hJktA L^lZi ^

NATIONAL ADVISORV COMMITTEE FOR AERONAUTICS

WARTIME REPORT

ORIGINALLY ISSUED

J;me 19U4 as
Memorandum Report

AEROCniAMIC TESTS OF AU AH-M-65-AZOH 1000-PODHD

RADIO -COHTROLIED BOMB IN THE LMAL I6-FOOT

HIGH-SPEED TUMEL

By E. 0. Pearson, Jr.

Langloy Memorial Aeronautical Laboratory
Langley Field, Va.

NACA

WASHINGTON

NACA WARTIME REPORTS are reprints of papers originally issued to provide rapid distribution of
advance research results to an authorized group requiring them for the war effort. They were pre-
viously held under a security status but are now unclassified. Some of these reports were not tech-
nically edited. All have been reproduced without change in order to expedite general distribution. I

L - 131

DOCUMENTS DEPARTMENT

Digitized by tlie Internet Archive

in 2011 witli funding from

University of Florida, George A. Smathers Libraries with support from LYRASIS and the Sloan Foundation

http://www.archive.org/details/aerodynamiccharaOOIang

-Ill f3^ u

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

Ma?^ORAND:J"[ REPORT
for the
Army Air Forces, Materiel Command
AERODYNAMIC TESTS OF AN AN-M-65-AZON 1000-POUND
RADIO-CONTROLLED BOMB IN THE LMAL I6-FOCT
HIGH-SPEED TUNNEL
33'- E. 0. Pearson, Jr.

STO.MARY

Tests were T.ade in the U.IAL l6-foot high-speed
tionnel to determine the aerod^mainic characteristics
of a 1000-pound AN-Iv'i-65-AZON radio-controlled bomb at
Mach nwiibers ranging from 0.2 to 0,6,

Over the Mach number range covered in the tests
the hinge-moment coefficients, yawing-moment coefficients,
and lateral-force coefficients exhibited no important
changes with increasing speed. The drag coefficients
increased gradually with Increasing Mach number but no
sudden increases v/ere observed. The effect on the bomb
aerod3mai:iJ c characteristics of antenna struts mounted
on the bomb tail was found to be small.

The rudder and aileron operating mechanisms wore
found to be capable of supr)l:'ring several times the
required torques for maximum control deflections at a
Mach number of 0.6 at sea level. The operating mecha-
nism is also adequate for maximum control deflections
at a Mach number of 1.0 provided that no appreciable
increases in hinge-moment coefficient occur between
M = 0,6 and M = 1.0. Hov;ever, because of uncertainty
as to the value of the hinge-moment coefficient at or
near M = 1,0, the desirability of providing m.ore
powerful control mechanisms was indicated.

INTRODUCTION

Tests have been conducted in the LL'AL l6-foct
high-speed tunnel to determine the aerodynamic

characteristics of a 1000-pound _'\i\[-M-65-AZON radio-
controlled bomb, Measure.mentG of rudder hinge moment,
yav;lng inoment, lateral force, end drag were made at a
number of tunnel speeds up to a Mach number of about
0.6 v\rhich corresponds to a speed of 67O feet per second
at sea level.

Aerodjniamlc tests of a two-thirds scale model of
a similar bomb were conducted previously at the Daniel
Guggenheim Airship Institute at Akron, Ohio. These
tests, however, were made at low airspeeds and, there-
fore, uncertain extrapolations of the data to the higli-
speed operating conditions were necessary in the design
of the control mechanism and in performance computations.

In drop tests of the bombs it was found that, vvhile
some of the experimental bombs performed satisfactorily,
a large percentage of the production version failed to
respond properly to the control. The present investiga-
tion was undertaken principally to determine if the
lack of control of the production bombs was due to
adverse compressibility effects. It was also desired
to obtain high-speed test data upon which to base per-
formance calculations and the design of the control
mechanism.

The investigation was undertaken at the request
of the Army Air Forces, Materiel Oommand.

3YLIECL3 AND DEFINITIONS

V free -stream velocity, feet per second

a speed of sound in air, feet -ner second

M Mach number (V/a)

p mass density of air, slugs per cubic foot

q d3niariiic pressure, pounds per square foot

ipv^)

Mtj hinge moh.enfc acting on one rudder, inch-pounds

Positive hinge moments tend to change the
ruddei' angle in a positive direction. (See
sketch in figure 1 illustrating the sign
conventions . )

distance from rudder hinge axis to trailing edge
of rudder, i^. .8l inches

area of one rudder, 0.232 square foot

hinge -moment coe f f i ci ent ( 'ViT-i/qSd )

N yawing-mcment, round -feet

r-csitive yawing moments tend to change the
angle of sideslip in a positive direction.
(See fig. 1. )

P area of rnaximvim cross section of bomb, I.91S square

feet

I over-all length of borab, ^,ol^ feet

C^ yawing-raoment coefficient (N/qPl)

Y lateral force, pounds

Cy lateral-force coefficient (Y/'qF)

D d^ag, pounds

Gq drag coefficient (D/qF)

v}/ angle of sideslio, degrees

Per positive angles the boiub nose is to the
right of the line of flight.

5r- rudder angle v;'th respect to the neutral position,
degress

For Dositive an^sles the trailing edge is to the
left'.

The dimensions given above, together with other
di'Tiensions which .night nrove useful, are Ghcj\rn in fig-
ure 2 .

DESCRIPTION OF BOMB ^WO APPARATUS

The AN-M-65-AZON bomb consists of a standard
fv!-D5 bomb case to -.vhich is fitted a special tail unit
equipped with novable control surfaces and housing a
control mechanism capable of being operated remotely

by radio. The bc'Ti'b cen be controlled onl:'- in the left
and right or azimuth direction, hence, the designation
AZON an abbreviation of "azlimith only." A gyroscopic
control mechanism operates the horizontal control sur-
faces differentially as ailerons to prevent the bomb
from rolling in flight.

For these tests part of bhe control apparatus was
removed from the tail and strain-gage equipment for
measuring rudder hinge moments vi/as Installed. Since
the ailerons and rudders were identical, hingo-mom.ent
data apnly eqtially to both. For this reason no provision
was made 'f'or measuring aileron hinge moments directly
and the ailerons were locked in tho neutral position
throughout the tests.

In order to facilitate Installation and testing
the bomb was m.ounted in the tunnel In a position 9^°
in roll from its normal flight position so that rudders

The bom:b vi^as supported on the tunnel center line
by means of a single vertical strut 8 inches in chord
and of NACA section 16-OO9. The strut was shielded by
a fairing of tiie same section to within about 10 inches
of the bomb case. In addition, four 0.030-inch wires
were attached to the bomb to provide lateral siipport.
The wires and support strut were mounted on the balance
frame and were included in the force measurements. A
photograph of the bomb installed in the tunnel is shown
in figure 5 .

The angle of sideslip of the bom.b was variable
from -20'^ to 2.0^^ throt^gh fixed increments by moans of
an internal indexing mechanism while the rudder angle
was continucusl.^'- variable from, -20^ to 20*^ by means
of a slotted plate arrangement.

As received, the bomb tall was fitted with four
struts wh:ch, in addition to serving as braces for the
tail surfaces, also served as the radio antenna. The
strirbs may be seen in figure 3.

TEST PROGSDuRE

The test procedure consisted, of measuring rudder
hinge moinent, yawing mcment, lateral force, and drag
at a numoer of speeds up to a Llach nuiuher of approxi-
mately 0.6 for each combination of rudder angle and
angle of sideslip. In datcrmining the bomb charac-
teristics only the negative range of sideslip angles
(0° to -20") was investigated in order to keep the bomb
tail outside the wake of the support strut and its
fairing. The angular range of Qo to 20° was employed
in determining the support-strut tares. The range of
rudder angles tested v/as from -15° to 20°. Most of the
test r-uns were made with the antenna tail struts
installed but a few runs were made with the struts
removed for purposes of comparison. This information
on the effect of the antenna struts was specifically
requested by the Army. The resiilts tliroughout the report
are given for the struts-installed configuration unless
otherwise specifically noted.

During the tests deflectionii of the bomb in the
direction of the air flow v/ere measured and corrections
to the yawing moment applied to accoijuat for the change
in position. In addition, a calibration was later made
to determine the angular deflections of bomb and rudders
vinder the influ.^nce of the uerodjmamic loads.

Strut tare ^orces were measured v/ith the aid of an
image strut mounted as shov/n in figure I4., Wire tare
forces were determined simply by making measurements
with the wires rem.oved.

Yawing moment, lateral force, and drag data have
been corrected both for tares and for the angular
deflections of bomb and rudders under aerodj'/namic loads.

Corrections for the deflections have not been applied
to the hinge -m.oment data. As presented the hinge -moment
coefficient is aboi.it 9 percent too low at the greatest
negative bomb and rudder angle and at the highest tunnel
speed, which is the extreme case.

RESULTS AND DISCUSSION

Curves of jraYi^ing-moment coefficient, lateral-force
coefficient, and drag coefficient versus Mach number are
presented in figures 5 through l3. It should be pointed

out that the ^^ = -0.5^ and Cp ~ 0° curves shown in
figures 5i 11, and 17 aro helieved to be slightly in error
due to friction in the Dalanco sjsten durjng that particu-
lar test run. This is in.dicated by the scatter of the
test points ■nartloularly at uhe lower speeds.

Derived curves of yaving r.Oinent, lateral force, and
drag coefficients versus sideslip angle are given in fig-
ures IQ, 20, and 21. Figure 22 sho\/s the lateral force
and drag coefficients at tri:in as, a function of the rudder
angle. The ei'fect of center-of-gravity location on the
stability and trim angles of the bomb is shown in fig-
ure 25. Einga-moinent data are presented in figures ?1\.
through 5.1. F'igure 52 is a photograph of a rudder after
structural fail^ire has occurred.

Ya"/irg norent .- Yawing-moment coefficient data are
presentad in figures 5 through 10. Turing the tests
yav/ing moments were measured about a point 22.58 inches
from the bomb nose. In the data nresented in the report
the moments were transferred to a point 2b. 60 inches
from the nose of the bomb, which was the center-of-
gravity location of sand-filled bombs used in the drop
tests, mentioned earlier in the report.

The figures sliow that the variation of the yawing-
moment coefficient with Mach laumber is sm.all for all the
rudder angles and angles of sideslip over the range of
speeds cox'ered in this investigation. Figures 9 ^^c' 10
show the i-ariation of yav/ing-mom.ent coefficient with
Mach mxnber with the antenna struts removed. From a
comparison of these figures Vifith figures 5 and 7 it
will be seen that the struts have -oractically no effect
on the yawlng-mioment coefficient.

Later a l force .- Curves of lateral-force coefficient
versus Mach nuriiter for the various rudder angles and
angles of sidoslio are presented in figures 11 through 16,
In general, the coefficient increases negatively with
increasing Mach nixnber. The change :s slight, however.
Figures 15 and I6 show the variation of lateral-force
coefficient with Mach niuaber with the antenna struts
removed. As in the case of the yawing-moment coefficients
the effect of the struts on the lateral-force coefficient
is small.

Drag .- The variation of drag coefficient with Mach
number is shown in figure I7 for various angles of side-
slip and rudder angles. No sudden increases in the drag

coefficient occurred in the speed range of the tests.
Curves for a few of the sice slip and rudder angles have
not been plotted in figure I7 to avoid excessive con-
gestion and overlapping. Gurvos of the drag coefficient
versus Maoh number with the anbenna tail struts removed
are ujiven in figure I8. A coTioai':! son of this figure with
figure 17 shows that for a sideslip angle of -0.5° and a
rudder angle of 0° removal of the struts results In a
decrease xn the drag coefficient of O.OO9 or about 5 per-
cent of the minimUiTi drag.

BOTib characters 3 1 i C3__as_a ^ft xn.c bi or of sidosli p
angle" .- Curves of yawTng-monient coefriciexit, lateral-
forct! coefficient, and drag coefficient versus the angle
of sideslip are presented in figures I9, 20, and 21,
respectively. These curves are cross plots of the faired
yawing -moment, Isteral-force, and drag coefficient curves
prsv?. ously presented. Since the variation cf the coeffi-
cients v;ith Maoh nuinber was not appreciable except for
the drag coefficient only the curves for a Maoh number
cf 0.6 are shown. It will be noted from figure I9 that
the bomb stability decreases slightly as the angle of
sideslip approaches zero. At the Icirger negative side-
slip angles and positive rudder angles there is also
some decrease in stability with increasing angle of
sideslip.

Trim conditions.- Figure 22. shov/s the lateral -force

and drag coefficients obta^'ning jxt triir. for various
rudder deflections. It will be seen that i'or tl-;e maxi-
mum rudder deflection of 20° and at a Maoh number of 0.6
the bomb trims in an attitude ^or which the lateral-force
coefficient is -0.Ii.22 and the. drag coefficient is O.56O.
Reference to figure l*^) shov/s that the corresponding angle
of sideslip for trim js -10.5°,

Bomb maneuverability .- The following table is pre-
sented to illustrate roughly the magnitude of the lateral
deviations possible rrtien the bomb is dropoed from dif-
ferent altitudes. It is assumed that the bomibing air-
plane is fl^d-ng at a constant indicated airspeed of
175 ^'iiiles per hour and that the maximum Dorab rudder
deflection of 20° is maintained over the entire flight
path. The left-hand column givds the height above sea
level v/hich is also considered to be ground level and
the right-hand column gives the approximate value of
the maximum lateral deviation possible when the bomb is
released at the corresponding altitude.

Altitude of releaso
(ft)

5,000

10,000
IS, 000
20,000
25,000

Apppoxiinate snaximu.-a lateral
deviation
(ft)

6 00
i6oo
2800

lj.200

5300

location. -

E f f e c t of cb anges in the ce nter- of -gravity

gure ~?3 sViOViTs the effect unon trim angle

and stability of a change in the center-of-gravity
location 2 Inches hackiivard or forv/ard from, the 23.6-inch
point. A more rearv;ard location of the center of gravity
results in a reduction in stability and an increase in the
angle of sideslip for trijn with a consequent increase in
the lateral-force coefficient. A location forward of the
23.6-inch point results, of course, in the opposite
effect.

7/ith the data presented yawing-moment coefficients
may be transferred from the 2o. 6-inch point to any new
point by the relation

^n ~

(cy

cos

*)f

C-

where

the new no
The value o''^
roar of

is the distance fron the 28,6-inch point to
int measured along the bomb longitudinal axis.
: is rositive if the new point Is to the
.6-inch noint and negative if to the front.

H i nge m oine n t » - Rudder hinge-moment data are presented
in figures 24 TFrough 31. Figures 2li through 27 shov; the
variation of rudder hinge-moraent coefficient with Mach
nijinber for various rudder aagles and angles of sideslip.
It may be seen that for the smaller angles of sideslip

and rudder angles bhero is little variation oj

moment coefficient v.'ith Mach nvimber.

^Qjop nrOnOU^^ -^.c*-^ r^r^7-> -t-Vi/.:! "l t-i^n^c^-\-* Q^->ri[»l

in any case

the hinge-

The change becomes
for the larger angles but is not important
over the range of speeds covei'ed in the tests.

It is clear that any loss of control of the oomb at Mach
numbers of 0.6 and belov/ cannot be attributed to radical
changes in hinge moments due to compressibility effects.
Figure 28 shows the variation of hlnge-mom.ent coefficient
with rudder angle for varioixs angles of sideslip and

PTach ntUTi'bers. Fi'2;ure3 29 a^^i 30 show the variation of
hinge -moment cosfricient with Mach n-umber with the antenna
struts removed. Figure Jl 3hows the variation of hinge-
moment coefficient vd th rudder angle for the same con-
figuration. A comparison of these figures with fig-
ures 2l|., 26, and 2^ shows that the struts have no
aDPreciable effect on the rudder hinge-mom.ent coefficient.

■'o ■

From tests in the LMAL Instnuiient Research Division
on tho rudder and aileron operating mechanisms it has
been determ.ined that the maximum torque available for
holding the rudde'^s in a given position is 200 inch-
pounds. For a rudder deflection of 20° and the corre-
sponding angle of sideslip for trim of -10.5^ the total
hinge moment acting on both rudders at a Mach number
of 0.6 at sea level is 52 inch-pounds. If it is assumed
that no change in the hinge-moir.ent coefficient occurs
between M = 0.6 and M = 1.0, the torque required at
a Mach munber of 1.0 is 1l|.5 Inch-po-unds . Such an assump-
tion is somewhat doubtful, however, and if appreciable
increases in the hinge -mcm.ent coefficieiit do occur, then
at or near a Mach number of 1.0 at sea level the torque
may be insufficient to maintain the maximum rudder
deflection of 20°.

For the maximum, aileron deflection of ±6° and an
angle of sideslip of -0.'^^ the torqiie required for one
aileron at a I.Iach number of 0.6 at sea level is 11 inch-
pounds. Assuming as before that no change in the hinge-
moment coefficient occurs betv;een M = 0.6 and M = 1.0
the torque roqiiired at I.' = 1.0 is 30 inch-pounds. The
maximum aileron torque available at the 6*^ deflection was
determined to be l\.0 inch-pounds, although the assuroption
of no change in the hinge-mom.ent coefficient between
M = 0.6 and V =1.0 is more reasonable for the case
of the small aileron deflection than for the larger
rudder deflection, the margin of torque available over
torque required at the higher speeds is not large in anj
case .

It would thus appear desirable from the foregoing
considerations to employ both aile^'on and rudder
ODeratlng mechanism.s m.ore conservative v/ith regard
to torque available.

Rudder failure .- During a routine Inspection of the
bomb upon coinple tlon of a test run It was found that
failure of one of the rudders had occurred along the

spot-welded skin joint at the hinge axis. Up to the
time the fail^ire v/as discovared 33 test runs, each of
about [j.5 T^iinu'^^es dM.ration, had been cornp?et9d. The
failure appeared to be due to fatigue. Figure 52 is a
photograph of tha broken rudder.

;oi:glusion^

Aorod^'namic tests cf a 1000-pound AZON bomb at
Mach nuiiibers ranging f"oir 0.2 to 0,6 have indicated
the following conclusions:

1. There we:^e no urpreciable compressibility effects
on the rudder hinge-moment coefficients, the yawing-
moraent coefr'icients , or on the lateral-force coefficients <
The drag coefficients increased gi'adually but no sudden
increases occurred.

2, Reiiioval of the antenna struts from the tail had
only a slight effect on the aerodynamic characteristics.

5. The torques supplied by the rudder and aileron
operating mechanisms were foijjnd to be several times
the torques required at a Mach mimber of 0.6 at sea
level for iaaximuin conti'ol doflecbions. If no appreciable
increases In the liinge-moment coefficient occur between
M = 0,6 and M = 1.0, the available torque will be
adequate at a Mach mjutf'^er of 1.0, Becai^se of some
uncertainty regarding the hinge -morr.ent coefficient at
a = 1,0, however, it would appear desirable to employ
nidder and aileron operating mechanisms having greater
available torque.

Langlev Memorial Aeronr-utical Laboratory

National AdTdsorv Co.-nmittee for Aeronaiatics
Langlev Field, Va., Jiune l6, 19-1^

Ernest 0. Pearson, Jr.
Aeronautical Engineer
Approved;

John Stack
Chief of Compressibility Research Division

FI'lTTRE LEGETOS

Figure 1.- Sketch showing sign conventions for angles,

forces and moments.

Figure 2.- Sketch showing dimensions cf AN-M-65-AZON hoTib,

Figure 5.- AN-?'-65-AZCN lOOC-poimd bomb Installed in the
16-foot high-speed tiinnel.

Figure 4..- Setup employed for determining support-strut

tares.

Figure 5«- Variation of yawing-moment coefficient v;1 th
Mach niunber i'or several rudder angles J/ ~ -.5°'

Figure 5.- Variation of yawing-racment coefficient v;lth
Mach number for several rudder angles ^ = -10.5^.

Figure ?•- Variation of yawing-moment coefficient vi/ith
Mach nui'flber for several rudder angles ^ = -15.5^^.

Figur-e 3.- Variation of yav/ing-raoment coefficient with
I'ach number ''or several rudder angles \|/ - -20.5^.

Figure 9«- Variation of :^'-awing-mo;nent coefficient with
Mech niimber for several rudder angles \|; = .5*^;
antenna struts removed.

Figure 10.- Variation of yawing-m^oment coefficient with
■lach number for several rudder angles ■]; = -15.5^;
antenna struts removed.

Figure 11.- Variation of latere.l-force coefficient v/ith
Mach number for several rudder angles '\f - -.5°.

Figure 12.- Variation of lateral-force coefficient v;ith
Mach number for several rudder angles \l/ = -10.5°.

Figure 13.- Variation of lateral-force coefficient with
Mach number for several rudder angles 'I' = -15.5°.

Figure 1I4..- Variation of lateral-force coefficient '-vith
Mach number for sevex'al rudder angles '<j - -20.5'^.

Figure IS.- Variation of lateral-force coefficient with
Mach n\imber for several rudder angles ^|/ = -.5°;
antenna struts removed.

Figure I6,- Variation of lateral-force coefficient ri th
Mach niainbGr for several ruddex' angles 1]/ = -15.5°;
antenna st'^uts rem^oved.

FI^HRE LEGFKD3 - Connluded

Figure I7.- Variatlcn of drag coef ■'"Icient with Mach number
for several angles of sideslip and rudder angles.

Figure I8,- Vc.riation of dra^ coefficient with Kach nimber;

antenna striits rer.ioved.

Figure 19 •" ■^'ariation of yawing-^onent coefficient with
angle of aideslip for several rudder angles M =0.6.

Figure 20,- Variation of lateral-:^orce coefficient v;/lth
angle of sideslip for sevort.l rudder angles M =0.6.

Figure 21,- "^.^ariaticn of drag coefficient with angle of
sideslip for several rudder angles 1,1 =0.6.

Figure 22,- Drag and lateral-i orc3 coefficients at trim
versus the rudder angle for several Mach nvjnbers.

Figure 25.- Variation of yawing-monent coefficient Viflth
angle of sideslip for three center-of-gravity locations
M = 0.6.

Figure 2i;.- Variation of rudder h:.n£e-:.noment coefficient
with Mach number for sevei'al rudder angles "^ = -0.5°.

Figure 25.- Variation of rudder hinge-noment coefficient
with Vaoh nixTiber .'"or sever.i^l rudder angles 'if = -10.5°.

Figure 26.- A,''ariation of rudder hinge-noment coefficient
with r.i'ach number for sevei'al rudder angles 'If - -IS. 5°.

Figure 27.- Variation of rudder hinge-moment coefficient
with Fach number Tos' several nadder singles •!/ = -20.5°.

Figure 2G.- Variation of rudder hinge-moment coefficient
with rudder angle for several tjigles of sideslip and
Kaoh numbers .

Figure 29.- Variation of rudder hinge -mom^ent coefficient
v;ith Mach num/oer for several rudder angles V; ::: -0.5°j
antenna struts removed.

Figure JO,- Variation of rudder h:Lnge-mcment coefficient
with Mach number for se
antenna struts rem.o\/ed.

With Mach number for several rudder angles ^i/ = -15.5°

.1 , _ I A_ _ _■»

Figure 31.- Variation of rudder hiiige-moment coefficient
with ru.dder angle; antenna struts rem.oved.

Figure 52.- Kudder failure.

H/r?g>e ax/s

/fu.dc/er

■/? \ NATIONAL ADVISORY

COMMintL FOR AERONAUTICS

/if^ure l.-S/refc/? s/iomng s/g/? converTf/io/ys for a/ygf/es.
forces. ar?c/ /7?o/7?enf6.

CO <^

=r OS

C3 Sc

_J o

<;i

<Z^ ttj

fet=

z s

•a

CO •
C .-I

E 3
O fJ

•a

XJ 0)
C QJ

CL I

1 x:

O tio
O -I

o s:

r-H

z o
o o

ts] <^

< I

1 lO

lO r^

I O)
I ♦^

< c

•H

I
to

Figure 4.- Setup employed for determining support-
strut tares.

WPgfPP

gftMMU Jim ^

;.| 1 1 1 1 1 1 1 1 14-1 p fiT

A1
/^/quLr<s. /7.- Vczr/cit/on of drag croefflcienf with

A^uoh number for several cui^Jes of

^id&3/ip and rudder an^Jes.

::;!:;:;;:::ti::::Ka::::::::iunu::::::sR:Knsn::::::;;»::::::::»:::::E::R::M:

TFr , V I

:::;RK::::::::a::h;!»:t:::r::::::::K:::!;:n:uKu:nt:a:»:'.::e::::;u:::::::h tu:uK^^

■ «■■■■■«■■■■■■( ■■■■■■■■£■■ ■■■■•■I ••■■■■■■■«•«■■«••■•■■■■■■■■■ ■■■■■i»ai •■■■■•■I ■■■■■^■■■■r •! ■■■«■■«•«■ ■■«««r -•■■■■■■■•■^-■k. vaaaai*

• ••

i-H

Li
(U

•a

CM
Li

UNIVERSITY OF FLORIDA

3 1262 08103 292 1

UNIVERSITY OF FLORIDA
DOCUMENTS DEPARTMENT
120 MARSTON SCIENCE UBRARY
RO. BOX 117011
GAINESVILLE. FL 32611-7011 USA