ir*t .,.

h i
C.P. No. 490 ;4,)Y.\L /$I,~--*‘, *jyj,-‘i ..hp;*?,; C.P. No. 490
(15,167) I ,.: “. (15,167)
A.R.C. Technical Report A.R.C. Technical Report

MINISTRY OF AVIAtION

AERONAUTICAL RESEARCH COUNCIL

CURRENT PAPERS

Effects of Design Speed

and Normal Acceleration
on Aircraft Structure Weight

M. E. Burt, B.A., A.F.R.Ae.S.

LONDON: HER MAJESTY’S STATIONERY OFFICE

I960

THREE SHILLINGS NET

 C.P. IL'O. 490

U.D.C. 110. 629.~3.071.5

Report No. Structures 130

June, 1952

RO&YiL AIRCRiU'T &?TAl%ISHiZ~?~

1. .*, 6 I

Effects of Design Speed ati Normal Acceleration

on Aircraft Struoture Veight

N.E. Burt, B.A., A.P.R.Ae.S,

Graphs show the structure wei.r&t of typical fighter and bomber

aircraft for ranges of design diving speed and maximum natmal acceleratim,
Varktions of wing load-ing and geometry, and the use of spar or box-beam
wing construction are investigated.
The calculations of structum weight are based upon methods developed
at the R.A.E.
 Pa&re

Introduction 4

Pr oceaum 4

Results 4

Design Sped (VD) and MomaL Acceleration (N) 5

wing~~aa'ulg ~3 Ge0metry 5

Sm-eepback
Typ of Construction
Aircraft Size
Conclusions
References

eend*.

Details of Structure T/eight Ca.lxxiL.ntiorm I

LIST OF '?ABZS
Table

Characteris-t$.cs of A5rcraft Investigatea I

Changes in Structure Weight (percentage of aircraft weight) II

for Changes in Design D5ving Sped (VD) and FadxredNomal
Acceleration Ccefficimt (N)

-2-
 LIST or' ILLT3Sx.UIom
Fimre
Fighter Aircraft - Variation of Structure Y/eight with V'
and Tli , for Values of A , t/c and X I
Fighter Aircraft - Variation of Structure T/eight with VD
and M , for Values of T::ing Loading (w) 2
Bar,ber A5rcraf't - Variation of Structure Teight with VD
and TJ' , for Values of A , t/c and h 3
RoTher Aircraft - Variation of Structure Keight with VI>
and N, for Values of Xng Loading (w) 4
Variation of Structure i'ie@t with Svee>back for Star&rd
Fighter 5
Wing Beight for Spar and Box-BearnDesigm - Standard Bomber 6
Variation of Stzucture height with Aircraft Size for Similar
Fighter Aticraft 7
Variation of Structure Keight -66th Aircraft Size for S5milar
Bomber Aircraft 8

-3-
1 Introduction
The specified values of design diving speed and maximum no?X&l
acceleration form the foundation of aircraft strength and stiffness
req#remcnts, and thus affect both airworthiness and structure vrcight.
The aim of this report is to give a general pictuj% of their effect on
structtire weight for use in project investigations.
Graphs show the struoture weight of @pica1 fighter and bomber
aircraft for ranges of design diving speed and normal acceleration, at
severs3 vs3ues of x5.n.g loading, aspect ratio, thickness/chord ratio and
taper ratio. Changes in structur e Veight for various conditions are
deduced. Tls effects of aircraft size ‘and of spar or box-beam sting con-
struction are considered. The effects of wing sweep-baok are also
considered.
The calculations of structure weight are based upon methods developed
at the R.A.E. and are &scribed in an Appendix. Simiiar graphs may be
constructed for cases not covered by this report.
2 Procedure
The structure weights of two "standard"* aircraft, a fighter and a
bcxnber, are calculated for ranges of design diving speed and factored
maximum normal acceleration coefficient. The oaloulations are repeated
with alternative values of some characteristics which may vary and alter
the structure weight. These are wing loading, aspect ratio, thickness/
chord ratio, taper ratio and s~cp. The standard values, and the other
values investigated, are listed in Table I.

All aircraft are assumed to be of conventional aluminium-alloy

oonstruotion. Two types of' vring construction are investigated. First
the "spar" type for which tie material providing bending strength is
ooncentrated in spar flanges, and is separate from the cover material
which provides torsional stiffness. The second is the "box-be&' type
for which tne material Troviding 'bending strength is distributed along
part of the top and bottom surfaces, so that the same material provides
both bending strength and torsional stiffness.

The total structure weight is *given by the sum of wing, fuselage,

tsJ.1 unit and undercarriage structure ?x5ghts. The ualculations of oom-
ponent structure wei$ts are based upon methods developed at the R.A.E?22$*
De-toils are given in Appendix I.
Tne airar,o.ft investigated are described throughout as the "fighter"
I
and the "bomber" , but the graphs represent the trends for aircraft with
any duQz v&ose characteristio s are similar to those assumed. In particular
the curves derived for the bomber apply to many civil and military tram-
port aircraft. Curves for aircr<a?t with charaoteristics not similar to
those investigxted can be constructed by the ssme methods.

3 Results
--
The results of the calculations are plotted in Figs.1 - 8, and are
discussed beloG.

* "Standard" aircrtit are defined atie those with chamoteristios

similar to fighters and bombers coming into production in 1952.

-4-
 The lighter form of 1tin.g construction, spar or box-beam, is assumed
in each case. The change from spar to box-besm construction is marked on
Figs.1 - & and is characterised by an abrupt reduction in slope of the
cuxves. The abrupt increases in slope occur when the weight of cover
required for torsion&l stiffness first exceeds that required for local
strength and stiffness under airloads, ard when the weight of shear webs
require-d for torsional stiffness first exceeds that for vertical shear
strength.
4 Design Speed (VD) and Normal Acceleration (N)

Figs.lA s..d 311give the structure weights of the standard fighter

end bcmber respectively, for ranges of VD and N , The curves for
total structure weight show these features:-

(4 structie weight increases with both VD and N ,

(b) the slope of the curves, which gives the rate of change of
structure weight with VD , increases as VD increases, but
is not greatly affected by the value of N .
(c) the spacing of the curves, tich gives the rate of change of
structure weightmith N, ia the saw for all values uf VD
and N when spar construction is used, but is reduced with
increase of VD when box-beam construction is used.
At average values of VD (600 knots for the fighter, 400 knots for
the amber) a change in structure weight af 1% of the aircraft TEight
corresponds to a ohange of about:-
(i) 35 knots in VD , or 1.5 in N for the standard fightir,
(ii) 35 knots in VD , or 0.8 in N for the standard bomber.

5 Ring Loading and Geometry

Figs.lB, C and D, and JE, C and D give curves similar to those
of Figs.lA and jh respctively, but :-rith values of v&ng aspect ratio,
thickness/chord ratio and taper ratio other than the standard values.
Figs.2A and C, and ,!& aizd C give similar curves for other values
of vring loading. Figs.lA end JA are reproduced as Figs.2B and @ for
convenience.
These cuzves all shav the general features noted in paragraph 4
for the standard aircreft, but the actual v&ights of the wing and total
structure are greatly affected by wing loading and geometry. This is
illustrated by Table II, vjhich gives changes in structure weight, expressed
as a percentage of the aircraft t?eight, for given changes in VD and N
under various conditions. The changes in structure weight increase rapidly
for increased aspot ratio, and for reduced wing loading and thickness/
tzkord. ratio.
Table II also illustrates the increasing importance of VD , relative
to that of N , as VD is increased.
6 Sweepback
In the raemod used for oalculating wing Weight, the weight of cover
material required for torsional stiffness, w!hich depends on VD , is
independent of sweep, so that variations of structure weigh-t with VD
are not affeoted.
 The weight of bending material, which depends on N , does varty,
and Fig.5 shows hi the ting, tail unit and tot,il. structure vIei.ght vary
with s~eepb%&. for the standard fighter. The se curves stho~ld Ix regarded
as approximate because the method bloc‘3 not allots for the high flcxur~al
stiffness Qhioh may be necessary for adequate stability and control for
sune stings of high svJeep and aspect ratio.

7 a of Constn~otion
At lovf v,n;Lues of VD the wLng cover is thin, and the lightest way
to provide ben&i.ng strength Is to use concentrated spar flanges. As VD
is increased the cover weight rapidly ' ecomes greater, because the required
torsional stiffness inoreases v,$.th VD!! , and above some "oritical" value
Of VD it is lighter to rcinforco the cover to provide the bending strength
( i.e. to uso box-beam construction). The critical value of VD is reduced
as N , wing loading and thiclmess/ohora ratio are reduced. Fi .6 shows
the full spar and box-beam curve s for three values of thickness 7 chord ratio.
The critical value of VD is lo-;~r for the bomber than for the fighter,
because there is a larger proportion of relief loads in the bcmber %'ing,
and because, as size increases, the weight of material requi~d for torsion
increases more rapidly than that roqaired for bending.

These facts explain several features noted in paragraph 4.. For spar
oonstruotion the bending material, affected by N , Luzd.the torsion
material, sffected by VD~ , are separate. The curves have the same slope,
~~hich increases as VD is i;ncreaseci, and the same spaoing. For box-hem
construction bending and torsion materijl are oorrmon. Usually the outer
part of the semi-span is designed by torsional requirements, BM the inner
part by bending. The outer part has a reserve of strength in bending and
a @Ten change in N involves less change in structure Weight than for
spar construotion. The survcs thus tend to oonverge abcrre the critical
v&he of vD .

8 Aircraft Size-
The tot&i. structure v?eight, expressed as a percentage of the airGraft
Treight, varies as the aircraft size is varied. The relative weights of the
various stXuc2;ural qononts also cknge. Cam is therefore necessary in
applying the results af this report to aircraft of sizes different from
those examined.
The structure weights Crc a fa&?ily of fighters ayld a family of bombers
are o&culated as a guide and sho~vn on Figs,7 and 8 respectively. These
airoraft have varying wfng are.2 but llstLandaral~ oharacteristics in other
respects, i.e. the ~-me '"r, , N , x 9 A , t/c , h, and the same proportion
of relief ILoGs, fuselage an3 tail areas, et3.
The curves suggest that the total structure weight of similar air-
oraft varies with (&.rcraPt ?Yeight)'*2C, instead aF (aircraft Weight)'o5o
as suggested by the theoretical "square-oube" l&W, vh?re’by aircraft weight
varies with the square, and the structure weight with the cube of the
Zinear dimension.

9 Conclusion
(1) A change in structure [email protected] of 1% of airoraft weight oorres-
ponds to change of about:-
(i) 35 knots in VT) ., or 1.5 in N for the flstanda~' fighter,

(ii) 35 knOtS in VD , or 0.8 in N for the f'standard" bomber.

-6-
 (2) These values are greatly affected ‘bTj variations in ting
loabing and geometry, and by the value of 'I'I, at lfiich the ohanges are
made.

(3) The importance of VD , relative to that of W , increases

as vD is increased.

(41 Spar type wing construction is lighter than box-beam for

low values of VD . A%ove scme vdue, which depends upon wing loading
a-nd geoine try, and aircraft size, kox-bcm construction is the lighter.
(5) The structure ~i&t of similar aircraft is calculated to
vary with (aircraft might, instead of (aircraft weight)"50 as
suggested Iy the theoretical "s&are-cube" layr.

-2
NO Author Title, etc.

1 E .L Ripley A method of wing weight prediction.

A.R.C. 14,269. &y, 195-l.
.
2 K.E. Burt and Prediction of fuselage and hull structuz-e
J. PhilXps ;~eQ~t;: ~
A.R.C, '15,421. ;;arch, 9952,

3 E.L. Ripley A siziple ~~~ethjd fcr tail unit stmoture

3eight predktim.
A.R.C, $3,710, Stmt,lti, Hwember, 1950.

-7-
 Jle-tails of Structure Weight Calculations_

1 wing weipglt
1.1 The method used is substcentially due to Ripley', but with some
simplifications in the treatment of relief loads and weight allowances
for speci~el features. It is described as briefly as possible exoept for
the dcprtures from the full method.
The %i.ng structure weight is given by

valere t+j,B = Weight of basio v&q struofzwe

= WC f WB + i$ + v;Is

wcl = v&i.ght of Oover ma teria1

WB = T&3ight of bending material

WS = Weight of shearmaterial

wIs = v;eight of internal struclture

WeisrF_t of speoial features
k-l =
might uf basic3 King structure

The use of the factor k is the first simplific\ation of the full

method. The v&&t of the ,sp%i.al feature, a is 0alculated by multiplying
the Weight CL' the basio tdng by a factor based upon experience of tigs
similar to that under investigation, not by individual a~sessment~

i.2 The weight of cover material for both sRar and box-beam construo-
tions is given by

v?hePe 7gTC = Weight of torsion box cover

WpJj = w&&-t cf remainder of uover

The Tom v&,/sq,ft (CT) of tne torsion box cover is given by

Equation 3.1~9 , provided that CT 4 CR as by the "smooth finish"
ourve of Fig.25. The uniform wts/sq.ft (CR of Xhe leading edge end
trailing edge are obtained fmn Fig.2.5, using the "smooth finish" curve
for the leading edge.
The torsion box is assumed continuous across the fuselage, but the
leading and trai1i.q edges are rmoved.

0 All references in this Appendix to Equations, Paragraphs and Figures

are to those cf Ref.1.
I.3 The weF$-h of bending material (WB) for spar type construction is
obtained from Equations 2."r 5, 2.22 and a simplified form of 2.21, which
is

K5 = x(m-wlq 19

There WR = weight of wing group (i.e. structure, engines fuel,etc.).

This is the second simplifioation. It is assumed that the spanwise
weight distribution of the items which relieve the airloads on the wing
is tne same as that of the airloads. Account of the relief loads is then
taka by subtracting their total weight (WP) from the aircraft weight in
the equation for KS .

The "two spar" curve of Fig.24 is used.

For box-beam type construction the spanwise distribution of the
material required f'or bending snd the torsion material effective in
bending are plotted as in Sketch 4.3. The weight of' bending material
is the additional material required for bending and shown hatched
?kcetch 4 3 The spanwise distribution of the net lnaterial required
for bending'i~ given by a modification of Equation 2.09.

%I = K7 N (W - WR) 7I2 19 I4 54

The spanwise distribution of the torsion material effective in

bending is obtained by assuming t&at three-quarters of the total torsion
box cover is effective in bending.
1.4 The weight of material (WS) rewired for vertical shear strength
is oalculated from Equations 5.10, 5.15 and 5.13 modified. The modifica-
tion to Equation 5.13 is to introduce the same assumption on distribution
of relief loads which was made for the bending material. The equation
beacmes
6.2
K;! = -y= b N set A (F - WP) I6
IO2

This value cf Ws is used unless over-ridden by that required in

the webs for torsional shear stiffness. Thii is given by

IiBBh
I-J, = (P + s) do 2 b set A CT

where p and q are the ratios of the depth of the aerofoil section at
the front and rear web positions to the maximum depth.

I.5 The weight of internal structure (WI,) is oakulated as given in

paragraPh 3.7.
2 Fuselage Weight

2.1 The method used is a simplif'ioation of the method given by Burt

and Phillips2. Examination of ike components of the fuselage structure
of a number of fighters and bombers shams that, for the purpose of this
report, fuselage structure weight may be expressed as
wllere WG = gross fuselage shell weight, calculated as in Ref.2 and
a, 2, are constants obtained from examination of existing aeropbnes.
Their values are given in Table I,

3 Tail U&t Weight

3.1 The tail unit structure weight j.s calculated Y~J Ripley's methcd3.
ALlowance is made for svveep'back and la% added to the total for mass-
ba.2.ance vtei pjht 8.

4 Undercqriage Wei&t

4.1 The undercarriage %eight is assumed to be 0.05Win all cases.

 TABLE I

CHARACiXfiISTICS OF AIRCRAFT IW!WPIG4TED

I

Fighter Aircraft Bomber Alrcraf t

I
Item i symbol standard 3ther Values wndard Ither Values
Value Considered Value Conslderad

Wing Area sq.ft. 300 loo. 200 2,000 500, 1000

500, 700 3000, 4000

King Loading lb/sq.ft. 50 25, 100 50 25, 100

Aircraft Weight lb 15,000 100,000

Design Diving Speed knots E.A.S. 600 400 r 500 400 200, 300
700, 800 500, 600

Factored Maxhnum Normal Acceleraff on I

Coefficient at Weight W
I 12.0 6.0, 9.0 4.5 3.0
6.0, 7.5
(Manoeuvre or gust)
t
i
Aspect Ratio A 4 2, 6 5 39 7

ThicknessiiZhord Ratio
t.
/c 11 1% .%, 1% 10% 5% 15%

Taper Ratio h I 0.5 0.25, 0.75 0.5 3.25, 0.75

Sweepbaok (on O. 25 chard line) h 4o" O”, 20°, 60' 25'

Torsion Box Forward

and Aft Limits
1
1
0.i 0
0.6 c
I 0.2
0.6
d
c

Effective Bending Depth 1 Spars 0.85

I 0.85
Y
Maximum Local Wing Depth ) Box-Beam 0.94 0.94

Weight of Wing Group WR 0.2 0.4

Aeroplane Weight T

Kefeht of u@ocial Features t 0.35

Basic Wing Weight
-
I
Fuselage
-’ Area
t

I
I
I

1.60
T 1.25
I
Wing Area I
Fuselage Fineness Ratio / 2.5

I a I 1.50 I 1.55
constants
Equation,
in Fuselage Xoight
Appendix I b 1 0.02 i 0.02
I
I’ 4.0 1 10.0
Fuselage llidth at Wing Level ft.
II L --
A--- I
i
Tailplane am Elevator Area ,/ 0.16 0.16
wm Area , 1 /
i /
Fin end Rudder Area
I
/ ;
0.08 , 0.08
Wing Area ,
 TAEiII

Chan,es in Sfxuctwre '7,,ei&t (l?ercentage of AixcraI"t Beight) for Changes in Desis

Diving Speed (Yj-,) and Yactorcd Normal Acceleratiori Coefficient (r;l) ‘

I
Changes
VD at
iii
I
Clknge in
1J fmm y iiit
t
-I1 1.7 i
; 4.6 I.9 I.3

1i ----I-
1.6 2.6 I.3 1.0

] 0’ 1 2.5 1 4.2 / 3.0 1 3.2 11 I 7

I 3.4 1 6.7 2.7 I.4

Thic&xess/Chord ) j% 4.0 4.9 1 3.1 i 2.7 I Thic'kness/Chord 57; 4.3 14c.3 3.1 0.4
Ratio Ratio
15% 1.3 2.0 I1 -I.7 q-7 I 1 13% 1.6 2.5 1.6 I.4

Taper Ratio lo.25 1.3 1 2.7 1.6 -1.6 / Taper Ra4io / C.25 1.8 0.6

I
,0.75 4.7 3.2 2.9 2.3 I i 0.75 1.9 5.0 2.3 1.6

Wing Loading Ving Loading

i 2.1 0.8
25
i
100
! 1.8 I1 A.4
 , . .

(o>
65 65-
6- WRlATlffl OF TAPER RATIO
STANDARD FIWTER U’ARIATION OF TH(TKt.ESS-CHORD RATIO
5 = 300 5 = 300 5=‘00 wJ;,$i,ooo
w = IE$OOO w= 15.000
60. w = 50 3 60- w= 50 w= 50 I
A= 4 A=4 A= 4
% = IO% A so5 ‘t/c= IO%
55 - A= 05
55-
- A=2 - A=025
--- --- A=075
50

50/
45 45 -

*9

‘“IN I2
40-

s 354
6
.I2
.9

‘O- STRUCTURE
6

.I2

,6
3
3 20

ii
ii
3
; ‘5

IO IO-

0 O-
4co SO0 700 4oc
KNOT:

FIG. l (A-6) FIGHTER AIRCRAFT -VARIATION OF STRUCTURE WEIGHT WITH V, & N, FOR VALUES OF A,% & )?
 6 (A) 65 (B> 65 cc>
I
w = so w = 15,000 4Ir =I00
s w = 30,000
= 300 S = 300 I
45a 9x060 lecco60
A=4 A :4

‘E/c = IO% 3 = IO%

55 5.5
hi05 A =05

IE STANDARD FIGHTER OF flG IA

37x 7500 50. I5003 50

I I I

45

soot IZCOD 40

35.

2250 3000 30

25. !
TOTAL
STRUCTURE
-
1500 mo 20-/-e/ I
A’

IS

750

FUSELAGE
5 UNDERCARRIAGE
5---s-------------
TAIL UNDERCARRIAGE

TAIL
0 0 - I
1 “I
.--
400 500 603 700 000 400 500 600 700 0C)o 4-o 500 600 700 000
DESIGN DIVING WEED - vr, - KNOTS

FIGZ(A-dFIGHTER AIRCRAFT-VARIATION OF STRUCTURE WEIGHT WITH V. & N, FOR VALUES OF WING LOADING (0).

l * .
 .

65 (A)
STANDARD
BOtmER
65
VARIATION
(B)
Of ASPECT RATIO
65-
VARIATION Of
cc>
TUICKNESS-
65
1 VARIATION Of
(0)
TAPER RATIO
CHOP.0 RATIO 5 8 2000
5 = 2000
s D 2000 w * 100,000 w = 100,000
5 i 2000
w = 10q000 w= 50 1 w .100,ooa W’ 50
4&l= 50 y= 50 ) 60
60 w- A= 5
flc = IO%
A= s A =os I: 25 % ’ IO%
‘t/c = IO% - A=3 - A=025
55 - 55
55 A -05 -- A=7 -- A -075

I
TOTAL ,
50
““FT-r-r- so-

45 .,I! 45-

Rw
40-

35 -
/

/
30-

25 :

2Oi
,
75
I5 - 60
45
z 30

IO,

5. I I I
FUSELAGE, UNDERCARRIAGE 5 FUSELAGE, UNDERCARRIAGE &
TAIL A5 STANDARD. TAIL A5 STANDARD.

O-
W 0 300 400 500 600 200 300 400 500
DESIGN DIVING SPEED - b- KNOT5

FIG&+D)BOMER AIRCRAFT-VARIATION OF STRUCTURE WEIGHT WITH V, & N, FOR VALUES OF A, %i & x.

 A-5
60 =4 = IO”/ 0

55w I TOTAL 1 ///f/ 1

5.5

50
I E STANDARD SOWER Of F,G 3A

45

4D N

--
35 30

30

25

YI I I I
I 5 L-----w/l :I,
75
30 20=Mi 44@JfJ

20,owJ

TAIL I
TAIL

zoo 400 600

01
200
I
300
I
400
I
500 600
3 o-
0
zoo 300 400
I
500 6m
OESIW ONING SPEED - v,, - KNOTS

FIG~(A-c)BOMBER AIRCRAFT-VARIATION OF STRUCTURE WEIGHT WITH V. 8, N, FOR VALUES OF WING LOADING (w).
1 * .
 IO.

O-

SWEEPBACK

FIGS. VARIATION OF STRUCTURE WEIGHT

WITH SWEEPBACK FOR STANDARD
FIGHTER.
 55-

iO-

SPARS
-B---b 00X BEAM

-5 - u= 50
S= 2000
A=5
x * 03
ro - N=4.5 _
lf
/
/
/I
I5 - /
I
I

O-

s-

3,
/
7’//1’
7,//’ /’

5-

L
3-

5-

I- I
2oc 1 300 4 500 6
OESl4N DIVING SPEEO - vn - KNOTS.
FIG.6 l WING WEIGHT FOR-SPAR AND BOX
BEAM DESIGNS. STANDARD BOMBER.
 ‘URE .

/WC

-FUSE -AGE.

-UNDERCARRIAGE.
I
-TAIL

AAl = 50 ‘URE.
= 4
$i = IO%
-A = l5
q-J = 600 KNOTS.
N = I2

-I- f WING

/UNDERCARRIAGE.

l=lG .7. VARlATI6~&%?hikTURE WEIGHT WITH

AIRCRAFT SIZE FOR SIMILAR FGHTER AIRCRAFT*
 WlNS

FUSELAGE

UNDERCARRIAGE
TAIL

u =50 TOTAL
STRUCTURE
A =5
SO- . 7C = lO”/a
h = o-5
v, =400 KNOTS
N = 4%.
50 -
i

WING.

FUSELAGE

UNDERCARRIAGE.

TAIL.

0 IO00 2000 3000 4000

WING AREA- SQ. FT.
FIG.8. VARIATION OF STRUCTURE WEIGHT WITH
AIRCRAFT SIZE FOR SIMILAR BOMBER AIRCRAFT
a
 3
.

c
cc
 C.P. No. 490
(15,167)
A.R.C. Technical Report

@ Crown Copyright 1960

Published by
HER MAJESTY’S STATIONERY OFFICE
To be purchased from
York House, Kingsway, London w.c.2
423 Oxford Street, London w.1
13~ Castle Street, Edinburgh 2
109 St. Mary Street, Cardiff
39 King Street, Manchester 2
Tower Lane, Bristol 1
2 Edmund Street, Birmingham 3
80 Chichester Street, Belfast
or through any bookseller

Printed in England

S.O. Code No. 23-90 I I-90

C.P. No. 490