- .. , .. .

APPROVED FOR PUBLIC RELEASE

(x
,

#:,,,.
,. .. .

,.,. . ., .
L% ,,,Jil;y
. k NOT TO BE TAKEN FROM -s iqf$
.
t.%
. .
~rNo.35 & tw

.,

txk
t-
*

VERIFIED UNCLASSIFIED

*~
C.&
-

-.

1.. __ _ I

APPROVED FOR PUBLIC RELEASE

L .-
 I
APPROVED FOR PUBLIC RELEASE
I
. .
I

I
,. I

.- .
VERIFIED UNCLASSIFIED

,*

,.,.,,. . . .. . . .. .-
ti-:. r .. ..... -.1:

1 - ..

*
. ti

APPROVED FOR PUBLIC RELEASE


APPROVED FOR PUBLIC
.e . 9** . .0
RELEASE *.
..00
.0::
. ,. 00
:: -
**
,-,,a@*
.-L- . . *a
!.

.*. 00 .:. *.
.?oa~ : .; : .
0: :00 *a
**
-. be
.. 000 . .
.

. *

.
A5STRAC T.
.

It is shown that, whatever the temperature to which a sectior,

of the a+mo .phere may be heate?, no seli-propagatin~ chain of nuclear rFwLCti Q~i6

. is Ijkely to be started. The energy losses to radiation alwuys overcompensate

*-
tke gains due to the reactions. This is true even with rather extravagant

nssumgtions
. concerning the reactivity of the nitrogen nuclei of tke air. The

t}~]y d~sq~iet~~g feature is t~t the safety factorq, i.e. the ratio Of ~osSeS

whi.~h greatly exceed the bombs now under consideration. but even if bombs

required voluze (i.e., greater thsn 1000 cubic meters; are employed,

transfer from electroris to light i]uanta by Compt.un scattering will pro-

further safety f~ctor and will make u chair. reaction in air impossible.
;

9.

UMMSSIFIEI)

.
.

.*e
O.
#

06 . *: .=
.4
.:::.: 9.*
APPROVED FOR PUBLIC RELEASE
be. . . m..
9** . .

t
 APPROVED FOR PUBLIC RELEASE
,,

F.

1., Introduction
.-

The Letonatian of a fission or thurmonuc]ear bomb produces a high

.
. t.omper.ature
which will stimulate the reaction of atomle nuclei of the air

with each other. If an ignition point exists and is surpassed, tho thermo-

nuole8r re~~tiorA my be prop~gated to all parts of the atmosphere.

.
. Propagation of the raa:tim demands that the energy production iu

each newly entered region exceed the losses from that region. The energy

. &Fpears in the farm of the kinetie energy of Particles which are pr~duats of

e~~grg~c nuclear reactj.ofis. The product particles, through collisions, share

their energy with the particles of the air an(? help maintain the temperature.

9n the o:her hand, the share of the energy given the electrons is rapidly
,,,
ruliated away. Ahis constitutes the chief energy lass. ~he radiation c&.r.-

not help to ]~inkin temperature becaustiof the great transparency of air,

afid %ecause, even if the heated volume is great enough to lvoontai.anthe radia-

tlon, the heat capacity of space fo:-ra<iation is so.great that the energy

produced in the reaction is many orders ofmapyit.ude too small to maintain the

needed radiation temperature.

.
On R first evaluation of the danger or igniting the atmosphere, one

. cun assame that the reactio~. ?r~auct partio]es dcl not effectually disperse

. their energy but deposit it where they aro produced. Then Sne can compare this
*
rate of ?evositton with the radiation rate~ The temperature, if such exists,
+.
.
at which the energy production rake equals the racliakive loss rate will be the

temperature of ignition. gecause of the uncertainties in the knowledge of

--
these proc~~ses, the policy should be ado?ted of exaggerating the dangers
.
b

APPROVED FOR PUBLIC RELEASE

 ,Oe ..0 ..
e.0.
APPROVED FOR PUBLIC
RELEASE
.9* ..ao
O.

tiNtiASSIFIEn 8
** 0*
~o::
@*

.ea 8*
.**
.** O.*
.*

.

,
-4- e . . . .e ..: .:~ O.
.0. :

.0 : ~
-*:ew
m .0

1:

.
*
. so: O.
:*:..

is eifeoted with an Unh9Wli cross seot ion, and 6ince the transfer 5e%ween
.

nuclei which %-;e comparable KWSS is uslalYy more raD~cl than the transfer to

electrons, all th~ eri9rgy goes first to esta91i6hing a .

naol ear temp2rnture.

+ trsnslar ra$e to electroas and radiation rate by the electrons.

. This tempera-

zure rzy be considerably lcw;er Aan the nuolesr temperature and thus increase

.
%.
Tpe ~recg.tt.~gc~nsidsrations will yie?d an evaluation ofthe mrgin

0 !7tiaiety as a fuastion of the temperature o;the atmosphere. ~ fir~l section

will be devoted to the behavior of a hct air mass OF finite volume and to the

besting whi$h a fisflion or themotiuclear born bmayactually be able to supply.

2. THE ENERGY FR(XXICTUMl I

The nuolear reaztions most to be feareci in air are the rti2?~iMkS cf

p i rti Ji nitrogen nucleic Other notable constituents of the citmosphere, oxygcn ,- ~

t>ar~ar:
and rare gas nuclei~ are much more stable than nitro~en, yielding

09inprSt~Vf31y little erwrgy. There i~ perhaps one other nucleus that shm~lc! be

@fe9 care!ul attention: tha proton. It.is ordinarily much less abundant in

the atmosphere than is nitrogen, but clouds of steam raised over khe weans by

intense heating, may raciically nltf3r that situation. This danger will be

::ivenattention after the tlsoussion of the reactions of nitro%en nuclei with

etichother has been campleted.

-.
Tk,e ~,ltroqea-nikro~eflreuctions iriwhich stiffizient er&er4Y is

procllx>edto enable the product particles to surmount each others Coulomb

. . barriers we the following:

.
.

.
0.
99* .
. :: : :.
O
O
:: 9

0 ..: ** ..:

0 : : *.
.O.
-.


::. :
APPROVED FOR PUBLIC RELEASE 9*


0:. 8*.
0:

;
.
m
 APPROVED FOR PUBLIC RELEASE

1 .

#
a
In parenthesis foil owing the r~actians are given the heights ofthe Ooulomb

barriers of the prod~.letoarticl es. In a~cor~ with the policy of atiopting the
+

m9st dtifi~sroazassump%iozs, it will be assumed that 17.7 Mev per reaction is

xwlet.see, tis is characteristic of the alpha emissi~n. (The .x~ture oi nitrogen

by nitrogen, the first reaction listed, is certaicly infreqaqnt compared to

;artic.le e,nissim, as r6Sctions of this t:{pealways are, ~~oreover, energy in

the form of y-rays will be uziavailau}efor carr:ying on the chain.)

The cross sec$i>n for nitrogen encounters is not knowfi from obi3erva.

. tion. if the geometric cross section is adopted , and a reaction attributed ta

.
The geometric cross seccion amounts to just ak=2 barrls. The assumption that

this is a constant cross section for all energies will be referred to as tho

canst.ant~assu-mptim and will be denatod.by the subscript k. Calculation will

also bo undertaken with the assumption that the geomtric oress section
k
.-
.is reached only at energies surpassing the tarrier height, BZ8.5 WeV, between

nitrogefi nuclei. For relutive energies lower than this barrier height, a
*
CYOSS section proportional to the G~mow penetratiori probability is assumed.

Me sub6cript G will denote tnis alternative set of ass~~mptions:

-.
,
.
= (Jk = 2 barns , E >9.6Mev.
.

APPROVED FOR PUBLIC RELEASE

..
 APPROVED FOR PUBLIC RELEASE

I .

nitrogen nuclei. E is the rel~tive energy, E =(1/2\w V2, as determinecl with

. .

barns.

The rate at whi.zh the nuclear reactions will prod~ce energy is

.-

i~~ energy units pm seoond per rlit.rorennacleus. N is the atomic density of

zi+rogefi in air, N * 4(10j~9 nuc~eiicm$O fly UT is meant an average over

the &xwell distribution of the prcid~xt o!cress sestioa anti relative velseity:

of T is written for kx tem~eratl-u-e. 1? ff is simply 2 barns at all energies,

.

(x) ~ = 2 4m- tarns.cm!seco

FOY great-enough temperatures also

At low temperatures, with T in &v,

.

e.~94/T1/3
(X) G%lo&(loj~O barns.om/sec (T$O.Z Mev)

. Iiesults for intermediate taparatur~s were obtained by nmerical integration.

.-
. Th9 contribution of the collisions with er,ergy greater tkn the barrier height
.

APPROVED FOR PUBLIC RELEASE

 APPROVED FOR PUBLIC RELEASE

. Using the above results for (~) , one obta ins the energy mwi-dction
4.
rnte per nitrogen nucleus :

..
(dE/dtjk = 0.42 ~ (lfcYJ/t.L8fK)
Tin M,v.

wi,th tlx ico~stant cross Sectim assumption. This is also the high-temperature

a~ymptot~ in the case of the C-amow.factor assumption, For low temperatures,

the latter assumption yields.:

. Fig. 1 exhibits the variak.ion of the production rate with temperature for the
.
two assumptions.
.
The pGCsible contributions by reactions of proton~ with nitrogen

mtll now be discussed, ~he onYy ones of these yielding enou&h energy to all}w

the product particles to surmount the %ulumb barrier are :

-. Tho saLAl]abundance cfN]5weig!.> hgavily ~.%inst the second of these reaotions.

.
The crass section of & protxm against nitrogen at high energies oannok exceed

th~ > barns asstumd for 1~ + N, sinse tha gwxnetrical cross section zmur is ex-
. .
.
. peeted to be about 1 barn. Xhe ~eometrica] cross section shoul~ persist down

to an energy at which the protui VA.G-I.M,::]is of thf~r2&~~~ t~@:,qtiJ410e&r

*.ae :.

UNCLASSIFIED:.:.:
. ::
;:~f;~:*
:0: 0.9
9*.
8
APPROVED FOR PUBLIC RELEASE
. 9

m. . .
,,
 APPROVED FOR PUBLIC RELEASE
UNCLASSIFIED
.

4s the asswned nitrogen-nitrlt~en rea.~tions.

..
.. There remoins the possibility that the products of tho N-+Y

rw.ctiGzs will themselves react with fresh

gains. This cossillility is enhanced during

product oartieles Lave not yet lost all the

m-mrgy to contrit.ute to fresh reections* %e magnitude of the eserEy release

in such secondary reactiozs will be of the sama order 6s for the proton reactions

discussed in the pr~cedir,g par&gra@. As a matter of fict, one of the most

prowi~e~t of thct primry Froducts are protons, The arguments of the preceding

par&graph, which iridicated that the proton remtiofis csAnnot add significantly

t,> the energy production attribuke~ to the N~N reactions, are valid also he?e.

Ore has available alsc these edeitionnl facts: the cress section of nitrogen for

noutrvns and for alplti particlec (both typos cf Particlt)s are also products of,

the primary reactions) have beefi measured up tc ecer~i6s of le- Nev. The
,
rosultfi ~b.owwidely spaced resonances which even a+. peak value are only of tho
.-
crder oian eighth of a tarn. This is much less than the two barns attributed to
.
%k.e primary N + X nrocass.
Jn. 8 i p cross sections should not diff~r much from N+ n cro;s sactions, which have
.-
. been measured UD to -1.- Mev. The cxasureaents indicate resonance in ;he cross
, section which even at peek value aru only ofthe crder of ari eighth of a barn.

APPROVED FOR PUBLIC RELEASE

 :00
APPROVED
FOR PUBLIC RELEASE
.0.
..

UNWNSIFIED:.::.,~~:.O j:::;

*.:9-. .e **... ..

:::
:0 :..::

: a ::

L: 9O -a:.0,*a
z,. Tiih LNERGY LC.SEO

..

The rain mcrgy loss is duo tc Brcmsstrahlung b:{the e] ectrnns, wt,ich

I
has the rate:

&
<

..

correction factora E being the kinetic energ~ of the electrcn. f9r Iuw eneugl-,

- (dE/dt)E = 1.7 ~
t
.
In Fenera),
.
-
LC
v(l+E! me-) ~
R ;Ij
[i/e ) Ho (i/e) 4H$1) (i/0)

e = T/mc 2 and, ~(1) , HI(I) are .Hankel f[mctions of the !jrs% ldnd,

. of zero ant first order, respectively ~ho 13remsstrahlur;grate is shown in fig. 1,

tic a !uncti.on of the electrcn tem.pera?.ure. A distinct.toriis ma?e here krtween

. put +Zheccrr~spc,nding electrori an 2 nuclear tam[)erature at the same 0.13scissa.

. .
.

. .
..

c
9
*W
9**
**
00


**
90
000

:0
,0
* * ** V< UNCMSSIFIEDV --=:
APPROVED FOR PUBLIC RELEASE *. -
 9*.
:*.
80..
*. O.
: : 8*9
APPROVED FOR
O.
90. PUBLIC RELEASE

***
~h!cLAS51FlED

.
, .OOOO:
..
-IG a..
* :
*m

C!a6n-=

.
..

averagec over a Aaxwell distribution of temperatur~ T@. For lcv~;temper~tt~res,

For higher temperatures, the relativistic .Maxwell distrikuticn must be used , giving :
.

(m) =(w)(awq
C?+mw+!l e+e
.
\l) (1)
[ (L;6\i{,, ( i;ej .2E, (i;$:.
1
-.
The nuclear ant? electron temperatures which wil? co-exist will be given by the
.
equti.kionDf the trer.sfer rate to the !lrcmsstrahlung rate. This yields :

APPROVED FOR PUBLIC RELEASE

 APPROVED FOR PUBLIC RELEASE

b .

..

0s . )2*otkler hULd the margin of 8afcty teomnes only -2.5 It seems des i~~t~le to

,)
btain a bettor experimental knowledge ~f the reaction cress section even though

temperatures approaching 10 l!ev seem hardly imp~ssible..

k.
.. Thcl temperate to w?,ich air

v$lure of air to be hetited. One can seek to estimate t.k,

e size or this VOIZ.W ty
can be hea tie<by a bcml? tegenc?~ on tho

examining the mechanism of enwgy transfer frm bomb to air. It W;ll kij sufr5-

Cient, ant nl~h less diff~~uli, to discuss the oossibil ities Uf hea t,ing +,he mini. -

cum volume wt,ich must be braught up to a high temperat we in order tc instire

, propagation of the react

ior.*

. The heating of too smll a volume will add enormously to t he c?ralnti orI

...
...-
..
,
O *e 9 O*
0
00 *m
** .*

.*
meo
*..*:
:0 0
. .. ... UNCLASSIFIED
0 **
APPROVED FOR PUBLIC RELEASE
., . .
 APPROVED
:00 FOR*e
O.PUBLIC
: O. RELEASE
* *9

u~!LAJ~lFIEf) m..
0.

*.

00
* .OO :.O : 9*.
*9:
,. . .em...

I..

ti-

.
,
...
.-..
. SIAfe, however, we assume thut 1 .? barns is tho efiergy..l~sscross section, i.e. ,

9
00
**
**

e@
*9

e**
. . . . . OO


**
APPROVED

ma*a
e RELEASE
FOR PUBLIC UIICLASSIF{[,

*: e,
9*e


 APPROVED
:00

FOR
*9 PUBLIC


0
: e. . RELEASE
0....
0.

0. ** :

,...: ;W
*89..
:0:
900:
0

@=9!a-

free pth in air of

retain a substanti~l ~rt

,>f the ewr~y prodaced. At lenst such u vclume must be heated ifiordar that the

. theraonuc)ear reac$ian ho su.s+~i~d.

An air sphere ~i $7 meters rfi,iius C:3n&AirIt4 about L(lO)X1 nUel Pj9 anli
.
,y~
7 3(lo)- 91eo trons. For it to ke heated to a nuclear temperature Tn = 10 Xev
\W
(Te= C).~?KISr)requires about. ).5(1G; h!ev of energy.

A *isaion bomb ~rodacae abou.. ,

+ {10)26 Mm per kg of active tmaterial,

when working with 100Y. efficieficy. If also all the produced energy goes to
4
nr~dueetile air temper%t~lre, theu 1.5(IO)J Kg of material would have to be

huri!ed up in order to obtaiz the d~ngero:ls 10 !!kv temperature. nctllelly at most

1 %CI the bo~b energy m~y be ~v~ila~le tO prudwe the temperature; the rest

A s:ipcr-bornb, buynin~ deuteriwn, may be mnre dangerous than a fission

. bomi, becauae its el!e~~!:is not put into radiation as it is within the fission.

!3Cm3. The deu+.eri.un

nuclei b~ve tcc small cmrte to CSUSe efficient ra.iiation.

.
,.
.

..

.9* 00 0

%.7+;+,

e
y
. 9
00 :00 *
b
.
9*
. *9
me :
bem
. 9**
* **

APPROVED FOR PUBLIC RELEASE

 APPROVED FOR PUBLIC RELEASE

.
.
.

.
%. .
Tk.ea~pareut]y dangerous sltuaticn discu~sed in the precs~ing para-

Severs] further effects exist which may nnke less of the aner~y

le?eksed by bombs available for carryin~ on a reaction in the air. I%erw may be
.

$rit.h palicy
till-l 01 making the most ,:angsro~s assumption wherever any uncert.ain$y

P e,tiEts, then the effeots in question must bs left out

APPROVED FOR PUBLIC RELEASE

 am
APPROVED
m. FOR PUBLIC RELEASE
**
...:*: *.


9

.**:e: ** . . s.:
**
::: 0.

6iiiia-
m

* :0 :00 00 coo .0

.gl+
:*.::
:
:* ::*


OO :00

inverse Comptofi effect acts to quench nuclear disintegration of

.
.
. as soon as this reaction has ex$ondsl over a raclius of a few

If a relatively still air mass Ls heated, t!;e radiation emitted will

simply ecwzpe. nut if a grtiter sphe~e or air is hsted, the quajl~a of

. e!ectrorr~~gneticradiation are likely to nuLke Ga.mpton CO1 Ilsloris with tha elec-
,.
trons Defore leaving ttie sphere of hot air. In such collisions the quantu will,

on the avera~:e, gain energy from the hct electror,s. The end effect is that

more energy is going into radiation and more omsrgy will become unavailable to

t!lethermor~ticloarre.a~tiona It cafibe shown% tnat the ratio of the Coinpton losses

to the bremsstrahlung losses for an electron at temperature T in air is

e

108s

since

.
e*
.-.-----P-
** *9 9*
---- 1

eae:e
::

:0 :060
.0

* Uee IA4J11 0 om 9* 99* 8*O 9

O9

:Yl




9*
*9
OO
*9
mmee*
* m.


O
*9
*
e om
APPROVED FOR PUBLIC RELEASE
 APPROVED FOR
9
0 :PUBLIC
.O
. .
m* . RELEASE 0

. ... :y{.e
**9.
me m.
a.
.

00 b..

th!? Cornptcm effect. rilw, i;s

j::+i
9*O : :. 0
000
**
. .

.
. .

.

by Lhe reacti:l:.
. I
.

%0 metersu After sono time this shol~ oan be traiteri as a ~lanei detonation
I

.
.

-1
become K +.iK5& the sqc~re of the meen free path ?,. For t,k.e
Jifi%sicri velocity

APPROVED FOR PUBLIC RELEASE

 APPROVED FOR PUBLIC RELEASE

m *. 9** 99* om 9
.* 8..
:

. With A s L& meters one see$e~.h.~:.

~he %u~fic~.~g by Conpton effect mist be.
. 99 :; ** *
eo~.e important as soon SS the ietorxiticn wave has-tr~velleci rvIOb me%ers.

Shcb this len@h is quite mall com~ared tc the depth of the atmosphere .

there ia

,
&nd thus
.
Wile the above wmccrs may be somewhat in error, th~ quenchir~ must

eventfully take place as Icng as the aetorition wsve is a simplo plane pertar~

btitiar,. Ihere renains the distant ;r~L&bility that some other less simple

mode ofburning may maintain itself in the atmospk.ere,

hen if the reuction is stopped within a sphere of a few hundred

meters raaius, the resultant e~rth-shock gnd the radioactive contamination of

the atmosphere might neoome catastrrprlic on a world-wide scale.

One f!!y ccncl;de that the arguments of this paper mke it unreason-
.
able to expect that the N Y N reaction could propagate. An unlimited propaga-

. tion is even ?*s6 likely. However, the complexity of the argument and the

absence of-satisfactory experimental foundations makes further work on the

sub$eot hig,hljr
cesj.rabt?.

?
.

*m b :e 00

e
O
-00
O**,
* :1 :0 ::
bo 0
* O* a- ** 9* *

APPROVED FOR PUBLIC RELEASE

 APPROVED FOR PUBLIC RELEASE
0 :
e*
:O.
*e..

O. . 9*
9**
*e. : **

m
*D
. . .

.
@*
99..**
CO:
9
.. .:0 ..* *9 *. . .
. .

.&fl.,..DATE
.:0,.,.:,:04
. :. A:.: O
0.45
o.\ 0.25 0.2
...
. ;,-
.
!- i:~,~ .j tL:+, + .1.:: .1;..[
-. T.- ,
; ; . . ;4
., I ...;! . ...
....-
,.
r
..
... -.. ,.; _,._f. ;.. 1!
; _,_,- i.! j ~ .:..:
. .. . LtC kON .iE ,.!-
,!!..
I J. ~ ;;. ~: ..r.~:~ .;: -.~ ;; ;
1. -llilli- -J$HTHT(M!U) ...
Fyr ,~ W(J y : >;,
k
, l~;-
::
,,;
!. 1 .; 1, L,, ._ ..: ......-!
.....
:; in T -:-111 ,
H.l
-:
;1 i
I ;,, ~ _ .~ L!-w-44 +t-
104 . . . ... ..- .+ .,,..
+-i- ;..
4.. - ,...4.
.I .? ..
-. . . .. . .1.,,
r-,..
.;. 7 .

Y~.,.:_ .- .-. .

]Qf
,il: ,[, .
1!
N
.. ,. ,,
1:

,,........ ...
-!..

. . .--

--

.~, ..
.,
-1

.-.

.4
-

.-

i_
-i!;-.,[ : ,;
1111 -

10, ,..- . ,. ,. ...

-. :
, --- .. I_. ~
-.-..
---,.-l.,
h -.-L. ..:..
. 1,...l_.
100C ,7. _LL_!iL .
L
- +

10(
>1--: .L-
~_r_. -.8. L.L
Ir-:
.. -

i:
,,
;:-[
. . ~ +j..$

: : ;:--/
-. -i

:: :-;
.:..!

@ .!.
,-1
!
-t-; }

-, ,_:../
.,

1-+
.;..
~-~~~
i
..
s 1- :s: I
!ll~
4 -f ..
,. .,., . ... . ., ,, .-. .
,,
,
1:
,,
:,, I
1,!. ,:
;: !.
,-
b

1(
;,,
,, ii;! I

11,
. . . . . . . ,.. .: . n=j= .-, -:- /.
,- r--- 3
. . .. :.-.,. 1.
: r ., .-..
. . . . .1..! -.1
,. .. :-:Nl .:. ,-
r . -, -- ,..
, ; ;-; -,t-
. ..-. . . .. :.,. ,. .
,. II
.-. . .. . L . ..L ,

o
1

1
w ~::1 -
2 3
IK.ku_Mk
4
LIliLJM 5.

:6 : 2:
*00*

2:910
* ** ** 9* ,*O *
O
!

O*
9*
b9b
*9
.* **
., .

9*
0,0
*000:
:
**
om
ee

b APPROVED FOR PUBLIC RELEASE e 9*9 .3 O

,- .,..
,- .,. . ..*.. .-. . ..- . . . .. . - . .. . . . .
 .
APPROVED FOR PUBLIC RELEASE
l-.----A- -4-!::!; :!!:!!,!!:>
3- - :..-.,,:: ,.:.,-----
,,
G;. .L. t. .-: r ,...
,., ,.-.J .- :=,:.; -.4_-L ;:.;, :
! .: ..: ... .:. .L.:. .-.2 :., ,!..-.:.,. . . . . . },. ~ ,.,,
,. .:.. .. -
1
w-. ----k--t~ .~,,: .~i,,, , l.--l_--._1 ! I L.l I I I
l_
9
8
7
G
~..
.,
.-- .,-
... ....
..:. --! -n+ Lv.:::
=j
r --+, --- ---. -.. .. . . . . . . . . . . . . . . . . . . . . . . . . . . ,, .,., -.
I
I . ... ---- . .-.== -, -.-< ,: ..:
4_
3
L---- L--L-IA
1. -.. .J-.. --, - L-L -t J:--L
I ; .!.. J_- I..l_!..l. .. . . L. Ll. LL.L.. L_l.; .l_..
1
I-
1 , 1 .,~,!
. .. . .. . . . .
I 1 !
. . . .
I
W-i-&
. . . ..
9. -t__--: i-. . f ,:. ,J..- .,~ t.
.
.. . ..- ,.. . .. . . . . ,,, { I
. . . .. .. . . . . . .. .. . !. . ..-. . . . ... ::.
,
?
,
.-
1.
?_
L
***O
Q o **..: ::0 o
,D
d APPROVED
cd FOR PUBLIC
@@*a
RELEASE. 0 mO
,., ,. . . .. .... . .. . :. ..: -,.-. . . .. . .. ..
APPROVED FOR PUBLIC RELEASE

APPROVED FOR PUBLIC RELEASE

 mm*
APPROVED FOR PUBLIC RELEASE

9: .

.9. *O btl

s:* b

:
0: :0 :

?
,<i

I DOCUMX( R(KIM I

REC. FROM ------------

DATE.___ ------------

REC...- NO.RX -<.

r
f > 3[46

i,

0 .0. . . .

m.

.*. .
9.
9
: .
::. : .:.

e .-
0 99* :0 m. .:* . .

** . *..

. . .
a.
: :0:
,0

9

:: *:
: Q.@:
00.. e.
*m .0

APPROVED FOR PUBLIC RELEASE