p}s.t
p{t}rt.~u.pvt.x.
xv}xuxrp}{.wxvw.qpts.~}.wt.spp.
tt}ts.u~|.e~{z~.t.p{[email protected]}s.
wt.A?@C.]bSdW.
t=.
RQ!11111
Hto!11183
Hov!5812
Uhov!5813
Hcevqt!7<!Yjcv-!kh!cp{-!Tkum!Vjgtg!ku!vq!
vjg!Rwdnke!Jgcnvj!
X}.x.tr~||t}spx~};.wt.WWb.
sxrts.q{xr.wtp{w.xz.p~rxpts.
xw.prt.p}s.rw~}xr.|pxyp}p.t.x}.
Upr~.E=._q{xr.wtp{w.xz.p.
|tpts.q.t|tvt}r.stp|t}.
xx.p}s.sv.tp|t}.ps|xx~}.pt.
sxrts.q.WWb.p}s.STP.x}.Upr~.
@;.C;.p}s.D=.bx|x{p{;.Upr~.A.sxrt.
|pxyp}p .wp|pr~{~v.p}s.tt}.
~|t.~u.wt.ps
tt.wtp{w.tuutr.
p~rxpts.xw.t=.\pxyp}p.t.|p.
puutr.wt.wxrp{.p}s>~.rw~{~vxrp{.
u}rx~}x}v.~u.p}.x}sx
xsp{.t;.q.
|p.p{~.wp
t.q~pst.q{xr.x|pr.
x}r{sx}v.sx
x}v.x|px|t}.p}s.
upp{xxt.u~|.rp.prrxst}=.
Tkumu!Htqo!Cewvg!Wug!qh!Octklwcpc!
P.sxrts.x}.wt.WWb.t
xt.
s~r|t}.6WWb;.A?@D7;.prt.pvt.~u.
|pxyp}p.x|px.rw~|~~.
tu~|p}rt.x}r{sx}v.|~~.r~}~{.
p}s.x|{x
x;.xz.pzx}v.p}s.
ttrx
t.u}rx~}.6ap|ptzt.t.p{=;.
A??CJ.ap|ptzt.t.p{=;.A??E7=.X}.p.
G<^HT^HO^23CWR3/UIO
23CWR3
> e~{= G@; ]~= @DE > Uxsp; Pv @A; A?@E > _~~ts a{t
|x}~x ~u.x}sx
xsp{.x}v.|pxyp}p;.
sw~xp;.~{~}vts.p}xt;.p}s.
rw~{~vxrp{ sxt.|p.qt.~qt
ts.
6Wp}t.t.p{=;.@HHH7=.cwt.STP.uwt.
}~t.p.trt}.t
xt.~u.prt.|pxyp}p.
tuutr.6fx{zx}~}.t.p{=;.A?@C7.wp.
t~ts.x|pxts }t~{~vxrp{.u}rx~}.
x}r{sx}v.p{tts.trtx~};.pp}~xp;.
st{pts.t~}t.x|t;.p}s.|t|~.
stuxrx=
X}.x tr~||t}spx~};.WWb.
tutt}rt p.|tp<p}p{x.r~}srts.q.
[x.t.p{.6A?@A7.wtt.wt.pw~.
r~}r{sts wp rw~|~~
x|px|t}.p~rxpts.xw.prt.
|pxyp}p.pvt.wp
t.p{~.qtt}.
p~rxpts.xw.x}rtpts.xz.~u.rp.
prrxst} xw.x}sx
xsp{.ttxt}rx}v.
prt.|pxyp}p x}~xrpx~}.6[x.t.p{=;.
[email protected];.A?@D7= cwt.STP.uwt.
}~t.|~t.trt}.sxt.tp|x}x}v.wt.
xz.p~rxpts.xw.|pxyp}p.t.p}s.
sx
x}v=.h~}vt.sx
t.6}st.A@7.
wp
t qtt} rwpprtxts p wt wxvwt
xz.v~ p~rxpts.xw.|pxyp}p.
t.p}s.sx
x}v.6fwxtwx{{.t.p{=;.A?@C7=.
Uwt|~t;.x}.A?@B;.|pxyp}p.p.
u~}s.x}.@B3.~u.wt.sx
t.x}
~{
ts.x}.
p~|~qx{t<t{pts upp{.prrxst}.
6\rRp;.A?@D7=.cwt.~t}xp{.xz.~u.
p~|~qx{t.prrxst}.p~rxpts.xw.
|pxyp}p.t.ptp.~.qt.x}rtpx}v.
x}rt.wtt wp.qtt}.p.tps.x}rtpt.x}.
x}sx
xsp{ x}~xrpts xw |pxyp}p
~
t wt.p.A?.tp.6fx{~}.t.p{=;.
A?@C7=.W~t
t;.p.trt}.s.
r~||xx~}ts q.wt.]px~}p{.Wxvwp.
cpuuxr.bput.Ps|x}xpx~}.6]WcbP7.
t~ts.wp.wt}.psyts.u~.
r~}u~}st.6p{r~w~{.t;.pvt;.
vt}st;.tw}xrx7;.wtt.p.}~.p.
xv}xuxrp}.x}rtpt.x}.rpw.xz.6upp{.
p}s.}~}upp{;.}.L.A;EGA7.p~rxpts.xw.
|pxyp}p t 6R~|~} p}s Qt}x}v;
cwt.STP.p{~.}~t.trt}.sxt.
tp|x}x}v.}x}t}x~}p{.t~t.~u.
rwx{st}.~.|pxyp}p.6fp}v.t.p{=;.
[email protected]?@C7=.fp}v.t.p{=.6A?@B7.t
xtts.
t|tvt}r.stp|t}.6TS7.
xx.p.p.
rwx{st} .w~xp{.x}.R~{~ps~.u~|.
Yp}p.@;.A??D.~.Strt|qt.B@;.A?@@=.
P.pts.q.wt.pw~;.x}.A???.
R~{~ps~.pts.P|t}s|t}.A?.wxrw.
p{{~ts u~.wt.t.~u.|pxyp}p=.
U~{{~x}v wt.ppvt.~u.p.}t.Yxrt.
Stp|t}.~{xr .x}rx}v.utstp{.
~tr~ }~.~.ttz.pt.~u.|tsxrp{.
|pxyp}p.t.p}s.{xt.p.{~}v.p.
wt r~}u~|.~.pt.{p .6p.pts.x}.
fp}v.t.p{=;.A?@B7;.@C.pxt}.x}.
R~{~ps~.}st.wt.pvt.~u.@A.tt.
ps|xts.~.wt.TS.u~.wt.}x}t}sts.
t.~u.|pxyp}p.~
t.p.AF.|~}w.
tx~s=._x~.~.wt.ppvt.~u.wx.
~{xr;.u~|.Yp}p.@;.A??D.~.
btt|qt.B?;.A??H.6DF.|~}w7;.wtt.
tt }~ tsxpxr TS
xx st ~
}x}t}x~}p{.|pxyp}p.t~t.
6fp}v.t.p{=;.A?@B7=.cwt.STP.p{~.}~t.
p.{pvt.rp{t.t
p{px~}.~u.tsxpxr.
t~t x}v wt ]px~}p{ _~x~}
Spp.bt| 6fp}v.t.p{=;.A?@C7=.cwp.
s.t~ts.wp.wtt.tt.HGD.
}x}t}x~}p{ |pxyp}p t~t x}
rwx{st}.6H.tp.p}s.~}vt7.qttt}.
Yp}p.@;.A??D.~.Strt|qt.B@;.A?@@=.
cwt pw~ pxuxts wt TS
xx q
pt.xw.{p.p{{~x}v.|tsxrp{.t.~u.
|pxyp}p;.pt.p}xx~}x}v.~.
{tvp{xpx~} u~ |tsxrp{ t; p}s pt
xw.}~.rw.{p=.^.~u.wt.HGD.
t~t;.CHD.tt.x}.}~}<{tvp{.pt.
6} L BB pt7; HB x} p}xx~}p{ pt
6}.L.G.pt7;.p}s.BHE.x}.{tvp{ .pt.
6}.L.H.pt7=.cwt.pw~.t~ts.wp.
wtt p p ~u~{s x}rtpt 6^a L A=@7
x}.|~stpt.~.|py~.tuutr.x}.rwx{st}.
xw.}x}t}x~}p{.|pxyp}p.t.p}s.p.
wttu~{s x}rtpt 6^a L B=C7 x}
ps|xx~}.~ rxxrp{.rpt.}x.x}.pt.
p{{~x}v |tsxrp{.t.~u.|pxyp}p;.x}.
r~|px~} ~ }~}<{tvp{ pt=
cwt.WWb.}~ts.wp.p.|py~.xz.u~|.
rw~}xr.|pxyp}p.t.x.p.sxx}rx
t.
xwspp{.}s~|t;.p.strxqts.x}.
[email protected]\¤D=.cwt.WWb.p}p{x.p{~.
~ts.wt.u~{{~x}v.strxx~}.~u.xz.
p~rxpts.xw.|pxyp}p.jrp}}pqxl.
pqt.u~|.wt.Sb\¤DI.
cwt.WWb.pts.wp.rw~}xr.
|pxyp}p t ~srt prt p}s
rw~}xr.ps
tt.tuutr.~}.wt.
txp~ t|;.|t|~.p}s.
{tp}x}v= atv{p |pxyp}p |~zx}v
rp}.~srt p.}|qt.~u.{~}v<t|.
{|~}p.r~}tt}rt;.x}r{sx}v.
rw~}xr.r~vw.p}s.x}rtpts.|.
6Psp|.p}s.\px};.@HHE7;.p}s.
wx~pw~{~vxr.pq}~|p{xxt x}.
q~}rwxp{.txwt{x|.6Psp|.p}s.
\px};.@HHE7=.
Octklwcpc!cu!c!Icvgyc{!Ftwi!
cwt.WWb.t
xtts.wt.r{x}xrp{.
sxt.t
p{px}v wt.vptp.
w~wtx x}.|pxyp}p.p}s.u~}s.
wt| ~.qt.{x|xts=.cwt.x|p.tp~}.
ttI.6@7 atrxts.pxrxp}.tt.
x}u{t}rts q ~rxp{; qx~{~vxrp{; p}s
tr~}~|xr.upr~.wp.r~}xqt.~.
tt}x
t sv.pqt.6Wp{{.p}s.[}zt;.
A??D7;.p}s.6A7.|~.sxt.tx}v.wt.
vptp.sv.w~wtx.u~.|pxyp}p.
t.wt.stt|x}px
t.|tpt.
pwt.wp}.p{x}v.
Sb\¤D.rxtxp.u~.sv.pqt.~.
cwt.WWb.rxts.t
tp{.sxt.wtt.
|pxyp}p.t.sxs.}~.{tps.~.~wt.
x{{xrx.sv.t.6Zp}st{.p}s.Rwt};.A???J.
~}.bs~.t.p{=;.A??AJ.]prt.t.p{=;.
@HFD7= c~ tppt {~}vxsx}p{
sxt.xw.ps~{trt}.x}v.
|pxyp}p.sxs.}~.st|~}pt.p}.
p~rxpx~}.xw.t.~u.~wt.x{{xrx.
sv.6Zp}st{.p}s.Rwt};.A???J.
~}.
bs~.t.p{=;.A??A7=.
X p.}~ts.q.wt.WWb.wp;.wt}.
t
p{px}v.wt.vptp.w~wtx;.
sxuutt}rt.ptp.wt}.tp|x}x}v.t.
t.pqt.~.stt}st}rt.~u.~wt.
x{{xrx.sv=.ep}.V}s.p}s.atqt{{~}.
6A?@?7.t~ts.wp.wtt.p.p.
r~t{px~} qttt}.|pxyp}p.t.x}.
ps~{trt}rt p}s ~wt x{{xrx sv t x}
tp{.ps{w~~s;.q.wt}.tp|x}ts.x}.
t|.~u.sv.pqt.~u.~wt.x{{xrx.
sv;.pvt<{x}zts.t~.p}s.~rxp{.
~{t.tt.r~}u~}st.x}.wt.
p~rxpx~}=.Stvt}wps.t.p{=.6A??H7.
t~ts.wp.|pxyp}p.t.~ut}.
trtst.t.~u.~wt.x{{xrx.sv;.q.
stt}st}rt.x}
~{
x}v.sv.~wt.wp}.
|pxyp}p.utt}{.r~t{pts.xw.
wxvwt.{t
t{.~u.x{{xrx.sv.pqt=.
Uwt|~t;.Stvt}wps.t.p{=.6A?@?7.
t~ts.wp.x}.r~}xt.xw.{~t.
t
p{t}rt ~u.|pxyp}p.pvt;.t.~u.
~wt.x{{xrx.sv.qtu~t.|pxyp}p.p.
~ut} s~r|t}ts=.
Qpts.~}.wtt.sxt.p|~}v.~wt;.
wt.WWb.r~}r{sts.wp.p{w~vw.|p}.
x}sx
xsp{.xw.p.sv.pqt.sx~st.
|p.wp
t.ts.|pxyp}p.p.~}t.~u.wtx.
ux.x{{xrx.sv;.wx.s~t.}~.|tp}.
wp.x}sx
xsp{ x}xxpts.xw.
|pxyp}p.x}wtt}{ x{{.v~.~}.~.
qtr~|t.tv{p t.~u.~wt.x{{xrx.
sv=.
64871!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
X.x.qt{xt
ts.wp.{prz.~u.~{tp}rt.~.
rw~prx
t.tuutr.|p.t{pt.~.
t{tr~wx~{~vxrp{.spp.st|~}px}v.
wp.rw~}xr.% <cWR.ps|x}xpx~}.
s~t.}~.puutr.x}rtpts.}t~}p{.uxx}v.
x}.wt.
t}p{.tv|t}p{.ptp;.p.qpx}.
tvx~}.wp.{p.p.rxxrp{.~{t.x}.sv.
tx}u~rt|t}.p}s.tps.6f.p}s.
Ut}rw;.A???7=.W|p}.rp}.st
t{~.
~{tp}rt.~.|pxyp}p .rpsx~
pr{p;.
p~}~|xr;.p}s.qtwp
x~p{.tuutr.6Y~}t.
t.p{=;.@HG@7=.c~{tp}rt.~.~|t.
qtwp
x~p{.tuutr.ptp.~.st
t{~.
xw.wtp
.p}s.rw~}xr.t;.q.}~.
xw.~rrpx~}p{.pvt=.ap|ptzt.t.p{=.
6A??H7.t~ts.wp.u~{{~x}v.prt.
ps|x}xpx~}.~u.|pxyp}p;.~rrpx~}p{.
|pxyp}p.t.x{{.twxqxts.
x|px|t}.x}.przx}v.p}s.pt}x~}.
pz.wttp.tu~|p}rt.~u.wtp
.
t.~}.wt.wtt.pz.p.}~.
puutrts=.X}.p.u~{{~<.s.xw.wt.
p|t.qytr.wp.pxrxpts.x}.wt.
s.q.ap|ptzt.t.p{=.6A??H7;.p.
}t~wx~{~vxrp{.pt|t}.p.
r~}srts.wtt.t
t}<t{pts.
~t}xp{.6Ta_7.tt.|tpts.x}v.
t{tr~t}rtwp{~vpw.6TTV7.
6cwt}xt}.t.p{=;.A?@A7=.bx|x{p.~.wt.
tp{xt.t{;.wt.wtp
.|pxyp}p.
t.6}.L.@@J.p
tpvt.~u.BC?.|pxyp}p.
t.t.tp7.wps.}~.rwp}vt.x}.wtx.
Ta_.xw.wt.prt.|pxyp}p.
t~t=.W~t
t;.~rrpx~}p{.t.6}.
L.@?J.p
tpvt.~u.DD.|pxyp}p.t.t.
tp7.wps.xv}xuxrp}.strtpt.x}.wt.
p|{xst.~u.p}.Ta_.r~|~}t}.
6rptv~xts.p._@??7.~}.przx}v.p}s.
pt}x~}.pz.p}s.Ta_.p|{xst.
rwp}vt.x.x}sxrpx
t.~u.p.rwp}vt.x}.qpx}.
prx
x.6cwt}xt}.t.p{=;.A?@A7=.
cwt.WWb.x}sxrpts.wp.s~}<.
tv{px~}.~u.rp}}pqx}~xs.trt~.|p.
qt.p.~xq{t.|trwp}x|.u~.~{tp}rt.~.
|pxyp}p .tuutr.6Wx
~}t}.t.p{=;.
[email protected]~}p{t.t.p{=;.A??DJ.a~sxvt.
st.U~}trp.t.p{=;.@HHCJ.^
xts~.t.p{=;.
@HHB7=.
P.x}sxrpts.q.wt.WWb;.wt.|~.
r~||~}.xwspp{.|~|.x}.
wtp
;.rw~}xr.|pxyp}p.t.pt.{tt.
sxuuxr{xt;.strtpts.ptxt.~.
Hcevqt!8<!Kvu!Ru{ejke!qt!Rj{ukqnqikecn!
txvw.{~;.xxpqx{x;.p}vt;.p}xt.~.
Fgrgpfgpeg!Nkcdknkv{!
}t
~}t;.p}s.t{t}t.6Qs}t.
Rj{ukqnqikecn!)Rj{ukecn*!Fgrgpfgpeg!kp! p}s.Wvwt;.A??EJ.Wp}t.t.p{=;.@HHH7=.
Jwocpu!
P.t~ts.q.WWb;.|~.|pxyp}p.
cwt.WWb.pts.wp.wtp
.p}s.
xwspp{.|~|.qtvx}.xwx}.AC¤.
rw~}xr.t.~u.|pxyp}p.rp}.{tps.~.
CG.w~.~u.sxr~}x}px~};.tpz.
wxrp{.stt}st}rt.6Sb\¤D;.A?@BJ.
xwx}.C¤E.sp;.p}s.{p.u~.@¤B.ttz=.
cwt.WWb.~x}ts.~.wp.wt.
Qs}t.p}s.Wvwt;.A??EJ.Wp}t.t.p{=;.
@HHH7=.c~{tp}rt.x.st
t{~ts.u~{{~x}v. P|txrp}._rwxpxr.P~rxpx~} .
6P_P 7.Sxpv}~xr.p}s.bpxxrp{.
ttpts.ps|x}xpx~}.~u.|pxyp}p.
p}s.xwspp{.|~|.pt.~qt
ts. \p}p{.~u.\t}p{.Sx~st<D.6Sb\¤D7.
x}r{sts.p.{x.~u.xwspp{.|~|.
p.u~{{~x}v.sxr~}x}px~}.~u.
u~{{~x}v.|pxyp}p.jrp}}pqxl.t.
|pxyp}p.pvt.6WWb;.A?@D7=.
6Sb\¤D;.A?@B7=.cwt.STP.}~t.wp.p.
cwt.WWb.|t}x~}ts.wp.~{tp}rt.
rp}.st
t{~.~.~|t.~u.|pxyp}p .
Sb\¤D.~zx}v.v~.t~.x}sxrpts.
tuutr;.q.s~t.}~.ptp.~.st
t{~.
wp.|pxyp}p.xwspp{.|~|.
xw.ttr.~.wt.rw~prx
t.tuutr=. tt.pssts.~.Sb\¤D.6wt.tt.}~.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Rgvkvkqpgtu!Oclqt!Eqoogpv!kp!Tgncvkqp!
vq!Hcevqt!7!cpf!vjg!Iqxgtpogpvu!
Tgurqpugu!
[email protected].txx~}t.r~||t}ts.wp.
|pxyp}p.s~t.}~.xv}xuxrp}{.x|pr.
~rxp{.qtwp
x~.x}.s~|px}.rw.p.
|~x
px~};.sx
x}v;.pvvtx~};.~.
w~x{x.6Twxqx.Q;.pvt.B?¤C@7=.
cwt.WWb.r~}r{sts.wp.\pxyp}p .
prt.tuutr.rp}.xv}xuxrp}{.x}tutt.
xw.p.t~} .pqx{x.= = =.~.~tpt.
|~~.
twxr{t= .6WWb;[email protected].
|t}x~}ts.x}.wx.upr~;.wtt.x.p.
xv}xuxrp}.xz.xw.|pxyp}p.t.p}s.
sx
x}v=.\pxyp}p.p.u~}s.x}.@B3.~u.
sx
t.x}
~{
ts.x}.p~|~qx{t.upp{.
prrxst}.6\rRp;.A?@D7=.Uwt|~t;.
x}.p.|tp<p}p{x.r~}srts.q.[x.t.p{=.
6A?@@7;.p}.p~rxpx~}.p.xst}xuxts.
qttt}.|pxyp}p.t.q.wt.sx
t.
p}s.p}.x}rtpts.xz.~u.vtx}v.x}~.p.
rp.prrxst}=.
cwt.STP.}~t.wp.wt.txx~}t.
~}{.r~}xstts.wtwt.|pxyp}p.
rtpt.~rxp{.~q{t|;.p}s.sxs.}~.
r~}xst.wx~{~vxrp{.rwp}vt.p}s.
x|pr.wp.p{~.w~{s.qt.t
p{pts.x}.
stt|x}x}v.wt.xz.~.q{xr.wtp{w=.
cwt.WWb.p}s.STP.r~}xstts.wt.
q{xr.wtp{w.x|pr.~u.rw.
wx~{~vxrp{.tuutr;.p.sxrts.x}.
wx.upr~.p}s.~wt.pq~
t=.\pxyp}p.
|p.t{.x}.prt.rpsx~
pr{p.
~xrx.p.x}sxrpts.q.trt}.t
xt.
tp|x}x}v.wtt.p~rxpx~}.6Wprzwp|;.
A?@DJ._p}px~xst;.A?@D7=.cwtt.x.p.
~xq{t.p~rxpx~}.qttt}.utt};.
{~}v<t|.|pxyp}p.t.p}s.x}rtpts.
xz.~u.txr{p.vt|.rt{{.rp}rt.p}s.
~|t.t
xst}rt.wp.rw~}xr.|pxyp}p.
t.|p.{tps.~.{}v.rp}rt.p{w~vw.
wt.t
xst}rt.x.x}r~}xt}=.
Uwt|~t;.p.|~t.trt}.xz.x.wt.
x}rtpt.x}.TS.
xx.~u.rwx{st}.
}x}t}x~}p{{.t~ts.~.|pxyp}p.
xw.x}rtpts.xz.upr~.u~.|py~.
ps
tt.tuutr.~.ps|xx~}.~.rxxrp{.
rpt.}x.x}.pt.wp.wp
t.{tvp{xts.
|pxyp}p.u~.|tsxrp{.~t.6fp}v.
t.p{=;.A?@C7=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11185
Hov!5812
Uhov!5813
t
x~{.x}r{sts.x}.Sb\¤Xe7.
qtrpt.|pxyp}p.xwspp{.wp.}~.
qtt}.t{xpq{.tt}ts.x}.t
tp{.
sxt.6Wpx}.t.p{=;.A?@B7=.X}.w~;.
|pxyp}p.xwspp{.xv}.pt.t~ts.
x}..~.~}t<wxs.~u.tv{p.t.p}s.
qttt}.D?3.p}s.H?3.~u.wtp
.t.
6Wpx}.t.p{=;.A?@B7=.Prr~sx}v.~.Sb\¤.
D.rxtxp;.x}.~st.~.qt.rwpprtxts.p.
wp
x}v.|pxyp}p.xwspp{;.p}.
x}sx
xsp{.|.st
t{~.p.{tp.wtt.~u.
wt.t
t}.|~|.xwx}.~}t.ttz.~u.
strtpx}v.~.~x}v.wt.wtp
.p}s.
~{~}vts.t.6Sb\¤D;.A?@B7=.cwtt.
t
t}.|~|.p[email protected]xpqx{xJ.
p}vt.~.pvvtx~};.6A7.}t
~}t.~.
p}xt;.6B7.{tt.sxuuxr{;.6C7.
strtpts.ptxt.~.txvw.{~;.6D7.
t{t}t;.6E7.strtpts.|~~s;.6F7.
~|pxr.|~|.rpx}v.xv}xuxrp}.
sxr~|u~.6Sb\¤D;.A?@B7=.
Ru{ejqnqikecn!)Ru{ejke*!Fgrgpfgpeg!kp!
Jwocpu!
Wxvw.{t
t{.~u.rw~prx
t.tuutr.
rw.p.~xx
t.tx}u~rt|t}.r~t{pt.
xw.x}rtpts.|pxyp}p.pqt.p}s.
stt}st}rt.6brwtt.t.p{=;.A??HJ.
itxvt.t.p{=;.A?@?7=.Txst|x~{~vxrp{.
|pxyp}p.t.spp.t~ts.q.]bSdW;.
\cU;.p}s.cTSb.~.wx.ptx~}.
p.tt}ts.x}.wt.WWb.A?@D.t
xt.~u.
|pxyp}p.p}s.spts.q.wt.STP=.
Prr~sx}v.~.wt.ux}sx}v.x}.wt.A?@C.
]bSdW.
t;.p}.tx|pts.H=A.
|x{{x~}.x}sx
xsp{.@A.tp.p}s.~{st.
ts.|pxyp}p.spx{.~.p{|~.spx{.6A?.
~.|~t.sp.xwx}.wt.p.|~}w7=.X}.
wt.A?@D.\cU.t~;.spx{.|pxyp}p.
t.6A?.~.|~t.sp.xwx}.wt.p.B?.
sp7.x}.Gw;.@?w;.p}s.@Aw.vpst.x.
@=@3;.B=?3;.p}s.E=?3;.ttrx
t{=.
cwt.A?@C.]bSdW.t~.pts.wp.
C=A.|x{{x~}.t~}.tt.r{pxuxts.xw.
stt}st}rt.~}.~.pqt.~u.|pxyp}p.x}.
wt.p.tp.6ttt}x}v.@=E3.~u.wt.
~p{.~{px~}.pvt.@A.~.~{st;.p}s.
DH=?3.~u.w~t.r{pxuxts.xw.x{{xrx.
sv.stt}st}rt.~.pqt7.qpts.~}.
rxtxp.trxuxts.x}.wt.Sxpv}~xr.p}s.
bpxxrp{.\p}p{.~u.\t}p{.Sx~st;.
Cw.tsxx~}.6Sb\¤Xe7=.Uwt|~t;.~u.
wt.ps|xx~}.~.{xrt}ts.qp}rt.
pqt.uprx{xxt;.p.tt}ts.x}.cTSb;.
|pxyp}p>wpwxw.p.wt.x|p.
qp}rt.~u.pqt.u~J.@G=B3.6BDA;AHF7.
~u.A?@@.ps|xx~}J.@F=D3.6B@D;A??7.~u.
[email protected]|xx~}J.p}s.@E=G3.6AG@;HH@7.
[email protected]|xx~}=.^u.wt.AG@;HH@.
ps|xx~}.x}[email protected]~.|pxyp}p>
wpwxw.p.wt.x|p.qp}rt;.
AC=B3.ts.|pxyp}p>wpwxw.spx{=.
P|~}v.ps|xx~}.~.tp|t}.u~.
|pxyp}p>wpwxw.p.wt.x|p.
qp}rt.x}.A?@B;.AF=C3.tt.pvt.@A.
~.@F.tp.p}s.AH=F3.tt.pvt.A?.~.
AC.tp=.
G<^HT^HO^23CWR3/UIO
23CWR3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
cwt.WWb.r~}r{sts.wp.|pxyp}p.
wp.p.wxvw.~t}xp{.u~.pqt.qpts.~}.
p.{pvt.}|qt.~u.t~{t.tv{p{.x}v.
|pxyp}p;.x.xsttps.t;.p}s.wt.
p.p|~}.~u.|pxyp}p.wp.x.
p
px{pq{t.w~vw.x{{xrx.rwp}}t{=.
\pxyp}p.x.wt.|~.pqts.p}s.
puuxrzts.x{{xrx.qp}rt.x}.wt.d}xts.
bpt=.P~x|pt{.AA=A.|x{{x~}.
x}sx
xsp{.x}.wt.d}xts.bpt.6G=C3.
~u.wt.d}xts.bpt.~{px~}7.tt.
p.|~}w.t.~u.|pxyp}p.prr~sx}v.
~.wt.A?@C.]bSdW.
t=.P.A?@D.
}px~}p{.
t.6\~}x~x}v.wt.Ut7.
wp.prz.sv.t.t}s.p|~}v.wxvw.
rw~~{.st}.w~ts.wp.q.@Aw.
vpst;.A@=B3.~u.st}.t~ts.x}v.
|pxyp}p.x}.wt.p.|~}w;.p}s.E=?3.
t~ts.wp
x}v.ts.x.spx{.x}.wt.p.
|~}w=.X}.A?@@;.bP\WbP .Sv.Pqt.
fp}x}v.]t~z.6SPf]7.t~ts.wp.
|pxyp}p.p.|t}x~}ts.x}.BE=C3.~u.
x{{xrx.sv<t{pts.t|tvt}r.
stp|t}.6TS7.
xx;.r~t~}sx}v.~.
CDD;EEG.~.~u.p~x|pt{.@=AD.
|x{{x~}.
xx=.cwt.ctp|t}.Tx~st.
Spp.bt.6cTSb7.w~ts.wp.@E=G3.~u.
}~}<x
pt.qp}rt<pqt.tp|t}.
uprx{x.ps|xx~}.x}.A?@B.tt.u~.
|pxyp}p.p.wt.x|p.sv=.
\pxyp}p.wp.s~t<stt}st}.
tx}u~rx}v.tuutr.wp.t}r~pvt.x.
pqt=.Q~w.r{x}xrp{.p}s.tr{x}xrp{.
sxt.wp
t.st|~}pts.wp.
|pxyp}p.p}s.x.x}rx{t.
rw~prx
t.r~}xt};.% <cWR;.
~t.wt.wp|pr~{~vxrp{.pxqt.
p~rxpts.xw.sv.~u.pqt=.cwt.
u}rx~}.p.sxrx|x}px
t.x|{x.p}s.
p.~xx
t.tx}u~rt.~.|px}px}.sv.
t.p}s.sv<ttzx}v.qtwp
x~=.
Pssxx~}p{{;.t.~u.|pxyp}p.rp}.
t{.x}.rw~{~vxrp{.stt}st}rt=.
A=.\pxyp}p.wp.}~.rt}{.
prrtts.|tsxrp{.t.x}.tp|t}.x}.wt.
d}xts.bpt=.
cwt.WWb.pts.wp.wt.USP.wp.}~.
p~
ts.p}.]SP.u~.|pxyp}p=.cwt.
WWb.}~ts.wp.wtt.pt.~~}xxt.
u~.rxt}x.~.r~}sr.r{x}xrp{.
ttprw.xw.|pxyp}p.p}s.wtt.pt.
prx
t.X]S.u~.|pxyp}p;.q.|pxyp}p.
s~t.}~.wp
t.p.rt}{.prrtts.
Hcevqt!9<!Yjgvjgt!vjg!Uwduvcpeg!ku!cp!
|tsxrp{.t.x}.wt.d}xts.bpt;.}~.
Koogfkcvg!Rtgewtuqt!qh!c!Uwduvcpeg!
s~t.x.wp
t.p}.prrtts.|tsxrp{.t.
Cntgcf{!Eqpvtqnngf!Wpfgt!vjg!EUC!
xw.t
tt.txrx~}=.
USP.p~
p{.~u.p}.]SP.x.}~.wt.
\pxyp}p.x.}~.p}.x||tsxpt.
~{t.|tp}.w~vw.wxrw.p.sv.rp}.qt.
tr~.~u.p}~wt.r~}~{{ts.
stt|x}ts.~.wp
t.p.rt}{.
qp}rt=.
prrtts.|tsxrp{.t .}st.wt.RbP=.
Fgvgtokpcvkqp!
P{x}v.wt.ux
t<p.t.||pxts.
Put.r~}xstpx~}.~u.wt.txvw.upr~. qt{~;.p.sv.wp.p.rt}{.prrtts.
sxrts.pq~
t.p}s.~u.wt.WWb .
|tsxrp{.t.xu.p{{.~u.wt.u~{{~x}v.ux
t.
atr~||t}spx~};.wt.STP.ux}s.wp.
t{t|t}.wp
t.qtt}.pxuxts=.P.
stpx{ts.x}.wt.WWb.t
p{px~}.p}s.p.
|pxyp}p.|tt.wt.wtt.rxtxp.u~.
{prx}v.p.qp}rt.x}.rwts{t.X.~u.wt. t.u~w.qt{~;.}~}t.~u.wtt.t{t|t}.
RbP.}st[email protected]=b=R=.G@A6q76@7I.
wp.qtt}.u{ux{{ts.u~.|pxyp}pI.
@=.\pxyp}p.wp.p.wxvw.~t}xp{.u~.
x=.Vjg!ftwiu!ejgokuvt{!owuv!dg!
pqt=.
mpqyp!cpf!tgrtqfwekdng/!
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Rgvkvkqpgtu!Oclqt!Eqoogpv!kp!Tgncvkqp!
vq!Hcevqt!8!cpf!vjg!Iqxgtpogpvu!
Tgurqpug!
[email protected].txx~}t.pts;.Vjgtg!ku!
pq!ugxgtg!rj{ukecn!ykvjftcycn!
u{pftqog!cuuqekcvgf!ykvj!ecppcdku/!
Ecppcdku!cffkevkqp!ku!cogpcdng!vq!
vtgcvogpv/!6Twxqx.Q;.pvt.@?7=.cwt.
txx~}t.uwt.x}sxrpts.wp.
|pxyp}p.oc{!dg!ru{ejqnqikecnn{!
cffkevkxg-!dwv!owej!nguu!uq!vjcp!qvjgt!
Uejgfwngf!jxrl.KK!ftwiu- .6Twxqx.Q;.
pvt.@?7.p}s.wp.wtt.x.p.{~.xz.~u.
stt}st}rt.p~rxpts.xw.|pxyp}p.
t=._txx~}t.uwt.pts.x}.Twxqx.
Q;.•pvt.AB;.Ecppcdku!jcu!nqy!tgncvkxg!
fgrgpfgpeg!tkum!cpf!fqgu!pqv!tgcej!vjg!
ugxgtkv{!cuuqekcvgf!ykvj!qvjgt!ftwiu/ .
cwt.WWb.pt.wp.|pxyp}p.
xwspp{.}s~|t.ptp.~.qt.
|x{s.r~|pts.~.r{pxrp{.p{r~w~{.p}s.
qpqxpt.xwspp{.}s~|t .p}s.
x.x|x{p.x}.|pv}xst.p}s.x|t.r~t.
~.~qprr~.xwspp{.}s~|t=.
Sb\¤D.}~.tr~v}xt.p}s.strxqt.
p.|pxyp}p.jrp}}pqxl.xwspp{.
}s~|t=.cwt.{xutx|t.xz.~u.
stt}st}rt.~.|pxyp}p.x.
p~x|pt{.H3.p|~}v.wtp
.~.{~}v<.
t|.t.6e~{z~.t.p{=;.A?@C7=.
\pxyp}p.t{.x}.~{tp}rt.p}s.
xwspp{.p.strxqts.tp{xt.x}.wx.
Upr~.F=.cwt.spp.u~|.]bSdW.x}sxrpt.
wp.wtt.x.r~}p}.stxt.u~.
|pxyp}p.p.}~ts.q.wt.r~}xt}{.
wxvw.}|qt.~u.rt}.spx{.t.x}.
ps{.p}s.ps~{trt}=.\pxyp}p.t.
p{~.tx.stxt.~q{t|.
p~rxpts.xw.wt.sv=.Rwp}vt.x}.X`.
wp
t.qtt}.}~ts.x}.ps~{trt}<~}t;.
rw~}xr.~.stt}st}.|pxyp}p.t;.
x}.pssxx~}.~.xwspp{.|~|=.
W~t
t;.|pxyp}p.t.wp.}~.
str{x}ts.x}.wt.x|t.wp.pvt.~u.wx.
sv.wp.qtt}.|~}x~ts=.Pssxx~}p{{;.
wtt.wp.qtt}.p}.x}rtpt.x}.r~}t}.~u.
wt.x|p.rw~prx
t.rwt|xrp{;.% <.
cWR;.x}.|pxyp}p.p|{t.p}p{ts.q.
wt.d}x
tx.~u.\xxxx ._~t}r.
\~}x~x}v._~ytr;.vvtx}v.
tutt}rt.u~.|pxyp}p.px}.xw.
wxvwt.{t
t{.~u.% <cWR=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11186
Hov!5812
Uhov!5813
64872!
Rwt|xrp{.r~}xt}.x}r{sx}v.% <.
cWR.p}s.~wt.rp}}pqx}~xs.x}.
|pxyp}p.
p.xv}xuxrp}{.x}.sxuutt}.
|pxyp}p.px}=.X}.pssxx~};.wt.
r~}rt}px~}.~u.% <cWR.p}s.~wt.
rp}}pqx}~xs.|p.
p.qttt}.px}=.
cwttu~t.wt.rwt|xrp{.r~|~xx~}.
p|~}v.sxuutt}.|pxyp}p.p|{t.x.
}~.t~srxq{t=.St.~.wt.
pxpx~}.~u.
wt.rwt|xrp{.r~|~xx~}.x}.|pxyp}p.
px};.x.x.}~.~xq{t.~.stx
t.p.
p}spsxts.s~t=.cwt.WWb.s~t.
ps
xt.wp.xu.p.trxuxr.Ecppcdku!px}.
x.r{x
pts.p}s.~rtts.}st.
r~}~{{ts.r~}sxx~};.wt.{p}.
rwt|x.|p.qt.r~}xt}.t}~vw.~.
stx
t.p}spsxts.s~t=.
xx=.Vjgtg!owuv!dg!cfgswcvg!uchgv{!
uvwfkgu/!
cwtt.pt.}~.pstpt.put.sxt.
~}.|pxyp}p.u~.t.x}.p}.trxuxr;.
tr~v}xts.|tsxrp{.r~}sxx~}=.cwt.
r~}xstpq{t.
pxpx~}.x}.wt.rwt|x.
~u.|pxyp}p.t{.x}.sxuutt}rt.x}.
put;.qx~{~vxrp{;.wp|pr~{~vxrp{;.p}s.
~xr~{~vxrp{.pp|tt.p|~}v.wt.
px~.|pxyp}p.p|{t=.
xxx=.Vjgtg!owuv!dg!cfgswcvg!cpf!ygnn.!
eqpvtqnngf!uvwfkgu!rtqxkpi!ghhkece{/!
cwtt.pt.}~.pstpt.p}s.t{{<.
r~}~{{ts.sxt.wp.stt|x}t.
|pxyp}p .tuuxrpr=.X}.p}.x}stt}st}.
t
xt.tu~|ts.q.wt.USP.~u.
q{xr{.p
px{pq{t.r{x}xrp{.sxt.~}.
|pxyp}p.6USP;.A?@D7;.USP.r~}r{sts.
wp.wtt.sxt.s~.}~.wp
t.t}~vw.
x}u~|px~}.~.ewttgpvn{!rtqxg!ghhkece{!
qh!octklwcpc .u~.p}.wtptxr.
x}sxrpx~}=.
x
=.Vjg!ftwi!owuv!dg!ceegrvgf!d{!
swcnkhkgf!gzrgtvu/!
P.wx.x|t;.wtt.x.}~.r~}t}.~u.
~x}x~}.p|~}v.tt.r~}rt}x}v.wt.
|tsxrp{.x{x.~u.|pxyp}p.u~.t.x}.
tpx}v.trxuxr.tr~v}xts.sx~st=.
=.Vjg!uekgpvkhke!gxkfgpeg!owuv!dg!
ykfgn{!cxckncdng/!
cwt.rt}{.p
px{pq{t.spp.p}s.
x}u~|px~}.~}.|pxyp}p.x.}~.
uuxrxt}.~.psst.wt.rwt|x;.
wp|pr~{~v;.~xr~{~v;.p}s.
tuutrx
t}t=.cwt.rxt}xuxr.t
xst}rt.
tvpsx}v.|pxyp}p .rwt|x.xw.
tvps.~.p.trxuxr.rp}}pqx.px}.wp.
r~{s.qt.u~|{pts.x}~.p}spsxts.
p}s.t~srxq{t.s~t.x.}~.rt}{.
p
px{pq{t=.
B=.cwtt.x.p.{prz.~u.prrtts.put.
u~.t.~u.|pxyp}p.}st.|tsxrp{.
t
xx~}=.
Rt}{;.wtt.pt.}~.USP<p~
ts.
|pxyp}p.~sr=.cwt.WWb.p{~.
r~}r{sts.wp.|pxyp}p.s~t.}~.wp
t.
p.rt}{.prrtts.|tsxrp{.t.x}.
tp|t}.x}.wt.d}xts.bpt.~.p.
rt}{.prrtts.|tsxrp{.t.xw.
t
tt.txrx~}=.Prr~sx}v.~.wt.
WWb;.wt.USP.x.}pq{t.~.r~}r{st.
wp.|pxyp}p.wp.p}.prrtpq{t.{t
t{.~u.
G<^HT^HO^23CWR3/UIO
23CWR3
64873!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
rwx~wt}xpI.P.{~}vxsx}p{.s.~u.
btsxw.r~}rx=.Ncpegv!@ICGB¤@CGB=.
@@=.Pt}sx}~.V;.Rwxp}tt.V;.cpv{xp{pt{p<.
brpupx.^.6A?@@7=.Rp}}pqx}~xsI.
^rrt}rt.p}s.|tsxrx}p{.rwt|x=.
Ewtt!Ogf!Ejgo!@G6F7I@?GD¤@?HH=.
@A=.Prw.aW;.b|xw.RV;.bx{t<Zw~s.c\;.
_pttx}.RY.6@HG@7=.Tuutr.~u.st{p.H<.
tpws~rp}}pqx}~{.sx}v.wt.
u~{{xr{p.wpt.~u.wt.wt.|~}zt.
6\prprp.|{pp7=.L!Enkp!Gpfqetkpqn!
Ogvcd!DA6@7ID?¤DD=.
@B=.Qp{t.a[.6@HH@7=.Sv.pqt.~t}xp{.
t
p{px~}.x}.p}x|p{=.Dt!L!Cffkev!
GE6@A7I@DCH¤@DDG=.
@C=.Qp{t.a[;._tr~.fa.6@HHA7=.St{pH<.
ctpws~rp}}pqx}~{.sxrx|x}px~}.x}.
p.p.p.|~st{.u~.rp}}pqx.x}~xrpx~}=.
Pgwtquek!Dkqdgjcx!Tgu!@E6@7IDD¤EA=.
@D=.Qp{t.a[;.Qxvt{~.VT.6A??B7=.Vxst{x}t.
Tghgtgpegu!
p}s.|tw~s~{~vxrp{.t
xt.r~}rt}x}v.
sv.pqt.{xpqx{x.pt|t}=.Ftwi!
@=.Pqp|.SX;.Wx{~}.YU;.atxt.aY;.bwpst.bQ;.
Cneqjqn!Fgrgpf!F?Ib@B¤bC?=.
T{qtxz.cP;.Pttzp.Uc;.Qt}~x.][;.
@E=.Qp}t.V;.[xrz~.e;.cw~|~}.c.6@HGD7=.
Qts.Q\;.Z~t{.Q;.Pqtv.YP;.Sttz.bV;.
Qtwp
x~p{.wp|pr~zx}txr.~u.
\xrwt{{.cU;.\{{xvp}.Z;.Qprrwtx._;.
|pxyp}p=.Ru{ejqrjctoceqnqi{!
\rR}t.Y\;.brwp|qt{p}.\.6A??B7=.
GD6@7ID@¤DE=.
bw~<t|.tuutr.~u.rp}}pqx}~xs.x}.
@F=.Qt}~x.][;.Y~}t.ac.6@HFD7=.
pxt}.xw.WXe¤@.x}utrx~}I.P.
Rpsx~
pr{p.tuutr.~u.~{~}vts.
p}s~|xts;.{prtq~<r~}~{{ts.r{x}xrp{.
st{p<H<tpws~rp}}pqx}~{.x}vtx~}=.
xp{=.Cpp!Kpvgtp!Ogf!@BH6C7IADG¤AEE=.
Enkp!Rjctoceqn!Vjgt!@G6B7IAGF¤AHF=.
A=.Psp|.XQ;.\px}.Qa.6@HHE7=.Rp}}pqxI.
_wp|pr~{~v.p}s.~xr~{~v.x}.p}x|p{. @G=.Qt}~x.][;.Y~}t.ac.6@HG@7=.
p}s.w|p}=.Cffkevkqp!H@I@DGD¤@E@C=.
Rpsx~
pr{p.p}s.|tpq~{xr.
B=.Pvt{{.b;.Stt.f[;.p}s.fx{{t.aT.
r~}xstpx~}.x}.~{~}vts.rp}}pqx}~xs.
6ts=7.6@HGC7=.Vjg!Ecppcdkpqkfu<!
ps|x}xpx~}.x}.|p}=.L!Enkp!Rjctoceqn!
Ejgokecn-!Rjctoceqnqike-!cpf!
A@6G¤H.b{7IA@Cb¤AABb=.
Vjgtcrgwvke!Curgevu/!]t.h~zI.
@H=.Q{p}r~.R;.P{st~}.S;.^vq}.T;.Vp}.
Prpst|xr._t=.
QU;.]}t.Te;.Wpt}qtw{t.\[;.Wpx}.
C=.Pvt{{.b;.Wp{{sx}.\;.[x}svt}.YT;.^w{~}.
Sb.6A??F7=.Rwp}vt.x}.wt.t
p{t}rt.~u.
P;.fxs|p}.\;.Vx{{txt.W;.W~{{xt.[.
}~}<|tsxrp{.trxx~}.sv.t.p}s.
6@HGE7=._wp|pr~zx}txr.p}s.
sv.t.sx~st.x}.wt.d}xts.bptI.
|tpq~{x|.~u.st{p<@<.
@HH@¤@HHA.p}s.A??@¤A??A=.Ftwi!Cneqjqn!
tpws~rp}}pqx}~{.p}s.~wt.
Fgrgpf!H?6A¤B7IADA¤AE?=.
rp}}pqx}~xs.xw.t|wpx.~}.|p}=.
A?=.Q{~rz.aX;.Upx}~.a;.brw{trwt.YP.
Rjctoceqn!Tgx!BG6@7IA@¤CB=.
6@HH@7=.Tuutr.~u.rw~}xr.|pxyp}p.t.
D=.P|txrp}.R~{{tvt.~u._wxrxp}.jPR_l=.
~}.t~t~}t;.{tx}xx}v.w~|~}t;.
6A??G7=.b~x}v.attprw.X}~.wt.
u~{{xr{t.x|{px}v.w~|~}t;.~{prx}.
cwtptxr.a~{t.~u.\pxyp}p=.
p}s.r~x~{.x}.|t}.p}s.~|t}=.Ftwi!
_wx{pst{•wxpI.P|t•xrp}.R~{{tvt.~u.
Cneqjqn!Fgrgpf!AG6A7I@A@¤@AG=.
_wxrxp}J.A??GI._~xx~}._pt.
A@=.Q{~rz.aX;.Upx}~.a;.Qp
t|p}.Z.
6P
px{pq{t.u~|.P|txrp}.R~{{tvt.~u.
6@HHA7=.Prt.tuutr.~u.|pxyp}p.~}.
_wxrxp};.@H?.]=.X}stt}st}rt.\p{{.
r~v}xx~}I.t{px~}wx.~.rw~}xr.
ft;._wx{pst{wxp;._P.@H@?E7=.
tuutr.p}s.|~zx}v.trw}xt=.
E=.P|txrp}.\tsxrp{.P~rxpx~}.jP\Pl=.
Rjctoceqn!Dkqejgo!Dgjcx!CB6B7IH?F¤.
6A??Hp7=.dt.~u.Rp}}pqx.u~.\tsxrp{.
H@F=.
_~t=.Tgrqtv!qh!vjg!Eqwpekn!qp!
AA=.Q~{{p.ZX;.Q~}.Z;.T{stw.S;.cpt.Z;.
Uekgpeg!cpf!Rwdnke!Jgcnvj!4/!
Rpst.Y[.6A??A7=.S~t<t{pts.
F=.P|txrp}.\tsxrp{.P~rxpx~}.jP\Pl=.
}t~r~v}xx
t.tuutr.~u.|pxyp}p.t=.
6A??Hq7=.P\P._~{xrI.\tsxrp{.
Pgwtqnqi{!DHI@BBF¤@BCB=.
\pxyp}p=.W¤HD¤HDA.\tsxrp{.
AB=.Q~{{p.ZX;.T{stw.SP;.\p~rwxz.YP;.Rpst.
\pxyp}p=.
Y[.6A??D7=.]tp{.qpt.~u.up{.
G=.P|txrp}.\tsxrp{.P~rxpx~}.jP\Pl=.
strxx~}<|pzx}v.x}.pqx}t}.|pxyp}p.
6A?@[email protected]}tx|.\ttx}v=.P|txrp}.
t=.PgwtqKocig!AEICG?¤CHA=.
\tsxrp{.P~rxpx~}.W~t.~u.St{tvpt.
AC=.Q~}}t.d.6A?@B7=.Rw~}xr.rp}}pqx.pqt;.
6X¤@B7=.Prrtts.p.yyy/coc.cuup/qti0
st{p<H<tpws~rp}}pqx}~{.p}s.
cuugvu0oggvkpi03124k0k24.tgheqoom.!
w~xs.u}rx~}=.Rjctoceqru{ejkcvt{!
cppqvcvgf/rfh/!
CE6@7IBD¤BE=.
H=.P|txrp}.b~rxt.~u.Pssxrx~}.\tsxrx}t.
AD=.Q~pq~{p.\;.ax}p{sx.\;.Rpp~}._;.
jPbP\l=.6A?@C7=.Pgy![qtm!Vkogu!rp{{.
Rpx{{~}.R;.St{trw.Q;.bwxt.S;.[t.U.
u~.{tvp{xpx~}.~u.|pxyp}p;.PbP\.
V;.Rpt{{p._.6@HHB7=.Rp}}pqx}~xs<.
~}v{.~qytr=.Prrtts.p.
trt~.ttx~}.x}.w|p}.
yyy/cuco/qti0fqeu0fghcwnv.uqwteg0
{tz~rt=.Gwt!L!Dkqejgo!A@C6@7I@FB¤.
rtguutgngcugu0cuco.rtguu.tgngcug.!
@G?=.
p{vkogu.gfkvqtkcn.octklwcpc.3125.8.!
AE=.Qpxsp.S;.X~t.b;._tv~x}x.b;.bp{p.\.
38/rfhAuhxtup?3!
6A??C7=.St{pH<tpws~rp}}pqx}~{<.
@?=.P}stp~}.b;.P{{tqtrz._;.T}v~¶ |.P;.
x}srts.r~}sxx~}ts.{prt.tutt}rt.
asqtv.d.6@HGF7=.Rp}}pqx.p}s.
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
put.x}.t{px~}.~.x.tuutrx
t}t.x}.
tpx}v.p.trxuxr.p}s.tr~v}xts.
sx~st.st.~.{prz.~u.t
xst}rt.xw.
ttr.~.p.r~}xt}.p}s.t~srxq{t.
s~t.wp.x.r~}p|x}px~}.utt=.cwt.
WWb.x}sxrpts.wp.|pxyp}p.ttprw.
x}
txvpx}v.~t}xp{.|tsxrp{.t.
w~{s.x}r{st.x}u~|px~}.~}.wt.
rwt|x;.|p}uprx}v;.p}s.
trxuxrpx~}.~u.|pxyp}p=.cwt.WWb.
uwt.x}sxrpts.wp.p.~rtst.u~.
st{x
tx}v.p.r~}xt}.s~t.~u.
|pxyp}p.w~{s.p{~.qt.st
t{~ts=.
cwttu~t;.wt.WWb.r~}r{sts.wp.
|pxyp}p.s~t.}~.wp
t.p}.prrtpq{t.
{t
t{.~u.put.u~.t.}st.|tsxrp{.
t
xx~}=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
RQ!11111
Hto!11187
Hov!5812
Uhov!5813
p}s.x}prttq~
t}xr{p.t{u<.
ps|x}xpx~}.x}.p=.Gwt!L!Rjctoceqn!
D?E6@7IEB¤EH=.
AF=.Qxt
~vt{.Rb;.Rwx{st.ba.6A???7=.
Rp}}pqx}~xs.pv~}x.xv}p{.p}srx~}.
x}.p.qpx}I.r~|px~}.~u.rp}}pqx}~xs.
pv~}x.x}.trt~.qx}sx}v;.V<~tx}.
prx
px~};.p}s.pst}{{.rr{pt.
x}wxqxx~}=.L!Rjctoceqn!Gzr!Vjgt!
AHD6@7IBAG¤BBE=.
AG=.Q~}.cc;.S~q.Pb.6A??A7=.T}s~rx}t.
tuutr.~u.|pxyp}p=.L!Enkp!Rjctoceqn!
CA6@@.b{7IH?b¤HEb=.
AH=.Q~}t.aV;.ftx|p}.P.6@HG@7=.
Sxrx|x}px
t.x|{.~txt.~u.
st{p.H<tpws~rp}}pqx}~{I.
|trwp}xxr.sxt=.L!Enkp!Rjctoceqn!
A@6G¤H.b{7IAAFb¤ABCb=.
B?=.Qs}t.PY;.Wvwt.Ya.6A??E7=.cwt.
rp}}pqx.xwspp{.}s~|t=.Ewtt!
Qrkp!Ru{ejkcvt{!@H6B7IABB¤ABG=.
B@=.Rpx~x.a\;.U~{x}.af;.Qps.Ye;.
Uxrw|p}.\f.6@HGG7=.Tuutr.~u.
|pxyp}p.~}.wt.pz<t{xrxts.
wx~{~vxrp{.t~}t=.Ftwi!Cneqjqn!
Fgrgpf!A@6B7I@GB¤@GF=.
BA=.Rprx}x.U;.Pxt{{~.R;.Sx.cp}}p.V.6A?@A7=.
X}rtpx}v.st{p<H<tpws~rp}}pqx}~{.
%<H¤cWR7.r~}t}.x}.wtqp{.rp}}pqx.
~
t.x|tI.t|pxr.t
xt.p}s.|tp<.
p}p{x=.Ewtt!Ftwi!Cdwug!Tgx!D6@7IBA¤.
C?=.
BB=.Rwpx.[S;.Qzt.ZP.6@HHC7=._tutt}rt.u~.
wxvw .
t.{~<~t}r.|pxyp}p=.
Rjctoceqn!Dkqejgo!Dgjcx!FIBDF¤BEC=.
BC=.Rwtt.YU;.Zt}sp{{.SP;.\pst}.RP.
6A???7=.Rp}}pqx}~xs.trt~.p}s.
tps.x}.wt.pI.p.r~}sxx~}ts.{prt.
tutt}rt.s=.Ru{ejqrjctoceqnqi{!
)Dgtn*!@D@6@7IAD¤B?=.
BD=.R~|~}.a_;.Qt}x}v.P.6A?@D7=.Sv.p}s.
P{r~w~{.Rpw.axz=.Vtchhke!Uchgv{!Hcevu!
Tgugctej!Pqvg/!S^c.Wb.G@A.@@F=.
fpwx}v~};.SRI.]px~}p{.Wxvwp.
cpuuxr.bput.Ps|x}xpx~}=.
BE=.R~|~}.f\;.Vp}.QU;.R~{{x
t.YS;.
V{p}.\S;.bx}~}.Ub.6A??C7=.
_t
p{t}rt.~u.|pxyp}p.t.sx~st.x}.
wt.d}xts.bptI.@HH@¤@HHA.p}s.A??@¤.
A??A=.LCOC!AH@6@F7IA@@C¤A@A@=.
BF=.R~}t.TY;.Y~w}~}.aT;.\~~t.YS;.a~prwt.
YS.6@HGE7=.Prt.tuutr.~u.|~zx}v.
|pxyp}p.~}.w~|~}t;.qytrx
t.
tuutr.p}s.tu~|p}rt.x}.|p{t.w|p}.
qytr=.Rjctoceqn!Dkqejgo!Dgjcx!
AC6E7I@FCH¤@FDC=.
BG=.R~u~s.Y[;.hp|p|p.c.6A??D7=.
Rp}}pqx}~xs.p}s.wt.x||}t.t|I.
~t}xp{.u~.wt.tp|t}.~u.
x}u{p||p~.sxtptN.L!Pgwtqkoowpqn!
@EE6@¤A7IB¤@G=.
BH=.Sp{x}v.Ya;.S~~s.Sa;.b}.g;.cpqt.Q[;.
ftx.]b;.Rwt}.R;.Qxvv.\[;.bp.Ya;.
St.bZ;.brwp.b\.6A??H7=.
P~rxpx~}.~u.|pxyp}p.t.p}s.wt.
x}rxst}rt.~u.txr{p.vt|.rt{{.|~=.
Ecpegt/!@@D6E7I@A@D¤@AAB=.
C?=.Sp{~}.fb;.\p.a;.[t|qtvt.[;.a~ssp.
QT;.U~}t.aQ.6@HFE7=.X}u{t}rt.~u.
rp}}pqxsx~{.~}.st{p<H<.
tpws~rp}}pqx}~{.tuutr=.Enkp!
Rjctoceqn!Vjgt!@H6B7IB??¤B?H=.
C@=.Sp.T\;._x{~t.]b;.Ps{t.fW;.]pvt{.YT;.
[p}vt.fa.6@HGH7=.cwt.tuutr.~u.H<.
t}ttpws~rp}}pqx}~{.~}.w~|~}t.
t{tpt.p}s.x||}t.u}rx~}=.L!Uvgtqkf!
Dkqejgo!BC6@¤E7IAEB¤AF?=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
CA=.Sp.][;.fpvt}t.SZ;.cp{~._\.6@HGD7=.
\tpt|t}.~u.qp}rt.t.sx}v.
tv}p}rI.|tw~s~{~vxr.xt=.PKFC!
Tgu!Oqpqit!DHIBE¤CF=.
CB=.Sp.][;.[ttrw.b[;.V~{srw|xs.[.6A?@@7=.
cwt.tuutr.~u.t}pp{.|pxyp}p.
t~t.~}.st{x}t}.qtwp
x~.pt.
|tsxpts.q.|tpt.~u.}t~r~v}xx
t.
u}rx~}x}v=.Pgwtqvqzkeqn!Vgtcvqn!
BB6@7I@AH¤@BE=.
CC=.Sp.][;.V~{srw|xs.[;.Sp.a;.[pzq.
R;.axrwps~}.VP.6A?@D7=._t}pp{.
|pxyp}p.t~t;.pvt.~u.|pxyp}p.
x}xxpx~};.p}s.wt.st
t{~|t}.~u.
rw~xr.|~|.x}.~}v.ps{=.
Ru{ejqn!Ogf!CD6G7I@FFH¤@FGF=.
CD=.Stvt}wps.[;.Wp{{.f;.[}zt.\.6A??B7=.
ctx}v.w~wtt.pq~.wt.
t{px~}wx.qttt}.rp}}pqx.t.p}s.
rw~x=.Ftwi!Cneqjqn!Fgrgpf!
F@6@7IBF¤CG=.
CE=.Stvt}wps.[;.Wp{{.fS;.[}zt.\;.
\rVpw.Y;.\r[pt}.Y;.Rp{pqxp.Q;.
fwxtu~s.W;.e~.c.6A??H7=.bw~{s.
qst}.~u.sxtpt.tx|pt.x}r{st.
rp}}pqx.t.p.p.xz.upr~.u~.
rw~xN.RNqU!Ogfkekpg/!
E6H7It@???@BB=.
CF=.Stvt}wps.[;.Sxtzt.[;.Rwx.fc;.
\tsx}p<\~p.\T;.]t|pz.h;.bp|~}.
];.P{~}~.Y;.P}vt|tt.\;.P}w~}.YR;.
Quupt.a;.t.p{.6A?@?7=.T
p{px}v.wt.
sv.t. vptp .wt~.x}v.r~<.
}px~}p{.sppI.r~}xt}r.p}s.
p~rxpx~}.~u.wt.~st.~u.x}xxpx~}.~u.
sv.t.p|~}v.pxrxp}.x}.wt.fW^.
f~{s.\t}p{.Wtp{w.b
t=.Ftwi!
Cneqjqn!Fgrgpf!@?G6@¤A7IGC¤HF=.
CG=.St._t~rt{{x._[;.Sx.\p~.e.6A??H7=.P}.
x}~srx~}.~.wt.t}s~rp}}pqx}~xs.
t|I.u~|.wt.tp{.~.wt.{pt.
r~}rt=.Dguv!Rtcev!Tgu!Enkp!Gpfqetkpqn!
Ogvcd!AB6@7I@¤@D=.
CH=.Stp|t}.~u.Wtp{w.p}s.W|p}.
bt
xrt.jWWbl.6A?@D7=.Qpx.u~.wt.
tr~||t}spx~}.u~.|px}px}x}v.
|pxyp}p.x}.brwts{t.X.~u.wt.
R~}~{{ts.bqp}rt.Pr=.
D?=.St
p}t.fP;.Wp}.[;.Qtt.P;._ttt.
aV;.bt
t}~}.[P;.Vxuux}.V;.Vxq~}.S;.
\p}st{qp|.P;.Tx}vt.P;.\trw~{p|.a.
6@HHA7=.X~{px~}.p}s.rt.~u.p.qpx}.
r~}xt}.wp.qx}s.~.wt.rp}}pqx}~xs.
trt~=.Uekgpeg!ADG6D?H?7I@HCE¤@HCH=.
D@=.Stt.f[;.\px}.Qa;.\p.T[.6@HGC7=.
Rp}}pqx}~xs.tt~x~|tI.
wp|pr~{~vxrp{.tuutr=.X}.b|xw.SU=.
6Ts=7.ETE!Jcpfdqqm!qh!uvgtgqkuqogtu<!
ftwiu!kp!ru{ejqrjctoceqnqi{-!B@F¤AE.
6Q~rp.ap~};.U[;.RaR._t7=.
DA=.Sv.T}u~rt|t}.Ps|x}xpx~}.6A?@D7=.
Ftwiu!qh!Cdwug/!
DB=.Sb\¤D.6A?@B7=.Sxpv}~xr.p}s.bpxxrp{.
\p}p{.~u.\t}p{.Sx~st;.Uxuw.
Tsxx~}=.P|txrp}._rwxpxr.
P~rxpx~}=.fpwx}v~};.SRI.P|txrp}.
_rwxpxr._q{xwx}v=.
DC=.Txt}tx}.cZ;.\tx{t.YY.6A?@D7=.Tuutr.
~u.rp}}pqx}~xs.~}.c<rt{{.u}rx~}.p}s.
txp}rt.~.x}utrx~}=.L!Pgwtqkoowpg!
Rjctoceqn!@?6A7IA?C¤A@E=.
DD=.T{sxsvt.YR;.\w.[[;.[p}suxt{s._f.
6@HH@7=.Rp}}pqx}~xs.p}s.wt.
wx~rp|p{.v{r~r~xr~xs.trt~I.
trt}.ux}sx}v.p}s.~xq{t.xv}xuxrp}rt=.
Uvgtqkfu!DE6D7IAAE¤AB@=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
DE=.T{b~w{.\P;.b{pst.S.6A??D7=.Rwt|xrp{.
r~}xt}.~u.|pxyp}pI.cwt.r~|{t.
|xt.~u.}pp{.rp}}pqx}~xs=.Nkhg!Uek!
FGIDBH¤DCG=.
DF=.Up}.ae;.Wtxw|p}.bY;.Q}zt.TQ;.
_xrz~w.fQ.6@HHG7=.Prt.p}s.
txsp{.tuutr.~u.|pxyp}p.x}.w|p}=.
Rjctoceqn!Dkqejgo!Dgjcx!E?6C7IFFF¤.
FGC=.
DG=.Hgfgtcn!Tgikuvgt!6@HHA7=.\pxyp}p.
brwts{x}v._txx~}J.St}xp{.~u._txx~}J.
at|p}s ¥Sv.T}u~rt|t}.
Ps|x}xpx~};.Ux}p{.^st=.Hgf!Tgikuvt!
DF6DH7I@?CHH¤@?D?G=.
DH=.Hgfgtcn!Tgikuvgt!6@HHH7=.atrwts{x}v.~u.
wt.U~~s.p}s.Sv.Ps|x}xpx~}.
P~
ts._~sr.R~}px}x}v.b}wtxr.
S~}pqx}~{.j6¤7<st{p.H<6p}7<.
ctpws~rp}}pqx}~{l.x}.btp|t.^x{.
p}s.T}rp{pts.x}.b~u.Vt{px}.
Rp{t.U~|.brwts{t.XX.~.brwts{t.
XXXJ.Ux}p{.a{t; .Hgf!Tgikuvt!
EC6@AF7IBDHAG¤BDHB?=.
E?=.Hgfgtcn!Tgikuvgt!6A??@7=.]~xrt.~u.St}xp{.
~u._txx~}I.Qpx.u~.wt.
atr~||t}spx~}.u~.\px}px}x}v.
\pxyp}p.x}.brwts{t.X.~u.wt.
R~}~{{ts.bqp}rt.Pr; .Hgf!Tgikuvt!
EE6FD7IA??BG¤A??FE=.
E@=.Utv~}.S\;.W~~~s.[Y;.axsst.T\.
6A??D7=.ct.~u.rpp{.{x}zpvt.qttt}.
rp}}pqx.t.p}s.rw~xr.|~|=.
Cffkevkqp/!@??6B7IBDC¤BEE=.
EA=.U~}t.\P;.Q~{{p.ZX;.R}wp._Y;.P{|txsp.
__;.Y}vt|p}.U;.[pp}ytxp.aa;.Qtp}.
aP;.[prtsp.P[.6A?@@7=.Rp}}pqx.t.
qtu~[email protected]}s.qtt}.ttrx
t.
u}rx~}x}v=.Dt!L!Ru{ejkcvt{!@HG6E7ICCA¤.
CCF=.
EB=.Uxts._P;.fpzx}~}.Q;.Vp}.P;.Z}xvw.
a\.6@HG?7=.Rwp}vx}v.pt}.~u.~u.
sv.t.x~.~.p}s.sx}v.tv}p}rI.
p.~trx
t.s=.Ftwi!Cneqjqn!
Fgrgpf!E6D7IBAB¤BCB=.
EC=.Uxts._P;.fpzx}~}.Q.6@HH?7=.BE<.p}s.
CG<|~}w.}t~qtwp
x~p{.u~{{~<.~u.
rwx{st}.t}pp{{.t~ts.~.
|pxyp}p;.rxvptt.p}s.p{r~w~{=.L!Fgx!
Dgjcx!Rgfkcvt!@@ICH¤DG=.
ED=.Uxts._P;.fpzx}~}.Q;.Vp.a.6@HHA7=.P.
u~{{~<.s.~u.pt}x~}p{.qtwp
x~.
x}.E<tp<~{s.rwx{st}.t~ts.
t}pp{{.~.|pxwp}p;.rxvptt.p}s.
p{r~w~{=.Pgwtqvqzkeqn!Vgtcvqn!@CIAHH¤.
B@@=.
EE=.Uxts._P;.fpzx}~}.Q;.Vp.a.6@HHG7=.
Sxuutt}xp{.tuutr.~}.r~v}xx
t.
u}rx~}x}v.x}.H<.~.@A<tp.~{s.
t}pp{{.t~ts.~.rxvptt.p}s.
|pxwp}p=.Pgwtqvqzkeqn!Vgtcvqn!
A?6B7IAHB¤B?E=.
EF=.Uxts._P.6A??A7=.Ps~{trt}.t}pp{{.
t~ts.~.|pxyp}pI.tp|x}px~}.~u.
uprt.~u.r~|{t.qtwp
x~.p}s.
r~|px~}.xw.wt.x}u{t}rt.~u.x}.
t~.rxvptt=.L!Enkp!Rjctoceqn!CA6@@.
b{7IHFb¤@?Ab=.
EG=.Uxts._P;.fpzx}~}.Q;.Vp.a.6A??D7=.
]t~r~v}xx
t.r~}tt}rt.~u.
|pxwp}p¥p.r~|px~}.xw.t<sv.
•t•u~•|p}rt=.Pgwtqvqzkeqn!Vgtcvqn!
AF6A7IAB@¤ABH=.
EH=.U}v.\;.Vp{{pvwt.R;.\prwp.\.6@HHH7=.
[}v.p}s.pt~<sxvtx
t.rp}rt.x}.
~}v.|pxyp}p.|~zt=.Vwoqtk!
GD6A7I@C?¤@CA=.
RQ!11111
Hto!11188
Hov!5812
Uhov!5813
64874!
F?=.Vpt._;.Ypuut.P;.R~t{p}s.Y.6A?@C7=.
Rp}}pqx.|~zx}v.p}s.txp~.
wtp{wI.r~}xstpx~}.~u.wt.{xtpt=.
Tgurktqnqi{!@H6D7IEDD¤EEA=.
F@=.Vw~{p}s.b;.\pwt.Wf;.bx|~}x}.U;.
Ux{{x~{.S;.Zxtuut•.Q[;.\p{s~}ps~.a.
6A??A7=.\~x
px~}p{.tuutr.~u.
rp}}pqx}~xs.pt.|tsxpts.q.|<~x~xs.
p}s.zpp<~x~xs.trt~=.L!Pgwtquek!
AA6B7I@@CE¤@@DC=.
FA=.V~{s.[W;.Qp{t.a[;.Qpt.a[;.Qx.Sc;.
\px}.Qa.6@HHA7=.P.r~|px~}.~u.wt.
sxrx|x}px
t.x|{.~txt.~u.
st{p.H<tpws~rp}}pqx}~{.p}s.R_.
DD;HC?.x}.p.p}s.wt.|~}zt=.L!
Rjctoceqn!Gzr!Vjgt!AEA6A7ICFH¤CGE=.
FB=.V~{srw|xs.[;.axrwps~}.VP;.fx{{u~s.
YP;.Sp.][.6A??G7=._t}pp{.|pxyp}p.
t~t.p}s.x}t{{xvt}rt.t.
tu~|p}rt.p.pvt.E=.L!Co!Cecf!Ejknf!
Cfqngue!Ru{ejkcvt{!CF6B7IADC¤AEB=.
FC=.V~{srw|xs.[;.axrwps~}.VP;.fx{{u~s.
YP;.bt
t~}.bV;.Sp.][.6A?@A7=.brw~~{.
prwxt
t|t}.x}.@C<tp<~{s.~w.
t}pp{{.t~ts.~.|pxyp}p=.
Pgwtqvqzkeqn!Vgtcvqn!BC6@7I@E@¤@EF=.
FD=.V~}v.W.Y;.cpwzx}.S_;.bx||~}.\b;.
Rp{
ptt.Q;.bwpx~.QY.6@HGC7=.Prt.
p}s.qprt.q~}rwxp{.tuutr.~u.~p{.
rp}}pqx}~xs=.Enkp!Rjctoceqn!Vjgt!
BD6@7IAE¤BA=.
FE=.V~}v.Y_;.^}px
x.Tb;.Xwxv~.W;.[x.`a;.
cpv{xput~._P;.Qr~.P;.dw{.Va.6A??E7=.
Rp}}pqx}~xs.RQA.trt~I.
x||}~wx~rwt|xrp{.{~rp{xpx~}.x}.p.
q•px}=.Dtckp!Tgu!@?F@6@7I@?¤AB=.
FF=.V~}x~tz.f;.[}}.R;.Up}.g;.]p{p.b;.
[}st{{.S;.Wxzx}.af=.T}s~rp}}pqx}~xs.
A<pprwxs~}{.v{rt~{.x.p.u{{.pv~}x.
w~vw.w|p}.t.A.rp}}pqx}~xs.
trt~I.p}pv~}x|.q.p}p}sp|xst=.
Oqn!Rjctoceqn!DF6D7I@?CD¤@?D?=.
FG=.V~}p{t.a.6A??F7=.Prt.p}s.}~}<prt.
tuutr.~u.rp}}pqx.~}.qpx}.u}rx~}x}v.
p}s.}t~rw~{~vxrp{.tu~|p}rt=.
Pgwtqru{ejqn!Tgx!@F6B7IBCF¤BE@=.
FH=.V~}p{t.b;.Rtqtxp.\;.Ut}p}st<ax.Y.
6A??D7=.Rp}}pqx}~xs.~{tp}rt.p}s.
stt}st}rtI.p.t
xt.~u.sxt.x}.
{pq~p~.p}x|p{=.Rjctoceqn!Dkqejgo!
Dgjcx!G@6A7IB??¤B@G=.
G?=.V~t.a[;.Tp{tx}t.\.6A??F7=.
\pxyp}p .trtx
ts.pssxrx
t}tI.P.
t.~u.r{x}xrxp}.p}s.ttprwt=.X}.
\=.Tp{tx}t;.6Ts=7.Rqv!rqnkvkeu<!Vjg!
equv!qh!rtqjkdkvkqp/!]t.h~zI.^u~s.
d}x
tx._t=.
G@=.Vxuuxw<[t}stx}v.\U;.fxv|p}.Yc;.
_x}rt.
p}.[ttt}.P;.Wxyqtv.bR;.
Wxx}z.PR;.^|t{.Y;.etw{.UR;.
p}.
^.Y;.bppq.W;.e~{{tqtvw.fP.6A?@B7=.
Rp}}pqx.t.p}s.
{}tpqx{x.u~.
rw~x.x}.tp{.ps~{trt}rt¥p.
caPX[b.s=.Cffkevkqp!@?G6C7IFBB¤.
FC?=.
GA=.V~t}wt|t}.U.6A??B7=.
_wp|pr~zx}txr.p}s.
wp|pr~s}p|xr.~u.rp}}pqx}~xs=.
Enkp!Rjctoceqmkpgv!CA6C7IBAF¤BE?=.
GB=.Vqt.bP;.bpvp.ZP;.Spw{vt}.\Z;.
aprx}t.\;.[zp.bT.6A?@A7=.Pvt.~u.~}t.
~u.|pxyp}p.t.p}s.ttrx
t.u}rx~}=.
Ru{ejqn!Cffkev!Dgjcx!AE6B7ICHE¤D?E=.
GC=.Wprzp|.SV.6A?@D7=.Rp}}pqx.p}s.~ztI.
t|pxr.ppxp{.~u.rpt.t~=.
Uvtqmg!CE6B7IGDA¤GDE=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64875!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
GD=.Wp{{.fS;.[}zt.\.6A??D7=.X.rp}}pqx.
p.vptp.svN.ctx}v.w~wtt.
pq~.wt.t{px~}wx.qttt}.rp}}pqx.
t.p}s.wt.t.~u.~wt.x{{xrx.sv=.
Ftwi!Cneqjqn!Tgx!AC6@7IBH¤CG=.
GE=.Wp{{.f;.Stvt}wps.[.6A?@C7=.cwt.ps
tt.
wtp{w.tuutr.~u.rw~}xr.rp}}pqx.t=.
Ftwi!Vguv!Cpcn!E6@¤A7IBH¤CD=.
GF=.Wp}t.\;.fps.Pb;.R~|t.bS;.U~{x}.
af;.Uxrw|p}.\f.6@HHH7=.Pqx}t}rt.
|~|.u~{{~x}v.|~zts.|pxyp}p.
x}.w|p}=.Ru{ejqrjctoceqnqi{!)Dgtn*!
@C@6C7IBHD¤C?C=.
GG=.Wpst.b;.axtq~rz.b.6@HHF7=.
R~}rt}px~}<tuutr.t{px~}wx.~u.
st{p<H<tpws~rp}}pqx}~{.p}s.
tsxrx~}.~u.rw~~xr.tuutr.put.
|~zx}v.|pxyp}p=.Kpvgtpcvkqpcn!Lqwtpcn!
qh!Enkpkecn!Rjctoceqnqi{!cpf!
Vjgtcrgwvkeu/!BD6C7I@DD¤@DH=.
GH=.Wpx}.Sb;.^ Qxt}.R_;.Pxpr~|qt.\;.
Q~vt.V;.Qrw~{.Z;.Qs}t.P;.
R~|~}.f\;.R~{t.c;.[x}v.f;._t.
]\;.brwrzx.\;.Vp}.QU.6A?@B7=.Sb\¤.
D.rxtxp.u~.qp}rt.t.sx~stI.
tr~||t}spx~}.p}s.px~}p{t=.Co!L!
Ru{ejkcvt{!@F?6G7IGBC¤GD@=.
H?=.Wtxw|p}.bY;.Wtx.\P;.Wt}}x}vuxt{s.
YT;.R~}t.TY.6@HH?7=.Prt.p}s.txsp{.
tuutr.~u.|pxyp}pI.~ux{t.~u.{p|p.
cWR.{t
t{;.wx~{~vxrp{;.qytrx
t;.
p}s.tu~|p}rt.|tpt=.Rjctoceqn!
Dkqejgo!Dgjcx!BF6B7IDE@¤DED=.
H@=.Wtzt}wp|.\;.[}}.PQ;.[x{t.\S;.
Y~w}~}.\a;.\t{
x}.[b;.st.R~p.Qa;.
axrt.ZR.6@HH?7=.Rp}}pqx}~xs.trt~.
{~rp{xpx~}.x}.qpx}=.Rtqe!Pcvn!Cecf!Uek!
GFI@HBA¤@HBE=.
HA=.Wt•zt}wp|.\.6@HHA7=.Rp}}pqx}~xs.
trt~.{~rp{xpx~}.x}.qpx}I.
t{px~}wx.~.|~~.p}s.tps.
t|=.Cpp!P![!Cecf!Uek!EDCI@H¤BA=.
HB=.Wt}x}v.aX;.W~~zt.fS;.Y~}t.ac.6@HGE7=.
ctpws~rp}}pqx}~{.r~}t}.p}s.
sxuutt}rt.x}.|pxyp}p.|~zx}v.
qtwp
x~=.Ru{ejqrjctoceqnqi{!)Dgtn*!
H?6A7I@E?¤@EA=.
HC=.Wx
~}t}.Y;.V~~sx}.ab;.[x.Rc;.ct.VT;.
i~vwqx.bb;.\~t.R;._xzt.ef;.e~{z~.
]S;.Wtx.\P;.X}}x.aQ.6A?@A7=.
at
txq{t.p}s.tvx~}p{{.t{trx
t.
s~}tv{px~}.~u.qpx}.rp}}pqx}~xs.RQ.
@.trt~.x}.rw~}xr.spx{.rp}}pqx.
|~zt=.Oqn!Ru{ejkcvt{!@F6E7IECB¤ECH=.
HD=.Wx
t{.a[;.\~wt.fP;.W~uu|p}.Uf.
6@HEE7=.X~{px~}.~u.p}<%H<.
tpws~rp}}pqx}~{.u~|.|pxwp}p=.L!
Co!Ejgo!Uqe!GGI@GBA¤@GBB=.
HE=.W~{{xt.[T;.Vx{{txt.WZ.6@HFB7=.St{p<.
G<.p}s.st{p<H<tpws~rp}}pqx}~{.
r~|px~}.x}.|p}.q.~p{.p}s.
x}p
t}~.ps|x}xpx~}=.Enkp!
Rjctoceqn!Vjgt!@C6B7IBDB¤BDF=.
HF=.W~{{xt.[T.6@HGE7=.Wtp{w.ptr.~u.
rp}}pqx=.Rjctoceqnqikecn!Tgx!BI@¤A?=.
HG=.W~{{xt.[T.6@HGG7=.Rp}}pqx.6[xtpt.
t
xt7=.Cevc!Ru{ejkcvt!Uecpf!)Uwrrn*!
FGI@?G¤@@G=.
HH=.W~{t.PR;.Qtx
~vt{.Rb;.Rwx{st.ba;.
Stps{t.bP;.Wp|~}.aT;._~x}~.[Y.
6A??C7=.Rp}}pqx}~xs.wx~{~v.p}s.
wp|pr~{~vI.B?.tp.~u.~vt=.
Pgwtqrjctoceqnqi{!CF.b{IBCD¤BDG=.
@??=.Wp}v.hW;.iwp}v.iU;.cpwzx}.S_;.Ut}v.
Q;.bpxu.Z;.Wpwxqt.Z.6A?@D7=.P}.
txst|x~{~vxr.t
xt.~u.|pxyp}p.p}s.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
rp}rtI.p}.spt=.Ecpegt!Grkfgokqn!
Dkqoctmgtu!Rtgx!AC6@7I@D¤B@=.
@?@=.Wtx.\P;.bp|~}.PW;.W~{xrz.QY;.
Qt}}x}vuxt{s.YT;.R~}t.TY.6@HHAp7=.
Rwpprtxpx~}.~u.wt.pq~x~}.wpt.
~u.|pxyp}p.|~zx}v=.Enkp!Rjctoceqn!
Vjgt!DAIB@¤C@=.
@?A=.Wtx.\P;.Qt}}x}vuxt{s.YT;.R~}t.TY.
6@HHAq7=.Q{~~s.Rp}}pqx}~xs=.@=.
Pq~x~}.~u.cWR.p}s.u~|px~}.~u.@@¤.
?W¤cWR.p}s.cWR.R^^W.sx}v.p}s.
put.|~zx}v.|pxyp}p=.L!Cpcn!Vqzkeqn!
@E6D7IAFE¤AGA=.
@?B=.W}.RP;.Y~}t.ac.6@HG?7=.c~{tp}rt.
p}s.sx~xx~}.~u.tpws~rp}}pqx}~{.
x}.|p}=.L!Rjctoceqn!Gzr!Vjgt!A@D6@7IBD¤.
CC=.
@?C=.X{p}.PQ;.Vt
x}.P;.R~{t|p}.\;.T{b~w{.
\P;.st.fx.W.6A??D7=.
]t~wx~{~vxrp{.p}s.qytrx
t.
~ux{t.~u.|pxyp}p.xw.
px}v.
r~}rt}px~}.~u.rp}}pqx}~xs=.Dgjcx!
Rjctoceqn!@E6D¤E7ICGF¤CHE=.
@?D=.X}xt.~u.\tsxrx}t.6@HGA7=.Sx
xx~}.~u.
Wtp{w.brxt}rt._~{xr=.\pxyp}p.p}s.
Wtp{wI.at~.~u.p.bs.q.p.
R~||xtt.~u.wt.X}xt.~u.\tsxrx}t;.
Sx
xx~}.~u.Wtp{w.brxt}rt._~{xr=.
fpwx}v~};.SRI.]px~}p{.Prpst|.
_t;.@HGA=.
@?E=.X}xt.~u.\tsxrx}t.6@HHH7=.Sx
xx~}.~u.
]t~rxt}rt.p}s.Qtwp
x~p{.Wtp{w=.
\pxyp}p.p}s.\tsxrx}tI.Ptx}v.wt.
brxt}rt.Qpt=.fpwx}v~};.SRI.]px~}p{.
Prpst|._t;.@HHH=.
@?F=.Y~wp}~}.T;.Wp{{sx}.\\;.Pvt{{.b;.
W~{{xt.[T;.Vx{{txt.WZ.6@HGH7=.
ct|x}p{.t{x|x}px~}.{p|p.wp{u<{xut.~u.
st{p.@<tpws~rp}}pqx}~{.6st{p.@¤.
cWR7.x}.wtp
.t.~u.|pxyp}p=.Gwt!L!
Enkp!Rjctoceqn!BF6B7IAFB¤AFF=.
@?G=.Y~}t.ac;.Qt}~x.][;.Wt|x}v.aX.
6@HG@7=.R{x}xrp{.t{t
p}rt.~u.rp}}pqx.
~{tp}rt.p}s.stt}st}rt=.L!Enkp!
Rjctoceqn!A@I@CBb¤@DAb=.
@?H=.Y~}t.ac.6A??A7=.Rpsx~
pr{p.t|.
tuutr.~u.|pxyp}p=.L!Enkp!Rjctoceqn!
CA6@@b{7IDGb¤EBb=.
@@?=.Yx}~
p.i;.cp}sp.V;.atswx.VW;.
V~{sqtv.ba.6A??B7=.bt{u<ps|x}xpx~}.
~u.st{pH<tpws~rp}}pqx}~{.6cWR7.q.
sv.}p'¶
t.xt{.|~}zt=.
Ru{ejqrjctoceqnqi{!)Dgtn*!@EH6A7I@BD¤.
@C?=.
@@@=.Yx}~
p.i;.cp}sp.V;.\}p._;.
V~{sqt•v.ba.6A??C7=.cwt.~•x~xs.
p}pv~}x.}p{t~}t.tsrt.wt.
tx}u~rx}v.tuutr.~u.St{p.H.
tpws~rp}}pqx}~{.6cWR7.x}.xt{.
|~}zt=.Ru{ejqrjctoceqnqi{!)Dgtn*!
@FB6@¤A7I@GE¤@HC=.
@@A=.Zp}st{.S.6@HFD7=.bpvt.x}.ps~{trt}.
x}
~{
t|t}.x}.sv.t=.Uekgpeg!
@H?IH@A¤H@C=.
@@B=.Zp}st{.SQ;.Rwt}.Z.6A???7=.ct.~u.
|pxyp}p.t.q.{~}vxsx}p{.r~t=.
L!Uvwf!Cneqjqn!E@6B7IBEF¤BFG=.
@@C=.Zp}x~{.XV;.bwxpzpp.X;.Zpx}zx.];.
_utut•|p}.P;.Rp•{x}x.TP.6@HFC7=.
Rp}}pqxsx~{.x}tutt.xw.wt.tuutr.~u.
st{p.H<tpws~rp}}pqx}~{.x}.|p}=.Gwt!
L!Rjctoceqn!AG6@7I@FA¤@FF=.
@@D=.Zp}x~{.XV;.bwxpzpp.X;.cpzpwpwx.a];.
Z}~qt{.T;.\.aT.6@HFD7=.Tuutr.~u.
st{pH<tpws~rp}}pqx}~{.p}s.
rp}}pqx}~{.x}.|p}=.Rjctoceqnqi{!
@B6E7ID?A¤D@A=.
RQ!11111
Hto!11189
Hov!5812
Uhov!5813
@@E=.Ztt}.[.A}s;._ttxp.S;.[px|t.f.
6A?@C7=.bt{u<t~ts.{xutx|t.|pxyp}p.
t.p}s.x}t{tzx}<E.{t
t{.x}.|xss{t<.
pvts.Puxrp}.P|txrp}=.Ftwi!Cneqjqn!
Fgrgpf!@C?I@DE¤@E?=.
@@F=.Zxz.Y\;.st.fx.W.6@HHH7=.at~}t.~.
~p{.st{pH<tpws~rp}}pqx}~{.x}.
utt}.p}s.x}utt}.|pxyp}p.t=.
Rjctoceqn!Dkqejgo!Dgjcx!EB6@7I@BF¤.
@CA=.
@@G=.Ztt.a;.
p}.^.Y;.[xtq.a;.fxrwt}.
Wd;.Wt}t.R.6A?@@7=.S~.rp}}pqx.p}s.
qp}xrx.r~<pxrxpt.x}.rpx}v.
rw~xN.T
xst}rt.u~|.p.@?<tp.
u~{{~<.r~w~.s=.Ru{ejqn!Ogf!
C@6@?7IA@A@¤A@AH=.
@@H=.Zwp{t.X;.W||t.\;.\x{{t.R;.
bt}t<d}ttvt.Q;.V}wt.e;.
ftrws~}.W;.Qpxp.WY;.U{txrwwprzt.
ff.6@HHH7=.Tuutr.~u.rp}}pqx.t.~}.
r~v}xx
t.u}rx~}.p}s.sx
x}v.pqx{x=.L!
Enkp!Ru{ejkcvt{!E?6E7IBHD¤BHH=.
@A?=.[pr~}.YR;.Rp~{{.YS;.cp~}.T;.
Rpt{p~.TY;.Qt}tx}.[;.R~tx.eZ.
6A?@A7=._~{px~}<qpts.rpt<r~}~{.
s.~u.trtpx~}p{.sv.t.p}s.tx.
rp}rt.xz.r~}ux|.p}.p~rxpx~}.
qttt}.|pxyp}p.t.p}s.
}~}t|x}~|p.xz=.Ecpegt!@@GIDBFC¤.
DBGB=.
@A@=.[tt.\W;.Wp}r~.aY.6A?@@7=.Tuutr.~u.
|~zx}v.rp}}pqx.~}.{}v.u}rx~}=.Gzr!
Tgx!Tgur!Ogf!D6C7IDBF¤DCE=.
@AA=.[t|qtvt.[;.bx{qttx}.bS;.Pt{~s.Y;.
Z~x}.XY.6@HF?7=.\pxwp}pI.sxt.~}.
wt.sx~xx~}.p}s.|tpq~{x|.~u.st{p<.
H<tpws~rp}}pqx}~{.x}.|p}=.Uekgpeg!
F?I@BA?¤@BAA=.
@AB=.[t|qtvt.[;.ftx.Y[;.fpp}pqt.P\;.
Vp{p}t.X\;.fp.aY;.Rps~}._e.
6@HFAp7=.St{p<H<tpws~rp}}pqx}~{I.
t|~p{.r~t{px~}.~u.wt.rw~{~vxrp{.
tuutr.p}s.q{~~s.{t
t{.put.
px~.
~t.~u.ps|x}xpx~}=.Pgy!Gpi!L!Ogf!
AGE6@B7IEGD¤EGG=.
@AC=.[t|qtvt.[;.Rpqtt.aT;.a~t.W\.
6@HFAq7=.@@<Ws~<%H<.
tpws~rp}}pqx}~{I.wp|pr~{~v;.
sx~xx~}.p}s.|tpq~{x|.~u.p.|py~.
|tpq~{xt.~u.|pxwp}p.x}.|p}=.Uekgpeg!
FFIEA¤EB=.
@AD=.[t|qtvt.[;.aqx}.P.6@HFD7=.cwt.
wx~{~vxr.sx~xx~}.~u.|pxwp}p.x}.
|p}=.Nkhg!Uek!@FI@EBF¤@ECA=.
@AE=.[x.\¤R;.Qps.YT;.Sx\pvvx~.RY;.[psx.
Pa;.c~}v;.Zh;.[x.V.6A?@A7=.\pxyp}p.
t.p}s.|~~.
twxr{t.rpwt=.
Grkfgokqnqike!Tgx!BCIED¤FA=.
@AF=.[xv~x.P;.Vp~.R_;.a~qx}~}.YW.6@HHG7=.
Tuutr.~u.|pxyp}p.~}.tx{xqx|;.
rw~|~~.tu~|p}rt;.p}s.
x|{pts.sx
x}v=.Dgjcx!Rjctoceqn!
H6F7IDHH¤E?H=.
@AG=.[xspw{.Z\;._xrt.Yb.6A?@A7=.X}rtpts.
|pxyp}p.t.p}s.vt}st.tsxr.~~t.
r~v}xx
t.u}rx~}x}v.x}.ps~{trt}.p}s.
t|tvx}v.ps{=.L!Kpv!Pgwtqru{ejqn!Uqe!
@G6C7IEFG¤EGG=.
@AH=.[~}.\Y;.Qp.Y[;._p}x~}.\b;.c~~|t.
a;.Txt}.b;.gxp}.W;.cp}v.\c.6A??C7=.
]t~rw~{~vxrp{.r~}tt}rt.~u.
tv{p.|pxyp}p.tI.p.x}.s=.
Ru{ejqn!Ogf!BC6F7I@ABH¤@AD?=.
@B?=.\przxt.Z;.[px.h;.ftt}q~tz.a;.
\xrwt{{.a.6@HHD7=.Rp}}pqx}~xs.prx
pt.
p}.x}ps{.trxux}v.~px|.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
r~}srp}rt.p}s.x}wxqx.`<t.rp{rx|.
rt}.x}.PcA?.rt{{.p}utrts.xw.
p.qpx}.rp}}pqx}~xs.trt~=.L!
Pgwtquek!@D6@?7IEDDA¤EDE@=.
@B@=.\p{s~}ps~.a.6A??A7=.bs.~u.
rp}}pqx}~xs.stt}st}rt.x}.p}x|p{=.
Rjctoceqn!Vjgt!HD6A7I@DB¤@EC=.
@BA=.\p{x}~zp.Q;.Qpp}~zp<Zrz~.\;.
brw{xrzt.T.6A?@A7=.cxwpxr.q{~~s.
tt.t~}t.~.rp}}pqx}~xsI.s~.
t.}stp}s.wt.|trwp}x|N.Dt!L!
Rjctoceqn!@ED6F7IA?FB¤A?GG=.
@BB=.\p}xt<Vprxp.T;.ip||x.b;.Sp{|p}.
R;.Wt||x}v~}.c;.P}stp~}.b;.
P{{tqtrz._.6A?@A7=.Rp}}pqx;.
rwx~wt}xp.p}s.~wt.}~}<puutrx
t.
rw~tI.BD.tp.~u.u~{{~<.~u.p.
~{px~}<qpts.r~w~=.Ru{ejqn!Ogf!
CA6E7I@BA@¤@BAG=.
@BC=.\pt||p}x.X;.[ptx.P;._prx}x.\;.
[~
trxr.\;._{prxsx.VU;._tvx.V.6A??C7=.
Sxpv}~xr.p}s.|~|p~{~vxrp{.
utpt.x}.rw~}xr.rw~xr.pxt}.
prr~sx}v.~.rp}}pqx.t.p=.L!
Ru{ejqcevkxg!Ftwiu!BE6A7IABD¤AC@=.
@BD=.\rRp.Pc.6A?@D7=.\pxyp}p.p}s.
sx
x}v.x}.wt.d}xts.bptI.t
p{t}rt;.
xz;.p}s.{p=._tt}ts.p.wt.Rpp{.
Prpxp{.b~rxt.bx}v.\ttx}v=.
R~{~ps~.bx}v;.R^=.\p.@H;.A?@D=.
@BE=.\r\pw~}.[a;.Vx}qv.QR;.[p|q.aY.
6A??G7=.Rp}}pqx}~xs.pv~}x.
sxuutt}xp{{.qxt.u~.wt.
sxrx|x}px
t.x|{.tuutr.~u.
St{p6H7<tpws~rp}}pqx}~{.x}.RDFQ[>
EY.|xrt=.Ru{ejqrjctoceqnqi{!)Dgtn*!
@HG6C7ICGF¤CHD=.
@BF=.\r\pw~}.[a.6A??H7=.Ppt}.puux}x.
tx|pt.~u.x|~}pqp}.x}.r~|qx}px~}.
xw.p}p}sp|xst.p}s.rwt|xrp{.p}p{~v.
~u.p}p}sp|xst.x}.wt.|~}zt.
sxrx|x}px}v.St{pH<.
tpws~rp}}pqx}~{=.
Ru{ejqrjctoceqnqi{!)Dgtn*!A?B6A7IA@H¤.
AAG=.
@BG=.\trw~{p|.a.6@HFB7=.Rp}}pqx}~xs.
rwt|x=.X}.\trw~{p|;.a=.6ts=7.
\pxyp}p;.=A¤GG.6]t.h~z;.]h;.
Prpst|xr._t;.X}r=7=.
@BH=.\trw~{p|.a;.Qt}<bwpqp.b;.Wp}.[;.
[xv|z.\;.Zp|x}zx.]T=;.brwp.Pa;.
V~wt.P;.P{|~v.b;.\px}.Qa;.R~|~}.
Sa;._ttt.aV;.Vxuux}.V;.Qptxrw.\;.
Qpv.Y;.e~vt{.i.6@HHD7=.Xst}xuxrpx~}.~u.
p}.t}s~vt}~.A<|~}~v{rtxst;.tt}.
x}.rp}x}t.v;.wp.qx}s.~.rp}}pqx}~xs.
trt~=.Dkqejgo!Rjctoceqn!D?6@7IGB¤.
H?=.
@C?=.\tw|tsxr.i;.Rwp}s•p.b;.b{pst.S;.
St}wp|.W;.U~t.b;._pt{.Pb;.a~.bP;.
Zwp}.XP;.T{b~w{.\P.6A?@?7=._~t}r.
t}s.~u.% <cWR.p}s.~wt.rp}}pqx}~xs.
x}.r~}uxrpts.rp}}pqx.tppx~}.
u~|.@HHB.~.A??G=.L!Hqtgpuke!Uek!
DD6D7I@A?H¤@A@F=.
@C@=.\txt.\W;.Rpx.P;.P|q{t.P;.
Wpx}v~}.W;.W~.a;.Zttut.ab;.
\rS~}p{s.Z;.fps.P;._~{~}.a;.\~uux.
cT.6A?@A7=._txt}.rp}}pqx.t.
w~.}t~rw~{~vxrp{.str{x}t.u~|.
rwx{sw~~s.~.|xs{xut=.Rtqe!Pcvn!Cecf!Uek!
WUC!@?H6C?7ITAEDF¤TAEEC=.
@CA=.\t}st{~}.YW;.\t{{~.]Z.6@HGC7=.Tuutr.
~u.|pxyp}p.~}.}t~t}s~rx}t.
w~|~}t.x}.w|p}.|p{t.p}s.ut|p{t=.
PKFC!Tgu!Oqpqit!CCIHF¤@@C=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
@CB=.\tx}x.[;.Zxp}xs~.P;.\p{tupzx.b;.
_ppwp}p~~{~._.6A??E7=.
]t~rw~{~vxrp{.stuxrx.x}.{~}v<t|.
utt}.rp}}pqx.t=.Pgwtqnqi{!
EEIFBF¤FBH=.
@CC=.\x}~x.b;.Sp
~{x.\;.Qpvpv{x.P\;.
P|p~.[;.etrrwx.b;._trrx.RP.6A?@?7=.
P}.~
t
xt.~u.t|pxr.t
xt.~}.
rp}}pqx.p}s.rw~xI.sxrx}v.
ppt}{.r~}u{xrx}v.t{=.Ftwi!
Cneqjqn!Tgx!AH6B7IB?C¤B@F=.
@CD=.\x{t|p}.\P;.[tx.aP;.\pr{t.\;.
bwt~~s.YQ;.\{{t.YT.6A??@7=.
cxvvtx}v.|~rpsxp{.x}uprx~}.q.
|pxyp}p=.Ektewncvkqp!@?BIAG?D¤AG?H=.
@CE=.]prt.T_;.\tt.P[;.a~wqtv.Y\;.
\p{t~}.U.6@HFD7=.Pssxrts.p}s.
}~}pssxrts.sv.t=.P.r~|px~}.~u.
sv.pvt.pt}=.Ctej!Igp!Ru{ejkcvt{!
BA6@7IFF¤G?=.
@CF=.^
xts~.P;.V{~p.Y;.Wtzt}wp|.\.
6@HHB7=.Rw•~}xr.rp}}pqx}~xs.
ps|x}xpx~}.p{t.rp}}pqx}~xs.
trt~.qx}sx}v.x}.p.qpx}I.p.
p}xpx
t.p~psx~vpwxr.s=.
Dtckp!Tgu!E@EIAHB¤B?A=.
@CG=._prwt._;.Qpzpx.b;.Z}~.V.6A??E7=.cwt.
t}s~rp}}pqx}~xs.t|.p.p}.t|tvx}v.
pvt.~u.wp|pr~wtp=.Rjctoceqn!
Tgx!DG6B7IBGH¤CEA=.
@CH=._p}px~xst.X\.6A?@D7=.fwp.x.wt.
p~rxpx~}.qttt}.rp}}pqx.
r~}|x~}.p}s.rpsx~
pr{p.
r~|{xrpx~}=.Uwduv!Cdwug!HI@¤B=.
@D?=._t{p~<ctp}.Y\;.bpt<_x}x{{p._;.
Rwpsx.];.Rt~<_pr~~.Q.6A?@A7=.Vt}t<.
t}
x~}|t}.x}tprx~}.}st{x}v.wt.
tuutr.~u.rp}}pqx.x}.ux.tx~st.
rw~x=.Ewtt!Rjcto!Fgu!@G6BA7ID?AC¤.
D?BD=.
@D@=._x~|t{{x.S.6A??D7=.cwt.t}s~rp}}pqx}~xs.
t|I.p.sv.sxr~
t.ttrx
t=.
Ewtt!Qrkp!Kpxguvki!E6F7IEFA¤EFH=.
@DA=._{trwt.\Y;.exx}vw~uu.T;.Zp{wp}.a;.
axrw|p}.Y;.bpuu~s.\;.bxs}t.b;.[x}.U;.
Ztt.b.6A?@A7=.P~rxpx~}.qttt}.
|pxyp}p.t~t.p}s.{|~}p.
u}rx~}.~
t.A?.tp=.Lqwtpcn!qh!vjg!
Cogtkecp!Ogfkecn!Cuuqekcvkqp/!
B?F6A7I@FB¤@G@=.
@DB=._~{{p~.U;.cpv{xp{pt{p<brpupx.^;.P{{pp.
\;.\}~.T;.Sx.\p~.e;.St._t~rt{{x.
[;.Pt}sx}~.V.6A?@@7=.Qx~prx
t.
t}{~v~.rp}}pqx}~xs.u~|.uxqt.
wt|.6Rp}}pqx.px
p7=.L!Pcv!Rtqf/!
FC6H7IA?@H¤A?AA.
@DC=._~t.WV.Y;.Vqt.PY;.Ws~}.YX;.
Wtx.\P;.hvt{}<c~ss.S.6A??A7=.
R~v}xx
t.|tpt.x}.{~}v<t|.
rp}}pqx.t=.L!Enkp!Rjctoceqn!CA6@@.
b{7IC@b¤CFb=.
@DD=.apsp}.\\;.T{b~w{.\P;.b{pst.S;.
Pw|ts.bP;.Zwp}.XP;.a~.bP.6A??H7=.
Qx~{~vxrp{{.prx
t.rp}}pqx}~xs.u~|.
wxvw<~t}r.Rp}}pqx.px
p=.L!Pcv!Rtqf!
FA6D7IH?E¤H@@=.
@DE=.ap|ptzt.YV;.Qtvwp.V;.
p}.[pp.\;.
S||t.^W.6A??C7=.S~t.t{pts.xz.~u.
|~~.
twxr{t.rpwt.put.rp}}pqx.t=.
Ftwi!cpf!Cneqjqn!Fgrgpfgpeg/!
FB6A7I@?H¤@@H=.
@DF=.ap|ptzt.YV;.Zpt.V;.
p}.axt}qttz.
_;.cwt}xt}.T[;.brw}txst.T;.\~t{{t.
\a.6A??E7=.Wxvw<~t}r.|pxyp}p.
x|px.ttrx
t.u}rx~}.p}s.
x}wxqx~.|~~.r~}~{=.
RQ!11111
Hto!1118;
Hov!5812
Uhov!5813
64876!
Pgwtqru{ejqrjctoceqnqi{!B@6@?7IAAHE¤.
AB?B=.
@DG=.ap|ptzt.YV;.Zpt.V;.cwt}xt}.T[;.
c~t}}t.bf=;.\~t{{t.\a.6A??H7=.
]t~r~v}xx
t.tu~|p}rt.sx}v.
prt.cWR.x}~xrpx~}.x}.wtp
.p}s.
~rrpx~}p{.rp}}pqx.t=.L!
Ru{ejqrjctoceqn!AB6B7IAEE¤AFF=.
@DH=.axvv._Z;.epxsp.U;.a~x.bb;.b~zx}.[b;.
V~p.Q;.Vp}.X;.T{{x.aY.6A?@A7=.P.
x{~.s.~u.wt.tuutr.~u.rp}}pqx.~}.
ptxt.w~|~}t.x}.WXe<x}utrts.ps{.
|t}=.Dtckp!Tgu!@CB@ICE¤DA=.
@E?=.a~sxvt.st.U~}trp.U;.V~xx.\P;.
Ut}p}st<ax.YY;._p{~|~.c;.ap|~.YP.
6@HHC7=.S~}tv{px~}.~u.p.qpx}.
rp}}pqx}~xs.qx}sx}v.xt.put.rw~}xr.
st{p.H<tpws~rp}}pqx}~{.tp|t}=.
Rjctoceqn!Dkqejgo!Dgjcx!CF6@7IBB¤C?=.
@E@=.a~w.\S;.P~p.P;.Qpz.bW;.Z{tt.
TR;.bx||~}.\;.cpwzx}.S_.6@HHG7=.
Pxp.x}u{p||px~}.x}.~}v.
|pxyp}p.p}s.~qprr~.|~zt=.
Cogtkecp!Lqwtpcn!qh!Tgurktcvqt{!cpf!Etkv!
Ectg!Ogf!@DFIHAG¤HBF=.
@EA=.a~w.\S;.cpwzx}.S_;.fwxpzt.Z\;.
Rw~x.a;.Qp{sx}.VR.6A??D7=.
ctpws~rp}}pqx}~{.tt.
x||}t.u}rx~}.p}s.t}wp}rt.WXe.
t{xrpx~}.x}.wt.w_Q[¤bRXS.|~t=.
Nkhg!Uek!FF6@C7I@F@@¤@FAA=.
@EB=.a~.T;.\pwt.\[;.Q}t.P;.et{x}.a;.
Qprw._Y;.bp}rwt<ap|~.Y;.Zx{x}.ZP.
6A??@7=.Rw~}xr.rp}}pqx.t.x}.wt.
r~|px~}pt.x}
txvpx~}p{.}t.sv.
~vp|I.P}.tp|x}px~}.~u.qt}tux.p}s.
ps
tt.tuutr.~u.{tvp{.r{x}xrp{.rp}}pqx=.
L!Ecppcdku!Vjgt!AIB¤DF=.
@EC=.bpupp.b;.Pup.U;.Pswp|x.e\;.
\zwp.W.6A??D7=.Rp}}pqx}~xs.trt~.
p.p.}~
t{.pvt.u~.wt.tp|t}.~u.
~pt.rp}rt=.Ecpegt!Tgu!ED6D7I@EBD¤.
@EC@=.
@ED=.brwtt.YU;.Vp}.YS;.S}rp}.PT;.bp~.
RT;.Wpqt.Ya;.Ypr~q.c;.Qrw~{.ZZ.
6A??H7=.bqytrx
t.tuutr.~.rp}}pqx.pt.
p~rxpts.xw.t;.pqt.p}s.
stt}st}rt.put.psyx}v.u~.vt}txr.
p}s.t}
x~}|t}p{.x}u{t}rt=.Ftwi!
Cneqjqn!Fgrgpf!@?D6@¤A7IFE¤GA=.
@EE=.brwxuu|p}.Y;.]pzp|p.Q;.Tp{tx}t.
\Y;.[pQxt.Y.6A??D7=.b|~|.~u.
rwx~.trtst.rp}}pqx.t=.
Ru{ejkcvt{!Tgu!@BC6@7IBF¤CA=.
@EF=.brwx||t{|p}}.QV;.R~}._;.R~~}.
b\;.Zutrw|xs.b;.Zp~.P;.
brw{t<[t.U;.\rV~._S;.[p|qt.
\.6A?@@7=.Rp}}pqx.t.sx~st.p}s.pvt.
p.~}t.~u.rw~x¥p.s.x}.ux.
tx~st.pxt}=.Uejk|qrjt!Tgu!
@AH6@7IDA¤DE=.
@EG=.brwtx}t.P\;.S}}.\T.6A?@A7=.
atxsp{.tuutr.~u.rp}}pqx.t.~}.
}t~r~v}xx
t.tu~|p}rt.put.
~{~}vts.pqx}t}rtI.p.|tp<p}p{x=.
Gzr!Enkp!Ru{ejqrjctoceqn!A?6D7ICA?¤.
CAH=.
@EH=.bt~}.\;.Rspqprz.T;.Pqs{{pw.aP;.
Ux}}t{{.Y;.\xrw{t.[Y;.a~vp.\;.
[xrw|p}.PW;.bt{{p.].6A?@C7=.Rp}}pqx.
t.q.x}sx
xsp{.xw.|{x{t.
r{t~xI.tuutr.~}.trxuxr.x||}t.
pp|tt=.Kphncooqrjctoceqnqi{!
AA6D7IAHD¤B?B=.
@F?=.bxs}t.b.6A??A7=.Rpsx~
pr{p.
r~}tt}rt.~u.|pxyp}p.t=.L!Enkp!
Rjctoceqn!CA6@@b{7IECb¤F?b=.
G<^HT^HO^23CWR3/UIO
23CWR3
ouvqemuvknn!qp!FUM4I;V193RTQF!ykvj!RTQRQUCNU3
64877!
Hgfgtcn! Tgikuvgt > e~{=. G@;. ]~=. @DE > Uxsp;. Pv. @A;. A?@E > _~~ts. a{t.
@F@=.b~{x}p.\;._p}{x{x~.[e;.Yx}~
p.i;.
hpp.b;.V~{sqtv.ba.6A??E7=.dx}v.sv<.
sxrx|x}px~}.trw}xt.~.s.wt.
pqt<t{pts.tuutr.~u.rw~prx
t.
sv.x}.p=.Pcv!Rtqvqe!@6B7I@@HC¤@A?E=.
@FA=.b~{~xy.];.btwt}.ab;.a~uu|p}.aP;.
Qpq~•.c;.Zpsst}.a;.\x{{t•.\;.
Rwxxp}t}.Z;.\ratt.Q;et}stx.Y.
6A??A7=.\pxyp}p.ctp|t}._~ytr.
attprw.V~=.R~v}xx
t.u}rx~}x}v.~u.
{~}v<t|.wtp
.rp}}pqx.t.ttzx}v.
tp|t}=.Lqwtpcn!qh!vjg!Cogtkecp!
Ogfkecn!Cuuqekcvkqp!AGF6H7I@@AB¤@@B@=.
@FB=.bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~}.6A?@B7=.Ftwi!
Cdwug!Yctpkpi!Pgvyqtm-!3122<!Pcvkqpcn!
Guvkocvgu!qh!Ftwi.Tgncvgf!Gogtigpe{!
Fgrctvogpv!Xkukvu/!WWb._q{xrpx~}.]~=.
6b\P7.@B¤CFE?;.SPf].btxt.S¤BH=.
a~rz
x{{t;.\S=.
@FC=.bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~};.Rt}t.u~.
Qtwp
x~p{.Wtp{w.bpxxr.p}s.`p{x.
6A?@Dp7=.at{.u~|.wt.A?@C.]px~}p{.
b
t.~}.Sv.dt.p}s.Wtp{wI.Stpx{ts.
cpq{t=.a~rz
x{{t;.\S=.
@FD=.bqp}rt.Pqt.p}s.\t}p{.Wtp{w.
bt
xrt.Ps|x}xpx~};.Rt}t.u~.
Qtwp
x~p{.Wtp{w.bpxxr.p}s.`p{x.
6A?@Dq7=.Vtgcvogpv!Grkuqfg!Fcvc!Ugv!
)VGFU*<!31143124/!Pcvkqpcn!
Cfokuukqpu!vq!Uwduvcpeg!Cdwug!
Vtgcvogpv!Ugtxkegu/!QWbXb.btxt.b¤FD;.
WWb._q{xrpx~}.]~=.6b\P7.@D¤CHBC=.
a~rz
x{{t;.\S=.
@FE=.cpx.aY;.\prZx}}~}.P;.Rwxt}t}.W.
6A?@@7=.Rp}}pqx.t.p}s.r~v}xx
t.
u}rx~}x}vI.G<tp.pytr~.x}.p.~}v.
ps{.r~w~=.Cffkevkqp!@?E6@A7IA@HD¤.
AA?B=.
@FF=.cp}ptr.a;.R~}p}x}tr.Rb.6A?@?7=.
Rp}}pqx}~xs.p}s.wt.x||}t.t|I.
p}.~
t
xt=.Koowpqdkqnqi{!
A@D6G7IDGG¤DHF=.
@FG=.cp}sp.V;.\}p._;.V~{sqtv.ba.6A???7=.
bt{u<ps|x}xpx~}.qtwp
x~.x.
|px}px}ts.q.wt.rw~prx
t.
x}vtsxt}.~u.|pxyp}p.x}.xt{.
|~}zt=.Pcv!Pgwtquek!B6@@7I@?FB¤@?FC=.
@FH=.cp}sp.V;.V~{sqt•v.ba.6A??B7=.
Rp}}pqx}~xsI.tps;.stt}st}rt;.p}s.
}st{x}v.}t~rwt|xrp{.
|trwp}x|¥p.t
xt.~u.trt}.
tr{x}xrp{.spp=.Ru{ejqrjctoceqnqi{!
)Dgtn*!@EH6A7I@@D¤@BC=.
@G?=.cpwzx}.S_.6A??D7=.b|~zts.|pxyp}p.
p.p.rpt.~u.{}v.x}y=.Oqpcnfk!Ctej!
Ejguv!Fku!EB6A7IHB<{??=.
@G@=.cpwzx}.S_;.iwp}v.iU;.Vtt}{p}s.b;.
R~t}.f;.\prz.c\;.p}s.\~vt}t}.W.
6A??E7=.\pxyp}p.t.p}s.{}v.rp}rtI.
t{.~u.p.rpt<r~}~{.s=.Cogtkecp!
Vjqtceke!Uqekgv{!Kpvgtpcvkqpcn!
Eqphgtgpeg/!Pqpr.PFFF=.
@GA=.cwt}xt}.T[;.Zpt.VU;.c~t}}t.
bf=;.\~t{{t.\a;.bp|qtw.P;.Q{p}rwps.
\\;.ap|ptzt.YV.6A?@A7=.
]t~wx~{~vxrp{.u}rx~}x}v.~u.
~rrpx~}p{.p}s.wtp
.rp}}pqx.t.
sx}v.cWR.x}~xrpx~}=.
Ru{ejqrjctoceqnqi{!)Dgtn*!AA?6A7IBC@¤.
BD?=.
@GB=.cpqt.Q;.bxvs~}.PY;.btt}t.P\;.
b~|.bb;.\rV{}}.ZP.6A?@@7=.
\pxyp}p.t.p}s.txr{p.vt|.rt{{.
|~=.Ecpegt!@@F6C7IGCG¤GDB=.
XgtFcvg!Ugr>22@3125!
31<62!Cwi!22-!3127
Lmv!349112
@GC=.cxrwt{{.f;.Q~}.b;.\przxt.Z.6@HHF7=.
Rp}}pqx}~xs.x}wxqx.]<.p}s._>`<t.
rp{rx|.rwp}}t{.x}.r{ts.p.
wx~rp|p{.}t~}=.L!Pgwtqrj{ukqn!
FG6@7ICB¤D?=.
@GD=.d=b=.U~~s.p}s.Sv.Ps|x}xpx~}.
6USP7;.Rt}t.u~.Sv.p}s.T
p{px~}.
attprw;.R~}~{{ts.bqp}rt.bpuu.
6A?@D7=.cwt.\tsxrp{.P{xrpx~}.~u.
\pxyp}pI.P.at
xt.~u._q{xwts.
R{x}xrp{.bsxt=.\prw.@H;.A?@D=.
@GE=.
p}.st.\tt.UY;.et{w~.T;.\txyt.RY;.
\prwxt{t}.\f;.st.Wpp}.[.6A?@A7=.
Rp}}pqx.t.x}.pxt}.p.r{x}xrp{.wxvw.
xz.~u.rw~xI.x|pr.~}.~s~|p{.
|~|.p}s.p}xx~}.~.rw~x=.
Ewtt!Rjcto!Fgu!@G6BA7ID?BE¤D?CC=.
@GF=.
p}.Vpt{.fP;.fxv|p}.Yc;.
\~}w~t.Z;.Zpw}.ab;.
p}.^.Y;.Q~z.
\_;.e~{{tqvw.fP.6A?@A7=.Rp}}pqx.
t.p}s.qr{x}xrp{.~xx
t.rw~xr.
ttxt}rt.x}.tp{.ps~{trt}rtI.
ux}sx}v.u~|.p.Srw.
t=.Cffkevkqp!
@?F6A7IBG@¤BGF=.
@GG=.ep}.V}s.Z;.atqt{{~}.RY.6A?@?7=.P.
[xut<r~t._ttrx
t.~}.wt.Vptp.
W~wtx= .L!Jgcnvj!Uqe!Dgjcx!
D@6B7IACC¤ADH=.
@GH=.
p}.^.Y;.Qpz.\;.Wp}t}.\;.Qxy{.ae;.
st.Vppu.a;.ets~.W=.6A??A7=.Rp}}pqx.
t.p}s.rw~xI.p.{~}vxsx}p{.
~{px~}<qpts.s=.Co!L!Grk!
@DE6C7IB@H¤BAF=.
@H?=.ep}}.aT;.Vp|pvt.cU;.fp•}t•.YP;.
\pwp{{.T\;.cp{~.][;.\px}.Qa;.
fx{t.Y[.6A??G7=.Sx
tvt}.tuutr.~u.
rp}}pqxsx~{.~}.wt.sxrx|x}px
t.
x|{.p}s.{prt.r~}sxx~}x}v.tuutr.
~u.St{p6H7<tpws~rp}}pqx}~{=.Ftwi!
Cneqjqn!Fgrgpf!HC6@¤B7I@H@¤@HG=.
@H@=.e~{z~.]S;.Qp{t.aS;.R~|~}.f\;.
ftx.ba.6A?@C7=.Ps
tt.wtp{w.tuutr.
~u.|pxyp}p.t=.P!Gpin!L!Ogf!
BF?6AB7IAA@H¤AAAF=.
@HA=.
~}.bs~.Z;.[xtq.a;._uxt.W;.W~u{t.
\;.fxrwt}.Wd.6A??A7=.fwp.tsxr.
x}rxst}.t.~u.rp}}pqx.p}s.~vtx~}.
~.pqt.p}s.stt}st}rtN.P.C<tp.
~trx
t.tp|x}px~}.~u.xz.upr~.
x}.p.r~||}x.p|{t.~u.ps~{trt}.
p}s.~}v.ps{=.Ftwi!Cneqjqn!Fgrgpf!
EG6@7ICH¤EC=.
@HB=.fprwt{.ba;.T{b~w{.\P;.a~.bP;.
P|qt.Y;.st.fx.W.6A??A7=.R~|px~}.~u.
wt.qytrx
t.tuutr.~u.St{p6H7<.
tpws~rp}}pqx}~{.p}s.|pxyp}p.x}.
w|p}=.Ru{ejqrjctoceqnqi{!)Dgtn*!
@E@6C7IBB@¤BBH=.
@HC=.fpv}t.YP;.epvp.Z;.Z}~.V.6@HHG7=.
Rpsx~
pr{p.prx~}.~u.rp}}pqx}~xs.
p}s.wtx.vt}tpx~}.sx}v.w~rz=.L!Oqn!
Ogf!FE6@A7IGAC¤GBE=.
@HD=.fp}v.Vb;.a~~t
t{.V;.Wtps.Z.6A?@B7=.
_tsxpxr.|pxyp}p.t~t.x}.p.
|tsxrp{.|pxyp}p.pt=.LCOC!Rgfkcvt!
@EF6F7IEB?¤EBB=.
@HE=.fp}v.Vb;.a~~t
t{.V;.[t.[px.\R;.
\px}t.T\;.Qrwt<Qpt{~}.Q;.
Q~}tx}.PR;.Wtps.Z.6A?@C7=.
P~rxpx~}.~u.}x}t}sts.tsxpxr.
t~t.xw.strx|x}p{xpx~}.~u.
|pxyp}p.x}.wt.d}xts.bpt=.Cpp!
Gogti!Ogf!EB6E7IEGC¤EGH=.
@HF=.fp}v.c;.R~{{t.Y_;.bwpx~.b;.fpt.\P.
6A??G7=.Ps
tt.tuutr.~u.|tsxrp{.
rp}}pqx}~xsI.p.t|pxr.t
xt=.EOCL!
@FG6@B7I@EEH¤@EFG=.
RQ!11111
Hto!11191
Hov!5812
Uhov!;;;1
@HG=.ft~}.Sa;.fpwq}._.6@HH?7=.Rt}.
pt}.~u.sv.pqt.wp.x}
~{
t.
|~zx}v=.PKFC!Tgu!Oqpqit!HHID¤@@=.
@HH=.fwxtwx{{.Y\;.ax
pp.U_;.\~t}~.\P.
6A?@C7=.\pxyp}p<x}v.sx
t;.p{r~w~{<.
x}v.sx
t;.p}s.wtx.pt}vtI.
t
p{t}rt.p}s.xz.upr~.p|~}v.
}stpvt.r~{{tvt.st}=.LCOC!Rgfkcvt!
@EG6F7IE@G¤EAC=.
A??=.fx{t.Y[;.Qpt.a[;.Qx.S[;.Qp{t.
a[;.\px}.Qa.6@HHB7=.Sxrx|x}px
t.
x|{.tuutr.~u.st{p.H<.
tpws~rp}}pqx}~{.p}s.st{p.H¤@@<.
tpws~rp}}pqx}~{.x}.p.p}s.wt.
|~}zt=.Pgwtqrjctoceqnqi{!
BA6C7IBDH¤BED=.
A?@=.fx{t.Y[;.Wuu|p}.Yf;.Qp{t.a[;.
\px}.Qa.6@HHD7=._wp|pr~{~vxrp{.
trxuxrx.~u.wt.sxrx|x}px
t.x|{.
tuutr.~u.st{p.H<tpws~rp}}pqx}~{.x}.
wt.|~}zt=.Ftwi!Cneqjqn!Fgrgpf!
C?6@7IG@¤GE=.
A?A=.fx{zx}~}.bc;.apswpzxw}p}.a;.
S b~p.SR.6A?@C7=.X|pr.~u.rp}}pqx.
t.~}.wt.st
t{~|t}.~u.rw~xr.
sx~st=.Ewtt!Cffkev!Tgr!@6A7I@@D¤@AG=.
A?B=.fx{~}.UP;.bx|~}.Y_;._pvp }.YP.
6A?@C7=.Upp{.rpwt.u~|.sx
t.tx}v.
~xx
t.u~.sv.x}.wt.d=b=;.@HHB¤.
A?@?=.Rwdnke!Jgcnvj!Tgr!@AH6C7IBCA¤BD?=.
A?C=.f.g;.Ut}rw.TS.6A???7=.Tuutr.~u.
rw~}xr.st{pH<tpws~rp}}pqx}~{.~}.
p.|xsqpx}.s~p|x}t.}t~}I.p}.
t{tr~wx~{~vxrp{.pt|t}=.
Pgwtqrjctoceqnqi{!BH6B7IBH@¤BHG=.
A?D=.itxvt.Yb;.Wpqtxrz.QR;.R~{t.a_;.
Twx}vt.\P;.R~{t.cY;.Wtx.YZ;.
W~ut.RY;.bp{{x}v.\R;.h~}v.bT=;.
awtt.bW.6A?@?7=.bqytrx
t.tuutr.~.
|pxyp}p.p~rxpts.xw.|pxyp}p.t.
x}.r~||}x.p}s.r{x}xrp{.qytr=.Ftwi!
Cneqjqn!Fgrgpf!@?H6@¤B7I@E@¤EE=.
A?E=.iwp}v.iU;.\~vt}t}.W;.bx.\a;.
cpwzx}.S_;.h.V_;.\pwp{{.Ya;.W.cR;.
brwp};.b_.6@HHH7=.\pxyp}p.t.p}s.
x}rtpts.xz.~u.p|~.rt{{.
rprx}~|p.~u.wt.wtps.p}s.}trz=.Ecpegt!
Grkfgokqn!Dkqoctmgtu!Rtgx!G6@A7I@?F@¤.
@?FG=.
A?F=.iwp}v.[a;.\~vt}t}.W;.Vtt}{p}s.b;.
Rwp}v.bR;.[pp._;.ctpt.\S;.f~{{._Y;.
^{~.X;.R~.Q.~}.qtwp{u.~u.wt.Rp}}pqx.
p}s.atxp~.Sxtpt.attprw.V~.
~u.]t.itp{p}s;.Qwp}t.h;.[x.V;.W}v.
aY.6A?@D7=.Rp}}pqx.|~zx}v.p}s.{}v.
rp}rt.xzI._~~{ts.p}p{x.x}.wt.
X}t}px~}p{.[}v.Rp}rt.R~}~x|=.
Kpv!L!Ecpegt!@BE6C7IGHC¤H?B=.
A?G=.ipsx.Pf;.bwxpzpp.X;.Ux}zt{upq.T;.
Zp}x~{.XV.6@HGA7=.Prx~}.~u.rp}}pqxsx~{.
~}.wt.p}xt.p}s.~wt.tuutr.
~srts.q.st{p.H¤cWR.x}.}~|p{.
qytr=.Ru{ejqrjctoeqnqi{!)Dgtn*!
FE6B7IACD¤AD?=.
jUa.S~r=.A?@E¤@FHDC.Ux{ts.G¤@@¤@EJ.GICD.p|l.
DKNNKPI! EQFG! 55211;R!
G<^HT^HO^23CWR3/UIO
23CWR3
Enclosure B: “The DEA’s Denial of Existing Medical Cannabis Research” (August 2016) The Dangers and Consequences of Misinformation on Marijuana February 22, 2017
The DEA’s Denial of Existing Medical Cannabis Research
A Peer-Reviewed Comparative Analysis of DEA’s
“Denial of Petition to Initiate Proceedings to Reschedule Marijuana”
Prepared by Americans for Safe Access (2016)
Reviewed by Jahan Marcu, Ph.D., Ethan Russo, MD, Jason Schechter, Ph.D., and Steph Sherer
ASA | DEA’s Denial of Existing Medical Cannabis Research
The research and analysis in this report was conducted by Americans for Safe Access Foundation, a 501(c)(3) non-profit organization.
Publication date July 2016 With over 100,000 active members in all 50 states, Americans for Safe Access (ASA) is the largest
national member-based organization of patients, medical professionals, scientists and concerned citizens promoting safe and legal
access to cannabis for therapeutic use and research. ASA works to overcome political and legal barriers by creating policies that
improve access to medical cannabis for patients and researchers through legislation, education, litigation, grassroots actions, advocacy
and services for patients and their caregivers, the medical cannabis industry, and governments.
Copyright © 2016 Americans for Safe Access Foundation™
All rights reserved. No part of this publication may be reproduced in any form by any electronic or mechanical means without
permission in writing from Americans for Safe Access.
Learn more about ASA at AmericansForSafeAccess.org.
1624 U Street NW, Suite 200,
Washington, D.C. 20009
Phone: 1-888-929-4367 fax: 202.857.4273
2 | A MER I CAN S F OR SAF E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
Table of Contents
I.
Foreword .................................................................................................................................................................................................................................... 4
II.
Introduction............................................................................................................................................................................................................................... 5
III.
Common Ground ..................................................................................................................................................................................................................... 6
Factor 1b: There is no significant diversion of the substance from legitimate drug channels................................................................... 6
Factor 1d: Cannabis is related to other approved drugs with acceptable safety profiles. ........................................................................... 7
Factor 6: That the “gateway” hypothesis is not supported by scientific evidence. ......................................................................................... 7
Factor 8: Cannabis is not an immediate precursor to a controlled substance................................................................................................... 8
IV.
Comparative Analysis of Available Data vs HHS Report ........................................................................................................................................ 8
V.
Evaluating the DEA’s Rationale for 1) Marijuana has a high potential for abuse. ................................................................................... 11
DEA’s Evidence............................................................................................................................................................................................................................ 11
Available Scientific Data .......................................................................................................................................................................................................... 12
VI.
Evaluating the DEA’s Statement 2) Marijuana has no currently accepted medical use in treatment in the Unites States. .. 12
Element (1) The drug's chemistry is known and reproducible. ........................................................................................................................... 13
DEA/FDA Evidence for Element 1 ................................................................................................................................................................................ 13
Available Scientific Data for Element 1 ...................................................................................................................................................................... 14
Element (2) There are adequate safety studies............................................................................................................................................................ 17
DEA/FDA Evidence for Element 2 ................................................................................................................................................................................ 18
Available Scientific Data for Element 2 ...................................................................................................................................................................... 18
Element (3) There are adequate and well-controlled studies proving efficacy. ............................................................................................ 19
DEA/FDA Evidence for Element 3 ................................................................................................................................................................................ 19
Available Scientific Data for Element 3 ...................................................................................................................................................................... 19
Element (4) The drug is accepted by qualified experts. ........................................................................................................................................... 20
DEA/FDA Evidence for Element 4 ................................................................................................................................................................................ 20
Available Scientific Data for Element 4 ...................................................................................................................................................................... 20
Element (5) The scientific evidence is widely available. .......................................................................................................................................... 23
DEA/FDA Evidence for Element 5 ................................................................................................................................................................................ 23
Available Scientific Data for Element 5 ...................................................................................................................................................................... 23
VII.
Evaluating the DEA’s Statement 3) Marijuana lacks accepted safety for use under medical supervision. ............................. 24
DEA’s Evidence Regarding Safety ....................................................................................................................................................................................... 24
Available Scientific Data Regarding Safety ..................................................................................................................................................................... 24
VIII.
Conclusion........................................................................................................................................................................................................................... 25
Recommendations ..................................................................................................................................................................................................................... 25
AMER IC AN S F OR S AF E ACCE SS | 3
ASA | DEA’s Denial of Existing Medical Cannabis Research
I.
Foreword
Today over 300 million Americans live in states with medical cannabis laws, and over 2 million
individuals are legally using medical cannabis under these state programs. However, all of these patients
and programs are in violation of federal laws. According to the Department of Justice (DOJ), this is due to
the fact that Congress determined that cannabis belongs in Schedule I of the Controlled Substances Act
(CSA).
However, the scheduling of cannabis has been a political – rather than scientific – establishment from the
very beginning. In 1970, cannabis was placed in Schedule I under the CSA as a placeholder, pending
evaluation by a government-appointed commission known as the National Commission on Marihuana
and Drug Abuse – since known as the Shafer Commission after the Commission's chairman, Raymond P.
Shafer. Even though the Shafer Commission recommended decriminalization of cannabis and medical
availability, these policies were rejected by President Nixon before the report could be published. Despite
numerous advances in science and research in the medical value of cannabis, due to political forces, as
well as Drug Enforcement Administration (DEA) and U.S. Food and Drug Administration (FDA) policies
that were designed for prescription drugs, cannabis has been stuck in Schedule I ever since.
Under these circumstances, the current rescheduling process will never allow cannabis to be
rescheduled. This is made clear in the DEA’s most recent “Denial of Petition to Initiate Proceedings to
Reschedule Marijuana,” which focuses on the fact that cannabis does not fit with current federal
regulations for a FDA approved drug, i.e. the medical value assigned to cannabis does not meet their
definition of “medicine,” not that cannabis has no medical value.
This is the 4th time in just over 4 decades that the DEA has denied a petition to reschedule cannabis. Not
only has the DEA taken several years to respond to each petition, but special rules for cannabis are
created and applied whenever there is data that does not support their policy. In the 1990s, the DEA
established a “5-element test” to determine if there was accepted medical use for a drug. However, the
consequences of not satisfying this test to fulfil the DEA’s definition of medicine have only been applied to
cannabis. Applying prescription drug standards – such as those required for FDA approval – to a
botanical drug is a case in point of special rules being applied where they wouldn’t be otherwise. Rather
than using the FDA guidelines for botanical drugs, cannabis is criticized as though it were a purified
pharmaceutical agent, and not a botanical medicine.
The rescheduling process has been designed for prescription drugs to move between the schedules, and
not for a Schedule I substance to enter into less restrictive schedules. This unworkable process for
botanical medicines, including but not limited to cannabis, has led 42 states plus the District of Columbia
to create their own definitions of medicine and distribution.
The DEA’s recent decision shows that the war against medical cannabis will unfortunately continue
unabated, and unaffected by either reason or scientific evidence. Until these policies can be changed, the
only viable solutions will require action by Congress.
4 | A MER I CAN S F OR SAF E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
II.
Introduction
In April 2011, the Department of Justice (DOJ) sent letters to governors of 9 medical cannabis states
“clarifying” that medical cannabis programs – and specifically regulated distribution programs – were in
violation of federal law, due to the Schedule I status of marijuana. In response, in November of the same
year, Governors Lincoln D. Chafee (RI) and Christine O. Gregoire (WA), petitioned the DEA to initiate
rulemaking proceedings under the rescheduling provisions of the CSA – to remove marijuana and
“related items” from Schedule I of the CSA and to reschedule as “medical cannabis” in Schedule II. After
nearly five years of review, on August 10, 2016 the DEA responded to the petition with a document
entitled Denial of Petition to Initiate Proceedings to Reschedule Marijuana (herein referred to as the "DEA
report").
The DEA concluded that “marijuana” (cannabis) should not be removed from the Schedule I status due to
the below 3 factors:
1) Marijuana has a high potential for abuse;
2) Marijuana has no currently accepted medical use in treatment in the United States; and
3) Marijuana lacks accepted safety for use under medical supervision.
DEA chief Chuck Rosenberg stated that this decision was based heavily on the FDA’s determination if
marijuana is “a safe and effective medicine.” This determination was based upon input from the
Department of Health and Human Services (HHS), which was conducted in consultation with the National
Institute on Drug Abuse (NIDA).
The DEA report cited the following in making their determination:
1. U.S. Department of Health and Human Services (HHS) – Basis for the Recommendation for
Maintaining Marijuana in Schedule I of the Controlled Substances Act (page 7).
2. While not listed in their cover letter as a submitted document, a review article added at the end of
the bibliography of the HHS report, entitled The Medical Application of Marijuana: A Review of
Published Clinical Studies prepared by the U.S. Food and Drug Administration (page 66).
3. U.S. Department of Justice - Drug Enforcement Administration (DEA), Schedule of Controlled
Substances: Maintaining Marijuana in Schedule I of the Controlled Substances Act, Background,
Data, and Analysis: Eight Factors Determinative of Control and Findings Pursuant to 21 U.S.C.
812(b) (Page 115).
While we do not agree with the DEA’s final determination that marijuana is not a safe and effective
medicine, we do appreciate the time and resources the DEA put into making this decision. We are pleased
to see a few areas of agreement between their report and the available scientific data on cannabis.
Generally, our analysis found that the DEA admits that cannabis satisfies several criteria regarding the 8Factor analysis.
However, the DEA report included both inaccurate and unclear background materials pertaining to the
scheduling process of cannabis, conjoined to misinterpretations of the CSA in general. In one clear
AMER IC AN S F OR S AF E ACCE SS | 5
ASA | DEA’s Denial of Existing Medical Cannabis Research
example of this, the report states there are no known standardized cannabis products. The DEA chose to
use a misinterpretation of the CSA to exclude any clinical research conducted with standardized cannabis
extracts from the HHS report. The report defines cannabis/marijuana in the CSA as including derivatives
and extracts of cannabis/marijuana such as purified THC, CBD, and nabiximols. However, in the DEA’s
political view, these resinous hash oils do not count as standardized cannabis products, nor do the
cannabis cigarettes that NIDA themselves produce according to DEA (and FDA) guidelines and mandate.
Clinical studies with resinous hash oil extractions were systematically excluded in the DEA’s denial of
rescheduling report.
Actual standardized “cannabis medicines” include purified THC, purified CBD, THC/CBD mixtures, and
nabiximols (commonly known as Sativex®). Purified CBD and Sativex® are FDA approved under IND for
pediatric epilepsy, and in Phase III clinical trials in the U.S., respectively. Marinol® is an FDA approved
cannabis product known as dronabinol. There exists no evidence of significant abuse, nor black market or
diversion issues, with currently available standardized medicinal cannabis products – including
dronabinol, nabiximols, or NIDA’s cannabis products. Such persistent misinterpretation of existing law –
coupled to apparent lack of knowledge of prevailing scientific investigations concerning both general
safety and medicinal usefulness – suggests that an uninformed and unbalanced opinion of cannabinoidbased medicine is being advanced.
In anticipation of the DEA’s pending decision on the scheduling of medical cannabis, Americans for Safe
Access (ASA) coordinated world experts on cannabis to draft an independent 8-Factor Analysis based on
all available data that concluded that cannabis does not meet the requirements for a Schedule I substance
under the CSA. The following memo is a comparative analysis of the research and findings used by the
DEA to make their determination that cannabis remain a Schedule I drug. The references in this memo
refer to DEA materials and ASA’s 8-Factor analysis.
III.
Common Ground
The DEA report claims that cannabis satisfies some sections of the 8-factor analysis. This means there are
sections where we all agree that cannabis meets the criteria for rescheduling. In short, we agree with the
DEA that cannabis satisfies Factors 1b, 1d, 2, 3, 6, and 8 (of the 8-factor analysis). For example, the DEA
cites research demonstrating that there is no evidence for long term harms associated from the chronic
use of cannabis to satisfy Factors 2 and 3.
Below are the Factors and the statements from the DEA to which we agree regarding cannabis as a
medicine and its rescheduling.
Factor 1b: There is no significant diversion of the substance from legitimate drug channels.
Factor 1b definition: “There is significant diversion of the substance from legitimate drug channels.”
On page 11, the DEA states, “There is a lack of evidence of significant diversion of marijuana from
legitimate drug channels.”
6 | A MER I CAN S F OR SAF E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
We agree with the FDA and DEA that legal cannabis products have not suffered from significant diversion
and additionally that cannabis is not a precursor for another schedule drug. Pure THC has been FDA
approved since the 1980s and no significant black market for Marinol is known to exist.
Factor 1d: Cannabis is related to other approved drugs with acceptable safety profiles.
Factor 1d definition: “The substance is not so related in its action to a substance already listed as having a
potential for abuse to make it likely that it will have the same potential for abuse as such substance, thus it is
not reasonable to assume that there may be significant diversions from legitimate channels, significant use
contrary to or without medical advice, or that it has a substantial capability of creating hazards to the
health of the user or to the safety of the community.”
On page 12 the DEA states, “FDA has approved two drug products containing cannabinoid compounds that
are structurally related to the active components in marijuana. These two marketed products are controlled
under the CSA.” Furthermore, the DEA goes on, “FDA approved Marinol in 1985 for the treatment of nausea
and vomiting associated with cancer chemotherapy in patients who failed to respond adequately to
conventional anti-emetic treatments. In 1992, FDA approved Marional [sic] for anorexia associated with
weight loss in patients with acquired immunodeficiency syndrome (AIDS). Secondly, in 1985, FDA approved
Cesamet, a drug product containing the Schedule II substance nabilone, for the treatment of nausea and
vomiting associated with cancer chemotherapy.”
We agree with the DEA and FDA that cannabis is a substance related in action to Marinol and Cesamet.
THC (marinol) and Cesamet are two FDA approved drugs with acceptable safety profiles (i.e., low abuse
potential) and no evidence of any significant diversion. Factors 2 and 3: Scientific Evidence for the
Pharmacological Effects and the State of Current Scientific Knowledge Regarding the Drug or Other
Substance.
On page 12, the DEA report states, “Abundant scientific data are available on the neurochemistry,
toxicology, and pharmacology of marijuana.”
On page 20, the DEA report states, “cannabis-associated cognitive deficits are reversible and related to
recent cannabis exposure, rather than irreversible and related to lifetime use.”
On page 22, the DEA report states, “At present, the available data do not suggest a causative link between
marijuana use and the development of psychosis.”
We agree with the DEA that the effects of cannabis are non-toxic and have no long-term consequences on
the human brain. Available data show that the chemistry of cannabis is well understood and does not
cause significant harm to the adult brain.
Factor 6: That the “gateway” hypothesis is not supported by scientific evidence.
Factor 6 definition: “What, if any, risk there is to public health.”
On page 43, the DEA report states, “Overall, research does not support a direct causal relationship between
regular marijuana use and other illicit drug use.”
AMER IC AN S F OR S AF E ACCE SS | 7
ASA | DEA’s Denial of Existing Medical Cannabis Research
On page 44, the DEA report states, “the gateway hypothesis only addresses the order of drug use initiation,
the gateway hypothesis does not specify any mechanistic connections between drug "stages" following
exposure to marijuana and does not extend to the risks for addiction.”
On page 162, the DEA report states, “The HHS reviewed the clinical studies evaluating the gateway
hypothesis in marijuana and found them to be limited.” The DEA goes on to say, “The HHS cited several
studies where marijuana use did not lead to other illicit drug use.”
On page 162, the DEA report states, “Based on these studies among others, the HHS concluded that
although many individuals with a drug abuse disorder may have used marijuana as one of their first illicit
drugs, this does not mean that individuals initiated with marijuana inherently will go on to become regular
users of other illicit drugs.”
Over 40 years ago the “gateway” hypothesis of cannabis was proposed. The report concludes predictably,
that the gateway theory of cannabis is not supported by the evidence. We agree that the hypothesis
attempted but failed to predict that cannabis use leads to the addiction of other drugs. Furthermore, no
clinically significant adverse public health effects related to rescheduling cannabis were provided to by
the DEA.
Factor 8: Cannabis is not an immediate precursor to a controlled substance.
Factor 8 definition: “Whether the substance is immediate precursor of a substance already controlled under
the article.”
On page 46, the DEA report states, “Marijuana is not an immediate precursor of another controlled
substance.”
We agree that cannabis is not an immediate precursor of another controlled substance.
While not sufficient for the DEA to reschedule, these statements show an evolution in the DEA’s opinions
on cannabis. All federal conversations about cannabis should begin with the above information.
IV.
Comparative Analysis of Available Data vs HHS Report
The 2016 HHS evaluation and the additional data gathered by the DEA constitute a document, entitled
“Basis for the Recommendation for Maintaining Marijuana in Schedule I of the Controlled Substances
Act.” This document supporting the basis of the DEA recommendation was preliminarily scrutinized by
ASA through use of a comparative reference analysis, in which we categorized and characterized each
reference in the DEA’s basis article according to multiple criteria (each references can have more than
one category selected). Our goal in doing this was to compare the proportion and type of research article
utilized in forming the DEA decision with that of the current available data that ASA used to write their 8Factor Analysis.
Criteria/categories are as follows:
Peer Reviewed (Peer reviewed research articles of any type)
8 | A MER I CAN S F OR SAF E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
Non-Peer Reviewed (Agency and policy documents, journalistic pieces, no independent 3rd party
analysis)
Clinical Research (Clinical research with controlled dosing looking for therapeutic effect)
Safety Studies (may or may not have controlled dosing, not investigating therapeutic effects but
safety)
Animal (Animal based research, rats, mice and their brains)
Surveys (Sociology and epidemiology research, survey based research articles)
Human Brain (Research pertaining to the human brain, disease, and toxicology to neuronal tissue)
Reviews (Review type article and reference manuals)
Original Publication (Original research article cited, opposite of review article)
<2000 (published during the year 2000 or earlier)
>2001 (Published in the year 2001 or later)
Product Safety Related (Research on medical cannabis programs, product safety, traffic and
fatality research in states with medical cannabis programs)
F IGURE 1 P ROPORTION AND T YPE OF REFERENCES U SED IN REPORTING
AMER IC AN S F OR S AF E ACCE SS | 9
ASA | DEA’s Denial of Existing Medical Cannabis Research
Table 1. Number, Type and Percentage of Citations Used
ASA (558 Citations)
Title
HHS Report (207 Citations)
Proportion of Research
Meeting Criteria
Peer Reviewed
Number of
Citations
Proportion of Research
Meeting Criteria
Number of
Citations
93.55%
522
90.38%
188
Non-Peer reviewed
5.56%
31
9.13%
19
Clinical Research
7.17%
40
0.96%
2
Safety Studies
15.59%
87
15.38%
32
Animal
10.04%
56
13.94%
29
Surveys
18.10%
101
33.17%
69
Human Brain
11.11%
62
19.23%
40
Reviews
36.92%
206
32.21%
67
Original Publication
56.63%
315
61.06%
127
<2000
29.21%
163
38.94%
81
>2001
67.56%
377
61.06%
127
Product Safety Related
12.54%
70
2.40%
5
Data was generated by adding together all qualifying studies listed in each criterion, then dividing the
total number of articles to generate a percentage or proportion.
In comparison, ASA’s 8-Factor analysis utilized a significantly higher proportion and number of clinical
research references and product safety related publications in its determinations. The DEA devoted less
than 1% of their referenced work to addressing clinical cannabinoid trials, and consists of almost 10%
non-peer reviewed publications, as compared to ASA’s 5%.
Fully, one-third of the DEA’s report is based only on surveys, regarding sociology and epidemiology.
These types of studies largely lack any clear clinical applications or scientific relevance. For example, the
DEA repeatedly cites surveys about cannabis use and suggested associations with psychosis, while
completely disregarding clinical correlations such as research from the last 10 years demonstrating that
suicide risks are not significantly increased with use.
The discrepancy between pre- and post-2001 literature in the analysis requires additional emphasis.
Almost 40% of the DEA report relies primarily on outdated research articles, many of which have not
been reproduced by the scientific community. In contrast, the ASA 8-Factor analysis uses almost 70% of
research citations that have been published within this century, conducted with modern scientific
instrumentation and controls. Several of the research articles used in the DEA report are so dated, that
they do not provide any practical information to address current issues. Research studies primarily
10 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
published in the last 15 years, focus more on clinical studies using standardized cannabis products and
biomedical breakthroughs in multiple sclerosis, cancer, regenerative and personalized medicine.
Pertaining to safety, the DEA report does not include any research regarding more recent standards of
safety. For instance, there is no mention of the volume of product safety research that exists on cannabis
and botanical medicine regulations today. Nor does it mention any relevant medical cannabis research on
edible products from John Hopkins University, which was prominently published in the Journal of the
American Medical Association (JAMA) and covered by well over 200 media outlets upon its publication.
The DEA report also ignores the book on the quality control and quality assurances of medical cannabis
products published by the Research Triangle Institute (RTI).
This analysis provides a characterization of the DEA’s basis report.
In summation:
DEA’s basis report had only 207 citations, as compared to ASA’s 558.
ASA’s report was submitted for peer-review to external third parties; there is no evidence that the
DEA basis report was peer reviewed, there are no listed authors, and thus no accountability at the
either the FDA, DEA, or HHS.
DEA’s basis report is deficient in addressing clinically relevant harms associated with cannabis.
DEA’s basis report is deficient in addressing clinical trials with existing standardized cannabisbased medicines (2 citations; representing <1% of the citations).
Nearly 1/10th of the DEA basis report comes from non-peer reviewed sources.
Fully 1/3rd of the DEA’s basis comes from epidemiologic and survey based research, many of
which do not bare clinical significance or do not demonstrate long term harm.
The DEA’s report was deficient in its analysis and reporting of medical cannabis products, i.e.,
9,000 patient/years of placebo-controlled clinical research with nabiximols (i.e., cannabis
extracts) was not even mentioned.
While the DEA devoted a higher proportion of citations to the human brain (19%), it represents
only 40 citations. While ASA cited 62 studies on the subject, which represents about 11% of ASA’s
558 citations.
V. Evaluating the DEA’s Rationale for 1) Marijuana has a high potential for abuse.
DEA’s Evidence
“The HHS evaluation and the additional data gathered by the DEA show that marijuana has a high potential
for abuse.”
AMER IC AN S F OR S AF E ACCE SS | 11
ASA | DEA’s Denial of Existing Medical Cannabis Research
Available Scientific Data1
If medical cannabis and related products had a high potential for abuse, there would exist a significant
black market for both FDA and non-FDA approved medical cannabis products, such as FDA approved
Marinol (pure THC), and the IND approved cannabis products Sativex®, Epidiolex®, and NIDA’s
catalogue of cannabis products for research (i.e., cannabis cigarettes). However, despite decades of
availability, there is virtually no identifiable black market for NIDA’s cannabis products, FDA approved
Marinol, or the cannabis extracts Epidiolex and Sativex.
Marinol is pure THC, and can legally be created as a generic drug from the THC isolated from cannabis
plants. Epidiolex and Sativex are standardized resinous extracts from cannabis plants. According to GW
Pharmaceuticals’ website and their widely available peer-reviewed clinical publications, their cannabis
extract Sativex has already been utilized in Phase II and III clinical trials in the U.S. for almost 10 years,
and without any abuse or diversion. Furthermore, biochemical fingerprinting of this standardized
cannabis extract has been adequate for FDA CMC (Chemistry, Manufacturing, and Control) approval.
Another cannabis extract, marketed under the name Epidiolex, is part of a national clinical study in the
U.S., investigating the role of this standardized product as frontline treatment in pediatric epilepsy. The
University of Mississippi has been producing whole plant cannabis products for decades, and shipping
about 300 cannabis cigarettes a month to IND patients since 1970, yet no report exists of finding these on
the black market. GW Pharmaceuticals has produced more tonnage of cannabis than any other
organization, legal or illegal, yet their cannabis extracts are simply not found on the black market. There
exists no case whereupon either a user or abuser has arrived to a clinic for treatment of addiction related
to the abuse of NIDA cannabis cigarettes, despite decades of use by IND patients. Further, neither Europe
nor the UK have reported any significant development of a black market for medical cannabis products
such as Sativex, Marinol, or pharmaceutical grade cannabis produced by Bedrocan®.
The DEA provides substantial evidence from surveys, that a great number of people report having used
cannabis at some point within the last year. However, these surveys cited by the HHS report do not point
to any relevant or significant negative public health outcome from these patterns of mass use. Indeed,
cannabis is physiologically non-toxic (there is no known LD50 for cannabis) and is not associated with
causing any long-term negative health consequences.
VI. Evaluating the DEA’s Statement 2) Marijuana has no currently accepted
medical use in treatment in the Unites States.
“Based on the established five-part test for making such determination, marijuana has no currently accepted
medical use,” because:
For Available Scientific Data references (i.e., [553]) please refer to the bibliography of ASA’s peer reviewed 8-Factor analysis,
available at: http://www.safeaccessnow.org/8_factor_analysis_on_cannabis.
1
12 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
“As detailed in the HHS evaluation, the drug’s chemistry is not known and reproducible; there are no
adequate safety studies; there are no adequate and well-controlled studies proving efficacy; the drug is not
accepted by qualified experts; and the scientific evidence is not widely available...This five-element test,
which the HHS and DEA have utilized in all such analyses for more than two decades, has been upheld by the
Court of Appeals. ACT, 15 F.3d at 1135.”
- Drug Enforcement Administration, August 12, 2016, Denial of Petition to Initiate Proceedings to
Reschedule Marijuana
The above statement from the DEA defines that a drug has a "currently accepted medical use" if all of the
following five elements have been satisfied:
1. the drug's chemistry is known and reproducible;
2. there are adequate safety studies;
3. there are adequate and well-controlled studies proving efficacy;
4. the drug is accepted by qualified experts; and
5. the scientific evidence is widely available.
In the absence of a New Drug Application (NDA) or Abbreviated New Drug Application (ANDA) approval,
DEA has established a “five-element test” for determining whether the drug has a currently accepted
medical use in treatment in the United States. Under this test, a drug will be considered to have a
currently accepted medical use only if all five elements are satisfied.
The following are intact and unaltered quotes from the FDA’s submitted report regarding cannabis and
the five elements. While FDA maintains that cannabis does not meet the 5-element test, we think the
evidence points to the contrary.
Element (1) The drug's chemistry is known and reproducible.
Definition: “The substance’s chemistry must be scientifically established to permit it to be reproduced
into dosages which can be standardized. The listing of the substance in a current edition of one of the
official compendia, as defined by section 201(j) of the Food, Drug and Cosmetic Act, 21 U.S.C. 321(j), is
sufficient generally to meet this requirement.” 57 Fed. Reg. 10499, 10506 (March 26, 1992).
DEA/FDA Evidence for Element 1
“Marijuana, as defined in the petition, includes all Cannabis strains. (For purposes of the CSA, marijuana
includes all species of the genus Cannabis, including all strains therein). Based on the definition of marijuana
in the petition, the chemistry of marijuana is not reproducible such that a standardized dose can be created.
Chemical constituents including Δ9-THC and other cannabinoids vary significantly in marijuana samples
derived from different strains (Appendino et al., 2011). As a result, there will be significant differences in
safety, biological, pharmacological, and toxicological parameters amongst the various marijuana samples.
Due to the variation of the chemical composition in marijuana samples, it is not possible to reproduce a
standardized dose when considering all strains together. The HHS does advise that if a specific Cannabis
AMER IC AN S F OR S AF E ACCE SS | 13
ASA | DEA’s Denial of Existing Medical Cannabis Research
strain is cultivated and processed under controlled conditions, the plant chemistry may be consistent enough
to derive reproducible and standardized doses.”
Available Scientific Data for Element 1
There are two blatant issues with the DEA’s statement on element 1. First, “strains,” as listed by the DEA
report, is not a technical or botanical term, it is a vague term and not appropriate. The terms that are
appropriate to use are chemovar or chemotype (i.e., chemical variety). A chemovar is often defined as a
particular species of plants, the chemical composition of which varies from the average because of
different environmental growing conditions.
Second, the DEA report states above, “Due to the variation of the chemical composition in marijuana
samples, it is not possible to reproduce a standardized dose when considering all strains together.” This
statement is scientifically indefensible. No product or company is responsible for the scientific, nor
manufacturing shortcomings of their predecessors. The fact that confiscated drug samples vary widely in
potency across the nation, should bear no weight when discussing the products produced by licensed and
pharmaceutical manufacturers. The DEA is implying that cannabis cannot be standardized based solely
on data from their confiscated drug samples, which of course are not uniform in content. Illicit street
cannabis varies widely in content but this has no relevance to developing standardized medical products
and again it must be stressed that this is a scientifically indefensible statement from DEA.
The chemistry of cannabis is both known and reproducible. Complete cannabis monographs have been
published, including one by the American Herbal Pharmacopoeia (AHP), setting clear, peer-reviewed
guidance for standards of identity, analysis, quality control, administration, and dosing of cannabinoidbased medicine. The AHP monographs themselves are based on FDA and the United States Pharmacopeia
(USP) guidelines for all botanical medicines. Additionally, standardized cannabis products are available
from the NIDA-funded University of Mississippi marijuana farm for the FDA’s IND program – a program
that has provided standardized cannabis cigarettes to the same participants, every month, for decades.
Furthermore, the Research Triangle Institute (A NIDA-funded, DEA-compliant organization) has also
released a quality control manual for cannabis, entitled The Analytical Chemistry of Cannabis – Quality
Assessment, Assurance, and Regulation of Medicinal Marijuana and Cannabinoid Preparations.
Internationally, private companies have completed controlled clinical studies and successfully marketed
standardized cannabis products (flowers, extracts, and nabiximols) in 27 countries. In the last decade, the
U.S. has approved over 550 studies of marijuana or cannabis, 144 with dronabinol or
tetrahydrocannabinol (THC), and 96 with pure CBD or a CBD-rich cannabis extract, according to
clinicaltrials.gov.
While cannabis is dispensed in pharmacies throughout Europe and at state-regulated dispensaries in the
U.S., many conform to standards that would qualify cannabis products as botanical medicines, based on
existing safety guidelines from the FDA, AHP, and the U.S. Department of Agriculture (USDA). The quality
and safety of medical cannabis and its derivatives are adequately addressed by extant national and local
standards. These standards also address best-practices for cannabis operations – such as manufacturers,
cultivation sites, laboratories, and dispensaries.
Botanical medicines and herbal products are regulated. A diverse set of local, national and international
botanical safety standards are directly applied to medical cannabis and cannabis products. Several
14 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
countries have made significant regulatory efforts to enact the existing national and local level standards
for cannabis production and distribution [57,214,543]. Various countries have published monographs
(i.e., Czech Republic, Holland, U.S., and Canada) to specifically address quality control of cannabis,
including methodology. Trade associations, internationally, have published best practices for cultivation,
dispensing, manufacturing, and laboratory practices [544]. Furthermore, an abundance of national and
international guidance documents provide quality control standards that address nearly every aspect of
quality control and product safety for botanical substances, such as cannabis and its derivatives.
One hurdle to quality control of medical cannabis products is the existing control status of cannabis in
countries such as the U.S., as well as controls under the conventions. National and international controls
prevent adequate product testing in U.S. cannabis programs, and may therefore inadvertently jeopardize
public health. To date, there has only been a single study that examined labeling accuracy (i.e., potency) of
those cannabis products’ accessed through three state programs in the U.S. – A study that demonstrated
that medical cannabis product labels can be inaccurate [545]. However, this U.S. study also demonstrated
that the current national controls for cannabis serve to impair the ability to address public health
concerns concerning medical cannabis and its derivatives.
It is difficult to address public health issues regarding medical cannabis products while it remains in
Schedule I status. As the DEA tightly controls the release of analytical-quality standards for calibrating
scientific instruments, cannabinoid compounds can only be purchased in necessary amounts if the
operation has received a Schedule I license from the DEA. However, the DEA will not grant a Schedule I
license to a state sponsored medical cannabis laboratory, because the laboratory would receive medical
cannabis samples for analysis from non-DEA licensed sources (such as state licensed manufacturers,
distribution centers, cultivation sites, patients, or doctors that recommend cannabis to patients).
Therefore, the Schedule I status of Cannabis blocks most laboratories from determining the precise
potency of a product. In contrast, testing for clinically relevant contaminants – such as heavy metals,
bacteria, and fungus – can proceed without requiring DEA licensure, but this product safety testing is just
as vulnerable to DEA or federal interference due to the scheduling status.
A potential normalizing factor for a medicine like cannabis in the U.S. could be for the USP to create a
cannabis monograph; these standards could then be adopted to regulate cannabis as a medicinal product
nationally [546]. However, such an action would grant pharmacists in the U.S. the ability to work with
cannabis, which is forbidden by the DEA. Hence, the USP cannot create a cannabis monograph and still
maintain compliance with the DEA. Presently, the USP defers to the AHP monograph as the current
standard for cannabis products in the U.S. [7]. A recent meeting of the USP suggested that drafting of the
document will not begin until cannabis is rescheduled – at least to a status that recognizes its medicinal
use and outstanding safety profile. This lack of a permitted monograph (i.e., from the USP) is one of the
issues that is directly responsible for the horrendous dereliction of responsibility in the industry to
produce well-characterized, non-toxic products. A terrible public health threat has resulted from this
policy. The best illustration is the pesticide contamination of legal cannabis in the Washington State
market, that many patients now have no option but to utilize.
The standards issued by the AHP monograph and American Herbal Products Association (AHPA) have
been adopted by 16 U.S. states to regulate product safety for their respective medical cannabis programs.
Furthermore, AHPA – the trade association for the herbal products industry – has issued its medical
cannabis manufacturing guidelines, completing its series of recommendations for state regulators in the
areas of manufacturing, packaging and labeling, cultivation, dispensary operations, and laboratory
AMER IC AN S F OR S AF E ACCE SS | 15
ASA | DEA’s Denial of Existing Medical Cannabis Research
practices. Another example of medicinal cannabinoid production with outstanding quality
assurances/controls exists in the Dutch program for medicinal cannabis. Produced under responsibility
of the Ministry of Health, the program meets a number of quality requirements including, but not limited
to: consistent strength on THC and composition of secondary cannabinoids, absence of microbiological
contamination, pesticides and heavy metals, and humidity. Where there is a norm provided in the
European Pharmacopoeia, this norm is followed [547].
The next sections below briefly discuss published resources and guidance documents being utilized by
world governments to provide proper quality control and product safety for agricultural products and
botanical medicines, including cannabis.
Good Agricultural and Collection Practices
The quality of raw material for botanical medicine can be safeguarded by using Good Agricultural and
Collection Practices (GACP, aka GAP) to the extent possible in all aspect of growing, harvesting, and
storage [548]. Specific guidelines for regulators regarding cannabis cultivation practices in the U.S. have
been published by the AHPA. These standards include requirements for standard operating procedure
documentation, employee safety training, security, and batch tracking [544]. Similarly, the American
Herbal Pharmacopoeia has also released standards of quality control for cannabis cultivation.
In the Netherlands, Czech Republic, and Italy, medicinal cannabis must be produced under GMP-like
conditions. All products must to be fully tested (by an independent laboratory) for cannabinoid content,
absence of heavy metals, aflatoxins, pesticides (residue), and microbes to a level of <10 cfu.
Standardization of cannabis and cannabis derivatives – according to the monograph of herbal medicines
of the European Medicine Agency (EMA) – is mandatory and must be proven for each batch produced.
In Austria (AGES) and the UK (GW Pharmaceuticals, Ltd), cannabis is required to be produced under GAP,
but any derivatives of this cannabis must be produced under GMP. Finished products must be
standardized according to regular pharmaceutical products.
Good Manufacturing Practice for Cannabis
Many guidance documents are available for reference and use in the manufacturing of plant medicines
and products, and any facility manufacturing products for human consumption should follow GMP. The
World Health Organization has published guidelines on manufacturing botanical and herbal medicines,
and the U.S. FDA has published guidance documents as well [549-552]. The AHPA manufacturing
guidelines have a specific procedure for the recall of medical cannabis products, in the case of cannabis
materials that do not meet “appropriate standards of identity, purity, strength, and composition and their
freedom from contamination or adulteration.” The AHP cannabis monograph also sets limits for residues
such as solvents and pesticides, heavy metals, bacteria, and fungi [214].
Good Laboratory Practices
Methods used to determine potency should be scientifically validated by laboratories for several criteria
including, but not limited to: specificity, linearity, accuracy, precision, and ruggedness. The FDA and other
organizations (i.e., AHPA, USP, and AHP) have provided extensive guidance documents that represent the
current thinking on method validation and other aspects of good laboratory practices. There are further
16 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
international standards for analyzing medical cannabis products, which have been issued, for example, by
the UN’s Office of Drugs and Crime in their document, entitled Recommended Methods for the
Identification and Analysis of cannabis and cannabis products [553].
Below are a few examples of applicable guidance from a regulatory perspective, for analytical method
validation for new methods, or methods not outlined in existing international and national regulatory
documents:
USP–NF, Validation of Compendial Methods; USP pharmacopeia 35, United States Pharmacopeia
Convention, Inc., Rockville, MD. May 1, 2012 – December 1, 2012.
U.S. FDA, Center for Drug Evaluation and Research (CDER), Reviewer Guidance on Validation of
Chromatographic Methods, November 1994.
American Herbal Pharmacopoeia Cannabis Inflorescence. Standards of Identity, Analysis, and
Quality Control (2013).
Quality control and quality standards for medicinal cannabis have been developed and adopted by 16 U.S.
states and many countries, including Canada, Israel, the Netherlands, and the Czech Republic. Current
standards are presently being appropriately applied or implemented through third party licensed
certification bodies, for regulating cannabis and cannabis-related products for human consumption.
Both the AHP and AHPA documents point to Patient Focused Certification (PFC) for implementation of
these standards. PFC has offices in Washington, DC and the Czech Republic. PFC is the only international
program that can verify that a country, state, or region’s cannabis standards are being followed.2 PFC
conducts both physical (i.e. site or facility) and documentation audits of the operation, to generate an
audit report that is submitted to a review board. PFC’s review board features experts that have served in
regulatory and scientific roles in U.S. presidential administrations, at the USDA, in quality control
laboratories, and related disciplines. PFC audited its first cannabis operations in the U.S. in 2013 and in
Europe in 2015, and is now an option for regulators in every country, state, or region with medical
cannabis access programs.
An undeniably successful public health outcome of product safety regulations has been demonstrated
through numerous successful product recalls in Canada and the U.S. Recalls required the cooperation of
government, manufacturers, and 3rd party certifying bodies, resulting in consumer protection [554-560].
To address public health concerns regarding the increasing availability of medical cannabis products, the
scheduling status of cannabis needs to be thoughtfully and deliberately rescheduled (or descheduled), in
order for producers, cultivators, manufacturers, laboratories, clinicians, researchers, and regulators to
fully implement quality control standards for medical cannabis products.
Element (2) There are adequate safety studies.
Definition: “There must be adequate pharmacological and toxicological studies, done by all methods
reasonably applicable, on the basis of which it could fairly and responsibly be concluded, by experts
2
For more information about the PFC program, see: www.patientfocusedcertification.org.
AMER IC AN S F OR S AF E ACCE SS | 17
ASA | DEA’s Denial of Existing Medical Cannabis Research
qualified by scientific training and experience to evaluate the safety and effectiveness of drugs, that the
substance is safe for treating a specific, recognized disorder.” 57 Fed. Reg. 10499, 10506 (March 26,
1992).
DEA/FDA Evidence for Element 2
“The HHS stated that there are no adequate safety studies on marijuana. As indicated in their evaluation of
Element #1, the considerable variation in the chemistry of marijuana complicates the safety evaluation. The
HHS concluded that marijuana does not satisfy Element #2 for having adequate safety studies such that
medical and scientific experts may conclude that it is safe for treating a specific ailment.”
Available Scientific Data for Element 2
Cannabis products have been on the market for decades, and have shown clearly acceptable safety
standards for use under medical supervision. Smoked, vaporized, or ingested cannabinoid medicine can
deliver consistent amounts of active constituents, while toxic and/or lethal overdose of cannabis is not
achievable and remains undocumented in either scientific or medical literature.
Sixteen states have adopted the national standards and guidance provided by the AHPA Cannabis Best
Practices documents and the American Herbal Pharmacopoeia Cannabis Inflorescence Standards of
Identity, Analysis, and Quality Control monograph. Federal standards are not available for cannabis and
will not be produced by the USP while the plant is Schedule I, because the USP would thusly fall out of
compliance with Drug Enforcement Administration (DEA) standards. Meanwhile, the FDA has approved
several cannabis studies and a new IND program with a cannabis extract (marketed as Epidiolex),
currently being administered to children in hospitals across the U.S with positive results.
While street marijuana arguably has a higher potential for abuse, standardized cannabis products
accessed through a regulated program do not appear to have such high societal potential of abuse.
Standardized cannabis-based medicines have been on the market for decades in the U.S. (Marinol and
Nabilone), and whole-plant cannabis medicines are now available in 27 other countries (Bedrocan and
nabiximols) [60]. Common sense dictates that self-administration of unstandardized, untested street
drugs possesses a high potential for abuse, but the data addressing cannabis does not report, document,
nor support the notion of significant abuse or divergence with standardized cannabis products. Cannabis
should therefore be rescheduled because standardized preparations show very low potential for abuse
and, therefore, possess minimal street value or resale value.
Based on current understanding of basic toxicity research – sedation, cytotoxicity, genotoxicity, etc. –
cannabis and its components have a uniquely wide safety margin [36-39]. To date, there has never been a
single well-documented case of human fatality attributable to an overdose of cannabis or its components,
and no experimental or non-extrapolated LD50 can be attributed to a toxic or lethal overdose of cannabis
or a preparation thereof. No scientifically significant negative neuropsychological sequelae have yet been
attributable to cannabis usage. The meta-analytical study of long-term cannabis use on neurocognitive
functioning, results failed to find any substantial, systematic effect on users who were not concurrently
intoxicated. Claims of brain damage and cerebral atrophy are not supported by current evidence. When
controlling for pertinent variables such as age, gender, and history of alcohol use, research has not been
able to show any association between the use of cannabis and changes in brain structures [59].
18 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
Short-term use of existing standardized medical cannabis and cannabis products appear to increase the
risk of non-serious adverse events. Risks associated with long-term cannabis use are poorly
characterized in published clinical trials and observational studies; however, the cognitive effects
observed in long-term users do not appear to be permanent in nature [40]. With the exception of very
limited studies on synthetic endocannabinoid system modulators, cannabis medicines do not appear to
cause significant serious adverse events.
Arguably, some prior studies remain limited by a number of factors that need to be controlled in future
investigations. Primarily, cannabis use and dosing needs to be confirmed in users with biological and
chemical tests, as issues of dosing and patterns of use are confounding factors when not adjusted for.
Element (3) There are adequate and well-controlled studies proving efficacy.
“There must be adequate, well-controlled, well-designed, well-conducted and well- documented studies,
including clinical investigations, by experts qualified by scientific training and experience to evaluate the
safety and effectiveness of drugs, on the basis of which it could be fairly and responsibly concluded by
such experts that the substance will have the intended effect in treating a specific, recognized disorder.”
57 Fed. Reg. 10499, 10506 (March 26, 1992).
DEA/FDA Evidence for Element 3
“As indicated in the HHS’s review of marijuana (HHS, 2015), there are no adequate or well-controlled studies
that prove marijuana’s efficacy. The FDA independently reviewed (FDA, 2015) publicly available clinical
studies on marijuana published prior to February 2013 to determine if there were appropriate studies to
determine marijuana’s efficacy (please refer to FDA, 2015 and HHS, 2015 for more details). After review, the
FDA determined that out of the identified articles, including those identified through a search of
bibliographic references and 566 abstracts located on PubMed, 11 studies met the a priori selection criteria,
including placebo control and double-blinding. FDA and HHS critically reviewed each of the 11 studies to
determine if the studies met accepted scientific standards. FDA and HHS concluded that these studies do not
“currently prove efficacy of marijuana” for any therapeutic indication due to limitations in the study designs.
The HHS indicated that these studies could be used as proof of concept studies, providing preliminary
evidence on a proposed hypothesis involving a drug’s effect.”
Available Scientific Data for Element 3
To date, more than 30,000 modern peer-reviewed scientific articles on the chemistry and pharmacology
of cannabis and the cannabinoids have been published. More than 1,500 articles investigating the body's
naturally-occurring endocannabinoids are published every year. In recent years, modern gold-standard
placebo-controlled human trials have also been conducted.
At the time of writing this document, according to clinicaltrials.gov, there are hundreds of approved
human research studies utilizing cannabinoids – A total of 144 are approved for THC, 96 are approved for
CBD, and 559 are approved for cannabis. These studies are currently either completed, recruiting,
approved, or in process. Due to the Schedule I status, however, medical cannabis preparations such as
nabiximols and CBD-rich extracts are imported and cannot be manufactured in the U.S., even though they
are licensed pharmaceutical products.
AMER IC AN S F OR S AF E ACCE SS | 19
ASA | DEA’s Denial of Existing Medical Cannabis Research
A 2009 review of clinical studies conducted over a 38-year period found that “nearly all of the 33
published controlled clinical trials conducted in the U.S. have shown significant and measurable benefits
in subjects receiving the treatment,” [148]. The review's authors made particular effort to note that
cannabinoids have the capacity for analgesia through neuromodulation in ascending and descending pain
pathways, neuroprotection, and by anti-inflammatory mechanisms – all of which indicate that the
cannabinoids found in cannabis have applications in significantly managing chronic pain, muscle
spasticity, cachexia, and other variously debilitating conditions.
There is a wealth of clinical information available on the uses of standardized medical cannabis products.
The FDA has approved new drug applications for cannabis products. For example, a CBD-rich extract
(marketed as Epidiolex) is an imported, purified cannabis extract that has been approved for clinical use
in children and is currently in clinical practice across several institutions in the U.S. Additionally, an
inhaled cannabis study has recently been approved for investigating therapeutic effects in PTSD.
Cannabis currently has accepted medical uses in 42 states and the District of Columbia and,
appropriately, its products have mandatory testing requirements. A cannabis nabiximols (Sativex), a
whole-plant ethanolic extract, has generated more than 9,000 patient/years of modern clinical data for
the treatment of chronic pain [126].
Currently, cannabis is most often recommended as a complementary or adjunctive medicine. However,
there exists a substantial consensus amongst experts in the relevant disciplines – including the American
College of Physicians – that cannabis and cannabinoid-based medicines have undeniable therapeutic
properties that could potentially treat a wide spectrum of serious and chronic illnesses.
Element (4) The drug is accepted by qualified experts.
Definition: “[A] consensus of the national community of experts, qualified by scientific training and
experience to evaluate the safety and effectiveness of drugs, accepts the safety and effectiveness of the
substance for use in treating a specific, recognized disorder. A material conflict of opinion among experts
precludes a finding of consensus.” 57 Fed. Reg. 10499, 10506 (March 26, 1992).
DEA/FDA Evidence for Element 4
“The HHS concluded that there is currently no evidence of a consensus among qualified experts that
marijuana is safe and effective in treating a specific and recognized disorder. The HHS indicated that
medical practitioners who are not experts in evaluating drugs cannot be considered qualified experts (HHS,
2015; 57 FR 10499, 10505). Further, the HHS noted that the 2009 American Medical Association (AMA)
report entitled, “Use of Cannabis for Medicinal Purposes” does not conclude that there is a currently
accepted medical use for marijuana. HHS also pointed out that state-level “medical marijuana” laws do not
provide evidence of such a consensus among qualified experts.”
Available Scientific Data for Element 4
In ASA’s 8-Factor analysis, under the section entitled “List of Medical and Scientific Organizations that
have Issued Letter of Support for Medical Cannabis,” there are over 200 medical, scientific, health
professionals, religious and community organizations who accept cannabis as a medicine and have issued
letters in support of this medicine
20 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
In April 2016, the Federation of State Medical Boards (FSMB) adopted “Model Guidelines for the
Recommendation of Marijuana in Patient Care.”3
The National Cancer Institute – one of 11 federal agencies under the National Institutes of Health –
changed its website to include cannabis as a Complementary Alternative Medicine, with possible benefits
for people living with cancer.4
Statements from Qualified Experts and Medical Organizations
“Based on much evidence, from patients and doctors alike, on the superior effectiveness and safety of whole
Cannabis (marijuana) compared to other medicines for many patients — suffering from the nausea
associated with chemotherapy, the wasting syndrome of AIDS, and the symptoms of other illnesses … we
hereby petition the Executive Branch and the Congress to facilitate and expedite the research necessary to
determine whether this substance should be licensed for medical use by seriously ill persons.” - American
Academy of Family Physicians
The American Medical Association “urges that marijuana’s status as a federal Schedule I substance be
reviewed with the goal of facilitating the conduct of clinical research and development of cannabinoid-based
medicines.”
The American College of Physicians “urges an evidence-based review of marijuana’s status as a Schedule
I controlled substance to determine whether it should be reclassified to a different schedule.”
The American Public Health Association “adopted a resolution [...] which urged federal and state drugs
laws to exclude Marijuana as a narcotic drug,” and “conclude[d] that Cannabis was wrongfully placed in
Schedule I of Controlled Substances, depriving patients of its therapeutic potential.”
“Marijuana should be available for appropriate medicinal purposes, when such use is in accordance with
state law, and that physicians who recommend and prescribe marijuana for medicinal purposes in states
where such use is legal, should not be censured, harassed, prosecuted or otherwise penalized by the federal
government.” - American Preventive Medical Association
“The Texas Medical Association supports (1) the physician's right to discuss with his/her patients any and all
possible treatment options related to the patients' health and clinical care, including the use of marijuana,
without the threat to the physician or patient of regulatory, disciplinary, or criminal sanctions; and (2)
further well-controlled studies of the use of marijuana with seriously ill patients who may benefit from such
alternative treatment.” - Texas Medical Association
The Rhode Island Medical Society has stated that “[T]here is sufficient evidence for us to support any
physician-patient relationship that believes the use of marijuana will be beneficial to the patient.”
“The definitive review of scientific studies ... found medical benefits related to pain relief, control of nausea
and vomiting, and appetite stimulation ... While there are a variety of ways of supplying marijuana for
3
4
See www.medicalCannabis.com/about/health-care-professionals/supporting-organizations.
See http://www.cancer.gov/about-cancer/treatment/cam/hp/Cannabis-pdq - section/all.
AMER IC AN S F OR S AF E ACCE SS | 21
ASA | DEA’s Denial of Existing Medical Cannabis Research
medical use, serious consideration should be given to the 1997 recommendation ... that the FDA reclassify
marijuana from Schedule I and provide a consistent, safe supply.” - New York County Medical Society
“The American Medical Student Association strongly urges the United States Government … to meet the
treatment needs of currently ill Americans by restoring the Compassionate (Investigational New Drug)
program for medical marijuana, and … reschedule marijuana to Schedule II of the Controlled Substances
Act, and … end the medical prohibition against marijuana.” - American Medical Student Association
“The National Nurses Society on Addictions urges the federal government to remove marijuana from the
Schedule I category immediately, and make it available for physicians to prescribe. NNSA urges the
American Nurses’ Association and other health care professional organizations to support patient access to
this medicine.” - National Nurses Society on Addictions
“The American Cancer Society supports the need for more scientific research on cannabinoids for cancer
patients, and recognizes the need for better and more effective therapies that can overcome the often
debilitating side effects of cancer and its treatment. The Society also believes that the classification of
marijuana as a Schedule I controlled substance by the US Drug Enforcement Administration imposes
numerous conditions on researchers and deters scientific study of cannabinoids. Federal officials should
examine options consistent with federal law for enabling more scientific study on marijuana.” - American
Cancer Society
“The Society supports the rights of people with MS to work with their MS health care providers to access
marijuana for medical purposes in accordance with legal regulations in those states where such use has
been approved. In addition, the Society supports advancing research to better understand the benefits and
potential risks of marijuana and its derivatives as a treatment for MS.” - National Multiple Sclerosis
Society
“The Epilepsy Foundation supports the rights of patients and families living with seizures and epilepsy to
access physician directed care, including medical marijuana. Nothing should stand in the way of patients
gaining access to potentially life-saving treatment. If a patient and their healthcare professionals feel that
the potential benefits of medical marijuana for uncontrolled epilepsy outweigh the risks, then families need
to have that legal option now — not in five years or ten years. For people living with severe uncontrolled
epilepsy, time is not on their side. This is a very important, difficult, and personal decision that should be
made by a patient and family working with their healthcare team.” - Epilepsy Foundation
“(T)he Leukemia & Lymphoma Society supports legislation to remove criminal and civil sanctions for the
doctor-advised, medical use of marijuana by patients with serious physical medical conditions.” - Leukemia
& Lymphoma Society
Medical schools are teaching required coursework which includes the endocannabinoid system and the
therapeutic applications of cannabis. One example, theanswerpage.org, a Harvard University based CME,
is educating physicians about the benefits of the medical uses of cannabis. This has led to the creation of
22 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
clinical cannabis certification for physicians; an educational program that is required for physicians to
recommended medical cannabis in states such programs.5
Element (5) The scientific evidence is widely available.
“In the absence of NDA approval, information concerning the chemistry, pharmacology, toxicology, and
effectiveness of the substance must be reported, published, or otherwise widely available, in sufficient
detail to permit experts, qualified by scientific training and experience to evaluate the safety and
effectiveness of drugs, to fairly and responsibly conclude the substance is safe and effective for use in
treating a specific, recognized disorder.” 57 Fed. Reg. 10499, 10506 (March 26, 1992).
DEA/FDA Evidence for Element 5
“The HHS concluded that the currently available data and information on marijuana is not sufficient to
allow scientific scrutiny of the chemistry, pharmacology, toxicology, and effectiveness. In particular,
scientific evidence demonstrating the chemistry of a specific Cannabis strain that could provide
standardized and reproducible doses is not available.”
Available Scientific Data for Element 5
One of the criteria preventing the rescheduling of cannabis is the notion that information about this
medicine is not widely available. There are tens of thousands of peer reviewed articles available through
online portals, journal websites, and other resources for health professionals to access clinical
information about cannabis, including but not limited to: Springer, Wiley, Pubmed, Public Libraries,
medical and graduate school libraries, and websites of expert groups such as Americans for Safe Access,
theAnswerpage.org, and the International Cannabis and Cannabinoid Institute.
The Internet has also revolutionized cannabinoid research and science, by allowing the generation of, and
access to, large amounts of information that would have previously been nearly impossible to obtain.
People across the globe can now access innumerable sources (a search for ‘cannabis research’ through
web of science alone yields 120,000 separate articles) of previously unavailable scientific and clinical
information.
Furthermore, the nabiximol Sativex is extracted from two fully-characterized, standardized cannabis
chemovars, one of which is called Skunk No.1. It is odd, therefore, that the FDA would claim, “scientific
evidence demonstrating the chemistry of a specific cannabis strain that could provide standardized and
reproducible doses is not available.” While according to NIDA, DEA, FDA, and RTI, University of
Mississippi researchers have grown several types of cannabis strains for decades, which are allegedly
turned into standardized products for clinical research under the supervision and participation of NIDA,
DEA, FDA and RTI6.
5
For more information about Cannabis Care Certification, see http://cannabiscarecertification.org.
See Thomas, B. F., & ElSohly, M. (2015). The Analytical Chemistry of Cannabis.Quality Assessment, Assurance, and Regulation
of Medicinal Marijuana and Cannabinoid Preparations. Elsevier.
6
AMER IC AN S F OR S AF E ACCE SS | 23
ASA | DEA’s Denial of Existing Medical Cannabis Research
It is simply disingenuous for an organization to state that no standardized cannabis product exists, while
simultaneously licensing both the production and distribution of such products.
VII. Evaluating the DEA’s Statement 3) Marijuana lacks accepted safety for use
under medical supervision.
DEA’s Evidence Regarding Safety
“At present, there are no marijuana products approved by the U.S. Food and Drug Administration (FDA), nor
is marijuana under a New Drug Application (NDA) evaluation at the FDA for any indication. The HHS
evaluation states that marijuana does not have a currently accepted medical use in treatment in the United
States or a currently accepted medical use with severe restrictions. At this time, the known risks of
marijuana use have not been shown to be outweighed by specific benefits in well-controlled clinical trials
that scientifically evaluate safety and efficacy.”
Available Scientific Data Regarding Safety
According to the CSA statute, as cited by the DEA in their evaluation:
“The CSA defines marijuana as the following:
All parts of the plant Cannabis Sativa L., whether growing or not; the seeds thereof; the resin extracted from
any part of such plant; and every compound, manufacture, salt, derivative, mixture, or preparation of such
plant, its seeds or resin. Such term does not include the mature stalks of such plant, fiber produced from such
stalks, oil or cake made from the seeds of such plant, any other compound, manufacture, salt, derivative,
mixture, or preparation of such mature stalks (except the resin extracted therefrom), fiber, oil, or cake, or
the sterilized seed of such plant which is incapable of germination (21 U.S.C. 802(16)).”
This definition means that THC and CBD isolated from the plant are “resins”. Yet, the DEA states clearly
under item 3:
“At present, there are no marijuana products approved by the U.S. Food and Drug Administration (FDA), nor
is marijuana under a New Drug Application (NDA) evaluation at the FDA for any indication.”
This statement is incorrect. Marinol, an FDA approved form of pure THC, can now be generically made
from THC isolated from cannabis plants, such as those from the University of Mississippi. Marinol started
out as synthetic THC, but can now be plant-derived, however the DEA report is implying that cannot
occur. No companies have admitted to pursuing this path, but it is an approved generic form of Marinol
by the FDA. As defined by the CSA, both Epidiolex® and Sativex® are resinous cannabis extracts, and are
presently undergoing clinical studies in the United States. According to GW Pharmaceutical’s website,
Phase III trials got underway in 2015, utilizing the cannabis extract Sativex® with FDA approval. Both
standardized cannabis extracts marketed by GW Pharmaceutical (Epidolex® and Sativex®) continue to be
imported and are undergoing clinical study in the United States.
This arbitrary interpretation of the CSA is used to simultaneously and systematically prevent any
discussion of the nearly 100 clinical trials completed with cannabis products while, at the same time,
24 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
THC, CBD, Sativex, and NIDA-generated cannabis cigarettes are considered “marijuana” if the user is
prosecuted.
This is another example of how the DEA report seems to follow a more politically-driven agenda, rather
than one of modern science and medicine. By attempting to redefine the CSA as meaning only whole plant
cannabis, when it was intended to include derivatives and extracts thereof, the DEA is allowed to
generate reports and statements that are not based on scientific research. The systematic use of biased
methods to generate reports on scientific data leaves large swaths of modern cannabinoid research
unheeded. Hence, the clinical references in the HHS 8-factor analysis consists of less than 1% of the
discussed research. If the DEA report had included more than two clinical studies in their HHS 8-factor
analysis, this would be a different conversation.
VIII. Conclusion
The goal of this comparative analysis is to objectively examine the data used in the DEA’s determination
of their denial to allow a petition to reschedule cannabis, and to compare it to the prevailing scientific
data on the medical value of cannabis. While we agree with portions of the DEA report – such as the lack
of evidence to support either diversion or black market sales or the “gateway” hypothesis, we do not
agree with either the process or the evidence upon which their denial was based. By applying politics and
ideology, while excluding current scientific information, the DEA can only further the passage of truly
inaccurate statements, which might then then be used to establish inaccurate laws regarding health and
medicine.
Ideology and politics should never be allowed to eclipse the available scientific and clinical truth in
matters of medicine or the health of our citizenry. This DEA report highlights how the use of engrained,
historically inaccurate political beliefs to arbitrarily interpret the CSA has been exploited at the expense
of public health. This stems from the fact that the DEA alone, inexplicably, has been allowed to determine
how “medicine” is defined in this country, with little to no accountability.
Unfortunately, cannabis will never be rescheduled under these Catch-22-like circumstances. The CSA is
arbitrarily used, on one hand, to exclude all medical research on derivatives of cannabis from their
report…while, on the other hand, it is used to prosecute anyone in possession of those derivatives.
Persistent misinterpretation of existing laws, coupled to lack of scientific knowledge, results in a very
dangerous and socially destructive policy for a government enforcement agency.
The documents submitted in the report for the denial of the petitions are contradictory, and would
appear to have little or no relevance to either contemporary cannabinoid science or medicine. Even so, as
there were no clear negative public health implications relating to moving cannabis out of Schedule I
status presented therein, it would appear that the DEA has chosen a disingenuous, overtly biased
response to legitimate medicinal cannabinoid progress.
This type of response is responsible for the pitfalls of the current cannabis market by preventing the
implementation of suitable controls. Such as addressing the pesticide contamination in the legal adult use
AMER IC AN S F OR S AF E ACCE SS | 25
ASA | DEA’s Denial of Existing Medical Cannabis Research
markets as a key case in point. Interference with product safety that results directly from ideological
policies, is a dereliction of responsibility that supports a major public health threat.
Recommendations
Pass CARERS
Congress should pass The Compassionate Access, Research Expansion, and Respect for States (CARERS)
Act (S. 683, H.R. 1538) as introduced in 2015 which, in addition to rescheduling cannabis and removing
cannabidiol (CBD) from the schedule entirely, allows states to establish medical cannabis access laws and
product safety regulations without interference by the federal government, and removes current
obstacles to research. The CARERS Act is currently stalled in the Senate Judiciary Committee, with
Chairman Chuck Grassley (IA) refusing to hold a vote.
Update Information on DEA Website and Educational Materials
We also recommend that the DEA update the following on their website and in education materials
provided online. The updates should be made to reflect the information from the current DEA report.
1. DEA statements regarding adverse health effects related to cannabis
● “[According to an Australian study,] there is now conclusive evidence that smoking cannabis hastens
the appearance of psychotic illnesses by up to three years.”7
● “Marijuana’s effects on these abilities may last a long time or even be permanent.”8
Requested change to reflect current information from the DEA’s report:
On page 12, the DEA report states, “Abundant scientific data are available on the neurochemistry,
toxicology, and pharmacology of marijuana.”9
On page 20, the DEA report states, “cannabis-associated cognitive deficits are reversible and related to
recent cannabis exposure, rather than irreversible and related to lifetime use.”10
From a document entitled Dangerous and Consequences of Marijuana Abuse (page 12):
https://www.dea.gov/docs/dangers-consequences-marijuana-abuse.pdf
7
And from From “DrugFacts: Marijuana.” Link available through DEA website/a DEA resource site:
https://www.drugabuse.gov/publications/drugfacts/marijuana
8
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-81
9
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-123
10
26 | AMER ICAN S F OR SA F E AC CE SS
ASA | DEA’s Denial of Existing Medical Cannabis Research
On page 22, the DEA report states, “At present, the available data do not suggest a causative link between
marijuana use and the development of psychosis.”11
2. Statements from DEA regarding the “gateway theory”12
● “Teens who experiment with marijuana may be making themselves more vulnerable to heroin
addiction later in life, if the findings from experiments with rats are any indication. Cannabis has very
long-term, enduring effects on the brain…”(pg. 37)
● “Marijuana use in early adolescence is particularly ominous. Adults who were early marijuana users
were found to be five times more likely to become dependent on any drug, eight times more likely to
use cocaine in the future, and fifteen times more likely to use heroin later in life.” (pg. 38)
● “Marijuana is a frequent precursor to the use of more dangerous drugs and signals a significantly
enhanced likelihood of drug problems in adult life.”(pg. 37)
Below are the requested changes to reflect the current information from the DEA report:
On page 43, the DEA report states, “Overall, research does not support a direct causal relationship between
regular marijuana use and other illicit drug use.”13
On page 44, the DEA report states, “the gateway hypothesis only addresses the order of drug use initiation,
the gateway hypothesis does not specify any mechanistic connections between drug "stages" following
exposure to marijuana and does not extend to the risks for addiction.”14
On page 162, the DEA report states, “The HHS reviewed the clinical studies evaluating the gateway
hypothesis in marijuana and found them to be limited.” The DEA goes on to say, “The HHS cited several
studies where marijuana use did not lead to other illicit drug use.”15
On page 162, the DEA report states, “Based on these studies among others, the HHS concluded that
although many individuals with a drug abuse disorder may have used marijuana as one of their first illicit
Link to paragraph:
https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiate-proceedings-to-reschedulemarijuana#p-131
11
12https://www.dea.gov/docs/marijuana_position_2011.pdf
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-237
13
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-242
14
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-956
15
AMER IC AN S F OR S AF E ACCE SS | 27
ASA | DEA’s Denial of Existing Medical Cannabis Research
drugs, this does not mean that individuals initiated with marijuana inherently will go on to become regular
users of other illicit drugs.”16
Over 40 years ago the “gateway” hypothesis of cannabis was proposed. The report concludes predictably,
that the gateway theory of cannabis is not supported by the evidence. We agree that the hypothesis
attempted but failed to predict that cannabis use leads to the addiction of other drugs.
3. Statements from the DEA regarding cannabis and cancer17
“Marijuana smoking has been implicated as a causative factor in tumors of the head and neck and of the
lung.”(pg.34)
“Marijuana takes the risks of tobacco and raises them. Marijuana smoke contains more than 400 chemicals
and increases the risk of serious health consequences, including lung damage.”(pg 36)
Below are the requested changes to reflect the current information from the DEA report:
“However, in a large clinical study with 1,650 subjects, no positive correlation was found between marijuana
use and lung cancer (Tashkin et al., 2006). This finding held true regardless of the extent of marijuana use
when both tobacco use and other potential confounding factors were controlled. The HHS concluded that
new evidence suggests that the effects of smoking marijuana on respiratory function and cancer are
different from the effects of smoking tobacco (Lee and Hancox, 2011).”18
“The DEA further notes the publication of recent review articles critically evaluating the association
between marijuana and lung cancer. Most of the reviews agree that the association is weak or inconsistent
(Huang et al., 2015; Zhang et al., 2015; Gates et al., 2014; Hall and Degenhardt, 2014). Huang et al. (2015)
identified and reviewed six studies evaluating the association between marijuana use and lung cancer and
the authors concluded that an association is not supported most likely due to the small amounts of
marijuana smoked in comparison to tobacco. Zhang et al. (2015) examined six case control studies from the
US, UK, New Zealand, and Canada within the International Lung Cancer Consortium and found that there
was a weak association between smoking marijuana and lung cancer in individuals who never smoked
tobacco, but precision of the association was low at high marijuana exposure levels...overall association is
weak between marijuana use and lung cancer especially when controlling for tobacco use.” 19
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-959
16
From DEA Position on Marijuana document:
https://www.dea.gov/docs/marijuana_position_2011.pdf
17
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-860
18
Link to paragraph: https://www.federalregister.gov/articles/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-861
19
28 | AMER ICAN S F OR SA F E AC CE SS
Enclosure C: “IQA Request for Correction of Information Disseminated by DEA Regarding Marijuana (Cannabis)” The Dangers and Consequences of Misinformation on Marijuana February 22, 2017
BEFORE THE UNITED STATES DEPARTMENT OF JUSTICE
INFORMATION QUALITY GUIDELINES STAFF
_____________________________________________
)
Re: DEA’s “The Dangers and Consequences )
of Marijuana Abuse” and “Drugs of Abuse”
)
)
______________________________________ )
REQUEST FOR CORRECTION OF INFORMATION DISSEMINATED
BY DEA REGARDING MARIJUANA (CANNABIS)
INFORMATION QUALITY ACT REQUEST FOR CORRECTION
DATE: DECEMBER 5, 2016
SUBMITTED BY: AMERICANS FOR SAFE ACCESS FOUNDATION
Attorneys for Petitioner
Executive Director for Petitioner
Vickie L. Feeman
[email protected]
Rick Fukushima
[email protected]
Alex Fields
[email protected]
Orrick, Herrington & Sutcliffe LLP
1000 Marsh Road
Menlo Park, CA 94025
Phone: (650) 614-7400
Steph Sherer
1624 U Street, NW
Suite 200
Washington, D.C. 20009
Phone: (202) 857-4272
Fax: (202) 618-6977
[email protected]
Americans for Safe Access
Request for Correction Pursuant to the DOJ’s Information Quality Guidelines
ISSUE
The Drug Enforcement Agency’s (“DEA”) website (dea.gov) contains inaccurate statements that
do not meet the standards of quality required by the Department of Justice (“DOJ”) and Office of
Management and Budget (“OMB”) under the Information Quality Act (“IQA”). In particular,
the DEA continues to disseminate certain statements about the health risks of medical cannabis
use that have been incontrovertibly refuted by the DEA itself in its recent “Denial of Petition to
Initiate Proceedings to Reschedule Marijuana” (the “DPR”), issued August 12, 2016. In fact, the
DEA’s recent statements confirm scientific facts about medical cannabis that have long been
accepted by a majority of the scientific community. Accordingly, Americans for Safe Access
(“ASA”) requests that the DEA correct or remove from the dea.gov website the inaccurate
statements described below in Section II (a)-(d). At minimum, the corrections should comport
with the DEA’s statements in the DPR.
PETITIONER
Americans for Safe Access Foundation (“ASA”), a non-profit advocacy group that represents the
interests of medical cannabis patients and caregivers, files this Request for Correction of
inaccurate information, disseminated by the DEA, relating to certain purported health effects of
cannabis use. ASA brings this action on behalf of patients, their families, medical providers,
scientists, and veterans across the United States who are deeply and immediately affected by the
DEA’s controverted statements. The seriously ill patients that ASA represents suffer variously
from cancer and the side-effects of its treatments, multiple sclerosis, HIV/AIDS, spinal injury,
chronic seizures, and other medical conditions that produce chronic pain, nausea, loss of appetite
and spasticity. Many of these persons who use medical cannabis to treat these symptoms do not
respond to conventional treatment options, cannot tolerate certain medications, or have serious
health needs not treatable by pharmaceutical medicine. If patients, who currently have access to
medical cannabis under state programs, were to lose access, they would be irreparably harmed.
And, patients in need of medical cannabis, but without access, are already being seriously
harmed.
The DEA’s misinformation informs the opinions and actions of Congress. As a result of this
misinformation, there is a substantial risk that Congress will fail to reauthorize the RohrabacherFarr Medical Cannabis Amendment (“the Amendment”) (discussed below)—failure to
reauthorize would encourage the DOJ to dismantle state medical cannabis systems and prosecute
medical cannabis users and providers throughout the nation. Furthermore, the CARERS Act
(discussed below) has yet to receive a vote, due in part to the dissemination of DEA
misinformation. ASA’s members reside in every United States Congressional District—they
have been negatively affected by Congress’ continuing refusal to hold a vote on the CARERS
Act, and they will be negatively affected by Congress’ failure to reauthorize the Amendment.
1
RELIEF REQUESTED
ASA requests corrections to DEA disseminated information as described in Section II (a)-(d).
ASA files this Request for Correction pursuant to the Information Quality Act amendments to
the Paperwork Reduction Act, 44 U.S.C. § 3516 Statutory and Historical Notes, P.L. 106-554
(“Information Quality Act”), as implemented through the Office of Management and Budget’s
“Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility, and Integrity of
Information Disseminated by Federal Agencies,” 67 Fed. Reg. 8452 (Feb. 22, 2002) (“OMB
Guidelines”), and the “DOJ Information Quality Guidelines,”
https://www.justice.gov/iqpr/information-quality (“DOJ Guidelines”).
FACTUAL BACKGROUND
For years, the DEA has published scientifically inaccurate information about the health effects of
medical cannabis, directly influencing the action – and inaction – of Congress. The
Compassionate Access, Research Expansion, and Respect States Act (“CARERS”) is a prime
example. Three senators introduced CARERS in March 2015 and an identical bill was
introduced in the House later that month. The legislation seeks to protect patient access to
medical cannabis in states with existing medical cannabis programs from federal intervention,
thereby codifying the collection of DOJ memoranda that currently govern federal policy of
medical cannabis enforcement against the states.1 Notably, CARERS would also reschedule
cannabis from Schedule I to Schedule II status, thus easing current restrictions on medical and
scientific research of the substance.2 Furthermore, the Act would exclude cannabidiols (cannabis
derivatives with less than 0.3% THC content) from the definition of cannabis entirely,3 permit
businesses acting in conformity with state cannabis laws to access banking services,4 mandate
the issuance of additional licenses to cultivate cannabis for FDA approved research,5 and grant
VA dependent veterans access to state medical cannabis programs.6
Since the CARERS Act was introduced in March of 2015, it has received additional support in
the Senate and House, but it seems unlikely that there will be a formal vote on the bill before the
new administration commences in January 2017. Proponents of the Act believe that it is less
likely to pass once the new Congress is sworn in and the new administration takes control. The
House bill is sitting in four committees and subcommittees; the Senate analog sits in the Senate
Judiciary Committee.7 Committee leadership in both chambers have denied the respective bills a
1
https://www.congress.gov/bill/114th-congress/senate-bill/683/text, at Section 2 (The Controlled Substances Act,
“shall not apply to any person acting in compliance with State law relating to the production, possession,
distribution, dispensation, administration, laboratory testing, or delivery of medical marihuana.”).
2
Id. at Section 3.
3
Id. at Section 4.
4
Id. at Section 6.
5
Id. at Section 7.
6
Id. at Section 8.
7
H.R. 1538 has been assigned to the (1) House Energy and Commerce Subcommittee on Health; (2) House
Judiciary Subcommittee on Crime, Terrorism, Homeland Security, and Investigations; (3) House Financial Services
2
hearing. House leadership has been hostile to medical cannabis legislation with the surreptitious
removal of a medical cannabis amendment to the Military Construction and Veterans Affairs
Appropriations Act in June 2016, after being approved by votes from the Senate Appropriations
Committee and House Floor.8 Changes in the Senate Judiciary Committee for the 115th
Congress include the ascension of CARERS opponent Dianne Feinstein to Ranking Member of
the Senate Judiciary Committee, while fellow CARERS opponent Chuck Grassley remains
committee chair. Representatives and senators that have commented unfavorably on the bills
have cited, implicitly and explicitly, the inaccurate DEA information on the supposed dangers of
medical cannabis.
The CARERS Act is not the only attempt to protect medical cannabis patients. In 2014,
Congress included the Amendment in the Commerce, Justice, and Science Appropriations Bill.9
The Amendment prevents the DOJ from spending federal funds to inhibit the implementation of
state medical cannabis laws. Without the Amendment, the DOJ could restrict or eliminate
patients’ access to medicine legally available to them under their states’ laws. The Amendment
was reauthorized in 2015, and a functionally identical amendment was introduced in April 2016
as part of the 2017 Commerce, Justice, Science, and Related Appropriations Act.10 While the
Amendment was approved by the Senate Appropriations Committee in May 2016 by a vote of
21-8, it has yet to receive a vote in the House for Fiscal Year 2017. Congress’ failure to pass the
CARERS Act or to reauthorize the Amendment, could destroy patients’ access to vital medicine
in states where medical cannabis is currently legal and available. Also, even if patients are not
the direct target of federal enforcement actions, they can be caught in harm’s way during a raid.
And, even if they are not present at the raid, losing access to their dispensary means a disruption
in their supply of medicine that may not be restored through access to another dispensing facility.
As a result, patients are terrified of losing access to essential medicine and providers live in
constant fear of federal criminal prosecution.
Elected representatives in Congress are using inaccurate DEA published information to inform
their votes on the CARERS Act and the Amendment. In the Denial of Petition to Initiate
Proceedings to Reschedule Marijuana11 (“DPR”), the DEA directly contradicted a multitude of
previously disseminated statements, which are currently available on the dea.gov website. The
following sections detail (1) the inaccurate information and requested changes, (2) how the
inaccurate information adversely impacts affected persons (i.e. ASA’s members), and (3) how
the requested changes will benefit affected persons.
Committee; and (4) House Veterans' Affairs Subcommittee on Health; available at
https://www.congress.gov/bill/114th-congress/house-bill/1538/all-actions.
8
See http://www.militarytimes.com/story/veterans/2016/06/28/marijuana-provision-stripped-veterans-affairsfunding-bill/86471448/.
9
10
https://www.congress.gov/bill/113th-congress/house-bill/4660/text, at Section 558.
https://www.congress.gov/bill/114th-congress/senate-bill/2837/text, at Section 537.
11
See https://www.federalregister.gov/documents/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-81.
3
ARGUMENT
I.!
LEGAL STANDARDS
Passed as an amendment to the Paperwork Reduction Act, 44 U.S.C. § 3501, the Information
Quality Act requires administrative agencies to devise guidelines to ensure the “quality,
objectivity, utility, and integrity of information” they disseminate and to “[e]stablish
administrative mechanisms allowing affected persons to seek and obtain correction of
information maintained and disseminated by the agency that does not comply with the
guidelines.”12
The DOJ Guidelines quote the OMB Guidelines, which define “quality” as “an
encompassing term comprising utility, objectivity, and integrity.”13 The term “utility” refers to
the “usefulness of the information to be disseminated to the public,” achieved by “continuously
monitoring information needs and developing new information sources or by revising existing
methods, models, and information products where appropriate.”14 “Objectivity” assures that, as
a “matter of substance and presentation,” disseminated information is “accurate, reliable, and
unbiased.”15 In short, the agency is required, prior to dissemination of information, to ensure
“compliance with the OMB and DOJ Guidelines” and “that the information fulfills the intentions
stated and that the conclusions are consistent with the evidence.”16
Additionally, where the agency is responsible for disseminating “influential” scientific or
statistical information, the DEA has heightened responsibilities under the Act to ensure that such
disseminated information is reproducible and accurate. Indeed, the accuracy of this information
is “significant due to the critical nature of these decisions.”17 “Influential information” is that
which is “expected to have a genuinely clear and substantial impact at the national level, or on
major public and private policy decisions as they relate to federal justice issues.”18 To determine
that there is a clear and substantial impact, the agency must “have greater certainty than would
be the case for many ordinary factual determinations that the impact is occurring or will occur.”19
Furthermore, the DOJ Guidelines require that statistical information disseminated by the
agency be based on the promotion of sound statistical methods. “Sound” scientific methods
“produce information (data and analysis results) that is accurate, reliable, and unbiased.
Guidelines to promote sound statistical methods would cover the planning of statistical data
12
44 U .S.C. § 3516, Statutory and Historical Notes.
13
https://www.justice.gov/iqpr/information-quality, at “Standards for Disseminated Information.”
14
Id. at “Utility.”
15
Id. at “Objectivity.”
16
Supra Note 11.
17
Id. at “For Influential Information.”
18
Id.
19
Id.
4
systems, the collection of statistical data, and the processing of statistical data (including
analysis).”20
II.!
THE DEA’S STATEMENTS ABOUT MEDICAL CANNABIS IN THE DPR
DIRECTLY CONTRADICT STATEMENTS CURRENTLY BEING MADE BY
THE DEA ELSEWHERE
Each of the DEA’s statements about medical cannabis set forth below have been directly
refuted by the DEA’s own statements in the DPR. Given its own recent contradiction of these
statements, the DEA cannot credibly maintain that they are “accurate,” “reliable,” “unbiased,” or
“reproducible.” Moreover, the statements are based on scientifically inaccurate data and result in
denying patients access to vital medicine. Accordingly, each of these statements violate the
IQA’s utility and objectivity standards and should be corrected.
ASA requests that the DEA replace the following scientifically inaccurate statements –
currently disseminated by the DEA on its website in publications entitled “The Dangers and
Consequences of Marijuana Abuse”21 and “Drugs of Abuse”22 – with the DEA’s own
scientifically accurate statements made in the DPR.
a.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements about cannabis’ alleged capacity to induce psychosis
The DEA is disseminating information about cannabis use and psychosis that lacks both
objectivity and utility. At the time the inaccurate statements were originally made, they may
have been supported by some evidence. But, the DEA recently admitted that the only association
between cannabis use and psychotic illness is in cannabis’ potential to increase the risk for
psychosis among individuals already predisposed to develop a psychotic disorder.23 Thus, in
light of numerous statements made by the DEA in the DPR, information suggesting that cannabis
use causes psychosis no longer satisfies the objectivity and utility standards required by the DOJ
and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis’ alleged capacity
to induce psychosis and psychotic illness:
20
Id. at “Sound Statistical Methods.”
21
https://www.dea.gov/docs/dangers-consequences-marijuana-abuse.pdf.
22
https://www.dea.gov/pr/multimedia-library/publications/drug_of_abuse.pdf#page=73.
23
Supra Note 9, at 53696-97 (citing Andreasson et al., Curr Med Chem. 18(7): 1085-99 (2011); Schimmelmann
et al., Schizophr Res 129(1): 52-56 (2011); Schiffman et al., Psychiatry Res.134(1): 37-42 (2005); Pelayo et al.,
Curr Pharm Des 18(32): 5024-35 (2012); Degenhardt et al., Drug and Alcohol Depend 71(1): 37-48 (2003)) (“The
authors concluded that marijuana use increased the risk for psychosis only among individuals predisposed to
develop the disorder […] Additionally, the conclusion that the impact of marijuana may manifest only in individuals
likely to develop psychotic disorders has been shown in many other studies.”) (emphasis added).
5
1.! “According to an Australian study, there is now conclusive evidence that smoking
cannabis hastens the appearance of psychotic illnesses by up to three years […] it
makes it very clear that cannabis is playing a significant role in psychosis.”24
2.! “Evidence of the damage to mental health caused by cannabis use—from loss of
concentration to paranoia, aggressiveness and outright psychosis—is mounting and
cannot be ignored.”25
3.! “Marijuana use can worsen depression and lead to more serious mental illness such as
schizophrenia, anxiety, and even suicide.”26
4.! “[T]eenage cannabis users are more likely to suffer psychotic symptoms and have a
greater risk of developing schizophrenia in later life.”27
5.! “Dr. John MacLeod, a prominent British psychiatrist states: ‘If you assume such a
link (to schizophrenia with cannabis) then the number of cases of schizophrenia will
increase significantly in line with increased use of the drug.’ He predicts that cannabis
use may account for a quarter of all new cases of schizophrenia in three years’
time.”28
6.! “Compared with those who had never used cannabis, young adults who had six or
more years since first use of cannabis were twice as likely to develop a non-affective
psychosis (such as schizophrenia) […] They were also four times as likely to have
high scores in clinical tests of delusion.”29
7.! “Researchers have also found an association between marijuana use and increased
risk of depression, an increased risk and earlier onset of schizophrenia, and other
psychotic disorders, especially for teens that have a genetic predisposition.”30
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
24
Supra Note 21, at 12 (quotations omitted).
25
Id. at 8.
26
Id. at 10.
27
Id.
28
Id. at 12.
29
Id.
30
Supra Note 22, at 73.
6
1.! “At present, the available data do not suggest a causative link between marijuana use
and the development of psychosis.”31
2.! “Numerous large, longitudinal studies show that subjects who used marijuana do not
have a greater incidence of psychotic diagnoses compared to those who do not use
marijuana.”32
3.! “[M]arijuana per se does not appear to induce schizophrenia in the majority of
individuals who have tried or continue to use marijuana. However, in individuals with
a genetic vulnerability for psychosis, marijuana use may influence the development of
psychosis.”33
b.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements about cannabis’ alleged capacity to induce lung cancer
and cause damage comparable to that caused by tobacco use
The DEA is disseminating information about cannabis use and lung cancer that lacks both
objectivity and utility. At the time the inaccurate statements were originally made, they may
have been supported by some evidence. But, the DEA recently admitted that the worst possible
respiratory effects associated with long-term cannabis use are “chronic cough, increased sputum,
as well as increased frequency of chronic bronchitis and pharyngitis.”34 Thus, in light of
numerous statements made by the DEA in the DPR, information suggesting that cannabis use
causes lung cancer and tobacco-like respiratory damage no longer satisfies the objectivity and
utility standards required by the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis’s alleged
capacity to induce lung cancer and cause damage comparable to that caused by tobacco use:
1.! “Marijuana smoking has been implicated as a causative factor in tumors of the head
and neck and of the lung.”35
2.! “Marijuana takes the risks of tobacco and raises them. Marijuana smoke contains
more than 400 chemicals and increases the risk of serious health consequences,
including lung damage.”36
31
Supra Note 11, at 53696.
32
Id.
33
Id. at 53696-97.
34
Id. at 53751 (citing HHS 2015; Adams and Martin, Addiction 91(11): 1585-1614 (1996); Hollister,
Pharmacological Rev 38, 1-20 (1986)).
35
Supra Note 21, at 16.
36
Id.
7
3.! “A study from New Zealand reports that cannabis smoking may cause five percent of
lung cancer cases in that country.”37
4.! “According to researchers at the Tale School of Medicine, long-term exposure to
marijuana smoke is linked to many of the same kinds of health problems as those
experienced by long-term cigarette smokers.”38
5.! “Smoking marijuana can cause changes in lung tissue that may promote cancer
growth, according to a review of decades of research on marijuana smoking and lung
cancer.”39
6.! “Nevertheless, researchers indicate […] that smoking pot could indeed boost lung
cancer risk.”40
7.! “The Foundation warned that smoking one cannabis cigarette increase the chances of
developing lung cancer by as much as an entire packet of 20 cigarettes.”41
8.! “Like tobacco smokers, marijuana smokers experience serious health problems such
as bronchitis, emphysema, and bronchial asthma. Extended use may cause
suppression of the immune system. Because marijuana contains toxins and
carcinogens, marijuana smokers increase their risk of cancer of the head, neck, lungs,
and respiratory tract.”42
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “The DEA further notes the publication of recent review articles critically evaluating
the association between marijuana and lung cancer. Most of the reviews agree that the
association is weak or inconsistent.” 43
2.! “The HHS concluded that new evidence suggests that the effects of smoking
marijuana on respiratory function and cancer are different from the effects of smoking
tobacco.” 44
37
Id. at 14.
38
Id. at 15.
39
Id.
40
Id.
41
Id. at 18.
42
Supra Note 22, at 73.
43
Supra Note 11, at 53751 (internal citation omitted).
44
Id. (internal citation omitted).
8
3.! “[O]verall association is weak between marijuana use and lung cancer especially
when controlling for tobacco use.” 45
4.! “[I]n a large clinical study with 1,650 subjects, no positive correlation was found
between marijuana use and lung cancer. This finding held true regardless of the extent
of marijuana use when both tobacco use and other potential confounding factors were
controlled.”46
5.! “The authors reported that occasional use of marijuana (7 joint-years for lifetime or
1 joint/day for 7 years or 1 joint/week for 49 years) does not adversely affect
pulmonary function.”47
c.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements regarding the “gateway theory” and cannabis
The DEA is disseminating information about cannabis use and the gateway theory that lacks
both objectivity and utility. The “gateway theory” – that cannabis use causes users to abuse
more serious drugs in the future – was never supported by epidemiological scientific evidence.48
And, in light of numerous statements made by the DEA in the DPR, information suggesting that
cannabis is a “gateway drug,” no longer satisfies the objectivity and utility standards required by
the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis and the gateway
theory:
1.! “Legalization of marijuana, no matter how it begins, will come at the expense of our
children and public safety. It will create dependency and treatment issues, and open
the door to use of other drugs, impaired health, delinquent behavior, and drugged
drivers.”49
2.! “Teens who experiment with marijuana may be making themselves more vulnerable
to heroin addiction later in life, if the findings from experiments with rats are any
indication.”50
3.! “Marijuana is a frequent precursor to the use of more dangerous drugs and signals a
significantly enhanced likelihood of drug problems in adult life.”51
45
Id. (internal citation omitted).
46
Id. (internal citation omitted).
47
Id.
48
Id. at 53705.
49
Supra Note 21, at 6.
50
Id. at 22.
51
Id.
9
4.! “[T]eens who used marijuana at least once in the last month are 13 times likelier than
other teens to use another drug like cocaine, heroin, or methamphetamine and almost
26 times likelier than those teens who have never used marijuana to use another
drug.”52
5.! “Marijuana use in early adolescence is particularly ominous. Adults who were early
marijuana users were found to be five times more likely to become dependent on any
drug, eight times more likely to use cocaine in the future, and fifteen times more
likely to use heroin later in life.”53
6.! “Healthcare workers, legal counsel, police and judges indicate that marijuana is a
typical precursor to methamphetamine.”54
7.! “Teens past month heavy marijuana users [sic] are significantly more likely than
teens that have not used marijuana in the past to: use cocaine/crack (30 times more
likely); use Ecstasy (20 times more likely); abuse prescription pain relievers (15 times
more likely); and abuse over the counter medications (14 times more likely).”55
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “Overall, research does not support a direct causal relationship between regular
marijuana use and other illicit drug use.”56
2.! “The HHS cited several studies where marijuana use did not lead to other illicit drug
use. Two separate longitudinal studies with adolescents using marijuana did not
demonstrate an association with use of other illicit drugs.”57
3.! “Little evidence supports the hypothesis that initiation of marijuana use leads to an
abuse disorder with other illicit substances. For example, one longitudinal study of
708 adolescents demonstrated that early onset marijuana use did not lead to
problematic drug use.” 58
4.! “Although many individuals with a drug abuse disorder may have used marijuana as
one of their first illicit drugs, this fact does not correctly lead to the reverse inference
52
Id.
53
Id. at 22-23.
54
Id. at 23.
55
Id.
56
Supra Note 11, at 53705.
57
Id. (internal citations omitted).
58
Id.
10
that most individuals who used marijuana will inherently go on to try or become
regular users of other illicit drugs.”59
5.! “[B]ecause the gateway hypothesis only addresses the order of drug use initiation, the
gateway hypothesis does not specify any mechanistic connection between drug
‘stages’ following exposure to marijuana and does not extend to the risks for
addiction.”60
6.! “Degenhardt et al. (2009) examined the development of drug dependence and found
an association that did not support the gateway hypothesis. Specifically, drug
dependence was significantly associated with the use of other illicit drugs prior to
marijuana use.” 61
d.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements regarding the alleged permanency of
cannabis-associated cognitive deficits
The DEA is disseminating information about the alleged permanency of cannabis-associated
cognitive deficits that lacks both objectivity and utility. At the time the inaccurate statements
were originally made, they may have been supported by some evidence. But, the DEA recently
noted that cannabis associated cognitive deficits are not apparent in those who initiate use after
the age of 15 years.62 Thus, in light of numerous statements made by the DEA in the DPR,
information suggesting that cannabis use causes permanent cognitive deficits no longer satisfies
the objectivity and utility standards required by the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding the alleged permanency of
cannabis-associated cognitive deficits:
1.! “Those who started using marijuana regularly after age 18 showed minor [cognitive]
declines.”63
2.! “Memory, speed of thinking, and other cognitive abilities get worse over time with
marijuana use.”64
59
Id.
60
Id.
61
Id.
62
Id. at 53695 (citing Fontes, et al., Br. J Psychiatry 198(6): 442-7 (2011)) (“Individuals with a diagnosis of
marijuana misuse or dependence who were seeking treatment for substance use, who initiated marijuana use before
the age of 15 years, showed deficits in performance on tasks assessing sustained attention, impulse control, and
general executive functioning compared to non-using controls. These deficits were not seen in individuals who
initiated marijuana use after the age of 15 years.”) (emphasis added).
63
Supra Note 21, at 8.
64
Id. at 11.
11
3.! “This study is the first to show that long-term cannabis use can adversely affect all
users, not just those in the high-risk categories such as the young, or those susceptible
to mental illness, as previously thought.”65
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “[T]he adult-onset chronic marijuana users showed no significant changes in IQ
compared to pre-exposure levels whether they were current users or abstinent for at
least 1 year.” 66
2.! “[C]annabis-associated cognitive deficits are reversible and related to recent cannabis
exposure, rather than irreversible and related to cumulative lifetime use.” 67
3.! “The effects of chronic marijuana use do not seem to persist after more than 1 to 3
months of abstinence. After 3 months of abstinence, any deficits observed in IQ,
immediate memory, delayed memory, and information processing speeds following
heavy marijuana use compared to pre-drug use scores were no longer apparent.”68
4.! “Similarly, following abstinence for a year or more, both light and heavy adult
marijuana users did not show deficits on score of verbal memory compared to nonusing controls.”69
5.! “According to a recent meta-analysis looking at non-acute and long-lasting effect of
marijuana use on neurocognitive performance, any deficits seen within the first month
following abstinence are generally not present after about 1 month of abstinence.”70
III.!
THE INACCURATE DEA INFORMATION LACKS BOTH OBJECTIVITY
AND UTILITY MAKING IT THE PROPER SUBJECT OF A REQUEST FOR
CORRECTION UNDER THE IQA
The overwhelming majority of the objective scientific studies – including studies cited by the
DEA in the DPR71 – disprove the inaccurate DEA statements described in Section II (a)-(d).
65
Id.
66
Supra Note 11, at 53695.
67
Id.
68
Id. (internal citation omitted).
69
Id.
70
Id.
71
Minozzi et al., Drug Alcohol Rev 29(3): 304-317 (2010); Fergusson et al., Addiction 100(3): 354-366 (2005);
Kuepper et al., Psychol Med 41(10): 2121-2129 (2011); Van Os et al., Am J Epidemiol 156(4): 319-327 (2002);
American Medical Association, AMA Policy: Medical Marijuana H-95-952 (2009); Degenhardt et al., Drug Alcohol
Depend 71(1): 37-48 (2003); Department of Health and Human Services, Basis for the recommendation for
maintaining marijuana in Schedule I of the Controlled Substances Act (2015); Huang et al., Cancer Epidemiol
12
Because the DEA itself made statements in the DPR that directly contradict information in “The
Dangers and Consequences of Marijuana Abuse” and “Drugs of Abuse,” it is undeniable that the
DEA information at issue lacks utility and objectivity.72
The DEA information lacks utility. Utility requires that information disseminated by the
DEA be useful to the public. Information that is admittedly incorrect – such as the DEA’s
statements regarding the gateway hypothesis and that marijuana causes psychosis, lung cancer
and permanent cognitive deficits – inherently lacks usefulness. While there may be some
demonstrable negative effects associated with cannabis abuse, the presentation of scientifically
unfounded information alongside scientifically accurate information obscures and diminishes the
utility of the accurate information and can jeopardize public health. Furthermore, the
disingenuous presentation of the inaccurate information described above makes it difficult for
public officials and medical providers to make informed decisions regarding the viability of
medical cannabis treatment options.
Utility also requires continuous monitoring of information and the correction and updating of
information where appropriate. The statements made by the DEA in the DPR described above,
as well as the studies cited by the DEA, demonstrate that the DEAs statements on its website
regarding the gateway theory, psychosis, lung cancer and permanent cognitive deficits need to be
corrected and updated.
The DEA information lacks objectivity. The information described in Section II (a)-(d) is
neither accurate, reliable, nor unbiased, as evidenced by the DEA’s contradictory statements in
the DPR. For example, as demonstrated above, the DEA makes numerous inaccurate, unreliable
and biased statements regarding the gateway theory and the health risks of marijuana use,
including that it causes psychosis, lung cancer and permanent cognitive deficits. The DEA itself
has disproven each of these statements in the DPR as described above. The contradictory
statements made in “The Dangers and Consequences of Marijuana Abuse” and in “Drugs of
Abuse,” evince a strong bias against medical cannabis and represent a dereliction of
responsibility. The documents cite outdated and unreliable studies, and fail to discuss contrary
authorities or the documented benefits of medical cannabis.
Biomarkers Prev 24(1): 15-31 (2015); Zhang et al., Int J Cancer 136(4): 894-903 (2015); Gates et al., Respirology
19(5): 655-662; Hall and Degenhardt, Drug Test Anal 6(1-2):39-45; Tashkin et al., American Thoracic Society
International Conference A777 (2006); Lee and Hancox Exp Rev Resp Med 5(4): 537-546 (2011); Kandel and Chen
J Stud Alcohol 61(3): 367-378 (2000); von Sydow et al., Drug Alcohol Depend 68(1): 49-64 (2002); Nace et al.,
Arch Gen Psychiatry 32(1): 77-80 (1975); Degenhardt et al., Alcohol Depend 108(1-2): 84-97 (2010); Vanyukov
et al., DrugAlcohol Depend 123 Suppl 1:S3-17 (2012); Degenhardt et al., PLoS Medicine 6(9): e1000133 (2009);
Meier et al., Proc.Natl.Acad.Sci U.S.A 109(40): E2657-E2664 (2012); Fried et al., Neuotoxicol Teratol 27(2):
231-239 (2005).
72
See https://www.justice.gov/iqpr/information-quality (“Utility: DOJ components will assess the usefulness of the
information to be disseminated to the public. Utility is achieved by continuously monitoring information needs and
developing new information sources or by revising existing methods, models, and information products where
appropriate. Objectivity: DOJ components will ensure disseminated information, as a matter of substance and
presentation, is accurate, reliable, and unbiased. Objectivity is achieved by using reliable data sources, sound
analytical techniques, and documenting methods and data sources.”).
13
Moreover, as discussed in the next section, the DEA has a heightened burden of ensuring the
accuracy of its statements regarding the risk of marijuana use because the information is highly
influential and affects national public policy. The DEAs failure to update and correct admittedly
outdated and incorrect information does not meet this heightened burden. Moreover, because of
the need for greater certainty for influential information, the results of any studies and
information relied on by the DEA must be reproducible. The DPR demonstrates that the studies
and information relied on by the DEA for each of the categories discussed above is not
reproducible.
Because the inaccurate information is neither useful nor objective, it must be changed to
more accurately reflect the current scientific consensus surrounding medical cannabis. At the
very least, the DEA should update its public information to comport with the statements it made
in the DPR—namely, that (1) the gateway drug hypothesis is invalid; (2) cannabis use does not
cause irreversible cognitive decline in adults; and cannabis use does not cause (3) psychosis or
(4) lung cancer.
IV.!
THE INACCURATE DEA STATEMENTS REQUIRE A HIGHER LEVEL OF
SCRUTINY BECAUSE THEY ARE “INFLUENTIAL INFORMATION”
AFFECTING NATIONAL PUBLIC POLICY
The DOJ Guidelines require an “added level of scrutiny” for information deemed
“influential.”73 The responsibility for determining whether information is influential lies with
the component of the DOJ responsible for disseminating the information.74 Here, because the
relevant DOJ component (the DEA) has not designated medical cannabis information as a
“class” of information that is “influential,” the DEA must determine whether information is
influential on a case-by-case basis.75 As stated above, the Guidelines define “influential”
information as that which has a “genuinely clear and substantial impact at the national level, or
on major public and private policy decisions as they relate to federal justice issues.”76 The DEA
should find that the inaccurate information described in Section II has a “clear and substantial
impact” if it is firmly convinced that the information has a high probability of impacting public
or private “policy, economic, or other decisions.”77
The incorrect information on medical cannabis published by the DEA clearly meets this
standard. The DEA is one of the most respected and influential federal agencies providing
information on drug use, drug abuse, and the health risks surrounding drug use. Unsurprisingly,
many elected officials rely on DEA information in making policy decisions and in educating
their colleagues regarding the risks and rewards of medical cannabis. In fact, members of the
House of Representatives have repeatedly cited to “The Dangers and Consequences of Marijuana
Abuse,” which is the primary subject of this request for correction. As such, the maintenance of
the inaccurate DEA information described in Section II has a genuinely clear and substantial
73
Supra Note 13, at “For Influential Information.”
74
Id.
75
Id.
76
Id.
77
Id.
14
impact at the national level and on important public policy decisions related to federal justice
issues.
Indeed, the “high probability” of impact has already materialized – via Congress’ continuing
failure to pass the 2015 CARERS Act– and is likely to continue occurring given the incoming
administration’s stance on medical cannabis. Recent statements made on the floor of the House
of Representatives indicate that elected officials are being directly influenced to vote against the
interests of medical cannabis patients as a result of the DEA’s inaccurate statements. During a
May 28, 2014 House discussion regarding the “Commerce, Justice, Science and Related
Agencies Appropriation Act of 2015,” Representatives John Fleming (R-LA) and Frank Wolf
(R-VA)78 directly cited to the DEA’s document “The Dangers and Consequences of cannabis
Abuse,” to support inaccurate propositions regarding the gateway theory and cannabis’ health
effects:
“I would like to close by reading the following statement from the Drug Enforcement
Agency's DEA May 2014 booklet on the ugly truth about marijuana: ‘Legalization of
marijuana, no matter how it begins, will come at the expense of our children and public
safety. It will create dependency and treatment issues and opens the door to use of other
drugs, impaired health, delinquent behavior, and drugged drivers.’ I think the DEA got it
right. It is time for the rest of the Justice Department to do their job and enforce current
U.S. law that recognizes marijuana's devastating impact on our children and society. I am
hopeful that my amendment will encourage DOJ to take steps necessary to correct any
misunderstanding regarding the Federal enforcement of the CSA and the BSA. I now
urge my colleagues to join me in supporting this amendment.” 79
…
“[M]arijuana is highly addictive, is closely linked to altered brain development;
schizophrenia; mental illness […]”80
…
“I was just reading the dangers and consequences of marijuana abuse. What is happening
to our country? […] I strongly support the amendment.” 81
78
Frank Wolf retired in January 2015.
79
https://www.congress.gov/congressional-record/2014/5/28/house-section/article/h48681?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=4, at H4907.
80
Id.
81
Id.
15
…
“And trust me, my friend, I will tell the gentleman that whether it is marijuana or heroin
or methamphetamines, as a drug addict once told me: All addicting substances are
gateways to other addicting substances.” 82
These opinions were directly influenced by the inaccurate statements in the “Dangers and
Consequences of Marijuana Abuse,” discussed in Section II above. 83 The Congressmen were
speaking in support of Rep. Fleming’s proposed amendment to H.R. 4660, which would have
reduced the DOJ’s general legal account by $866,000 until the Attorney General enforced the
Controlled Substances Act (“CSA”) by prosecuting medical cannabis providers and patients
operating under State laws.84 Because outspoken and active members of the House use the
aforementioned DEA statements in support of federal criminal justice legislation, the subject
information is highly influential and can be expected to have a genuinely clear and substantial
impact at the national level on important public policy decisions related to federal justice issues.
While this particular amendment did not pass, Congress could pass a similar amendment or
simply refuse to reauthorize the Rohrabacher-Farr Medical Cannabis Amendment85—an
amendment that prohibits the DOJ from using funds under the Act to interfere with providers and
patients acting in accordance with state medical cannabis laws. This injury could occur as soon
as December 2016 when Congress passes 2017 appropriations acts. It is highly likely that
Congress will (1) refuse to reauthorize the Amendment; and/or (2) refuse to pass the CARERS
Act.
Similar statements made by other US representatives demonstrate the pervasiveness of
inaccurate beliefs regarding medical cannabis that are being perpetuated by DEA
misinformation.
In a July 2016 Hearing, the House Subcommittee on Crime and Terrorism discussed
researching the potential medical benefits and risks of cannabis. Representative Lindsey
Graham, the Chairman of the subcommittee, made statements about the refuted gateway drug
theory:
“I also hear about how marijuana is a gateway drug that gets people going down the
wrong road.” 86
82
Id.
83
See generally supra Note 21.
84
Supra Note 79, at H4906.
85
https://www.congress.gov/bill/113th-congress/house-bill/4660/text, at Section 558.
86
http://www.judiciary.senate.gov/meetings/researching-the-potential-medical-benefits-and-risks-of-marijuana, at
30:40.
16
…
“I have also been a prosecutor and I understand that this has been a gateway drug.” 87
While these statements do not explicitly reference DEA documents, they mirror DEA
misinformation and strongly suggest that Sen. Graham believes that the gateway theory
surrounding cannabis remains scientifically accurate. As a former prosecutor, it is likely that
Sen. Graham was influenced by inaccurate DEA information in forming his opinions about the
gateway theory. Yet, as a CARERS Act cosponsor, Sen. Graham believed he was presenting a
balanced view regarding the potential benefits and harms of medical cannabis. This hearing took
place approximately one month prior to the DEA’s August 2016 acknowledgement that the
gateway theory is not supported by science. Had Sen. Graham been aware of the invalidity of
the gateway theory, it is likely that he would have presented more nuanced and fact-based
evaluation of the risks and benefits associated with medical cannabis and the CARERS Act.
Additionally, Sen. Graham has a major influence on public policy and on other
representatives (especially republicans). And, while he seems willing to consider the medical
potential of cannabis and cannabis derivatives, his willingness to support (1) research using
federal funds, (2) institutional access to cannabis for research, or (3) medicinal access for
patients in need is stymied by his belief in the gateway theory. Declining to allow or fund
medical research at a national level certainly qualifies as a major public policy decision. As
such, Rep. Graham’s statements suggest that inaccurate DEA information about the gateway
theory has a genuinely clear and substantial impact at the national level on important public
policy decisions.
In a June 24, 2015 Senate Drug Caucus Hearing on Barriers to Cannabidiol Research,
Senator Dianne Feinstein (D-CA) stated:
“It concerns me greatly because young people use it … it is also a gateway drug … they
go onto other things … and it’s problematic.” 88
Sen. Feinstein is the Co-Chair of the Senate Drug Caucus, and she is under the impression
that cannabis is a gateway drug that leads users to abuse more serious drugs. Again, while the
Senator did not directly reference DEA materials, it is likely that the DEA’s dissemination of
inaccurate information regarding cannabis and the gateway theory contributed to her incorrect
views. And, it is highly likely that she would reconsider her beliefs about the gateway theory if
she were exposed to correct information from a nationally trusted source like the DEA. As the
Co-Chair on the Senate Drug Caucus, Sen. Feinstein is in a unique position to influence federal
drug policy and national research efforts; thus, her statements suggest that inaccurate DEA
information about the gateway theory has a genuinely clear and substantial impact at the national
level on important public policy decisions related to federal justice issues.
Senator Chuck Grassley’s (R-IA) views further demonstrate the “high probability” of impact
posed by DEA misinformation. For example, Sen. Grassley’s spokeswoman noted specific
87
Id. at 01:05:21.
88
http://www.drugcaucus.senate.gov/content/drug-caucus-hearing-barriers-cannabidiol-research-0, at 02:00:51.
17
reasons that Sen. Grassley did not support the CARERS Act, stating that he believes “marijuana
users [are] much more likely to take up heroin and other serious drugs than non-users.”89 The
impact of Sen. Grassley’s belief in the gateway theory is particularly acute – as the Chairman of
the Senate Judiciary Committee, Sen. Grassley is the proverbial gatekeeper to any Senate hearing
on the CARERS Act. And, given his general support for research into cannabidiol medicines,90
Sen. Grassley’s belief in the gateway theory is likely a primary impediment preventing him from
facilitating a vote on the CARERS Act.
At the April 5, 2016 Drug Caucus hearing, Senator Jeff Sessions (R-AL) made several
references to the gateway theory without specifically mentioning the theory by name. In a
conversation with hearing witness Benjamin B. Wagner, U.S. Attorney for the Eastern District of
California, Sen. Sessions asserted that “good people do not smoke marijuana” and described the
damage that could ensue if more people use cannabis:
“You can see that it is in fact a very real danger, you can see the accidents traffic deaths
related to marijuana jumped by 20%. These are the kind of things we’re going to see
throughout the country and you’ll see cocaine and heroin increase more than it would
have I think had we not talked about it […]”91
…
“Lives will be impacted, families will be broken up, children will be damaged because of
the difficulties their parents have, and people may be psychologically impacted the rest of
their lives with marijuana. And if they go on to more serious drugs which tends to
happen, and you can deny it if you want to, but it tends to happen […]”92
As the probable incoming attorney general, Sen. Sessions will dictate whether the DOJ does
or does not interfere with state medical cannabis systems. He clearly harbors a strong hatred for
cannabis generally; nevertheless, his erroneous views on the gateway theory and the alleged
permanency of cannabis associated cognitive deficits are likely informed by DEA
misinformation, as Sen. Sessions has displayed a sense of trust in the opinions of “the Drug Czar
and the DEA leadership.”93 Notably, Sen. Sessions’ comments were made approximately four
months before the DEA formally acknowledged that the gateway theory is not supported by
science. Because Sen. Sessions – the apparent incoming attorney general – likely draws his
opinions about the gateway theory from DEA misinformation, the maintenance of such
89
http://beeherald.com/news/local-farmer-taking-grassley-over-medical-marijuana.
90
http://www.grassley.senate.gov/news/news-releases/bill-introduced-expand-research-potential-medical-benefitscannabidiol-and.
91
https://www.youtube.com/watch?v=gg0bZvIS0K8&feature=youtu.be&t=38m47s, at 39:48.
92
Id. at 42:13.
93
Id. at 42:35.
18
inaccurate information has a genuinely clear and substantial impact at the national level on
important public policy decisions related to federal justice issues.
During a May 29, 2014 House discussion regarding the “Commerce, Justice, Science and
Related Agencies Appropriation Act of 2015,” Representative Andy Harris (R-MD) stated:
“This is dangerous for [children]. How do we know this? The health risks: brain
development, schizophrenia, increased risk of stroke.”94
As part of the House Committee on Appropriations, Representative Harris is charged with
allocating dollars to federal agencies. As such, he has power to influence DOJ enforcement of
federal cannabis laws by withholding DOJ funds. 95 Rep. Harris believes that cannabis causes
schizophrenia, an admittedly false fact96 currently being promulgated by DEA literature.
Moreover, Rep. Harris believes in the gateway theory, as demonstrated by his statements at a
National Rx Drug Abuse Summit on April 8, 2015:
“That's not the way we should deal with such a dangerous drug […] marijuana is pretty
clearly a gateway drug that has not been shown to be safe or medically effective.”97
Because of his belief in the psychosis and gateway theories, Rep. Harris opposed the
Amendment.98 Rep. Harris’ statements suggest that currently accessible DEA information
continues to promote the unfounded psychosis and gateway theories, thus creating a genuinely
clear and substantial impact at the national level on important public policy decisions related to
federal justice issues.
During a June 2, 2015 House discussion regarding the “Commerce, Justice, Science and
Related Agencies Appropriation Act of 2016,” Representative John Fleming (R-LA) stated:
“It [marijuana] is known to have brain development alterations; schizophrenia and other
forms of mental illness, psychosis; heart complications; and an increased risk of
stroke.”99
94
https://www.congress.gov/congressional-record/2014/5/29/house-section/article/h49682?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=3, at H4983.
95
See e.g., supra Note 79, at H4906.
96
See supra Note 11, at 53696.
97
http://halrogers.house.gov/news/documentsingle.aspx?DocumentID=398203.
98
“I rise to oppose the amendment.” Supra Note 94.
99
https://www.congress.gov/congressional-record/2015/6/2/house-section/article/h37002?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=2, at H3746.
19
…
“It means the younger a child is exposed to it, the more likely that child will later become
an addict to something else, like methamphetamine, prescription drugs, heroin.” 100
As the Co-Chair of the Addiction, Treatment, and Recovery Caucus, Rep. Fleming is charged
with raising awareness and increasing education regarding substance abuse and addiction
treatment. As such, he is in a unique position to educate other members of Congress and the
public about the dangers and benefits of medical cannabis. As illustrated by his statements in the
May 28, 2014 and June 2, 2015 House discussions, 101 he is directly influenced by inaccurate
DEA information and promulgates this shoddy information in support of strict anti-medical
cannabis laws and stronger enforcement of the CSA amongst the states. It is clear that inaccurate
DEA information regarding the gateway theory and cannabis’ alleged ability to cause psychosis
has a genuinely clear and substantial impact at the national level on important public policy
decisions related to federal justice issues.
Representative Frank Wolf (R-VA) opposed the Amendment in a floor speech on May 9,
2012 discussing the Commerce, Justice, Science, and Related Agencies Appropriations Act of
2013.102 Representative Jerrold Nadler (D-NY) pointed out why this was the case:
“I heard [Rep. Wolf] say that the DEA says there is no medical use for marijuana. That’s
true that they’ve said it. The DEA has no credibility with people who have looked at
[medical cannabis] . . . We know that, for people suffering pain, for people suffering
nausea from AIDS and cancer, marijuana is the only thing that produces relief and
enables them to eat and get sustenance and to regain weight and to, perhaps, regain
health. . . . The DEA doesn’t know [this] because it refuses to see it and refuses to allow
systematic research.” 103
Rep. Wolf’s opposition to the Amendment is directly influenced by DEA misinformation, as
he has directly cited104 to the DEA’s faulty document: “The Dangers and Consequences of
Marijuana Abuse.” The statement above lends further credence to the fact that DEA
misinformation has a genuinely clear and substantial impact at the national level on important
public policy decisions related to federal justice issues.
Due to the widespread acceptance of inaccurate DEA information amongst the United States
Congress, the information at issue has a genuinely clear and substantial influential impact on
federal public policy decisions. This is especially true when considering DEA statements which
100
Id. at H3747.
101
See Supra Notes 79-80 & 99-100.
102
https://www.congress.gov/congressional-record/2012/5/9/house-section/article/h25153?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=1, at H2525.
103
Id. at H2526.
104
Supra Note 81.
20
perpetuate the false notions that cannabis use causes psychosis and acts as a gateway drug to
more serious drug abuse. Affected persons (i.e. ASA members) have already been affected by
Congress’ continuing refusal to hold a vote on the CARERS Act, and they will be further
affected if the Amendment is not reauthorized. Because the information at issue is “influential
information” within the meaning of the Guidelines, the DEA should review the inaccurate DEA
information with an added level of scrutiny, to ensure that it is reproducible.
V.!
ASA REPRESENTS SERIOUSLY ILL “AFFECTED PERSONS” WHO ARE
DEEPLY AND IMMEDIATELY AFFECTED BY THE DEA’S INCORRECT
AND CONTROVERTED STATEMENTS
a.! ASA’s members are “affected persons” within the meaning of the DOJ’s
Information Quality Guidelines
According to the DOJ and OMB Guidelines, affected persons are allowed to “seek and
obtain, where appropriate, timely correction of information maintained and disseminated by the
agency that does not comply with OMB or agency guidelines.”105 And, an “affected person” is
an “individual or entity that may use, benefit, or be harmed by the disseminated information at
issue.”106 ASA is composed of the following affected persons: (1) patients who are unable to
access medical cannabis or are at risk of losing access; (2) doctors who are unable to recommend
medical cannabis or are at risk of losing their ability to recommend it; (3) patients and providers
who have been criminally prosecuted or are at risk of prosecution; and (4) scientists who are
unable to obtain cannabis for research or are at risk of losing access.107 On behalf of these
affected persons, ASA seeks to obtain correction of DEA information that fails to comply with
the Guidelines. ASA and its individual members are currently being harmed by – and are at risk
of future harm from – the DEA’s dissemination of inaccurate information regarding medical
cannabis. Specifically, the DEA’s aforementioned statements regarding the gateway theory,
cannabis’ supposed tendency to induce psychosis and lung cancer, and the alleged permanency
of cannabis associated cognitive deficits have harmed and continue to harm ASA and its
members. The harm results because the inaccurate information obfuscates legitimate medical
cannabis research, which would otherwise inform our elected official’s opinions and actions.
As described in Section III, elected officials across the nation rely on DEA information when
forming opinions about the safety and efficacy of medical cannabis. These officials have made
public policy decisions based, at least in part, on inaccurate DEA information. These policy
decisions include failing to reschedule cannabis via passage of the CARERS Act, which has the
effect of denying patients access to medical cannabis, preventing doctors from prescribing
medical cannabis, and criminally prosecuting medical cannabis users/providers. And, while
there are many states that have implemented their own medical cannabis systems, medical
cannabis remains federally illegal, in part due to elected officials’ inaccurate perceptions that
105
Supra Note 13, at “Introduction and Purpose.”
106
Id. at “Process for Citizen Complaint.”
107
ASA has members residing in every United States Congressional District.
21
cannabis is a gateway drug and that it causes psychosis, lung cancer, and permanent cognitive
deficits. The federal status of medical cannabis has prevented multiple states from allowing
healthcare providers to recommend medical cannabis in those states. Furthermore, there is a
substantial risk that a misinformed Congress will either repeal or refuse to reauthorize the
Amendment, thereby urging the DOJ to enforce the CSA in states with legal medical cannabis
systems.
The inaccurate perceptions of at least several outspoken United States Congressmen originate
from DEA information lacking both objectivity and utility. These representatives often push for
stricter enforcement of the CSA in the states and maintenance of cannabis as a Schedule I drug.
A correction of the erroneous DEA information would benefit ASA, its members, and millions of
medical cannabis patients by shifting US representatives’ perceptions of the true risks of medical
cannabis. Such a shift could result in many benefits, including but not limited to: (1) patients’
continued access to medical cannabis in states that currently permit its use;108 (2) patients’ access
to medical cannabis in states which currently prohibit its use;109 (3) elimination of criminal
penalties for medical cannabis physicians and patients;110 and (4) more federal funding and
access to cannabis for medical research. 111
108
There were approximately 2,045,888 registered medical cannabis patients as of Dec. 2015, based on available
patient registry statistics compiled by ASA. Available at https://american-safeaccess.s3.amazonaws.com/documents/EstimatedNumberOfMMJPatientsDec2015.pdf.
109
There are currently 6 states with no medical cannabis and an additional 15 states with limited CBD-focused laws.
Only one of the CBD-focused laws allows for patients to obtain the medical cannabis-derived products from a
dispensary in the state, all other CBD-focused laws only protect patients from arrest if they obtain and possess
products acquired from a state with licensed distribution and reciprocity access.
110
According to the FBI, there were 643,121 cannabis arrests in 2015, over 89% of which were for possession alone
– this is the crime patients are most likely to violate. However, the FBI does not provide any information on how
many of those arrests involved a defendant claiming medical necessity. While medical cannabis physicians are
rarely targeted for arrest, the chilling effect of its Schedule I status creates stigma that suppresses the number of
physicians who are willing to recommend medical cannabis under state law. Available at https://ucr.fbi.gov/crimein-the-u.s/2015/crime-in-the-u.s.-2015/home.
111
Researchers have commented on the lack of federal funding available for medical cannabis research. University
of Pennsylvania professor Marcel Bonn-Miller said, “[f]rom the National Institutes of Health to the VA to whatever,
there was nothing,” referring to the available funding for medical cannabis research. Ethan Russo, Former GW
Pharmaceuticals researcher and current medical director at the Los Angeles biotechnology firm Phytecs, elaborated
on the problem facing medical cannabis researchers: “Traditionally, if you had a compelling reason to do research,
you could get funding … Now nothing is getting funded unless you have something really sexy. And marijuana is
like kryptonite.” Between 1999 and 2012, the number of studies approved for funding dropped from 34% to 19%.
Available at http://www.ibtimes.com/marijuana-news-2016-scientists-frustrated-funding-shortfalls-launch-institute2379921.
22
23
BEFORE THE UNITED STATES DEPARTMENT OF JUSTICE
INFORMATION QUALITY GUIDELINES STAFF
_____________________________________________
)
Re: DEA’s “The Dangers and Consequences )
of Marijuana Abuse” and “Drugs of Abuse”
)
)
______________________________________ )
REQUEST FOR CORRECTION OF INFORMATION DISSEMINATED
BY DEA REGARDING MARIJUANA (CANNABIS)
INFORMATION QUALITY ACT REQUEST FOR CORRECTION
DATE: DECEMBER 5, 2016
SUBMITTED BY: AMERICANS FOR SAFE ACCESS FOUNDATION
Attorneys for Petitioner
Executive Director for Petitioner
Vickie L. Feeman
[email protected]
Rick Fukushima
[email protected]
Alex Fields
[email protected]
Orrick, Herrington & Sutcliffe LLP
1000 Marsh Road
Menlo Park, CA 94025
Phone: (650) 614-7400
Steph Sherer
1624 U Street, NW
Suite 200
Washington, D.C. 20009
Phone: (202) 857-4272
Fax: (202) 618-6977
[email protected]
Americans for Safe Access
Request for Correction Pursuant to the DOJ’s Information Quality Guidelines
ISSUE
The Drug Enforcement Agency’s (“DEA”) website (dea.gov) contains inaccurate statements that
do not meet the standards of quality required by the Department of Justice (“DOJ”) and Office of
Management and Budget (“OMB”) under the Information Quality Act (“IQA”). In particular,
the DEA continues to disseminate certain statements about the health risks of medical cannabis
use that have been incontrovertibly refuted by the DEA itself in its recent “Denial of Petition to
Initiate Proceedings to Reschedule Marijuana” (the “DPR”), issued August 12, 2016. In fact, the
DEA’s recent statements confirm scientific facts about medical cannabis that have long been
accepted by a majority of the scientific community. Accordingly, Americans for Safe Access
(“ASA”) requests that the DEA correct or remove from the dea.gov website the inaccurate
statements described below in Section II (a)-(d). At minimum, the corrections should comport
with the DEA’s statements in the DPR.
PETITIONER
Americans for Safe Access Foundation (“ASA”), a non-profit advocacy group that represents the
interests of medical cannabis patients and caregivers, files this Request for Correction of
inaccurate information, disseminated by the DEA, relating to certain purported health effects of
cannabis use. ASA brings this action on behalf of patients, their families, medical providers,
scientists, and veterans across the United States who are deeply and immediately affected by the
DEA’s controverted statements. The seriously ill patients that ASA represents suffer variously
from cancer and the side-effects of its treatments, multiple sclerosis, HIV/AIDS, spinal injury,
chronic seizures, and other medical conditions that produce chronic pain, nausea, loss of appetite
and spasticity. Many of these persons who use medical cannabis to treat these symptoms do not
respond to conventional treatment options, cannot tolerate certain medications, or have serious
health needs not treatable by pharmaceutical medicine. If patients, who currently have access to
medical cannabis under state programs, were to lose access, they would be irreparably harmed.
And, patients in need of medical cannabis, but without access, are already being seriously
harmed.
The DEA’s misinformation informs the opinions and actions of Congress. As a result of this
misinformation, there is a substantial risk that Congress will fail to reauthorize the RohrabacherFarr Medical Cannabis Amendment (“the Amendment”) (discussed below)—failure to
reauthorize would encourage the DOJ to dismantle state medical cannabis systems and prosecute
medical cannabis users and providers throughout the nation. Furthermore, the CARERS Act
(discussed below) has yet to receive a vote, due in part to the dissemination of DEA
misinformation. ASA’s members reside in every United States Congressional District—they
have been negatively affected by Congress’ continuing refusal to hold a vote on the CARERS
Act, and they will be negatively affected by Congress’ failure to reauthorize the Amendment.
1
RELIEF REQUESTED
ASA requests corrections to DEA disseminated information as described in Section II (a)-(d).
ASA files this Request for Correction pursuant to the Information Quality Act amendments to
the Paperwork Reduction Act, 44 U.S.C. § 3516 Statutory and Historical Notes, P.L. 106-554
(“Information Quality Act”), as implemented through the Office of Management and Budget’s
“Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility, and Integrity of
Information Disseminated by Federal Agencies,” 67 Fed. Reg. 8452 (Feb. 22, 2002) (“OMB
Guidelines”), and the “DOJ Information Quality Guidelines,”
https://www.justice.gov/iqpr/information-quality (“DOJ Guidelines”).
FACTUAL BACKGROUND
For years, the DEA has published scientifically inaccurate information about the health effects of
medical cannabis, directly influencing the action – and inaction – of Congress. The
Compassionate Access, Research Expansion, and Respect States Act (“CARERS”) is a prime
example. Three senators introduced CARERS in March 2015 and an identical bill was
introduced in the House later that month. The legislation seeks to protect patient access to
medical cannabis in states with existing medical cannabis programs from federal intervention,
thereby codifying the collection of DOJ memoranda that currently govern federal policy of
medical cannabis enforcement against the states.1 Notably, CARERS would also reschedule
cannabis from Schedule I to Schedule II status, thus easing current restrictions on medical and
scientific research of the substance.2 Furthermore, the Act would exclude cannabidiols (cannabis
derivatives with less than 0.3% THC content) from the definition of cannabis entirely,3 permit
businesses acting in conformity with state cannabis laws to access banking services,4 mandate
the issuance of additional licenses to cultivate cannabis for FDA approved research,5 and grant
VA dependent veterans access to state medical cannabis programs.6
Since the CARERS Act was introduced in March of 2015, it has received additional support in
the Senate and House, but it seems unlikely that there will be a formal vote on the bill before the
new administration commences in January 2017. Proponents of the Act believe that it is less
likely to pass once the new Congress is sworn in and the new administration takes control. The
House bill is sitting in four committees and subcommittees; the Senate analog sits in the Senate
Judiciary Committee.7 Committee leadership in both chambers have denied the respective bills a
1
https://www.congress.gov/bill/114th-congress/senate-bill/683/text, at Section 2 (The Controlled Substances Act,
“shall not apply to any person acting in compliance with State law relating to the production, possession,
distribution, dispensation, administration, laboratory testing, or delivery of medical marihuana.”).
2
Id. at Section 3.
3
Id. at Section 4.
4
Id. at Section 6.
5
Id. at Section 7.
6
Id. at Section 8.
7
H.R. 1538 has been assigned to the (1) House Energy and Commerce Subcommittee on Health; (2) House
Judiciary Subcommittee on Crime, Terrorism, Homeland Security, and Investigations; (3) House Financial Services
2
hearing. House leadership has been hostile to medical cannabis legislation with the surreptitious
removal of a medical cannabis amendment to the Military Construction and Veterans Affairs
Appropriations Act in June 2016, after being approved by votes from the Senate Appropriations
Committee and House Floor.8 Changes in the Senate Judiciary Committee for the 115th
Congress include the ascension of CARERS opponent Dianne Feinstein to Ranking Member of
the Senate Judiciary Committee, while fellow CARERS opponent Chuck Grassley remains
committee chair. Representatives and senators that have commented unfavorably on the bills
have cited, implicitly and explicitly, the inaccurate DEA information on the supposed dangers of
medical cannabis.
The CARERS Act is not the only attempt to protect medical cannabis patients. In 2014,
Congress included the Amendment in the Commerce, Justice, and Science Appropriations Bill.9
The Amendment prevents the DOJ from spending federal funds to inhibit the implementation of
state medical cannabis laws. Without the Amendment, the DOJ could restrict or eliminate
patients’ access to medicine legally available to them under their states’ laws. The Amendment
was reauthorized in 2015, and a functionally identical amendment was introduced in April 2016
as part of the 2017 Commerce, Justice, Science, and Related Appropriations Act.10 While the
Amendment was approved by the Senate Appropriations Committee in May 2016 by a vote of
21-8, it has yet to receive a vote in the House for Fiscal Year 2017. Congress’ failure to pass the
CARERS Act or to reauthorize the Amendment, could destroy patients’ access to vital medicine
in states where medical cannabis is currently legal and available. Also, even if patients are not
the direct target of federal enforcement actions, they can be caught in harm’s way during a raid.
And, even if they are not present at the raid, losing access to their dispensary means a disruption
in their supply of medicine that may not be restored through access to another dispensing facility.
As a result, patients are terrified of losing access to essential medicine and providers live in
constant fear of federal criminal prosecution.
Elected representatives in Congress are using inaccurate DEA published information to inform
their votes on the CARERS Act and the Amendment. In the Denial of Petition to Initiate
Proceedings to Reschedule Marijuana11 (“DPR”), the DEA directly contradicted a multitude of
previously disseminated statements, which are currently available on the dea.gov website. The
following sections detail (1) the inaccurate information and requested changes, (2) how the
inaccurate information adversely impacts affected persons (i.e. ASA’s members), and (3) how
the requested changes will benefit affected persons.
Committee; and (4) House Veterans' Affairs Subcommittee on Health; available at
https://www.congress.gov/bill/114th-congress/house-bill/1538/all-actions.
8
See http://www.militarytimes.com/story/veterans/2016/06/28/marijuana-provision-stripped-veterans-affairsfunding-bill/86471448/.
9
10
https://www.congress.gov/bill/113th-congress/house-bill/4660/text, at Section 558.
https://www.congress.gov/bill/114th-congress/senate-bill/2837/text, at Section 537.
11
See https://www.federalregister.gov/documents/2016/08/12/2016-17954/denial-of-petition-to-initiateproceedings-to-reschedule-marijuana#p-81.
3
ARGUMENT
I.!
LEGAL STANDARDS
Passed as an amendment to the Paperwork Reduction Act, 44 U.S.C. § 3501, the Information
Quality Act requires administrative agencies to devise guidelines to ensure the “quality,
objectivity, utility, and integrity of information” they disseminate and to “[e]stablish
administrative mechanisms allowing affected persons to seek and obtain correction of
information maintained and disseminated by the agency that does not comply with the
guidelines.”12
The DOJ Guidelines quote the OMB Guidelines, which define “quality” as “an
encompassing term comprising utility, objectivity, and integrity.”13 The term “utility” refers to
the “usefulness of the information to be disseminated to the public,” achieved by “continuously
monitoring information needs and developing new information sources or by revising existing
methods, models, and information products where appropriate.”14 “Objectivity” assures that, as
a “matter of substance and presentation,” disseminated information is “accurate, reliable, and
unbiased.”15 In short, the agency is required, prior to dissemination of information, to ensure
“compliance with the OMB and DOJ Guidelines” and “that the information fulfills the intentions
stated and that the conclusions are consistent with the evidence.”16
Additionally, where the agency is responsible for disseminating “influential” scientific or
statistical information, the DEA has heightened responsibilities under the Act to ensure that such
disseminated information is reproducible and accurate. Indeed, the accuracy of this information
is “significant due to the critical nature of these decisions.”17 “Influential information” is that
which is “expected to have a genuinely clear and substantial impact at the national level, or on
major public and private policy decisions as they relate to federal justice issues.”18 To determine
that there is a clear and substantial impact, the agency must “have greater certainty than would
be the case for many ordinary factual determinations that the impact is occurring or will occur.”19
Furthermore, the DOJ Guidelines require that statistical information disseminated by the
agency be based on the promotion of sound statistical methods. “Sound” scientific methods
“produce information (data and analysis results) that is accurate, reliable, and unbiased.
Guidelines to promote sound statistical methods would cover the planning of statistical data
12
44 U .S.C. § 3516, Statutory and Historical Notes.
13
https://www.justice.gov/iqpr/information-quality, at “Standards for Disseminated Information.”
14
Id. at “Utility.”
15
Id. at “Objectivity.”
16
Supra Note 11.
17
Id. at “For Influential Information.”
18
Id.
19
Id.
4
systems, the collection of statistical data, and the processing of statistical data (including
analysis).”20
II.!
THE DEA’S STATEMENTS ABOUT MEDICAL CANNABIS IN THE DPR
DIRECTLY CONTRADICT STATEMENTS CURRENTLY BEING MADE BY
THE DEA ELSEWHERE
Each of the DEA’s statements about medical cannabis set forth below have been directly
refuted by the DEA’s own statements in the DPR. Given its own recent contradiction of these
statements, the DEA cannot credibly maintain that they are “accurate,” “reliable,” “unbiased,” or
“reproducible.” Moreover, the statements are based on scientifically inaccurate data and result in
denying patients access to vital medicine. Accordingly, each of these statements violate the
IQA’s utility and objectivity standards and should be corrected.
ASA requests that the DEA replace the following scientifically inaccurate statements –
currently disseminated by the DEA on its website in publications entitled “The Dangers and
Consequences of Marijuana Abuse”21 and “Drugs of Abuse”22 – with the DEA’s own
scientifically accurate statements made in the DPR.
a.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements about cannabis’ alleged capacity to induce psychosis
The DEA is disseminating information about cannabis use and psychosis that lacks both
objectivity and utility. At the time the inaccurate statements were originally made, they may
have been supported by some evidence. But, the DEA recently admitted that the only association
between cannabis use and psychotic illness is in cannabis’ potential to increase the risk for
psychosis among individuals already predisposed to develop a psychotic disorder.23 Thus, in
light of numerous statements made by the DEA in the DPR, information suggesting that cannabis
use causes psychosis no longer satisfies the objectivity and utility standards required by the DOJ
and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis’ alleged capacity
to induce psychosis and psychotic illness:
20
Id. at “Sound Statistical Methods.”
21
https://www.dea.gov/docs/dangers-consequences-marijuana-abuse.pdf.
22
https://www.dea.gov/pr/multimedia-library/publications/drug_of_abuse.pdf#page=73.
23
Supra Note 9, at 53696-97 (citing Andreasson et al., Curr Med Chem. 18(7): 1085-99 (2011); Schimmelmann
et al., Schizophr Res 129(1): 52-56 (2011); Schiffman et al., Psychiatry Res.134(1): 37-42 (2005); Pelayo et al.,
Curr Pharm Des 18(32): 5024-35 (2012); Degenhardt et al., Drug and Alcohol Depend 71(1): 37-48 (2003)) (“The
authors concluded that marijuana use increased the risk for psychosis only among individuals predisposed to
develop the disorder […] Additionally, the conclusion that the impact of marijuana may manifest only in individuals
likely to develop psychotic disorders has been shown in many other studies.”) (emphasis added).
5
1.! “According to an Australian study, there is now conclusive evidence that smoking
cannabis hastens the appearance of psychotic illnesses by up to three years […] it
makes it very clear that cannabis is playing a significant role in psychosis.”24
2.! “Evidence of the damage to mental health caused by cannabis use—from loss of
concentration to paranoia, aggressiveness and outright psychosis—is mounting and
cannot be ignored.”25
3.! “Marijuana use can worsen depression and lead to more serious mental illness such as
schizophrenia, anxiety, and even suicide.”26
4.! “[T]eenage cannabis users are more likely to suffer psychotic symptoms and have a
greater risk of developing schizophrenia in later life.”27
5.! “Dr. John MacLeod, a prominent British psychiatrist states: ‘If you assume such a
link (to schizophrenia with cannabis) then the number of cases of schizophrenia will
increase significantly in line with increased use of the drug.’ He predicts that cannabis
use may account for a quarter of all new cases of schizophrenia in three years’
time.”28
6.! “Compared with those who had never used cannabis, young adults who had six or
more years since first use of cannabis were twice as likely to develop a non-affective
psychosis (such as schizophrenia) […] They were also four times as likely to have
high scores in clinical tests of delusion.”29
7.! “Researchers have also found an association between marijuana use and increased
risk of depression, an increased risk and earlier onset of schizophrenia, and other
psychotic disorders, especially for teens that have a genetic predisposition.”30
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
24
Supra Note 21, at 12 (quotations omitted).
25
Id. at 8.
26
Id. at 10.
27
Id.
28
Id. at 12.
29
Id.
30
Supra Note 22, at 73.
6
1.! “At present, the available data do not suggest a causative link between marijuana use
and the development of psychosis.”31
2.! “Numerous large, longitudinal studies show that subjects who used marijuana do not
have a greater incidence of psychotic diagnoses compared to those who do not use
marijuana.”32
3.! “[M]arijuana per se does not appear to induce schizophrenia in the majority of
individuals who have tried or continue to use marijuana. However, in individuals with
a genetic vulnerability for psychosis, marijuana use may influence the development of
psychosis.”33
b.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements about cannabis’ alleged capacity to induce lung cancer
and cause damage comparable to that caused by tobacco use
The DEA is disseminating information about cannabis use and lung cancer that lacks both
objectivity and utility. At the time the inaccurate statements were originally made, they may
have been supported by some evidence. But, the DEA recently admitted that the worst possible
respiratory effects associated with long-term cannabis use are “chronic cough, increased sputum,
as well as increased frequency of chronic bronchitis and pharyngitis.”34 Thus, in light of
numerous statements made by the DEA in the DPR, information suggesting that cannabis use
causes lung cancer and tobacco-like respiratory damage no longer satisfies the objectivity and
utility standards required by the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis’s alleged
capacity to induce lung cancer and cause damage comparable to that caused by tobacco use:
1.! “Marijuana smoking has been implicated as a causative factor in tumors of the head
and neck and of the lung.”35
2.! “Marijuana takes the risks of tobacco and raises them. Marijuana smoke contains
more than 400 chemicals and increases the risk of serious health consequences,
including lung damage.”36
31
Supra Note 11, at 53696.
32
Id.
33
Id. at 53696-97.
34
Id. at 53751 (citing HHS 2015; Adams and Martin, Addiction 91(11): 1585-1614 (1996); Hollister,
Pharmacological Rev 38, 1-20 (1986)).
35
Supra Note 21, at 16.
36
Id.
7
3.! “A study from New Zealand reports that cannabis smoking may cause five percent of
lung cancer cases in that country.”37
4.! “According to researchers at the Tale School of Medicine, long-term exposure to
marijuana smoke is linked to many of the same kinds of health problems as those
experienced by long-term cigarette smokers.”38
5.! “Smoking marijuana can cause changes in lung tissue that may promote cancer
growth, according to a review of decades of research on marijuana smoking and lung
cancer.”39
6.! “Nevertheless, researchers indicate […] that smoking pot could indeed boost lung
cancer risk.”40
7.! “The Foundation warned that smoking one cannabis cigarette increase the chances of
developing lung cancer by as much as an entire packet of 20 cigarettes.”41
8.! “Like tobacco smokers, marijuana smokers experience serious health problems such
as bronchitis, emphysema, and bronchial asthma. Extended use may cause
suppression of the immune system. Because marijuana contains toxins and
carcinogens, marijuana smokers increase their risk of cancer of the head, neck, lungs,
and respiratory tract.”42
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “The DEA further notes the publication of recent review articles critically evaluating
the association between marijuana and lung cancer. Most of the reviews agree that the
association is weak or inconsistent.” 43
2.! “The HHS concluded that new evidence suggests that the effects of smoking
marijuana on respiratory function and cancer are different from the effects of smoking
tobacco.” 44
37
Id. at 14.
38
Id. at 15.
39
Id.
40
Id.
41
Id. at 18.
42
Supra Note 22, at 73.
43
Supra Note 11, at 53751 (internal citation omitted).
44
Id. (internal citation omitted).
8
3.! “[O]verall association is weak between marijuana use and lung cancer especially
when controlling for tobacco use.” 45
4.! “[I]n a large clinical study with 1,650 subjects, no positive correlation was found
between marijuana use and lung cancer. This finding held true regardless of the extent
of marijuana use when both tobacco use and other potential confounding factors were
controlled.”46
5.! “The authors reported that occasional use of marijuana (7 joint-years for lifetime or
1 joint/day for 7 years or 1 joint/week for 49 years) does not adversely affect
pulmonary function.”47
c.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements regarding the “gateway theory” and cannabis
The DEA is disseminating information about cannabis use and the gateway theory that lacks
both objectivity and utility. The “gateway theory” – that cannabis use causes users to abuse
more serious drugs in the future – was never supported by epidemiological scientific evidence.48
And, in light of numerous statements made by the DEA in the DPR, information suggesting that
cannabis is a “gateway drug,” no longer satisfies the objectivity and utility standards required by
the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding cannabis and the gateway
theory:
1.! “Legalization of marijuana, no matter how it begins, will come at the expense of our
children and public safety. It will create dependency and treatment issues, and open
the door to use of other drugs, impaired health, delinquent behavior, and drugged
drivers.”49
2.! “Teens who experiment with marijuana may be making themselves more vulnerable
to heroin addiction later in life, if the findings from experiments with rats are any
indication.”50
3.! “Marijuana is a frequent precursor to the use of more dangerous drugs and signals a
significantly enhanced likelihood of drug problems in adult life.”51
45
Id. (internal citation omitted).
46
Id. (internal citation omitted).
47
Id.
48
Id. at 53705.
49
Supra Note 21, at 6.
50
Id. at 22.
51
Id.
9
4.! “[T]eens who used marijuana at least once in the last month are 13 times likelier than
other teens to use another drug like cocaine, heroin, or methamphetamine and almost
26 times likelier than those teens who have never used marijuana to use another
drug.”52
5.! “Marijuana use in early adolescence is particularly ominous. Adults who were early
marijuana users were found to be five times more likely to become dependent on any
drug, eight times more likely to use cocaine in the future, and fifteen times more
likely to use heroin later in life.”53
6.! “Healthcare workers, legal counsel, police and judges indicate that marijuana is a
typical precursor to methamphetamine.”54
7.! “Teens past month heavy marijuana users [sic] are significantly more likely than
teens that have not used marijuana in the past to: use cocaine/crack (30 times more
likely); use Ecstasy (20 times more likely); abuse prescription pain relievers (15 times
more likely); and abuse over the counter medications (14 times more likely).”55
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “Overall, research does not support a direct causal relationship between regular
marijuana use and other illicit drug use.”56
2.! “The HHS cited several studies where marijuana use did not lead to other illicit drug
use. Two separate longitudinal studies with adolescents using marijuana did not
demonstrate an association with use of other illicit drugs.”57
3.! “Little evidence supports the hypothesis that initiation of marijuana use leads to an
abuse disorder with other illicit substances. For example, one longitudinal study of
708 adolescents demonstrated that early onset marijuana use did not lead to
problematic drug use.” 58
4.! “Although many individuals with a drug abuse disorder may have used marijuana as
one of their first illicit drugs, this fact does not correctly lead to the reverse inference
52
Id.
53
Id. at 22-23.
54
Id. at 23.
55
Id.
56
Supra Note 11, at 53705.
57
Id. (internal citations omitted).
58
Id.
10
that most individuals who used marijuana will inherently go on to try or become
regular users of other illicit drugs.”59
5.! “[B]ecause the gateway hypothesis only addresses the order of drug use initiation, the
gateway hypothesis does not specify any mechanistic connection between drug
‘stages’ following exposure to marijuana and does not extend to the risks for
addiction.”60
6.! “Degenhardt et al. (2009) examined the development of drug dependence and found
an association that did not support the gateway hypothesis. Specifically, drug
dependence was significantly associated with the use of other illicit drugs prior to
marijuana use.” 61
d.! The DEA’s statements in the DPR directly contradict its scientifically
inaccurate statements regarding the alleged permanency of
cannabis-associated cognitive deficits
The DEA is disseminating information about the alleged permanency of cannabis-associated
cognitive deficits that lacks both objectivity and utility. At the time the inaccurate statements
were originally made, they may have been supported by some evidence. But, the DEA recently
noted that cannabis associated cognitive deficits are not apparent in those who initiate use after
the age of 15 years.62 Thus, in light of numerous statements made by the DEA in the DPR,
information suggesting that cannabis use causes permanent cognitive deficits no longer satisfies
the objectivity and utility standards required by the DOJ and OMB Guidelines.
The DEA is making the following inaccurate statements regarding the alleged permanency of
cannabis-associated cognitive deficits:
1.! “Those who started using marijuana regularly after age 18 showed minor [cognitive]
declines.”63
2.! “Memory, speed of thinking, and other cognitive abilities get worse over time with
marijuana use.”64
59
Id.
60
Id.
61
Id.
62
Id. at 53695 (citing Fontes, et al., Br. J Psychiatry 198(6): 442-7 (2011)) (“Individuals with a diagnosis of
marijuana misuse or dependence who were seeking treatment for substance use, who initiated marijuana use before
the age of 15 years, showed deficits in performance on tasks assessing sustained attention, impulse control, and
general executive functioning compared to non-using controls. These deficits were not seen in individuals who
initiated marijuana use after the age of 15 years.”) (emphasis added).
63
Supra Note 21, at 8.
64
Id. at 11.
11
3.! “This study is the first to show that long-term cannabis use can adversely affect all
users, not just those in the high-risk categories such as the young, or those susceptible
to mental illness, as previously thought.”65
The following statements, taken directly from the DPR, contradict the aforementioned
statements. Thus, in order to maintain the objectivity and utility standards, ASA requests that the
DEA replace the aforementioned inaccurate statements with the following accurate statements,
or in the alternative, delete the inaccurate statements in their entirety:
1.! “[T]he adult-onset chronic marijuana users showed no significant changes in IQ
compared to pre-exposure levels whether they were current users or abstinent for at
least 1 year.” 66
2.! “[C]annabis-associated cognitive deficits are reversible and related to recent cannabis
exposure, rather than irreversible and related to cumulative lifetime use.” 67
3.! “The effects of chronic marijuana use do not seem to persist after more than 1 to 3
months of abstinence. After 3 months of abstinence, any deficits observed in IQ,
immediate memory, delayed memory, and information processing speeds following
heavy marijuana use compared to pre-drug use scores were no longer apparent.”68
4.! “Similarly, following abstinence for a year or more, both light and heavy adult
marijuana users did not show deficits on score of verbal memory compared to nonusing controls.”69
5.! “According to a recent meta-analysis looking at non-acute and long-lasting effect of
marijuana use on neurocognitive performance, any deficits seen within the first month
following abstinence are generally not present after about 1 month of abstinence.”70
III.!
THE INACCURATE DEA INFORMATION LACKS BOTH OBJECTIVITY
AND UTILITY MAKING IT THE PROPER SUBJECT OF A REQUEST FOR
CORRECTION UNDER THE IQA
The overwhelming majority of the objective scientific studies – including studies cited by the
DEA in the DPR71 – disprove the inaccurate DEA statements described in Section II (a)-(d).
65
Id.
66
Supra Note 11, at 53695.
67
Id.
68
Id. (internal citation omitted).
69
Id.
70
Id.
71
Minozzi et al., Drug Alcohol Rev 29(3): 304-317 (2010); Fergusson et al., Addiction 100(3): 354-366 (2005);
Kuepper et al., Psychol Med 41(10): 2121-2129 (2011); Van Os et al., Am J Epidemiol 156(4): 319-327 (2002);
American Medical Association, AMA Policy: Medical Marijuana H-95-952 (2009); Degenhardt et al., Drug Alcohol
Depend 71(1): 37-48 (2003); Department of Health and Human Services, Basis for the recommendation for
maintaining marijuana in Schedule I of the Controlled Substances Act (2015); Huang et al., Cancer Epidemiol
12
Because the DEA itself made statements in the DPR that directly contradict information in “The
Dangers and Consequences of Marijuana Abuse” and “Drugs of Abuse,” it is undeniable that the
DEA information at issue lacks utility and objectivity.72
The DEA information lacks utility. Utility requires that information disseminated by the
DEA be useful to the public. Information that is admittedly incorrect – such as the DEA’s
statements regarding the gateway hypothesis and that marijuana causes psychosis, lung cancer
and permanent cognitive deficits – inherently lacks usefulness. While there may be some
demonstrable negative effects associated with cannabis abuse, the presentation of scientifically
unfounded information alongside scientifically accurate information obscures and diminishes the
utility of the accurate information and can jeopardize public health. Furthermore, the
disingenuous presentation of the inaccurate information described above makes it difficult for
public officials and medical providers to make informed decisions regarding the viability of
medical cannabis treatment options.
Utility also requires continuous monitoring of information and the correction and updating of
information where appropriate. The statements made by the DEA in the DPR described above,
as well as the studies cited by the DEA, demonstrate that the DEAs statements on its website
regarding the gateway theory, psychosis, lung cancer and permanent cognitive deficits need to be
corrected and updated.
The DEA information lacks objectivity. The information described in Section II (a)-(d) is
neither accurate, reliable, nor unbiased, as evidenced by the DEA’s contradictory statements in
the DPR. For example, as demonstrated above, the DEA makes numerous inaccurate, unreliable
and biased statements regarding the gateway theory and the health risks of marijuana use,
including that it causes psychosis, lung cancer and permanent cognitive deficits. The DEA itself
has disproven each of these statements in the DPR as described above. The contradictory
statements made in “The Dangers and Consequences of Marijuana Abuse” and in “Drugs of
Abuse,” evince a strong bias against medical cannabis and represent a dereliction of
responsibility. The documents cite outdated and unreliable studies, and fail to discuss contrary
authorities or the documented benefits of medical cannabis.
Biomarkers Prev 24(1): 15-31 (2015); Zhang et al., Int J Cancer 136(4): 894-903 (2015); Gates et al., Respirology
19(5): 655-662; Hall and Degenhardt, Drug Test Anal 6(1-2):39-45; Tashkin et al., American Thoracic Society
International Conference A777 (2006); Lee and Hancox Exp Rev Resp Med 5(4): 537-546 (2011); Kandel and Chen
J Stud Alcohol 61(3): 367-378 (2000); von Sydow et al., Drug Alcohol Depend 68(1): 49-64 (2002); Nace et al.,
Arch Gen Psychiatry 32(1): 77-80 (1975); Degenhardt et al., Alcohol Depend 108(1-2): 84-97 (2010); Vanyukov
et al., DrugAlcohol Depend 123 Suppl 1:S3-17 (2012); Degenhardt et al., PLoS Medicine 6(9): e1000133 (2009);
Meier et al., Proc.Natl.Acad.Sci U.S.A 109(40): E2657-E2664 (2012); Fried et al., Neuotoxicol Teratol 27(2):
231-239 (2005).
72
See https://www.justice.gov/iqpr/information-quality (“Utility: DOJ components will assess the usefulness of the
information to be disseminated to the public. Utility is achieved by continuously monitoring information needs and
developing new information sources or by revising existing methods, models, and information products where
appropriate. Objectivity: DOJ components will ensure disseminated information, as a matter of substance and
presentation, is accurate, reliable, and unbiased. Objectivity is achieved by using reliable data sources, sound
analytical techniques, and documenting methods and data sources.”).
13
Moreover, as discussed in the next section, the DEA has a heightened burden of ensuring the
accuracy of its statements regarding the risk of marijuana use because the information is highly
influential and affects national public policy. The DEAs failure to update and correct admittedly
outdated and incorrect information does not meet this heightened burden. Moreover, because of
the need for greater certainty for influential information, the results of any studies and
information relied on by the DEA must be reproducible. The DPR demonstrates that the studies
and information relied on by the DEA for each of the categories discussed above is not
reproducible.
Because the inaccurate information is neither useful nor objective, it must be changed to
more accurately reflect the current scientific consensus surrounding medical cannabis. At the
very least, the DEA should update its public information to comport with the statements it made
in the DPR—namely, that (1) the gateway drug hypothesis is invalid; (2) cannabis use does not
cause irreversible cognitive decline in adults; and cannabis use does not cause (3) psychosis or
(4) lung cancer.
IV.!
THE INACCURATE DEA STATEMENTS REQUIRE A HIGHER LEVEL OF
SCRUTINY BECAUSE THEY ARE “INFLUENTIAL INFORMATION”
AFFECTING NATIONAL PUBLIC POLICY
The DOJ Guidelines require an “added level of scrutiny” for information deemed
“influential.”73 The responsibility for determining whether information is influential lies with
the component of the DOJ responsible for disseminating the information.74 Here, because the
relevant DOJ component (the DEA) has not designated medical cannabis information as a
“class” of information that is “influential,” the DEA must determine whether information is
influential on a case-by-case basis.75 As stated above, the Guidelines define “influential”
information as that which has a “genuinely clear and substantial impact at the national level, or
on major public and private policy decisions as they relate to federal justice issues.”76 The DEA
should find that the inaccurate information described in Section II has a “clear and substantial
impact” if it is firmly convinced that the information has a high probability of impacting public
or private “policy, economic, or other decisions.”77
The incorrect information on medical cannabis published by the DEA clearly meets this
standard. The DEA is one of the most respected and influential federal agencies providing
information on drug use, drug abuse, and the health risks surrounding drug use. Unsurprisingly,
many elected officials rely on DEA information in making policy decisions and in educating
their colleagues regarding the risks and rewards of medical cannabis. In fact, members of the
House of Representatives have repeatedly cited to “The Dangers and Consequences of Marijuana
Abuse,” which is the primary subject of this request for correction. As such, the maintenance of
the inaccurate DEA information described in Section II has a genuinely clear and substantial
73
Supra Note 13, at “For Influential Information.”
74
Id.
75
Id.
76
Id.
77
Id.
14
impact at the national level and on important public policy decisions related to federal justice
issues.
Indeed, the “high probability” of impact has already materialized – via Congress’ continuing
failure to pass the 2015 CARERS Act– and is likely to continue occurring given the incoming
administration’s stance on medical cannabis. Recent statements made on the floor of the House
of Representatives indicate that elected officials are being directly influenced to vote against the
interests of medical cannabis patients as a result of the DEA’s inaccurate statements. During a
May 28, 2014 House discussion regarding the “Commerce, Justice, Science and Related
Agencies Appropriation Act of 2015,” Representatives John Fleming (R-LA) and Frank Wolf
(R-VA)78 directly cited to the DEA’s document “The Dangers and Consequences of cannabis
Abuse,” to support inaccurate propositions regarding the gateway theory and cannabis’ health
effects:
“I would like to close by reading the following statement from the Drug Enforcement
Agency's DEA May 2014 booklet on the ugly truth about marijuana: ‘Legalization of
marijuana, no matter how it begins, will come at the expense of our children and public
safety. It will create dependency and treatment issues and opens the door to use of other
drugs, impaired health, delinquent behavior, and drugged drivers.’ I think the DEA got it
right. It is time for the rest of the Justice Department to do their job and enforce current
U.S. law that recognizes marijuana's devastating impact on our children and society. I am
hopeful that my amendment will encourage DOJ to take steps necessary to correct any
misunderstanding regarding the Federal enforcement of the CSA and the BSA. I now
urge my colleagues to join me in supporting this amendment.” 79
…
“[M]arijuana is highly addictive, is closely linked to altered brain development;
schizophrenia; mental illness […]”80
…
“I was just reading the dangers and consequences of marijuana abuse. What is happening
to our country? […] I strongly support the amendment.” 81
78
Frank Wolf retired in January 2015.
79
https://www.congress.gov/congressional-record/2014/5/28/house-section/article/h48681?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=4, at H4907.
80
Id.
81
Id.
15
…
“And trust me, my friend, I will tell the gentleman that whether it is marijuana or heroin
or methamphetamines, as a drug addict once told me: All addicting substances are
gateways to other addicting substances.” 82
These opinions were directly influenced by the inaccurate statements in the “Dangers and
Consequences of Marijuana Abuse,” discussed in Section II above. 83 The Congressmen were
speaking in support of Rep. Fleming’s proposed amendment to H.R. 4660, which would have
reduced the DOJ’s general legal account by $866,000 until the Attorney General enforced the
Controlled Substances Act (“CSA”) by prosecuting medical cannabis providers and patients
operating under State laws.84 Because outspoken and active members of the House use the
aforementioned DEA statements in support of federal criminal justice legislation, the subject
information is highly influential and can be expected to have a genuinely clear and substantial
impact at the national level on important public policy decisions related to federal justice issues.
While this particular amendment did not pass, Congress could pass a similar amendment or
simply refuse to reauthorize the Rohrabacher-Farr Medical Cannabis Amendment85—an
amendment that prohibits the DOJ from using funds under the Act to interfere with providers and
patients acting in accordance with state medical cannabis laws. This injury could occur as soon
as December 2016 when Congress passes 2017 appropriations acts. It is highly likely that
Congress will (1) refuse to reauthorize the Amendment; and/or (2) refuse to pass the CARERS
Act.
Similar statements made by other US representatives demonstrate the pervasiveness of
inaccurate beliefs regarding medical cannabis that are being perpetuated by DEA
misinformation.
In a July 2016 Hearing, the House Subcommittee on Crime and Terrorism discussed
researching the potential medical benefits and risks of cannabis. Representative Lindsey
Graham, the Chairman of the subcommittee, made statements about the refuted gateway drug
theory:
“I also hear about how marijuana is a gateway drug that gets people going down the
wrong road.” 86
82
Id.
83
See generally supra Note 21.
84
Supra Note 79, at H4906.
85
https://www.congress.gov/bill/113th-congress/house-bill/4660/text, at Section 558.
86
http://www.judiciary.senate.gov/meetings/researching-the-potential-medical-benefits-and-risks-of-marijuana, at
30:40.
16
…
“I have also been a prosecutor and I understand that this has been a gateway drug.” 87
While these statements do not explicitly reference DEA documents, they mirror DEA
misinformation and strongly suggest that Sen. Graham believes that the gateway theory
surrounding cannabis remains scientifically accurate. As a former prosecutor, it is likely that
Sen. Graham was influenced by inaccurate DEA information in forming his opinions about the
gateway theory. Yet, as a CARERS Act cosponsor, Sen. Graham believed he was presenting a
balanced view regarding the potential benefits and harms of medical cannabis. This hearing took
place approximately one month prior to the DEA’s August 2016 acknowledgement that the
gateway theory is not supported by science. Had Sen. Graham been aware of the invalidity of
the gateway theory, it is likely that he would have presented more nuanced and fact-based
evaluation of the risks and benefits associated with medical cannabis and the CARERS Act.
Additionally, Sen. Graham has a major influence on public policy and on other
representatives (especially republicans). And, while he seems willing to consider the medical
potential of cannabis and cannabis derivatives, his willingness to support (1) research using
federal funds, (2) institutional access to cannabis for research, or (3) medicinal access for
patients in need is stymied by his belief in the gateway theory. Declining to allow or fund
medical research at a national level certainly qualifies as a major public policy decision. As
such, Rep. Graham’s statements suggest that inaccurate DEA information about the gateway
theory has a genuinely clear and substantial impact at the national level on important public
policy decisions.
In a June 24, 2015 Senate Drug Caucus Hearing on Barriers to Cannabidiol Research,
Senator Dianne Feinstein (D-CA) stated:
“It concerns me greatly because young people use it … it is also a gateway drug … they
go onto other things … and it’s problematic.” 88
Sen. Feinstein is the Co-Chair of the Senate Drug Caucus, and she is under the impression
that cannabis is a gateway drug that leads users to abuse more serious drugs. Again, while the
Senator did not directly reference DEA materials, it is likely that the DEA’s dissemination of
inaccurate information regarding cannabis and the gateway theory contributed to her incorrect
views. And, it is highly likely that she would reconsider her beliefs about the gateway theory if
she were exposed to correct information from a nationally trusted source like the DEA. As the
Co-Chair on the Senate Drug Caucus, Sen. Feinstein is in a unique position to influence federal
drug policy and national research efforts; thus, her statements suggest that inaccurate DEA
information about the gateway theory has a genuinely clear and substantial impact at the national
level on important public policy decisions related to federal justice issues.
Senator Chuck Grassley’s (R-IA) views further demonstrate the “high probability” of impact
posed by DEA misinformation. For example, Sen. Grassley’s spokeswoman noted specific
87
Id. at 01:05:21.
88
http://www.drugcaucus.senate.gov/content/drug-caucus-hearing-barriers-cannabidiol-research-0, at 02:00:51.
17
reasons that Sen. Grassley did not support the CARERS Act, stating that he believes “marijuana
users [are] much more likely to take up heroin and other serious drugs than non-users.”89 The
impact of Sen. Grassley’s belief in the gateway theory is particularly acute – as the Chairman of
the Senate Judiciary Committee, Sen. Grassley is the proverbial gatekeeper to any Senate hearing
on the CARERS Act. And, given his general support for research into cannabidiol medicines,90
Sen. Grassley’s belief in the gateway theory is likely a primary impediment preventing him from
facilitating a vote on the CARERS Act.
At the April 5, 2016 Drug Caucus hearing, Senator Jeff Sessions (R-AL) made several
references to the gateway theory without specifically mentioning the theory by name. In a
conversation with hearing witness Benjamin B. Wagner, U.S. Attorney for the Eastern District of
California, Sen. Sessions asserted that “good people do not smoke marijuana” and described the
damage that could ensue if more people use cannabis:
“You can see that it is in fact a very real danger, you can see the accidents traffic deaths
related to marijuana jumped by 20%. These are the kind of things we’re going to see
throughout the country and you’ll see cocaine and heroin increase more than it would
have I think had we not talked about it […]”91
…
“Lives will be impacted, families will be broken up, children will be damaged because of
the difficulties their parents have, and people may be psychologically impacted the rest of
their lives with marijuana. And if they go on to more serious drugs which tends to
happen, and you can deny it if you want to, but it tends to happen […]”92
As the probable incoming attorney general, Sen. Sessions will dictate whether the DOJ does
or does not interfere with state medical cannabis systems. He clearly harbors a strong hatred for
cannabis generally; nevertheless, his erroneous views on the gateway theory and the alleged
permanency of cannabis associated cognitive deficits are likely informed by DEA
misinformation, as Sen. Sessions has displayed a sense of trust in the opinions of “the Drug Czar
and the DEA leadership.”93 Notably, Sen. Sessions’ comments were made approximately four
months before the DEA formally acknowledged that the gateway theory is not supported by
science. Because Sen. Sessions – the apparent incoming attorney general – likely draws his
opinions about the gateway theory from DEA misinformation, the maintenance of such
89
http://beeherald.com/news/local-farmer-taking-grassley-over-medical-marijuana.
90
http://www.grassley.senate.gov/news/news-releases/bill-introduced-expand-research-potential-medical-benefitscannabidiol-and.
91
https://www.youtube.com/watch?v=gg0bZvIS0K8&feature=youtu.be&t=38m47s, at 39:48.
92
Id. at 42:13.
93
Id. at 42:35.
18
inaccurate information has a genuinely clear and substantial impact at the national level on
important public policy decisions related to federal justice issues.
During a May 29, 2014 House discussion regarding the “Commerce, Justice, Science and
Related Agencies Appropriation Act of 2015,” Representative Andy Harris (R-MD) stated:
“This is dangerous for [children]. How do we know this? The health risks: brain
development, schizophrenia, increased risk of stroke.”94
As part of the House Committee on Appropriations, Representative Harris is charged with
allocating dollars to federal agencies. As such, he has power to influence DOJ enforcement of
federal cannabis laws by withholding DOJ funds. 95 Rep. Harris believes that cannabis causes
schizophrenia, an admittedly false fact96 currently being promulgated by DEA literature.
Moreover, Rep. Harris believes in the gateway theory, as demonstrated by his statements at a
National Rx Drug Abuse Summit on April 8, 2015:
“That's not the way we should deal with such a dangerous drug […] marijuana is pretty
clearly a gateway drug that has not been shown to be safe or medically effective.”97
Because of his belief in the psychosis and gateway theories, Rep. Harris opposed the
Amendment.98 Rep. Harris’ statements suggest that currently accessible DEA information
continues to promote the unfounded psychosis and gateway theories, thus creating a genuinely
clear and substantial impact at the national level on important public policy decisions related to
federal justice issues.
During a June 2, 2015 House discussion regarding the “Commerce, Justice, Science and
Related Agencies Appropriation Act of 2016,” Representative John Fleming (R-LA) stated:
“It [marijuana] is known to have brain development alterations; schizophrenia and other
forms of mental illness, psychosis; heart complications; and an increased risk of
stroke.”99
94
https://www.congress.gov/congressional-record/2014/5/29/house-section/article/h49682?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=3, at H4983.
95
See e.g., supra Note 79, at H4906.
96
See supra Note 11, at 53696.
97
http://halrogers.house.gov/news/documentsingle.aspx?DocumentID=398203.
98
“I rise to oppose the amendment.” Supra Note 94.
99
https://www.congress.gov/congressional-record/2015/6/2/house-section/article/h37002?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=2, at H3746.
19
…
“It means the younger a child is exposed to it, the more likely that child will later become
an addict to something else, like methamphetamine, prescription drugs, heroin.” 100
As the Co-Chair of the Addiction, Treatment, and Recovery Caucus, Rep. Fleming is charged
with raising awareness and increasing education regarding substance abuse and addiction
treatment. As such, he is in a unique position to educate other members of Congress and the
public about the dangers and benefits of medical cannabis. As illustrated by his statements in the
May 28, 2014 and June 2, 2015 House discussions, 101 he is directly influenced by inaccurate
DEA information and promulgates this shoddy information in support of strict anti-medical
cannabis laws and stronger enforcement of the CSA amongst the states. It is clear that inaccurate
DEA information regarding the gateway theory and cannabis’ alleged ability to cause psychosis
has a genuinely clear and substantial impact at the national level on important public policy
decisions related to federal justice issues.
Representative Frank Wolf (R-VA) opposed the Amendment in a floor speech on May 9,
2012 discussing the Commerce, Justice, Science, and Related Agencies Appropriations Act of
2013.102 Representative Jerrold Nadler (D-NY) pointed out why this was the case:
“I heard [Rep. Wolf] say that the DEA says there is no medical use for marijuana. That’s
true that they’ve said it. The DEA has no credibility with people who have looked at
[medical cannabis] . . . We know that, for people suffering pain, for people suffering
nausea from AIDS and cancer, marijuana is the only thing that produces relief and
enables them to eat and get sustenance and to regain weight and to, perhaps, regain
health. . . . The DEA doesn’t know [this] because it refuses to see it and refuses to allow
systematic research.” 103
Rep. Wolf’s opposition to the Amendment is directly influenced by DEA misinformation, as
he has directly cited104 to the DEA’s faulty document: “The Dangers and Consequences of
Marijuana Abuse.” The statement above lends further credence to the fact that DEA
misinformation has a genuinely clear and substantial impact at the national level on important
public policy decisions related to federal justice issues.
Due to the widespread acceptance of inaccurate DEA information amongst the United States
Congress, the information at issue has a genuinely clear and substantial influential impact on
federal public policy decisions. This is especially true when considering DEA statements which
100
Id. at H3747.
101
See Supra Notes 79-80 & 99-100.
102
https://www.congress.gov/congressional-record/2012/5/9/house-section/article/h25153?q=%7B%22search%22%3A%5B%22marijuana%22%5D%7D&resultIndex=1, at H2525.
103
Id. at H2526.
104
Supra Note 81.
20
perpetuate the false notions that cannabis use causes psychosis and acts as a gateway drug to
more serious drug abuse. Affected persons (i.e. ASA members) have already been affected by
Congress’ continuing refusal to hold a vote on the CARERS Act, and they will be further
affected if the Amendment is not reauthorized. Because the information at issue is “influential
information” within the meaning of the Guidelines, the DEA should review the inaccurate DEA
information with an added level of scrutiny, to ensure that it is reproducible.
V.!
ASA REPRESENTS SERIOUSLY ILL “AFFECTED PERSONS” WHO ARE
DEEPLY AND IMMEDIATELY AFFECTED BY THE DEA’S INCORRECT
AND CONTROVERTED STATEMENTS
a.! ASA’s members are “affected persons” within the meaning of the DOJ’s
Information Quality Guidelines
According to the DOJ and OMB Guidelines, affected persons are allowed to “seek and
obtain, where appropriate, timely correction of information maintained and disseminated by the
agency that does not comply with OMB or agency guidelines.”105 And, an “affected person” is
an “individual or entity that may use, benefit, or be harmed by the disseminated information at
issue.”106 ASA is composed of the following affected persons: (1) patients who are unable to
access medical cannabis or are at risk of losing access; (2) doctors who are unable to recommend
medical cannabis or are at risk of losing their ability to recommend it; (3) patients and providers
who have been criminally prosecuted or are at risk of prosecution; and (4) scientists who are
unable to obtain cannabis for research or are at risk of losing access.107 On behalf of these
affected persons, ASA seeks to obtain correction of DEA information that fails to comply with
the Guidelines. ASA and its individual members are currently being harmed by – and are at risk
of future harm from – the DEA’s dissemination of inaccurate information regarding medical
cannabis. Specifically, the DEA’s aforementioned statements regarding the gateway theory,
cannabis’ supposed tendency to induce psychosis and lung cancer, and the alleged permanency
of cannabis associated cognitive deficits have harmed and continue to harm ASA and its
members. The harm results because the inaccurate information obfuscates legitimate medical
cannabis research, which would otherwise inform our elected official’s opinions and actions.
As described in Section III, elected officials across the nation rely on DEA information when
forming opinions about the safety and efficacy of medical cannabis. These officials have made
public policy decisions based, at least in part, on inaccurate DEA information. These policy
decisions include failing to reschedule cannabis via passage of the CARERS Act, which has the
effect of denying patients access to medical cannabis, preventing doctors from prescribing
medical cannabis, and criminally prosecuting medical cannabis users/providers. And, while
there are many states that have implemented their own medical cannabis systems, medical
cannabis remains federally illegal, in part due to elected officials’ inaccurate perceptions that
105
Supra Note 13, at “Introduction and Purpose.”
106
Id. at “Process for Citizen Complaint.”
107
ASA has members residing in every United States Congressional District.
21
cannabis is a gateway drug and that it causes psychosis, lung cancer, and permanent cognitive
deficits. The federal status of medical cannabis has prevented multiple states from allowing
healthcare providers to recommend medical cannabis in those states. Furthermore, there is a
substantial risk that a misinformed Congress will either repeal or refuse to reauthorize the
Amendment, thereby urging the DOJ to enforce the CSA in states with legal medical cannabis
systems.
The inaccurate perceptions of at least several outspoken United States Congressmen originate
from DEA information lacking both objectivity and utility. These representatives often push for
stricter enforcement of the CSA in the states and maintenance of cannabis as a Schedule I drug.
A correction of the erroneous DEA information would benefit ASA, its members, and millions of
medical cannabis patients by shifting US representatives’ perceptions of the true risks of medical
cannabis. Such a shift could result in many benefits, including but not limited to: (1) patients’
continued access to medical cannabis in states that currently permit its use;108 (2) patients’ access
to medical cannabis in states which currently prohibit its use;109 (3) elimination of criminal
penalties for medical cannabis physicians and patients;110 and (4) more federal funding and
access to cannabis for medical research. 111
108
There were approximately 2,045,888 registered medical cannabis patients as of Dec. 2015, based on available
patient registry statistics compiled by ASA. Available at https://american-safeaccess.s3.amazonaws.com/documents/EstimatedNumberOfMMJPatientsDec2015.pdf.
109
There are currently 6 states with no medical cannabis and an additional 15 states with limited CBD-focused laws.
Only one of the CBD-focused laws allows for patients to obtain the medical cannabis-derived products from a
dispensary in the state, all other CBD-focused laws only protect patients from arrest if they obtain and possess
products acquired from a state with licensed distribution and reciprocity access.
110
According to the FBI, there were 643,121 cannabis arrests in 2015, over 89% of which were for possession alone
– this is the crime patients are most likely to violate. However, the FBI does not provide any information on how
many of those arrests involved a defendant claiming medical necessity. While medical cannabis physicians are
rarely targeted for arrest, the chilling effect of its Schedule I status creates stigma that suppresses the number of
physicians who are willing to recommend medical cannabis under state law. Available at https://ucr.fbi.gov/crimein-the-u.s/2015/crime-in-the-u.s.-2015/home.
111
Researchers have commented on the lack of federal funding available for medical cannabis research. University
of Pennsylvania professor Marcel Bonn-Miller said, “[f]rom the National Institutes of Health to the VA to whatever,
there was nothing,” referring to the available funding for medical cannabis research. Ethan Russo, Former GW
Pharmaceuticals researcher and current medical director at the Los Angeles biotechnology firm Phytecs, elaborated
on the problem facing medical cannabis researchers: “Traditionally, if you had a compelling reason to do research,
you could get funding … Now nothing is getting funded unless you have something really sexy. And marijuana is
like kryptonite.” Between 1999 and 2012, the number of studies approved for funding dropped from 34% to 19%.
Available at http://www.ibtimes.com/marijuana-news-2016-scientists-frustrated-funding-shortfalls-launch-institute2379921.
22
23
Enclosure D: ASA's IQA Deadline Letter to DEA The Dangers and Consequences of Misinformation on Marijuana February 22, 2017
Enclosures E: Relevant research studies from ‘Denial of Petition to Initiate Proceedings to Reschedule Marijuana” I. II. III. IV. V. Pope, H. G., Gruber, A. J., Hudson, J. I., Huestis, M. A., & Yurgelun-‐Todd, D. (2001). Neuropsychological performance in long-‐term cannabis users. Archives of General Psychiatry, 58(10), 909–915. Degenhardt, L., Chiu, W. T., Conway, K., Dierker, L., Glantz, M., Kalaydjian, A., et al. (2009). Does the “gateway” matter? Associations between the order of drug use initiation and the development of drug dependence in the National Comorbidity Study Replication. Psychological Medicine, 39(1), 157–
167. http://doi.org/10.1017/S0033291708003425 Fergusson, D. M., Horwood, L. J., & Ridder, E. M. (2005). Tests of causal linkages between cannabis use and psychotic symptoms. Addiction, 100(3), 354–366. Degenhardt, L., Hall, W., & Lynskey, M. (2003). Testing hypotheses about the relationship between cannabis use and psychosis. Drug Alcohol Depend, 71(1), 37–48. Kandel, D. B., & Chen, K. (2000). Types of marijuana users by longitudinal course. J Stud Alcohol, 61(3), 367–
378. http://doi.org/10.15288/jsa.2000.61.367http://doi.org/10.1017/S00332917080
03425 VI. VII. VIII. Lee, M. H. S., & Hancox, R. J. (2011). Effects of smoking cannabis on lung function. Expert Review of Respiratory Medicine, 5(4), 537–46– quiz 547. http://doi.org/10.1586/ers.11.40 Kuepper, R., van Os, J., Lieb, R., Wittchen, H. U., Hofler, M., & Henquet, C. (2011). Continued cannabis use and risk of incidence and persistence of psychotic symptoms: 10 year follow-‐up cohort study. BMJ, 342,, 738–738. http://doi.org/10.1136/bmj.d738 Minozzi, S., Davoli, M., Bargagli, A. M., Amato, L., Vecchi, S., & Perucci, C. A. (2010). An overview of systematic reviews on cannabis and psychosis: Discussing apparently conflicting results. Drug and Alcohol Review, 29(3), 304–317. http://doi.org/10.1111/j.1465-‐3362.2009.00132.x The Dangers and Consequences of Misinformation on Marijuana February 22, 2017
ORIGINAL ARTICLE
Neuropsychological Performance
in Long-term Cannabis Users
Harrison G. Pope, Jr, MD; Amanda J. Gruber, MD; James I. Hudson, MD, SM;
Marilyn A. Huestis, PhD; Deborah Yurgelun-Todd, PhD
Background: Although cannabis is the most widely used
illicit drug in the United States, its long-term cognitive
effects remain inadequately studied.
Methods: We recruited individuals aged 30 to 55 years
in 3 groups: (1) 63 current heavy users who had smoked
cannabis at least 5000 times in their lives and who were
smoking daily at study entry; (2) 45 former heavy users
who had also smoked at least 5000 times but fewer than
12 times in the last 3 months; and (3) 72 control subjects who had smoked no more than 50 times in their
lives. Subjects underwent a 28-day washout from cannabis use, monitored by observed urine samples. On days
0, 1, 7, and 28, we administered a neuropsychological
test battery to assess general intellectual function, abstraction ability, sustained attention, verbal fluency, and
ability to learn and recall new verbal and visuospatial information. Test results were analyzed by repeated-
D
From the Biological Psychiatry
Laboratory, McLean Hospital,
and the Department of
Psychiatry, Harvard Medical
School, Belmont, Mass
(Drs Pope, Gruber, Hudson,
and Yurgelun-Todd); and the
Intramural Research Program,
National Institute on Drug
Abuse, Baltimore, Md
(Dr Huestis).
measures regression analysis, adjusting for potentially confounding variables.
Results: At days 0, 1, and 7, current heavy users scored
significantly below control subjects on recall of word lists,
and this deficit was associated with users’ urinary 11nor-9-carboxy-!9-tetrahydrocannabinol concentrations at study entry. By day 28, however, there were virtually no significant differences among the groups on any
of the test results, and no significant associations between cumulative lifetime cannabis use and test scores.
Conclusion: Some cognitive deficits appear detectable
at least 7 days after heavy cannabis use but appear reversible and related to recent cannabis exposure rather
than irreversible and related to cumulative lifetime use.
Arch Gen Psychiatry. 2001;58:909-915
OES LONG-TERM heavy use
of cannabis cause residual neuropsychological deficits? The literature has long been divided
on this question.1 A recent investigation by
our laboratory found deficits on memory
of word lists and on mental flexibility
among 65 heavy-smoking college students, compared with 64 infrequent smokers after 1 day of abstinence from cannabis. 2 Fletcher et al 3 found significant
differences between 17 older heavy cannabis users and 30 matched nonusers on
memory of word lists and on selective and
divided attention tasks after 72 hours of abstinence. However, these authors found no
significant differences between 37 younger
users and 49 matched nonusers. Another
group found electroencephalographic abnormalities in chronic cannabis users after 24 hours of abstinence,4,5 but found no
significant alteration in auditory or visual
P300 responses in another study of cannabis users, after controlling for potentially
confounding variables. 6 By contrast,
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
909
Solowij7 found significant delays in auditory P300 responses in heavy cannabis users examined after at least 12 hours of abstinence. Cannabis users also displayed
significantly slower reaction times and reduced accuracy on a selective attention task.
However, it is difficult to determine
whether such deficits, observed after only
12 to 72 hours of abstinence, are temporary
(eg, due to a residue of cannabinoids in the
brainortoacutewithdrawaleffectsfromcannabis) or long-lasting (due to a neurotoxic
effect of long-term cannabis exposure). On
this critical latter question, the data are meagerandconflicting.Lyketsosandcolleagues,8
examining 1318 participants younger than
age 65 in the Epidemiologic Catchment Area
Study, found no significant differences
among heavy cannabis users, light users, and
nonusers in the degree of cognitive decline
on the Mini-Mental State Examination during the course of 12 years. By contrast, Struve
and colleagues9 tentatively suggested that
electroencephalographicabnormalitieswere
more pronounced in longer-duration cannabis users, even when adjusting for the
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
SUBJECTS AND METHODS
SUBJECTS
We recruited individuals aged 30 to 55 years in 3 groups:
(1) current long-term heavy users reporting at least 5000
lifetime episodes of cannabis smoking (to be counted as
separate, episodes had to be at least 1 hour apart), and currently smoking at least 7 times per week; (2) former longterm heavy users reporting at least 5000 episodes of smoking, but no more than 12 episodes during at least the last 3
months; and (3) control subjects reporting that they had
smoked at least once, but no more than 50 times in their
lives, and no more than once during the past year.
Our threshold of 5000 episodes for “heavy use” was
equivalent to smoking at least once a day for at least 13 years.
We considered recruiting controls who had never smoked
cannabis, but elected to choose subjects who had tried the
drug at least once, because individuals who had never tried
cannabis might differ from individuals who had in ways
that might be associated with cognitive performance. All
subjects were studied at McLean Hospital, Belmont, Mass,
and were required to sign informed consent for the study,
which was approved by the McLean Hospital institutional
review board.
Subjects qualifying on telephone screening were evaluated by one of us (H.G.P. or A.J.G.) at a baseline (day 0)
interview, which included demographic questions, detailed questions about frequency of use of cannabis and other
drugs throughout the subject’s lifetime, the Structured Clinical Interview for DSM-IV,10 assessment for history of attention-deficit/hyperactivity disorder (ADHD) using the
Wender Utah Rating Scale11 and a modified ADHD rating
scale,12,13 semistructured questions regarding family history of DSM-IV Axis I psychiatric disorders,14 and laboratory tests for standard chemistries, hematology, and urinalysis. Ratings of ADHD were introduced only during the
second year of the study and, hence, were limited to 109
of the 180 subjects (33 current users, 31 former users, and
45 controls). We calculated a conduct disorder score by
adding the scores on 4 items on the Wender Utah Rating
Scale: “ran away from home”; “get in fights”; “trouble with
authorities, trouble with school, visits to the principal’s office”; and “trouble with the police, booked, convicted.”
We excluded subjects who reported (1) use of any other
class of drugs of abuse (such as hallucinogens, cocaine,
stimulants, or opiates) more than 100 times in their lives;
(2) a history of DSM-IV alcohol abuse or dependence; (3)
a current DSM-IV Axis I disorder other than simple phobia or social phobia; (4) a history of a head injury with loss
of consciousness requiring hospitalization; (5) current use
of any psychoactive medication; or (6) a medical, psychiatric, or neurological condition that might affect cognitive
greater age of these subjects. Most ominously, Solowij7
found a strong correlation between duration of cannabis
use and increased processing negativity to complex irrelevant stimuli in a selective attention task, even in users
with a mean of 2 years’ abstinence.
To augment these limited data on the cognitive consequences of long-term cannabis use, we examined neuropsychological performance in 108 long-term heavy users of cannabis throughout 28 days of monitored
abstinence from the drug.
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
910
function. We also screened urine by immunoassay (EMIT
II; Behring Diagnostics, Cupertino, Calif) for 11-nor-9carboxy-!9-tetrahydrocannabinol (THCCOOH), creatinine, cocaine metabolites, benzodiazepines, barbiturates,
phencyclidine hydrochloride, opioids, and amphetamines, and by enzymatic assay for ethanol. The immunoassay threshold for detection of cannabinoids was 20
ng/mL; ethanol detection was considered positive if it exceeded 0.02 g/dL. Samples positive for THCCOOH were then
tested by gas chromatography-mass spectroscopy to obtain
quantitative THCCOOH and creatinine concentrations.
Samples showing evidence of ethanol levels above 0.02 g/dL,
or evidence of any of the other 6 classes of drugs listed, were
also confirmed by gas chromatography–mass spectroscopy.
ABSTINENCE PERIOD
Following the baseline evaluation, subjects were required to
abstain from cannabis and other drugs of abuse for 28 days,
monitored by observed urine samples daily (current users)
or every other day (former users and controls). All subjects
were permitted to consume caffeine and tobacco, and up to
2 alcoholic drinks (defined as 12 oz of beer, 4 oz of wine, or
11⁄2 oz of distilled liquor) per day. Subjects were withdrawn
from the study if urine samples indicated noncompliance with
these requirements. Current users, who by definition were
smoking regularly up until day 0, were judged to be abstinent provided that their urinary THCCOOH concentrations, normalized to urinary creatinine concentrations, decreased in a manner consistent with residual drug excretion
in the absence of any new cannabis use.15
NEUROPSYCHOLOGICAL TESTING
On days 0, 1, 7, and 28, an investigator, blinded to the subjects’ group status, administered the neuropsychological tests
described in this subsection. To maintain blindness, the
tester worked in a separate building. Before testing, subjects were instructed not to reveal to the tester any information about their prior cannabis use or current frequency of urine samples.
Day 0
At baseline, subjects were administered the vocabulary subtest of the Wechsler Adult Intelligence Scale–Revised, a measure correlated with general intellectual ability16 and relatively insensitive to cortical insults.17
Days 0, 1, 7, and 28
On all 4 testing days, subjects were administered (1) a computerized Continuous Performance Test (Conners’
RESULTS
On telephone screening, 246 subjects appeared to meet criteria for 1 of the 3 study groups. Of these, 66 were either
excluded at the baseline interview or subsequently withdrawn during the study (Figure), leaving 180 evaluable
subjects. The 3 groups (63 current users, 45 former users,
and 72 control subjects) were similar in age, ethnic distribution, and sex (the latter because of matching) (Table 1).
Interestingly, subjects in all groups reported similar eduWWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
version 3.0)18; (2) an Auditory Continuous Performance
Test19 to assess measures of attention; and (3) the Buschke Selective Reminding Test (BSRT)20 to assess verbal
learning and memory. On days 0, 7, and 28, subjects also
received the Benton Revised Visual Retention Test21 to assess visuospatial memory. The BSRT and Benton Revised
Visual Retention Test were administered in alternate forms
to minimize learning effects.
Day 28
On the final testing day, subjects were administered 6 additional tests: (1) the Wisconsin Card Sorting Test22; (2) the
Wechsler Memory Scale23; (3) the block design subtest of
the Wechsler Adult Intelligence Scale–Revised16; (4) the
Controlled Oral Word Association Test (often known as the
“FAS” test)24; (5) the Stroop Test25; and (6) the Raven Progressive Matrices.26 These measures of attentional and executive functions and verbal and visuospatial memory were
chosen because of their known sensitivity to various forms
of brain dysfunction17,24 and because they had demonstrated possible deficits in heavy cannabis users in previously published studies.1-3 Because these 6 tests were not available in multiple versions, they could be administered on only
a single occasion and, thus, were reserved for day 28.
STATISTICAL ANALYSIS
For baseline demographic characteristics, we compared
groups using the Fisher exact test for binary variables and
the Wilcoxon rank sum test for continuous variables. For
neuropsychological test scores, we compared current users and former users separately with controls via multivariate linear regression analysis. We used 2 sets of adjustments for possible confounding variables. Analysis 1
adjusted only for variables that could not have been affected by cannabis use: sex, age, ethnicity (white vs nonwhite), mother’s and father’s educational level, parents’
household income, presence of substance abuse or dependence in a first-degree relative, and presence of any other
psychiatric disorder in a first-degree relative. Analysis 2 adjusted for verbal IQ (VIQ), as determined by the vocabulary subtest of the Wechsler Adult Intelligence ScaleRevised in addition to the other variables.
Because VIQ is generally well preserved despite cortical insults,16,17 analysis 2 was intended to adjust for the
effects of premorbid intelligence. This adjustment is potentially important, because the heavy users displayed lower
VIQs than did controls (see the “Results” section). However, we cannot exclude the possibility that the lower VIQs
of heavy users might be partially a consequence, rather than
an antecedent, of cannabis use. Therefore, the 2 analyses
effectively provide upper and lower bounds for the
cational levels and household income in their families of
origin, whereas the subjects themselves differed markedly
on these same indices, with users reporting much lower
educational attainment and income than controls.
Of the 4 neuropsychological tests performed serially during the 28 days of abstinence, 2 (the Auditory Continuous Performance Test and Continuous Performance Test) revealed no significant differences between
control subjects and current users, in analyses with and
without VIQ adjustment, on any of the 4 testing days on
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
911
neuropsychological effects of cannabis use: analysis 1 (VIQunadjusted) assumes that the lower VIQ of heavy users is
entirely a consequence of cannabis use and entirely unrelated to premorbid differences in intelligence, while analysis 2 (VIQ-adjusted) assumes that lower VIQ is entirely a
consequence of premorbid differences and entirely unrelated to cannabis use. If one assumes that the truth lies somewhere between these extremes, then the VIQ-unadjusted
analysis would be expected to overestimate the true neuropsychological deficits associated with heavy cannabis use,
whereas the VIQ-adjusted analysis would tend to underestimate such deficits.
For tests involving serial measures at different time
points, we used the methods of longitudinal analysis
with generalized estimating equations, with compound
symmetry as a working covariance structure, to account
for correlation of observations within individuals.27 We
used appropriate transformations for variables in which
there appeared to be a dependence of the variance on the
mean.
We also tested the association between neuropsychological measures and lifetime use of cannabis in current and
former users, and between these measures and baseline
THCCOOH-creatinine ratio. For these analyses, we used
multivariate linear regression as already described in this
subsection, except that we restricted the analysis to a single
group and entered as predictor variables lifetime use (modeled as log of the total number of lifetime episodes of use)
and baseline THCCOOH-creatinine ratio. Using this ratio
allowed us to correct for differences in the concentration
of urine samples provided by subjects at day 0 and, thus,
provided a rough approximation of the subject’s recent exposure to cannabinoids. We modeled this value as log (ratio + 1).
We had complete information on the most important
covariates: age, sex, ethnicity, and VIQ. For the small number of missing observations for other covariates, we assigned the median value for the total sample for purposes
of analysis.
We also fitted a model that included terms for scores
on the ADHD rating scale and the conduct disorder scores
calculated as described in the “Subjects” subsection of the
“Subjects and Methods” section. This was a secondary analysis, because these data were limited to 109 subjects and because we could not exclude the possibility that some features of ADHD and conduct disorder represented effects
of cannabis use.
All tests were 2-tailed. The large number of correlated outcome measures makes proper adjustment for multiple comparisons difficult. To control partially for the effects of multiple comparisons, we set the " level at .01.
We used commercially available statistical software
(Stata 6.028) for all analyses.
any of the measures tested (total correct responses and
total errors). On the Benton Revised Visual Retention Test,
the groups did not differ significantly at any time point
on the number of correct responses, but current users
made more errors on day 0, although this difference met
our proposed "=.01 only in the analysis without VIQ adjustment (adjusted mean difference [SE], 1.2 [0.3],
P = .001 without VIQ adjustment; 0.8 [0.3], P = .02 with
VIQ adjustment). However, memory of word lists on the
BSRT more consistently distinguished the current users
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
from control subjects at days 0, 1, and 7, although generally not at day 28 (Table 2). The former users, by contrast, were not significantly different from controls on all
measures of all 4 tests at all time points, in the VIQadjusted and VIQ-unadjusted analyses.
Scores on the 6 neuropsychological tests administered exclusively at day 28 appeared consistent with these
findings. We found no significant differences between either the current or former users and the control subjects,
using either the VIQ-adjusted or VIQ-unadjusted analySubjects Selected as Candidates for Study
on the Basis of Telephone Screen
N = 246
Current Heavy Users
n = 100
Former Heavy Users
n = 62
Comparison Group
n = 84
Excluded at Screen:
Other Substance Use n = 15
Axis I Disorder n = 1
Other n = 5
Excluded at Screen:
Other Substance Use n = 3
Axis I Disorder n = 5
Medical Disorder n = 2
Other n = 4
Excluded at Screen:
Axis I Disorder n = 3
Medical Disorder n = 2
Other n = 2
Current Heavy Users
n = 79
Former Heavy Users
n = 48
Comparison Group
n = 77
Withdrawn During Study:
2 Poor Compliance
3 Withdrew Consent
1 Onset of Severe Depression
4 Urine Samples Showed Cocaine
1 Urine Sample Showed Alcohol
2 Urine Samples Showed
Increasing THC
3 Urine Samples Showed No THC
Withdrawn During Study:
1 Poor Compliance
1 Withdrew Consent
1 Urine Sample Showed
Cocaine
Withdrawn During Study:
1 Poor Compliance
1 Withdrew Consent
1 Urine Sample Showed Alcohol
2 Urine Samples Showed
Amphetamine
Completed Subjects:
n = 63
Completed Subjects:
n = 45
Completed Subjects:
n = 72
Flow sheet showing subjects recruited and withdrawn in the 3 study groups.
THC indicates tetrahydrocannabinol.
ses, on the standard measures generated by these tests, as
shown in Table 3. In addition to the measures shown in
Table 3, we also failed to find significant differences in any
of these same comparisons on times for word reading and
color naming on the Stroop Test; immediate and delayed
memory for stories, figures, and pairs on the Wechsler
Memory Scale; digit span on the Wechsler Memory Scale;
and total categories achieved on deck 1 of the Wisconsin
Card Sorting Test. On categories achieved on deck 2 of the
Wisconsin Card Sorting Test, we found one significant difference: in the VIQ-unadjusted analysis, current users
achieved fewer categories than did controls (estimated difference [SE], −0.5 [0.2] categories, P = .003). However, this
difference largely disappeared in the VIQ-adjusted analysis (−0.2 [0.2], P = .25) and failed to achieve significance
in the comparison of former users vs controls (VIQunadjusted, −0.3 [0.2], P = .09; VIQ-adjusted, −0.3 [0.2],
P = .17). Overall, these findings suggest that cognitive deficits associated with cannabis use persisted at least 7 days,
but could not be detected with our measures after 28 days.
We then performed additional analyses to test the
impression that reduced cognitive performance was associated with recent exposure to cannabis, rather than
total lifetime use of the drug. First, as described in the
“Subjects and Methods” section, we examined the association between subjects’ estimated lifetime number of
episodes of use and performance at day 28 on all of the
measures shown in Tables 2 and 3. Subjects’ lifetime cannabis use varied more than 10-fold, from 5000 to more
than 70000 estimated episodes, thus permitting a test of
the association between total use and test measures. In
current and former users, however, none of these asso-
Table 1. Demographic Features of Current Users and Former Users vs Control Subjects*
Demographic Feature
Age [range], y
Male
White
High school education or less
Annual household income #$30 000
Mother’s education high school or less§
Father’s education high school or less!
Parents’ annual household income #$30 000¶
Family history of any Axis I disorder#
Lifetime episodes of cannabis use
Years smoking cannabis $7 times per week
Lifetime alcoholic drinks
Lifetime packs of cigarettes
Lifetime caffeinated drinks
Conduct disorder score‡‡
Attention-deficit/hyperactivity disorder score‡‡
Verbal IQ
Current Users
(n = 63)
Former Users
(n = 45)
Controls
(n = 72)
36 [32-41]
55 (87)
54 (86)
18 (29)†
32 (51)‡
37 (61)
26 (46)
16 (25)
37 (60)
18 720 [11 700-27 000]**
19 [15-24]**
4700 [2100-7700]
730 [0-5100]**
13 800 [3000-23 200]
1 [1-3]††
10 [4-14]
106 [95-118]**
41 [37-48]
30 (67)
39 (87)
6 (13)‡
23 (51)‡
17 (40)
22 (50)
11 (25)
21 (50)
11 000 [8400-16 000]**
15 [11-19]**
3900 [1100-10 100]
420 [0-4400]**
15 300 [3200-26 100]
1 [0-2]
10 [7-13]
115 [99-127]
39.5 [34-44]
61 (85)
60 (83)
0
19 (26)
42 (58)
27 (38)
13 (18)
26 (37)
10 [5-25]
0
2800 [1100-5500]
0
12 400 [3600-20 000]
0 [0-1]
7.5 [5-15]
115 [110-126]
*Data are given as number (percentage) for proportions and as median [interquartile range] for continuous variables. P values are significance of differences vs
controls. All statistical tests were 2-tailed. Numbers of users and controls vary because of missing data.
†P#.001, Fisher exact test.
‡P#.01, Fisher exact test.
§61 current users and 43 former users.
!56 current users, 44 former users, and 71 controls.
¶44 former users.
#62 current users, 42 formers users, and 70 controls.
**P#.001, Wilcoxon rank sum test.
††P#.01, Wilcoxon rank sum test.
‡‡Thirty-three current users, 31 former users, and 45 controls.
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
912
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
Table 2. Scores of Study Groups on the Buschke Selective Reminding Test on Successive Testing Days
Estimated Mean Differences (SE) Between Groups*
Mean (SD) Scores
Current Users vs Controls
Current Users
(n = 63)
Former Users
(n = 45)
Controls
(n = 72)
Day 0
Day 1
Day 7
Day 28
104.5 (15.0)
106.7 (17.0)
111.7 (15.4)
116.4 (12.9)
109.1 (13.4)
114.9 (11.3)
118.7 (12.9)
117.9 (13.5)
Total Recall
113.6 (16.3)
115.5 (15.7)
120.9 (13.6)
121.1 (13.6)
Day 0
Day 1
Day 7
Day 28
96.5 (22.7)
99.4 (22.5)
105.7 (22.0)
112.4 (18.6)
104.5 (21.1)
108.2 (16.5)
115.1 (18.3)
112.4 (19.6)
Day 0
Day 1
Day 7
Day 28
58.0 (29.2)
62.0 (33.4)
70.8 (32.8)
79.4 (31.2)
Day 0
Day 1
Day 7
Day 28
8.8 (2.3)
8.3 (2.7)
8.6 (2.6)
9.1 (2.3)
Former Users vs Controls
With VIQ
Adjustment
Without VIQ
Adjustment
With VIQ
Adjustment
Without VIQ
Adjustment
−6.1 (2.7)
−5.8 (2.8)
−6.3 (2.4)†
−1.8 (2.4)
−9.1 (2.6)†
−8.8 (2.8)†
−9.3 (2.4)‡
−4.8 (2.1)
−3.0 (2.4)
0.9 (2.2)
−0.6 (2.1)
−1.7 (2.3)
−3.9 (2.5)
0.0 (2.2)
−1.6 (2.2)
−2.6 (2.2)
Long-term Storage
108.6 (21.9)
−8.5 (4.1)
109.0 (21.2)
−6.0 (3.9)
117.2 (16.6)
−7.8 (3.5)
117.2 (19.0)
−1.2 (3.5)
−11.9 (3.8)†
−9.4 (3.7)
−11.2 (3.3)†
−4.5 (3.1)
−1.5 (3.8)
1.7 (3.3)
0.4 (2.9)
−2.3 (3.4)
−2.8 (3.8)
0.4 (3.2)
−0.8 (3.0)
−3.6 (3.2)
64.4 (27.7)
75.0 (28.4)
86.9 (30.9)
82.2 (32.6)
Consistent Long-term Retrieval
77.1 (33.0)
−13.7 (5.4)
81.7 (36.3)
−14.3 (6.0)
91.5 (35.2)
−15.3 (5.6)†
91.9 (31.7)
−7.1 (5.6)
−19.3 (5.1)‡
−19.9 (5.7)‡
−20.9 (5.4)‡
−12.7 (5.0)
−10.3 (5.1)
−4.3 (5.1)
−2.2 (5.4)
−7.3 (5.3)
−12.4 (5.2)
−6.3 (5.3)
−4.3 (5.5)
−9.4 (5.4)
9.4 (2.2)
9.6 (2.4)
9.9 (2.3)
9.2 (2.5)
30-Minute Delayed Free Recall
9.7 (2.4)
−0.6 (0.4)
9.6 (2.6)
−0.8 (0.4)
10.1 (2.0)
−1.1 (0.4)†
10.2 (2.2)
−0.8 (0.4)
−1.0 (0.4)
−1.2 (0.4)†
−1.5 (0.4)‡
−1.1 (0.4)†
−0.3 (0.4)
0.0 (0.4)
−0.2 (0.4)
−0.9 (0.4)
−0.4 (0.4)
−0.1 (0.5)
−0.2 (0.4)
−1.0 (0.4)
*VIQ indicates verbal IQ. P values are significance of differences vs controls.
†P#.01.
‡P#.001.
Table 3. Scores of Study Groups at Day 28 on Representative Test Measures
Estimated Mean Differences (SE) Between Groups*
Mean (SD) Scores at Day 28
Test Score
Total score on Wechsler
Memory Scale
Raw score on Controlled Oral
Word Association Test
Total perseverations on
Wisconsin Card Sorting
Test†
Scaled score on block design
subtest of the Wechsler
Adult Intelligence
Scale−Revised
Color interference time on
Stroop Test, s
Total score on Raven
Progressive Matrices
Current Users vs Controls
Former Users vs Controls
Current Users
(n = 63)
Former Users
(n = 45)
Controls
(n = 72)
With VIQ
Adjustment
Without VIQ
Adjustment
With VIQ
Adjustment
Without VIQ
Adjustment
69.3 (8.4)
68.9 (6.8)
70.3 (6.2)
0.2 (1.3)
−1.6 (1.2)
−0.1 (1.2)
−1.2 (1.2)
47.1 (10.8)
48.2 (10.3)
51.4 (11.0)
−2.3 (2.3)
−4.9 (2.0)
−2.7 (2.2)
−3.2 (2.2)
2.4 (0.8)
2.4 (0.8)
2.1 (0.7)
0.1 (0.1)
0.3 (0.1)
0.1 (0.2)
0.2 (0.1)
11.7 (2.5)
11.4 (2.5)
11.9 (2.6)
0.1 (0.5)
−0.5 (0.4)
−0.1 (0.5)
−0.3 (0.5)
105.5 (26.5)
107.4 (24.1)
101.5 (23.5)
5.5 (5.3)
7.4 (4.5)
3.5 (5.1)
3.8 (4.8)
49.3 (6.5)
49.4 (6.9)
51.1 (6.7)
−0.5 (1.3)
−2.3 (1.1)
−1.1 (1.3)
−1.9 (1.3)
*None of these differences achieved statistical significance. VIQ indicates verbal IQ.
†Shown and analyzed as logarithm of total perseverations because of right-skewed distribution.
ciations proved significant in either the VIQ-adjusted or
VIQ-unadjusted analyses.
Turning to the issue of recent cannabis exposure,
we also examined the association between baseline
THCCOOH-creatinine ratios and the neuropsychological measures at each time point for the current users. This
analysis, with VIQ adjustment, produced significant associations between baseline ratios and BSRT Total Re(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
913
call at day 1 (estimated decrease in words recalled for every increase of 1 in log of ratio [SE], −5.7 [2.0], P = .005)
and Consistent Long-term Retrieval on day 1 (−11.8 [4.3],
P = .006). Without VIQ adjustment, we also found significant associations with BSRT Total Recall at day 1 (−6.6
[2.1], P = .002), Consistent Long-term Retrieval at day
1 (−13.3 [4.4], P = .002) and day 7 (−11.8 [4.2], P = .005),
and 30-Minute Delayed Recall at day 28 (−0.9 [0.3],
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
P=.003). However, we found no significant association
between baseline ratios and scores on the other 3 serial
tests or on the 6 tests given exclusively at day 28.
We also examined the effects of sex. On all of the measures in Tables 2 and 3, we found no significant genderby-group interaction. However, the power of this analysis was limited by the small number of female subjects.
Given evidence that ADHD and antisocial behavior
may be associated with neurocognitive deficits,29-34 we
also performed analyses adjusting for ADHD and for conduct disorder scores among the 109 subjects for whom
we possessed these data. However, adjustment for these
variables produced only small changes in the estimate of
mean effect of group on each of the neuropsychological
measures and did not alter any qualitative conclusions
(ie, whether a result was statistically significant).
COMMENT
In a study of cognitive function among long-term heavy
cannabis users, we found deficits on memory of word lists,
detectable at least 7 days after discontinuing the drug and
related to initial urinary concentrations of THCCOOH. After 28 days of abstinence, however, users showed virtually no significant differences from control subjects on a
battery of 10 neuropsychological tests. Former heavy users, who had consumed little or no cannabis in the 3 months
before testing, showed no significant differences from control subjects on any of these tests on any of the testing days.
The paucity of significant differences between the cannabis and control groups at day 28, together with the lack of
significant associations between test scores and lifetime cannabis consumption, suggests that cannabis-associated cognitive deficits may be reversible phenomena associated with
recent drug exposure, rather than irreversible phenomena associated with cumulative lifetime use.
Deficits on memory of word lists, persisting for days
after discontinuing cannabis use, might be attributable to
cannabinoids lingering in the central nervous system or to
withdrawal from abruptly stopping use. Although we cannot clearly discriminate between these hypotheses, measures of aggression35 and subjective indices36,37 in the users suggest that withdrawal-associated agitation, often lasting
at least 7 days, may have compromised their neuropsychological performance. A withdrawal hypothesis might explain why deficits on the BSRT in current users were at least
as great on day 7 as on days 0 and 1 (Table 2).
Our findings are generally congruent with those of previous studies1-6 showing neuropsychological deficits within
the first few days after cannabis use is stopped. Also, in agreement with another recent study,8 we failed to find an association between cumulative lifetime use of cannabis and
cognitive deterioration. Only the findings by Solowij7 appear somewhat discrepant with ours, in that she found significantly increased processing negativity to irrelevant
stimuli in former heavy users after a mean of 2 years’ abstinence, whereas we found little evidence of neuropsychological deficits after 28 days of abstinence. Possibly, cannabis produces irreversible effects detectable on
electroencephalographic measures, but too subtle to be detected on our neuropsychological test battery. Alternatively, the differences between the 2 studies may have been
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
914
because of unmeasured or inadequately controlled confounding variables.
The cannabis users and controls in our study reported similar educational levels and income in their families of origin, whereas the users themselves exhibited significantly lower educational attainment, income, and
estimated VIQ than controls. We cannot determine
whether these differences are because of premorbid attributes of the users or because of cannabis effects. Even
if cannabis produces little or no irreversible cognitive deficit, chronic cannabis intoxication might still compromise educational ambitions, income potential, and the
acquisition of new verbal information.
Several limitations of our study should be considered. The first is a possible selection bias caused by our
study requirements. For example, users with severe neuropsychological deficits might have been less likely to enter the study, although a similar bias might also have affected the control group. In any event, we cannot exclude
the possibility that we might have underestimated the cognitive deficits associated with cannabis use because severely impaired individuals were underrepresented.
A second limitation is the possibility of residual confounding, because of either unmeasured confounders or
inadequate adjustment for measured confounders. However, it seems unlikely that such confounders could explain the lack of differences between users and controls
at day 28, because the most plausible unmeasured confounding variables in the users—such as undetected psychopathologic conditions, unrecognized premorbid cognitive deficits, unreported prior use of other drugs, or
undetected surreptitious use of cannabis during the study—
would all be expected to militate against our finding of an
absence of differences. Similarly, users’ greater lifetime
consumption of alcoholic drinks and cigarettes would also
be expected to militate against our finding, barring the remote possibility that nicotine from possible compensatory cigarette smoking among abstinent users might actually improve neuropsychological performance.38
Third, subjects’ histories, including information on
cannabis and other drug use, were obtained by selfreport without external validation. However, as mentioned in the “Subjects” subsection of the “Subjects and
Methods” section, subjects were interviewed about their
drug histories without knowledge of the answers necessary to gain acceptance into the study. Furthermore, previous studies39-41 have suggested that self-reports of use
of cannabis and other drugs are fairly reliable. Finally,
our principal positive findings—the initial cognitive deficit of the current users and its association with THCCOOH
concentrations at study entry—were largely independent of self-report, because THCCOOH concentrations
were measured on observed urine samples, using a sophisticated method likely to detect all but the most minimal levels of surreptitious cannabis use.15
Fourth, it might be argued that we should have chosen control subjects who had never used cannabis, as opposed to individuals who had used the drug 1 to 50 times.
However, we reasoned that “minimal-user” controls would
more closely resemble the heavy users on possible confounding variables (measured and unmeasured) than
would “never-used” controls, while still differing more
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
than 1000-fold from the heavy users in their median level
of exposure to cannabis (Table 1).
Fifth, our study design included only a limited assessment of premorbid intellectual functioning, based on
the vocabulary subtest of the Wechsler Adult Intelligence Scale–Revised. Although this measure has been
shown to provide reliable estimates of premorbid IQ in
other populations,17,24 it is possible that lower VIQ is, at
least partly, a consequence, rather than an antecedent,
of long-term cannabis use. As discussed in the “Statistical Analysis” subsection of the “Subjects and Methods”
section, we addressed this question by performing analyses with and without adjustment for VIQ, thus providing upper and lower bounds for our estimate of the neuropsychological deficits associated with cannabis use.
However, in the non–VIQ-adjusted analysis, which would
be expected to be the least favorable to cannabis users,
we still found virtually no significant differences at day
28 between users and controls on the test measures.
Sixth, it is possible that long-term cannabis use might
produce long-term cognitive deficits, but that our neuropsychological tests were not sufficiently sensitive to detect them. For example, practice effects on the BSRT, combined with a possible ceiling effect, might have reduced
the ability of this instrument to detect differences between groups on the fourth administration, on day 28. The
sensitivity of the study is also limited by its sample size.
For example, in the VIQ-adjusted analysis for current users, the 99% confidence intervals for the day 28 test measures shown in Tables 2 and 3 do not exclude an effect of
0.4 to 0.8 (median, 0.6) SD units (the estimated difference between groups divided by the SD in the control
group). Therefore, the possibility remains that more sophisticated neurocognitive assessment measures, such as
electroencephalographic or functional magnetic resonance imaging measures, might reveal deficits in longterm cannabis users below the threshold detectable with
our neuropsychological test battery.
In summary, our findings do not support the hypothesis that long-term heavy cannabis use causes irreversible cognitive deficits, at least at the level detectable
with our test instruments and our sample size. However, in agreement with previous reports, we found evidence that heavy users exhibit some cognitive deficits lasting for many days, and possibly for weeks, after
discontinuing cannabis use.
Accepted for publication May 1, 2001.
This study was supported in part by grant 5 R37 DA10346 from the National Institute on Drug Abuse, Rockville, Md.
Corresponding author: Harrison G. Pope, Jr, MD,
McLean Hospital, Harvard Medical School, 115 Mill St,
Belmont, MA 02478.
REFERENCES
1. Pope HG Jr, Gruber AJ, Yurgelun-Todd D. The residual neuropsychological effects of cannabis. Drug Alcohol Depend. 1995;38:25-34.
2. Pope HG Jr, Yurgelun-Todd D. The residual cognitive effects of heavy marijuana
use in college students. JAMA. 1996;275:521-527.
3. Fletcher JM, Page JB, Francis DJ, Copeland K, Naus MJ, Davis CM, Morris R,
Krauskopf D, Satz P. Cognitive correlates of long-term cannabis use in Costa Rican
men. Arch Gen Psychiatry. 1996;53:1051-1057.
(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, OCT 2001
915
4. Struve FA, Straumanis JJ, Patrick G, Leavitt J, Manno JE, Manno BR. Topographic quantitative EEG sequelae of chronic marihuana use. Drug Alcohol Depend. 1999;56:167-179.
5. Patrick G, Struve FA. Reduction of auditory P50 gating response in marihuana
users: further supporting data. Clin Electroencephalogr. 2000;31:88-93.
6. Patrick G, Straumanis JJ, Struve FA, Fitz-Gerald MJ, Manno JE. Early and middle
latency evoked potentials in medically and psychiatrically normal daily marihuana users. Clin Electroencephalogr. 1997;28:26-31.
7. Solowij N. Cannabis and Cognitive Functioning. Cambridge, England: Cambridge University Press; 1998.
8. Lyketsos CG, Garrett E, Liang KY, Anthony JC. Cannabis use and cognitive decline in persons under 65 years of age. Am J Epidemiol. 1999;149:794-800.
9. Struve FA, Patrick G, Straumanis JJ, Fitz-Gerald MJ, Manno J. Possible EEG sequelae of very long duration marihuana use. Clin Electroencephalogr. 1998;29:
31-36.
10. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for
DSM-IV Axis I Disorders. New York: Biometrics Research Dept, New York State
Psychiatric Institute; 1996.
11. Ward MF, Wender PH, Reimherr FW. The Wender Utah Rating Scale. Am J Psychiatry. 1993;150:885-890.
12. DuPaul GJ. Parent and teacher ratings of ADHD symptoms: psychometric properties in a community-based sample. J Clin Child Psychol. 1991;20:245-253.
13. Findling RL, Schwartz MA, Flannery DJ, Manos MJ. Venlafaxine in adults with
attention-deficit/hyperactivity disorder. J Clin Psychiatry. 1996;57:184-189.
14. Hudson JI, Pope HG Jr, Jonas JM, Yurgelun-Todd D, Frankenburg FR. A controlled family history study of bulimia. Psychol Med. 1987;17:883-890.
15. Huestis MA, Cone EJ. Differentiating new marijuana use from residual drug excretion in occasional marijuana users. J Anal Toxicol. 1998;22:445-454.
16. Wechsler D. Wechsler Adult Intelligence Scale–Revised Manual. Cleveland, Ohio:
Psychological Corp; 1981.
17. Luria A. Higher Cortical Functions in Man. New York, NY: Basic Books; 1966.
18. Conners CK, and Multi-Health Systems Staff. Conners’ Continuous Performance Tests. North Tonawanda, NY: Multi-Health Systems Inc; 1995.
19. Weintraub S, Mesulam M-M. Mental state assessment of young and elderly adults
in behavioral neurology. In: Mesulam M-M, ed. Principles of Behavioral Neurology. Philadelphia, Pa: FA Davis Co; 1985:71-123.
20. Buschke H. Selective reminding for analyses of memory and learning. J Verbal
Learning Verbal Behav. 1973;12:543-550.
21. Benton AL. The Revised Visual Retention Test. 4th ed. New York, NY: Psychological Corp; 1974.
22. Heaton R. Wisconsin Card Sorting Test Manual. Odessa, Fla: Psychological Assessment Resources; 1981.
23. Wechsler D. A standardized memory scale for clinical use. J Clin Psychol. 1945;
19:87-95.
24. Lezak MD. Neuropsychological Assessment. 3rd ed. New York, NY: Oxford University Press; 1995.
25. MacLeod CM. Half a century of research on the Stroop effect: an integrative review. Psychol Bull. 1991;109:163-203.
26. Burke HR. Raven Progressive Matrices (1938): more on norms, reliability, and
validity. J Clin Psychol. 1985;41:231-235.
27. Diggle PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data. Oxford, England:
Oxford University Press; 1994.
28. Stata Statistical Software. Release 6.0. College Station, Tex: StataCorp; 1999.
29. Pennington BE, Ozonoff S. Executive functions and developmental psychopathology. J Child Psychol Psychiatry. 1996;37:51-87.
30. Barkley RA. Behavioral inhibition, sustained attention, and executive functions:
constructing a unifying theory of ADHD. Psychol Bull. 1997;121:65-94.
31. Aronowitz B, Liebowitz M, Hollander E, Fazzini E, Durlach-Misteli C, Frenkel M,
Mosovich S, Garfinkel R, Saoud J, DelBene D. Neuropsychiatric and neuropsychological findings in conduct disorder and attention-deficit hyperactivity disorder. J Neuropsychiatry Clin Neurosci. 1994;6:245-249.
32. Lueger RJ, Gill KJ. Frontal-lobe cognitive dysfunction in conduct disorder adolescents. J Clin Psychol. 1990;46:696-706.
33. Gorenstein EE. Cognitive-perceptual deficit in an alcoholism spectrum disorder.
J Stud Alcohol. 1987;48:310-318.
34. Morgan AB, Lilienfeld SO. A meta-analytic review of the relation between antisocial behavior and neuropsychological measures of executive function. Clin Psychol Rev. 2000;20:113-136.
35. Kouri EM, Pope HG Jr, Lukas SE. Changes in aggressive behavior during withdrawal from long-term marijuana use. Psychopharmacology. 1999;143:302308.
36. Haney M, Ward AS, Comer SD, Foltin RW, Fischman MW. Abstinence symptoms following smoked marijuana in humans. Psychopharmacology. 1999;141:
395-404.
37. Kouri EM, Pope HG Jr. Abstinence symptoms during withdrawal from chronic
marijuana use. Exp Clin Psychopharmacol. 2000;8:483-492.
38. Levin ED, Rezvani AH. Development of nicotinic drug therapy for cognitive disorders. Eur J Pharmacol. 2000;393:141-146.
39. Harrison ER, Haaga J, Richards T. Self-reported drug use data: what do they reveal? Am J Drug Alcohol Abuse. 1993;19:423-441.
40. Brown J, Kranzler HR, Del Boca FK. Self-reports by alcohol and drug abuse inpatients: factors affecting reliability and validity. Br J Addict. 1992;87:1013-1024.
41. Rouse BA, Kozel NJ, Richards LG, eds. Self-Report Methods of Estimating Drug
Use: Meeting Current Challenges to Validity. Washington, DC: Government Printing Office; 1985. National Institute on Drug Abuse Research Monograph 57.
WWW.ARCHGENPSYCHIATRY.COM
©2001 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ on 04/22/2014
Psychological Medicine (2009), 39, 157–167. f 2008 Cambridge University Press
doi:10.1017/S0033291708003425 Printed in the United Kingdom
O R I G I N AL A R T IC L E
Does the ‘ gateway ’ matter? Associations between
the order of drug use initiation and the development
of drug dependence in the National Comorbidity
Study Replication
L. Degenhardt1*, W. T. Chiu2, K. Conway3, L. Dierker4, M. Glantz3, A. Kalaydjian5, K. Merikangas5,
N. Sampson2, J. Swendsen 6 and R. C. Kessler2
1
National Drug and Alcohol Research Centre, University of NSW, Sydney, NSW, Australia
Department of Health Care Policy, Harvard Medical School, Boston, MA USA
3
Division of Epidemiology, Services and Prevention Research, National Institute on Drug Abuse, US National Institutes of Health,
Bethesda, MD USA
4
Department of Psychology, Wesleyan University, Middletown, CT, USA
5
National Institute of Mental Health, Bethesda, MD, USA
6
Psychopathology Research Laboratory, University of Bordeaux, Bordeaux, France
2
Background. The ‘gateway’ pattern of drug initiation describes a normative sequence, beginning with alcohol and
tobacco use, followed by cannabis, then other illicit drugs. Previous work has suggested that ‘ violations’ of this
sequence may be predictors of later problems but other determinants were not considered. We have examined the role
of pre-existing mental disorders and sociodemographics in explaining the predictive effects of violations using data
from the US National Comorbidity Survey Replication (NCS-R).
Method. The NCS-R is a nationally representative face-to-face household survey of 9282 English-speaking respondents
aged 18 years and older that used the World Health Organization (WHO) Composite International Diagnostic Interview
(CIDI) to assess DSM-IV mental and substance disorders. Drug initiation was estimated using retrospective age-of-onset
reports and ‘ violations ’ defined as inconsistent with the normative initiation order. Predictors of violations were
examined using multivariable logistic regressions. Discrete-time survival analysis was used to see whether violations
predicted progression to dependence.
Results. Gateway violations were largely unrelated to later dependence risk, with the exception of small increases
in risk of alcohol and other illicit drug dependence for those who initiated use of other illicit drugs before cannabis.
Early-onset internalizing disorders were predictors of gateway violations, and both internalizing and externalizing
disorders increased the risks of dependence among users of all drugs.
Conclusions. Drug use initiation follows a strong normative pattern, deviations from which are not strongly predictive
of later problems. By contrast, adolescents who have already developed mental health problems are at risk for deviations
from the normative sequence of drug initiation and for the development of dependence.
Received 13 September 2007 ; Revised 27 February 2008 ; Accepted 15 March 2008 ; First published online 9 May 2008
Key words : Alcohol, cannabis, dependence, gateway, illicit drugs, National Comorbidity Survey Replication, tobacco.
Introduction
There has been considerable debate about the significance of the ‘gateway effect ’. This describes a progression into polydrug use beginning with tobacco
and alcohol use, moving on to cannabis and then
‘harder ’ illicit drugs (Kandel & Faust, 1975 ; Kandel,
1984 ; Kandel et al. 1986, 1992). Debates have typically
centred on whether the predictive association between
* Address for correspondence : Professor L. Degenhardt,
National Drug and Alcohol Research Centre, University of NSW,
Sydney NSW 2052, Australia.
(Email : [email protected])
cannabis and other illicit drug use is causal or reflects
confounding factors (Kandel & Faust, 1975 ; Kandel,
1984 ; Morral et al. 2002 ; Hall & Lynskey, 2005 ;
Fergusson et al. 2006 ; Kandel & Yamaguchi, 2006 ;
MacCoun, 2006).
Irrespective of the mechanisms behind the association, the gateway pattern describes the typical
sequence of progression to greater polydrug use. Obviously, several factors affect such a sequence, including drug availability and background prevalence.
Some illicit drug use is significantly more common
among more recent birth cohorts and the available
drugs have changed. It is perhaps not surprising, then,
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
158
L. Degenhardt et al.
that the concept of gateway drugs has been applied to
ecstasy (Reid et al. 2007) and oxycodone (Grau et al.
2007) in the USA, and that a ‘reverse gateway’ has been
described for cannabis in Australia (where cannabis
use has been linked to increased risk of subsequent
initiation to tobacco use and dependence) (Patton et al.
2005).
There have been investigations of the extent and significance of violations of normative patterns. Studies
in the USA of problematic drug users (Golub &
Johnson, 1994a, b, 2002 ; Mackesy-Amiti et al. 1997) and
homeless youths (Ginzler et al. 2003) have found that
significant proportions had not progressed through the
typical pattern of progression, with many beginning
cannabis use before they had first used alcohol,
and some starting other illicit drug use before using
alcohol or cannabis. In those studies, individuals with
‘atypical ’ patterns of progression were found to come
from more disadvantaged backgrounds (MackesyAmiti et al. 1997), be from different birth cohorts
(Golub & Johnson, 1994 a, b ; Mackesy-Amiti et al. 1997),
and be heavier polydrug users (Mackesy-Amiti et al.
1997 ; Ginzler et al. 2003) than users who followed the
normative progression.
This suggests that violations of normative patterns
of progression may be important markers of subsequent risk of progression. The above studies provided
interesting data, yet were in most cases limited to unrepresentative samples of heavy drug users ; typically
presented limited bivariate associations with other
characteristics ; did not adjust for pre-morbid mental
health or demographic factors that might have been
related to progression ; and did not consider the impact of such atypical progressions for the later development of dependence. In this paper, we consider all
of these possibilities using data from a representative
sample of the US adult population, from the National
Comorbidity Survey Replication (NCS-R).
Method
Participants and study procedures
As described in detail elsewhere (Kessler &
Merikangas, 2004), the NCS-R is a nationally representative household survey of English speakers
aged o18 years in the contiguous USA. Respondents
were selected from a multistage clustered area probability sample of households and face-to-face interviews carried out from February 2001 to April 2003 by
professional interviewers from the Institute for Social
Research at the University of Michigan (U-M). The
response rate was 71 %. The survey was administered
in two parts. Part 1 included a core diagnostic assessment (n=9282). Part 2 included assessed risk factors,
consequences, correlates, and assessments of additional disorders that were administered to all Part 1
respondents who met lifetime criteria for any disorder
plus a probability subsample of other respondents
(n=5692). Interviewers explained the study and
obtained verbal informed consent prior to beginning
the survey. Recruitment, consent and field procedures
were approved by the Human Subjects Committees of
Harvard Medical School and U-M.
Diagnostic assessment
Drug use modules
Drug use modules in the Part II sample were administered following a positive response to screening
questions inquiring whether the respondent had ever
used (1) tobacco (cigarettes, cigar or pipe) ; (2) alcohol ;
(3) cannabis, hashish ; (4) cocaine ; (5) tranquillizers,
stimulants, painkillers or other prescription drugs ;
or (6) any other illicit drug including heroin, opium,
glue, LSD or peyote. Detailed analyses of drug use and
associations with demographic variables from this
dataset have been reported previously (Degenhardt
et al. 2007c).
Assessments of DSM-IV mental and substance use
disorders were based on responses to the World
Health Organization (WHO) Composite International
Diagnostic Interview (CIDI ; Kessler & Ustun, 2004), a
fully structured lay-administered diagnostic interview
used to generate DSM-IV diagnoses.
Drug use disorders
Any positive responses to drug use were followed
with a detailed assessment of lifetime use of that drug,
including age of onset of use, progression, and symptoms of abuse and dependence. Assessment of dependence was conducted separately for tobacco and
alcohol. For other drugs, assessment of dependence
was carried out with participants responding to dependence symptoms attributed to any of the drugs
they reported having used. This is consistent with the
DSM category for ‘dependence not otherwise specified’, whereby a person may meet criterion A1 for
cannabis, A2 for cocaine and A3–4 for yet another
drug, but does not meet full criteria for dependence on
any single drug ; they would nonetheless be classified
as meeting criteria for ‘drug dependence ’. Included
here are people who meet full criteria for dependence,
and where the symptoms are associated with the use
of either one particular drug or multiple drugs.
This method of assessment of drug dependence was
the same as that used in the Epidemiological Catchment Area (ECA) study and the National Comorbidity
Survey (NCS). Good concordance has been reported in
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Associations between order of initiation of drug use and drug dependence
an NCS-R clinical reappraisal subsample between
diagnoses of substance use disorders based on the
CIDI and diagnoses based on blinded clinical reappraisal interviews (Kessler et al. 2004 a) using the
Structured Clinical Interview for DSM-IV (SCID ; First
et al. 1996).
DSM-IV internalizing disorders
These included specific phobia, social phobia, panic
disorder, agoraphobia with panic disorder, generalized anxiety disorder with hierarchy, post-traumatic
stress disorder, and major depressive disorder with
hierarchy or dysthymia with hierarchy. They were
aggregated into a summary variable reflecting the
number of internalizing disorders that were reported
to have occurred as of the age of 15 (range 0–7).
DSM-IV externalizing disorders
These included bipolar disorder, oppositional-defiant
disorder with hierarchy, conduct disorder, attention
deficit hyperactivity disorder, and intermittent explosive disorder with hierarchy. They were aggregated
to reflect the number that had occurred as of the age
of 15 (range 0–5). Assessment of the disorders requiring childhood onset of symptoms (separation anxiety
disorder, oppositional-defiant disorder, conduct disorder, attention deficit hyperactivity disorder) was
limited to those under 45 years at the time of interview
to reduce recall bias.
Order of onset and violations of the ‘ typical ’
gateway progression
Different onset orders, as determined by retrospective
age-of-onset reports, were evaluated separately. The
violations were :
(1) the use of cannabis prior to both alcohol and
tobacco use ;
(2) other illicit drug use prior to both alcohol and
tobacco use ;
(3) other illicit drug use prior to cannabis use.
Initiation of cannabis and/or other illicit drug use (a)
prior to alcohol use (but not tobacco) and (b) prior to
tobacco use (but not alcohol) was considered. These
were post hoc and, given they are not ‘ true’ violations
of the gateway sequence, were not considered in
further analyses.
Statistical analyses
Weights were used to adjust for variation in Part II
probabilities described earlier, as well as withinhousehold probability of selection, non-response,
and differences between the sample and the 2000
159
Census on sociodemographic variables. Further detail
has been provided in previous work (Kessler et al.
2004b).
Cumulative incidence proportions of gateway violations were estimated, with standard errors derived
using the Taylor series linearization (TSL) methods
implemented in SUDAAN version 9 (SAS Institute,
Cary, NC, USA) to adjust for the effects of weighting
and clustering on the precision of estimates. Regression
coefficients were estimated and then exponentiated
for interpretation as odds ratios (ORs). When p
values are reported or indicated ( by an asterisk), they
are from Wald tests obtained from TSL design-based
coefficient variance–covariance matrices (a=0.05, twotailed).
Regression analysis was carried out to examine the
association with age, sex and early-onset mental disorders with gateway ‘ violations ’ among users of each
drug type. Predictors of gateway violations among
users of each drug were examined using multivariable
logistic regression models.
Discrete-time survival models among users of a drug
examined predictors of dependence onset. Predictors
included sex, age cohort (defined by age at interview :
18–29, 30–44, 45–59, o60 years), number of externalizing and internalizing disorders by age 15, age of
onset of use of the drug concerned, years since first
onset of use (a time-varying covariate), a variable indicating whether there was a gateway violation (three
dummy variables defined as outlined above), tobacco
use (a time-varying covariate), alcohol use (a timevarying covariate), and the number of other drugs
used (a time-varying covariate). The resulting ORs
represent the estimates of risk of first-onset dependence in a given year.
Results
Overall, 5.2 % of participants initiated substance use in
an order that violated the gateway sequence (Table 1).
The most common violation was initiation of other illicit drugs before cannabis (3.7 %), followed by cannabis use before alcohol and tobacco use (1.6 %).
Prevalence differed significantly across birth cohorts.
Respondents in the o60 years group were extremely
unlikely to report illicit drug use before alcohol and
tobacco, whereas the three younger age groups were
more likely to do so.
Table 2 specifies the types of illicit drugs used before alcohol and tobacco among those who violated
the gateway sequence. Cannabis was the most common drug initiated before that time (69.2 % of the
group). Cocaine was more commonly initiated prior to
alcohol and tobacco for the 18–29 years age group
(18.7 %) compared to older groups.
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
130.9*
955
0.6
14.5
21
0.5
1.6
223
1.2
13.3
437
1.6
21.0
Standard error.
A statistically reliable estimate could not be made.
* Significant at 0.01 level, two-tailed test.
S.E.,
a
1.5
19.6
274
163
471
94
31
161
57
252
341
0.2
0.4
0.3
0.1
0.3
0.1
0.2
0.3
2.5
7.0
1.6
0.4
2.2
0.8
3.7
5.2
–
–
–
0.3
–
–
0.5
0.5
1.0
1.1
0.7
0.1
0.4
0.3
0.6
0.9
4.7
9.5
2.0
0.2
2.0
1.2
3.9
5.9
01. Cannabis before alcohol, not tobacco
02. Cannabis before tobacco, not alcohol
03. Cannabis before alcohol and tobacco
04. Other illicit drugs before alcohol, not tobacco
05. Other illicit drugs before tobacco, not alcohol
06. Other illicit drugs before alcohol and tobacco
07. Other illicit drugs before cannabis
Any violation of the gateway order of
initiation (03, 06, 07)
Any of 01 to 07
51
144
23
5
35
18
64
87
3.6
10.2
3.0
0.7
3.7
1.1
4.6
7.4
0.5
1.0
0.8
0.2
0.5
0.3
0.6
1.0
87
222
50
16
86
25
91
137
1.1
6.8
1.1
0.3
2.6
0.9
4.4
5.4
0.3
0.9
0.2
0.1
0.7
0.3
0.6
0.7
24
102
20
7
40
14
81
100
–a
–a
–a
0.5
–a
–a
1.3
1.3
1
3
1
3
0
0
16
17
n
S.E.
%
n
S.E.
%
n
S.E.
n
%
Order of onset of use
S.E.
%
S.E.
n
%
Total (n=5692)
o60 yr (n=974)
45–59 yr (n=1521)
30–44 yr (n=1826)
18–29 yr (n=1371)
Table 1. Distribution of each violation of the gateway pattern of drug use initiation by age cohort. Data from the National Comorbidity Survey Replication (NCS-R), 2001–2003
88.2*
138.2*
45.7*
5.6
75.9*
50.9*
22.9*
35.3*
L. Degenhardt et al.
Significance
160
Table 3 presents the results of regressions examining predictors of gateway violations. Sex was not related to the initiation of illicit drug use prior to both
alcohol and tobacco, but was related to initiation of
other illicit drugs prior to cannabis, with females less
likely than males to have done so. Age was strongly
related to violations of all three kinds, with younger
age groups significantly more likely than the oldest
age group to have initiated substance use out of the
gateway sequence.
Mental disorders by age 15 years were unrelated to
the precocious initiation of cannabis use (i.e. before
alcohol and tobacco use). Internalizing disorders were
related to precocious initiation of other illicits (defined
as cocaine, sedatives/stimulants/analgesics or other
drugs including heroin). With each additional internalizing disorder, the likelihood of initiating such
drug use before alcohol and tobacco increased by 40 %
on average [OR 1.4, 95 % confidence interval (CI)
1.1–1.8], and of initiating such drug use before cannabis use by 50 % on average (OR 1.5, 95 % CI 1.2–1.8).
Externalizing disorders by 15 years were unrelated to
initiation order.
Table 4 shows the results of multivariable survival
analyses examining the risk of incident dependence
among users of each drug. When other factors were
controlled, gateway violations were unrelated to the
risk of developing nicotine dependence, or drug
dependence among cannabis and cocaine users.
Initiation of any other illicit drugs (cocaine, sedatives/
stimulants/analgesics or other drugs) before cannabis
use was significantly related to the risk of incident
alcohol dependence among alcohol users (OR 1.5, 95 %
CI 1.0–2.2), and drug dependence among sedative/
stimulant/analgesic/other drug users (OR 2.3, 95 % CI
1.4–3.9).
Consistently significant predictors of transitioning
to dependent use in a given year were : earlier age of
onset of use, recency since onset of use, and the extent
of illicit drug use to date. Further analyses were conducted to evaluate the possibility that precocious initiation into illicit drug use might also reflect greater
polydrug use, such that gateway violations were related to the number of drug types used. Additional
analyses were conducted without controlling for
the number of illicit drugs used by that age (see
Appendix). In almost all cases, there was no difference
in the significance of the observed associations. Two
notable exceptions were the risk of incident nicotine
dependence among tobacco users, where initiation of
cannabis use prior to tobacco/alcohol use predicted
incident nicotine dependence, and dependence among
cannabis users, where initiation of other illicit drugs
prior to cannabis predicted incident dependence. In
both cases, inclusion of the number of illicit drugs
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Associations between order of initiation of drug use and drug dependence
161
Table 2. Drugs used among those who had used any illicit drugs prior to alcohol and tobacco, by age. Data from the National
Comorbidity Survey Replication (NCS-R), 2001–2003
Among those who had begun using any illicit drugs before alcohol and tobacco (n=141),
those who had already used…
Cannabis
Age group
(years)
n
Other illicit drugsa
Cocaine
%
S.E.
%
S.E.
%
S.E.
18–29
30–44
45–59
o60
39
68
33
1
64.9
77.0
57.1
–b
10.0
7.9
7.8
–
18.5
3.9
0.0
–b
7.7
2.0
0.0
–
24.7
26.9
45.9
–b
7.9
8.0
8.5
–
Total
141
69.2
5.0
7.5
2.7
30.2
5.4
S.E.,
a
Standard error.
Includes sedatives/stimulants/analgesics and any other drugs.
b
A statistically reliable estimate could not be made.
Table 3. Multivariable predictors of violation of the gateway sequence of drug use initiation. Data from the National Comorbidity
Survey Replication (NCS-R), 2001–2003
Cannabis before both
alcohol and tobacco
aOR (95 % CI)
Female
Age at interview (years)
18–29
30–44
45–59
o60
No. internalizing disorders by 15 yearsa
No. externalizing disorders by 15 yearsb
1.1 (0.6–2.1)
53.8* (6.3–459.7)
82.4* (10.2–667.1)
29.4* (3.7–232.5)
1
1.2 (0.8–1.7)
1.0 (0.6–1.6)
Other illicit drugs
before both alcohol
and tobacco
aOR (95 % CI)
1.1 (0.5–2.1)
5457.0* (3255.8–9146.5)
5115.8* (2726.3–9599.8)
4130.1* (2064.8–8261.5)
1
1.4* (1.1–1.8)
1.1 (0.7–1.6)
Other illicit drugs
before cannabis
aOR (95 % CI)
0.7* (0.5–0.9)
2.7* (1.1–6.3)
3.2* (1.3–7.4)
3.2* (1.4–7.4)
1
1.5* (1.2–1.8)
1.1 (0.8–1.4)
aOR, Adjusted odds ratio ; CI, confidence interval.
Results based upon multivariable logistic regression models.
‘Other illicit drugs’ include cocaine, opioids, analgesics, sedatives, and ‘other drugs’.
a
DSM-IV internalizing disorders included : panic disorder, agoraphobia without panic disorder, social phobia, specific
phobia, generalized anxiety disorder with hierarchy, post-traumatic stress disorder, and major depressive disorder with
hierarchy/dysthymia with hierarchy.
b
DSM-IV externalizing disorders included : bipolar disorder, oppositional-defiant disorder with hierarchy, conduct disorder,
attention deficit hyperactivity disorder and intermittent explosive disorder with hierarchy.
* OR significant at 0.05 level, two-tailed test. x2 statistics are available upon request.
used made this association non-significant, suggesting
that violation of the gateway order of onset in these
cases was related to a higher likelihood of using
a greater number of illicit drugs, and also related to
incident dependence.
Finally, a greater degree of psychiatric co-morbidity
by 15 years was associated with risk of incident
dependence. The odds of users transitioning to
dependent use increased by 20 % (nicotine) to 50 %
(alcohol, other drugs) with each additional internalizing disorder by 15 years ; and similarly with each additional externalizing disorder (20 % for nicotine to
60 % for alcohol). As noted in Table 3, pre-existing internalizing disorders were also significant predictors
of gateway violations, meaning that failure to control
for these disorders would allow a spuriously positive
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
No. internalizing disorders by 15 yearsa
No. externalizing disorders by 15 yearsb
Age of onset of usec
Years since first onset of usec
Tobacco use
Alcohol use
Number of illegal drugs usedd
None
1
2
3
4
‘Gateway violation’
Cannabis use before tobacco and alcohol
Other illicit drugs before tobacco and alcohole
Other illicit drugs before cannabise
Alcohol dependence
among alcohol users
Tobacco dependence
among tobacco users
Drug dependence
among cannabis users
Drug dependence
among cocaine users
Drug dependence
among other drug userf
aOR
95 % CI
aOR
95 % CI
aOR
95 % CI
aOR
95 % CI
aOR
95 % CI
0.5*
0.4–0.7
1.1
0.9–1.2
0.9
0.6–1.2
1.1
0.7–1.9
1.0
0.7–1.5
1.0
0.6
0.9
1
0.5–1.8
0.3–1.0
0.5–1.6
1.8*
0.8
1.0
1
1.4–2.4
0.7–1.0
0.8–1.2
0.5
0.4
0.5
1
0.1–2.5
0.1–2.0
0.1–2.3
0.4
0.3
0.3
1
0.1–2.4
0.1–1.9
0.0–1.8
0.7
0.6
0.6
1
0.1–4.2
0.1–3.5
0.1–3.4
1.5*
1.6*
0.9*
0.4*
2.5*
1.4–1.6
1.4–1.8
0.9–0.9
0.3–0.6
1.8–3.4
1.2*
1.2*
0.7*
1.0*
1.2–1.3
1.1–1.4
0.6–0.9
1.0–1.0
2.5*
1.9–3.4
1.4*
1.4*
0.4*
0.8*
2.0*
1.6
1.3–1.6
1.2–1.7
0.2–0.7
0.8–0.9
1.1–3.9
0.5–5.4
1.4*
1.4*
0.5*
0.8*
1.3
3.0
1.2–1.7
1.1–1.7
0.3–1.0
0.7–0.9
0.6–2.7
0.3–35.1
1.5*
1.3*
0.5*
0.8*
1.8
5.4
1.2–1.7
1.1–1.6
0.3–0.9
0.8–0.8
0.9–3.6
0.9–32.0
1
3.1*
6.5*
5.8*
10.4*
2.1–4.5
3.6–11.9
3.2–10.6
5.5–19.8
1
1.9*
2.6*
3.5*
3.8*
1.6–2.2
2.2–3.1
2.8–4.5
2.8–5.3
1
6.9*
13.3*
33.7*
4.2–11.3
8.3–21.4
18.8–60.5
1
0.7
1.0
2.4
0.1–4.2
0.2–4.5
0.6–10.6
1
3.3*
7.0*
18.2*
1.1–9.9
2.2–22.1
5.8–57.6
0.7
0.4*
1.5*
0.3–1.5
0.1–1.0
1.0–2.2
1.3
0.9
0.9
0.8–2.0
0.4–1.7
0.6–1.2
1.2
0.7
1.4
0.5–2.8
0.2–1.9
0.8–2.3
0.8
0.6
1.7
0.3–2.3
0.2–2.4
0.7–3.9
1.1
1.5
2.3*
0.4–2.8
0.4–5.6
1.4–3.9
aOR, Adjusted odds ratio ; CI, confidence interval.
Results based upon multivariable discrete time survival models.
‘ Onset of dependence’ refers to onset of the full dependence syndrome.
a
DSM-IV internalizing disorders included : panic disorder, agoraphobia without panic disorder, social phobia, specific phobia, generalized anxiety disorder with hierarchy,
post-traumatic stress disorder, and major depressive disorder with hierarchy/dysthymia with hierarchy.
b
DSM-IV externalizing disorders included : bipolar disorder, oppositional-defiant disorder with hierarchy, conduct disorder, attention deficit hyperactivity disorder and intermittent
explosive disorder with hierarchy.
c
Age of onset, or years since onset, of the drug use concerned.
d
This is a time-varying covariate and refers to the number of illicit drugs (grouped as cannabis, cocaine, sedatives/stimulants/analgesics, or ‘other’) the person had used by a given
year.
e
Other illicit drugs includes any of cocaine, sedatives/stimulants/analgesics, or ‘other’.
f
‘Other drugs’ includes sedatives/stimulants/analgesics and ‘other’.
* OR significant at 0.05 level, two-tailed test. x2 statistics are available upon request.
L. Degenhardt et al.
Female
Age at interview (years)
18–29
30–44
45–59
o60
162
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Table 4. Multivariable predictors of onset of dependence by drug type. Data from the National Comorbidity Survey Replication (NCS-R), 2001–2003
Associations between order of initiation of drug use and drug dependence
association between gateway violations and subsequent dependence.
Discussion
This study examined the order of onset of drug use,
and considered the possible association between deviations from the normative (gateway) pattern of drug
progression with subsequent onset of substance dependence in a representative sample of US adults.
Three violations were examined : (a) cannabis use
before alcohol and tobacco ; (b) other illicit drug use
before alcohol and tobacco, and (c) other illicit drug
use before cannabis. Importantly, using a person-years
framework, this study was able to consider the risk
of first developing dependent use across each year of
life for the participants in this study. In doing so, we
could also control at each year for the age of onset of
that drug use ; time since initiation of such use ; the participant’s lifetime-to-date use of other drugs ; and comorbid mental health problems developed by early
adolescence. This approach to the analysis of gateway
patterns and their predictive associations with subsequent dependent use represents a significant advance,
as previous studies of this issue have concentrated on
unrepresentative samples of problematic drug users,
with insufficient capacity to conduct detailed investigations of risk for problems while controlling for
the important confounding variables considered here.
Deviations from the gateway order of onset were
found to occur only for a minority of persons (5.2 %).
The most common violation was other illicit drug use
before cannabis (3.7 %), and the least common was
other illicit drugs before both alcohol and tobacco use
(0.8 %). There were some strong cohort differences in
the likelihood of these violations : they were less common among the oldest age group than the younger
ones. These findings are consistent with historical
trends in drug use ; cannabis use is much more common in more recent birth cohorts (Degenhardt et al.
2000 ; Johnston et al. 2003), so it is not surprising that
cannabis is also more likely to occur earlier in the sequence of drug use for some younger people.
Previous studies have found that, among disadvantaged samples of drug users, many of whom
had co-morbid mental health problems, violations of
the gateway order of initiation involving precocious
initiation into illicit drug use (such as cocaine use very
early on in their drug use career) were common
(Golub & Johnson, 1994a, b, 2002 ; Mackesy-Amiti et al.
1997). The current study demonstrated that one significant predictor of such deviations was the early
development of internalizing mental disorders such as
depression, post-traumatic stress disorder, social
phobia or generalized anxiety disorder. This suggests
163
that pre-morbid mental health problems are related to
precocious initiation of illicit drug use.
This same deviation, the use of other illicit drugs
(cocaine, sedatives, stimulants, opioids or other drugs)
before cannabis use, was the only one significantly
associated with the risk of subsequently developing
dependent use. Among cocaine and other illicit drug
users, risk for dependent use was elevated among
those who had initiated use of these drugs before
cannabis use and was significant after controlling for
important potential mediators of dependence risk and
common causes of the violation and dependence. This
finding is consistent with the finding in studies of
persons who have developed serious illicit drug use
problems that high rates of atypical patterns of progression through stages of drug use exist in such
samples, usually involving initiation of illicit drugs
before cannabis or other drug use (Golub & Johnson,
1994a, b, 2002 ; Mackesy-Amiti et al. 1997).
Why do violations of normative patterns of illicit
drug use onset play some part in the development of
drug dependence, but others do not ? This is the first
study that has investigated this issue using a survival
analytic framework, so few comparable data exist.
One rather obvious possibility is that deviations from
normative patterns matter much more for drugs that
are infrequently used than for drugs that are in themselves much more normative to use. Thus, alcohol,
tobacco and cannabis are by far the most frequently
used drugs in the USA ; by comparison, cocaine and
other illicit drugs are used by far fewer people
(Anthony et al. 1994 ; Johnston et al. 2003 ; Degenhardt
et al. 2007c). This supports the view that the significance of a gateway sequence is not related to a particular order of the initiation of particular drugs, but
rather to a reflection of relative social or psychiatric
deviance, and perhaps a pattern of escalating deviance.
A second possibility is that the violation documented here, the onset of cocaine or other illicit drug
use before cannabis use, reflects a greater and earlier
prominence of these drugs earlier in the user’s drug
history, irrespective of the age of onset of use. The
multitude of studies examining the risks of early-onset
cannabis use have never been able to tease apart the
possible contributions of the primacy of this drug in
many people’s illicit drug use careers. The fact that
cannabis typically begins first makes it difficult to
know whether associations of early-onset cannabis use
with later drug use problems reflect the order of onset
or a specific drug effect (Degenhardt et al. 2007d). The
findings of the current study suggest that both the type
and order of onset of drug use may be influential in
conferring risk upon the development of dependent
use.
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
164
L. Degenhardt et al.
Finally, it is very plausible that gateway violations
reflect important individual characteristics. Young
people who choose to use drugs are more likely to be
impulsive and take risks ; the gateway violation that
was a significant marker of dependence risk here was
that which involved premature entry into illicit drug
use. The finding that violations reflecting precocious
entry into drug use were associated with elevated risks
for later dependence would be consistent with the
possibility that violation of gateway patterns reflects a
broader underlying vulnerability to drug problems. It
also suggests that the nature of this gateway sequence
does not matter ; it is a description of a normative
sequence of entry into drug use that differs across
countries and time (Patton et al. 2005 ; Grau et al. 2007 ;
Reid et al. 2007), violations of which (or adherence to)
reflect other factors, including individual characteristics (Shedler & Block, 1990 ; Morral et al. 2002), that
may ultimately matter more for the development of
dependence.
Mental health appeared to be important for both the
order of initiation of illicit drug use and particularly
for the development of dependent use once use had
begun. In this study, those who had early-onset (by age
15 years) internalizing disorders were more likely to
deviate from the normative order of onset of illicit
drug use. Early-onset mental disorders, early-onset
drug use and more extensive polydrug use were all
important moderators of risk for developing dependent use, and were more important risk factors than
violations of the ‘normative ’ order of onset of drug
use.
The finding that adolescents with both externalizing
and internalizing disorders were at elevated risk of
developing drug use problems later in life if they began using such drugs is consistent with prospective
cohort studies, which have found that early-onset
drug use and mental health problems are risk factors
for later dependent drug use (Toumbourou et al. 2007),
and that mental health problems escalate risk of developing dependent use. Detailed investigation of the
specific mental disorders related to drug dependence
was beyond the scope of the current paper, but further
work is under way to investigate in more detail the
nature of these co-morbidities, particularly to tease
apart possible differences across different internalizing and externalizing disorders in their importance for
predicting incident substance dependence.
There are clear public health and clinical implications, nonetheless, of the broad findings documented
here. Adolescents with mental health problems are a
particular risk group for the development of dependent use should they begin using legal or illegal
drugs. Preventive interventions that address multiple
areas of risk for both drug use and mental health
among young people, including family social
disadvantage, early school engagement and social
inclusion, are effective (Patel et al. 2007 ; Toumbourou
et al. 2007).
Limitations
Any cross-sectional retrospective survey research
has limitations (Wu et al. 2003). Some of the observed
cohort differences might be traced to higher mortality
among individuals in the older cohorts who began
drug use at an early age. Nonetheless, we believe that
differential mortality is unlikely to explain the fairly
large differences in cumulative incidence for illegal
drug use across adjacent age groups given that mortality associated with cannabis use is highly unlikely
to be substantial (Hall et al. 2001). Conversely, the
evidence of tobacco-related premature mortality is
substantial, but tobacco use showed the least prominent age-associated variation.
Retrospective reporting of age of first drug use may
be subject to error, given that respondents are being
asked about events that, for older persons, may have
occurred decades ago. Although it is likely that some
proportion of participant reports contained an element
of recall bias, longitudinal studies of adolescents have
found that estimates of the age of first use do tend to
increase upon repeat assessment (i.e. as people age),
but the rank ordering for different drugs does not
change (Henry et al. 1994 ; Engels et al. 1997 ; Labouvie
et al. 1997).
One possible limitation of the study relates to potential underestimation of dependence because the
NCS-R used a ‘gated ’ assessment of dependence,
whereby dependence was only assessed among those
who met criteria for abuse. We examined the impact of
a ‘gated ’ assessment approach upon alcohol, cannabis
and illicit drug dependence prevalence estimates in
the USA (Degenhardt et al. 2007a, b, 2008). We found a
very modest attenuation of the prevalence of past year
cannabis dependence (0.26 % v. 0.32 %), but not for
cannabis use disorders (Degenhardt et al. 2007b) ; the
reduction was greater for alcohol dependence (2.5 %
v. 3.8 %) (Degenhardt et al. 2007a). There was no appreciable reduction of cocaine dependence prevalence
estimates, and for other drugs estimates were so
low that there was insufficient power to detect any
difference at a general population level, even with
a sample of over 40 000 persons (Degenhardt et al.
2008). Relationships with demographic variables of
interest remained remarkably consistent across the
gated and ungated assessment approaches, suggesting
that any attenuation of estimated prevalence was not
strongly concentrated within certain subpopulations
(Degenhardt et al. 2007a, b, 2008).
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Associations between order of initiation of drug use and drug dependence
Conclusions
Deviations from normative patterns of drug use initiation that involve the initiation of illicit drug use
earlier than usual in the gateway pattern of initiation
may carry small risks for dependence, but other factors seem to be more important in the development of
drug dependence. Drug use and initiation are clearly
nested within a social normative context, yet neither
adherence nor deviation from this order signals highly
elevated risks of drug problems in and of themselves,
although some violations are predicted by pre-existing
mental disorders that seem to be more powerful
risk factors for subsequent substance dependence.
Although a gateway violation might be a marker of
such risk factors, their associations with gateway violations are relatively modest. In targeting intervention
efforts, it would probably be more productive to
screen directly for these factors (i.e. internalizing disorders, early-onset substance use) than to screen for
gateway violations.
Acknowledgements
The US National Comorbidity Survey Replication
(NCS-R) is supported by the National Institute of
Mental Health (NIMH ; U01-MH60220) with supplemental support from the National Institute of Drug
Abuse (NIDA), the Substance Abuse and Mental
Health Services Administration (SAMHSA), the Robert
Wood Johnson Foundation (RWJF ; Grant 044708), and
the John W. Alden Trust. Data analyses and manuscript preparation were undertaken with support from
an Intramural Research Program of the National
Institutes of Health, NIMH (Kalaydjian, Merikangas),
grants K01 DA15454 and R01 DA012058 from the
NIDA (Dierker), and an Investigator Award from the
Patrick and Catherine Weldon Donaghue Medical
Research Foundation (Dierker). The National Drug
and Alcohol Research Centre is funded by the
Australian Government Department of Health and
Ageing (Degenhardt), and Degenhardt is the recipient
of an Australian National Health and Medical
Research Council (NH&MRC) Senior Research
Fellowship. The NCS-R is carried out in conjunction
with the World Health Organization World Mental
Health (WMH) Survey Initiative. We thank the WMH
staff for assistance with instrumentation, fieldwork
and data analysis. These activities were supported
by the US NIMH (R01MH070884), the John D.
and Catherine T. MacArthur Foundation, the Pfizer
Foundation, the US Public Health Service (R13MH066849, R01-MH069864, and R01 DA016558), the
Fogarty International Center (FIRCA R01-TW006481),
the Pan American Health Organization, Eli Lilly and
165
Company, Ortho-McNeil Pharmaceutical, Inc.,
GlaxoSmithKline, and Bristol–Myers Squibb. A complete list of WMH publications can be found at
www.hcp.med.harvard.edu/wmh/.
Declaration of Interest
Professor Kessler has been a consultant for Astra
Zeneca, Bristol–Myers Squibb, Eli Lilly and Co.,
GlaxoSmithKline, Pfizer, Sanofi-Aventis, and Wyeth
and has had research support for his epidemiological
studies from Bristol–Myers Squibb, Eli Lilly and
Company, Ortho-McNeil, Pfizer, and the Pfizer
Foundation. Professor Degenhardt has received an
untied educational grant from Reckitt Benckiser to
examine the diversion and injection of opioid substitution treatment in Australia.
References
Anthony JC, Warner L, Kessler R (1994). Comparative
epidemiology of dependence on tobacco, alcohol,
controlled substances, and inhalants : basic findings from
the National Comorbidity Survey. Experimental and Clinical
Psychopharmacology 2, 244–268.
Degenhardt L, Bohnert KM, Anthony J (2007 a). Case
ascertainment of alcohol dependence in general
population surveys : ‘ gated’ versus ‘ungated’ approaches.
International Journal of Methods in Psychiatric Research 16,
111–123.
Degenhardt L, Bohnert KM, Anthony J (2008). Assessment
of cocaine and other drug dependence in the general
population : ‘ hated’ versus ‘ ungated’ approaches.
Drug and Alcohol Dependence 93, 227–232.
Degenhardt L, Cheng H, Anthony J (2007b). Assessing
cannabis dependence in community surveys :
methodological issues. International Journal of Methods in
Psychiatric Research 16, 43–51.
Degenhardt L, Chu W-T, Sampson N, Kessler RC, Anthony
JC (2007c). Epidemiological patterns of extra-medical drug
use in the United States : evidence from the National
Comorbidity Survey Replication, 2001–2003. Drug and
Alcohol Dependence 90, 210–223.
Degenhardt L, Coffey C, Moran P, Carlin JB, Patton GC
(2007d). The predictors and consequences of adolescent
amphetamine use : findings from the Victorian Adolescent
Health Cohort Study. Addiction 102, 1076–1084.
Degenhardt L, Lynskey M, Hall W (2000). Cohort
trends in the age of initiation of drug use in Australia.
Australian and New Zealand Journal of Public Health 24,
421–426.
Engels R, Knibbe R, Drop M (1997). Inconsistencies in
adolescents’ self-reports of initiation of alcohol and tobacco
use. Addictive Behaviors 22, 613–623.
Fergusson D, Boden J, Horwood LJ (2006). Cannabis use
and other illicit drug use : testing the cannabis gateway
hypothesis. Addiction 101, 556–569.
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
166
L. Degenhardt et al.
First MB, Spitzer RL, Gibbon M, Williams JB (1996).
Structured Clinical Interview for DSM-IV Axis I Disorders,
Clinician Version (SCID-CV). American Psychiatric Press :
Washington, DC.
Ginzler J, Cochran B, Domenech-Rodriguez M, Cauce A,
Whitbeck L (2003). Sequential progression of substance
use among homeless youth : an empirical investigation
of the gateway theory. Substance Use and Misuse 38,
725–758.
Golub A, Johnson BD (1994a). Cohort differences in
drug-use pathways to crack among current crack
abusers in New York City. Criminal Justice and Behavior
21, 403–422.
Golub A, Johnson BD (1994b). The shifting importance of
alcohol and marijuana as gateway substances among
serious drug abusers. Journal of Studies on Alcohol 55,
607–614.
Golub A, Johnson BD (2002). The misuse of the ‘Gateway
Theory’ in US policy on drug abuse control : a secondary
analysis of the muddled deduction. International Journal of
Drug Policy 13, 5–19.
Grau L, Dasgupta N, Harvey A, Irwin K, Givens A, Kinzly
M, Heimer R (2007). Illicit use of opioids : is OxyContin a
‘ gateway drug’? American Journal on Addictions 16,
166–173.
Hall W, Degenhardt L, Lynskey M (2001). The Health and
Psychological Consequences of Cannabis Use. Australian
Publishing Service : Canberra.
Hall W, Lynskey M (2005). Is cannabis a gateway drug?
Testing hypotheses about the relationship between
cannabis use and the use of other illicit drugs. Drug and
Alcohol Review 24, 39–48.
Henry B, Moffitt T, Caspi A, Langley J, Silva P (1994).
On the ‘ remembrance of things past’ : a longitudinal
evaluation of the retrospective method. Psychological
Assessment 6, 92–101.
Johnston LD, O’Malley PM, Bachman JG (2003). National
Survey Results on Drug Use from the Monitoring the Future
Study, 1975–2003. National Institute on Drug Abuse :
Rockville, MD.
Kandel D, Davies M, Karus D, Yamaguchi K (1986).
The consequences in young adulthood of adolescent
drug involvement. Archives of General Psychiatry 43,
746–754.
Kandel D, Faust R (1975). Sequence and stages in patterns of
adolescent drug use. Archives of General Psychiatry 32,
923–932.
Kandel D, Yamaguchi K (2006). Testing the gateway
hypothesis. Addiction 101, 470–472.
Kandel DB (1984). Marijuana users in young adulthood.
Archives of General Psychiatry 41, 200–209.
Kandel DB, Yamaguchi K, Chen K (1992). Stages
of progression in drug involvement from
adolescence to adulthood : further evidence
for the gateway theory. Journal of Studies on Alcohol 53,
447–457.
Kessler RC, Abelson J, Demler O, Escobar JI, Gibbon M,
Guyer ME, Howes MJ, Jin R, Vega WA, Walters EE,
Wang P, Zaslavsky A, Zheng H (2004a). Clinical
calibration of DSM-IV diagnoses in the World Mental
Health (WMH) version of the World Health Organization
(WHO) Composite International Diagnostic Interview
(WMHCIDI). International Journal of Methods in Psychiatric
Research 13, 122–139.
Kessler RC, Berglund P, Chiu WT, Demler O, Heeringa S,
Hiripi E, Jin R, Pennell BE, Walters EE, Zaslavsky A,
Zheng H (2004b). The US National Comorbidity Survey
Replication (NCS-R) : design and field procedures.
International Journal of Methods in Psychiatric Research 13,
69–92.
Kessler RC, Merikangas KR (2004). The National
Comorbidity Survey Replication (NCS-R) : background
and aims. International Journal of Methods in Psychiatric
Research 13, 60–68.
Kessler RC, Ustun TB (2004). The World Mental Health
(WMH) Survey Initiative Version of the World Health
Organization (WHO) Composite International Diagnostic
Interview (CIDI). International Journal of Methods in
Psychiatric Research 13, 93–121.
Labouvie E, Bates M, Pandina R (1997). Age of first use : its
reliability and predictive utility. Journal of Studies on Alcohol
58, 638–643.
MacCoun R (2006). Competing accounts of the gateway
effect : the field thins, but still no clear winner. Addiction
101, 473–474.
Mackesy-Amiti ME, Fendrich M, Goldstein PJ (1997).
Sequence of drug use among serious drug users : typical vs
atypical progression. Drug and Alcohol Dependence 45,
185–196.
Morral A, McCaffrey D, Paddock S (2002). Reassessing the
marijuana gateway effect. Addiction 97, 1493–1504.
Patel V, Fisher AJ, Hetrick S, McGorry P (2007). Mental
health of young people : a global public-health challenge.
Lancet 369, 1302–1313.
Patton G, Coffey C, Carlin J, Sawyer SM, Lynskey M (2005).
Reverse gateways? Frequent cannabis use as a predictor of
tobacco initiation and nicotine dependence. Addiction 100,
1518–1525.
Reid L, Elifson K, Sterk C (2007). Ecstasy and gateway
drugs : initiating the use of ecstasy and other drugs. Annals
of Epidemiology 17, 74–80.
Shedler J, Block J (1990). Adolescent drug use and
psychological health : a longitudinal inquiry. American
Psychologist 45, 612–630.
Toumbourou J, Stockwell T, Neighbors C, Marlatt G,
Sturge J, Rehm J (2007). Interventions to reduce harm
associated with adolescent substamce use. Lancet 369,
1391–1401.
Wu LT, Korper S, Marsden M, Lewis C, Bray R (2003). Use of
incidence and prevalence in the substance use literature : a
review. Substance Abuse and Mental Health Services
Administration, Office of Applied Studies : Rockville, MD.
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Associations between order of initiation of drug use and drug dependence
167
Appendix
Table A1. Comparison of the association between gateway violations and incident drug dependence, with and without control for the
number of drug types used. Data from the National Comorbidity Survey Replication (NCS-R), 2001–2003
Alcohol
dependence
among
alcohol users
Tobacco
dependence
among
tobacco users
Drug
dependence
among
cannabis users
Drug
dependence
among
cocaine users
Drug
dependence
among
other drug
usersa
OR
95 % CI
OR
95 % CI
OR
95 % CI
OR
95 % CI
OR
95 % CI
A. Cannabis use before tobacco and
alcohol
B. Cannabis use before tobacco and
alcohol without controlling for the
number of illicit drugs usedb
0.7
0.3–1.5
1.3
0.8–2.0
1.2
0.5–2.8
0.8
0.3–2.3
1.1
0.4–2.8
1.2
0.5–2.5
1.9*
1.2–2.9
1.0
0.4–2.1
0.7
0.2–2.0
1.2
0.5–2.5
A. Other illicit drugs before tobacco
and alcoholc
B. Other illicit drugs before tobacco
and alcoholc without controlling for
the number of illicit drugs usedb
0.4*
0.1–1.0
0.9
0.4–1.7
0.7
0.2–1.9
0.6
0.2–2.4
1.5
0.4–5.6
0.4
0.2–1.2
0.9
0.5–1.8
0.6
0.2–1.9
0.6
0.2–2.3
1.2
0.4–3.7
A. Other illicit drugs before cannabisc
B. Other illicit drugs before cannabisc
without controlling for the number of
illicit drugs usedb
1.5*
2.5*
1.0–2.2
1.6–3.9
0.9
1.2
0.6–1.2
0.9–1.7
1.4
3.7*
0.8–2.3
2.2–6.3
1.7
1.9
0.7–3.9
1.0–3.6
2.3*
1.0
1.4–3.9
0.7–1.5
OR, Odds ratio ; CI, confidence interval.
Results based upon multivariable discrete time survival models.
Model A presents the coefficients from the model as shown in Table 4. Model B was the same analysis as model A with the
exception that the number of illicit drug types used was removed from the model.
‘Onset of dependence’ refers to onset of the full dependence syndrome.
a
‘Other drugs’ includes sedatives/stimulants/analgesics and ‘other’.
b
This is a time-varying covariate and refers to the number of illicit drugs (grouped as cannabis, cocaine, sedatives/
stimulants/analgesics, or ‘ other’) the person had used by a given year.
c
Other illicit drugs : includes any of cocaine, sedatives/stimulants/analgesics, or ‘other’.
* OR significant at 0.05 level, two-tailed test. x2 statistics are available upon request.
Downloaded from https:/www.cambridge.org/core. The University of Montana Missoula, on 19 Feb 2017 at 14:19:32, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291708003425
Blackwell Science, LtdOxford, UKADDAddiction0965-2140© 2005 Society for the Study of Addiction
100
Original Article
Cannabis use and psychotic symptoms
David M. Fergusson
et al.
RESEARCH REPORT
Tests of causal linkages between cannabis use and
psychotic symptoms
David M. Fergusson, L. John Horwood & Elizabeth M. Ridder
Christchurch Health and Development Study, Christchurch School of Medicine, Christchurch, New Zealand
Correspondence to:
Professor David Fergusson
Christchurch Health and Development Study
Department of Psychological Medicine
Christchurch School of Medicine and Health
Sciences
PO Box 4345
Christchurch
New Zealand
Tel: 64 3372 0406
Fax: 64 3372 0405
E-mail: [email protected]
Submitted 31 August 2004;
initial review completed 15 October 2004;
final version accepted 5 November 2004
RESEARCH REPORT
ABSTRACT
Aim To examine possible causal linkages between cannabis use and psychosis
using data gathered over the course of a 25-year longitudinal study.
Design A 25-year longitudinal study of the health, development and adjustment of a birth cohort of 1265 New Zealand children (635 males, 630 females).
Setting The Christchurch Health and Development Study, a general community sample.
Participants A total of 1055 participants from the Christchurch Health and
Development Study (CHDS) cohort for whom data on cannabis use and psychotic symptoms were available on at least one occasion from 18, 21 and
25 years.
Measurements As part of this study, data were gathered on frequency of cannabis use and psychotic symptoms at ages 18, 21 and 25 years.
Findings Regression models adjusting for observed and non-observed confounding suggested that daily users of cannabis had rates of psychotic symptoms that were between 1.6 and 1.8 times higher (P < 0.001) than non-users of
cannabis. Structural equation modelling suggested that these associations
reflected the effects of cannabis use on symptom levels rather than the effects of
symptom levels on cannabis use.
Conclusions The results of the present study add to a growing body of evidence suggesting that regular cannabis use may increase risks of psychosis. The
present study suggests that: (a) the association between cannabis use and psychotic symptoms is unlikely to be due to confounding factors; and (b) the direction of causality is from cannabis use to psychotic symptoms.
KEYWORDS Cannabis, longitudinal study, psychosis, psychotic symptoms, structural equation modelling.
INTRODUCTION
Over the last decade there has been growing research
into the linkages between the use of cannabis and the
development of psychosis and psychotic symptoms (for
reviews see [1–3]). This research has resulted in a growing body of evidence that suggests that the use (and particularly heavy use) of cannabis may be associated with
increased risks of psychosis or psychotic symptoms. This
conclusion has been supported by evidence from a series
of longitudinal studies, all of which have found increased
risks of psychosis or psychotic symptoms among can© 2005 Society for the Study of Addiction
nabis users after control for confounding factors [4–7].
Epidemiological research linking cannabis use and psychosis has also been underwritten by laboratory-based
research examining the psychogenic effects of cannabis
(e.g. [8–11]) and by increasing evidence on the effects of
cannabis on brain chemistry and functioning (e.g. [12–
14]). Collectively, this evidence has provided growing
support for the hypothesis that heavy cannabis use may
precipitate or exacerbate psychosis or psychotic symptoms in vulnerable individuals. None the less, considerable uncertainly still remains about this topic and there is
a clear need for further evidence to confirm the causal
doi:10.1111/j.1360-0443.2005.001001.x
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
355
contribution of cannabis to psychosis and to develop a
clearer understanding of the underlying pathways by
which the consumption of cannabis may be transformed
into an increased susceptibility to psychosis. The aims of
this paper are to test further the causal linkages between
cannabis use and psychotic symptoms by applying statistical modelling methods to the results of a longitudinal
study of cannabis use in a birth cohort studied into adulthood. The background to this analysis is developed below.
have an increased susceptibility to using cannabis as a
consequence of their psychological state.
The above suggests that to clarify further the role of
cannabis in the development of psychotic symptoms and
psychosis there is a need for further research to address
issues relating to the control of residual confounding and
reverse causality in the association between cannabis use
and psychosis. Below we describe methods using longitudinal data to address each of these problems.
Key issues in determining the causal role of cannabis in
psychosis
Controlling residual confounding with the fixed effects
regression model
It has now been well established that the use of cannabis
is statistically linked to increased risks of psychosis. In a
review of five studies, Arsenault et al. [1] found that all
the studies were in agreement that the use of cannabis
increases the risk of subsequent schizophrenia and psychotic symptoms. Similarly, in a parallel review of this
topic Smit et al. [3] concluded that cannabis use is associated with the onset of psychosis, especially in those
prone to developing schizophrenia, and also makes a
unique contribution to the risk of developing schizophrenia. However, the extent to which these statistical associations reflect a cause and effect association in which the
consumption of cannabis leads to an increased susceptibility to psychosis/psychotic symptoms remains open to
debate. There are two potential major threats to validity
that need to be addressed.
Although it is often believed that epidemiological
research can control only for the effects of observed confounders, in fact this is not strictly correct and there are a
number of analytical approaches that permit the control
of non-observed confounders in non-experimental
research. Perhaps the best-known of these is the so-called
discordant twin design, in which monozygotic twins who
are discordant for some exposure variable (e.g. cannabis
use) are compared on an outcome measure (e.g. psychosis). Because the twin pairs share both common genes
and common environment, this comparison controls for
these factors even though the common genes and common environment are not observed [16,17].
The principles underlying the discordant twin design
can also be applied to longitudinal data on singletons via
the fixed effects regression model. In particular, subject to
the availability of longitudinal data, it proves possible to
estimate the associations between a time-varying exposure variable (such as cannabis use) and a time-varying
outcome measure (such as psychosis) net of any nonobserved factors that are associated with the outcome
and that may be correlated with the exposure variable
[18]. The underlying logic of the fixed effects regression
model is described later in Statistical methods. In effect,
this model makes it possible to eliminate one major
source of confounding from fixed factors. However, the
model does not address the issue of confounders that may
vary over time and to control for such confounding, the
fixed effects model needs to be augmented by observed
time-dynamic confounding factors.
Residual confounding
The largest threat to the validity of causal conclusions in
this area comes from the possibility of uncontrolled residual confounding. In reviewing this issue, Macleod et al.
[15] concluded that while a number of studies had shown
linkages between cannabis use and mental health that
persisted following control for confounders the possibility
remained that these linkages reflected uncontrolled residual confounding rather than the causal effects of cannabis use on psychotic symptoms. There is thus a need for
more searching methods for controlling confounding
factors.
Reverse causality
However, even if it were possible to establish that an association existed between cannabis use and psychosis net of
confounders, this evidence would not establish the direction of causation. In particular, there are potentially two
causal pathways that may link cannabis use and psychosis. First, cannabis use may lead (via changes in brain
chemistry) to an increased susceptibility to psychotic
symptoms. Alternatively, those developing psychosis may
© 2005 Society for the Study of Addiction
Ascertaining causal direction using structural equation
modelling
Establishing that cannabis use and psychosis are related,
even following control for confounding, is an important
step in ascertaining a causal relationship between cannabis use and psychosis. However, such analysis does not
resolve the issue of the direction of causality between cannabis use and psychosis: does cannabis use cause psychosis or does psychosis lead to an increased use of cannabis?
Addiction, 100, 354–366
356
David M. Fergusson et al.
Answering such questions proves to be difficult and even
with well collected longitudinal data, establishing which
factor is antecedent and which factor is consequent
proves difficult [1,19]. Furthermore, there is a possibility
that cannabis use and psychosis are related to each other
reciprocally by a feedback loop in which the use of cannabis increases risks of psychosis while at the same time
the onset of psychosis leads to an increased consumption
of cannabis. Structural equation models provide one
means of addressing such a complex issue by devising statistical models that permit reciprocal relationships
between cannabis use and psychosis and using these
models to provide a guide to probable patterns of causation. An account of the ways in which structural equation modelling may be employed to examine reciprocal
pathways is given in the Statistical methods section of this
paper.
Aims of the present study
The present study seeks to examine these issues using
extensive data collected on the development of cannabis
use and psychotic symptoms in a birth cohort of New
Zealand young people studied throughout adolescence
and young adulthood. The aims of this study were
twofold:
1 To control the association between cannabis use and
psychotic symptoms using a range of statistical methods including fixed effects regression to control for
non-observed confounding factors.
2 To employ structural equation modelling methods to
explore the direction of any causal influence between
the use of cannabis and psychotic symptoms.
More generally, the aims of the paper are to apply
complex multivariate methods to an extensive body of
data on cannabis use and psychotic symptoms to address
issues relating to both residual confounding and causal
direction.
METHOD
Participants
The data described in this report were gathered during
the course of the Christchurch Health and Development
Study (CHDS). The CHDS is a longitudinal study of an
unselected birth cohort of 1265 children (635 males,
630 females) born in the Christchurch (New Zealand)
urban region in mid-1977. This cohort has now been
studied at birth, 4 months, 1 year and at annual intervals
to age 16 years, and again at ages 18, 21 and 25 years.
As part of the study, information has been gathered from
a range of sources including: parental interview, teacher
reports, psychometric testing, self-reports, and medical
© 2005 Society for the Study of Addiction
and police records. The present analysis is based on a
sample of 1055 participants for whom information on
cannabis use and psychotic symptoms was available for
at least one assessment from age 18, 21 or 25 years. All
phases of data collection were subject to written,
informed consent from study participants. The following
measures were used in the analysis.
Psychotic symptomatology
At ages 18, 21 and 25 years, sample members were
administered a comprehensive mental health interview
designed to assess a number of aspects of the individual’s
mental health and psychosocial adjustment. As part of
this interview, participants were questioned on current
(over the past month) psychotic symptomatology using
items from the Symptom Checklist 90 (SCL-90) [20]. A
series of 10 items were selected as representative of psychotic symptoms [5]. These items spanned the following
symptoms: hearing voices that other people do not hear;
the idea that someone else can control your thoughts;
other people being aware of your private thoughts; having thoughts that are not your own; having ideas and
beliefs that others do not share; the idea that something is
seriously wrong with your body; never feeling close to
another person; the idea that something is wrong with
your mind; feeling other people cannot be trusted; feeling
that you are watched or talked about by others. Confirmatory factor analysis of the item set has shown previously that the items formed a unidimensional scale
reflecting the extent of psychotic symptomatology [5].
Scale scores were estimated by summing the number of
symptoms reported by each participant at each age. Reliability was assessed using coefficient alpha, a = 0.74
(18 years), a = 0.73 (21 years) and a = 0.75 (25 years).
Frequency of cannabis use
At each assessment from 18 to 25 years, sample members were questioned about their use of cannabis use
since the previous interview. As part of this questioning,
information was obtained on the frequency of cannabis
use over the previous 12-month period. This information
was used to classify sample members on a five-point scale
reflecting the average level of cannabis use throughout
the year. This scale was: 1 = non-user; 2 = used cannabis
on less than a monthly basis; 3 = used cannabis on at
least a monthly basis; 4 = used cannabis on at least a
weekly basis; 5 = used cannabis on a daily basis. To examine the accuracy of reports of cannabis use, data on the
participant’s cannabis use were also obtained from a
nominated informant. There was good agreement
between respondent and informant reports (r = 0.68;
P < 0.001).
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
Time-dynamic covariate factors
To control the associations between cannabis dependence and psychotic symptoms for time-varying sources
of confounding the following measures were selected
from the database of the study.
Prior history of cannabis use/psychotic symptoms
To control for the individual’s prior history of cannabis
use and psychotic symptoms, measures of the frequency
of cannabis use and psychotic symptoms at the time of
the preceding assessment were included as confounding
factors. Thus, for 18 years, psychotic symptoms and cannabis use at age 16 years were controlled, for 21 years
psychotic symptoms and cannabis use at age 18 years
were controlled and for 25 years psychotic symptoms and
cannabis use at age 21 were controlled.
Concurrent/prior mental disorders
As part of the mental health interviews administered at
ages 16, 18, 21 and 25 years, questioning was conducted
to assess standardized diagnostic criteria for a range of
mental disorders. At age 16, questioning was conducted
using an interview that combined components of the
Diagnostic Interview Schedule for Children [21], the SelfReport Delinquency Inventory [22], the Rutgers Alcohol
Problems Index [23] and custom-written survey items to
assess Diagnostic and Statistical Manual version III–revised
(DSM-III-R) symptom criteria. From age 18 onwards the
interview combined components of the Composite International Diagnostic Interview [24], the Self-Report Early
Delinquency Scale [25] and custom-written survey items
to assess relevant DSM-IV diagnostic criteria. Using these
data, sample members were classified on the following
DSM disorders at each age: major depression in the past
12 months; anxiety disorders (including generalized
anxiety disorder, panic disorder/agoraphobia, social
phobia and specific phobia); alcohol and illicit drug
dependence in the past 12 months; current nicotine
dependence; conduct and/or antisocial personality disorder. For the purposes of the present analysis, measures
of both concurrently assessed disorders and disorders at
the time of the previous assessment were included as
covariates.
Other factors
Parallel to questioning on mental health, information
was also obtained on other time-dynamic aspects of the
individual’s life-style, including the extent of affiliations
with deviant peers and exposure to adverse life events. At
each age sample members were questioned on a series of
items concerning the extent to which their friends used or
© 2005 Society for the Study of Addiction
357
had problems associated with alcohol, tobacco or illicit
drugs, had problems with aggression or were involved in
criminal offending. These items were combined to derive
a scale score measure of the extent of deviant peer affiliations at each age [26]. The reliability of all three measures, assessed using coefficient alpha, was 0.85. In
addition, at each assessment sample members were questioned about exposure to adverse life events over the past
12 months using a scale based on the life events scale
described by Henderson, Byrne & Duncan-Jones [27]. At
each age, the number of life events reported was summed
to provide a measure of the extent of adversity experienced in the previous 12 months.
Fixed covariate factors
A wide range of measures of social, family and individual
functioning that were assessed prior to age 18 and were
correlated with either cannabis use or psychotic symptoms were considered in the analysis. These factors
included the following.
Measures of family socio-economic circumstances
(a) Maternal education at the time of the survey child’s
birth was classified in three levels according to the
mother’s highest level of educational attainment (no formal qualifications; high school qualifications; and tertiary qualifications). (b) Maternal age was coded in whole
years at the time of the survey child’s birth. (c) Family
socio-economic status was assessed at the point of birth
using the Elley–Irving [28] scale of socio-economic status
for New Zealand. This index classifies families into six levels on the basis of paternal occupation. (d) Family living
standards (0–10 years): The quality of family living standards was assessed at annual intervals from age 1–
10 years on the basis of interviewer ratings made on a
five-point scale from very good to very poor. These ratings
were averaged over the 10-year period to provide a global
measure of the family’s averaged standard of living over
this period.
Measures of family functioning
(a) Changes of parents (0–15 years): as part of the study
detailed information was obtained at annual intervals
from birth to age 15 years on any changes in family composition. An index of family instability during childhood
was constructed on the basis of a count of the total
number of changes of parents experienced by the child up
to age 15 years. (b) Parental attachment (15 years): the
quality of parental attachments during adolescence was
assessed at age 15 years using the Armsden & Greenberg
[29] Parental Attachment Scale. The reliability of this
Addiction, 100, 354–366
358
David M. Fergusson et al.
scale, assessed using coefficient alpha, was 0.87. (c)
Parental history of depression/anxiety (15 years): when
sample members were aged 15, parents were questioned
about their history of depression or anxiety problems:
29.9% of the sample had at least one parent who reported
problems of depression or anxiety. (d) Parental criminality
(15 years): when sample members were aged 15, parents
were questioned about their history of involvement in
criminal offending: 13.3% of the sample had at least one
parent who reported a history of criminality. (e) Parental
alcohol problems (15 years): when sample members were
aged 15, parents were questioned about their history of
alcoholism or problems with alcohol: 12.1% of the sample had at least one parent who reported alcohol problems. (f) Parental illicit drug use (11 years): when sample
members were aged 11 years, parents were questioned
about their use of cannabis or other illicit drugs: 24.9% of
the sample had at least one parent with a history of illicit
drug use.
Measures of child abuse
(a) Childhood sexual abuse (0–16 years): at ages 18 and
21 years sample members were questioned concerning
their experience of childhood sexual abuse prior to age
16 years, and the nature/context of any episodes of
abuse. Using these data, a four-level classification of the
severity of abuse experience was constructed based on
the worst episode of abuse reported at either age [30].
This classification was: no sexual abuse (86.0% of the
sample); non-contact sexual abuse only (2.7%); contact
sexual abuse not involving attempted or completed intercourse (5.1%); attempted or completed intercourse
(6.2%). (b) Childhood physical abuse (0–16 years): the
extent of childhood physical abuse was assessed on the
basis of the young person’s reports of the extent of parental use of physical punishment during their childhood
(prior to age 16 years), also obtained when sample members were aged 18 years and 21 years. The extent of
physical punishment was coded on a four-point scale
based on the highest level of physical punishment
reported at either age [30]: parents never used physical
punishment (4.5% of the sample); parents rarely used
physical punishment (78.2%); at least one parent regularly used physical punishment (11.3%); at least one parent used physical punishment too often or too severely
(6.0%).
Measures of individual characteristics
(a) Gender. (b) Child neuroticism (14 years): this was
assessed using a short-form version of the neuroticism
scale of the Eysenck Personality Inventory [31] administered when sample members were aged 14 years. The
© 2005 Society for the Study of Addiction
reliability of this scale, assessed using coefficient alpha,
was 0.80. (c) Novelty seeking (16 years): the extent of novelty seeking behaviours was assessed using the novelty
seeking subscale of the Tridimensional Personality Inventory [32] administered when sample members were aged
16 years. The reliability of this scale, assessed using coefficient alpha, was 0.76. (d) Self-esteem (15 years): a measure of self-esteem was obtained at age 15 years using the
Coopersmith Self-Esteem Inventory [33]. The full scale
score was used in the present analysis and this measure
had reliability (alpha) of 0.76. (e) Child IQ (8 years): when
sample members were aged 8 years, children were
assessed on the Revised Wechsler Intelligence Scale for
Children [34]. The full scale score was used in the present
analysis. The reliability of this scale, assessed using split
half methods, was 0.93.
Statistical analysis
Associations between frequency of cannabis use and psychotic
symptoms
The first stage of the analysis reports the bivariate associations between the extent of cannabis use over the age
intervals 17–18, 20–21 and 24–25 years and rates of
psychotic symptoms reported at ages 18, 21 and 25. The
association between the level of cannabis use and the rate
of psychotic symptoms in each year was assessed using a
negative binomial regression model in which the rate of
psychotic symptoms was modelled as a log-linear function of the level of cannabis use. The negative binomial
model provides a useful alternative to Poisson regression
for count data in the presence of overdispersion, that is
where the variance of the outcome variable is greater
than would be expected of a true Poisson [35]. In each
case the significance of the association was assessed
using the log likelihood ratio c2 statistic for the effect of
cannabis use from the fitted model. Tests were conducted
using both linear models and design variates to assess the
impact of cannabis use. These tests showed that, in all
cases, the linear model provided the best fit to the
observed data.
Covariate adjustment models
To adjust the associations between cannabis use and psychotic symptoms for confounding factors, a series of
covariate adjustment models were fitted to the joint data
over the three measurement periods. These models were
as follows.
Model 1: the population averaged model. In this model the
rate of psychotic symptoms at each time was modelled as
a log-linear function of (a) the level of cannabis use in the
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
past year, (b) the set of observed fixed covariates described
above and (c) the set of observed time-dynamic covariates
described above. The kernel of this model was a Poisson
regression model of the form:
Log (Yit) = B0 + B1 Xit + S Bj Zj + SBk Zkt
where Yit was the rate of psychotic symptoms for the ith
participant at time t, Xit was the corresponding measure
of cannabis use at time t, Zj were the set of observed fixed
covariates and Zkt the set of observed time-dynamic
covariates. In this model, the coefficient B1 represents
the effect of cannabis use on the rate of psychotic
symptoms after adjustment for covariates. This
coefficient gives an estimate of the averaged effect of
cannabis use on psychotic symptoms after adjustment
for covariates obtained by pooling observations over the
three measurement periods. To take account of the
correlations between repeated measures for the same
participant over time the model also assumed an
unstructured covariance matrix of the model disturbances over time.
Model 2: the random effects model. This model also
adjusted the pooled association between frequency of
cannabis use and psychotic symptoms for observed fixed
and time-dynamic covariates. However, the model differed from Model 1 in that it also permitted an individual
specific intercept term. The general form of this model
was:
Log (Yit) = ai + B1 Xit + S Bj Zj + SBk Zkt
where ai was the individual specific intercept and all
other variables were as defined above. The random effects
model assumes that the individual intercept terms are
independent of each other and are uncorrelated with the
other predictors in the equation [36].
Model 3: the conditional fixed effects model. The general form of this model was:
Log (Yit) = ai + B1 Xit + SBk Zkt
In this model the ai are individual specific terms that are
assumed to reflect the effects of all fixed sources of variation in the outcome Yit. These effects are assumed to be
constant over time and may be correlated with other predictors in the model. The major advantage of the fixed
effects model is that it can adjust for all sources of fixed
covariate effects, including non-observed fixed confounders [37]. Thus, for example, the fixed effects model can
adjust for such non-observed factors as fixed genetic factors that influence the risks of both cannabis use and psychotic symptoms.
More detailed accounts of the differences between
these three models can be found in [35–37]. In the first
© 2005 Society for the Study of Addiction
359
instance, all three models were fitted to the data using
Poisson regression methods. The analyses were then
repeated using equivalent negative binomial regression
models to account for overdispersion in the distribution of
psychotic symptoms. Both sets of analyses produced the
same conclusions, and the negative binomial results are
reported in the paper. All models were fitted using Stata
6.0 [38].
From the fitted models, estimates of the adjusted
incidence rate ratios (IRRs) of psychotic symptoms for
varying levels of cannabis use were calculated relative to
non-users of cannabis. For a given model, the adjusted
IRR for a one-level increase in the frequency of cannabis
use was given by eB1, where B1 was the regression coefficient associated with cannabis use in the fitted model and
e is the base of natural logarithms.
Structural equation modelling
Although the covariate adjustment models above address
sources of confounding, these models do not provide tests
of the direction of causality (if any) between cannabis use
and psychotic symptoms. To explore this issue, a series of
structural equation models were fitted to the data. These
models are depicted in Figs 1 and 2.
The model in Fig. 1 assumes that: (a) the observed
measures of cannabis use (Ct, t = 1,2,3) over the three
time periods are linked by an autoregressive structure in
which past cannabis use predicts future cannabis use; (b)
the observed measures of psychotic symptoms (Pt,
t = 1,2,3) are also linked by a similar autoregressive
structure in which past symptoms predict future symptoms; (c) within time periods cannabis use and psychotic
symptoms are potentially reciprocally related so that (i)
current cannabis use may influence current psychotic
symptoms and (ii) current psychotic symptoms may
influence current cannabis use. These reciprocal effects
are assumed to be constant over time. The model specification is:
Model equations
C3 = B1 P3 + B3 C2 + B5 C1 + n3 P3 = B2 C3 + B6 P2 + B8 P1 + t3
P2 = B2 C2 + B7 P1 + t2
C2 = B1 P2 + B4 C1 + n2
C1 = B1 P1 + n1
P1 = B2 C1 + t1
Model assumptions
Cov (nr, ts) = Cov (nr, ns) = Cov (tr, ts) = 0 for r π s
Cov (Cr, ns) = Cov (Pr, ns) = 0 for r < s
Cov (Cr, ts) = Cov (Pr, ts) = 0 for r < s
In terms of assessing the direction of causality between
cannabis use and psychotic symptoms, the values of the
parameters B1, B2 may provide important information
about both the size and direction of this influence.
A limitation of the autoregressive model in Fig. 1 is
that this model does not take into account common conAddiction, 100, 354–366
360
David M. Fergusson et al.
founding factors that may influence both cannabis use
and psychotic symptoms. This issue is addressed in the
model in Fig. 2 which includes fixed effects factors to take
into account: (i) fixed factors that influence cannabis use
and (ii) fixed factors that influence psychotic symptoms.
Specifically, the model assumes that: (a) the observed
measures of cannabis use (Ct) are influenced by fixed
sources of variance (C) that are constant over time and
time-dynamic sources of variation (Ut); (b) the observed
measures of psychotic symptoms (Pt) are also influenced
by fixed sources of variation (P) that are constant over
time and time-dynamic sources of variation (Wt); (c) the
fixed factors C and P are permitted to be correlated; (d)
the time-dynamic components of cannabis use (Ut) and
psychotic symptoms (Wt) are linked by autoregressive
processes in which past cannabis use predicts future cannabis use and past psychotic symptoms predict future
psychotic symptoms, respectively; (e) the time dynamic
components of cannabis use and psychotic symptoms are
reciprocally related so that current Ut influences current
Wt and vice versa. These reciprocal effects are assumed to
be constant over time. The specification for this model as
follows:
Model equations
Ct = C + Ut (t = 1, 2, 3)
U3 = B1 W3 + B3 U2 + n3
U2 = B1 W2 + B4 U1 + n2
U1 = B1 W1 + n1
Pt = P + Wt (t = 1, 2, 3)
W3 = B2 U3 + B5 W2 + t3
W2 = B U2 + B6 W1 + t2
W1 = B2 U1 + t1
Missing data and sample bias
Model assumptions
Cov (C, Ut) = Cov (C, Wt) = Cov (C, nt) = Cov (C, tt) = 0
(t = 1, 2, 3)
ν3
Cov (P, Ut) = Cov (P, Wt) = Cov (P, nt) = Cov (P, tt) = 0
(t = 1, 2, 3)
Cov (nr, ts) = Cov (nr, ns) = Cov (tr, ts) = 0 for r1 s
Cov (Cr, ns) = Cov (Cr, ts) = Cov (Pr, ns) = Cov (Pr,
ts) = 0 for r < s
Cov (Ur, ns) = Cov (Ur, ts) = Cov (Wr, ns) = Cov (Wr,
ts) = 0 for r < s
The advantage of the model specification in Fig. 2 is that
it estimates the pathways between cannabis use and psychotic symptoms, taking into account non-observed fixed
factors associated with these measures.
The models depicted in Figs 1 and 2 were fitted to the
observed measures of cannabis use and psychotic symptoms at age 18, 21 and 25 years. As the observed measures were markedly non-normally distributed the
models were fitted to the covariance matrix of the
observed data using the method of weighted least
squares. All models were fitted using LISREL 8 [39].
Model goodness of fit was assessed on the basis of a number of indices including: (a) the log-likelihood ratio c2
statistic; (b) the root mean squared error of approximation (RMSEA). Values of RMSEA less than 0.05 are
assumed to be indicative of a well-fitting model; (c) the
standardized root mean squared residual correlation
(RMSR) between the observed measures. Values of RMSR
close to zero indicate a well-fitting model. (d) The Comparative Fit Index (CFI). This index varies between 0 and
1 with values close to 1 indicating a well-fitting model
[39].
As noted above, the analysis is based on the sample of
1055 participants for whom data on cannabis use and
τ3
B1
C3
P3
B2
B3
ν2
τ2
B6
B1
B5
C2
P2
B8
B2
τ1
ν1
B4
B7
B1
P1
C1
B2
© 2005 Society for the Study of Addiction
Figure 1 Autoregressive model of cannabis use and
psychotic symptoms with reciprocal paths between
cannabis use and psychotic symptoms. Ct = cannabis
use at time t; Pt = psychotic symptoms at time t;
nt = disturbance term for Ct; tt = disturbance term
for Pt
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
n3
t3
1
C3
361
1
B1
U3
W3
P3
B2
1
1
B3
1
B1
1
P
P2
W2
U2
C2
B5
t2
1
1
C
n2
B2
n1
B4
1
B1
1
C1
1
B6
t1
1
U1
W1
P1
B2
Figure 2 Auroregressive model of cannabis use and psychotic symptoms incorporating fixed effects and reciprocal paths between dynamic
components of canabis use and psychotic symptoms. Ct = cannabis use at time t; Pt = psychotic symptoms at time t; C =fixed effects component of Ct; P = fixed effects component of Pt; Ut = time dynamic component of Ct; Wt = time dynamic component of Pt;
nt = disturbance term for Ut; tt = disturbance term for Wt
psychotic symptoms were available on at least one occasion from 18, 21 and 25 years. However, as not all participants were assessed at all ages the observed sample
numbers vary between age 18 (n = 1025), age 21
(n = 1011) and age 25 (n = 1003). These samples represented between 79% and 81% of the initial cohort of
1265 participants. In addition, as a result of missing data
on some covariates the sample number included in the
covariate adjustment analyses was reduced to approximately 900.
To examine the implications of sample attrition and
missing data for study conclusions a series of additional
analyses were undertaken. First, regression imputation
methods were used to impute estimates for the missing
data on covariate factors, and the covariate adjustment
analyses were repeated with the missing data replaced by
their imputed values. The regression imputation was
conducted using the impute procedure of Stata 6.0 [38].
Secondly, to adjust for possible sample selection bias
resulting from sample attrition, the methods described
by Carlin et al. [40] were used. These methods involved a
two-stage analysis process. In the first stage of the analysis, a sample selection model was constructed by using
data gathered at birth to predict participation at each
age. This analysis showed that there were statistically
significant (P < 0.05) tendencies for the obtained sample
at each age to under-represent children from more
socially disadvantaged backgrounds (low parental edu© 2005 Society for the Study of Addiction
cation, low socio-economic status, single-parent family).
On the basis of the fitted selection models, the sample
was then poststratified into a series of groups and the
probability of study participation estimated for each
group at each age.
In the second stage of the analysis the data were reanalysed by fitting a negative binomial regression model
to the full data with the observations for each individual
weighted by the inverse of the probability of study participation at each age to adjust for sample selection bias. All
analyses produced essentially identical conclusions to the
findings reported here, suggesting that the effects of missing data and possible sample selection bias on the results
were likely to be minimal.
RESULTS
Associations between cannabis use and rates of psychotic
symptoms at 18, 21 and 25 years
Table 1 shows the relationship between reported rates of
cannabis use in the past 12 months at ages 18, 21 and
25 years, and self-reported psychotic symptoms at these
ages. Each comparison is tested for statistical significance
using the log likelihood ratio c2 statistic derived from a
negative binomial regression model. The analysis shows
that at all ages there were clear and highly statistically
Addiction, 100, 354–366
362
David M. Fergusson et al.
significant (P < 0.0001) trends for increasing use of cannabis to be associated with increasing rates of psychotic
symptoms: young people who were daily users of cannabis had rates of psychotic symptoms that were between
2.3 and 3.3 times higher than the rates for those who did
not use cannabis.
Adjustments for covariate factors
As explained in the Methods, the associations between
cannabis use and psychotic symptoms were adjusted for
observed covariates using three approaches to covariate
adjustment: (a) a population averaged model using
observed fixed and time dynamic covariates; (b) a random
effects model using observed fixed and time-dynamic
covariates; and (c) a fixed effects model that took into
account both non-observed fixed factors and observed
time dynamic covariates. The results of these analyses are
given in Table 2, which shows the incidence rate ratios
(IRRs) of psychotic symptoms and corresponding 95%
confidence intervals associated with each model after
adjustment for covariates. In each case the IRRs show the
rate of psychotic symptoms for a given level of cannabis
use relative to non-users. All models yield highly consis-
tent estimates that suggest that those who used cannabis
daily had rates of psychotic symptoms that were in the
region of 1.6–1.8 times higher than those who did not
use cannabis. Furthermore, the findings suggest that the
adjustments for observed covariates in Models 1 and 2
produce conclusions that are consistent with the adjustments for non-observed covariates using the fixed effects
model.
Results from reciprocal causes models
The findings in Table 2 are consistent with the view that
cannabis use and psychotic symptoms may be linked by a
cause-and-effect model. However, the analysis does not
establish that this association is one in which increasing
frequency of cannabis use leads to increased psychotic
symptoms. To address this issue, the data were analysed
using the reciprocal cause structural equation models
described in the Methods. These models include the
autoregressive model shown in Fig. 1 and the autoregressive model including fixed effects factors shown in Fig. 2.
The key findings of this analysis are summarized in
Table 3 which shows: (a) estimates of the fitted model
Table 1 Mean psychotic symptoms (number of subjects) by frequency of cannabis use (past 12 months) at 18, 21 and 25 years.
Frequency of cannabis use (past 12 months)
Age (years)
Never
Less than
monthly
At least
monthly
At least
weekly
Daily
P
18
21
25
0.64 (598)
0.69 (538)
0.60 (559)
0.95 (242)
1.00 (215)
0.89 (232)
1.07 (82)
1.14 (100)
0.93 (76)
1.93 (70)
1.48 (94)
1.15 (81)
1.64 (33)
1.61 (64)
1.95(55)
<0.0001
<0.0001
<0.0001
Table 2 Estimated incidence rate ratios (95% CI) of psychotic symptoms by level of cannabis use after adjustment for covariates.
Frequency of cannabis use (past 12 months)
Covariate
adjustment
model
At least
weekly
Daily
P
Model 1: population averaged (observed fixed1 and time dynamic2 covariates)
IRR
1 (95% CI) 1.12 (1.05–1.20)
1.25 (1.09–1.43)
1.40 (1.14–1.71)
1.56 (1.20–2.04)
<0.001
Model 2: random effects (observed fixed1 and time dynamic2 covariates)
IRR
1 (95% CI) 1.12 (1.05–1.19)
1.24 (1.10–1.41)
1.39 (1.15–1.68)
1.55 (1.21–1.99)
<0.001
Model 3: fixed effects (non-observed fixed and observed time dynamic covariates)
IRR
1 (95% CI) 1.15 (1.06–1.25)
1.33 (1.13–1.56)
1.53 (1.20–1.95)
1.77 (1.28–2.44)
<0.001
Never
Less than
monthly
At least
monthly
2
1
Observed fixed covariates included: gender; parental education; family socio-economic status; family living standards; changes of parents; parental alcohol problems; parental illicit drug use; parental depression/anxiety; parental criminality; childhood sexual abuse; childhood physical abuse; neuroticism;
novelty seeking; self-esteem; parental attachment; child IQ. 2Observed time dynamic covariates included: prior psychotic symptoms; prior frequency of
cannabis use; concurrent and prior mental disorders (major depression, anxiety disorders, alcohol dependence, nicotine dependence, illicit drug dependence, conduct disorder/aspd); adverse life events; deviant peer affiliations.
© 2005 Society for the Study of Addiction
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
parameters and standard errors for the effects of cannabis use on psychotic symptoms and the effects of psychotic symptoms on the frequency of cannabis use; (b)
measures of model fit including the log likelihood ratio c2
test statistic, the (RMSEA), the standardized root mean
squared residual correlation (SRMR) and the comparative fit index (CFI). The results of the structural equation
models suggest the following conclusions:
1 For both models, cannabis use had a positive and significant effect (P < 0.001) on psychotic symptoms,
implying that increasing cannabis use was associated
with increased symptom levels.
2 For both models, the effect of psychotic symptoms on
cannabis use was negative and, for Model 2, statistically non-significant. These results imply that it was
unlikely that the development of psychotic symptoms
led to increased use of cannabis and that, if anything,
the development of these symptoms may have inhibited rather than encouraged cannabis use.
3 Both models proved to be well fitting on the basis of a
range of goodness of fit measures.
Collectively, the results in Tables 2 and 3 are consistent
with two major conclusions. First, the use of cannabis
and rates of psychotic symptoms were related to each
other, independently of observed/non-observed fixed
covariates and observed time dynamic factors (Table 2).
Secondly, the results of structural equation modelling
suggest that the direction of causation is that the use of
cannabis leads to increases in levels of psychotic symptoms rather than psychotic symptoms increasing the use
of cannabis. Indeed, there is a suggestion from the model
results that increases in psychotic symptoms may inhibit
the use of cannabis.
DISCUSSION
This analysis has used data gathered over the course of a
25-year longitudinal study to address two issues regarding the linkages between the use of cannabis and psy-
363
chotic symptoms. The first issue concerned the extent to
which the association between cannabis use and psychotic symptoms reflected uncontrolled confounding factors. The second issue addressed the direction of causality
(if any) between cannabis use and psychotic symptoms.
The findings of these analyses are reviewed below.
The effects of confounding factors
One of the more controversial issues regarding linkages
between cannabis use and psychosis/psychotic symptoms has concerned the extent to which these linkages
reflect uncontrolled residual confounding [15]. In this
paper we have attempted to address this problem by
adjusting these associations using two approaches to
covariate control. In the first approach we controlled for
observed confounders using extensive prospectively collected covariate data. In the second approach we used
fixed effects regression to control for non-observed fixed
sources of confounding. Both methods of adjustment
gave similar results and suggested the presence of a dose–
response relationship between the frequency of cannabis
uses and rates of psychotic symptoms. It was estimated
that daily users of cannabis had rates of these symptoms
that were 1.6–1.8 times higher than non-users of cannabis even after both observed and non-observed sources
of confounding were taken into account.
These results add to a growing body of evidence that
now suggests that the linkages between cannabis use and
psychotic symptoms are likely to be causal and are
unlikely to be due to sources of uncontrolled residual
confounding. None the less, the possibility of residual
confounding cannot be dismissed entirely because
although the regression models used in this analysis controlled for both observed and non-observed fixed factors,
the possibility of confounding by (non-fixed) timedynamic factors remains. A further issue concerns the
assessment of psychotic symptoms. In this study we have
used a scale measure based on a count of symptoms.
However, it could be suggested that this measure differs
Table 3 Estimated reciprocal effects of frequency of cannabis use and psychotic symptoms for alternative structural equation models.
Model
Effect of cannabis use on psychotic symptoms
Effect of psychotic symptoms on cannabis use
B (SE)
P
B (SE)
P
<0.001
- 0.094 (.047)
<0.05
- 0.045 (.043)
>0.25
Model 1: autoregressive model on observed variables
0.154 (.044)
Model 2: autoregressive model incorporating non-observed fixed effects
<0.001
0.352 (.087)
Goodness of fit indices: (a) for model 1, LR c2 (4) = 7.6, P > 0.10; RMSEA = 0.03, P > 0.80; SRMR = 0.029; CFI = 0.998. (b) For model 2, LR c2 (5) = 4.00,
P > 0.50; RMSEA = 0.00, P > 0.98; SRMR = 0.017; CFI = 1.00.
© 2005 Society for the Study of Addiction
Addiction, 100, 354–366
364
David M. Fergusson et al.
from diagnostic classification and also may not disclose
all aspects of psychosis. While measurement issues are a
potential threat to validity in studies of cannabis and psychosis, this threat does not appear to be large. As recent
reviews [1–3] have shown, authors using range of
measures including diagnostic classifications and scale
dimensions have been able to show linkages between the
use of cannabis and rates of psychosis/psychotic symptoms. Despite these caveats we believe that the weight of
the evidence is now firmly in favour of the view that cannabis use and psychosis/psychotic symptoms are likely to
be causally related.
Direction of causality
The demonstration that cannabis use and psychotic
symptoms remain associated even following control for
confounding suggests a causal linkage, but does not
establish the direction of causality. There are potentially
two causal pathways by which cannabis use and psychosis may be linked. First cannabis use may lead (via biochemical changes in the brain) to increased rates of
psychotic symptoms amongst susceptible users. Alternatively, those prone to psychosis or psychotic symptoms
may be more prone to use cannabis as a consequence
of their condition and perhaps as an attempt at selfmedication [41,42]. Resolving this issue is clearly critical
to understanding the causal role that cannabis use may
play in psychosis. To address this issue we have employed
methods of structural equation modelling that permit
estimation of reciprocal causal pathways. Two models
were fitted, with the first using a relatively simple autoregressive structure to identify model parameters and the
second incorporating fixed effects models for cannabis
use and psychotic symptoms. Both models led to similar
conclusions about the possible causal linkages between
cannabis use and psychotic symptoms. First, there was
clear evidence to suggest that increasing use of cannabis
was associated with statistically significant increases in
the risks of psychotic symptoms. Secondly, increasing
psychotic symptoms were not positively associated with
increased rates of cannabis use and indeed the fitted
autoregressive model suggested that the association
between psychosis and cannabis use may be negative, so
that increasing psychotic symptoms were associated with
a decline in the use of cannabis. The weight of the evidence from the SEM approach clearly suggests the presence of a causal process in which increasing use of
cannabis is associated with increasing rates of psychotic
symptoms.
Of course, these conclusions rest upon some of the relatively strong assumptions (see Methods) required to
identify these models, but it is important to note that
these assumptions did not favour finding a particular
© 2005 Society for the Study of Addiction
causal pathway between cannabis use and psychotic
symptoms.
Does cannabis use cause psychosis?
Finally, the present study needs to be seen in the context
of a wider literature that has explored the issue of cannabis use and psychosis. This literature is beginning to
provide the foundations of a coherent picture that supports the view that cannabis use may contribute to psychosis or psychotic symptoms in individuals vulnerable to
these conditions. This evidence includes:
1 The growing epidemiological evidence (including the
present study) that suggests evidence of dose–
response relationships between the extent of cannabis
use and subsequent psychosis/psychotic symptoms
even following control for sources of confounding and
possible reverse causality [4–7].
2 Evidence from clinical studies suggesting that cannabis use is associated with an increased relapse rate
in individuals with schizophrenia [43,44].
3 Growing neuropsychological evidence on the multiple
effects of cannabis on the brain and brain biochemistry [12,13].
4 Evidence from laboratory-based studies suggesting
that the acute effects of cannabis intoxication may
create psychotic-like symptoms and may be used as a
‘model’ psychosis [8,11].
Although each of these lines of evidence is subject to
uncertainty and debate, the weight of the evidence
clearly suggests that the use of cannabis (and particularly the heavy use of cannabis) may alter underlying
brain chemistry and precipitate the onset of psychosis/
psychotic symptoms in vulnerable individuals. The
present study adds to this evidence by showing: (a) it is
unlikely (although not impossible) that the association
between cannabis use and psychotic symptoms in a
population sample was due to confounding factors, and
(b) the predominant direction of causality is likely to
involve a path from cannabis use to psychotic symptoms
rather than a path from psychotic symptoms to cannabis use.
Acknowledgements
This research was funded by grants from the Health
Research Council of New Zealand, the National Child
Health Research Foundation, the Canterbury Medical
Research Foundation and the New Zealand Lottery
Grants Board.
References
1. Arseneault, L., Cannon, M., Witton, J. & Murray, R. M.
(2004) Causal association between cannabis and psychosis:
Addiction, 100, 354–366
Cannabis use and psychotic symptoms
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
examination of the evidence. British Journal of Psychiatry,
184, 110–117.
Hall, W. & Degenhardt, L. (2000) Cannabis use and
psychosis: a review of clinical and epidemiological evidence. Australian and New Zealand Journal of Psychiatry, 34,
26–34.
Smit, F., Bolier, L. & Cuijpers, P. (2004) Cannabis use and the
risk of later schizophrenia: a review. Addiction, 99, 425–430.
Arseneault, L., Cannon, M., Poulton, R., Murray, R., Caspi,
A. & Moffitt, T. E. (2002) Cannabis use in adolescence and
risk for adult psychosis: longitudinal prospective study.
BMJ, 325, 1212–1213.
Fergusson, D. M., Horwood, L. J. & Swain-Campbell, N. R.
(2003) Cannabis dependence and psychotic symptoms in
young people. Psychological Medicine, 33, 15–21.
van Os, J., Bak, M., Hanssen, M., Bijl, R. V., de Graaf, R. &
Verdoux, H. (2002) Cannabis use and psychosis: a longitudinal population-based study. American Journal of Epidemiology, 156, 319–327.
Zammit, S., Allebeck, P., Andreasson, S., Lundberg, I. &
Lewis, G. (2002) Self reported cannabis use as a risk factor
for schizophrenia in Swedish conscripts of 1969: historical
cohort study. BMJ, 325, 1199.
D’Souza, D. C., Cho, H.-S., Perry, E. B. & Krystal, J. H. (2004)
Cannabinoid ‘model’ psychosis, dopamine–cannabinoid
interactions and implications for schizophrenia. In: Castle,
D. & Murray, R., eds. Marijuana and Madness: Psychiatry and
Neurobiology, pp. 142–165. Cambridge: Cambridge University Press.
Emrich, H. M., Leweke, F. M. & Schneider, U. (1997)
Towards a cannabinoid hypothesis of schizophrenia: cognitive impairments due to dysregulation of the endogenous
cannabinoid system. Pharmacology, Biochemistry and Behavior, 56, 803–807.
Isbell, H., Gorodetzsky, C. W., Jasinski, D., Claussen, U.,
von Spulak, F. & Korte, F. (1967) Effects of (–)delta-9trans-tetrahydrocannabinol in man. Psychopharmacologia,
11, 184–188.
Voruganti, L. N., Slomka, P., Zabel, P., Mattar, A. & Awad,
A. G. (2001) Cannabis induced dopamine release: an in-vivo
SPECT study. Psychiatry Research, 107, 173–177.
Dean, B., Sundram, S., Bradbury, R., Scarr, E. & Copolov, D.
(2001) Studies on [3H]CP-55940 binding in the human
central nervous system: regional specific changes in density
of cannabinoid-1 receptors associated with schizophrenia
and cannabis use. Neuroscience, 103, 9–15.
Leroy, S., Griffon, N., Bourdel, M. C., Olie, J. P., Poirier, M. F.
& Krebs, M. O. (2001) Schizophrenia and the cannabinoid
receptor type 1 (CB1): association study using a single-base
polymorphism in coding exon 1. American Journal of Medical
Genetics, 105, 749–752.
Mathew, R. J., Wilson, W. H., Coleman, R. E., Turkington,
T. G. & DeGrado, T. R. (1997) Marijuana intoxication and
brain activation in marijuana smokers. Life Sciences, 60,
2075–2089.
Macleod, J., Oakes, R., Copello, A., Crome, I., Egger, M.,
Hickman, M., Oppenkowski, T., Stokes-Lampard, H. &
Smith, G. D. (2004) Psychological and social sequelae of
cannabis and other illicit drug use by young people: a systematic review of longitudinal, general population studies.
Lancet, 363, 1579–1588.
Kendler, K. S. & Prescott, C. A. (1998) Cannabis use, abuse
and dependence in a population-based sample of female
twins. American Journal of Psychiatry, 155, 1016–1022.
© 2005 Society for the Study of Addiction
365
17. Lynskey, M. T., Heath, A. C., Nelson, E. C., Bucholz, K. K.,
Madden, P. A. F., Slutske, W. S., Statham, D. J. & Martin, N.
G. (2002) Genetic and environmental contributions to cannabis dependence in a national young adult twin sample.
Psychological Medicine, 32, 195–207.
18. Fergusson, D. M., Horwood, L. J. & Swain-Campbell, N. R.
(2002) Cannabis use and psychosocial adjustment in adolescence and young adulthood. Addiction, 97, 1123–1135.
19. Johns, A. (2001) Psychiatric effects of cannabis. British
Journal of Psychiatry, 178, 116–122.
20. Derogatis, L. R., Lipman, R. S. & Covi, L. (1973) SCL-90: an
outpatient psychiatric rating scale—preliminary report.
Psychopharmacology, 9, 13–28.
21. Costello, A., Edelbrock, C., Kalas, R., Kessler, M. & Klaric, S.
A. (1982) Diagnostic Interview Schedule for Children (DISC).
Bethesda, MD: National Institute of Mental Health.
22. Moffitt, T. R. & Silva, P. A. (1988) Self-reported delinquency: results from an instrument for New Zealand. Australian and New Zealand Journal of Criminology, 21, 227–240.
23. White, H. L. & Labouvie, E. W. (1989) Towards the assessment of adolescent problem drinking. Journal of Studies on
Alcohol, 50, 30–37.
24. World Health Organization (1993) Composite International
Diagnostic Interview (CIDI). Geneva: World Health
Organization.
25. Elliott, D. S. & Huizinga, D. (1989) Improving self-reported
measures of delinquency. In: Klein, M. W., ed. Cross-National
Research in Self-Reported Crime and Delinquency, pp. 155–
186. The Netherlands: Kluwer Academic Publishers.
26. Fergusson, D. M., Swain-Campbell, N. R. & Horwood, L. J.
(2002) Deviant peer affiliations, crime and substance use: a
fixed effects regression analysis. Journal of Abnormal Psychology, 30, 419–430.
27. Henderson, S., Byrne, D. G. & Duncan-Jones, P. (1981)
Neurosis and the Social Environment. Sydney: Academic
Press.
28. Elley, W. B. & Irving, J. C. (1976) Revised socio-economic
index for New Zealand. New Zealand Journal of Educational
Studies, 11, 25–36.
29. Armsden, G. C. & Greenberg, M. T. (1987) The inventory of
parent and peer attachment: individual differences and
their relationship to psychological well-being in adolescence. Journal of Youth and Adolescence, 16, 427–454.
30. Fergusson, D. M., Horwood, L. J. & Woodward, L. J. (2000)
The stability of child abuse reports: a longitudinal study of
young adults. Psychological Medicine, 30, 529–544.
31. Eysenck, H. M. & Eysenck, S. B. G. (1964) Manual of the
Eysenck Personality Inventory. London: London University
Press.
32. Cloninger, C. R. (1987) A systematic method for clinical
description and classification of personality variants. A proposal. Archives of General Psychiatry, 44, 573–588.
33. Coopersmith, S. (1981) SEI—Self Esteem Inventories. Palo
Alto, CA: Consulting Psychologists Press.
34. Wechsler, D. (1974) Manual for the Wechsler Intelligence
Scale for Children—Revised. New York: Psychological
Corporation.
35. Hausman, J., Hall, B. H. & Griliches, Z. (1984) Econometric
models for count data with an application to the patents—R
& D relationship. Econometrica, 52, 909–938.
36. Cameron, A. C. & Trivedi, P. K. (1998) Regression Analysis of
Count Data. Cambridge: Cambridge University Press.
37. Hamerle, A. & Ronning, G. (1995) Panel analysis for qualitative variables. In: Arminger, G., Clogg, C. C. & Sobel, M.
Addiction, 100, 354–366
366
38.
39.
40.
41.
David M. Fergusson et al.
E., eds. Handbook of Statistical Modeling for the Social and
Behavioral Sciences, pp. 401–451. New York: Plenum Press.
StataCorp (1999) Stata Statistical Software: Release 6.0.
Texas: Stata Corporation, College Station.
Joreskog, K. G. & Sorbom, D. (1993) Structural Equation
Modelling with the SIMPLIS Command Language. Chicago:
Scientific Software.
Carlin, J. B., Wolfe, R., Coffey, C. & Patton, G. C. (1999)
Tutorial in biostatistics. Analysis of binary outcomes in longitudinal studies using weighted estimating equations and
discrete-time survival methods: prevalence and incidence of
smoking in an adolescent cohort. Statistics in Medicine, 18,
2655–2679.
Peralta, V. & Cuesta, M. J. (1992) Influence of cannabis
© 2005 Society for the Study of Addiction
abuse on schizophrenic psychopathology. Acta Psychiatrica
Scandinavica, 85, 127–130.
42. Dixon, L., Haas, G., Weiden, P., Sweeney, J. & Frances, A.
(1990) Acute effects of drug abuse in schizophrenic
patients: clinical observations and patients’ self-reports.
Schizophrenia Bulletin, 16, 69–79.
43. Negrete, J. C., Knapp, W. P., Douglas, D. E. & Smith, W. B.
(1986) Cannabis affects the severity of schizophrenic symptoms: results of a clinical survey. Psychological Medicine, 16,
515–520.
44. Linszen, D. H., Dingemans, P. M. & Lenior, M. E. (1994)
Cannabis abuse and the course of recent-onset schizophrenic disorders. Archives of General Psychiatry, 51, 273–
279.
Addiction, 100, 354–366
Drug and Alcohol Dependence 71 (2003) 37 /48
www.elsevier.com/locate/drugalcdep
Testing hypotheses about the relationship between cannabis use and
psychosis
Louisa Degenhardt *, Wayne Hall, Michael Lynskey
National Drug and Alcohol Research Centre, University of NSW, Sydney, NSW 2052, Australia
Received 20 August 2002; accepted 11 February 2003
Abstract
Aim: To model the impact of rising rates of cannabis use on the incidence and prevalence of psychosis under four hypotheses
about the relationship between cannabis use and psychosis. Methods: The study modelled the effects on the prevalence of
schizophrenia over the lifespan of cannabis in eight birth cohorts: 1940 /1944, 1945 /1949, 1950 /1954, 1955 /1959, 1960 /1964,
1965 /1969, 1970 /1974, 1975 /1979. It derived predictions as to the number of cases of schizophrenia that would be observed in
these birth cohorts, given the following four hypotheses: (1) that there is a causal relationship between cannabis use and
schizophrenia; (2) that cannabis use precipitates schizophrenia in vulnerable persons; (3) that cannabis use exacerbates
schizophrenia; and (4) that persons with schizophrenia are more liable to become regular cannabis users. Results: There was a
steep rise in the prevalence of cannabis use in Australia over the past 30 years and a corresponding decrease in the age of initiation of
cannabis use. There was no evidence of a significant increase in the incidence of schizophrenia over the past 30 years. Data on trends
the age of onset of schizophrenia did not show a clear pattern. Cannabis use among persons with schizophrenia has consistently
been found to be more common than in the general population. Conclusions: Cannabis use does not appear to be causally related to
the incidence of schizophrenia, but its use may precipitate disorders in persons who are vulnerable to developing psychosis and
worsen the course of the disorder among those who have already developed it.
# 2003 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Cannabis use; Psychosis; Schizophrenia
1. Introduction
Clinical research has shown that high proportions of
persons with schizophrenia report regular cannabis use
and meet criteria for cannabis use disorders (Fowler et
al., 1998; Mueser et al., 1990; Ziedonis and Trudeau,
1997). Epidemiological studies have also found an
association between cannabis use and psychosis in the
general population (Anthony and Helzer, 1991; Cuffel et
al., 1993; Degenhardt and Hall, 2001; Tien and
Anthony, 1990).
There has been considerable debate about the reasons
for this association (Batel, 2000; Blanchard et al., 2000;
* Corresponding author. Tel.: /61-2-9385-0230; fax: /61-2-93850222.
E-mail address: [email protected] (L. Degenhardt).
Gruber and Pope, 1994; Hall, 1998; Hall and Degenhardt, 2000; McKay and Tennant, 2000; Mueser et al.,
1998; Rosenthal, 1998; Thornicroft, 1990). Depending
upon the nature of the relationship between cannabis
use and psychosis, changes in the prevalence of cannabis
use may potentially lead to changes in the incidence,
prevalence or age of onset of psychosis.
In Australia, there has been a dramatic increase in the
prevalence of cannabis use since the early 1970s
(Degenhardt et al., 2000; Donnelly and Hall, 1994;
Makkai and McAllister, 1998; McCoy, 1980). The
present report assesses the evidence for four hypothesised relationships between cannabis use and psychosis,
which would each predict different effects of increased
cannabis use on the incidence, prevalence and age of
onset of schizophrenia and the prevalence of chronic
cannabis use among persons with the disorder.
03765-8716/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved.
doi:10.1016/S0376-8716(03)00064-4
38
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
1.1. Hypothesis 1: Cannabis use causes psychosis
According to this hypothesis there is a causal link
between cannabis use and schizophrenia in the sense
that cannabis use causes cases of the disorder that would
not otherwise have occurred. This hypothesis has arisen
from reports of ‘cannabis psychoses’ (Basu et al., 1999;
Bernardson and Gunne, 1972; Carney et al., 1984;
Chopra and Smith, 1974; Eva, 1992; Solomons et al.,
1990; Tennant and Groesbeck, 1972; Wylie et al., 1995)
and there is some evidence that cannabis users are more
likely to report psychotic symptoms (e.g. Andreasson et
al., 1987; Degenhardt and Hall, 2001; Tien and
Anthony, 1990).
If this hypothesis is correct, then an increase in
prevalence of cannabis use among young adults should
increase the incidence and ultimately the prevalence of
schizophrenia. Since there has been a dramatic rise in
the prevalence of cannabis use in Australia, this
hypothesis predicts an increase in the greater number
of cases of schizophrenia among recent birth cohorts.
Since the age of cannabis initiation has also declined,
this hypothesis also predicts that the age of onset of
schizophrenia would decline in recent birth cohorts.
This hypothesis also predicts a rising prevalence of
cannabis use among persons with schizophrenia.
1.2. Hypothesis 2: Cannabis use precipitates psychosis
among vulnerable individuals
A second hypothesis is that regular cannabis use
precipitates schizophrenia among vulnerable individuals ,
that is, among persons who would have developed the
disorder regardless of whether they used cannabis or not
(Hall, 1998). This is supported by evidence that: (a)
persons with first-episode schizophrenia who use cannabis are younger than those who do not (Linszen et al.,
1994; Mathers et al., 1991; Rolfe et al., 1993); (b)
cannabis use usually precedes the development of
psychotic symptoms (Allebeck et al., 1993; Hambrecht
and Haefner, 2000; Linszen et al., 1994); and (c) among
first-episode cases of psychosis, those who used cannabis
were more likely to have a family history of psychosis
(McGuire et al., 1995).
According to this hypothesis, an increase in regular
cannabis use in the general population would not affect
the incidence of schizophrenia but it would reduce the
age of onset of psychotic illness among those who used
cannabis. That is, the incidence rates of persons using
cannabis would be ‘brought forward’. If this led to more
chronic psychotic disorders (e.g. because earlier onset
cases are more likely to relapse) the prevalence of
chronic cases of psychosis would increase. This would
increase the prevalence of regular cannabis use among
persons with schizophrenia.
1.3. Hypothesis 3: Cannabis use worsens the prognosis of
persons with schizophrenia
According to this hypothesis, cannabis use would
worsen the prognosis of schizophrenic persons by
increasing relapse to schizophrenia. It is supported by
evidence that persons with schizophrenia who use
cannabis are more likely to suffer a relapse (Jablensky
et al., 1991; Linszen et al., 1994). This hypothesis does
not predict an increased incidence of schizophrenia
among regular cannabis users. Instead, it predicts that
persons with schizophrenia who are regular cannabis
users will be more likely to have a relapse after their
initial episode. This could increase the number of
persons in the population with chronic schizophrenia.
It would not affect the age of onset of psychosis. The
prevalence of cannabis use among persons with schizophrenia would increase because there would be more
cannabis users among chronic cases.
1.4. Hypothesis 4: Regular cannabis use is more likely
among persons with psychosis
According to this hypothesis, persons with schizophrenia are more likely to become regular cannabis
users, if they use the drug (Mueser et al., 1998). There is
no causal relationship between cannabis use and psychosis, so increasing rates of cannabis use will have no
effect upon the incidence or prevalence of schizophrenia
and there would not be a change in age of onset. There
would be an increased prevalence of cannabis use
among persons with psychosis.
The predictions generated from each of these four
hypotheses are summarised in Table 1; notably, each
hypothesis has a unique set of predictions. We used
mathematical modelling to assess the plausibility of
these four hypotheses. We combined empirically derived
information about the epidemiology of cannabis use and
psychosis to predict trends in incidence, prevalence and
age of onset of schizophrenia according to each of these
hypotheses. These predicted trends were compared with
evidence on observed trends in schizophrenia and
cannabis use.
2. Method
2.1. Parameters for schizophrenia
2.1.1. Incidence
It was assumed that schizophrenia does not develop
before the age of 15 years (Goldstein et al., 1984), and
that new cases do not occur after the age of 54 years
(Goldstein et al., 1984). Separate specific incidence rates
were used for males and females because males have an
earlier onset of schizophrenia on average than females
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
39
Table 1
Predicted trends in schizophrenia, and in cannabis use among persons with schizophrenia, given an increase in the prevalence of regular cannabis use
in the general population
Trends in schizophrenia
(1)
(2)
(3)
(4)
Causal
Precipitation
Worse prognosis
Increased risk of dependence
Trends in cannabis use
Incidence
Prevalence
Age of onset
Prevalence among those with schizophrenia
/
0/
0/
0/
/
/
/
0/
¡/
¡/
0/
0/
/
/
/
/
Note: 0/ /no change; / /increase; ¡/ /decrease.
(Jablensky et al., 1991; Jones and Cannon, 1998).
Estimates of the average incidence rate of schizophrenia
per 100 000 population per year by age and gender were
obtained from a case register in New South Wales,
Australia (Goldstein et al., 1984) that covered a period
when cannabis was not widely used in Australia
(Donnelly and Hall, 1994).
2.1.2. Chronicity of schizophrenia
An earlier expert consensus view on the long-term
outcomes of schizophrenia in Australia (Hall et al.,
1985) was that: 25% of persons with schizophrenia
would have a ‘good’ outcome (a single episode with 60
days in hospital); 40% would have a ‘median’ outcome
(an average of 0.08 admissions to hospital per year for
life); and 35% would have a ‘poor’ outcome (0.16
admission per year for life).
More recent evidence has supported the estimate that
25% of patients will not relapse after long follow-up
periods (Eaton et al., 1992a,b; Mason et al., 1996). A
number of other studies have found that relapse tends to
be in the first few years after the initial episode, with
rates levelling off afterwards (Carone et al., 1991; Eaton
et al., 1992a,b; Mason et al., 1996). For the purposes of
the present study, we assumed that over a period of 4
years, 75% of incident cases of schizophrenia will relapse
and that 25% of cases will have a ‘good’ outcome (Eaton
et al., 1992a,b; Hall et al., 1985; Mason et al., 1996).
Relapse rates are higher for cases with an earlier age
of onset (Eaton et al., 1992a,b). Studies using a case
register in Victoria, Australia, found that persons with
the earliest age of onset (15 /19 age group) were most
likely to relapse, with the following relative risks
(compared to the 15 /19 age group) for older age
groups: 0.84 (20 /29), 0.73 (30 /39), 0.68 (40 /49), 0.59
(50 /59) (Eaton et al., 1992a,b). These data were used to
predict the probability of relapse (assuming an overall
relapse rate of 75%; Fig. 1).
2.1.3. Mortality in schizophrenia
A meta-analysis by Brown (1997) estimated that the
aggregate crude mortality rate of schizophrenia was 189
deaths per 10 000 population per year. The population
mortality rate was obtained from Australian Bureau of
Statistics and the rate in the general population was
assumed equal to the rate among non-schizophrenic
persons. For males the average rate of death was nine
per 10 000 males per year and for females it was five per
10 000 females per year.
2.2. Parameters for cannabis use
We examined the natural history of cannabis use
because it changes during a person’s lifetime. We
therefore needed estimates of prevalence of cannabis
use at each age over the life span of each of the birth
cohorts. By combining the estimates of the cumulative
lifetime prevalence of cannabis use with the pattern of
persistence of cannabis use from the cohort study, we
could estimate the number of people at each age in the
birth cohort who were still using cannabis.
Data on lifetime patterns of cannabis use were
obtained from two sources: a longitudinal study of the
natural history of cannabis use (Chen and Kandel,
1995); and an analysis of birth cohorts trends in drug
use (Degenhardt et al., 2000) derived from the Australian National Drug Strategy Household Survey
(NDSHS).
2.2.1. Natural history of cannabis use
Data on patterns of cannabis use in a longitudinal
study of cannabis use in the USA (Chen and Kandel,
1995) were used to estimate of the prevalence of monthly
cannabis use in Australian birth cohorts using data on
the lifetime prevalence of cannabis use in the 1998
NDSHS of the Australian population. The proportion
using cannabis at least monthly for each birth cohort
was estimated by multiplying the above rates by the
ratio of the proportion of persons in the birth cohort
who had used cannabis to the proportion in Chen and
Kandel’s cohort.
The modelling also took account of the substantial
decline in the age of first cannabis use among successive
birth cohorts in Australia (Degenhardt et al., 2000). The
mean age of first reported use of cannabis has decreased
40
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
Fig. 1. Estimated percentage of cases of schizophrenia relapsing according to age of onset of first episode.
by approximately 2 years with each successive birth
cohort. The following assumptions were made:
. that the curve for each birth cohort was that observed
by Chen and Kandel (1995);
. that each of these curves moved to the left by 2 years
for each successive birth cohort;
. that the absolute position of these curves could be
estimated by anchoring the birth cohort that was the
same as the cohort in Chen and Kandel’s study (i.e.
the 1955 /1959 birth cohort). The peak periods of
cannabis use for the 1965 /1969, 1970 /1974, 1975/
1979 birth cohorts between were estimated to be
between the ages of 15 and 20, compared to 17 /22
years for the 1960/1964 birth cohort, 19 /24 years for
the 1955/1959 birth cohort and so on;
. that there were no differences between birth cohorts
in the duration of monthly cannabis use. There were
no good data on birth cohort trends in the peak
period of use of cannabis, so this simpler assumption
was made. It is likely to reduce differences between
birth cohorts;
. it was assumed that the prevalence of weekly or more
frequent cannabis use was half of the proportion
reporting monthly or more frequent use.
2.2.2. Mortality of cannabis users
Our analyses assume that there was no increase in
mortality among cannabis users. Research has failed to
find increased mortality among cannabis using males
aged 34/36 years, after adjusting for alcohol and other
drug use (Andreasson and Allebeck, 1990) or among
cannabis using males and females aged 15 /49 years
(Sidney et al., 1997) over 8 years of follow up.
Details of formulae used to generate the models and
their predictions are provided in Appendix A.
2.3. Application to Australian population numbers
The size of each birth cohort (by gender) was
estimated from data published by the Australian Bureau
of Statistics on June 30th of each year. The cohort sizes
were estimated from the number in each year of birth
who were still alive at 15 years.
3. Results
3.1. Modelling the natural history of cannabis use
Fig. 2 shows the estimated natural history of cannabis
use in each of the birth cohorts. The peak prevalence of
regular cannabis use occurs earlier in recent birth
cohorts while peak prevalence of weekly use was higher
for earlier birth cohorts.
3.2. Modelling the prevalence of schizophrenia
Fig. 3 shows the estimated prevalence of schizophrenia among Australian males and females according to
age. The prevalence of schizophrenia by age 54 was
1.17% for males, and 1.08% for females. This is at the
higher end of the estimated prevalence of schizophrenia
(Jablensky et al., 1991; Robins and Regier, 1991) but it
corresponds to a point prevalence of schizophrenia in
1998 of 0.7% for the population born between 1940 and
1979. This is very similar to previous estimates of the
population prevalence of schizophrenia (Jablensky et
al., 2000, 1991; Robins and Regier, 1991).
3.3. Modelling the hypothesised relationships
3.3.1. Hypothesis 1: Causal relationship
On this hypothesis, the prevalence of schizophrenia by
age 25 years is estimated to be 0.38% among those in the
1940 /1944 birth cohort, compared to 0.43% in the
1975 /1979 birth cohort. The difference of 0.05% is a
14% increase in prevalence. At age 20 years, the
difference between the oldest and youngest birth cohorts
in the number of cases of schizophrenia */caused by
cannabis use */is 125 cases. The total would increase
from 736 males aged 20 years in the 1940 /1944 birth
cohort, to 861 in the 1975 /1979 birth cohort. This is an
increase of 17% (between the calendar years 1960 /1964
and 1995 /2000) in the number of cases aged 20 years
with schizophrenia coming to the attention of treatment
services.
Table 2 shows these results in terms of the number of
additional incident cases that would have occurred by
age 35 years on this hypothesis. Among the more recent
birth cohorts */those born from the 1960s and later */by
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
41
Fig. 2. Modelled prevalence of schizophrenia over the lifespan according to gender.
the time they were 35 years old, there would be an
additional 1225/1438 cases of schizophrenia per birth
cohort. This would be an increase in the number of
incident cases of schizophrenia of around 10% for each
birth cohort. The number of new cases in the later
cohorts (1225 cases) is almost 10 times larger than those
in the oldest birth cohort (180 additional cases).
3.3.2. Hypothesis 2: Cannabis precipitates schizophrenia
among vulnerable individuals
Table 3 shows the number of cases in each birth
cohort whose onset would occur a year earlier if
cannabis use precipitated schizophrenia. The age at
which this would have the most marked effect would
be age 14, when the only incident cases would be among
Fig. 3. Model of the natural history of cannabis use among Australian males and females by birth cohort.
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
42
Table 2
Hypothesis 1 */Modelled number of incident cases of psychosis by 35 years caused by cannabis use by the age of 35 years, by gender and birth cohort
1940 /44
1945 /49
1950 /54
1955 /59
1960 /64
1965 /69
1970 /74
1975 /79
Total incident cases by 35 years
assuming no link
Total incident cases by 35 years if cannabis use
caused psychosis
Number of incident cases by 35 years caused
by cannabis
Males
Females
Males
Females
Males
Females
3891
5444
5870
6572
7181
6995
7625
6948
2898
3903
4380
4896
5301
5197
5664
5128
3990
5759
6482
7373
7984
7768
8480
7689
2979
4051
4613
5289
5831
5748
6268
5631
99
315
612
801
803
773
856
742
81
147
233
387
511
532
582
483
Table 3
Hypothesis 2 */Modelled number of additional cases that would be
precipitated 1 year earlier by cannabis use by gender and birth cohort
1940 /1944
1945 /1949
1950 /1954
1955 /1959
1960 /1964
1965 /1969
1970 /1974
1975 /1979
Table 4
Hypothesis 3 */Modelled number of additional chronic cases of
psychosis due to cannabis use observed by the age of 35 years, by
gender and birth cohort
Cases precipitated at 14
years
Cases precipitated at 19
years
Additional chronic cases by age 35 years
Males
Females
Males
Females
Males
Females
0.2
0.4
1
10
31
49
54
48
0.01
0.05
0.07
1
16
16
18
17
2
13
32
46
44
44
49
44
0.5
2
5
13
19
20
22
20
12
38
68
82
77
70
76
63
10
17
25
39
49
50
54
43
cannabis users, and at age 19, when persons using
cannabis regularly have higher incidence rates. In the
1940 /1944 cohort, less than one case would have been
precipitated at age 14 years whereas this would rise to 50
cases by age 14 among the younger male cohorts.
3.3.3. Hypothesis 3: Cannabis worsens prognosis
According to the hypothesis that cannabis use worsens prognosis, there would be an additional 106 /130
chronic cases of schizophrenia caused by cannabis use
by age of 35 years in the more recent birth cohorts
(Table 4). However, these would comprise only 1% of all
chronic cases by this age. This is because relapse rates
among young adults were already very high so most
cases would relapse regardless of whether they used
cannabis use or not.
3.3.4. Hypothesis 4: Regular cannabis use is more likely
among persons with psychosis
Fig. 4 shows the predicted prevalence of weekly
cannabis use among persons with schizophrenia if such
persons are twice as likely as those in the general
population to become weekly users if they use cannabis
in the past year. The prevalence of weekly cannabis use
1940 /44
1945 /49
1950 /54
1955 /59
1960 /64
1965 /69
1970 /74
1975 /79
increases markedly among successive birth cohorts:
among males 5% of those aged 20 years, among the
1940 /1944 birth cohort would report weekly cannabis
use, compared to over 40% of those born after 1965. A
similar pattern is observed among females with schizophrenia.
3.4. Evaluation of the four hypotheses
The sections below discuss the available data on
trends in the incidence and prevalence of psychosis, in
the age of onset of psychosis, and in the prevalence of
cannabis use among persons with psychosis; and compare these data with the predictions of the four
hypotheses. Table 5 summarises the results of these
comparisons.
3.4.1. Trends in the incidence of psychosis
Numerous studies conducted in many countries,
including Australia (Parker et al., 1985), have reported
declines in the incidence of schizophrenia over the past
30 years (Eagles and Whalley, 1985; Geddes et al., 1993;
Joyce, 1987; Kendell et al., 1993; Munk-Jorgensen,
1995; Munk-Jorgensen and Mortensen, 1992; Suvisaari
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
43
Fig. 4. Hypothesis 4 */Modelled prevalence of weekly cannabis use among persons with schizophrenia by age and birth cohort.
with this evidence: all predicted that increases in the
prevalence of cannabis use would have little or no effect
upon the incidence of psychosis.
The evidence that the incidence of psychosis has
remained stable is consistent with hypothesis 2. So too
is: recent evidence that more cases of schizophrenia are
diagnosed as ‘drug-induced’ (although this could reflect
clinicians’ assumptions that substance use is precipitating the disorder, Brewin et al., 1997); and the fact that
cases in more recent birth cohorts have a younger
average age of onset (DiMaggio et al., 2001).
et al., 1999). This has not been universal, however, with
some reporting stable or increased rates (Bamrah et al.,
1992; Castle et al., 1991; Haefner and an der Heiden,
1986; Harrison et al., 1991). One study concluded that
incidence rates of psychosis in Australia had not
changed in the period 1848 /1978 (Haefner, 1987). It
appears unlikely that there has been an increase in the
incidence of schizophrenia in Australia. Given uncertainty about whether there has been a decrease in
incidence, the most conservative conclusion is that the
incidence rates of schizophrenia have remained stable and
possibly decreased over the past several decades. It is
unlikely that they have increased.
As hypothesis 1 predicted an increase in the incidence
of psychosis, the available evidence does not support
hypothesis 1. The other three hypotheses were consistent
3.4.2. Trends in the prevalence of psychosis
The data presented above also suggested that the
prevalence of psychosis has not increased. This was not
consistent with the increased prevalence of hypothesis 1,
Table 5
Consistency of predicted and actual trends in schizophrenia, and in cannabis use among persons with schizophrenia
Trends in schizophrenia
(1)
(2)
(3)
(4)
Causal
Precipitation
Worse prognosis
Increased risk of dependence
Trends in cannabis use
Incidence
Prevalence
Age of onset
Prevalence among those with schizophrenia
X
ª
ª
ª
X
?
ª
ª
?
?
?
?
?
?
?
?
Note: ª /evidence appeared to support the prediction of this hypothesis; X/evidence did not appear to support the prediction of this hypothesis;
?/there was insufficient evidence to determine the nature of the trends.
44
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
something that would certainly have been noted by case
registers.
It is more difficult to assess the validity of hypothesis
3, which predicted a very small (at most 1%) increase in
the number of chronic cases of schizophrenia by age 35
years. An increase of this size would be difficult, if not
impossible, to detect using existing epidemiological and
clinical data. This means that even if cannabis use
increases the rate of relapse, it will make a very small
difference to the number of persons with chronic
schizophrenia.
Hypotheses 3 and 4 did not predict any change in the
prevalence of psychosis. These predictions are also
consistent with the limited data on the prevalence of
schizophrenia.
3.4.3. Trends in the age of onset of schizophrenia
A recent study of first episode psychosis found a
lower age of onset in more recent birth cohorts
(DiMaggio et al., 2001). The evidence on the age of
onset among first-episode cases of schizophrenia who
use cannabis use is less certain. Some studies have found
that such cases had a significantly younger age of onset
than non-users of cannabis (Rolfe et al., 1993) but a
number of studies have not done so (Gut-Fayand et al.,
2001; McGuire et al., 1995).
The limited evidence on the average age of onset of
schizophrenia makes it difficult to draw any conclusions
about this indicator. Some evidence that the age of onset
of schizophrenia has decreased in more recent birth
cohorts is consistent with hypotheses 1 and 2 but clinical
samples of first-episode psychosis have not consistently
found that cannabis use is associated with an earlier
onset of psychosis. Better controlled studies may clarify
this issue.
3.5. Trends in the prevalence of cannabis use among
persons with schizophrenia
It is difficult to interpret evidence on changes in the
prevalence of regular cannabis use among schizophrenic
persons over the past three decades. First, many studies
report rates of cannabis use disorders in the lifetime
rather than the past year. Second, selection biases in
clinical samples (Berkson, 1946; Caron and Rutter,
1991; Galbaud Du Fort et al., 1993) make it difficult
to know whether variations in prevalence across different samples reflect changes in referral processes or
changes in prevalence of use. Third, there are few data
on the prevalence of cannabis use among persons with
schizophrenia in the Australian population. These have
shown higher rates of lifetime (Fowler et al., 1998) and
past year prevalence of dependence (Fowler et al., 1998)
than in the Australian population (Hall et al., 1999).
Because these studies are all recent, there is not much
that can be concluded about trends in the prevalence of
cannabis use among persons with psychosis in Australia.
A conservative assumption is that the prevalence of
cannabis use has increased among persons with schizophrenia at a similar rate to that in the general population of Australia over this period (Degenhardt et al.,
2000).
The data on trends in the prevalence of cannabis use
among persons with psychosis in Australia or anywhere
else are so limited that it is impossible to draw any
conclusions. The high rates predicted by hypothesis 4
are consistent with the findings in recent studies
(Degenhardt et al., 2000; Fowler et al., 1998; Jablensky
et al., 2000) but we do not know what rates of cannabis
use were in previous years. If we make the reasonable
assumption that rates of lifetime cannabis use have gone
up among persons with schizophrenia at the same rate
as in the Australian community (Donnelly and Hall,
1994), then these recent data are consistent with the
hypothesis that persons with schizophrenia are more
likely to become regular cannabis users than peers
without the disorder.
4. Discussion
4.1. Does cannabis use cause psychosis?
The hypothesis that cannabis causes schizophrenia
was not supported by the data on trends in the incidence
of this psychosis in Australia. There was no evidence
that there has been an increase in incidence over the past
30 years of the magnitude predicted by the hypothesis.
This suggests that cannabis use has not caused cases of
psychosis that would not otherwise have occurred . Even if
regular cannabis use did double the risk of users
developing schizophrenia (the ‘doubling’ of risk being
the best estimate), the prevalence of schizophrenia in the
population would increase from 1 to 2%. An increase of
1000 cases per birth cohort */as was predicted by our
modelling */would have been noticed in clinical settings.
The widespread discussion of apparent declines in the
incidence of schizophrenia suggests that this hasn’t
occurred. Even if some of the environmental risk factors
for schizophrenia have been reduced, such as poor
maternal nutrition, infectious disease, and poor antenatal and perinatal care (Eagles, 1991; Takei et al., 1996), it
seems unlikely that the decline in incidence from these
causes would have exactly offset an increase of 1000
incident cases per birth cohort predicted by the hypothesis that cannabis causes schizophrenia.
4.2. Does cannabis use precipitate psychosis?
This hypothesis is consistent with the evidence of a
reduction in the age of onset of psychosis among
persons born in more recent cohorts (DiMaggio et al.,
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
2001) and with some findings that first episode psychosis
cases who used cannabis were younger than non-users.
It would also explain the recent increase in the diagnosis
of ‘drug-induced’ psychoses (Brewin et al., 1997).
4.3. Does cannabis use worsen prognosis?
The third hypothesis made surprisingly little difference to the number of chronic cases that would be seen
by age 35 years. It is consistent with the elevated rates of
cannabis use among persons with psychotic illnesses,
and with the results of prospective studies that have
been carried out evaluating this issue.
4.4. Is regular cannabis use more likely among persons
with psychosis?
This hypothesis is consistent with the high prevalence
of cannabis use in Australian samples of persons with
psychosis. If we assume that rates of cannabis use
among persons with schizophrenia have gone up in
parallel with those in the Australian community (Donnelly and Hall, 1994), then these recent data are
consistent with this hypothesis.
4.5. Study limitations
Modelling of any trends such as those examined here
has limitations, since it is based upon assumptions that
may not be completely accurate. In the case of the
present paper, two issues in particular must be noted.
The first is potential changes in the potency of
cannabis use over time. This is an issue that has been
a matter of some debate in recent years in Australia, as
in other countries, with some claims that the THC
content of cannabis has increased 30-fold over the past
three decades. The data on this issue have been
examined by Hall and Swift, who concluded that the
limited evidence available suggested that the THC
content of cannabis may have increased by 3/4% over
this period (Hall and Swift, 2000). In any rate, if
cannabis use were a cause of psychosis de novo, an
increase in the potency of cannabis would be expected to
result in an increase in the prevalence of psychosis, given
the rise in cannabis use over the same period.
The second concerns changes in the classification of
schizophrenia over the period examined here. Over time,
the criteria used to define schizophrenia have become
increasingly based upon empirically validated and
rigorous definitions of the disorder. In particular, there
has been increasing precision with which subtypes of
psychotic illness have been defined. Unfortunately,
given the limitations of the data available, it is not
possible to examine trends in the clinical subtypes of
psychosis with any degree of confidence.
45
5. Conclusions
This study has used modelling (incorporating databased parameters) to predict what changes we would
expect to see in the incidence and prevalence of
schizophrenia if each of four hypotheses about the
relationships between cannabis use and psychosis were
true. The claim about cannabis and psychosis is widely
understood in the popular media and public debate in
Australia to imply that cannabis use has increased the
number of cases of psychosis in the population (in the
sense of causing cases of psychosis that would not
otherwise have occurred). It is therefore interesting that
using plausible assumptions, the present modelling
exercise suggests that (a) cannabis use as a cause of
cases of psychosis does not fit the data; and (b) it would
be difficult to detect any increases even if cannabis use
was a cause of incidence among those vulnerable to the
disorder.
Notably, if there were a common causal mechanism
for the association between cannabis use and psychosis,
whereby common factors increased the likelihood of
both cannabis use and psychosis, we would expect
to see increases in psychosis along with increases in
cannabis use. Since this was not the case, there does not
appear to be strong support for common causes
completely explaining the association that has been
observed.
The other three hypotheses provided a better
fit to the available data but because of data limitations
it was difficult to decide between them. If cannabis
use acts as a precipitant of psychosis, we would
have seen small increases in the number of early onset
cases. If cannabis use made relapse to psychotic
symptoms, we would have seen small increases in the
number of chronic cases. Finally, if persons with
psychosis were more likely to become regular
cannabis users, we would expect to see only a higher
prevalence of regular cannabis use in this population.
Future research needs to test these hypotheses in
prospective studies. The results of this study suggest
that persons at risk of psychosis may be advised of
this possible relationship and counselled against using
cannabis.
A similar approach to modelling may be useful in
empirically assessing the plausibility of hypotheses
about relationships between risk factors and the incidence and prevalence of other mental disorders in the
population.
Appendix A: Equations
C prevalence of regular cannabis use
I age-specific incidence rate of schizophrenia
R age-specific relapse rate of schizophrenia
46
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
Hypothesis 1: causal relationship
N(chronic cases atyear 3)
N(incident cases year 2)R0:25
It was hypothesised that weekly cannabis use doubled
the risk of developing schizophrenia */in other words,
that regular cannabis users had an incidence rate of
schizophrenia that was double that among persons who
did not use cannabis. This risk ratio is based on previous
work by Tien and Anthony (1990), Andreasson et al.
(1987), and the NSMHWB (Degenhardt and Hall,
2001).
N(incident cases atyear n)
(IC
n 2
n
I
n (1Cn ))N(without schizophrenia atyear n)
N(chronic cases atyear 2)
N(incident cases year 1)R0:5
N(chronic cases atyear 4)
N(incident cases year 3)R0:25
N(incident cases year 2)R0:5
N(incident cases year 1)R0:75
N(chronic cases atyear 5)
N(incident cases year 4)R0:25
N(incident cases year 3)R0:5
N(incident cases year 2)R0:75
N(incident cases year 1)R
N(incident cases year 1)R0:25
N(chronic cases atyear 3)
N(incident cases year 2)R0:25
Hypothesis 3: Cannabis use worsens prognosis
N(incident cases year 1)R0:5
N(chronic cases atyear 4)
N(incident cases year 3)R0:25
N(incident cases year 2)R0:5
N(incident cases year 1)R0:75
N(chronic cases atyear 5)
N(incident cases year 4)R0:25
N(incident cases year 3)R0:5
N(incident cases year 2)R0:75
N(incident cases year 1)R
It was assumed that the chance of relapse (i.e. the
occurrence of further psychotic episodes) was increased
by 2.5 times among weekly cannabis users. This was
based upon the findings of the Linszen and colleagues
study, which found that those using cannabis at least
weekly were 2.5 times more likely to relapse to psychotic
symptoms (Linszen et al., 1994).
The model also assumed that (a) there is no association between cannabis use and precipitation of psychosis; and (b) that the percentage of persons using
cannabis is initially the same among schizophrenic and
non-schizophrenic persons.
N(incident cases atyear n)
Hypothesis 2: Cannabis use precipitates psychosis among
vulnerable individuals
schizophrenia)
IN(without
n
N(chronic cases atyear 2)
This hypothesis assumes that there is no effect of
regular cannabis use upon overall incidence or chronicity of psychosis, but that among persons who use
cannabis there is a reduced age of onset of psychosis.
It was assumed that persons using cannabis develop
the illness 1 year earlier than those who do not use
cannabis regularly. This estimate was taken from the
study of Linszen and others in which those who used
cannabis were on average 1 year younger than those
who did not use cannabis (Linszen et al., 1994).
N(incident cases year 1)(2RCR(1C))=4
N(chronic cases atyear 3)
N(incident cases year 2)(2RCR(1C))=4
N(incident cases year 1)(2RC
R(1C))=2
N(chronic cases atyear 4)
N(incident cases year 3)(2RCR(1C))=4
N(incident cases atyear n)
schizophrenia atyear n)
IN(without
n
N(chronic cases atyear 2)
N(incident cases year 1)R0:25
N(incident cases year 2)(2RC
R(1C))=2
N(incident cases year 1)(2RC
R(1C))0:75
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
N(chronic cases atyear 5)
N(incident cases year 2)(2RCR(1C))=4
N(incident cases year 2)(2RC
R(1C))=2I(2RCR(1C))0:75
2
N(incident cases year 2)(2RCR(1C))
Hypothesis 4: Regular cannabis use is more likely among
persons with psychosis
This hypothesis assumes that there is no effect of
cannabis use upon either incidence or outcome (chronicity) of psychosis. The prevalence of regular (weekly)
cannabis use among persons with psychosis was assumed to be double that in the general population. This
is taken from research suggesting that regular or
dependent cannabis use is twice as likely among persons
who meet criteria for psychosis (Andreasson et al., 1987;
Tien and Anthony, 1990). This hypothesis assumes that
there is no effect of cannabis use upon either incidence
or outcome (chronicity) of psychosis. The prevalence of
regular (weekly) cannabis use among persons with
psychosis will be assumed to be double that in the
general population. This is taken from research suggesting that regular or dependent cannabis use is twice as
likely among persons who are likely to meet criteria for
psychosis (Andreasson et al., 1987; Tien and Anthony,
1990).
References
Allebeck, P., Adamsson, C., Engstrom, A., Rydberg, U., 1993.
Cannabis and schizophrenia: a longitudinal study of cases treated
in Stockholm County [published erratum appears in Acta Psychiatr. Scand. 1993 Oct; 88 (4) 304]. Acta Psychiatr. Scand. 88, 21 /
24.
Andreasson, S., Allebeck, P., 1990. Cannabis and mortality among
young men: a longitudinal study of Swedish conscripts. Scand. J.
Soc. Med. 18, 9 /15.
Andreasson, S., Allebeck, P., Rydberg, U., 1987. Cannabis and
schizophrenia: a longitudinal study of Swedish conscripts. Lancet
2, 1483 /1486.
Anthony, J.C., Helzer, J., 1991. Syndromes of drug abuse and
dependence. In: Robins, L.N., Regier, D.A. (Eds.), Psychiatric
Disorders in America. The Free Press, New York, pp. 116 /154.
Bamrah, J., Freeman, H., Goldberg, D., 1992. Epidemiology of
schizophrenia in Salford, 1974 /84: changes in an urban community over 10 years. Br. J. Psychiatry 159, 802 /810.
Basu, D., Malhotra, A., Bhagat, A., Varma, V.K., 1999. Cannabis
psychosis and acute schizophrenia */a case-control study from
India. Eur. Addict. Res. 5 (2), 71 /73.
Batel, P., 2000. Addiction and schizophrenia. Eur. Psychiatry 15 (2),
115 /122.
Berkson, J., 1946. Limitations of the application of fourfold table
analysis to hospital data. Biometr. Bull. 2, 47 /53.
Bernardson, G., Gunne, L.M., 1972. Forty-six cases of psychosis in
cannabis abusers. Int. J. Addict. 7, 9 /16.
47
Blanchard, J., Brown, S., Horan, W., Sherwood, A., 2000. Substance
use disorders in schizophrenia: review, integration, and a proposed
model. Clin. Psychol. Rev. 20 (2), 207 /234.
Brewin, J., Cantwell, R., Dalkin, T., Fox, R., Medley, I., Glazebrook,
C., Kwiecinski, R., Harrison, G., 1997. Incidence of schizophrenia
in Nottingham: a comparison of two cohorts, 1978 /1980 and
1992 /1994. Br. J. Psychiatry 171 (8), 140 /144.
Brown, S., 1997. Excess mortality of schizophrenia. A meta-analysis.
Br. J. Psychiatry 171, 502 /508.
Carney, M.W.P., Bacelle, L., Robinson, B., 1984. Psychosis after
cannabis use. Br. Med. J. 288, 1047.
Caron, C., Rutter, M., 1991. Comorbidity in child psychopathology:
concepts, issues and research strategies. J. Child. Psychol. Psychiatry 32 (2), 1063 /1080.
Carone, B., Harrow, M., Westermeyer, J., 1991. Posthospital course
and outcome in schizophrenia. Arch. Gen. Psychiatry 48, 247 /253.
Castle, D., Wesseley, S., Der, G., Murray, R., 1991. The incidence of
operationally defined schizophrenia in Camberwell, 1965 /84. Br. J.
Psychiatry 159, 790 /794.
Chen, K., Kandel, D.B., 1995. The natural history of drug use from
adolescence to the mid-thirties in a general population sample [see
comments]. Am. J. Public Health 85 (1), 41 /47.
Chopra, G.S., Smith, J.W., 1974. Psychotic reactions following
cannabis use in East Indians. Arch. Gen. Psychiatry 30, 24 /27.
Cuffel, B.J., Heithoff, K.A., Lawson, W., 1993. Correlates of patterns
of substance abuse among patients with schizophrenia. Hospital
Community Psychiatry 44, 247 /251.
Degenhardt, L., Hall, W., 2001. The association between psychosis and
problematical drug use among Australian adults: Findings from
the National Survey of Mental Health and Well-Being. Psychol.
Med. 31 (4), 659 /668.
Degenhardt, L., Lynskey, M., Hall, W., 2000. Cohort trends in the age
of initiation of drug use in Australia. Aust. N. Z. J. Public Health
24 (4), 421 /426.
DiMaggio, C., Martinez, M., Menard, J.-F., Petit, M., Thibaut, F.,
2001. Evidence of a cohort effect for age at onset of schizohprenia.
Am. J. Psychiatry 158 (3), 489 /492.
Donnelly, N., Hall, W., 1994. Patterns of cannabis use in Australia.
NCADA Monograph Series No. 27. Australian Government
Publishing Service, Canberra.
Eagles, J., 1991. Is schizophrenia disappearing? Br. J. Psychiatry 158,
834 /835.
Eagles, J., Whalley, L., 1985. Decline in the diagnosis of schizophrenia
among first admissions to Scottish mental hospitals from 1969 to
1978. Br. J. Psychiatry 146, 151 /154.
Eaton, W.W., Mortensen, P.B., Herrman, H., Freeman, H., Bilker,
W., Burgess, P., Wooff, K., 1992a. Long-term course of hospitalization for schizophrenia: Part I. Risk for rehospitalization.
Schizophr. Bull. 18 (2), 217 /228.
Eaton, W.W., Mortensen, P.B., Herrman, H., Freeman, H., Bilker,
W., Burgess, P., Wooff, K., 1992b. Long-term course of hospitalization for schizophrenia: Part II. Change with passage of time.
Schizophr. Bull. 18 (2), 229 /240.
Eva, J., 1992. Cannabis psychosis. Psychiatr. Bull. 16, 310 /311.
Fowler, I., Carr, V., Carter, N., Lewin, T., 1998. Patterns of current
and lifetime substance use in schizophrenia. Schizophr. Bull. 24 (3),
443 /455.
Galbaud Du Fort, G., Newman, S., Bland, R., 1993. Psychiatric
comorbidity and treatment seeking: sources of selection bias in the
study of clinical populations. J. Nerv. Ment. Dis. 181 (8), 464 /474.
Geddes, J., Black, R., Whalley, L., Eagles, J., 1993. Persistence of the
decline in the diagnosis of schizophrenia among first admissions to
Scottish hospitals from 1969 to 1988. Br. J. Psychiatry 163, 620 /
626.
Goldstein, G., Hall, W., Andrews, G., 1984. The incidence of
schizophrenia in New South Wales, Australia. Acta Psychiatr.
Scand. 70, 220 /227.
48
L. Degenhardt et al. / Drug and Alcohol Dependence 71 (2003) 37 /48
Gruber, A.J., Pope, H.G., 1994. Cannabis psychotic disorder: does it
exist? Am. J. Addict. 3, 72 /83.
Gut-Fayand, A., Dervaux, A., Olie, J.-P., et al., 2001. Substance abuse
and suicidality in schizophrenia: a common risk factor linked to
impulsivity. Psychiatry. Res. 102, 65 /72.
Haefner, H., 1987. Epidemiology of schizophrenia. In: Haefner, H.,
Gattaz, W., Janzarik, W. (Eds.), Search for the Causes of
Schizophrenia. Springer-Verlag, Berlin.
Haefner, H., an der Heiden, W., 1986. The contribution of European
case registers to research on schizophrenia. Schizophr. Bull. 12,
26 /51.
Hall, W., 1998. Cannabis use and psychosis. Drug Alcohol Rev. 17,
433 /444.
Hall, W., Degenhardt, L., 2000. Cannabis use and psychosis: a review
of clinical and epidemiological evidence. Aust. N. Z. J. Psychiatry
34, 26 /34.
Hall, W., Swift, W., 2000. The THC content of cannabis in Australia:
evidence and implications. Aust. N. Z. J. Public Health 24, 503 /
508.
Hall, W., Goldstein, G., Andrews, G., Lapsley, H., Bartels, R., Silove,
D., 1985. Estimating the economic costs of schizophrenia. Schizophr. Bull. 11, 598 /611.
Hall, W., Teesson, M., Lynskey, M., Degenhardt, L., 1999. The 12month prevalence of substance use and ICD-10 substance use
disorders in Australian adults: Findings from the National Survey
of Mental Health and Well-Being. Addiction 94 (10), 1541 /1550.
Hambrecht, M., Haefner, H., 2000. Cannabis, vulnerability, and the
onset of schizophrenia: an epidemiological perspective. Aust. N. Z.
J. Psychiatry 34, 468 /475.
Harrison, G., Cooper, J., Gancarczyk, R., 1991. Changes in the
administrative incidence of schizophrenia. Br. J. Psychiatry 159,
811 /816.
Jablensky, A., McGrath, J., Herrman, H., Castle, D., Gureje, O.,
Evans, M., Carr, V., Morgan, V., Korten, A., Harvey, C., 2000.
Psychotic disorders in urban areas: an overview of the Study on
Low Prevalence disorders. Aust. N. Z. J. Psychiatry 43, 221 /236.
Jablensky, A., Sartorius, N., Ernberg, G., 1991. Schizophrenia:
manifestations, incidence and course in different cultures. A World
Health Organization Ten-Country Study. Psychological Medicins
Supplement No. 20.
Jones, P., Cannon, M., 1998. The new epidemiology of schizophrenia.
Psychiatr. Clin. North Am. 21 (1), 1 /25.
Joyce, P., 1987. Changing trends in first admissions and readmissions
for mania and schizophrenia in New Zealand, 1974 /1984. Aust. N.
Z. J. Psychiatry 21, 82 /86.
Kendell, R., Malcolm, D., Adams, W., 1993. The problem of detecting
changes in the incidence of schizophrenia. Br. J. Psychiatry 162,
212 /218.
Linszen, D.H., Dingemans, P.M., Lenior, M.E., 1994. Cannabis abuse
and the course of recent-onset schizophrenic disorders. Arch. Gen.
Psychiatry 51, 273 /279.
Makkai, T., McAllister, I., 1998. Patterns of drug use in Australia,
1985 /1995. Australian Government Publishing Service, Canberra.
Mason, P., Harrison, G., Glazebrook, C., Medley, I., Croudace, T.,
1996. The course of schizophrenia over 13 years. A report from the
International Study on Schizophrenia (ISoS) coordinated by the
World Health Organization. Br. J. Psychiatry 169 (5), 580 /586.
Mathers, D., Ghodse, A., Caan, A., Scott, S., 1991. Cannabis use in a
large sample of acute psychiatric admissions. Br. J. Addict. 86,
779 /784.
McCoy, A., 1980. Drug Traffic: Narcotics and Organised Crime in
Australia. Harper and Row, Sydney.
McGuire, P., Jones, R., Harvey, I., Williams, M., McGuffin, P.,
Murray, R., 1995. Morbid risk of schizophrenia for relatives of
patients with cannabis associated psychosis. Schizophr. Res. 15,
277 /281.
McKay, D.R., Tennant, C.C., 2000. Is the grass greener? The link
between cannabis and psychosis. Med. J. Aust. 172 (6), 284 /286.
Mueser, K., Yarmnold, P., Levinson, D., Singh, H., Bellack, A., Kee,
K., Morrison, R., Yadalam, K., 1990. Prevalence of substance
abuse in schizophrenia: demographic and clinical correlates.
Schizophr. Bull. 16, 31 /56.
Mueser, K.T., Drake, R.E., Wallach, M.A., 1998. Dual diagnosis: a
review of etiological theories. Addict. Behav. 23 (6), 717 /734.
Munk-Jorgensen, P., 1995. Decreasing rates of incident schizophrenia
cases in psychiatric services: a review of the literature. Eur.
Psychiatry 10, 129 /141.
Munk-Jorgensen, P., Mortensen, P., 1992. Incidence and other aspects
of the epidemiology of schizophrenia in Denmark, 1971 /1987. Br.
J. Psychiatry 161, 489 /495.
Parker, G., O’Donnell, M., Walter, S., 1985. Changes in the diagnoses
of the functional psychoses associated with the introduction of
lithium. Br. J. Psychiatry 146, 377 /382.
Robins, L.N., Regier, D.A. (Eds.), 1991. Psychiatric disorders in
America: The Epidemiological Catchment Area study. New York:
The Free Press.
Rolfe, M., Tang, M., Sabally, S., Todd, J., Sam, E., Hatib N’Jie, A.,
1993. Psychosis and cannabis abuse in the Gambia: a case-control
study. Br. J. Psychiatry 163, 798 /801.
Rosenthal, R., 1998. Is schizophrenia addiction prone? Curr. Opin.
Psychiatry 11, 45 /48.
Sidney, S., Beck, J.E., Tekawa, I.S., Quesenberry, C.P., Jr, Friedman,
G.D., 1997. Marijuana use and mortality. Am. J. Public Health 87
(4), 585 /590.
Solomons, K., Neppe, V.M., Kuyl, J.M., 1990. Toxic cannabis
psychosis is a valid entity. S. Afr. Med. J. 78, 476 /481.
Suvisaari, J., Haukka, J., Tanskanen, A., Lonnqvist, J., 1999. Decline
in the incidence of schizophrenia in Finnish birth cohorts born
from 1954 to 1965. Arch. Gen. Psychiatry 56, 733 /740.
Takei, N., Lewis, G., Sham, P., Murray, R., 1996. Age-period-cohort
analysis of the incidence of schizophrenia in Scotland. Psychol.
Med. 26, 963 /973.
Tennant, F.S., Groesbeck, C.J., 1972. Psychiatric effects of hashish.
Arch. Gen. Psychiatry 27, 133 /136.
Thornicroft, G., 1990. Cannabis and psychosis: is there epidemiological evidence for an association? Br. J. Psychiatry 157, 25 /33.
Tien, A.Y., Anthony, J.C., 1990. Epidemiological analysis of alcohol
and drug use as risk factors for psychotic experiences. J. Nerv.
Ment. Dis. 178 (8), 473 /480.
Wylie, A.S., Scott, R.T.A., Burnett, S.J., 1995. Psychosis due to
‘skunk’. Br. Med. J. 311, 125.
Ziedonis, D., Trudeau, K., 1997. Motivation to quit using substances
among individuals with schizophrenia: implications for a motivation-based treatment model. Schizophr. Bull. 23, 229 /238.
Typesof Marijuana Usersby Longitudinal Course*
DENISE B. KANDEL,
PH.D.,? AND KEVIN CHEN, PH.D., M.P.H.?
New YorkStatePsychiatricInstitute,1051 RiversideDrive, Unit 20, New York,New York10031
ABSTRACT. Objective:Taxonomiesof alcoholismand antisocial
behaviorsbasedon developmentalcourseconvergeon two-group
classifications
that emphasizeearly and late onset.Typologiesfor
usersof illicit drugsremain to be developed.This article proposesa
developmental
taxonomyof marijuanausers.Method:Clusteranalysis was applied to a representativecommunitysampleof 708 (364
male, 344 female) marijuanausersfollowed from adolescenceto age
34-35. The Ward method,followed by relocation,was usedto classify marijuana users into different types based on age of onset,
chronicityof heavy use and persistence
of use. ANOVA and logit
analyseswereutilizedto describetheclustersolutionandexaminethe
correlatesof clustermembership.Results:Four marijuanauseclusterswereidentified:earlyonset-heavy
use,earlyonset-lightuse,mid
onset-heavyuseand late onset-lightuse.The groupsdifferedfrom
HESTUDY
OFdevelopmental
trajectories
ofinvolve-
ment in the useof drugscanbe approachedusingthree
strategies.One strategyidentifiespathwaysof progression
from one drugclassto another,includinglegal, illegal and
medicallyprescribed
psychotropic
drugs(Ensmingeret al.,
1982; Fleming et al., 1989; Huba, 1983; Kandel and Yamaguchi, 1993; Kandel et al., 1992; Mills and Noyes, 1984;
Windie et al., 1989). A secondstrategyinvestigatesincreasinginvolvementwithinonedrugclass,fromexperimentation
to casualuse, abuse,dependence,desistenceand relapse
(Chen and Kandel, 1995; Raveis and Kandel, 1987). A third
strategyinvestigatestaxonomiesbased on developmental
course.
Much research has been carded out to define taxonomies
for alcoholism (Babor, 1996; Babor et al., 1992, 1994;
Cloningeret al., 1981, 1986; Weber et al., 1989; Zucker et
al., 1994, 1996)and antisocialbehavior (Achenbach,1993;
eachotherin degreeof involvementin marijuanaandotherdrugs,sociodemographic
and lifestylecharacteristics.
The majorityof those
with early onset did not become heavily involved in marijuana.
Uniquefactorswereassociated
with membership
in eachgroup.Factorsdifferentiating
earlyfrommid-onsetheavyuseincludedassociation with marijuana-using
peersand havinghad a mentaldisorder.
Peerdelinquency
wasanadditionalfactordifferentiating
earlyinitiatorswho becameheavyusersfrom thosewho did not. Conclusions:A
simpletwo-typeclassificationfails to take into accountthe heterogeneityof earlyandlateonsetgroups.By itself,earlyonsetintomarijuanawill not lead to problematicuseor rapid progression
into the
useof otherdrugs.Motivationunderlyinguseanddysfunctional
behaviorsare associatedwith the developmentof problematicdruguse
anddependence.
(J. Stud.Alcohol61: 367-378, 2000)
DiLalla and Gottesman,1989; Hinshaw et al., 1993; Moffitt,
1993;Moffitt et al., 1996;Quay, 1987).The formertypically
hasbeenconducted
onindividualsdependent
onalcohol.For
bothalcoholismandantisocialbehaviorthereis convergence
on two-groupclassifications
that emphasizeearly andlate
onset(Babor,1996).Cloningeret al. (1986) describetwo geneticsubtypesof alcoholism:
Type I Milieu-Limitedalcoholism,characterized
by adultonsetand no criminalityand
Type II Male-Limitedalcoholism,
characterized
by teen-age
onset and criminal involvement. Similarly, Babor et al.
(1992) identifyA andB alcoholtypesdifferentiated
by age
of onset,childhoodriskfactorsandcurrentpsychopathology.
Zucker(1994) proposes
hisownversionof theA-B classification:AntisocialAlcoholismandDevelopmentallyLimited
Alcoholism. Moffitt's (1993) distinctionbetween "LifeCourse Persistent" and "Adolescence-Limited"
antisocial
behaviorconstitutes
a very similartypologyfor antisocial
behavior.
Received:February 18, 1999. Revision:June 11, 1999.
*This researchwas supported,in part, by National Instituteon Drug
Abuse (NIDA) grants DA01097, DA02867, DA03196, DA04866,
DA09110
and NIDA
Research Scientist Award
DA00081
to Denise B.
Kandel.Partialsupportfor computercostswasprovidedby NationalInstitute of Mental Health/Mental Health Clinical ResearchCenter grant
MH30906 to theNew York StatePsychiatricInstitute.
*DeniseB. Kandelis alsowith theDepartmentof PsychiatryandSchool
of PublicHealthat ColumbiaUniversity,New York, NY. Kevin Chen,formerlywith theDepartmentof PsychiatryandSchoolof PublicHealthat ColumbiaUniversity,is nowwiththeDepartmentof Psychiatry,
UMDNJ-New
JerseyMedical School,Newark,NJ.
367
Taxonomiesfor usersof illicit drugsremainto be developed.No cleartheoretical,clinicalor empiricalhypotheses
are availableto guidework in this area.We relied on clusteringtechniques
to identifya developmental
taxonomyof
marijuanausersbasedonlongitudinal
course.Theresultsdependon the variablesincludedto define the clusters,the
methodandcriteriausedto indexdissimilaritybetweenindividualsanddefinethe clusters,the procedures
usedto validate the clusters,and the interpretationof the results
(BergmanandMagnuson,1987).Cliniciansandepidemiologistsdiffer, philosophically,
regardingthe variablesto be
368
JOURNAL
OF STUDIES
ON ALCOHOL
included as a basis •or classification. The differences are
rootedin the dissimilargoalsof cliniciansand epidemiologists:treatmentmatchingin one case,elucidationof causal
processes
in the other.Cliniciansinclude,amongthe distinguishingcharacteristics
of thetypes,variablesthatarecauses
or consequences
of the behaviorsof interest,therebyprecludinggaininga clearunderstanding
of theiretiologies.
One
differencebetweenthe two typesof alcoholismdefinedby
Cloningeret al. (1981, 1986)andothers,for example,is the
presence
or absenceof delinquency.
The variablesselected
for the presentclusteranalysiswere restrictedto thosedescribingtheindividual'shistoryof marijuanainvolvement.
The goalsof our studyare: (1) to identifydevelopmental
taxonomiesof marijuanausein a generalpopulationsample
and (2) to examine factors associatedwith each identified
subtype.We relied on etiologicalresearchon adolescent
drug use (Brook et al., 1990; Clayton, 1992; Costaet al.,
1999; DeWit et al., 1995; Felix-Ortiz and Newcomb, 1992;
Hawkins et al., 1992; Jessoret al., 1995; Kandel, 1984; Kan-
del et al., 1978; Stacyet al., 1992) to selectvariablesthat
wouldbe associated
with membership
in a specificdevelopmentalcluster.Factorspredictiveof substance
usebelongto
fourdomainsof variables:individual,family,peergroupand
more distalcontextualfactors.It is assumedthat strongattachmentandcommitmentto the major socialinstitutionsin
the youth'slife (i.e., family and school)fosterthe adoption
of conventionalnormsand beliefs and protectagainstrisk
factorsfor drug use, includinginvolvementwith deviant
drug-usingpeersandparticipationin delinquencyactivities.
Familial influencesincludeoverall family climate,the quality of parent-childinteraction,parentaldrug use and psychopathology.
The role of deviantpeersin the etiologyand
maintenance
of druguseis oneof thebestreplicatedfindings
in drugresearch.Characteristics
of the child, in particular
difficult temperament,
early childhoodaggression
and adolescentdepressive
symptoms,are associated
with the developmentof drug use. Overall, four risk factorshave been
foundto be mostsalient(directlyor indirectly)in the developmentof adolescent
substance
use:poorparenting,particularly lack of monitoringand low closeness;
parentaldrug
use;association
with a drug-using
peergroup;andthechild's
priorbehavioraldifficultiesanddelinquency.
We considered
theserisk andprotectivefactorsin relationto marijuanacluster membership,althoughnot all relevantvariableswere
availablein our study.
Method
Sample
Data are derived from adults(N = 708; 364 male, 344 fe-
male) who ever usedmarijuanaat least 10 times from the
New York StateFollow-Up Cohort.The cohortconstitutes
a
representativesampleof former adolescents,enrolled in
/ MAY
2000
grades10 and 11 in 18 New York Statepublichighschools
in 1971-72,who were followedover 19 years,to age34-35
(Kandel et al., 1976). Studentswho had not participatedin
the initial surveyand, presumably,were chronicabsentees,
werealsoselected
topermitunbiased
estimates
Oftheformer
studentpopulationat thefollow-ups.Respondents
werefirst
contactedin 1971 (age 15-16) and reinterviewedin 1980
(age24-25), 1984(age28-29) and1990(age34-35). Of the
initial targetsampleof adolescents
still alive, 1,160(71.7%)
werereinterviewedin 1990.Informedconsentwasgivenfor
participationin the study.
Datawereobtainedthroughstructured
personalhousehold
interviewsthat includedtwo chartsdesignedto reconstruct
the respondent's
life and drughistorieson a monthlybasis.
The drughistorieswereobtainedfrom respondents
who had
everusedeachdrugat least10 times.At eachinterviewrespondents
were askedabouttheirfrequencyandquantityof
useof eachdrugduringthepast12months,themonthswhen
they usedeachdrug sincethe last interview,the periodsof
highestuse,andfrequencies
andquantitiesusedin thosepehods.Two analyticalsampleswereused.Descriptiveanalyseswerebasedon subjects(N = 708) who usedmarijuanaat
least10 timesandreportedan ageof firstmarijuanause;this
numberrepresents81.4% of all thosewho reportedever using marijuanaby ages34-35. By the time of the initial survey, 40% hadstartedusingmarijuana.Multivariateanalyses
were basedon thosewho participatedin the initial school
surveyandprovidedanswersat eachof thethreeinterviews
(n = 589).
Analytical strategies
Clusteringprocedure.The clusteranalysiswas basedon
three variables that describedmarijuana use from adolescenceto middle adulthood:age of onset,monthsof near-
dailymarijuanauseandusein thelastyearpreceding
thelast
interview.
The choiceof a clusteringmethodis a criticalissue.MilliganandCooper(1985) reviewedthe30 mostpopularcluster methods in a Monte Carlo simulation and concluded that
no existing procedureprovided sufficient information to
identify the numberof clustersin the data. Three statistics
performedbestfor recoveringthetrueclusterstructurein the
artificial data:a pseudoF statistic(Calinskiand Harabasz,
1974); the J½(2)/J½(1)
statistic(Duda and Hart, 1973), transformedby SAS into a pseudot2 statistic;andSarle's(1983)
cubicclusteringcriterion(CCC). The statisticsindicatethe
statisticalsignificanceof the changedue to each specific
combination
but not which solution is the best in absolute
terms. A good-fittingsolutionshouldexplain at least two
thirdsof the variancein the clusteringvariables(Bergman,
1998).
Ward's (1963) minimum-variance hierarchical method
was implementedand estimatedby SAS. The procedure
KANDEL
AND
combines the two closest cases or clusters in terms of their
standardized
scores
onthecriteriavariables,
step-by-step
until all casesare groupedinto a singlecluster.The estimated
similaritycoefficientis thesquared
Euclideandistanceto the
cluster means. These distances are summed for all the cases.
At eachstep,the two casesor clustersthatmergeare those
that result in the smallest increase in the overall sum of the
squaredwithin-clusterdistances.
At eachstep,the threestatisticsdescribed
abovearealsoprovided.A changeof directionor a largechangein valuesof thethreestatistics
signala
significantdifferencebetweentwo solutions(Aldenderfer
and Blashfield, 1984; SAS Institute,1990). To furthermini-
mize the distancewithin eachclusterandincreasethe proponion of varianceexplainedby the solutionselected,we
appliedthek-meanclustertechnique
to relocatesomeof the
cases,while retainingthe clustercentergeneratedby the
Ward method(Bergman,1998).To validatethe solution,we
examinedthe differentiationof the clustersby individuals'
sociodemographic
characteristics,
and patternsand consequencesof marijuanause(ANOVA).
Multivariate
models
We estimatedmultivariatelogisticregressions
to identify
the uniquepredictorsof membership
in eachcluster.Since
manypredictors
weremeasured
at age 15-16,the multivariate modelswere estimatedon the 589 marijuanauserswho
everusedmarijuanaat least10 timesandparticipated
in the
initial schoolsurvey.As comparedwith theseparticipants,
userswhohadbeenabsentfrom schoolat theinitial survey
were more likely to be male (61% vs 49%, p < .05), black
(18% vs 6%, p < .01) andof lowereducation(meanyearseducation= 13.2vs 14.5, p < .01).Theytendedto reportmore
extensivelifetime experiencewith marijuanathan participants(used1,000+times:39.5% vs 27.7%, p < .05), to be
morelikely to havebeennear-dailyusers(53.8% vs 42.9%,
p < .05), andto haveexperienced
a greaternumberof drugrelatedproblemsby 1990 (p < .05); theyhad lower school
gradesand were moredelinquentin high school(p < .05).
There were no significant differencesbetween the two
groupson agesof onsetof marijuana,alcoholor cigarette
use,proportionof lastyear marijuanausers,and numberof
monthsusedmarijuananear-daily.
CHEN
369
centrelationships
in adolescence;
drug usein the person's
immediatesocialnetworkat ages15-16 and24-25; conventionality/deviance
in adolescence;
personalitycharacteristics
and psychiatricproblem(seeTables 1-3). All cumulative
drugconsumption
measuresfrom 1971to 1990 were calculated on the retrospective
monthlydrug use historiesobtainedat eachfollow-up interviewin 1980, 1984 and 1990.
Marijuana near-daily was definedas use4 or more daysa
week.Highestfrequencyof marijuanausewascodedfrom 1
= 1-2 timesa yearto 8 = daily. Ever usedillicit drugsother
than marijuana (OLD) includedever usedat least 10 times
cocaine/crack,
heroin,psychedelics
or, nonmedically,minor
tranquilizers,sedativesand stimulants.Numberof drug-related problems counteddrug-specificproblemschecked
from a list of 11 problems(e.g., with health, hurt performancein schooland/oron the job, lessenergy,made depressed, got into trouble with the police). Last-year
dependence
on alcoholwas basedon approximatemeasure
of DSM-IV alcoholdependence(Kandelet al., 1997). Most
importantreasons
for usingmarijuanain thepast12 months
classified11 reasonsinto threetypes.Eachrespondent
was
scoredpositivelyfor thattypeif checkedanycomponent
reason:(1) for socialreason,2 items (e.g., "to go alongwith
whatmy friendsaredoing");(2) to reducenegativefeelings,
3 items(e.g., "to overcomedepression");
(3) to enhancepositive feeling,4 items(e.g., "to get pleasure,feel good,get
high"). Frequencyof churchattendancecoded1 = neverto
7 = almostdaily. Last-yeardelinquencyindexcountedpositive responses
to a list of 15 delinquentbehaviors(e.g.,gottenintoa seriousfightin schoolor at work,takensomething
from a storewithoutpaying for it, forgedor passedbad
checks).Depressivesymptoms(Kandel and Davies, 1982)
wasa six-itemscale(e.g.,feelingtootiredto do things,feelingunhappy,feelingnervous).With theexceptionof lifetime
experience
of psychiatric
problemsfor selfandfamily members,reasonsfor usingmarijuanaanddruguseby peersand
delinquencyat ages 24-25, the predictivevariableswere
measuredat age 15-16.The reasonsunderlyingusewerenot
availablein 1971 for all users,sinceonly 38.4% had started
usingmarijuanaby that time. Variablesmeasured
in adulthoodcouldbe predictorsor consequences
of belongingto a
particularcluster.
Results
Measures
A four-type clustersolution
Selectedsociodemographic
and drug historyvariables
were used to describe and validate the clusters in ANOVA.
Variables,mostlymeasuredin adolescence,
wereincludedas
predictorsof clustermembership
in multivariatelogisticregressions.
There were 19 variablesmeasuring:sociodemo-
graphiccharacteristics;
drugusehistory;family historyof
druguseandpsychiatric
disorders;
qualityof parent-adoles-
Three marijuanause variableswere usedto identify the
clusters:age of onsetinto marijuana,extentof chronicuse
and persistenceof use by age 34-35 (see above).Of total
users(N = 708), 45% ever usedmarijuananear-daily;23%
werestill usingmarijuanaby age34-35. The two-clusterand
three-cluster
solutionsdifferedsignificantlyfrom eachother
370
JOURNAL
OF STUDIES
ON ALCOHOL
by thethreestatistics
generated
by theWard minimumvariance method, and the four-cluster solution differed from the
three-cluster
solutionby two statistics
(pseudoF andpsuedo
t2).The four-clustersolutionexplainedtwo thirds(67.8%) of
thetotalvariance,while thethree-cluster
solutionexplained
55.2%. We selectedthe four-cluster solution, which distin-
guishedtwo typesof early onsetusers:thosewho became
heavyusersandthosewhoremainedlightusers.Thesetwo
groupswerecombinedin the three-cluster
solution.After kmeanrelocationof the four-clustersolution,the overall R2increasedto 70.9%; 9.7% of caseswere relocated.
To examinethe reliabilityof the clustersolution,we repeatedthe analyseson two randomsplit subsamples,
and
amongmenandwomenseparately.
The four-cluster
solution
was one of the bestin termsof the critical statistics(datanot
shown).While the clusteringpatternswere identicalin the
varioussubsamples,
the proportionsof individualsin each
cluster varied. The characteristics of individuals in each clus-
terweresimilarin eachsplitsubsample.
We alsoapplieddifferentclustering
procedures
(i.e.,simplelinkageandaverage
linkagebetweengroupsandobserved
similarfour-cluster
solutions). These resultsincreasedour confidencein the robusiness of the four-cluster solution.
Therearetwo early-onsetgroups;bothstartedaroundage
15, but they had differentsubsequent
lifetime patternsof
marijuanause(Table 1). In onegroup(Group 1, early onsetheavy use), all becamenear-dailyusersand half (49.2%)
were still usingmarijuanaat age 34-35. By contrast,in the
other group (Group 2, early onset-lightuse), only half
(44.0%) becamenear-dailyusersandonly 10% persisted
in
usingmarijuanaby age34-35. A thirdgroup(Group3, mid
onset-heavyuse) started 1 year later than the early-onset
groups(aroundage 16); two thirds (67.2%) becameneardaily usersandall were still usingmarijuanaat age 34-35.
The fourthgroup(Group4, lateonset-lightuse)startedusing
marijuanaat age 19.5 (4.5 yearslater than the early onset
groups);a muchlowerproportion(21.3%) of thatgroupthan
of any other becamenear-dailyusersand only 0.8% remaineda userinto adulthood.Most usersfell in thelight-using groups,evenly dividedbetweenearly (35.3%) and late
(37.1%) starters.A minority(8.6%) fell in the early onsetheavyusegroup.
No detaileddrug historywas obtainedfrom marijuana
userswho everusedmarijuanafewerthan10 timesandthey
werenotincludedin theclusteranalysis.By definition,these
werevery light users;only 0.6% were still usingmarijuana
at age 34-35.
Characteristicsof thefour clustersof users
Drug histories.The four groupsdiffered significantly
from eachotheron 6 of the 17 drughistorycharacteristics
thatwere not usedin the clustering(Table 1). Heavy users
/ MAY
2000
differedaccording
to early-or mid-onseton 8 of theremaining 11 variables;lightusersdifferedaccording
to ageof onseton 4 of these11 variables.To placethesemarijuanausers
in a broadercontext,we also displaythe characteristics
of
thosewho only ever usedmarijuanaat most 9 times, and
thosewho neverusedmarijuana.
The early onset-heavy
usegroup(Group1) startedexperimentingwith marijuanathe earliestandits membersexperiencedthe highestfrequencyof marijuanause.All became
near-dailyusersanddidsoby age17.5,earlierthananyother
group.Exceptfor thelatestarters(Group4), theintervalbetweentheagesof firstandnear-dailyuseincreased
by about
a yearfor eachsuccessive
groupof users.Amonglate-onset
marijuanausers(Group4), a minority progressed
to neardailyuserapidly.Thetwoheavy-using
groups(Groups1 and
3) had highernumbersof near-dailymarijuanause spells
thanthetwo light-usinggroups.The chronicityof the spells
of near-dailyusein theearlyonset-heavy
usegroupis striking: 93.1 monthson averageversus18.5 to 26.1 monthsin
the otherthreegroups.
Thefourgroupsalsodifferedsignificantly
fromeachother
regardingtheirinvolvementin alcohol,cigarettesandother
illicit drugs(OIDs). Theearly-heavygroupbecamethemost
involvedin all substances,
exceptheavydrinking.The latelight groupwas the leastinvolvedin almosteveryclassof
drugs, including cigarettesand dependenceon alcohol.
Heavymarijuanausers(Groups1 and3) weremostlikely to
belong-termheavysmokers
anddrinkers,irrespective
of age
of onsetintomarijuana.The groupsareconsistently
arrayed
from heaviestto lightestinvolvementin OID use. While
morethan90% of theearly-heavygroupeverexperimented
with OIDs and had doneso by age 16.9 yearson average,
fewer thana quarterof the late-lightgroupever did so, and
they did so 5 yearslater (21.8 years).Progression
from the
use of a legal drugto the useof marijuanatook longerfor
eachsuccessive
group,fromearly-heavy(3 years),to earlylight (2.9 years),mid-heavy(4 years)andlate-lightgroup(6
years).Thegroupsalsodifferedsignificantly
fromeachother
in age of onsetand numberof yearsthey usedOIDs. Furthermore,theearly-heavygroupreporteda highernumberof
drug-relatedproblems,andmarijuanaproblemsspecifically,
thanthemid-heavyuserswhostarteda yearlater.Theywere
alsomorelikely thanany othergroupto reporteverhaving
felt addictedto any drug,to meetcriteriafor alcoholdependence,to reportthe longestperiodof heavysmoking,treatment for a drug-relatedproblem, and having had a
psychiatric
problem.
The two light-usinggroups(Groups2 and4) weresignificantlydifferentfromeachotherin theproportions
smoking
cigarettesheavilyand havingbeentreatedfor drug-related
problems.
With rare exceptions,thosewho usedmarijuanafewer
than 10 timesand were not includedin the clusteringwere
KANDEL AND CHEN
TABLE 1.
371
Drugusehistory
anddrug-related
problems
inthefourclusters
of marijuana
(mja)users
(N= 708)
Mja userclusters
Group1
Earlyheavy
Drug usehistory
Clusteringvariables
Any mja use
by age34-35 (%)
Monthsusedmja
near-daily2
49.2 a,b
(SD)
Age of first mja
use(yrs)
Mja usehistory
Ever usedmja
near-daily(%)
Age startednear-daily
mja use2
(mean)
(SD)
10.0•
Group3
Midheavy
100•.b
Group
4
Latelight
Overall
F tesff
0.8•
1496.04,
268.95,
131.1a.*
27.8•
41.8•,•
21.4•
(36.2)
(19.2)
(25.9)
(18.1)
15.0•
15.1b
16.1•,b
19.5•.b
(2.2)
(1.4)
(2.1)
(2.1)
44.0•
67.2a
21.3a
69.13*
18.6•,•
(3.0)
20.3•
(3.7)
20.9•
(2.4)
18.03'
-
100•
(mean)
(SD)
Group2
Earlylight
17.5•,•
(2.4)
Used
lessthan
10times
Nonusers
0.6
257.03*
20.0
(3.2)
Interval from onset to
near-daily2
Numberof near-daily
usespells2
Mean lengthnear-daily
spells
(mean
yrs)
2.5•,a
3.6c,a
4.4•,b
1.6a,c
10.81'
(SD)
(2.8)
(3.3)
(3.5)
(2.1)
-
(mean)
(SD)
2.F,a
(2.0)
1.5•.a
(1.1)
2.0•,•
(1.9)
1.3•,•
(0.8)
5.09*
-
(meanmths)
(SD)
93.1•,•,c
(54.4)
21.6•
(16.3)
26.1•
(18.5)
18.5•
(16.8)
102.4'
-
Highestfrequencyever
used
(mean)
7.7•
5.8a
6.8•
4.7•
(0.4)
(1.9)
(1.6)
(2.0)
91.8•
50.8•
70.9a
23.9•
56.15*
25.6
7.3
(mean yrs)
(SD)
16.9•
18.0•
19.3•
21.8•
40.16*
22.3
23.3
(3.4)
(3.3)
(3.9)
(3.7)
(3.9)
(5.2)
7.4a
2.2a
4.5•
0.6•
(SD)
(5.6)
(3.4)
(4.8)
(1.6)
(mean)
(SD)
6.5b
4.5•
5.5•
(6.7)
(5.9)
(5.8)
(5.3)
63.9•,a
45.8•,a
72.4a.c
47.5•,•
1.5•,b
(2.2)
0.3•
(1.1)
0.9a,•
(1.3)
(SD)
Use of other substances
% ever used OID 3
10+ times
Age of first OID
use
Yearsany OID used
monthly
62.57*
-
83.66*
-
0.1
(0.9)
0.005
(0.5)
Years smoked 1+
packa day
3.5•,•
6.55*
2.8
1.3
(4.7)
(3.7)
10.93,
25.9
18.0
0.3•
(0.9)
21.22'
-
0.0
(0.0)
1.1•
24.03*
-
% ever drank
alcoholneardaily
Problems
No. mja-related
problems
(mean)
(SD)
No. anydrug-related
problems(except
cigarettes)
(1980-90)
(mean)
(SD)
3.7•
1.5a
2.3•
(3.1)
(2.4)
(2.5)
13.1
•,a
4.0•,a
9.7•,•
(1.9)
0.3
0.2
-
(0.8)
(0.8)
6.28*
0.6
0.3
% dependent
on alcohol
(1990)
2.2a'•
% felt addictedto drugs
(by1990)
55.7
•,b
24.4
•
41.0•
20.1
a
15.34*
14.2
5.9
23.0•,•
9.6•
9.0o
3.4•
8.93*
0.6
0.7
34.8
c
36.8
b
28.9
•
5.45*
32.1
% treatedfor drug
problem
(by1990)
% had psychiatric
problem
(by1990)
TotalN
55.7•,•,•
61
250
134
263
-
162
17.0
289
•F testof between-group
effects:
*p< .01;*p< .001.Student-Newman-Kuels
(SNK)testwasapplied
totestgroup
differences:
Identical
superscript
letters
indicate
significant
differences
inmeans
between
groups
(p<.05).Those
whoused
lessthan10timesandnonusers
arenotincluded
intheF test.
2Restricted
to near-dailyusers.
3OID = Illicit drugsotherthanmarijuana.
372
JOURNAL
OF STUDIES
ON ALCOHOL
very similarto the late onset-lightusegroup.However, the
formerwere lesslikely to haveconsumedalcoholon a near
daily basisand were alsolesslikely to have felt addictedto
any drug or been treatedfor a drug problemthan the latter
group.
Those who had never used marijuanawere much less
likely thanall the marijuanausegroupsto haveusedother
classesof drugs,whether tobacco,alcohol or illicit drugs
other than marijuana.A striking differencebetweenthat
groupandall themarijuanausegroups,includingthosewho
usedfewerthan 10 times,wasthemuchlowerproportionreportingeverhavingexperienced
a psychiatricproblem.
Demographicand lifestylecharacteristics
The four marijuanausertypesalsodifferedsignificantly
from each other with respectto sociodemographic
and
lifestylecharacteristics
(Table 2). Male usersstartedusing
marijuanaearlierand were more likely than womento becomeheavilyinvolved.Whereasthe two light-usinggroups
were almostevenlydividedbetweenmen and women,men
constitutedthe majority of the heavy-usinggroups,especially of theearly-heavygroup(75.4% earlyand62.7% mid).
Indeed,threetimesas many men as womenwere classified
in the early-heavygroup(12.6% vs 4.4%) andalmosttwice
asmanyin themid-heavygroup(23.1% vs 14.5%).The two
groupsof light userswereevenlydividedbetweenthe genders.
The sharpest
differenceswereobservedbetweentheearlyheavyandlate-lightgroups.On the sociodemographic
characteristicsthe two heavy- and the two light-usinggroups
were generallysimilarto eachother,irrespectiveof ageof
onsetinto marijuana.As comparedwith light users,heavy
usershad lower education,lower overallhigh schoolGPA
andlowerchurchattendance;
theywerealsomoredelinquent
in adolescence
and adulthood,more likely to take risksin
adulthood,morelikely to be in networksof marijuana-using
friendsin adolescence
andearlyadulthood,andchanged
jobs
more frequently.Among heavyusers,thosewho startedusing marijuanalater (mid-heavy)had higheroverall GPA in
high school,lower delinquentparticipationin adolescence
and adulthood,and fewer marijuana-using
friendsthanthe
early-heavygroup.The significantdifferencesin patternsof
the use of legal drugs and OIDs, drug-relatedproblems,
lifestyleandsociodemographic
factorsamongthefourtypes
of marijuanauserssupportthetypologyderivedon thebasis
of the clusteranalysis.
Again, the marijuanauserswho had ever usedmarijuana
fewerthan 10 timeswerevery similarto thelate onset-light
usegroup.Therewerethreedifferences:Comparedwith the
late-lightgroup,thosewho usedmarijuanafewer than 10
timesattendedchurchmorefrequentlyasyoungadults;they
were muchlesslikely to be embeddedin socialnetworksof
/ MAY
2000
marijuana-using
friendsor spouses;
andtheywerelesslikely
to haveparentswho wereproblemdrinkers.Thesesamefactors,with evenlowervaluespluslowerdelinquentparticipation in adolescence
andearly adulthood,differentiatedthose
who had neverusedmarijuanafrom the marijuanausers.
Uniquepredictorsof specifictypesof marijuanausers
To identifyuniquepredictorsof clustermembership,
we
estimatedfour sets of multivariatelogistic regressionsin
which we comparedtwo clustersat a time (Table 3). We
comparedtheheavyuserswho startedusingearlywith those
who startedlater (Groups 1 and 3, early vs mid). We comparedthe two early onsetgroupswith eachother(Groups1
and2, heavyvs light).To identifythedistinguishing
characteristicsof heavyversuslight usersirrespectiveof ageof onset,we comparedcombinedGroups1 and3 versuscombined
Groups2 and 4. Finally, to examinetypologiesparallelto
those described for alcoholism and criminal behavior, we
comparedthe two combinedearly-onsetgroups(Groups1
and2) with the two late-onsetgroups(Groups3 and4). The
analyseswererestrictedto userswho hadparticipatedin the
initial schoolsurvey(n = 589). The covariates(seeTable 3)
included a subset of variables in Tables 1 and 2.
We were mostsuccessful
in identifyingthe uniquecorrelatesof early versuslate onsetandof heavyversuslight involvement.Few uniquevariablesdifferentiatedheavyusers
who startedusingearly from thosewho startedlater. Two
factorsdid so (first column):perceivingmanyfriendsto be
usingmarijuanain youngadulthood,andeverhavinghad a
psychiatricdisorder.Earlyonset-heavy
userswereoverthree
timesmorelikely to everhavehada psychiatric
disorderthan
heavyuserswho starteda yearlater.
The strongestfactors differentiatingearly onset-heavy
usersfrom early onset-lightusers(secondcolumn) were:
having friends using marijuanain young adulthood,ever
havinghada psychiatricdisorder,andminordelinquencyin
adolescence.Additional predictors,all marginally significant(p <. 10), were beingmale, usingmarijuanato enhance
positivefeelingsandgrowingup in a nonintactfamily.
The factorsthat differentiatedthe two aggregatedheavyusing groupsfrom the two aggregatedlight-usinggroups
(thirdcolumn)were,with oneimportantexception,similarto
thosethatdifferentiatedheavyversuslight useamongthose
with early onset:being male, using marijuanato enhance
positivefeelings,havingmarijuana-using
friendsin young
adulthood,andbeingdelinquentin adolescence.
Havingever
had a psychiatricdisorder,however,was not significant.
The mostsignificantdifferencesbetweenearlyversuslate
onset(fourthcolumn),irrespectiveof degreeof involvement,
were: age of onsetinto alcohol and cigarettes,numberof
friendsusing marijuanain adolescence,
and GPA in high
school.Usingmarijuanato overcomenegativefeelingswas
associated
with reducedoddsof earlyinitiation.
KANDEL
TABLE 2.
AND CHEN
373
Sociodemographic
and
lifestyle
characteristics
offour
clusters
of•narijuana
(mja)
users
(N= 708)
Mja userclusters
Group1
Earlyheavy
Sociodemographic
and
lifestylecharacteristics
Group2
Earlylight
Group3
Midheavy
Group4
Latelight
Overall
F tesff
Used
lessthan
10times
Nonusers
Sociodemographic
Males(%)
75.4b,d
46.4c,d
62.7'•c
44.9a.b
9.63*
40.1
38.4
(mean)
(SD)
13.6a
(2.3)
14.3
(2.4)
14.1
(2.3)
14.5a
(2.5)
3.12*
-
14.3
(2.5)
14.0
(2.3)
(mean)
(SD)
74.8a,b
(8.6)
77.5a
(8.2)
77.9•
(8.4)
79.5a
(7.4)
6.76*
-
79.5
(7.7)
79.3
(8.3)
59.0
65.2
53.7 a
69.6 •
3.54*
73.5
68.2
6.9•
(3.8)
5.8•
(3.5)
6.8•
(3.7)
5.0•,•
(2.9)
4.9
(3.0)
4.3
(2.6)
15.2•
11.2
14.1a
27.7
37.7
11.59'
-
1.7
(1.5)
0.8
(1.2)
Highestyear of
schooling
Socialrole participation
Overall high
school GPA
Currentlymarried
(1990) (%)
Total no. of employer
spellsby 1990
(mean)
(SD)
10.71'
-
Commitment/deviance
Attended church 2-3/month
1.6a,b
or more(1980) (%)
Minor delinquency
(1971)2
Lastyear delinquency
(1980)
Last yeardelinquency
(1990)
2.94'
(mean)
(SD)
2.5b
(1.6)
1.61a
(1.4)
2.0•
(1.7)
1.3a,b
(1.3)
(mean)
(SD)
1.5•,o,c
(2.1)
0.7c
(1.4)
1.0•
(1.5)
0.6•
(1.3)
6.72'
-
0.4
(0.8)
0.2
(0.6)
(mean)
(SD)
0.8•,b
(1.6)
0.2•
(0.7)
0.5•,•
(0.9)
0.2a
(0.7)
10.08•
-
0.1
(0.5)
0.1
(0.4)
(mean)
(SD)
2.8•,a
(0.8)
2.5c,a
(0.7)
2.7•.c
(0.8)
2.4•,b
(0.8)
5.93'
-
2.2
(0.8)
2.1
(0.8)
(mean)
17.8
18.3
17.4
18.1
0.86
18.4
16.9
(SD)
(4.9)
(5.0)
(4.4)
(4.7)
-
(5.4)
(4.8)
54.1
51.2
49.3
48.3
0.30
42.6
43.9
11.5
10.0
12.7
11.8
0.25
8.6
8.3
11.5
7.2
5.2
5.9
1.08
3.1
2.8
41.0
32.4
33.6
30.8
0.51
30.9
22.1
40.0•
39.9•
29.3•
12.2b
16.56'
9.7
3.4
90.2a,
•
40.8•
60.9•,•
33.6•
29.36*
11.2
5.9
51.2
41.8
48.2
36.8
1.46
16.1
7.7
61
250
134
263
162
289
Personality/psychological
symptoms
Risk-taking
(1980)
Depressionindex
atage15-162
Family history
Mother/father
smoked when R s
wasin high school
Heavysmoking
(%)
Mother/father
drank when R s
wasin highschool
Heavydrinking(%)
Problemdrinking/
alcoholic
(%)
Parent/sib. ever treated for
emotional
disorder
(%)
Drug involvementin socialcontext
Most friends used
mja2 (1971)(%)
Most friends use
mja(1980)(%)
Spouse/partner
used
mja4 (1980)(%)
(TotalN)
-
•F testof between-group
effects:
* p < .05;*p< .001.Student-Newman-Kuels
(SNK)testwasapplied
totestgroupdifferences:
Identical
superscript
lettersindicatesignificant
differences
betweengroupsatp < .05.
2Basedon 589 casessurveyedin 1971.
3R = respondent.
4Basedon 441 caseswhoevermarriedor lived with a partnerby 1980.
374
JOURNAL
OF STUDIES
ON ALCOHOL
Discussion
Four clustersof marijuanauserswereidentifiedby longitudinalcoursein a cohortfollowed from adolescenceto age
34-35. Thesetypeswereidentifiedby includingthreeaspects
of marijuana-usehistoryin a clusteranalysis:age of first
marijuanause,durationof nearlydailyuseandpersistence
of
/ MAY
2000
useinto adulthood.Adolescentswith early onsetat age 1516 couldbe distinguished
by their subsequent
extensiveness
of involvementinto early onset-heavyuse and early onsetlight usegroups.The two othergroupswereonethatstarted
a year later and becameheavily involved, the mid onsetheavy usegroup,and one that started4 yearslater than the
early-onsetgroupsand neverbecameheavilyinvolved,the
TABLE3. Factorsassociated
with beingin specificmarijuana(mja) userclusters
a (basedon multivariatelogisticregressions)
(1 vs 3)
Early-heavy
(vs mid-heavy)
Covariates
Sociodemographic
variables
Gender(1 = male)
Race (1 = black)
Highestyearsof schooling
by eitherparent
AOR b
(1 vs 2)
Early-heavy
(vs early-light)
(l&3 vs 2&4)
Heavyuse
(vs light use)
AOR
AOR
(l&2 vs 3&4)
Earlyonset
(vs late onset)
AOR
1.58
0.19
2.53
0.42
2.36*
1.38
0.97
0.71
1.03
0.96
1.02
1.04
Drug usehistoryby age24-25
Age of onsetinto alcohol/cigarettes
Usedmja to overcomenegativefeelings
Usedmja to enhancepositivefeelings
Usedmja for socialreasons
0.98
1.73
3.09
0.66
1.02
1.98
7.78õ
0.68
0.96
1.32
2.57*
0.77
0.94*
0.62*
1.38
0.91
Familyhistoryof druguse/disorder
Parentalsmokingstatusat age 15-16c
Parentaldrinkingstatusat age 15-16
1.02
1.70
1.13
1.37
1.03
1.19
0.93
1.06
0.67
1.04
1.10
0.86
Parent-childrelationat age 15-16
Family intactness
in adolescence
(1 = intact)
Closeness
to parent
1.03
0.99
0.44
1.01
0.91
0.99
1.21
0.99
Peerdruginvolvement
No. of friendsusingmja at age 15-16
No. of friendsusingmja at age24-25
1.08
2.39*
0.84
1.04
1.56'
2.77*
1.68'
1.15
0.90
0.99
0.97
0.97
1.37'
0.99
0.88
1.22'
1.00
0.93
0.99
0.97*
1.03
0.98
0.98
0.996
3.21'
2.96*
1.43
1.22
108.4'
80.7*
Parent/sib. ever treated for
mental disorder
Commitment/deviance
at age 15-16
Church attendance
Minor delinquency
indexat age15-16
Last yearGPA in high school
Psychologicalsymptoms
Depression
indexat age15-16
Ever had psychiatricproblem
(by age34-35)
ModelX2d(20 df)
Total N
33.7*
152
75.8*
256
õp< .10; *p < .05; *p < .01; *p < .001.
aNfor early-heavy
= 45; early-light= 211; mid-heavy= 107;late-light= 226.
bAOR= adjustedoddsratios.
cMissing
categorycodedasa separate
dummyvariablebutnotshown.
•tDifference
betweengoodness-of-fit
chi squares
for thecurrentandbaselinemodels.
589
589
KANDEL
late onset-lightuse group.The early-heavygroup representeda minority of individualsand containedmore than
threetimesasmanymenaswomen.The majorityof usersremainedlight usersirrespectiveof their ageof initiationinto
marijuanause.Persistent
andheavyusersdid notnecessarily
initiatemarijuanauseat theearliestages.
The fourgroupsdifferedfromeachotherin theirdegreeof
involvementin marijuanaandotherdrugs.The early-heavy
groupconsisted
of the mostinvolvedandmostproblematic
marijuanausers.Surprisingly,this groupwas lesslikely to
persistin usingmarijuanainto adulthoodthanthe mid onsetheavyusegroupwho startedusingmarijuanausea yearlater.
The fact thatthe first groupwasmuchmorelikely thanany
otherto havebeenin treatmentmaypartiallyexplainthisresult.The lateonset-lightusetypewasoneof thetwo largest
clustersin thecohort,accounting
for overa thirdof the sample, and was characterized
by light marijuanainvolvement
and nonpersistent
use. In comparisonwith the other types
theseuserswere alsothe leastinvolvedin drugsotherthan
marijuana,they reportedthe fewestdrug-relatedproblems
and were more likely to haveconventionalparticipationin
the socialrolesof adulthood(e.g.,marriageandlaborforce
participation).
A moststrikingdifferencebetweenmarijuanauserswho
neverusedmarijuanamore than 10 timesand otherusersis
their greaterreligiosity in early adulthoodand lesserinvolvementin marijuana-using
networks.Thosewho never
usedmarijuanawere evenmorereligious,lessinvolvedin
marijuanausinggroupsand muchlesslikely to reportany
psychiatricproblemsthan thosewho ever usedmarijuana
AND CHEN
375
groupof marijuanausersthat did not becomeheavilyinvolved in using marijuana.Whether mental disordersare
consequences
of or riskfactorsfor earlyandheavyinvolvementin marijuanacannotbe determined
from ourdata.The
majorityof studieshave found that psychopathology
(e.g.,
majordepression,
anxietyand,especially,
conductdisorders)
precedes
substance
usedisorders
(Burkeet al., 1994;Christie
et al., 1988;Deykin et al., 1987;Giaconiaet al., 1994;Nelsonet al., 1996),althoughmooddisorders
havealsobeenreportedto follow the onsetof alcoholor drug abuseand
dependence
(Rohdeet al., 1991).Psychiatricsymptoms
and
disordersother than depression
probablyconstitutea risk
factorfor heavyinvolvementin marijuana.
Age of onsetinto alcoholand cigarettesdifferentiated
early from late onsetmarijuanause, but not extentof use
withineachtype.Thissuggests
that,whileuseof a legaldrag
facilitatesearlyprogression
to theuseof anillicit drug,other
factorscomeinto play to accountfor degreeof involvement
in thesedrags.The differentialimportanceof peers'marijuanause,whichdifferentiates
earlyfrom latemarijuanause
onsetin adolescence
andheavyfrom lightusein adulthood,
suggests
that differentetiologicalprocesses
comeinto play
at differentphasesof thelifecycle.Drug-usingpeernetworks
in adolescence
may predictthe timingof marijuanauseonset,while in adulthood
thenetworksmaybe associated
with
sustaining
extensiveness
of use.We havearguedelsewhere
thatextentof peerinfluenceconfounds
selectionandsocialization effects(Kandel, 1996). The selectionof drug-using
peersmightbe strongerin adulthood
thanin adolescence.
Existing
taxonomies
developed
for
alcoholismandcrimifewer than 10 times.
nality
emphasize
the
distinction
between
early andlate onSeveral findings deservecomments.Different factors
set.
The
ages
associated
with
early
onset
vary for different
wereimportantfor differentclusters.Risk factorsimportant
behaviors:
starting
to
drink
in
adolescence
for alcoholism,
for differentiatingat leastoneclusterfrom all othersincluded
manifestingconductproblemsin childhoodfor criminality.
beingmale,early ageof onsetinto alcoholor cigarettes,usMidadolescence
represents
earlyonsetfor marijuana,assuging marijuanato enhancepositivefeelings,membershipin a
gestedby the presentresults.As for alcoholism,delinquent
marijuanausinggroup,delinquentparticipationin adolesparticipationin adolescence
is associated
with early onset
cence,andeverhavinghad a mentaldisorder.Factorspreand
heavy
subsequent
involvement
in
marijuana.
While adodictingdelayeduseincludedsuperioracademic
performance
lescent
delinquency
did
not
differentiate
heavy
marijuana
in high schooland usingmarijuanato overcomenegative
users
according
to
age
of
marijuana
onset,
it
discriminated
feelings.Youthswho onsetlate may be especiallylikely to
do soto self-medicate.
Qualityof parentingin adolescence betweenall heavyandall light usersandbetweenearly onset userswho becameheavy usersand thosewho did not.
failedto havelong-termprotectiveeffectsonthechild'spatThus,
two contrastinggroupsof marijuanausers,the comtern of marijuanainvolvement.The protectiveness
of relibined
heavy
users(earlyonsetandmid onsetgroups)andthe
giosity observedat the univadatelevel disappeared
with
control for other factors.
lateonset-lightusegroup,correspond
to thedualtaxonomies
developed
for
early
and
late
onset
alcoholics
and antisocial
In conclusion,membershipin socialcontextsof drug-usindividuals.The characteristics
of the early and late onset
ing peers,usingdrugsto achieveeuphoria,the presenceof
psychiatricproblems,and personaldeviancefosterthe degroupsfor marijuanauseare very similarto thosefor alcovelopmentof early andheavymarijuanainvolvement.Havholismandantisocialbehavior.However,thepresentstudy,
ing experienceda psychiatricdisorder is an important
conducted
in a generalpopulationsample,highlightstheimdistinguishing
characteristic
of heavyuserswho startedthe
portanceof recognizingthat there can be heterogeneity
earliest.By correlation,the absenceof psychopathology withintheearlyandlateonsetgroupsthemselves
(for a simseemsto be a particularlydistinguishing
characteristic
of the
ilar pointregardingalcoholismseeMezzichet al., 1993).We
376
JOURNAL
OF STUDIES
ON ALCOHOL
/ MAY
2000
have identifiedan early onsetgroupthat doesnot become
heavily involvedin usingmarijuana.Early onsetby itself
will notleadto problematicuseor rapidprogress
intotheuse
of otherdrugs.The desireto enhanceone's moodby using
marijuanaanddysfunctional
behaviors(e.g.,delinquencyor
psychopathology)
are associatedwith the developmentof
problematicdruguseanddependence.
Motivationsunderlyingmarijuanauseplaya complexrole
andrequiredistinguishing
amongdifferenttypesof reasons
for use.While the enhancement
of positivemoodis a strong
motivatingfactor underlyingheavy involvementbut not
early onset,the reverseappearsto hold for the reductionof
negativemood. The use of marijuanato reducenegative
moodand deal with problemsreducesthe risk of heavyinvolvementbut doesnotappearto be a risk factorfor earlyonset. The importanceof marijuanause to enhancepositive
feelingsas a risk factorfor heavyinvolvementparallelsthe
motivationsdistinguishing
the two typesof alcoholicsidentified by Cloningeret al. (1986). The moreseverealcoholics
(Type II), characterized
by early onset,were morelikely to
drink to induceeuphoriathanType I, characterized
by later
onset.Type I alcoholicswere more likely than Type II to
drink to relieve anxiety.
Severallimitationsof the studymustbe noted.One limitationpertainsto theidentificationof theclustersandreflects
the stateof methodology
in the field. The methodsavailable
to identifyclustersdo notby themselves
provideuniqueand
unequivocalsolutions.Furthermore,theresultingtypologies
are functionsof the variablesselectedfor clustering.The developmentaltaxonomyof marijuanause proposedhere is
suggestiverather than definitive. The analysesneed to be
replicatedon other data setsto provideconfirmatoryevidencefor the typology.Anotherlimitationderivesfrom the
torsin the early lives of differenttypesof users.Sucha typologywill help researchers
and practitionersidentify potential types of marijuana users in the early stage of
psychosocial
development
anddevelopmoreeffectiveearly
preventionand interventionprogramstargetedtoward specific typesof substance
usersor abusers.It is clearthat simple two-typeclassifications
of deviantbehaviorfail to take
intoaccounttheheterogeneity
of earlyandlateonsetgroups.
We considerthedocumentation
of thisheterogeneity
to be a
majorcontributionof thisstudy.
As notedearlier,threedifferentstrategies
canbe pursued
to understandprogression
into variousforms of drug use.
These include investigationsof pathwaysof progression
fromoneclassof drugto another,degreeof involvementinto
a particularclassof drug,andtaxonomies
basedon developmentalcourse.The threestrategiesare complementary
and
not mutuallyexclusive.For instance,individualsat a particular stageof druguse are heavierusersof precedingdrugs
thanthoseremainingat the precedingstage(s)(Kandel and
Yamaguchi,1999; Kandelet al., 1992). A completedevelopmentalaccounting
of drugbehavioroverthelife courserequiresanintegrationof all threestrategies
anda specification
of the interrelationships
and pointsof intersectionamong
them.Sucha perspective
wouldbringuscloserto thepersoncenteredapproachadvocated
by Magnusson(Bergmanet al.,
1998; Magnusson,1998) for understanding
behavioraldevelopment,includingsubstance
useandabuse.
nature of the data. The measures of several constructs were
ACHENBACH,
T.M. EmpiricallyBasedTaxonomy:How to Use Syndromes
and Profile Types Derived from the CBCL/4-18, TRF, and YSR,
Burlington,VT: Departmentof Psychiatry,Universityof Vermont,
psychometrically
weak; in particular,measuresof psychiattic problems.In this prospectivestudy,the intervalsbetweenwavesof datacollectionwere long,rangingfrom 4 to
9 years.Much of the informationwasretrospective,
affecting to someunknowndegreereportsof the occurrenceand
timing of events.Given theselimitations,the regularityof
differencesobservedamongthegroupsonvariablesdescribing variousaspectsof drughistoriesis all the moreremarkable. It would be important to implement longitudinal
follow-upsin whichtheintervalsbetweenwavesof datacollection would be shorter,ideally no longerthan 1 year. Finally, sincethe studyis basedon a singlecohort,theresults
maybe affectedby historicalor cohort-specific
factors,such
asdifferencesin ageof onsetinto marijuanause.
To thebestof ourknowledge,thisstudyrepresents
thefirst
attemptto identifysubgroups
of marijuanauserson thebasis
of longitudinaldatafrom adolescence
to adulthood,spanning
the yearsof highestrisk for involvementin marijuana.We
justifiedthe typologyby identifyingrisk andprotectivefac-
Acknowledgment
The assistance
of ChristineSchaffranis gratefullyacknowledged.
References
1993.
ALDENDERFER,
M.S. ANDBLASHFIELD,
R.K. ClusterAnalysis,SageUniversityPaperSeries:QuantitativeApplicationsin the SocialSciences,
No. 07-044, ThousandOaks,CA: Sage,1984.
BABOR,T.F. The classification
of alcoholics:Typologytheoriesfrom the
19thcenturyto thepresent.AlcoholHlth Res.World 20: 6-14, 1996.
BABOR, T.F., HESSELBROCK,
V.M., MEYER, R.E. A•qO SHOEMAKER,W.
(Eds.)Typesof Alcoholics:Evidencefrom Clinical,Experimental,and
GeneticResearch,
Annalsof the New York Academyof Sciences,Vol.
708, New York: New York Academyof Sciences,1994.
BABOR,T.F., HOEMANN,M., DEL BOCA,F.K., HESSELBROCK,
V., MEYER,
R.E., DOLINSKY,
Z.S. ANDROUNSAVILLE,
B. Typesof alcoholics:I. Evidencefor an empiricallyderivedtypologybasedon indicatorsof vulnerabilityandseverity.Arch.Gen.Psychiat.49: 599-608,1992.
BERGMAN,
L.R. A pattern-oriented
approachto studyingindividualdevelopment:Snapshots
and processes.
In: Cairns,R.B., Bergman,L.R. and
Kagan,J. (Eds.)MethodsandModelsfor Studyingthe Individual:Essaysin Honor of Marian Radke-Yarrow,ThousandOaks, CA: Sage,
1998,pp. 83-122.
BERGMAN,
L.R. ANDMAGNUSSON,
D. A personapproach
to thestudyof the
developmentof adjustmentproblems:An empiricalexampleand some
KANDEL
AND
CHEN
377
research
strategy
considerations.
In: MAGNUSSON,
D. AND•)HMAN,A.
JESSOR,
R., VAN DEN BOS, J., VANDERRYN,J., COSTA,F.M. AND TURBIN,
(Eds.)Psychopathology:
An Interactional
Perspective,
SanDiego,CA:
AcademicPress,1987,pp. 383-401.
M. Protectivefactorsin adolescent
problembehavior:Moderatoreffects
anddevelopmental
change.Devel. Psychol.31: 923-933, 1995.
KANDEL,D.B. Marijuanausersin youngadulthood.Arch. Gen. Psychiat.
BROOK,J.S., BROOK,D.W., GORDON,A.S., WHITEMAN,m. ANDCOHEN,P.
Thepsychosocial
etiologyof adolescent
druguse:A familyinteractional
approach.
Genet.SocialGert.Psychol.Monogr.116:111-267, 1990.
BURKE,
J.D.,BURKE,
K.C. ANDRAE,D.S. Increased
ratesof drugabuseand
dependence
after onsetof moodor anxietydisordersin adolescence.
Hosp.CornTun.Psychiat.45: 451-455, 1994.
CALINSKI,T. ANDHARABASZ,
J. A dendritemethodfor clusteranalysis.
Communicat. Stat. 3: 1-27, 1974.
CHEN,K. ANDKANDEL,
D.B. Thenaturalhistoryof drugusefromadolescenceto themid-thirtiesin a generalpopulation
sample.Amer.J. Publ.
Hlth 85: 41-47, 1995.
CHRISTIE,
K.A., BURKE,J.D., REGIER,D.A., RAE,D.S. ET AL. Epidemiologicevidence
forearlyonsetof mentaldisorders
andhigherriskof drug
abusein youngadults.Amer. J. Psychiat.145: 971-975, 1988.
CLAYTON,R.R. Transitionsin drug use:Risk and protectivefactors.In:
GLANTZ,M. AND PICKENS,
R. (Eds.) Vulnerabilityto Drug Abuse,
Washington,
DC: AmericanPsychological
Association,
1992,pp. 1551.
CLONINGER,C.R., BOHMAN,M. ANDSlGVARDSSON,
S. Inheritanceof alco-
hol abuse:Cross-fostering
analysisof adoptedmen.Arch. Gert.Psychiat. 38: 861-868, 1981.
CLONINGER,C.R., SIGVARDSSON,
S., REICH, T. AND BOHMAN,M. Inheritanceof risk to developalcoholism.In: BRAUDE,
M.C. ANDCHAO,H.M.
(Eds.)GeneticandBiologicalMarkersin DrugAbuseandAlcoholism.
NIDA Research
MonographNo. 66, DHHS PublicationNo. (ADM) 861444),Washington:
Government
PrintingOffice, 1986,pp. 86-96.
COSTA,
F.M., JESSOR,
R. ANDTURBIN,M.S. Transitionintoadolescent
problem drinking:The role of psychosocial
risk andprotectivefactors.J.
Stud. Alcohol 60: 480-490, 1999.
DEWIT, D.J., SILVERMAN,G., GOODSTADT,
M. ANDSTODUTO,G. The constructionof risk andprotectivefactorindicesfor adolescentalcoholand
otherdruguse.J. Drug Issues25: 837-863, 1995.
DEYKIN,E.Y., LEVY,J.C. ANDWELLS,V. Adolescentdepression,
alcohol
anddrugabuse.Amer. J. Publ. Hlth 77: 178-182, 1987.
DILALLA,L.F. ANDGOTTESMAN,
I.I. Heterogeneityof causesfor delinquencyandcriminality:Lifespanperspectives.
Devel.Psychopathol.
1:
339-349, 1989.
DUDA,R.O. ANDHART,P.E. PatternClassification
and SceneAnalysis,
New York: JohnWiley & Sons,1973.
ENSMINGER,M.E., BROWN,C.H. AND KELLAM, S.G. Sex differences in an-
tecedentsof substance
useamongadolescents.
J. SocialIssues38: 2542, 1982.
FELIX-ORTIZ,
M. ANDNEWCOMB,
M.D. Riskandprotective
factorsfordrug
useamongLatinoandwhiteadolescents.
Hisp.J. Behav.Sci. 14: 291309, 1992.
FLEMING,
R., LEVENTHAL,
H., GLYNN,K. ANDERSHLER,
J. Theroleof cigarettesin the initiationandprogression
of early substance
use.Addict.
Behav. 14: 261-272, 1989.
GIACONIA,R.M., REINHERZ,H.Z., SIVERMAN,
A.B., PAKIZ,B., FROST,A.K.
ANDCOHEN,E. Agesof onsetof psychiatricdisordersin a community
populationof olderadolescents.
J. Amer. Acad.Child Adolesc.Psychiat. 33: 706-717, 1994.
HAWKINS,
J.D.,CATALANO,
R.F. ANDMILLER,J.Y. Riskandprotectivefactorsfor alcoholandotherdrugproblems
in adolescence
andearlyadulthood:Implicationsfor substance
abuseprevention.Psychol.Bull. 112:
64-105, 1992.
HINSHAW,S.P.,LAHEY,B.B. ANDHART,E.L. Issuesof taxonomyandcomorbidityin thedevelopment
of conduct
disorder.
Devel.Psychopathol.
5: 31-49, 1993.
HUBA,G.J.Cross-validation
of a drugusecausalmodelin theColumbialongitudinal study data. Bull. Soc. Psychol.Addict. Behav. 2: 128-132,
1983.
41: 200-209, 1984
KANDEL,D.B. The parentalandpeer contextsof adolescent
deviance:An
algebraof interpersonal
influences.
J. Drug Issues26:289-315, 1996.
KANDEL, D., ellEN, K., WARNER,L.A., KESSLER,R.C. AND GRANT, B.
Prevalenceanddemographic
correlatesof symptomsof lastyeardependenceon alcohol,nicotine,marijuanaandcocainein the U.S. population. Drug AlcoholDepend.44:11-29, 1997.
KANDEL,D.B. ANDDAVIES,M. Epidemiology
of depressive
moodin adolescents:
An empiricalstudy.Arch.Gen.Psychiat.39: 1205-1212,1982.
KANDEL, D.B., KESSLER,e.c. AND MARGULIES,R.Z. Antecedentsof ado-
lescentinitiationinto stagesof druguse:A developmental
analysis.J.
Youth Adolesc. 7: 13-40, 1978.
KANDEL,D., SINGLE,E. ANDKESSLER
R.C. The epidemiology
of druguse
amongNew York statehigh schoolstudents:
Distribution,lrends,and
changein ratesof use.Amer. J. Publ. Hlth 66: 43-53, 1976.
KANDEL,D.B. ANDYAMAGUCHI,K. From beer to crack:Developmental
patternsof druginvolvement.Amer.J. Publ.Hlth 83: 851-855, 1993.
KANDEL,D.B. ANDYAMAGUCHI,
K. Developmental
stagesof involvement
in substance
use.In: OTT, P.J. ANDTARTER,R.E. (Eds.) Sourcebookon
Substance
Abuse:Etiology,Epidemiology,
Assessment,
andTreatment,
NeedhamHeights,MA: Allyn andBacon,1999,pp. 50-74.
KANDEL,D.B., YAMAGUCHI,
K. ANDellEN, K. Stagesof progression
in
druginvolvementfrom adolescence
to adulthood:Furtherevidencefor
thegatewaytheory.J. Stud.Alcohol53: 447-457, 1992.
MAGNUSSON,
D. The logicandimplicationsof a person-oriented
approach.
In: CAIRNS,R.B., BERGMAN,
L.R. ANDKAGAN,J. (Eds.) Methodsand
Modelsfor Studyingthe Individual:Essaysin Honorof MarianRadkeYarrow,ThousandOaks,CA: Sage,1998,pp. 33-64.
MEZZICH, A., TARTER,e., KIRISCI, L., CLARK, D., BUCKSTEIN,O. AND
MARTIN,C. Subtypesof early ageonsetalcoholism.AlcsmClin. Exp.
Res. 17: 767-770, 1993.
MILLIGAN,G.W. ANDCOOPER,
M.C. An examinationof procedures
for determiningthe numberof clustersin a dataset.Psychometrika
50:159179, 1985.
MILLS, C.J. AND NOYES,H.L. Patternsand correlatesof initial and subse-
quentdruguseamongadolescents.
J.Cons.Clin. Psychol.52:231-243,
1984.
MOFFITT,T.E. Adolescence-limited
andlife-course-persistent
antisocialbehavior:A developmental
taxonomy.Psychol.Rev. 100: 674-701, 1993.
MOFFITT,T.E., CASPI,A., DICKSON,N., SILVA, P. ANDSTANTON,W. Child-
hood-onsetversusadolescent-onset
antisocialconductproblemsin
males:Naturalhistoryfrom ages3 to 18 years.Devel.Psychopathol.
8:
399-424, 1996.
NELSON,C.B., LITTLE, R.J.A., HEATH,A.C. ANDKESSLER,
R.C. Patternsof
DSM-III-R alcoholdependence
symptomprogression
in a generalpopulationsurvey.Psychol.Med. 26: 449-460, 1996.
QUAY,H.C. Patternsof delinquentbehavior.In: Quay,H.C. (Ed.)Handbook
of JuvenileDelinquency,
New York:JohnWiley & Sons,1987,pp. 118138.
RAVELS,
V.H. ANDKANDEL,D.B. Changesin drugbehaviorfrom the middle to the late twenties:Initiation,persistence
and cessationof use.
Amer. J. Publ. Hlth 77: 607-611, 1987.
ROHDE,P., LEWINSOHN,
P.M. ANDSEELE¾,
J.R. Comorbidityof unipolar
depression:
II. Comorbiditywith othermentaldisordersin adolescents
andadults.J. AbHorre.Psychol.100: 214-222, 1991.
SARLE,W.S. Cubic ClusteringCriterion,SAS TechnicalReport A-108,
Cary, NC: SAS Institute,1983.
SAS INSTITUTE.SAS/STAT User'sGuide,Version6, 4th Edition, Cary,
NC: SAS Institute, 1990.
STACY, A.W., NEWCOMB,M.D. AND BENTLER, P.M. Interactive and
378
JOURNAL
OF STUDIES
higher-ordereffectsof socialinfluenceson druguse.J. Hlth SocialBehay. 33: 226-241, 1992.
WARD,J.H. Hierarchicalgroupingto optimizean objectivefunction.J.
Amer. Stat. Assoc. 58.' 236-244, 1963.
WEBER,M.D., GRAHAM,J.W., HANSEN,W.B., FLAY, B.R. AND JOHNSON,
C.A. Evidencefor twopathsof alcoholuseonsetin adolescents.
Addict.
Behav. 14: 399-408, 1989.
WINDLE, M., BARNES,G.M. AND WELTE, J. Causal models of adolescent
substance
use:An examinationof genderdifferencesusingdistributionfreeestimators.
J. Pers.SocialPsychol.56: 132-142,1989.
ZUCKER,R.A. Pathwaysto alcoholproblemsand alcoholism:A developmentalaccountof theevidencefor multiplealcoholisms
andfor contextual contributionsto risk. In: ZUCKER,R., BOYD, O. AND HOWARD,J.
(Eds.)The Developmentof AlcoholProblems:Exploringthe Biopsy-
ON ALCOHOL
/ MAY
2000
chosocialMatrix of Risk. NIAAA ResearchMonographNo. 26, NIH
PublicationNo. 94-3495,Rockville,MD: Departmentof HealthandHuman Services,1994,pp. 255-289.
ZUCKER,R.A., ELLIS,D.A. AND FITZGERALD,
H.E. Developmentalevidencefor at leasttwo alcoholisms:
I. Biopsychosocial
variationamong
pathwaysinto symptomatic
difficulty.In: BABOR,
T.F., HESSELBROCK,
V.M., MEYER,R.E. ANDSHOEMAKER,
W. (Eds.) Typesof Alcoholics:
Evidencefrom Clinical,Experimental,andGeneticResearch,Annalsof
the New York Academyof Sciences,Vol. 708, New York: New York
Academyof Sciences,1994,pp. 134-146.
ZUCKER,R.A., ELLIS, D.A., FITZGERALD,H.E., BINGHAM,C.R. AND SANFORD,K. Other evidence for at least two alcoholisms: II. Life course
variationin antisociality
andheterogeneity
of alcoholicoutcome.Devel.
Psychopathol.
8: 831-848, 1996.
Expert Review of Respiratory Medicine
ISSN: 1747-6348 (Print) 1747-6356 (Online) Journal homepage: http://www.tandfonline.com/loi/ierx20
Effects of smoking cannabis on lung function
Marcus HS Lee & Robert J Hancox
To cite this article: Marcus HS Lee & Robert J Hancox (2011) Effects of smoking cannabis on lung
function, Expert Review of Respiratory Medicine, 5:4, 537-547, DOI: 10.1586/ers.11.40
To link to this article: http://dx.doi.org/10.1586/ers.11.40
Published online: 09 Jan 2014.
Submit your article to this journal
Article views: 1905
View related articles
Citing articles: 2 View citing articles
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ierx20
Download by: [75.104.66.102]
Date: 18 February 2017, At: 18:15
Review
CME
For reprint orders, please contact [email protected]
Effects of smoking cannabis on
lung function
Expert Rev. Respir. Med. 5(4), 537–547 (2011)
Marcus HS Lee1 and
Robert J Hancox†1,2
Department of Respiratory Medicine,
Waikato Hospital, Pembroke St.,
Hamilton, New Zealand
2
Department of Social and Preventative
Medicine, University of Otago,
PO Box 913, Dunedin, New Zealand
†
Author for correspondence:
[email protected]
1
Although cannabis (or marijuana) is the world’s most widely-used illicit drug, there has been
surprisingly little research into its effects on respiratory health. Part of the problem is the inherent
difficulty of studying the long-term effects of an illegal habit. It has often been assumed that
smoking cannabis will have similar long-term effects to smoking tobacco. Several recent
observational studies suggest that this is not the case and that cannabis has quite different
effects on the lung function. There are consistent findings that smoking cannabis is associated
with large airway inflammation, symptoms of bronchitis, increased airway resistance and lung
hyperinflation. The evidence that smoking cannabis leads to features of chronic obstructive
pulmonary disease, such as airflow obstruction and emphysema is not convincing. However,
there are numerous case reports of bullous emphysema among cannabis smokers. These findings
have not been confirmed in systematic analytical studies and probably represent uncommon
adverse effects in very heavy cannabis smokers. There is now additional controversial evidence
that cannabis is at least an occasional cause of respiratory malignancies, but again the evidence
is inconclusive.
KEYWORDS : bronchitis • bullae • cannabis • cigarettes • emphysema • lung cancer • lung function • marijuana
• respiratory • tobacco
Medscape: Continuing Medical Education Online
This activity has been planned and implemented in accordance with the Essential Areas and
policies of the Accreditation Council for Continuing Medical Education through the joint
sponsorship of Medscape, LLC and Expert Reviews Ltd. Medscape, LLC is accredited by the
ACCME to provide continuing medical education for physicians.
Medscape, LLC designates this Journal-based CME activity for a maximum of 1 AMA PRA
Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent
of their participation in the activity.
All other clinicians completing this activity will be issued a certificate of participation. To participate in this journal CME activity: (1) review the learning objectives and author disclosures;
(2) study the education content; (3) take the post-test with a 70% minimum passing score
and complete the evaluation at www.medscape.org/journals/expertrespiratory; (4) view/
print certificate.
Release date: August 22, 2011; Expiration date: August 22, 2012
Learning objectives
•
•
•
•
www.expert-reviews.com
Distinguish short-term effects of smoking cannabis on lung function
Evaluate the long-term effects of cannabis use on lung function
Describe the effects of cannabis use in promoting emphysema and lung bullae
Analyze the effects of cannabis use on the risk for cancer
10.1586/ERS.11.40
© 2011 Expert Reviews Ltd
ISSN 1747-6348
537
Review
CME
Lee & Hancox
Financial & competing interests disclosure
EDITOR
Elisa Manzotti
Editorial Director, Future Science Group, London, UK.
Disclosure: Elisa Manzotti has disclosed no relevant financial relationships.
CME Author
Charles P Vega, MD, Associate Professor; Residency Director, Department of Family Medicine, University of California, Irvine, CA, USA.
Disclosure: Charles P Vega, MD, has disclosed no relevant financial relationships.
Authors and Credentials
Marcus HS Lee
Department of Respiratory Medicine, Waikato Hospital, Pembroke St, Hamilton, New Zealand.
Disclosure: Marcus HS Lee has disclosed no relevant financial relationships.
Robert J Hancox
Department of Respiratory Medicine, Waikato Hospital, Pembroke St, Hamilton, New Zealand and Department Preventive & Social Medicine,
University of Otago, Dunedin, New Zealand.
Disclosure: Robert J Hancox has disclosed no relevant financial relationships.
Cannabis is the most widely used illicit drug worldwide [1] . It
has been hailed for its analgesic properties [2] and pilloried for
both its psychoactive and addictive nature [3] . It is illegal in
most countries but decriminalized in some. In our country, New
Zealand, cannabis is the third most commonly used drug after
tobacco and alcohol, and the most commonly used illegal drug.
Approximately three-quarters of New Zealanders have tried cannabis by age 25 and 13.7% of New Zealanders have used cannabis
in recent years [4] . It is most often smoked although it can also
be taken orally [5] .
The widespread use of cannabis has raised many concerns
over its long-term effects. Often these have been about the
effects of chronic cannabis use on mental health [6] . However,
the frequent practice of smoking cannabis also raises concerns
over its potential for adverse effects on the respiratory system.
Unfortunately, the illegal status of cannabis makes it difficult to
obtain reliable data on cannabis use and its effects. Self-reports
of cannabis consumption are likely to be inaccurate owing
to social desirability bias and the fear of legal consequences.
Furthermore, as most cannabis users also smoke tobacco, the
effects of cannabis on the respiratory system may be obscured
by the effects of tobacco.
Even if cannabis use is honestly reported by users, quantifying cannabis consumption is inherently difficult because, as an
illegal substance, there is no standardization of supply and significant variations in strengths and amounts of cannabis occur.
In addition, cannabis can be smoked via various methods including bongs and bubble pipes, as well as directly from a cannabis
cigarette. These different methods of smoking cannabis may
have influenced the quantity and composition of smoke inhaled.
Most studies quantify exposure to cannabis smoke as ‘joint-years’
whereby one joint-year is equivalent to one joint smoked daily for
a year. This approach is pragmatic but focuses on the frequency of
cannabis use and ignores differences in the quantity of cannabis
leaf in each joint and in the methods of smoking it. An internet
survey of cannabis users found that two measures of the quantity
of cannabis consumed – the amount of cannabis purchased each
538
month and the usual level of intoxication after using it – predicted
respiratory symptoms independently of and in addition to the
reported frequency of use [7] . Unfortunately, few studies have
gathered such detailed information.
Taken together, the difficulties in obtaining accurate information about cannabis use, the problem of quantifying consumption and the difficulty of separating the effects of cannabis from
those of tobacco have meant that the respiratory side effects
of cannabis have not been well studied. The likelihood that
smoking cannabis harms the respiratory systems has usually
been extrapolated from the well-documented effects of smoking tobacco. At face value, it seems reasonable to assume that
cannabis and tobacco would have similar effects, since, apart
from the main psychoactive ingredients of tetrahydrocannibinol
and nicotine, the substances contain a broadly similar mix of
chemicals [8] . However, recent reports suggest that the effects
of cannabis and tobacco on lung function may be quite different. The paucity of direct evidence on the respiratory effects
of chronic cannabis use therefore leaves a major gap in our
understanding of one of the world’s most commonly inhaled
substances. This article appraises recent evidence that cannabis
is harmful to lungs.
Cannabis & bronchitis
Numerous studies confirm that smoking cannabis can lead to
respiratory symptoms. These studies show that cough, increased
sputum production and wheeze are present in approximately a
fifth to a third of cannabis smokers [9–11] . Cannabis smoking is
also associated with dyspnea, pharyngitis, hoarsening of voice
and exacerbations of asthma [10] . These symptoms appear to
result from the toxic effects of cannabis smoke on the bronchial
mucosa. Bronchoscopic mucosal biopsies from 40 cannabisonly smokers and 31 tobacco-only smokers have demonstrated
that both cannabis and tobacco smoking cause significant
bronchial damage, with an increase in basal cell hyperplasia,
goblet cell hyperplasia, cell disorganization, nuclear variation,
and an increase in nuclear/cytoplasm ratio [12] . This study also
Expert Rev. Respir. Med. 5(4), (2011)
CME
Effects of smoking cannabis on lung function
demonstrated an increase in squamous cell metaplasia in cannabis smokers, raising the possibility that smoking cannabis may
be a risk factor for developing lung cancer.
Another report found that even asymptomatic cannabis smokers with normal physical examinations and spirometric function
have central airway inflammation under direct bronchoscopic
visualization, bronchial mucosal biopsies and bronchial lavage
fluid [13] . Those who smoked both cannabis and tobacco also
had distal airway inflammation. There was a high incidence of
erythema, edema and airway secretions in both exclusive cannabis smokers and exclusive tobacco smokers. These findings
demonstrate that routine physical examination and spirometry
may be insensitive measures of lung injury caused by cannabis.
While the finding that cannabis smoke causes mucosal damage
is not surprising, the most striking result of this study was the
fact that cannabis smokers of an average of a few joints a day
had the same degree of airway damage as tobacco smokers of
20–30 cigarettes a day. Moreover, this damage was present in
young and asymptomatic cannabis smokers.
Effect of cannabis on airflow obstruction
Acute effects
Cannabis has long been recognized as a bronchodilator. Indeed,
newspapers in New Zealand (and presumably many other countries) carried advertisements for imported cannabis cigarettes as
a treatment for asthma in the late 1800s [101] . Hence, inhaling
cannabis appears to predate inhaled adrenergic bronchodilator
therapy by at least half a century [14] . There appears to be no
doubt that smoking cannabis does have acute bronchodilator
effects: in a recent systematic review, 11 out of 12 studies demonstrated a bronchodilator effect of cannabis [10] . However, this
acute bronchodilator effect is modest and does not appear to be
sustained with continued use over 6–8 weeks [15] . It has been
shown to be of slower onset than salbutamol, which has greater
bronchodilator effects at 5 min compared to tetrahydrocannibinol [16] . The potential short-term therapeutic effects also need to
be weighed against the adverse effects of increased bronchitis and
exacerbations of asthma that have been associated with regular
cannabis use. Consequently, cannabis is not currently considered
to have a therapeutic role in acute bronchospasm (although this
is occasionally claimed by cannabis users to justify their habit)
and the acute effects will not be considered further in this article.
Long-term effects
Although it has often been assumed that chronic cannabis use will
have similar effects on the airways to tobacco, objective evidence
for this is lacking. Since the early 1970s, studies have looked for
evidence of airway obstruction in cannabis smokers. Most of these
have failed to show an association between chronic cannabis use
and forced expiratory volume (FEV1) values (TABLE 1) . A systematic
review by Tetrault et al. in 2007 found that the evidence that cannabis was associated with airflow obstruction was inconclusive [10] .
Since then, at least three further studies have explored the association between cannabis smoking and airflow obstruction and/or
chronic obstuctuve pulmonary disease (COPD).
www.expert-reviews.com
Review
Aldington et al. studied lung function in a convenience
sample of 339 people in Wellington, New Zealand, who were
either nonsmokers, smokers of either tobacco or cannabis
only, or smokers of both substances [11] . This study found that
unlike tobacco, cannabis smoking had no effect on FEV1 values,
although there was a borderline statistically significant trend
to lower FEV1/forced vital capacity (FVC) ratios in cannabis
smokers which appeared to show a dose-dependant relationship. There were also statistically significant dose-dependent
associations between a lifetime cumulative use of cannabis and
specific airway conductance (sGaw) as well as an association
between cannabis smoking and hyperinflation measured as
total lung capacity by body plethysmography. Among those
who smoked both substances, cannabis appeared to attenuate
the effect of tobacco smoking on measures of airflow obstruction including FEV1, FEV1/FVC ratios and mid-expiratory flow
values, although these effects were also of borderline statistical
significance [8] .
A Canadian population-based study of 878 individuals aged
40 years and over found no association between exclusive cannabis
smoking and COPD. Only four COPD patients were exclusive
current cannabis smokers, and this small number limits definite
conclusions [17] . However, there was a statistically significant
interaction with tobacco smoking: smokers of both cannabis
and tobacco had an increased risk of developing airflow obstruction compared to nonsmokers, suggesting a synergistic effect of
tobacco smoking and cannabis in the development of COPD.
Smoking tobacco alone was also associated with an increased
risk of COPD [17] .
The lack of association between cannabis use and airflow
obstruction was confirmed in a recent report from the Dunedin
Multidisciplinary Health and Development Study, which tracked
a population-based birth cohort of 1037 individuals with information on cannabis and tobacco use and lung function at 18,
21, 26 and 32 years of age [18] . Unlike tobacco, cannabis was not
associated with lower FEV1 values or with the FEV1/FVC ratios
once tobacco use had been adjusted for. Nor was there evidence of
airflow obstruction among cannabis smokers who did not smoke
tobacco. However, there was evidence of increased resistance to
flow in the central airways with significant associations between
cannabis use, lower sGaw and increased airway resistance. These
effects were much stronger for cannabis than for tobacco. There
was also a significant association between cannabis use and hyperinflation as measured by the FVC on spirometry (12 ml per jointyear [95% CI: 3.0–21.0]), total lung capacity (TLC; 25 ml per
joint-year [95% CI: 13.9–36.0]), functional residual capacity
(15.1 ml per joint-year [95% CI: 4.8–25.4] and residual volume
(12.6 ml per joint-year [95% CI: 7.0–18.3]) by plethysmography,
and alveolar volume (17.8 ml per joint-year [95% CI: 6.8–28.9])
by gas dilution. This association with hyperinflation was stronger
for cannabis than tobacco.
Like the Aldington study, earlier reports from the Dunedin
study had documented a borderline-significant association
between cannabis use and lower FEV1/FVC ratios [8,19] . It is now
apparent that this trend to lower FEV1/FVC ratios was owing to
539
Review
CME
Lee & Hancox
increases in the FVC rather than cannabis-induced decreases in
the absolute value of the FEV1 [18] . Taken together, the pattern of
findings from the studies by Aldington et al. and the Dunedin
study suggests that cannabis causes central airways resistance to
airflow (lower sGaw), associated with prominent symptoms of
bronchitis [8,10,18,19] and hyperinflation, but that there is little or no
effect on the FEV1 and airflow obstruction. Of note, the practice
of mixing cannabis and tobacco in the same joint is uncommon
in New Zealand, enabling the researchers to study the effects of
tobacco and cannabis separately.
In summary, there is currently no convincing evidence that smoking cannabis
Table 1. Epidemiological associations between cannabis use and
causes airflow obstruction. This may be
lung function.
surprising and appears to conflict with
the consistent evidence for increased resisStudy (year) Study
Subjects Results
Ref.
design
(n)
tance to airflow in the large airways. In
addition to the studies reported previously,
[18]
Hancox et al. Observational 919
Increased lung capacity and airway
four other studies have found that canna(2010)
cohort
resistance in marijuana smokers. No
evidence of airway obstruction, gas
bis smokers have increased levels of airway
trapping or impaired gas transfer
Raw and/or lower levels of SGaw than non[17]
users or tobacco smokers [11,18,20,21] . These
Tan et al.
Observational 878
Marijuana smoking not associated
(2009)
cohort
with increased bronchitic symptoms
findings suggest that cannabis has signifiand COPD
cant effects on large airway function asso[50]
ciated with bronchitis and mucous producSherrill et al. Observational 856
Pulmonary function was reduced
(1991)
cohort
in subjects reporting
tion, which are greater than those found
marijuana smoking
for tobacco, but has little or no effect on
[51]
airway obstruction and the risk of COPD.
Tashkin et al. Observational 394
No effect of marijuana smoking on
(1997)
cohort
Aldington
et al. (2007)
Crosssectional
339
Marijuana associated with airflow
obstruction, hyperinflation and
large airways impairment
[11]
Bloom et al.
(1987)
Crosssectional
990
No effect of marijuana on FEV1
or FVC
[45]
Cruickshank
(1976)
Crosssectional
60
No difference between marijuana
smokers and controls
[46]
Hernandez
et al. (1981)
Crosssectional
23
Normal spirometry in
marijuana smokers
[47]
Moore et al.
(2005)
Crosssectional
6728
marijuana use not associated with
decreased FEV1/FVC ratio
[48]
Sherman
et al. (1991)
Crosssectional
63
No significant difference in
FEV1/FVC and DLco in marijuana
smokers and nonsmokers
[49]
Tashkin et al.
(1980)
Crosssectional
189
Marijuana smokers had lower sGaw
compared with controls (p < 0.001)
[21]
Tashkin et al.
(1993)
Crosssectional
542
Marijuana smoking associated with
airway hyper-responsiveness with
lose-dose methacholine
[52]
Tilles et al.
(1986)
Crosssectional
68
Marijuana smoking regardless of
tobacco smoking, resulted in
reduction of single breath DLco
compared with nonsmokers
[53]
Tashkin et al.
(1987)
Crosssectional
446
Male marijuana smokers had
reduced sGaw compared with male
tobacco smokers. No difference
in DLco
[20]
Total:
FEV1 decline
12,613
Taylor et al. performed two studies (2000 [8] and 2002 [19]) on the same cohort that have been superseded
by Hancox et al. [18].
COPD: Chronic obstructive pulmonary diease; DLco: Diffusing capacity for carbon monoxide; FEV1: Forced
expiratory volume after 1 s; FVC: Forced vital capacity; sGaw: Specific airway conductance.
540
Emphysema & bullous disease
There are now at least 36 case reports of
bullous lung disease attributable to heavy
cannabis smoking in English literature.
These cases consistently report upper lobe
predominance with relatively preserved
lower lung parenchyma (TABLE 2) . Despite
the presence of bullae on high resolution CT scans, lung function tests and
chest x-ray appearances have largely been
unremarkable in these patients. Most of
these cases have been reported in young
adults under the age of 45 years. This age
distribution may reflect the fact that older
generations may not have smoked much
cannabis or may be owing to a reporting bias. How cannabis might cause such
severe lung damage is not clear. It has been
postulated that the methods of inhalation
of cannabis smoke may cause significant
barotrauma. Cannabis smokers tend to
hold their breath for up to four-times longer than cigarette smokers, with a nearly
70% increase in inspiratory volume [22] .
This high lung volume and breath holding results in the prolonged exposure to
inhaled particulates at very high temperatures, which in turn may be responsible for
epithelial injury and inflammation.
Currently, the evidence that smoking
cannabis causes emphysema and bullae is
limited to these case reports and therefore
Expert Rev. Respir. Med. 5(4), (2011)
CME
Effects of smoking cannabis on lung function
Review
remains anecdotal. Although Tashkin et al. demonstrated modest reports may be that bullous lung disease is a rare complicashort-term decreases in gas transfer (DLco) among 30 men allowed to tion. The number of cases reported in the literature is small in
smoke cannabis ad libitum for 94 days [15] , none of the population- relation to the widespread use of cannabis. It is possible that,
based studies have been able to confirm that cannabis consumption when compared to tobacco, only a relatively small proportion of
is associated with persistent impairment of DLco [11,15,16] . This is in people are susceptible to developing parenchymal lung damage
stark contrast to tobacco smoking, for which a reduction in DLco is from cannabis smoke and even then, only if they smoke a very
probably the most sensitive indicator of parenchymal lung damage. large amount. Hence, impairment of gas transfer and macroIn Aldington’s cross-sectional study, exclusive smokers of cannabis scopic evidence of emphysema are unlikely to be detected among
were much less likely to show evidence of emphysema on high-reso- general population samples. This explanation would require
lution CT scans than tobacco smokers, suggesting that macroscopic parenchymal lung damage to be caused by a process distinct
emphysema is not a common consequence of cannabis use [11] .
from the central inflammation that is observed in most regular
Even though cannabis smoking is infrequently associated with cannabis users.
emphysema in population-based studies, two studies have found a
In summary, the existing data are unable to confirm a definite
trend towards increased static lung volumes among cannabis users. link between cannabis and bullous emphysema. However, the case
Both the cohort study by Hancox et al. [18] and the cross-sectional reports support the likelihood that at least occasional heavy canstudy by Aldington et al. [11] found greater total lung capacities nabis smokers are susceptible to this disease. Further evidence from
among cannabis users, while Aldington also found evidence that systematic observational studies is required to confirm this.
cannabis was associated with hyperinflation on high-resolution
CT scans. This is consistent with other studies demonstrating Different to tobacco?
that cannabis is associated with statistically significant increases in The findings previously summarized suggest that smoking canFVC on spirometry [17] . It is difficult to interpret the significance nabis does have adverse effects on respiratory function, but conof these increases in static and dynamic lung volumes: whereas trary to what is often assumed, the pattern of damage in cannabis
hyperinflation is usually a feature of emphysema, this seems to smokers is different from that associated with tobacco. There is
be unlikely without evidence that cannabis causes either airflow now clear evidence that smoking cannabis causes inflammatory
obstruction (measured by FEV1/FVC ratios), impaired gas transfer changes in the central bronchi and a consistent trend to increased
(DLco), or parenchymal destruction on high-resolution CT scans. airway resistance (or reduced conductance). Surprisingly, this does
There are at least two reasons why these observational studies not appear to have a great impact on the FEV1. Trends to lower
conflict with numerous case reports of severe emphysematous bul- FEV1/FVC ratios have also been observed in several studies, but
lae among cannabis smokers. Perhaps the most likely explanation this seems to be due to an increase in the FVC, rather than a reducrelates to the dose of cannabis smoked. Most of the reported cases of tion in FEV1. The higher FVC observed among cannabis users
bullous emphysema have been in very heavy cannabis smokers. For is consistent with evidence of hyperinflation seen on plethysmoexample, in the largest series of patients (n = 17) the mean lifetime graphy and on CT scans. The patterns of effects associated with
consumption of cannabis was 54 joint-years [23] . Although cannabis tobacco and cannabis smoking in a cohort of 32 year olds are
use is very common, such prolonged heavy use is not. Even in large compared in TABLE 3.
population-based studies there may only be
a small number of heavy cannabis users. Table 2. Reports of bullous lung disease in cannabis users.
Indeed, in the cohort study by Hancox Study
Pateints Mean Mean
Results
Ref.
et al., none of the participants had accumu- (year)
(n)
age
joint-years
(years)
lated more than a 30 joint-year history by
the age of 32 [Hancox RJ; Unpublished data] . Beshay et al. 17
[23]
27
53
Upper lobe predominance
Purposeful samples, such as that used by (2007)
with bullae ranging from 0.3
to 12 cm
Aldington, may be more likely to identify
such heavy users, but it is important to note Johnson
[40]
4
38
NS
All had upper lobe bullae
that Aldington et al. applied very strict et al. (2000)
and normal lower lobes
exclusion criteria to their sample to exclude Gao et al.
[41]
1
23
NS
Bilateral upper lobe bullae,
the possibility of respiratory effects owing (2010)
more prominent on the right
to other illicit drugs. This may have also
[42]
Hii et al.
10
41
74
Upper and mid-zone
excluded the heaviest users of cannabis. The (2008)
emphysematous bullae
only exclusive cannabis smoker with macro[43]
26
>10
Extensive cystic and bullous
scopic emphysema on high-resolution CT Phan et al. 1
(2005)
lung changes primarily
scanning in their study had a 437 joint-year
affecting lower lobes
history [11] .
[44]
39
NS
Large apical lung bullae
The other reason why systematic stud- Thompson 3
et al. (2002)
ies have failed to identify the lung function changes reported in individual case NS: Not stated.
www.expert-reviews.com
541
Review
CME
Lee & Hancox
Why cannabis and tobacco should have different effects on
the lungs is not clear. As noted, other than nicotine and cannabinoids, smoke from the two substances contains a similar mix
of chemicals. It is possible that tetrahydrocannibinol, a known
short-term bronchodilator [10] , has long-lasting effects on lung
function (although the short-term bronchodilator effect seen in
single-dose studies does not persist during continued use [15]).
It is also possible that differences in the concentration of some
unidentified substance in the smoke results in these differences.
However, it seems more likely that the different methods of smoking cannabis compared to smoking tobacco are responsible for the
different effects on lung function. Cannabis is usually smoked
unfiltered and the smoke from cannabis is hotter compared to
filtered tobacco smoking [22] . Cannabis smokers also tend to take
much deeper breaths and employ breath-holding techniques to
increase the absorption of tetrahydrocannabinol as bioavailability
ranges from 18 to 50%, depending on the volume of air inhaled,
the depth of inhalation and the duration of retention of smoke in
the alveoli [24,25] . It is possible that by using Valsalva maneuvres to
increase the uptake of tetrahydrocannabinol, smokers also subject
themselves to hyperexpansion of the chest and the potential for
barotrauma. Interestingly, although prolonged breath-holding and
Valsalva maneuvres appear to be widely used by cannabis smokers,
studies indicate that is not necessary to perform these maneuvres
because the psychoactive effects of cannabis are similar if it is
smoked normally [26,27] .
Cannabis can also be taken in a variety of ways: either rolled
and smoked like cigarettes, inhaled through specialized devices
that use water filtration, bongs or such as vaporizers, and can
also be consumed in cakes, beverages and oils. To date, we are
not aware of any research looking specifically at the methods of
cannabis delivery to the lungs and their long-term impact on
lung function.
Importantly, whether cannabis and tobacco have synergistic
effects on lung function is a question that remains unanswered.
Most cannabis smokers also smoke tobacco and it seems likely that
they would be predisposed to a combination of effects. Tan et al.
Table 3. Differences in lung function associated
with cannabis and tobacco use.
Measure
Cannabis
Tobacco
FEV1
↔
↔/↓
FVC
↑
↔
FEV1/FVC ratio
↔
↓
TLC
↑
↔/↑
RV
↑
↑
DLco
↔
↓
sGaw
↓
↔/↓
↔: No association; ↑: Increase; ↓: Decrease.
DLco: Diffusing capacity for carbon monoxide; FEV1: Forced expiratory volume;
FVC: Forced vital capacity; RV: Residual volume; sGaw: Specific airway
conductance; TLC: Total lung capacity.
Data taken from [18].
542
found that although cannabis smoking alone was not associated
with an increased risk of COPD in their sample of older adults,
it appeared to increase the risk among those who also smoked
tobacco [17] . However, other studies have found little evidence
that cannabis modifies the effects of tobacco on lung function.
Rather, the pattern of abnormalities found in those who smoke
both substances suggests an additive effect or a combination of
the different tobacco and cannabis effects, rather than synergistic
action [11,12,18,19] .
The fact that there appears to be a difference in the pattern of
lung function abnormalities associated with tobacco and cannabis
does not necessarily mean that cannabis will not have a similar effect to tobacco for lung cancer and other health problems.
However, conflicting reports published in recent years have also
been unable to resolve the issue of whether cannabis smoking
causes lung malignancies.
While it has been found that cannabis condensates are more
cytotoxic, mutagenic and have a greater tendency to induce chromosomal damage and in a more erratic fashion compared with
tobacco [28] , a systematic review of the evidence by Mehra et al.
in 2006 failed to demonstrate a clear increased risk of lung cancer
among cannabis smokers after accounting for tobacco use. They
cite methodological deficiencies in the observational studies that
they reviewed and a lack of adjustment for tobacco smoking as
the main reason they were unable to reach the conclusion that
cannabis is a cause of lung cancer [29] . The evidence that cannabis
smoking causes lung cancer remains elusive [30] . For example, a
large American cohort study found no evidence of an increase
in overall cancer risk, and no increased risk of lung cancer in
particular among cannabis smokers [31] .
Recently, a New Zealand case–control study of 79 cases of
lung cancer showed a trend towards an increased risk of lung
cancer of about 8% for each joint-year smoked (compared with
a 7% increase in risk for each pack-year of cigarette smoking).
This increase in risk was only evident for the heaviest tertile
(>10.5 joint-years) of cannabis smokers who had a relative risk of
5.7 (95% CI: 1.5–21.6) after adjusting for cigarette smoking and
other potential confounding variables [32] . By contrast, a larger
case–control study of 2252 subjects in Los Angeles (CA, USA)
did not find an increased risk of lung cancer nor for oropharyngeal
cancers in cannabis smokers despite some subjects smoking very
large amounts of cannabis (in excess of 60 joint-years) [33] . Possible
reasons for the differences between these findings include differences in study design and the selection of controls, selection bias
of the cases, difficulty in quantifying cannabis use, and the difficulty in separating the effects of tobacco from those of cannabis
in people who smoke both [29,30,32] . Further data are urgently
required to resolve this issue.
The continuing uncertainty about the risk of lung cancer associated with cannabis highlights the problems associated with
studying the effects of an illegal and unstandardized substance
such as cannabis. On the other hand, these conflicting epidemiological findings are matched by contradictory biological data from
in vitro studies which have found that that cannabinoids have
both antineoplastic effects [34] and can also stimulate growth of
Expert Rev. Respir. Med. 5(4), (2011)
CME
Effects of smoking cannabis on lung function
lung cancer cells [35] . In fact, despite the similarities in chemical
properties, the pharmacological effects of cannabis and tobacco
smoke differ substantially and this may influence the carcinogenic
potential of the smoke [36] .
Expert commentary
For a substance that is so widely used, the paucity of evidence
on the respiratory effects of smoking cannabis is surprising. The
evidence that we have suggests that cannabis definitely does have
respiratory effects, but that these are different to tobacco. The relationship between cannabis smoking and the common smokingrelated problems associated with tobacco such as airway obstruction, emphysema and lung cancer is not clear. Notwithstanding
the difficulties in conducting research on illegal substances and the
problems of quantifying cannabis consumption, further studies
with large population samples and long-term follow-up are needed.
Case reports of bullous emphysema among cannabis smokers are difficult to reconcile with systematic observational data.
These cases are probably rare, although they may also be underrecognized because of under-reporting of cannabis use. They are
likely to represent the extreme end of the spectrum of cannabis-related lung disease, occurring only in very heavy smokers.
However, the evidence remains anecdotal and the development
of bullae and emphysema as a consequence of smoking cannabis
is not supported by the available systematic observational studies. Clearly, more needs to be done to confirm whether there is a
genuine cause-and-effect relationship between smoking cannabis
and lung bullae and, if such a relationship exists, a threshold at
which irreversible damage occurs.
We also need research into the methods of inhaling cannabis
and the influence that this may have on its respiratory effects.
Whether breath-holding and Valsalva maneuvres can explain
the association between cannabis use and lung hyperinflation
is intriguing: nothing in our understanding of lung physiology
appears to indicate that such simple maneuvres could make such
a marked difference to lung function.
A relatively unexplored area is whether cannabis has therapeutic
potential as an acute bronchodilator, either as an adjunct or an
alternative to current drugs. As noted, cannabis has a long history
as treatment for asthma [101] . It is unlikely that anyone would advocate smoking cannabis to treat obstructive airways disease, but
there may be less harmful ways to deliver the drug. Early research
investigated the effects of cannabis aerosols [15,37] . More recently,
vaporizers have been proposed as a method of inhaling ‘medical
cannabis’ in a smoke-free form. An internet survey suggested that
users of vaporizers had fewer respiratory symptoms but we are
not aware of any published long-term studies of their effects [38] .
Despite the continuing uncertainty regarding the effects of
cannabis on the lungs, we suggest that health practitioners routinely ask about cannabis use when taking a medical history.
Although medical students are taught to ask about illicit drugs
(particularly intravenous drugs), until recently, little attention has
been given to quantifying cannabis use [39] . Given the widespread
use of this substance in many countries, this should be carried
out far more often. It is particularly important for patients with
www.expert-reviews.com
Review
unexplained respiratory symptoms, apparently ‘idiopathic’ lung
bullae or pneumothorax, lung, and head and neck cancers. While
the relationship between cannabis and these diseases may still
be unproven, by raising awareness of cannabis use, we are more
likely to establish whether there is a causal relationship or not.
The research findings may also have implications for drug
policy. We have strong evidence that cannabis causes bronchial
inflammation, respiratory symptoms and affects lung function.
While we do not yet understand the full significance of the
pattern of lung function changes documented by the research,
it is clear that smoking cannabis is not harmless to the lungs.
Cannabis is also a controversial cause of lung cancer and emphysematous bullae in a small but uncertain number of users. It is
beyond the scope of this article to consider whether these harms
are best reduced by maintaining the illegal status of cannabis,
decriminalization, or by legalising and regulating its use. What
we can say is that cannabis is harmful to lungs and that drug
policies should take this into consideration. We also recommend
that future policies should encourage further research into the
health effects of smoking cannabis.
In conclusion, cannabis has been shown to have a range of
effects on lung function that are different to those found with
tobacco. Acute inhalation of cannabis produces bronchodilation,
but chronic use is associated with bronchitic symptoms, central
airway inflammation, and increased large airway resistance to
airflow. There is also evidence for lung hyperinflation, but no
convincing evidence that cannabis smoking leads to airflow
obstruction and COPD. Despite the case reports of emphysematous bullae among heavy cannabis users, it has not yet been proven
that cannabis causes emphysema. Cannabis also contains many
carcinogenic substances but it remains controversial whether it is
a cause of lung malignancies.
Five-year view
We have a great deal to learn about the effects of cannabis on the
lungs. Over the next 5 years we anticipate more studies examining
the effect of smoking cannabis on lung function:
• Large population-based cohort studies with longer periods of
follow-up. Hopefully these will include heavier cannabis smokers to clarify the effects of cannabis smoking on the risk of
developing COPD;
• Case–control studies of lung cancer to assess the link between
cannabis smoking and lung malignancies. The current evidence
is conflicting and further studies are urgently needed;
• Case–control studies of cannabis use in patients with bullous
emphysema and correlation to quantity of cannabis smoked. To
date, we only have anecdotal evidence from case reports and case
series linking cannabis to lung bullae. This contrasts with the
failure to demonstrate a link between cannabis exposure and
emphysema in population-based cohort studies. More case
reports will not resolve this issue; we need analytical studies of
cannabis exposure among people with bullous lung disease and
control subjects;
543
Review
CME
Lee & Hancox
• In addition, we hope to see research into the different ways
of inhaling cannabis, for example, comparing ‘bongs’ or water
filtration devices with unfiltered ‘joints’ (direct cigarette
smoking) and with other devices such as vaporizers. These
methods of inhalation may have markedly different effects
on the lung, but we are not aware of any systematic studies
of this;
• A better understanding of the long-term pulmonary effects of
repeated Valsalva maneuvres and deep breath holding commonly used by cannabis smokers is also needed. This is a difficult issue to study, but we hope that more imaginative
researchers than ourselves will find a way;
• We hope that there will be research into cannabis users perspectives of the health risks of cannabis smoking. How do they
decide which method to use for smoking it? What limits cannabis consumption – do users titrate the dose according to their
level of intoxication? This would identify educational needs
and the potential for harm reduction. This information may
also inform future drug policies;
• We expect medical marijuana to become more widely used in
the coming years and that more countries will legalise its use.
We need to know more about the potential adverse effects of this
and also of the potential effects of synthetic cannabinoids on the
lung. An intriguing possibility is that cannabinoids have an
unexploited potential as a bronchodilator. While the bronchodilator action of cannabis has been known for more than a
century, we still do not know if this could be useful in practice;
• Finally, we anticipate that taking and quantifying an individuals cannabis smoking history will become as routine in clinical
practice as recording tobacco exposure. Doctors should inform
their patients about the known effects of cannabis smoke in
causing bronchitis. It may be difficult to persuade users to stop
smoking cannabis but they need to be advised of the possible
risk of lung cancer.
Key issues
•
•
•
•
•
•
•
•
•
Cannabis is widely used throughout the world and is currently the most common illegal drug.
The pattern of lung function abnormalities among cannabis smokers is clearly different from those associated with tobacco smoking.
Cannabis smoke has potent effects on the bronchial mucosa and is associated with large airway inflammation and symptoms of bronchitis.
Systematic research into the long-term effects of smoking cannabis on lung function show increased large airways resistance
and hyperinflation.
By contrast there is no convincing evidence that smoking cannabis causes obstructive airways disease or emphysema.
The numerous case reports of bullous emphysema in the literature have not been replicated in systematic studies. It is likely that these
represent occasional complications among extremely heavy cannabis smokers.
More data are needed on the controversial issue of whether smoking cannabis causes lung cancer.
The advantages and disadvantages of the different methods of inhaling cannabis (joint, bong, pipe or vaporizer) are unknown.
Cannabis has acute bronchodilator effects but there is no evidence that this is clinically useful.
References
6
Papers of special note have been highlighted as:
•ofinterest
••ofconsiderableinterest
Hall W, Degenhardt L. Adverse health
effects of non-medical cannabis use. Lancet
17(374), 1383–1391 (2009).
•• Thoroughreviewonstudiesoftheeffects
ofcannabisonrespiratoryfunctionand
symptomsupuntil2005.
11
1
2
3
4
5
•
United Nations Office on Drugs and Crime
World Drug Report 2008. United Nations,
United Nations Office on Drugs and
Crime. United Nations Publication (2008).
Wide-rangingreviewofeffectsofcannabis
useonmajorbodysystems.
7
Noyes R Jr, Brunk SF, Avery DA et al.
Analgesic effect of d-9-tetrahydrocannabinol.
J. Clin. Pharmacol. 15(2–3), 139–143 (1975).
Walden N, Earleywine M. How high:
quantity as a predictor of cannabisrelated problems Harm. Reduct. J. 5, 20
(2008).
8
Hoffmann D, Brunnemann KD, Gori GB
et al. On the carcinogenicity of marijuana
smoke. Recent Adv. Phytochem. 9, 63–81
(1975).
9
Taylor DR, Poulton R, Moffitt TE et al.
The respiratory effects of marijuana
dependence in young adults. Addiction
95(11), 1669–1677 (2000).
Nocon A, Wittchen HU, Pfister H et al.
Dependence symptoms in young
marijuana users? A prospective
epidemiological study. J. Psychiatr. Res.
40(5), 394–403 (2006).
Boden JM, Fergusson DM, Horwood LJ.
Illicit drug use and dependence in a New
Zealand birth cohort. Aust. NZJ Psychiatry
40(2), 156–163 (2006).
Lange P. Marijuana and the lung. Thorax
62(12), 1036–1037 (2007).
544
10
Tetrault JM, Crothers K, Moore BA et al.
Effects of marijuana smoking on
pulmonary function and respiratory
complications: a systematic review. Arch.
Intern. Med. 167(3), 221–228 (2007).
Aldington S, Williams M, Nowitz M et al.
Effects of marijuana on pulmonary
structure, function and symptoms.
Thorax 62(12), 1058–1063 (2007).
•• Studyoftheeffectsofcannabisonlung
functioninaconveniencesampleof
regularcannabisuserscomparedwith
tobaccosmokers,mixedcannabisand
tobaccosmokersandnonsmokers.The
onlystudytousehigh-resolutionCT
scanningtodetectemphysema.
12
Fligiel SE, Roth MD, Kleerup EC et al.
Tracheobronchial histopathology in
habitual smokers of cocaine, marijuana,
and/or tobacco. Chest 112(2), 319–326
(1997).
13
Roth MD, Arora A, Barsky SH,
Kleerup EC, Simmons M, Tashkin DP.
Airway inflammation in young
Expert Rev. Respir. Med. 5(4), (2011)
CME
marijuana and tobacco smokers.
Am. J. Respir. Crit. Care. Med. 157(3), Pt 1,
928–937 (1998).
14
Camps PWL. A note on the inhalation
treatment of asthma. Guys Hosp. Report 79,
496 (1929).
15
Tashkin DP, Shapiro BJ, Lee YE et al.
Subacute effects of heavy marihuana
smoking on pulmonary function in healthy
men. N. Engl. J. Med. 294 (3), 125–129
(1976).
16
17
Tan WC, Lo C, Jong A et al. Marijuana and
chronic obstructive lung disease: a
population-based study. CMAJ 180(8),
814–820 (2009).
Hancox RJ, Poulton R, Ely M et al.
Effects of marijuana on lung function:
a population-based cohort study. Eur.
Respir. J. 35(1), 42–47 (2010).
•• Population-basedbirth-cohortstudy
of1037participantssuggestingthat
cannabisandtobaccousehave
differentpatternsofeffectson
respiratoryfunction.
19
20
21
22
•
23
24
25
Williams SJ, Hartley JP, Graham JD.
Bronchodilator effect of d1tetrahydrocannabinol administered by
aerosol of asthmatic patients. Thorax 31(6),
720–723 (1976).
•• Canadianstudyinanolderpopulationof
over40yearsofage,whichidentifiesthe
riskofchronicobstructivepulmonary
diseaseamongcannabisandtobacco
smokersandthosewhosmokeboth.
18
Effects of smoking cannabis on lung function
Taylor DR, Fergusson DM, Milne BJ et al.
A longitudinal study of the effects of
tobacco and marijuana exposure on lung
function in young adults. Addiction 97,
1055–1061 (2002).
Tashkin DP, Coulson AH, Clark VA et al.
Respiratory symptoms and lung function in
habitual heavy smokers of marijuana alone,
smokers of marijuana and tobacco, smokers
of tobacco alone, and nonsmokers.
Am. Rev. Respir. Dis. 135(1), 209–216
(1987).
Tashkin DP, Calvarese BM, Simmons MS
et al. Respiratory status of seventy-four
habitual marijuana smokers. Chest 78(5),
699–706 (1980).
Wu TC, Tashkin DP, Djahed B et al.
Pulmonary hazards of smoking marijuana
as compared with tobacco. N. Engl. J. Med.
318(6), 347–351 (1988).
Comparesdifferentmethodsofsmoking
cannabisandtobaccoandhighlightsthe
riskofinjuryfromthedifferential
methodsofinhalation.
www.expert-reviews.com
26
27
28
29
Beshay M, Kaiser H, Niedhart D et al.
Emphysema and secondary pneumothorax
in young adults smoking marijuana. Eur. J.
Cardiothorac. Surg. 32(6), 834–838
(2007).
Azorlosa JL, Greenwald MK, Stitzer ML.
Marijuana smoking: effects of varying
puff volume and breathhold duration.
J. Pharmacol. Exp. Ther. 272(2), 560–569
(1995).
Azorlosa JL, Heishman SJ, Stitzer ML
et al. Marijuana smoking: effect of
varying d 9-tetrahydrocannabinol
content and number of puffs.
J. Pharmacol. Exp. Ther. 261(1), 114–122
(1992).
Zacny JP, Chait LD. Breathhold duration
and response to marijuana smoke.
Pharmacol. Biochem. Behav. 33, 481–484
(1989).
Zacny JP, Chait LD. Response to
marijuana as a function of potency and
breathhold duration. Psychopharmacology
103, 223–226 (1991).
Maertens RM, White PA, Rickert W et al.
The genotoxicity of mainstream and
sidestream marijuana and tobacco smoke
condensates. Chem. Res. Toxicol. 22(8),
1406–1414 (2009).
Mehra R, Moore BA, Corthers K et al.
The association between marijuana
smoking and lung cancer – a systematic
review. Arch. Intern. Med. 166, 1359–1367
(2006).
•
Reviewupuntil2005ofstudiesof
cannabisanditsassociationwith
lungcancer.
30
Hashibe M, Straif K, Tashkin DP et al.
Epidemiologic review of marijuana use and
cancer risk. Alcohol 35(3), 265–275
(2005).
31
32
Sidney S, Quesenberry CP Jr, Friedman
GD et al. Marijuana use and cancer
incidence (California, United States).
Cancer Causes Control 8(5), 722–728
(1997).
Aldington S, Harwood M, Cox B et al.
Marijuana use and risk of lung cancer: a
case–control study. Eur. Respir. J. 31(2),
280–286 (2008).
•
Case–controlstudysuggestingthatthere
isanincreasedriskofdevelopinglung
cancerinheavycannabissmokers.
33
Hashibe M, Morgenstern H, Cui Y et al.
Marijuana use and the risk of lung and
upper aerodigestive tract cancers: results of
a population-based case–control study.
Cancer Epidemiol. Biomarkers Prev. 15(10),
1829–1834 (2006).
Review
•
Largecase–controlstudy(n=2252)
thatdemonstratednoincreasedriskof
developinglungcancerwithcannabisuse.
34
Munson AE, Harris LS, Friedman MA
et al. Antineoplastic activity of
cannabinoids. J. Natl Cancer Inst. 55,
597–602 (1975).
35
Hart S, Fischer OM, Ullrich A.
Cannabinoids induce cancer cell
proliferation via tumor necrosis factor
a-converting enzyme (TACE/ADAM17)mediated transactivation of the epidermal
growth factor receptor Cancer Res. 64,
1943–1950 (2004).
36
Melamede R. Cannabis and tobacco smoke
are not equally carcinogenic. Harm.
Reduction J. 2, 21 (2005).
•• Articlethatsuggeststhatcannabisis
notascarcinogenicastobaccobecause
offundamentaldifferencesinthe
pharmacologicpropertiesof
cannabinoidsandnicotine.
37
Van Dam NT, Earleywine M. Pulmonary
function in cannabis users: support for a
clinical trial of the vaporizer. Intern. J. Drug
Policy 21, 511–513 (2010).
38
Earleywine M, Barnwell SS. Decreased
respiratory symptoms in cannabis users
who vaporize. Harm Reduction J. 4, 11
(2007).
39
Douglas G, Nicol F, Robertson C.
Macleod’s Clinical Examination (12th
Edition). Churchill Livingstone, NY, USA
(2009).
40
Johnson MK, Smith RP, Morrison D
et al. Large lung bullae in
marijuana smokers. Thorax 55, 340–342
(2000).
41
Gao Z, Wood-Baker R, Harle R et al.
‘Bong lung’ in cystic fibrosis: a case report.
J. Med. Case Rep. 4, 371 (2010).
42
Hii S, Tam J, Thompson B et al. Bullous
lung disease due to marijuana. Respirology
13, 122–127 (2008).
43
Phan TD, Lau KK, Li X. Lung bullae and
pulmonary fibrosis associated with
marijuana smoking. Australas. Radiol. 49,
411–414 (2005).
44
Thompson CS, White RJ. Lung bullae and
marijuana. Thorax 57, 563 (2002).
45
Bloom JW, Kaltenborn WT, Paoletti P
et al. Respiratory effects of non-tobacco
cigarettes. Br. Med. J. (Clin. Res. Ed.) 295,
1516–1518 (1987).
46
Cruickshank EK. Physical Assessment of
30 chronic cannabis users and 30 matched
controls. Ann. NY Acad. Sci. 282, 162–167
(1976).
545
Review
CME
Lee & Hancox
47
Hernandez MJ, Martinez F, Blair HT et al.
Airway response to inhaled histamine in
asymptomatic long-term marijuana
smokers. J. Allergy Clin. Immunol. 67,
153–155 (1981).
50
Sherrill DL, Krzyzanowski M, Bloom JW
et al. Respiratory effects of non-tobacco
cigarettes: a longitudinal study in general
population. Int. J. Epidemiol. 20, 132–137
(1991).
48
Moore BA, Augustson EM, Moser RP et al.
Respiratory effects of marijuana and
tobacco use in a US sample. J. Gen. Intern.
Med. 20, 33–37 (2005).
51
Tashkin DP, Simmons MS, Sherrill DL
et al. Heavy habitual marijuana smoking
does not cause an accelerated decline in
FEV1 with age. Am. J. Respir. Crit. Care
Med. 155, 141–148 (1997).
49
Sherman MP, Roth MD, Gong H Jr et al.
Marijuana smoking, pulmonary function,
and lung macrophage oxidant release.
Pharmacol. Biochem. Behav. 40, 663–669
(1991).
546
52
Tashkin DP, Simmons MS, Chang P et al.
Effects of smoked substance abuse on
non-specific airway hyperresponsiveness.
Am. Rev. Respir. Dis. 147, 97–103 (1993).
53
Tilles DS, Goldenheim PD, Johnson DC
et al. Marijuana smoking as a cause
of reduction in single-breath carbon
monoxide diffusing capacity. Am. J. Med.
80, 601–606 (1986).
Website
101
PAPERSPAST – Star, Issue 6298, 1
October 1898, page 7
http://paperspast.natlib.govt.nz/cgi-bin/
paperspast
Expert Rev. Respir. Med. 5(4), (2011)
CME
Effects of smoking cannabis on lung function
Review
Effects of smoking cannabis on lung function
To obtain credit, you should first read the journal article. After
reading the article, you should be able to answer the following,
related, multiple-choice questions. To complete the questions
(with a minimum 70% passing score) and earn continuing medical education (CME) credit, please go to www.medscape.org/
journal/expertrespiratory. Credit cannot be obtained for tests
completed on paper, although you may use the worksheet below
to keep a record of your answers. You must be a registered user on
Medscape.org. If you are not registered on Medscape.org, please
click on the New Users: Free Registration link on the left hand
side of the website to register. Only one answer is correct for each
question. Once you successfully answer all post-test questions
you will be able to view and/or print your certificate. For questions regarding the content of this activity, contact the accredited
provider, [email protected]. For technical assistance, contact
[email protected]. American Medical Association’s Physician’s
Recognition Award (AMA PRA) credits are accepted in the
US as evidence of participation in CME activities. For further
information on this award, please refer to http://www.ama-assn.
org/ama/pub/category/2922.html. The AMA has determined
1.
that physicians not licensed in the US who participate in this
CME activity are eligible for AMA PRA Category 1 Credits™.
Through agreements that the AMA has made with agencies in
some countries, AMA PRA credit may be acceptable as evidence
of participation in CME activities. If you are not licensed in the
US, please complete the questions online, print the AMA PRA
CME credit certificate and present it to your national medical
association for review.
Activity Evaluation
Where 1 is strongly disagree and 5 is strongly agree
1 2 3 4 5
1. The activity supported the learning objectives.
2. The material was organized clearly for learning
to occur.
3. The content learned from this activity will
impact my practice.
4. The activity was presented objectively and
free of commercial bias.
42-year-old man presents with coughing. He quit smoking cigarettes 20 years ago but has smoked cannabis
several times per week over the last 15 years. He says that he is worried “that the pot is hurting my lungs.”
What should you consider in regard to the effects of cannabis on respiratory health?
£ A Cannabis is a potent bronchoconstrictor
£ B
The acute airway effects of cannabis do not appear to be sustained after 8 weeks of regular use
£ C
Many respiratory abnormalities related to cannabis use may be detected on physical examination
£ D Many respiratory abnormalities related to cannabis use may be detected on spirometry
2. Which of the following lung function values is most likely to be abnormal in this patient?
£ A Forced expiratory volume in 1 second (FEV1)
£ B
Forced vital capacity (FVC)
£ C
Airway resistance (Raw)
£ D FEV1/FVC
3. What can you tell this patient about the association between cannabis use and emphysema/bullous disease?
£ A Most adults with chronic cannabis use have evidence of pulmonary bullae
£ B
Bullae associated with cannabis are invariably located in the lower lobes
£ C
Chronic cannabis use reduces diffusion lung capacity for carbon monoxide (DLCO)
£ D Cannabis does not appear to promote emphysema
4. The patient is also concerned about the effects of cannabis on his risk for cancer. What can you tell him?
£ A There is conflicting evidence as to whether cannabis can promote lung cancer
£ B
Cannabis condensates are less mutagenic than tobacco condensates
£ C
Cannabis appears to increase the risk for oropharyngeal cancer but not lung cancer
£ D Bong smoking appears to confer a lower risk for lung cancer compared with smoking joints
www.expert-reviews.com
547
RESEARCH
Continued cannabis use and risk of incidence and
persistence of psychotic symptoms: 10 year follow-up
cohort study
Rebecca Kuepper, research psychologist,1 Jim van Os, professor,1 visiting professor,2 Roselind Lieb,
professor,3,4 Hans-Ulrich Wittchen, professor,4,5 Michael Höfler, research statistician,5 Cécile Henquet,
lecturer1
1
Department of Psychiatry and
Neuropsychology, South Limburg
Mental Health Research and
Teaching Network, EURON,
Maastricht University Medical
Center, Maastricht, Netherlands
2
King’s College London, King’s
Health Partners, Department of
Psychosis Studies, Institute of
Psychiatry, London, UK
3
Department of Psychology,
Division of Epidemiology and
Health Psychology, University of
Basel, Switzerland
4
Max Planck Institute of
Psychiatry, Munich, Germany
5
Institute of Clinical Psychology
and Psychotherapy, Technical
University Dresden, Germany
Correspondence to: J van Os,
Department of Psychiatry and
Neuropsychology, South Limburg
Mental Health Research and
Teaching Network, EURON,
Maastricht University Medical
Center, PO Box 616, NL-6200 MD,
Maastricht, Netherlands
[email protected]
Cite this as: BMJ 2011;342:d738
doi:10.1136/bmj.d738
ABSTRACT
Objective To determine whether use of cannabis in
adolescence increases the risk for psychotic outcomes by
affecting the incidence and persistence of subclinical
expression of psychosis in the general population (that is,
expres
© Copyright 2024 ExpyDoc
