Citation
The Ivory Trade

Material Information

Title:
The Ivory Trade
Creator:
Parker, Ian.S.C.
Publication Date:
Language:
English
Edition:
Volume 2
Physical Description:
typescript report plastic spiral bound

Subjects

Subjects / Keywords:
Ivory Trade
Africia wildlife

Notes

Abstract:
The Ivory Trade which consists of the commerce in ivory, biological aspects, discussions and recommendations and tables.
Preferred Citation:
Parker, I. [Ian] S.C. 1979. The Ivory Trade. [Four volumes, vol. four in three parts]. “Consultancy undertaken for Dr. Iain Douglas-Hamilton on behalf of the United States Fish & Wildlife Service of the Department of the Interior, and the International Union for the Conservation of Nature and Natural Resources, Morge, Switzerland. Available: http://ufdc.ufl.edu/AA00020117/00001.
General Note:
Ian Parker Collection Re: East African Wildlife Conservation.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the African Studies Curator (danrebo@uflib.ufl.edu) with any additional information they can provide.

UFDC Membership

Aggregations:
African Studies Collections

Downloads

This item has the following downloads:


Full Text
11 -, Tt 1 7,1-1 .. -- ::'..-- -,
, V ,: ,, -- :,,,,-, 1, n',,,' ,,, 7W 7w ....... ., ".: .,-.T., 7 1 1 ., -_ ",
" '7 _- 7 ., 11 "; -,., _-.-.; -_
,;., I 11 -11 I ..-, : 1. ., : S
-,", i e : l, _f,, 11 I I ': : 11 .,---,.. ., rl .:,
:, -, .: ,:'' ,' -* 11 4;,, I I ""., 1. ., '11- .., 'e- ,
,: '': i A I I ., -, ., 1: : -:%; I I I I I ,'.,, :"? -
F :.i- I ,.,. ." I .1 I I '- ", I. :. : ,:, I --- ,
- : I -- I ,-
- .. -,-- ,,i --', ,,,-, I ":,- .,!. ,., .2 .. I "I ": : I ,- -. ,: ;
I I ; n' I I : -:, : f I ., I .: .. ?
, ", I 11 I I I 11 I I : ,.. .1 1. "' ,- -, "
, I -- ,, -- "!: --,- V, -,,' : I- ; -,,
1 11' 1, "I ,, -,-,- ,-, -i -,- -- I "I. :''.-, I :,. -- ,,% ',
_, '' -,,.O, 1, .. 7 -, a ,.Illl
- *'., 11 11, ,,
,- I- ,-- -.,.. :, -1 -:. : I I : I I I ""
- '' I I -' rf .. ":.- : I 11 .1 ,, % -, -, ,, !; : I I -, ,', ".-"!"''
r, 1 '' -,- -, I -: I I I -.1 ;: .... ... 'I.,
-, ,- -, -,, 1, I I I ,. ',," -,_ 11 --- I ,. "," ,,, : -,
'- ,Ir, :, I I w ,,, ,
, 11-1, I.- ll-, I : I I I I., : _: I ,.,, ,- ,
- "-- I ,,, :. '' !"il
, --' '- ; ,,-, ,, ,-
I I ,,,'- '-,-- i I ,,, e, I I I .. ,.,
1, -, -1 1 I -- 1, .".". ,. : ',, .rl -, .... -
lll ; -, -,---- ""- !, "",-"- ,-,l e : I I I I I ., I --,.-,-,, ,, ,
,- I I rr : ,;. ,. ,. I, ,; ,., '. .,; e ,. I I I I I -- : I ,- I -Z
-, ,,-". Z .. 11 ,, I '. I -1 ; ., ,- .. ... ,. I -0 : _
, -, '' .1 ,,, I I. -. .,::. '' ., -, I I .1 -, I.: '"'. ;:., ,
%, -- -,,, -, ,; ,- I I ,
- ', '' I'L ,,- I '':, 1, I ", : j .., .. ': I I .,
'I, I I I '' (.- 1, .
, 1 ,I -11 -- l i 11
- ., Y', -- .. ', 4 I .1 I I I
, -1 : I i -- I ,;, I -: ':
, 1 I I I I ",- I I- : :.-.,. ..'. I ,. 1 ,:' .". -',,, I I I 11 .
"I I" I -" --' 11 "" i -,,, -- i", ', -, I I -ii.,,,. .." ,,, ,',-. :::,"',,::,.: : I I I w -- : ,
,:-., .. I -' .-,,
-, ,, ,- ,,,. .. ". ,--- ,-7. '. ,
, ,: :: % ., ],:, ,
, "". -, ', -,-,, ., : I ,, '. I 1. -- ; I I :. .- "-
I- ,- ,,- ,,,- ". ,: _71'.., .., 1. I .
, ,-
- I '. .
"', I I- I I .. : -, > --: ': "" ,., -, :,-. ;. ,, I I-- L', ,:- -_ ,::i: '. -, -, ': : leo .. I ., : ,
I I I I I 11 1. F: .,i '
o, -- -1- ... -1 z, .: ,,,, 7,:: : l ,. .1 I I : '. %, ,... .
I -, 1'1 I 1 "', : .:!,: :: ; : !:! ,-" "J. ...,
", : ,, -,% .,., I : % ,, .- 7, .",. ;;. ,,, I .. I '. 1, ,; I t., -'.'' '- I'll. ,,.",,: ,;. ': :., ',,
, .1 I -, -
'. I I :: .-,-
,.- _;, i: i I I 1.1; I ., ;" Z ,'
, ... -, ,- :-, h I ..;r. ,,-, ,.;-.::: I ,, :. ,,:- .?-- ,-.,
, I !,-; ,, -- ', .". .. ,:
2. 1 '.." ;. C, .- -. .. ..-,
- -lI.-l'-. .I-- '-. -,.. ..- 1" .1 :, '_.7'--,:: ,,', I -, r l '' I d.- I I 1
:: -, I ; 1- 1 ;. ,., .
,. I I I ", : :::: I .-, -'r
,%".; ,- < .,:.. 11 .. ...... I I I a ,
- ,; :, z : -1 --,":
:- -: .::", I -- --..If. 'I .
, i'..::.,:, : ',., ;.,: ..1, I I ., .- I 11 I I I z;. ;I,-.. :, ., .,
- ", .. I I I I ". L '. : 1. I I ; ---:T'. I I :
r I ... ,- ,- -, ... I .-... ." : '.
, -, : --,: -1 I. ,- ., : : -
. ;, i, -,,,,,. v,, ,, ," I : :,., ": ,,. ,,,,,":, ,-, -,,, :
,.,- : I lr I .. I I ,
- ,.. ,.
,. '. I I :
. I 'A. ', '' .- ,. -,. I I I I
,11 t, "i" ,,, ,' 11 7: ;- ,, -.
,, .... p ", X, ,'-- '. h "i ...... /:-- I I I I -- ,--- -, ;' ,
.. ". .- -. ,.;, -- :: .: -, ",, 1:1 I .1 : I -", --- : .-1'11,
, ,, ,,,,, -', ... : ,, I .1 --Ill. ,,,,
.. ,- : I I : : I I : m .., ,ml 1.
. --.-o !.,-, '. '.., ,-." ., ". '. ., .:: I I : r, -.''. .. ., ,, ; ..
-; '. ,, ... I '. :. .." -;:, .
, ', .:. "'. ": : I I ,- '. :, .... .. ,.- I -.
-, - : r -.- ..,- ,, _, ': ., ,,Z i I I I I '. I I -. .1 I I I I I I J ,-- .
, ,. !, ". '..: ., "" I I .1, ; ,,
: ,- -.,." ;- --- ,. ., I '," -- -, I I '.. ':
.;-", : I .,. -, % I. !, -,
1;1 ,;, '.. "'. -"; .,,;, I.., :: '-, : I I I -I '11
1: .1 z ,' .. '' ,. .. ,- ,.. L -, :,, -, I I I I -.. ., .. .." '.-
., ". : 1, I 1, 1, I .: 1.
.. .11.1 1, 1, ,, ,-,,. ''. I I I -. I ": I I :: : ., v I I
- I . I I I I ..
-1 ,,, .! I -.,-,, i,; -, '. ", ., : I I I : : ,
, .1 .. I 1. : I I I ., I I :: ''., ; ., .: i '. .
,. ,,- I -, ..:, ,, ,.,--- I.~ I a -1 ..' .,,-, : "', ,. I .. 11 (. .1. I
- ., I I I ...." : ,.
.., ..- ", ,- -.;- ,, '. .- I I I 1. J I I I I I "I _;%. :.:. '.
:- RI 'I,:: _,e -,; I ." :" ,:, ": ,,, ., ,. ". I _,, ,. 7 I I',- ; _e- ," ,,, I '', '-:. ,: 1, "I., I I q : -; ,, I I I I ,
, Z,!, ; z .., I I ;,, 1, ,
." ;; 11, '. I : '.... I I '.., : -,.,: I ., ; '. : : ... ... 1. I
. I I I
, "- ,,: ; e- .., ;, ; -, : : .1 I '. I 1, ., I .. I 11 1, I -
'. .; ". I '. I '' ". .11, I I., I I ... I I I .., -, ., I ''. 1, .1. I I I'. I I ". I -_ 11" .1
- ... ,:, i;_ ,, 2 ; I'll
., I -, ". I I : I I :.. : ., ., Z ., I I I ., I "'. I ,
I J, -,. -t.-, 1. ,- ,, -, : I I 1 .:. : I I -1.1 .
- ?-, --, ,' .-."-,7" ; : '' 1. 1. ,, ,: : .. ,
. I I 1-1 I I ''
-,, 7.''., .-. -." : ,:-:. : : ''. 1, ,:- .e., o 1 1 ". : I I 1. t I
I "-, ,,", ,- '' ,, '. .1
" ., I.. r I : ,- I I % .. .
I y, i, '. z ',
11 : 1. .11 I I I ; 11 m, ., I I I I I I 1. : .- : -, ": ;
. -- ..., 11 ", 7 .. ,;:--17 I I I 1, ,. I I -
., '-- ,_ ; l; ': ': I i ,:, ,,, ---.- .. .: I I ,. ':. ., .., ,
. ,, : ,:: ,, ,- I ; .,- .1 '3- I .- I '. _- I I : 1. .. "
' : I I ,
I -, I I I .., 11 I I ,
1, I ,..:; ,:, ,. f- : :ie : '. ,. -"" ."!-" .-, -: ", I I ..i
I :1 I : I, ., j ..... 'I, ,
" ,- I I I I I I I I ", 1 I '- I ,
.." I I I I I I I -. -, .-
:", I- -: ''. 4- , .,, I I I ,
a ,; ; -",. ,, -1 --,,' .-' '-' : -.,, ,:: ,% : :, .- I -1 i
- ; I : ;- ., ,- ", ". -.,., I 11 .-. I _.: I I : I 1 I .1 1, .. ,, 1. .. ,
11 I I -, -- .-, .. I I I I "I I I I I ..
. .-Z, I I '. 1. I I : : I ,, I 11
.. .i"! 1- ", I. "7 : ;" : I : I I I I I I I ', I ': : ; ": i
- ;.,,- ,' ; 1 -. I I % : I I 1 ': 1. 1". :. I ,
" I I n 11. -,. .... ..... ... .1 W11- 1 : : I ., : ; : : -, I I I I I .,. : :, ,
"r. : : I I I : I I i ,; :, -, : ,
I I, ,, ,, I I '' I : ,, -. I-- :, ':
,: ,,, .,,. ,, ,., I : -
-. ; ., ., --.,; ; ", : ::.,
. I ; ,
; j q .. -, ;. :. z I I I I -- -
- I -
I : ,. ..,:" ., I I ,. I I I 11 1 1 ., ,
,- a "- "', ;, I- :,: :,-:,. I I I I I I I I : : :, I I I r -- -- ': : :". -, I.:." I
. I I 1. I ....
,,- ,- ,, ...... : :1 I I I I I I I : I -, 7 i.,
- :,
"I -, :: !, : :, : 1 4 1 ... I I I I '-.. I e I '. ; ,:, I ,: ," ,%, "
I -, .., I I '. i ,t
.. I I ,. ., ;, ..,, ,, ,: ., ., :. 1. .. I "I I 11 I I ,- ,.- ,, ; X.- .. I 11 .,
- ,, .. I I I .. I I ; e J
11 I i il-,"L- 1 : I I I -1 I I I .1 1.
i : ,., : : O ...
'-' I ,. ,. :-- 11 I : : I 1. I I '. .., I I I ,; !:r e ,,, I I ,:, ,
- I ; % I I I : .. I : ,, ,- 11, ,, 1, : "I I .
I '% '... : ll 11 I '. I : ... I I I 1, : I .
I I I I I I .1 I I : I ., I I I ,
. i 7_ I : ,:: : : I I :, ; I A" '-,. ,.,! ,, ; ,, :; : I I I I I .1 .- I ., "I '. .,: .:
, ,;, 1. -- I I I I I 1. : % I -- ,, ., :1 I ;/ ,- .. 11 I I- I 11 .-, I .1 -
. I I ,. I ... I I ; I I I .1 I ., .1 1. .1 : -.-,.!
: ': 'l; : I I I -1 :. ; .. I 1, I I .1 I I I I 11 I I ,., ; ,., ,, .,.., .. :-, ,,, I I I ". I I I I I I I I I I : I I ,, ,i. -J, ..
.1 .1 I I I 1,
- ; ', :,., I I I I I ;_,. ; -
- I 4 .. I ., : .. .... I I I I 1. I '. 'I, i ", ,, : I." I I I 11 '. e, .,
. ,,, : I .1 I I t I I ". I
.. I I I .,! I ,,, ,- I I 4, 1.11 I 10 I .1. I I I '. ., 1. -'... I i
- I I : 4 ', I I ',,J I I li V -1 '. .1 I I I I 1 1, .
. -.,... I I
- ;., -M ,
, ,
1. .,l,',--.:.';: ,- ,'.-% : I'll ,- I I I 1 : e I I I I ". I .1 j,"'m ,
. I ".
l ,. I I 11
:, ;, I / I % I .. '' ,
I III .! ". -"'.' ::, E I .., 0 R ,.T ': I '. S'. "
- I I : H :: .- .' I ,Y /,,' .1 R "- D I.... ,_ ,
-.. : ': .. I -'. .1 ., :-
'.,.-- ,.l; 1, 1. ,,, -ri" :,-'- "
-, I I I I ., ,. !'' :., -,. .,4 1 '
. : I 1-1
I.- ,, r,*.,,:V. 71 f' I I 1: I : ,.-.1 I I ; ,
.7. I I I I 7. 1 I ; I "; ,,, ',
, ', I I I'. I I 1. I I I I I '' c '! -., 4-"
I .1 i I I I I I I : ., I.. I ,,
I I 1, ; 1, ,,, .' 'i, :, :', I .1 .1 I I -, I : ", ,. : : I : .. -- ,
1. .. I .. ,
. !, .- I I :1 I ; ., I :, -, -,' ..."'. ,,, J, ,
, ,. '! ,, : I : I I I I Ili I ,; I I I 2'. "I 1
:- ,7 -. -i ',' ', ,', I i .. .1 ; : % I ., I I ; I ; ,
. I I II.-, ." !. i I I ." .1. I ". 91 I 11 I I I I I : .: I I I I : :1 : I -' ,
, I : .1 : I 11 I 1, .1 I I I I I I .. I .., i.,, '. r I
. I I : I I I I I I I I z I 1, 'i. I -
, .1 I ,, 7-, r I I ,, 11 I I I I -- 1. .. l I i:
- I I I I I I
11 ., I, -i -, ; ': % ,:, .
1 1 ; -.1 .. ,,- l I 1. t .. ,,,- '%. 11 I 1, :. -11 I '
, I,% I : '. '' 11; ,: I I I I I I ":, I -- i "
1 I I gn- I i. I 11 I I ". ,, ; !, -, i't i,
. I ,I- I I I : I I I '. I I I ;, I 1, 11, I 1.
.. ., 11 .1 I I I I I -,,, -,- ',', 1 i. :.." ,
, ; ; .. I I I : I .."1 '. ;1 I : I I I I I : :1, : 1:1 I '. "" ,.- .. ,e
11 .. ,- -, ... % ; : I I ,,, I ;-, I
- I 11, 11 I I I I I ,
- I I 11 I I .. I I I -, ,,, ,,., -'. j 1, ,
' i z -' i ;;- .. ... ". I I I "I :. -.. ., ; 1.. .1 I I I I I I I ., .- ",-. ", -' s :, V .7, ': 14
, i I I I I ;, ,. ., ,
I I ,,, ,, I I I I I .1 1, I il, '' -- .1. 1
1 1, I ll. I j ',' ,, ", I I I I I ; ,I. -
-,T ,T-,, 7" -,. ., ., o 1 1, '. ", I I ... ,. if
! ; ,, I I I .., I ;'I,. ,.. ',
. : :: I ", 1, ,.. .,;
, "'.."'. ,, --%, :,.'.:-. 1, I I I I I I I ; ., : -1 I ,
'i ;.. ,- I I I I I I ,,, -,,
, ',' : : I I :: I I -,,' I -.. l I I :. ,;, w ,
. ; .'': .. '' 0 1 1 'U M e 2 '.... "- .' r- ", ', Ilf- i', ,
,-Y : ,,,- , .- ". ,. I I I I I I I .1 I I -,-, 4 ..., t I -,'j
' ... --- I., I I I I I I '' ,
. I I..." ',-, ", I 1- .-I" 1 I I .- ,.. I .I';,- ,- ," ;"'w-,"" ', '11- .-I I I I I I I -,, ,. ": "
, '' I I 11 ..!, i ,l,"-"-,:-,r".,ll-,
-11-Y-1 '. .. I I .l. I I %: : ,."-.","... ",.",; -,.,; -
,"' ',- -,''', I I ; t, ,,,,--14
11' 'J", ."' .' ', -.' ;- ,.- ,.f, ,-,
, ,-, %: I ", I ., '',. -,
, -.,' I 1 ,7 ,
- ,.":: I ,- ;-,i I' l l I I I .", ." ,
- I iiil ,- 1 1 I I 1 I E- I ., .: ,
;, '' ,-:.. :- 1. 111. 1 1 1 11 ;; .: ,
- -1., :;i ,,: ,,I',,.' :.-. ." :, I .1 1 1 1 '. I I ; ;., ._I ". ._ I'i
,:,4 I 1- 1:, :- I I'll ''I" I 1. : ,c .. -,_1 ", ,,,-,
,, :, ',- I I I I I I ,;,- ,,,-," ,
;V, i I ; :, I I : .. Z -: .. ,', :i,: ,", ,' p
I 11 -I.- 11'' 1'.!A 1 '',I .. '', I I -.11- I 11 .. 1. .- -," I ".
- .. ,, I .,. '" A"
I 11 1 "" ".'''.1 -:_ ., ,. I I t ,
-, s l- :ii'- 4
I- ''., : 1 -- ;, ,,,
,-,! I ; ;
'- :, .,. I r : ,,,
. 1, I,; -,, '' 0 i zc
. Bl' ,';,,',i'
.1 S .-p ..e :.,,c -..,i.:, 1, i,- ,, ,;R
7 I -,-.. '. '.-..--- 1. 11 I 11 I I ;:, ,:",
:: ,A, '- ,-, -: i I 9 I I 1. e -. -,,. ,',-
I I I 3 'I, ; -,",, -, "', ;",
- --,.". -,,, ",-_ -, ." I I .; I I., I 1, : ,-,. I'l .....,. "-- 1, I I I I 't- ,. ,. ,-,, ,-
-, I ,-,, --,,, t- : I .11 : I "I iz- ,'r"",- .', ,, ; 4"-"-
'.- I I .
1.1 !.,, ., i;; ,- ": -, 11 I 1: I '' I ,.".,. "i.- ', """, "'"
. ..''.,.. I '
llj 4" ,-V ,, : ,.,--. I -, ,! i ", -, .. ""'Y4, i. .- ., ,' ,, -
1- 1 ; I I I ,I -
'o --, ,,
-- z I'A.. : ,,, I I ., I ,
I I i:,-'l- i I .1, I : I" I I I I .1 ., ;:,, I' ,,; -,; -, '. -.."- "" I I I I I I I I.. _,, '. ,i '_,
. .,., I I I I : I '' '. ii".-I"-,
" ,., I i I I : I I I I .1 I 1 1 .. I 1, 1 '.I ,
, I.,., I 11 I ': :, ;.:, ,. I ,, e: ,, ": ,-_, ;.
- ;, Iil I i -11, I .-fl I .l I -.
e .. ;, ,,, -,,,-il -' ;, :"'': I I I I ; ,, I-" -
- I 1, 11 I ,.,, ,"-
,. ..., ,PL, I ''I "I .. 1, I.- --',,- l,-, I
- I I ,
-6- J-- I I r,.,.. .',, -
; E. "; ,, -,- ,, ,,,, ;;, ,,,,- I I : : :. I "I I I %- ', "--_.. ..!.', '' ",
, I I : ,- V
1-11 11 -,, --- -, : -, I .,. ... ... ,
. I ,' I'l- ll- "".7 I ': ,: "'. 0- ,.- ", L.
ll-.N ,,:",,,, 'I',, I ":. .. .. .. .. 11 ... I I -', ', ., "".1 : -%". -
1. I I I I ,,', ,, ." ,l 11 11 .I-
11 I Ir '' ;, 2 I I .1 I.. ". ':
,-, 11 I I I ....... I .1 I. % I I :, I '-tl'.'. 1, i.,
, .. _;_ I 11 r I I I I ';,:' 7 -' : I .1 I I I I ., '.. :,, ., I .. I I'll ,
. I ; I
.1 'I., r IT' I I .. : r ''. I I I I I I ', ., .. : -: ,-
'-.'-,. ,, i,- e: ,- ..: ,,- I -, I I J 2
.. i ., ''. : I I I I I _, ,, ,%;. : 1 --; 1"
I i 4 I I '. I : : I? -
, ,, ., '. I ,
-. I I I ; I I -
I ,, ,
1, I I I L I I ""
":,'-'- ,,,, ,,, ,-, e : I I I I :: 1 I I I % I I J -: .,.: ., i .. -
: I : I I '. 1 1 1 1 1 % I I I I I : : ., ,- ':-, ,
". 1: .11 7, -, .,.,,,- -, "-" % :, I I I I I I I : :4. 1, I 1:
I r : ., ,. ., I I I ,
I I I I 'r I I I I I I I I : I I !, I .1 I 1, I 1. I I I I !. ,
r : ,., 1.11 1. 11 I I I I I I I 1. I I ,; 11 I I" .1 11 ., i I,
:.. f 1 ,- : I I I I .
, 4, ,.
! : I I : J,:U N E 7 9 I I I I ,'. i.
" I 'll, "" I .,.,,. ,, ; ..,. : : I I I I I i I I !" : ;', "z I I. ;, ,
I 'I-, "". -!, : .- '' I I ., .. ', : I I ,:,
. I : '. I I 1. ,, I I ,'" .- .. : i,, %: I I I "'. '. 1. I I 1. I., 1. ., ,,.
% rl -( r.. .
, I ., ,,-,, ., '. -, I I -:, %...''
I ,i- ,- ,, ...-"- :Fl, .- .. Pl. ;: I I I I I. I I 11 I ,; .
- 1 : I I I n ,r,.' I I I ; r ,"'- ', -, 7...:, 1. ." -11 I
.. i : --*, ,, .- I I I I l-" .. I I I % 1.11, 1, I I I : .. .. .
. ,. : ,- ,,, ,:, -, .".;:2- I I ::, I .,
, rf 1. '.. :, I I ::,i'., I I I I K .. ..' ', % ,
, I,% :, .:- .-i .,. ''. 11 : '. I :., I '., I -
I I : ., ,.,. I : I 1 :
i .1 .. ,,', I I I I I
.., I I I -. .- I I ; I .. ; '' -. .
,i ,,,,,,'-, ... ., 1: I ,;! ", ,- 11 1, .1 '! -,, .111,
- t, '- ", ,:, r I -'- I .:., 4 -, : : I I Z :-,. -1-
; r ,,'. p 11 : : I I I I I .i-
1. "I : ,, ill ,: j.,,,',,.:"-,", -, I I ", -, "I, l ik I
- I I I I : .1 ,1, .1 1. I I I : I ,,
, '.."", r, "', I z ;." ,,,:, ,, : .
. .1 ,
, .. '; 111. "I',- ,." I I I .1 I I I -, 1 '. 7,: %,
" ; I'll I I ,,, ". .. .,
-. ; ,-,. ", ", I 1',- I 1 I ,- '; ., !, I : I 11 4 A I T ., .; :
,, I I- 41 711 ,, ,: .. ., .11 I I ... I I I ." .1 :-:.W- : I : ;,._, .1 ;.-
. .111 J .. I 1, .: -: -."- ) ... .. I 1, ; ;. : I I ; 11 I I ; : .1 I I I I I I : 'I, '. I :
... -"-,- ,-" ,, I 1. ,
. ,,,,r,:,,r,,- l, I I ,, -.' 4'.
.. .. I ,I T, ,. I .. ,. I : ; I
, ', -,-;', I" '1 7 1. ,'. I I .%: '.', ,. I .1 : ,% : 1 1 ;:
:
,l ;
I ,.I-
11 ,- .,"V -, '. I I .. L I I I :,
.1 I I I ,:: ,;: ""I, ,X, :, ., : -'.,, ,, .,, -, I I 1, I I I I ,,, ,,,
'' I 1 1 11:1 ,;:.- 1 4 I 1, -.- -',Vc. ll h, i ; ,4 :.- I 11 '. I : : .;, i -,I
. I : ,f ,; '..,
V "";! ., 11, :- .,, ,.-. I I I I I 1 ,
. I 11 A iAl ,r,
, ;1- I ., I., _4 I : ,,,
. I I I, q ':tP ,i % ,. I .. I I .i .: J, I : I I I I : % r -
- : 2 : ", ""..,I ", I ; 11 I : -..
'. .. ,, Y &- -- """-, : : .;!;" :- ; ,: I I.. : ; ; .,. I I I I I.. -
,. I I -,; ; :,- :.; : .: I 1, : ,; I '. :. 6 '. I I I'll I w il I ,% I I I ::% ,
, -, j I i 1 ", :, ., -%, .1. I ,- I ., I I I I ', ; .,
,,,, -1 Oil r ,,. ,, -i ., -- z: z :, '. 1, h I I I 1, I I I 11 I I I ... -
, ,;!, ,.. % ,, I I ., I I I" 1,
, ,,, '. .. ; I ; I I
I : ; ;: : : ., '' 1, Z I ,, I. : : 1. '. 1,, -.. i I I ,
r : :7 I --j I IL 1.1 ,
.. IE 4 1, ,,, f I e .1 I;.. '..: ,* -
" Y; ; :, ,,, ", I .. %; -. I ., I .1 Z ,, ', ,';,, I I ., 1. ,.", ., '. :.
I- i ..... j i'' 'j : n :, .-.- I .1 .. I 1. .
r I 4 't 'j ; ia : ;,: e. ... 4
.. I I I I I ": :: .:.;i, ;, ': .: r .. : -,- ,. I ., ; .. .t ., 1, .. ,,- I I r ".." .. ., :",e I I -, I I ': I 1. I 1. .,.''. ; ,,, j'., I "! I .
1". ,,, .. I ;: .
." : I I c .1 1. I I I ., ,. r,_.-, !". I I ". i .1 ..., I '
. I.. I i I ; ,. ; I I .!
"I ; I I I % 0 ,. ..
1, 11 ,: I I 6 '. ; -; I I ..:, I I .-,.% "- I I I -
4 1'1 I" I I ..,: : .. 1, .. .., I I I .., ) : I I I i. 1., I : -c
1, ; '. : e ''. e:- ;, I ,..' ., 7 ., : I ., I
, I I t .v : : %, :-,: : I ,Z%
,',,,-,, ,"' I ., I .'' -1 .., 1, I I ,: ",% z
I :, ,, ,. :. ;, ,-, :., : ,:, ', .1 I 1. 7-
,. el I j : ; .. ,:: I 1 .G l ., I I .1 .. I I I ; ,:,:. ;. I
'. I *- .; -- ,, I : ; : :: I I I I :1 I I I I ,
, 11, .1 I 11 1. I I v ,. .; :t ... -, : ': : I ":
. 1, I I ,; ; Y,:, ,. I I 1, 1, ,. I : : : ; I I 11 : i '. .. -.
11 11 ." t J, J, ," I I I % ,,. ,- '. I I I ., ', 1. V :. s :- I ,. I '' .- 1.
1, ,. -0 -11 -, I 11 ; .... .. I '. I I I", e '' ., % I I I., I ,. I I .:
- .. I I I lv J I !,
., ,,,, r. ., I.. J ,." .. I .. I 1. .1 "" .. I I I I i I 11, I "
' I ; ; :: : '' ": L I I A : : :, I I z I ;, -,
Z I., ,, -T, ., I 11 11 I., I I I .11 I I .
, i : ', I N I ,
,- Ij,' I I z I .: I '.. .
., ; C. : I I I I I .t
, I I k I I I L 7 ; ., I 1. I : I I -1.,:. ., .: 1. I I I I : I ,
. .... , J, i,, ., -" F.4. ", t I t I ..
. 1, r I I':- I I ^ .. I ( .. : ; %
- ., I ll -I. -: : I %.
, I ,, I :, L, I .. ". I I I I I I I '.
" I :- .,.;, :_ : :.. : I ". I I I I I .. % j
, I "i I e :.
- ... I : : :i: 1, ". ', ::.-, ,%,. : ,, !, I .
... ,,,, I % : .!.. ". I I I- -: : : ": % :, .- -:
r. ", I I ", ,, ..
.e.. I L 'r ;; ., I ;, ,!::, :i. ,
, L : '.I I.. : 1. I I
1. I .. I ,! "I .. i ': '. %-, ,. : ; : : :, I:z, :: I I I : I I
, : : ; ; I ; I I I
, ,, ", I : ..,,, I
. .., I I '. I I







VO0R Y

VolI

um e

lo g

calI

A sp ect

J U NE '79

Parker

TH E

TR AD E

Bio




147
VOLUME 2 (1) IVORY THE POPULATION INDEX
The following comments on tusks start with a synopsis of a
paper in preparation by Laws and Parker. As this will be
published shortly with data and detail, I shall forego repetition
here. Suffice it that the material was obtained from the tusks
of 2,900 elephant culled in Uganda, Kenya and Tanzania during
the years 1965-1969. Information on this culling and analysis
of other relevant data obtained have been published (e.g. Laws
1966, 1967a,b, 1969a,b, Laws and Parker 1968, Laws, Parker and
Johnstone 1975, Parker and Archer 1970) and further papers will
be forthcoming over the next decade (e.g. Laws Development,
growth and wear of mandibular teeth in the African elephant
Loxodonta africana is already written).
Elephant tusks are modified incisors which, in the African
species, are present in both males and females. A 'milk' tusk
is present in the late-term foetus and at birth which seems to
be-resorbed internally and never features as an externally
visible tooth. The permanent tusk which replaces it erupts at
an average age of 16 months, but may not be visible beyond the
lip until several months later. The age at tusk erupt4on is
variable, in *some cases it may not take place until the'
animal's third year. Tusks grow continuously throughout life,
albeit at declining rates from the middle of the 3rd decade,
in the form of a logarithmic spiral or helix. Both tusks and
molars exhibit pronounced 'layering' in their structure, which
correspond to seasonal changes in the environment, and which
(presumably) are related to nutrition. These act as an indicator
of growth rates. In the main they confirm the validity of
Laws' (1966) earlier age criteria.
Aspects of tusk growth are illustrated in Figures 32-35
in which total tusk length, length of that portion lodged in
the head, circumference at the lip and weight are respectively
plotted against age. Male and female tusks show markedly
different growth patterns, though with both, size increases




250-
200- .
L
E .
N
G
N .. ; ." .
G I ..
T 150- ,
H ; .
'' ..li
N
100-.
Black = Male
50- Orange = Female
ib 2b 36 4b 5b 6(
AGE IN YEARS
FIG.32 SCATTERS RELATING MALE AND FEMALE TUSK LENGTH TO AGE
(n ; males 705; females 827)




90-
80-
70-
60-
L
E
N
G..
T50-
H : .-
N ..
40-
C
M
S
30-
20-
10- Black = Male
Orange = Female
0
1'0 0 3'0 o 0o 0
AGE IN YEARS
FIG. 33 Scatters relating length of male and female
tusks in the atreolus
(n = male 147; female 170)




10 20 30 40 50
AGE IN YEARS

55-
50-

--
;
r
t;
~ -~ r
Ir 1'
i
i
~ e '*.
rl ~' i
* a

40-
30-
20-
10-

D
B
BI
~

Black = Male
Orange = Female




W--
E
H
T 40-
N
30-. ,"'
K :
G" ,
20- Black= Male: -
Orange Female
7 7
Is p
:'t__ -'
010 20 30 40 50
AGE IN YEARS
Fig. 35 Scatters relating male and female tusk weights to age
(N = male 1862, female 2733)




148

with age until the last decade of life when in some individuals
at least wear and breakage exceeds replacement through growth.
The incremental layers have permitted calculation of
potential cumulative increment in length and weight. These are
555 cm (c.18 'ft) and 80.8 kg (178 lbs) and 521 cm (c.17 ft) and
21.7 kg (47.8 lbs) at 60 years in males and females respectively.
Actual lengths and weights at 60 years average 250 cm (c.8 ft)
and 61.0 kg (134 lbs) in the male and 160 cm (c.5 ft) and
7.5 kg (16.5lbs) in the female. Thus male elephant in East
Africa attain only 75.5% of their tusk weight potential, and
females only 34.61. The loss is caused by breakage and wear
throughout the elephants' lives. The differential loss between
the sexes is almost certainly the consequence of the female tusk
being a smaller tool, relative to the size of the animal. The
potential weight of the male tusk (80.8 kg) is 1.5% of the
male's potential total body weight at 60 (5314 kg Laws et at.
1975), whereas the potential of the female tusk is 0.7% of the
female potential total body weight (2986 kg Laws et at. .1975).
While the mechanics can no doubt be described elegantly in
engineering terms, the female can at all times exert a
proportionately greater stress (body weight : potential tusk
weight) than males upon tusks of similar lengths (greater weight
on a thinner lever).
The differences between tusks of the sexes that are so
marked in Figs. 32-35 are easily discernible at a glance, for
the tusks also differ in overall shape. Almost all male tusks
within the normal range of female tusk weights are pronouncedly
conical from the base. Even if the tooth beyond the lip is
broken, the taper is still noticeable in that section of the
tusk that was within the elephant's head (the alveolus).
Female tusks appear cylindrical their taper in the first half
of the tusk being too gradual to perceive easily. Male tusks
only become cylindrical when well beyond the limits of female
growth. The relationship of circumference, length, weight and
shape permit male tusks to be separated from females' with ease:




149

a fact of which both hunter and trader have been aware for
centuries.
The dynamics of ele phant tusk growth are such that given
tusks' weights alone, it is possible to determine those which
are unlikely'to be female, i.e. all over 7.5 kg. Of those over
7.5 kg it is possible to ascribe ages to the bearing elephants
at the time of death. If, in addition to the weight, tusks can
be seen, their shape and weight at any size permit both sexing
and aging of the elephants which bore them. Similarly, on
their own, total lengths, circumferences and or the length of
tusk in the alveolus, are all useful indices of age. With
shape or in combination with any one other of the parameters
illustrated, they will determine sex and age. Precision
increases if combinations are made with two or more of the other
parameters. A fifth measurement on which Laws and Parker will
publish data in due course is the volume of the tooth pulp
('hollow' to the trade and 'nerve' to the hunter). Up to 35 years
of age this increases in volume up to 5.85 litres in the male,
and 0.82 litres in the female tusk. After this it regresses
down to 0.3 litres in very old females, being replaced with
ivory. The degree of male regression is not yet confirmed in
view of limited records. The replacement of pulp with ivory
must partially compensate for any slowing down in incremental
length growth, and for external wear and breakage.
The salient point is that tusk measurement can provide an
objective record of human/elephant interactions. If data on
weight, lengths and circumferences and can be collected they
will impart a great deal about the manner in'which hunting is
affecting populations, and also considerable detail on elephant
status.
The data from which our descriptions derive are exclusively
East African. It has yet to be shown that they are valid for
other African regions. Subjectively, I can recognise 4 types
of ivory: East African, South-Western, Sahelian and Cyclotiform.




150

Throughout this survey, when seeing ivory in Europe or Hong
Kong, I estimated its origin before asking the owner from whence
he obtained it. In all cases I ascribed it to the correct
region. Experienced traders are also able to make similar
distinctions as to ivory's origin, particularly when it is in
large lots so that common features stand out.
East African ivory occurs the whole length of the Indian
ocean seaboard and inland to include Somalia, southern Ethiopia,
southern Sudan east of the Nile, Uganda east of the western
rift, Kenya, Tanzania, eastern Zambia, Malawi (with reservations) ,
eastern Rhodesia, Mozambique and the Kruger Park of South Africa.
In trade terms this region produces East African "Soft" or
"white" ivory. In older animals it is long, and the tusks
generally bow and curve in several planes: it is this ivory
which grows in a noticeably logarithmic helix. The twist on a
large male tusk is very obvious.
South-Western ivory is from southern Angola, western Zambia,
Namibia, Botswana and western Rhodesia. In trade terms this
is South African "Soft" or "white" (and not so highly prized as
East African white. Subjectively I believe these elephants to
be substantially bigger than the East African form. However
their tusks seem much shorter, with a bigger proportion in the
alveolus, and seldom attain the same weight. It is far more
heavily sharedd" than most East African ivory. Their bluntness
and lack of length suggest very much higher breakage rates.
South-Western tusks seldom attain sufficient length to show
curvature in more than one.plane which, however, is Very
pronounced. The absence of an obvious twist in the grain
suggests that these tusks grow more in a logarithmic spiral
within one plane and less in helical form.
As the name implies, Sahelian ivory derives from.the sub-
Saharan zone from Senegal to the Nile. It is more similar to the
South-Western, than East African, but it does get much larger
that the former. Again there appears to be a very much higher




151

breakage rate than in East Africa particularly 'in the female.
A common phenomena in Sahelian ivory is a very fine 'herring-
bone' shake, which is so dense as to impart a grey appearance to
the tusk. This does occur in South-Wester n tusks, but not to the
same degree.
The East African, South-Western and Sahelian elephants are
all "bush" or "savanna" types, i.e. Loxodonta a. africana. The
fourth is the forest elephant of the equatorial forests in the
Zaire basin the West Africa: Loxodonta a. cyclotis. It is the
least known of all the African elephant forms. Suffice it
that it is much smaller than any of the savanna elephants. The
tusks are equally distinct, being slender (particularly in the
female), very straight with no twist; if there is curvature
it is slight and in one plane only. The proportion of a
cyclotis tusk in the alveolus is far less than an africana tusk
of similar weight (Fig. 36). In both sexes the tusks point
downwards rather than forwards as in the case of africana.
Further, the male tusks show far less taper than in the savanna
forms, although it is still quite discernible. Cyclotis tusks
have very few "shakes", are often black or deep mahogany in
colour (presumably from-vegetable staining) and where use
removes the staining, show a characteristic yellow patina. To
the trade they are "hard" or "yellow" ivory.
No clear-cut division separates the small forest cyclotis
from, savanna forms of elephant. Rather, all around the periphery
of the rain forests there are clines in which elephants are
intermediate, showing features of both forest and bush types.
It is in this "hybrid" zone that large individuals occur,
carrying cyclotiform ivory in all aspects but its large size.
Many of the biggest tusks I have seen, have been straight,
without shakes, and with the characteristic black colouration
of the forests, or forest edges. Traders refer to it as
"semi-hard" or "semi-soft".




~B
TRACINGS FROM A PHOTOGRAPH TO ILLUSTRATE THE
DIFFERENCE IN SHAPE AND DEPTH OF TUSK IN THE
ALVEOLUS BETWEEN'TWO 20 KG MALE TUSKS
A. FROM Loxodonda africana africana AND
B. FROM L. a. cyclotis. THE UNSHADED AREA
IS THAT WITHIN THE ALVEOLUS.

FIG. 36




152

Figure 37 illustrates the distribution of the four gross
geographical forms of ivory. It must be stressed that they are
based on broad generalisation. The characteristics are more
apparent when a number of tusks from one region are assembled
in proximity to a group from another. The individual tusks can
show wide variation. An example of intra-family similarities
and inter-family differences is given in Figure 38 from two
herds taken within the same East African population.
Objective data to support the contention of four ivory
regions are few. Elder (1970) examined some aspects of tusk
morphometry from South-Western, East African and Cyclotiform
ivory. He found no difference in the specific gravity of the
three forms either between sexes or between regions. He did
comment on the greater depth of tusk in the alveolus of the
South-Western elephant which suggests a bigger animal. He
mentioned a greater circumference at lip than in either East
African or Cyclotiform tusks, but this was not allied to age.
If his Botswana sample derived from older elephants, the
apparently greater circumference would be explained. I have
seen a great deal of Botswana ivory and, albeit subjectively,
have never gained the impression that it was noticeably thicker
than East African tusks. Extensive data have been collected
on western Rhodesia ivory which were not available to this study
and in due course their analysis will settle the matter.
The records of William Finaughty who hunted the South-
Western elephants extensively between 1864 and 1875, indicate
the heaviest tusks he got were two which weighed 56.7 kg each.
While these are big, they are not outstanding by East African
standards, but they were obviously exceptional to him.
Stevenson-Hamilton (1947), an accurate observer, wrote
"I am inclined to think that South African ivory can
never have equalled that carried by elephants in
equatorial Africa, where even to-day tusks of well
over 100 lbs (45 kg) are not uncommon."




FIG. 37

DIAGRAMMATIC ILLUSTRATION OF THE DISTRIBUTION
OF FOUR IVORY TYPES-IN AFRICA.




FIG. 38 TUSKS FROM TWO FAMILY UNITS IN THE SAME EAST AFRICAN
POPULATION TO ILLUSTRATE (i) INTRA-FAMILY SIMILARITY
AND (ii) INTER-FAMILY DIFFERENCES.
(Photos : R.M. Laws)




153
On such as this and my own observations I am inclined to
believe that weight in the South-Western elephant tusk is
lower at age than in East Africa, and that if we apply criteria
developed from the Laws and Parker material, we would under-
rather than over-age these elephants.
Sherry (1978) indicated asymptotic weights of 5.6 kg and
6.9 kg for female tusks at 60 years of age in two south-east
Rhodesian elephant populations. These are substantially less
than the 7.5 kg we have postulated from East African data.
However, we have evidence of differential breakage and wear
between out 5 East African populations, and these low
Rhodesian estimates may be rather more extreme expressions of
the same phenomenon.
I have many reservations about applying East African tusk
criteria on a pan-African scale. However they are the only
yardsticks available to this survey. In the following chapters
I shall apply them to material from the ivory trade with the
caution that the results may need some re-adjustment (particularly
those relating to Cyclotiform ivory) in the light of future,
more precise data.







154

VOLUME 2 (2) IVORY THE STANDING CROP
An elephant population's standing crop of ivory (the total
weight of ivory carried) is obviously the product of its age
and sex structure. Such structures vary widely, and appreciation
of this is necessary before ivory entering the trade can be
analysed. I
The elephant culling recorded by Laws et al. 1975 took
random samples that were representative of population age and
sex structure. Five populations 2 in Uganda, 1 in Kenya and
2 in Tanzania were sampled in the manner described. Below
I present 3 examples : 1) from Uganda (MFPS) which is
representative of an 'old' population, 2) from Tanzania (MKE)
representative of a 'young' unprotected population and 3) being
the sum of 1 and 2 and the 3 other populations intermediate
between them to give a Igeneralised' population.
The age structures of the MFPS population are presented
as 5 year classes in Figure 39 and Table 162, males and females
separately and combined. The sex ratio of the sample was 42:58
males to females. In the same figure and table are the
distributions of ivory weight in 5 year age classes to produce
'ivory shadows' of male, female and combined sectors of the
population. Of the gross ivory crop, 64.8% was held by males
and 35.2% by females an exaggerated reversal of the sex ratio.
The average male tusk was 10.08 kg and 68.7% of male ivory was in
animals of 26 years or older. The average female tusk weight
was 3.94 kg and 72.7% of female ivory was in elephants of 26
years or older. The.population's average tusk weight was 6.51 kg
and 70% of its ivory held by animals of 26 years or older.
The MFPS population had a very low recruitment rate
(insufficient to maintain.the population) with correspondingly
few young animals. Its range at the time of the culling had
been a sanctuary for 52 years which reduced human inflicted
mortality and permitted many individuals to approach their




SSO-
0
o
40-
4 .
o
0
co
10
-0

a 6-10 1 11-151 16-20 1 21-2!, 26-30 '31-35 136-40
AGE CLASSES

A el-Ca I 4-Ou I JI S u I lES
AGE CLASSES S

050 -
m40 -
o 1
0-
4 0
30-
0
lo
14&-50 ~2 15-5-5--I
so-
fd 40-
-p
0
4 10
4 30-
0
P,
D
+I
G

40-
rd
0
O
'44
.0
u 10-
r .05 6-10 I11"-15

S 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45
AGE CLASSE 8

A- E CLA-c Ou
A GE C L AS 9 E S

50-
S40-
41
0
O
130-
44
o
4 20-
0 10-
13 .
,

0-5 6-10 11-15 16-20 21-25 26-30 31-35 136-40 41-45 46-50

fmmum

' 21-25 25-30 31-35 1 35-40 '
AGE CLA 8 ES

CLA SES

AGE

FIG. 39

AN ILLUSTRATION OF THE DISTRIBUTION OF AGES IN 5 YR CLASSES AND THEIR CORRESPONDING
PROPORTIONS OF THE STANDING IVORY CROP IN AN OLD ELEPHANT POPULATION, MFPS.
A = MALES. B = MALE IVORY. C = FEMALES. D = FEMALE IVORY. E-= SEXES COMBINED.
F = IVORY SEXES COMBINED.




155

maximum potential longevity. These factors produced a very
'old' population that is unlikely to have many (if any)
counterparts elsewhere in Africa.
The age structures of the MKE population are presented in
Fig. 40 and Table 163, males and females both separately and
combined. The sex ratio of the sample was 49:51 males to
females. The corresponding ivory shadows for both sexes
separately and combined are also illustrated. Of the gross
standing crop of ivory, 59.9% was held by males and 40.1% by
females. The average male tusk weight was 4.63 kg and 25.7%
of male ivory was held in animals of 26 years or over. The'
average female tusk weight was 2.99 kg and 47.2% of female ivory
held in elephants of 26 years or more.. The population's
average tusk weight was 3.81 kg and only 34.4% of the population's
ivory was held in animals of 26 years or over.
The MKE population was characterised by very few old
elephants none at all in the last decade of life and only
2.6% over 41 years (cf. 11.8% in MFPS). Constant hunting and
harassment were the cause of this (Parker and Archer 1970) and
mortality was such that only 5.4% of males were 26 or older
and none reached 41. Recruitment was high, producing a
correspondingly large segment of-immatures. As such MKE was
a particularly 'young' population and it is doubtful whether
there would be younger groups elsewhere, other than in artificial
circumstances (e.g. as in the Akagera National Park of Rwanda
where the whole population is of translocated immatures.
The combined age structures and ivory shadows of all 5
East African samples are presented in Figure 41 and Table 164.
The sex ratio in this sample is 45:55 males to females. Of the
standing crop of ivory 63.4% was held by males and 36.6% by
females. The average male tusk weighed 7.26 kq and 55.1% of
male ivory was in elephants of 26 years or older. The average
female tusk weighed 3.36 kg with 65.4% being held in animals
of 26 years or more. The population tusk averaged 5.09 kg and




0-5 6-10 11-15 16-20 21-25 26-30 131-35 1364

41-45 14&50 1 51-551 56-0 I

0-5 1 6-10 1 11-15 16-20 1 21-25 26-30 131-35 13640 41-45 1 46-501 51-5 60

r

50.
r.
,-4
t40.
41
0
4u
30-
o
0)
S20-
u 10.

-
-
-

G-51515 56-60

050
,A
fO40
0~a
30 -
tw
o
320 -
S5 C10 11-15 120 21-25 263013135 3640 41-45 41051-55 6-6
0-5 6-0 11512022523 313 360414 4605155

CLASSES

CLASSES

AGE

AGE

so-

Xo --

D

0-5

46,50 51-5 IS-WO

CLASSES

AGE

I 21-25 2-30 1 31-35 I 36-40 41-
AGE CLASSES

Ss50 -
o
,4
40-
4J
30 -
41 30-
4
O
o
S20-
ul-
'1

u

4
O
40
0
o
30
20
o,

0-5 6-10 11-18 16-20 21-25 26-30 31-35 36-40 41-45 1-55 560

CLASSES

AGE

CLASSES

AGE

FIG. 40 AN ILLUSTRATION OF THE DISTRIBUTION OF AGES IN 5 YEAR CLASSES AND THEIR CORRESPONDING
PROPORTIONS OF THE STANDING IVORY CROP IN A YOUNG ELEPHANT POPULATION, MKE. A = MALES.
B = MALE IVORY. C = FEMALES. D = FEMALE IVORY. E = SEXES COMBINED. F = IVORY SEXES COMBINED.




31-35 36-40 141-45 I46-50 51-55I 56-01
CLASSES

O 21-25 126-30 31-35 36-40 1 41-45 46-60 51-55 6
AGE CLASSES

50-
C 40-
30-
0
i,
I
i20
10 -
50-
o30-
41
0
o
40-
O3
0 0

' 31-35I35-40 141-45 146
CLASSES

AGE

AGE

50-
.3
40-
rd
4J
o 30-
.0
g 20-
u
10-

0-5 6-10 1 11-15 16-20 21-25 1 26-30 31-35 136-40 141-45 46-501 51-551 56-
AGE CLASSES

T56-6

0-5 r1m-lm riml

1 21-25 26-30 31-35 I b-4au
AGE CLASSES

S40.-
030-
42
4 20-
V-

AGE CLASSES

FIG. 41.

AN ILLUSTRATION OF THE DISTRIBUTION OF AGES IN 5 YEAR CLASSES AND THEIR CORRESPONDING
PROPORTIONS OF THE STANDING IVORY CROP IN A GENERAL POPULATION, COMBINED EAST AFRICAN
DATA FROM 5 POPULATIONS. A = MALES. B = MALE IVORY. C = FEMALES. D = FEMALE IVORY.
E = SEXES COMBINED. F = IVORY SEXES COMBINED.

40
- 40-
4'~ 30-
o
30
,J
10
4)
Ak i




156
58.9% of its ivory is held in animals of 26 or older. The
distribution of ivory within the population is illustrated in
Figure 42 and Table 165.
The pattern presented in Figure 41 is the standard against
which I shall compare data obtained from the ivory trade in
subsequent chapters.




56-60
51-55

46-50
E41-45
36-40
31-35
S26-30
21-25
O16-20
11-15
6-10
0-5

[0% 10%
63.3% -

10% 20%
36.7%

FIG. 42 DISTRIBUTION OF ALL IVORY IN AGE CLASSES
FOR 5 EAST AFRICAN POPULATIONS COMBINED.







157

VOLUME 2 (3) HUNTING EFFECTS
Elephants are killed for a variety of reasons which involve
different techniques and have different influences upon
populations.
Perhaps the most widespread reason for hunting elephants is
still for meat. Even if tuskless they.would still be taken.
Thus in Malawi (Bell 1979), Zambia, Mozambique, Zaire and parts
of the Sudan, there are usually meat smoking racks and other
evidence of their use as food, in close proximity to carcasses of
hunted elephants. Even in Angola where ivory buys ammunition,
elephants are equally important for the meat that they provide
for guerilla troops (N. Steyn, pers. comm.).
In most subsistence situations there is insufficient
manpower available to make optimum use of all the meat an
elephant provides. Cutting it up for smoking and drying is time
consuming and laborious. Even an adolescent specimen provides
a surplus, and only animals killed relatively close to large
settlements can be fully utilised. Thus with subsistence
hunting for meat there is little incentive to select particularly
big elephants. If there is a bias in what is hunted, it
probably arises more from the elephant's social organisation
than any other factor. Females and young in family units show
considerable cohesion and are more dangerous to the lone hunter
with a primitive weapon. Lone males, in contrast, are among the
easiest to approach ofall game.
However while this bias will possibly result in more adult
males being selected and hunted when using spears, bows and
arrows, muskets and rifles, it would not apply where the more
economic and less dangerous techniques of trapping are employed.
Indeed here the bias would work the other way, for it will be the
weaker, less experienced young individuals who would be more
likely to succumb to a drop spear wound or less able to break
snares. overall, the effect of subsistence hunting is likely




158

to affect all ages and sexes of the population. To my knowledge
it has not been seriously suggested as a cause of elephant
extermination.
At an elementary level it is difficult to separate
subsistence from commercial hunting. For centuries the Watta of
eastern Kenya have taken elephants for both meat and ivory.
Ivory was the key to their social bond with the Galla
pastoralists, who provided protection in a somewhat overlord/
serf relationship. At this level they were in truth subsistence
ivory traders and in having a special use for ivory, were
appreciative of big tusks. A great deal of their hunting,
however, was opportunist: lying in wait at a waterhole, they
would ambush any elephants which came to drink. The volley of
arrows loosed would be aimed at the nearest to the bowmen and
this technique would probably have swamped the better hunters'
efforts to be selective of large male elephant.
As the 20th century unfolded, so pressures to enter the
monetary economy will have influenced the traditional
Subsistence' ivory hunter. Killing for meat will have receded
in importance and for ivory have made corresponding gains. With
the decline in the value of ivory relative to rising commodity
prices (Volume 1 Chapter 7) the need to kill more for the same
return would have been augmented by human population growth and
further demand upon elephants. Selectivity will almost certainly
have declined. Overall the commercial hunting will have reduced
life expectancies of elephants, the degree depending on hunting
intensity. The direction will have been toward an MKE population
of low age and high fecundity. The most extreme expression will
have been that of the-Somalis in Kenya in the present decade.
'Control' shooting is a widely used misnomer applied to the
destruction of elephants in defense of human interests. At its
more simple level it entails shooting those which venture out of
sanctuaries and damage crops (usually males in their late teens
and twenties). This is the most common form of control where




159

there is a stable in area sanctuary surrounded by human
settlement.
A different complexion is taken on by control work where
humans occur as islands in a sea of elephants. *In view of
human population increase, politics and economics the boundaries
of these islands are seldom constant. In this century they
have, for the most part, been expanding, but on occasion have
contracted e.g. with the 'villagisation' of peasants in
Tanzania (Rodgers, Lobo and Mupunda, 1978). However, whether
expanding or contracting, there is almost constant elephant
,attrition about their edges.- Thus in south-eastern Tanzania
over 100,000 elephants have been shot since 1930, and c.2,500
are still shot annually (Rodgers et at. 1978). Solutions have
ranged from a limited policy of shooting one or two members of a
herd in the vicinity of cultivation to 'teach' the survivors to
leave it alone, to grandiose quasi-military strategies to drive
the elephants 'back' and to establish frontiers which they
should not cross. ,
"In other words a system of defensive fronts has
been instituted to ward off the encroachments of
elephants" (Anon 1925).
This involved shooting virtually every elephant on the wrong
side of the 'frontiers'. At the same time there was a persistent
desire to leave big tuskers for sportsmen.
"Rangers were told to spare all bulls with warrantable
tusks" (Anon 1926).
The records of the Uganda Game Department from 1925-1959
provide the most cohesive documentation of the effect intense
and prolonged human activity had upon elephants. A synopsis of
the relevant data is presented in Table 166. It includes the
yearly number of elephants shot on 'control' (males and females);
their average tusk weight; the number of elephants shot on licence
and their average tusk weight; the number of tusks found both
in the battle grounds and elsewhere, with their average weights;
the total weight of ivory produced and overall Uganda average
tusk weight; the contribution tusks of less than 4.5 kg made to




160

the overall number of tusks; the weight of ivory claimed by the
Game Department to have been produced compared with Customs
records of export; and lastly a comparison of the number of rifle
permits and game licences issued over a 9 year period.
To gain-confidence in the record let us first compare Game
Department stated production with Customs records of export for
the 19 years 1939-1957. They should marry approximately as
Uganda had no internal ivory uses of consequence. The Game
Department 'produced' 434.5 tons when Customs claimed exports of
434.9 tons: an insignificant difference of 0.09%! With this
backing it is reasonable to accept Game Department volumes of
ivory as accurate.
The first set of columns in Table 166 concerns elephants
killed on control. These rose from an average of 705 a year
1925-1929, to a peak of 1351 a year 1935-1939. From this high
point the 5 year average declined to 667 annually between 1950
and 1954. From then it rose steeply back past 1027 a year in
1955-1959 and, though there is a gap in the record as the Game
Department stopped publishing reliable data after 19591 the
trend proceeded to the large-scale cropping described by Laws
et al. 1975 and culminated in the wholesale slaughter of
survivors under the dictatorship of Idi Amin.
It was all very well to decree from behind a desk that
control shooting should concentrate on 'unwarrantable bulls'.
It was quite another matter putting it into practice. Shooting
elephants in dense Pennisetum c.4 metres (12 ft) high was
unhealthy for the hunter. Grass this tall hides elephants as
effectively as a stand of wheat conceals rats. Invariably they
remain invisible until at very close quarters. Much shooting
was also done in thick forest which was equally uncomfortable.
It was extremely difficult to determine the sharp from the blunt
end of an elephant, let alone whether it was an unwarrantable
male'. The de facto system was shoot what you can. Thus control
kills were virtually random except that small immature tended




161

to be under-represented. They were more concealed by the
vegetation and bodies of other elephants.
In the circumstances, the offtake should have been similar
to a cross section as illustrated in the preceding chapter
- minus most'animals under c.5 years of age. Consequently the
expected tusk weight should have been slightly above the mean
population weight of 5.09 kg (Table 164). At first glance this
is indeed the case average control tusk weight remained
remarkably constant between 5.14 and 7.61 kg over the 35 years.
However this constancy is at complete variance with the sex
ratios given. The average tusk weights are reasonable given an
expected 45:55 male to female sex ratio, minus the youngest
classes of tusk bearers. They are not reasonable if the sex
ratio is reversed and exaggerated to an average of 65:35 males
to females as in the case in Table 166. Further examination
shows that the claimed sex ratios varied very widely between
52:48 and 84:16 always in favour of the male. Such variation
clashes with the constancy of average tusk weights. Either the
sex ratios or the average tusk weights are wrong. The manner
of control hunting, the known range of sex ratios in elephant
populations, and the independent agreement with Customs
statistics indicate that the weights are right and the sex ratios
wrong. From my own experience I know that many game scouts and
not a few game wardens cannot sex an elephant accurately. The
myth of the dominant "herd bull" dies hard and family unit
matriarchs, which are often substantially bigger than other
members are claimed as "lords" of the herds. This misidentification
of sex has also been reported from Tanzania (Nicholson pers. comm.
and Rodgers et at. 1978). Added to these genuine misidentifi-
cations, was a desire to claim males shot. In those days it was
the epitome of bad game-keeping to shoot females!
The constancy of average weights in animals shot on control
run somewhat contrary to expectations. Theoretically continuous
harrying should have reduced life expectancy towards an MKE
level (Table 163); 35 years of control should surely have




162

produced a declining tusk weight? In fact the reverse is true,
for it only rose to over 7 kg at the end of the period.
The elephant shot on licence bear out the expectation of a
declining tusk size. This was the subject of concern in the
Game Department shortly after the second World War (Pitman pers.
comm.). The records were analysed by Brooks and Buss (1962)
who documented that there had been a decline in tusk weight.
This was inversely related to the number of elephants taken on
licence: the more elephants taken, the lower the average weight
of tusk. The assumption is that the hunting demand exceeded the
population's supply of 'warrantable' bulls. (Note : at the time
it was not appreciated that control shooting the 'unwarrantablel
males today, removed the pFospect of 'warrantable' males
tomorrow.) The assumption of over-hunting was tackled in 1952
by raising the price of elephant licences. This produced a sharp
drop in the number bought and, as expected, a rise in the average
weight of tusk taken on licence. However, even though the number
of male elephant taken on licence dropped to below the 1938, 39,
40 and 41 figures, the average tusk weight taken did not return
to the previous level. From this one might deduce that the
consequence of all forms of hunting had so reduced male life
expectancy that the average tusk weights of the 1930s were no
longer attainable: the population had become 'younger'.
The average weights of 'found' tusks were slightly higher
than those obtained on control. However they show no obvious
trends. The combined Uganda average tusk weights are, from the
contribution of sport hunters, higher than those of either control
or found tusks. In view of this contribution, and the constancy
of control weights It would have seemed reasonable if the overall
tusk average had declined in sympathy with the fall in
sportsmen's tusk size. In fact this didn't happen. Overall
average weights increased. This could have come about if the
contribution of the lightest tusk classes declined at a greater
rate than the contribution of sportsmen's results. This in fact
occurred. In the mi d 1930s tusks of less than 4.5 kg constituted




163

over 40% (over 51% in 1934) of all tusks recorded. After 1948
the proportion never rose above 31%. A logical assumption would
be that the control officers became more selective. The truth
is more likely that there were fewer small elephants to shoot,
and I offer the following rationale.
Figure 43 presents graphs of 4 indices: 1) average tusk
weight taken on licence, 2) number of elephants killed on licence,
3) average tusk weight from all sources and 4) the contribution
that all tusks under 4.54 kg made to the total number of tusks.
The year 1939 is taken as 100 for all 4 indices (see also Table
167).
These illustrate graphically what was stated in the preceding
paragraphs. Tusk weights from licensed hunting declined as the
number of elephants taken went up. The big drop in such elephants
killed after 1952 produced a rise in average tusk weight, but it
never rose above an index level of 98. In the 1930s the index
had reached 122.
The overall average tusk weight rose more or less
continuously from 1933 to 1957 in a reverse trend to the decline
in licensed tusk weight. The cause of this is clear in the
diminishing contribution made by tusks of less than 4.54 kg.
The decline in licensed tusk size is easy to understand:
hunting exceeded the rate of replacement in the older age classes.
However, added to the control killing this might have brought
compensating mechanisms into play. Natural mortality should have
declined, and faster growth, increased fecundity and earlier
puberty should have tilted Uganda's elephant toward a MKE
situation. If this happened, the contribution of under 4.54 kg
tusks should have increased. It didn't because throughout the
period 1925-1959, the elephants were losing not only numbers but
range. Brooks and Buss (1962) documented the fact that at the
outset of this span, elephants occurred over 70% of Uganda's land
area. By 1959 they occurred in only 11%. At the same time as




FIG. 43 TRENDS IN 4 ELEPHANT POPULATION PARAMETERS
IN UGANDA 1926-1957. 1939 = 100

INDEX'OF : *
Average tusk weight of elephant
0-----O shot on licence.
No. elephant shot on licence.
Average tusk weight all sources.
Tusks under 4.54 kg as proportion
eeeooeoof annualtotal number of tusks.

/ -

A

Oi
%0

1940

1930

1950




164

the shooting, elephants were being displaced by expanding humans.
The situation changed from one in which humans were islands in a
sea of elephants, to its converse. As the humans increased and
their islands coalesced, elephant distribution became
discontinuous and fragmented islands constantly diminishing in
size. It is'here that Laws et al. (1975) work on MFPS elephants
is of particular relevance for it took place in one of the
declining islands. They described the progressive retraction of
the North Bunyoro elephant range from 1946-1968 at the rate of
a mile inwards a year (Laws et at. op. cit. p.18). They
summarised the situation thus (p.260) :
"Models for the MFPS population have been constructed
for 1946 (or earlier), 1966, and 1971. The adult
mortality rates appear not to have altered significantly,
but calf mortality has apparently increased by more than
50 per cent (perhaps as much as 100 per cent) since 1946.
Together with deferred maturity and reduced fecundity
(through overcrowding), this has led to a massive decline
in recruitment. It is estimated that relative natality
may have fallen to 34 per cent of the 1946 level and
relative recruitment at four years to 25 per cent of the
earlier level. PopuZation size is estimated to have
decreased from 22 000 in 1946 to 9400 in 1966 and to
7900 in 1971, that is by 64 per cent. Less than half of
this decrease was due to control shooting and sport
hunting. Natural regulatory mechanisms alone were
apparently responsible for a decrease in the residual
population of about 51 per cent up to 1971.11
I suggest that this is what happened wherever elephants
became diminishing islands in Uganda. The severity of effect
would of course have been related to the pace at which range
was depleted. A further factor worthy of note is that a
declining area in which to live would have caused an apparently
constant hunting effort to become progressively more severe in
effect. The rate of range loss was c.2% per annum. In the
years 1950-54 an annual average of 1,110 elephants were being
taken, as against an average of 1,540 p.a. in 1930-34. However
while the later elephant crop*was 72% of the earlier, it was
being taken on only 60% of the former range. The actual hunting
intensity had risen not declined.




165

The figures in Table 166 are of course minimal, for" they do
not take note of the tusks poached successfully, nor account of
all the deaths by wounding which were considerable (Laws et at.
1975). The discrepancy between the number of game licences and
firearms permits issued (Table 166) may be an indication of the
extent to which unlawful hunting took place. The Game
Department knew that a minimum of 45,583 elephants died in the
34 years and the real figure may well be double this amount.
The Game Department record is no more than an index of what
happened, yet it illustrates several important points. The first
is that the tusk weights of elephant taken on licence are useful
indices of declines in average age for as long as the licensees
seek the biggest elephants. The second is that an increase in
overall tusk weight may herald a worsening situation for the
elephants rather than an improvement. This is likely to be the
case if the elephants occupy a rapidly contracting range.
One Uganda experience deserves some prominence. In the
Budongo Forest the removal of elephants was decreed in the
interests of forestry (Laws et at. 1975). It was never achieved
despite intense effort and illustrates that in thick vegetation
elephants are difficult to exterminate.
In contrast to Uganda, the position reported by Rogers et at.
(1978) in Tanzania accords with the expected situation of a stable
or slightly declining tusk size. Here the elephants have not been
deprived of range; indeed they may have temporarily gained
ground. The attrition around the peripheries of the human islands
has produced an MKE effect and not an MFPS situation. The large
scale control shooting reduces numbers in the vicinity of human
settlement; biological compensating mechanisms come into play,
more food is available to the individual through reduced density;
there is lowered natural mortality, faster growth, earlier
puberty and increased fecundity. In short young, highly
productive populations have been induced.




VOLUME 2 (4) INFORMATION FROM TUSKS IN TRADE (A)
There have been a number of competent African elephant
studies in the past 3 decades e.g. Caughley 1973, Douglas-
Hamilton 1972, Laws 1969a, *Perry*1953, Sherry 1978
and Smuts 1975. They have produced a broad comprehension of the
species' biology. However through expense, the time necessary
to undertake them and a wide variety of human factors, studies
of this quality are of necessity local. They do not reveal
much about the continental status of elephant. What they have
shown is that, short of disciplined study, reliance on local
lore and guesses about the abundance of elephants is generally
not worthwhile. Efforts to overcome this through aerial
counting have been made, but on the basis of expanse and
consequent expense, the technique has limited application.
Added to this it is notoriously crude. To census elephants in
2
the 7,000,000 km (2.7 million square miles) (IUCN Elephant
Study Interim Report) of Africa where they still occur isf on
the basis of logistics alone, an impossible task.
In theory, tusk parameters should provide a relatively
cheap, practical and precise index of overall population trends.
The idea was approached by Quick (1963) and the principle
determining changes in age structure from characteristics of
trophies secured by hunting (e.g. the number of tines on deer
antlers etc.) is not new. The question at the outset of this
survey was merely whether access to large samples of tusks
could be obtained. Ideally, examination should be at the points
of the trade's greatest constriction the marts of Hong Kong
and Belgium. Of the two, Hong Kong was the obvious choice as
it receives ivory from a far wider base than Antwerp (refer
Fig. 5 Volume 1).
The Hong Kong traders were approached and, with one
exception, gave me access to their stocks of tusks. The only
stipulation was that I maintained normal business confidences
- a courtesy I was happy to comply with. A sample of 18,453




167

tusks weighing gross 185,942.81 kg, averaging 10.08 kg and
related to countries of origin (Table 168) was obtained through
examination of invoices. A separate and probably at least
partially overlapping sample of 22,260 tusk weights, grossing
214,829.79 kg and averaging 9.65 kg was obtained from official
import documents covering the period June-November 1978.
Examination of all or part of the stocks of 10 traders selected
at random provided data on 5,132 tusks. The following information
from each tusk was recorded: circumference at lip, sex, ivory-type
region or, if possible, country of origin and an estimate of
whether it was from elephants killed by manor which had died from
natural cause. The sheer volume of ivory to be examined and the
disruption of business entailed in moving it about, limited me to
measuring one parameter. I chose circumference at lip because it
was a) easy, b) uninfluenced by individual and regional patterns
of wear and c) uninfluenced by breakage in trade which often
affects the first part of the tusk's hollow section. Judgment as
to whether a tusk had come from man-induced or natural mortality
derived from personal experience in collecting ivory from a wide
variety of sources. My decisions were not entirely subjective
however.
An elephant tusk is fixed in the alveolus by a mass of
fibrous connective tissue. To remove it from a freshly killed
elephant, the surrounding maxillary bone has to be chopped away
with an axe. It is virtually impossible to do this without
leaving some slight evidence on the alveolar section of the
tusk. After the bone has been removed and the tooth pulp
extracted,-it is normal to try and remove such tissue as is
connected to the tusk. This adheres strongly and can only be
removed by scraping with a knife. Again, this leaves minute
but tell-tale marks on the tusk, which are easy to perceive.
The alveolar section of a tusk is longitudinally fluted and very
thin layers of tissue remain in the grooves of the fluting,
which dries quickly, remains tightly bonded to the tusk and is
easily seen.




168

It is of course possible to leave the tusks in situ until
decomposition has broken down the connective tissue which holds
them. Complete dissolution of the tissues takes at least ten
days and often longer. Usually they have broken down sufficiently,
within 4-7 days for the tusk to be loose. If it is wrenched at
and turned from side to side, the remaining fibres break and the
ivory can be drawn. However there is usually still some tissue
adhering and which dries onto the tusk if left. It is rather
noisome and commonly the hunter will endeavour to scrape away
the offensive material, again leaving tell-tale marks. Whether
a hunter waits till decomposition allows him to draw tusks or
not depends on a number of factors the most important being
whether he can afford the 4 day wait. Throughout this time he is at
some risk of losing his ivory to rivals or, to a lesser degree, to
officials. However much also depends on how far from his base the
kill was made the greater the distance, the less likely he is
to want to wait around. This is countered if the elephant was
killed for meat, which often takes a week or more to smoke. The
point is that whether tusks are cut out or drawn after 4-7 days,
the evidence of tissue still remains.
A third technique for extraction is to build a fire under
the fresh elephant head and literally cook the tusks out. I have
only heard of this being done in parts of Zaire. Again the
process seldom removes all tissue and the tusks often show signs
of over-heating characteristic lateral cracking and flaking.
When an elephant dies and is left to decompose without
human interference,--all tissue ultimately disappears. Tusks then
fall out of their sockets when a scavenger (or other elephants)
move the skull about. Such tusks have a distinctive greenish-
khaki stain on the alveolar section the product of advanced
decomposition. Rain and weathering will remove this, but replace
it with very fine hair-line shakes which never occur in a drawn
or hacked out tusk. Still later in time the tusk becomes cracked,
chalky and the surface flakes away seasonal fires, rain and
sunlight all combine in the process of weathering. Another




169
indication that tusks have lain unowned are the gnawings of
porcupines (Hystrix spp.). This occurs all over Africa but is
particularly common in the Cyclotiform ivory region. Moreover
porcupines don't just nibble ivory, they eat it. I have seen
many tusks of over 40 cms in circumference gnawed clean through.
The objective criteria for man-induced/natural mortality
with tusks were thus :
Killed by man Natural Mortality
The presence of : The absence of :
1) Any axe, knife or scrape 1) Axe, knife or scrape
mark irrespective of how marks;
trifling; 2) All tissue;
2) Any tissue no matter how 3) Blood stains.
little;
3) Any blood stain; The presence of
4) Any sign of burning without 4) The characteristic stain
other form of weathering, of putrescence;
5) Hair line shakes in the
alveolar section;
6) Weathering;
7) Burns and porcupine
gnawings in the absence
of 1), 2) and 3) in the
opposite column.
Tusks drawn and buried in moist earth would lose tissue
through bacterial action and, if they had not been scraped, would
appear to have been from natural mortality. Conversely, tusks
from natural mortality which were found before the carcass had
advanced in decay would appear to have come from killed elephants.
Despite such obvious sources of misidentification I feel that the
criteria provide a workable basis for separating killed from
found ivory.
Most of the 5,132 tusks examined and measured had been
acquired in the preceding 18 months. A minority were much older
dating back to the previous decade. Where possible my estimates




170

for region of origin were checked with the tusks' owners. In
no.case was an estimate of region incorrect and in several cases
country of origin was established. Thus all Sahelian ivory
examined was tied either to the Sudan or to the Central African
Empire. Identifications were aided in a number of cases by
characteristic markings made upon tusks by African Governments:
e.g. much ivory from'Angola bore 'a South African stamp with
'Rundul on it (a post on the Angola/Namibia border) The
Sudanese either brand tusks with a hot iron or stamp them with
metal punches in a characteristic manner.
The raw data on circumference, sex, origin and mortality
are presented in Tables 169-178. Where individual weights of
tusks were available en masse, they were collected and tabulated.

The following data were acquired :
* 1. A sample of Botswana Government 'found'
ivory.
* 2. A sample of tusks taken on licence by
Botswana hunters.
3. A sample of tusks taken on licence by
expatriate residents in Botswana.
t 4. A sample of tusks taken on licence by
tourists from the U.S.A.

(Table 179)
(Table 180)
(Table 181)
(Table 182)
(Table 183)
(Table 184)
(Table 185)
(Table 186)
(Table 187)
(Table 188)
(Table 189)
(Table 190)
(Table 191)

number
Kruger
number
Kruger
number
Kruger
number
Kruger
number
Kruger
number
Kruger
number

of tusks
National
of tusks
National
of tusks
National
of tusks
National
of tusks
National
of tusks
National
of tusks

and their weights
Park, 1972.
and their weights
Park, 1973.
and their weights
Park, 1974.
and their weights
Park, 1975.
and their weights
Park, 1976.
and their weights
Park, 1977.
and their weights
Park, 1978.

f rom
f rom
f rom
f rom
f rom
f rom
f rom

5. The
the
6. The
the
7. The
the
8. The
the
9. The
the
10. The
the
11. The

the Kruger National

*12. A sample of tusks and weights from the
Transvaal Nature Conservation Authority.
13. A sample of tusks and weights from the
Angolan/Namibian border.




171

*14. A sample of tusks and weights from
Angola/Namibia recorded in Johannesburg.
*15. A sample of tusks and weights from
Mozambique recorded in Johannesburg.
*16. A sample of tusks and weights auctioned
in Rhodesia 1977-78.
17. A sample of tusks and weights from Kasungu
National Park, Malawi.
18. A sample of tusks and weights from
Lilongwe, Malawi.
19. A sample of tusks and weights from
northern Malawi.
20. A sample of tusks and weights from
southern Malawi.

(Table 192)
(Table 193)
(Table 194)
(Table 195)
(Table 196)
(Table 197)
(Table 198)

* Indicates data supplied from company records;
t Provided by hunters;
The rest are supplied by Government authorities.
Both tusk circumferences and weights can be presented as
ivory 'shadows' of the elephants they came from. Even if the
sex of the elephants is unknown they can be compared to a
combined male + female distribution of tusk weights from known
populations to give some idea of the sectors which they
represent. When it is possible to determine sex as well as
circumference or weight, the picture becomes much clearer.
Conversion of tusk circumference and weight distributions
into sex and age distributions are based entirely on data
obtained from the elephant cropped in Uganda, Kenya and Tanzania.
In the first instance age was determined in the original
collections-by the method described by Laws (1966). Various
mathematical expressions were examined with the aid of a Wang
2200 computer and the following were selected as best fits for
male tusks.
Circumference
age(y) = 13.0439 + 0.81268 x (tusk circumference)
which explained 90% of variance between ages 0-49 (Figure 44).




*

31
3,
m 3
03

* e

Age (y) = a + bx

02
a2
4
F.
01
C)

*/*

FIG. 4 4 MALE TUSK CIRCUMFERENCE RELATED TO
AGE FROM 5 EAST AFRICAN POPULATIONS
COMBINED. (n = 701)

Age in Years




172

Weight :
log age(y) = log..-2.5312 + 1.5193 log x (tusk weight)
for ages 1-23 which explained 91% of variance (Figure 45), then
age(y) = -16.5286 + 1.1318 x (tusk weight)
for ages 24-52 which explained 63% of variance (Figure 46).
Data for males in their last decade of potential life
(50-60 years) are too few to make estimates of worth on either
tusk circumference or weight. However I would expect them to
call for different treatment to cater for senescent slow-down.
Within the limits of time and available funds it was not
possible to establish a satisfactory mathematical expression
relating either female tusk circumference or weight to age and
growth curves were drawn by eye to relate age from dentition to
mean tusk circumference and weight (Figures 47 and 48). This
produced adequate results for the purposes of this report, but
the defection will be rectified in the pending Laws and Parker
paper on tusk growth. '
Attribution of each tusk circumference or weight to an age
class seems adequate for present requirements viz. 0-5, 6-10,
11-15 etc. Accordingly the pivotal circumferences and weights
for each age class were calculated from the male regressions or
read by eye for females, both of which are presented in Table"199.
These pivotal measurements are rounded to the nearest 0.5 cm for
circumferences and 1.0 kg for weights.
If these relationships have retained their value as aging
criteria through the lumping into 5 year classes and rounding
into 0.5 cm or 1.0 kg classes, an elephant should still be
assigned to the same age class using dentition, tusk circumference
or weight. Further, if any two of these parameters are plotted
against each other, the slope should be 1.0. Samples are given
in Figures 49 a-c. The scatter along the diagonal is small
enough to have confidence in aging from either tusk circumference
or weight and that the treatment they have been subjected to has




4-
04
3- log age (y) =log a b b(log x)
2-
5 10 15 20 24
Age in years
FIG. 45 MALE TUSK WEIGHT RELATED TO AGE (1-23 YRS) FROM
5 EAST AFRICAN POPULATIONS COMBINED. (n=1862)




ip

- 2,

age (y) = a + bx

"Q

30 35
Years

Age

FIG. 46 MALE TUSK WEIGHT RELATED TO AGE (24-52 YRS) FROM
5 EAST AFRICAN POPULATIONS COMBINED. (n=527)
(See Figi 45 for Ages 1-23 Yrs)




29-
28-
27-
26-
C 25-
r 24-
u 23-
m
f 22-
e -
r
e 21-
n
c 20-
e
19-
17-
m '16-
15-
14-
13-
12-
2 4 6 8 0 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40, 42 44 46 48 50 52 54 56
A G E I N Y E A R S
FIG. 47 A CURVE RELATING FEMALE TUSK CIRCUMFERENCE TO AGE, DRAWN BY EYE, BASED ON
MEANS AT AGE FROM A SAMPLE OF 824.




9-
U 7 ---
s-
h 6 - - --- !
W
1 4 .-o,-,
n
3
g
2 45- 0 1 4 1 8 2 2 24 2 8 3 2 3 6 3 0 42 4 6 4 0 5 4 5 8 6
At
FIG 48 ACRE4--NG EAETS EGT OAE RW YEYBSDO EN
ATnEFO A1PEO 74




40 40
35- 35
30 30 W
25 -25
Ax X I
k" /X M h I u
" 20 20 '
15 15
10 -10
5- -5~
5 5Y X
5 10 15 20 25 3n 35 40 45
AGE DENTITIONN)
40 -40
-35
35-
30 30
x v..
H 25 25
S2020
15 51
0 10- -10 a
5i
~cU"
5 10 15 20 25 30 35 40 45
AE (D E N T I T 0 N)
FIG. 49a AGES DERIVED FROM TUSK WEIGHT, DENTITION AND TUSK
CIRCUMFERENCE PLOTTED AGAINST ONE ANOTHER TO SEE IF
THEY FALL ABOUT A SLOPE OF 1.0 AS A CHECK THAT LUMPING
INTO 5 YR CLASSES HAS NOT DISTORTED THEIR VALUES.
A = MFPS FEMALES (Tusk wt/Dent n=141; Circ./dent
n=137)- B = MKE FEMALES (Tusk wt/Dent n=212;
Circ./dent n=203).




Jv 50
45 45
40 40
35 -- 35 z
30 / 30
S25-- 25 u
20 / 20
15 / 15
YL<
A G E D E N I T 1
FIG. 49b AGES DERIVED FROM TUSK WEIGHT, DENTITION AND
TUSK CIRCUMFERENCE PLOTTED AGAINST ONE ANOTHER
TO SEE IF THEY FALL ABOUT A SLOPE OF 1.0 AS A
CHECK THAT LUMPING INTO 5 YR CLASSES HAS NOT
DISTORTED THEIR VALUES. MFPS MALES.




50-
45 -45
40 ---40
:35- -35 Z
30- -3C
S25- -425
15 -15
10 -10
5-5
5 10 15 20 25 30 35 40 45
A G E (D E N T I T 1 0 N)
FIG. 49c AGES DERIVED FROM TUSK WEIGHT, DENTITION AND
TUSK CIRCUMDERENCE PLOTTED AGAINST ONE ANOTHER
TO SEE IF THEY FALL ABOUT A SLOPE OF 1.0 AS A
CHECK THAT LUMPING INTO 5 YEAR CLASSES HAS NOT
DISTORTED THEIR VALUES. MKE MALES.




173

not seriously distorted the original direct relationship
between age by dentition/c.ircumference or weight at age.
As male tusks were separated from females' at the time
circumference data were collected in Hong Kong, conversion to
appropriate age distributions is merely a matter of referring
to Table 199.
Table 200 presents average female tusk weight for each
year class, sample size and the size of that class as a percentage
of the total sample (2,734), average tusk weight being read from
the graph in Figure 48. These data are also presented in Table
200 as 5 year running averages for female tusk weight to smooth
out fluctuations in simple year to year averages due to small
samples etc. In Figure 50 the running averages are presented as
cumulative percentages of the population, which illustrate that
tusks of 7.5 kg have a 94% chance of being male, a proportion
which rises steeply to 100% thereafter. For practical purposes
I take this weight as the arbitrary point above which all tusks
will be classified as male.
If we know, or can assume a sex ratio in our ivory
populations we can estimate what the male component of the
under 7.5 kg tusks might be through the formula in the following
example :
All tusks over 7.5 kg are male.
The assumed sex ratio is 40% male : 60% female.
The sample is 1,000 tusks.
1
m = 100 tusks >:7.5 kg and therefore male
m = male tusks <7.5 kg

f = total female tusks, all <7.5 kg




OQ e 0
4-
@0.
A
2 4 6 6 10 12 14 16 18 20 22 24 2!6 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 5A 60
A G E I N Y E A R S
100-
90o
70 o
60
d50 o -
40 -
30 -
20 I
10-
1 2 3 4 5 6 7 i
T U SK WEIGHT K G.
FIG. 50 FIVE YEAR RUNNING AVERAGES OF MEAN FEMALE
TUSK WtEIGHTS RELATED TO AGE (A), AND SAME
DATA EXPRESSED AS CUMULATIVE PERCENTAGE TO
ESTABLISH 7.5 KG AS CUT-OFF POINT BEYOND WHICH
ALL TUSKS HAVE BEEN ARBITRARILY CONSIDERED
MALE (B). (n = 2734)




174
then
(m 1+ m) = 400 and f =600
therefore 100 + m = 400
m = 300, which can be presented as a single
<7.5 kg class or broken down into 1 kg classes in proportion to
the distribution of tusk weight <7.5 kg.
The structure of the female segment of the population could
be calculated by yet further assumption, which in this report
will not be attempted. In the analyses which follow, the actual
(>7.5 kg) + hypothetical (<7.5 kg) males will be combined to
provide comparative population indices.
To provide pause, analysis of the circumference and weight
data presented in tables of this chapter, form the basis of the
following section.







175

VOLUME 2 (4) INFORMATION FRO14 THE TUSKS IN TRADE (B)
In this Chapter tusk circumference data.obtained in Hong
Hong (summed in Table 201) and weights collected in southern
Africa, have been transformed from 'ivory populations' to
elephant populations. The pictures derived are presented as
cumulative percentage curves and as an indulgence as
histograms, for some find one easier to follow than the other.
Presentation of these population pictures is more to
demonstrate what could be achieved with the type of information
the ivory trade could provide, than to give definitive statements
on the status of any elephant population. The data could be
subjected to more penetrating analysis, but as they stand should
not be stretched beyond their worth. This caution notwithstanding,
the information provided by ivory in the trade can give a more
objective insight into general trends in Africa's elephants,
than any other presented to date.
The circumference data have been transcribed into male and
female population structures of 5 year acTe classes, for man-
induced and natural mortality. The patterns from the man-induced
deaths should illustrate what humans in a particular region are
doing to their elephants. A cropping programme as that already
described in Chapter 1 of this volume, should represent all age
classes in the proportions in which they occur in a population;
the highest number being in the youngest categories, diminishing
progressively with age. Hunting for meat will produce abroad
spectrum of ivory from across the population, with only immature
being under'-represented. Hunting for ivory will show selection
of older classes of both sexes. Trophy hunting will-provide the
most extreme expression of this selectivity concentrating on
old males. Defence of property will tend to emulate sport-
hunting when it is largely concerned with destroying male
venturers from, elephant islands (national parks) in the midst of
humans, though somewhat younger classes ofmales would'be taken.
In situations such as occur in southern Tanzania, where humans




176

form islands in an ocean of elephants, killing will be
unselective and approach a random cross-section of the
population minus the youngest age classes.
The picture from natural mortality in a population not
subjected to human interference should approximate to the
structure of the living population mortality curves being
more or less reciprocals of survival curves. Laws 1969a and
Laws et at. 1975 have described the pattern of natural mortality
in elephants. In brief it is one of a high death rate in the
first year of life which declines steeply to a low and constant
rate by adolescence. In the female this trend of low mortality
persists through adulthood until the onset of senescence late
in the fifth decade, when it accelerates to achieve 100% at the
end of the sixth decade. In males the pattern is similar to the
females' until the mid-20s when there is a pronounced increase
in deaths. This appears to slow down slightly in the fourth and
early fifth decade, then accelerate to 100% by the end of the
sixth decade. The differential mortality between the sexes in
their third decade is such that it reduces the chances of any
significant number of males reaching the potential of 60 years.
Indeed very few ever pass 50. The tusk data should reflect the
expected mortality curve with a high proportion of young
animals, a descending but even distribution of adult female
tusks over the age span 15-45, then an upswing with rather more
old females in the last decade of life. The males should show
a marked bulge in the 20-30 year range, a slackening in the
30-40 class and then a secondary bulge at the end of life,
45-55 years.
In areas of heavy hunting with muzzle-loading guns, bows
and arrows, cable-snares and wholesale 'control' work, wounding
rates are usually very high (Laws et at. 1975, Rodgers et at.
1978). It is unfortunate that the provision of heavy calibre
rifles does not as a rule, reduce wounding rates. In many cases
it increases them, for the timorous hunter can shoot at greater
distance but correspondingly lower accuracy. Elephants dying




177

of bullet wounds may take months to succumb and when they do,
there will be no way to dif f erentiate their tusks from those
available from genuine 'natural' mortality. Such deaths will
cause the two mortality sources to produce similar pictures.
Tusk circumference
Sahelian ivory from the Central African Empire
The data are illustrated in Fig. 51 and Tables 169, 170 and
202. The sex ratio of 69.3% males and 30.7% females is a strong
reversal of natural sex ratios and evidence of deliberate
selection for males. The proportion of males taken in each age
class between 11-15 and 31-35 is similar. In the natural
population the 11-15 class would be far more abundant than 31-35
year olds. This too is evidence of a strong selection of larger
males, for.they form a greater proportion of the kill than they
are of the population.
The female picture is based on only 130 tusks, but it
suggests positive selection for animals over the age of 21.
The data from natural mortality are too few for speculative
analysis, other than to observe that they are in sympathy with
the hunting kills and lack the expected proportion of many young
elephants.
The average male tusk circumference of elephants killed is
32.28 cm (Table 169) which would convert to an age of c.23 years
and a tusk 'Weight of 9.3 kg.
The average female tusk circumference was 19.5 cm. which
converts to an age of c.19.5 years and weight of 2.75 kg. With
the sex ratio of 69.3% and 30.7% the average tusk weight of the
hunted elephants would be 7.29 kg.




AGE CLASSES

>40 1

0-5 6-10 11-15

.

n-

FIG. 51 SAHELIAN IVORY FROM CENTRAL AFRICAN EMPIRE. MAN-INDUCED MORTALITY AND NATURAL MORTALITY AS
CUMULATIVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES.

1001-

Natural mortality
(n=7)
1-20 21-25 6-30 3 m1 3 40

Natural mortality
V
/*
I
I
Male
O Female
I ("?' I I I I I I

Man-induced mortality

901-

S0-Z 6-10 11-15 16-20 21-25 26-30 31;-35 36-40 >40
AGE CLASSES

AGE CLASSES

M A L E
Man-induced mortality
(n -294)

F E M A L E
Man-induced mortality
(n=130)

50-
40-
a 3
o
a 2
1 0-
0-
40-
C
r
S30
a 20-
*
0
6-

50-
40-
u 30-
S20-
10-

0-5 A6-10 E11-15 16-20 21-25 26-30 31-35 164 >
AGE CLASSES

AG 1 6-10 1 11-15 1 16-20 1CLASSES21-25 1 26-303135 130 1
AGE CLASSES

50-
40-
30-
20-
10-
0-

Natural mortality
(n-67).

0-5 6-0 1 11-15 16-20 1 21-25 1 2630
AGE CLASSES




178
This is substantially less than the average of 17.75 from
1,050 tusks from the Central African Empire shown in Table 168.
However the two sources are not directly comparable as the data
in Table 168 will have included both Sahelian and Cyclotiform
ivory, both of which are exported from that country.
The small samples from male and female natural mortality
(Table 170) suggest older animals :a male average circumference
of 34.76 cm. = 27 years =14 kg tusk weight. The female average
circumference was 22 cm =26 years = 4.3 kg tusk weight.
Combining these with the killed sample raises the average
circumference only slightly : males =32.74 cm = 24 years=
9.9 kg, while females stay the same.
The overall sex ratios change to 72.5% male : 27.5% female
which, with the slight increase in male tusk size raises the
average tusk weight to 7.93 kg.
Sahelian ivory from the Sudan
The data are illustrated in Figure 52 and presented in
Tables 171, 172 and 203. The sex ratios of 47.6% male :52.4%
female in the hunted elephants and 48.5% male :51.5% female
in natural mortality are both close to what might be expected
in the population, and suggest only slight bias in selecting
males. In all sets of data, both male and female, man-induced
and natural mortality show proportionately far more immatures
than in the C.A.E. sample. The inference is that the hunting
is relatively indiscriminate and that tusks from natural
mortality are recovered with greater efficiency than in the
C.A.E.
The average male tusk circumference in those taken by man
is 27.22 cm = 18 years = 6.4 kg. The corresponding female tusk
circumference was 19.37 cm. = 19 years = 2.3 kg. The average
tusk weight of both sexes combined is 4.25 kg.




FIG. 52. SAHELIAN IVORY FROM SUDAN. MAN-INDUCED MORTALITY AND NATURAL MORTALITY AS CUMULATIVE FREQUENCY
GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES.
100 100
Man-induced mortality Natural mortality
90 90 (
80 80-
70 70
o O
. 60 60- Male
o Male
50 6 Female 50 O Female
40 4-
30 30-
20 / 20 0
10t 10-
S I I I I I I I I I I
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES AGE CLASSES
M A L E F E M A L E
50- Man-induced mortality 50- Man-induced mortality
(n-348) 40 (n-383)
40- 4 0
S30- v 30
a20- u 20
e e
10- 10
0 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40 0 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 40
AGE CLASSES AGE CLASSES
Natural mortality 50- Natural mortality
(n-94) 40 (n-100)
. 4,1
30- 30-
j. 20, 20-
o *
1 10
0 0-5 6-10 11-15 16-20 21-25 26-30 31-35 136-40 >40 0 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES AGE CLASSES




179

The average male tusk circumference from natural mortality
(Table 172) was 25 '.84 cm = 16 years = 5.37 kg. The corresponding
female circumference was 19.45 cm = 19.5 years = 2.75 kg. The
combined tusk weight from natural mortality would average
4.02 kg. Again this and the average from hunted elephants is
substantially' lower than the average of 11.17 from 3,405 tusks
in Table 168.
Cyclotiform ivory:
The data are presented in.Figure 53 and Tables 173, 174 and
204. The most striking feature of the illustrations in the
cumulative percentage curves and histograms in Figure 53 are the
markedly different patterns between the sexes. Both hunted and
natural mortality show similar patterns of selection for mature
males. Both female illustrations give a much broader spectrum
of age classes with far greater weight being placed upon young
animals. There may be some valid reason for differential
hunting of the sexes, but one would not expect it to be reflected
in the natural mortality. A more plausible explanation is that
in Loxodonta a. cyclotis female tusks are relatively more
slender at age (which subjectively I believe to be the case) than
in L. a. africana and that my circumference interpretations are
inappropriate in female cyclotiform ivory. If this is so, it
would explain the apparent abundance of younger female classes
and believing this to be likely, I am unable to use the material
for general comparisons.
The male data suggest marked selection for the large
adults :a point that is borne out by the sex ratio of 77.5%
males :22.5% females in hunter kills. That the pattern is
reflected closely in the natural mortality suggests a very high
wounding rate producing deaths to mask the population's
intrinsic rates. As in the two preceding cases the expected
high proportions of immatures are absent.




FIG. 53 CYCLOTIFORM IVORY FROM CENTRAL AFRICAN EMPIRE, CONGO, CAMEROUN, SOUTHERNMOST SUDAN AND
ZAIRE. MAN-INDUCED MORTALITY AND NATURAL MORTALITY AS CUMULATIVE FREQUENCY GRAPHS AND
HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES.

100r

Man-induced mortality

Natural mortality
0-
Female ()* Male
//
I I I I I I I I I

90

Male

Female
o/

0
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES

,

15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES

M A L E
50- Man-induced mortality .
(n=701)
10- q
20-
I0-
0 0-5 6-10 11-15 120 21-25 26-30 31-35 36-40 >40
AG E C LA SSEBS

F E M A L E
Man-induced mortality
(n-224)

0 0-5 6-10 11-15 16-20* 21-25 26-30 31-35 36-40 >40
AGE CLASSES

so-,
40-
20-
1o-

50-
40-
S3D-
1 2o-
2&*

Natural mortality
(n=267) .

Natural mortality
(n=147)

AGE CLASSES

0 0-5 6-10 11-15 16-20 12 630 31-35 36-40 >40
.AGE CLASSES




180

The average male tusk circumference from hunted animals
(Table 173) is 31.99 cm = 23 years = 9.3 kg of ivory. The
average from male natural mortality (Table 174) is 30.58 cm
22 years = 8.7 kg. Obviously the inclusion of female ivory
would drag these down yet further and widen the difference
between the average of 10.83 from 6,276 tusks shown in Table 168.
East African ivory :
The circumference data are presented in Figure 54 and
Tables 175, 176 and 205. The patterns of man-induc6d mortality
show larger proportions of animals under 20 years than in any
of the preceding figures, suggestive of more intense and general
hunting. This gains support from the hunting sex ratios of
53.4% males : 46.6% females which indicates relatively slight
male selectivity. For the first time in the natural mortality
patterns, we have a pattern approaching the expected high
proportion of immature. The suggestion is that they are sought
more diligently and successfully.
At this point it is perhaps apposite to comment on the
more or less consistent secondary (or in East Africa's case,
primary) peak in male mortality that occurs in the 11-15 year
old class. -It is in this age group thA young males reach a
height that renders them among the bigger animals in a family
unit. This relative size makes them a target for anyone hunting
animals from family units. At this age they are expelled from
family units (Douglas-Hamilton, 1972), the process of which is
protracted. During this period they tend to hang about the
outskirts of the herds, frequently tagging along some distance
behind, making them the most vulnerable members of family units
and many are shot. Once the break with the maternal unit has
been made, and the adolescent male is in the company of other
males, his (then) relatively smaller size switches selectivity
in favour of the larger males. At the same time the young male
is no longer such a herd-fringe animal and consequently less
vulnerable.




0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES

FIG. 54 EAST AFRICAN (TANZANIA AND KENYA) IVORY. MAN-INDUCED MORTALITY AND NATURAL MORTALITY AS
CUMULATIVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES.

Kan-induced mortality ol100
90
70
male 60
4o
30
1/ 2O
I I I 10

I

90
70
S60
50oi
40
30

Natural mortality
Male

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40
AGE CLASSES
M A L E
S Man-induced mortality
(n-641)

F E MAL E
Man-induced mortality
(n-559)
AGE CLASSES
Natural mortality
(n-110)

0-5 6-10 111 16-20 21-25 26-30 31-35 36-40 40
AGE CLASSES

0 0-5 6-10 11-15, 16-20'21-25 '26_30 '31-35 '3640I
AGE CLASSES

so-
Natural mortality
40- (n-148)
30-
20
v
10
0 0-5 6-10 11-15 16 20 21-25 26-30 31-35 36-40 40
AGE CLASSES




181

The carryover of the 11-15 year age class in the natural
mortality record is likely to be an indicator of fatal wounding
which are not recovered by hunters.
The average male East African tusk circumference from human
kills (Table 175) is 28.66 cm = 19 years = 7 kg tusk weight. The
average hunted female tusk circumference is 19.07 cm = 18.5 years
= 2.35 kg. The sexes combined in the ratio of their occurrence
in the sample would give an average tusk weight of 4.83 kg.
The corresponding data from natural mortality (Table 176)
are : male tusk circumference 24.80 cm = 14 years = 4.34 kg;
female tusk circumference 16.81 cm = 13.2 years = 2.0 kg. The
average tusk weight of the sexes combined in the ratio of their
occurrence (male 57.4% : female 42.6%) is 3.34 kg.
A sample of 574 sexed and weighed tusks was obtained in
June 1979 in Kenya. This is worth inclusion here for comparison
with the sexed circumference data. Details are presented in
Figure 55 and in Table 206. The average male tusk weight was
7.79 kg = 20 years = 29.30 cm. circumference; the average female
tusk weight was 3.19 kg = 21.5 years = 20.5 cm circumference.
The sex ratio was 42.9% male, 57.1% female and this gave a
combined sample tusk weight of 5.16 kg. The sample was
,predominantly.from poached elephants and the hunting was
indiscriminate (Somali). It illustrates a preponderance of
young animals.
-The range of average tusk weights derived from circumferences
are close to, but still lower than the weights from the weighed
and sexed sample or the average of 5.36 kg for Kenya or 7.08 kg
for Tanzania in Table 168.
south-Western ivory :
Tusk circumference data are presented in Figure 56 and
Tables 177, 178 amd 207. The kills illustrate that both family




MALES

40-
, 30
~20-
i5 10-
50-

41-45 46-50
41-45 46-50

0-5 :6-10 ) 11-15! 16-20' 21-25 26-30 31-35 36-40
AGE CLASSES

FEMALES

S30-
20-
1 -

I I I I I I I I I
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45
AGE CLASSES

FIG. 55 DISTRIBUTION OF AGES OF 246 MALES AND 328 FEMLES
FROM A SAMPLE OF 574 TUSKS OBTAINED IN KENYA IN
1979, MAINLY FROM POACHERS.




FIG. 56. SOUTH-WESTERN IVORY FROM ANGOLA AND SOUTH AFRICA. MAN-INDUCED MORTALITY AND NATURAL MORTALITY
AS CUMULATIVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES.
100 Man-induced mortality 100- Natural mortality ..-
80 -/ 0- O
70 70-
J 60- 60
550-
40 40-
30/ 30 Male
Male ) Female female
2020
10 l0 -
0-;5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 > 40 0-5 6-10 11-15 16-20 21-15 26-30 31-35 36-40 .40
AGE CLASSES AGE CLASSES
MALE FEMALE
50 Man-induced mortality 50- Man-induced mortality
(n-466) (n=306
40- 40-
*30- 30-
0
20- 20-
4 .0 .
10 a 10
0 0-5 6-10 11-15 16-20 21-25 i26-30 31-35 i36-40 40 0 0-5 6-10 11-15 16-20 2t-25 26-30 31-35 36-40
AGE CLASSES AGE CLASSES
50 50
Natural mortality Natural mortality
40 (n=87) 40 (n=53)
30 30
14 2p 20
10 10
0 0-5 6-o0 11-5 16-20 21-25 26-30 31-35 36-40 >40 0 0-5 6-10 11-15 16-20 21-25126-30 31-35 36-40
AGE CLASSES AGE CLASSES




182

units and independent males are hunted, selection in both cases
being for the bigger animals in these social components of the
population. The sex ratio of 60.4% males 39.6% females
bespeaks a positive selection of males. The data on natural
mortality (Table 178) are few, but suggest a moderately effective
recovery of what should be available, overlain with the
consequence of a high fatal wounding rate.
The average hunted male tusk circumference (Table 177) is
28.08 cm = 19 years = 7.0 kg tusk weight. The average female
circumference is.19.5 cm = 19.5 years = 2.75 kg tusk weight. In
the sex ratio of the sample, the average tusk weight from this
South-Western ivory would be 5.32 kg. This has no likeness in
Table 168 as the Botswana ivory though South-Western, is not
comparable (see later).
Before proceeding to examination of the unsexed weight
data, let me summarise the salient points from the circumference
material. That there are differences in hunting effect is plain.
In the Sahelian C.A.E. and Cyclotiform ivory there is heavy
selectivity of males. In East and South-Western Africa,
selectivity for males persists but at a lower level. The
hunters of the Sudan are the least discriminating in what they
take. A major inconsistency is apparent between the derived
average weights from circumference and those obtained from the
bigger samples in Table 168. The Hong Kong dealers were asked
why this should be. They pointed out that as a general rule
tusks of 14 kg or more were not held in Hong Kong. The majority
were extracted from each consignment shortly after arrival and
re-exported to Japan. The tusks I examined were influenced by
this selection process. Recourse to official export documents
confirmed this : a sample of 38 export shipments to Japan
grossed 19,641.41 kg consisting of 1,357 tusks. The average
exported tusk weighed 14.47 kg. It did include a number of
less than 14.0 kgs, but I accept the contention that stocks held
in Hong Kong will have been biased by the extraction of large
ivory. In consequence the male : female ratio will have been




183

biased toward an over-representation of females, and the patterns
of male mortality in particular will have been biased towards a
far younger structure than was the actual case.
Tusk weight
The tusk weight data which follow were obtained inside
Africa at concentration points in the ivory flow, i.e. the
records of general ivory buyers and official conservation
authorities. In view of the several assumptions which would
have to be made I have not tried to extrapolate female
population structures from the data. Instead male populations
are presented as comparative indices;*their under 7.5 kg
structure being devised by the formula outlined.in the preceding
Chapter, and in which one of three sex ratios was applied.
(1) 44.6% males when there were grounds to believe that
a general cross-section of the population was likely
to be represented in the ivory (e.g. when it derived
from cropping in national parks or generalised
'control' work); this male proportion being derived
from the East African cropping material of Laws et at.
(1975) combined with Smuts' (1975) South African
records from Kruger culling;
(2) 62.95% males when tusk weights were from general
hunting; the proportion being derived from a
combination of sex ratios from the examined Hong Kong
circumference-measured tusks;
(3) 100% males when the tusks were from sportsmen or
peripherally control shooting.
The first example is of Botswana found ivory that is
tusks handed in to Government from natural mortality. The data
are presented in Figure 57 and Tables 179 and 208. The sample
size is small (150) but the pattern is the converse of the
expected in natural mortality, i.e. numerous immature, few
matures and rather more senescent animals.




L I I I I I I I I I I I I

70-
60
50-
40
30
20
101

0-5 6-10 11-15 16-20 21-25

26-30 31-35 36-40 41-45

46--50 51-55

CL AS SE S

AG E

50-
P 40-
e
r
30-
.C
e 20-
n
t

36-40

. I 1I
41-45 46-50 51-55

16-20 21-25
AG E

26-30 31-35'

0-5 6-10 11-15

.CL AS SE S

MALE IVORY FROM NATURAL MORTALITY
AS CUMULATIVE FREQUENCY CURVE AND
RELATED TO AGE CLASSES. (n = 150)

FIG. 57

BOTSWANA.
EXPRESSED
HISTOGRAM




184
The second example is of 3 Botswana hunting results
Mi from Batswana hunters,
(ii) from expatriates resident in Botswana, and
(iii) from visiting U.S. sportsmen.
In these cases the weights were treated as 100% male, although
the Batswana hunters do take a few females. The results are
presented in Figure 58 and Tables 180 182 and 209. They
clearly illustrate three levels of selective hunting for male
elephants. The Batswana shoot elephant to make money from both
tusks and hide. They select the largest animals they can in
available time, but having a modest standard of living do not
seek such high rewards as the next group white expatriate
residents. Consequently they accept slightly smaller trophies
and their average tusk weight is c.14.4 kg (period 1974-1978).
The white expatriate residents also shoot elephants to make
money, but being much keener to get the most out of ivory, they
are more selective and take an average tusk weight of 16.5 kg
(period 1974-1978). The constraints upon the two local groups
of Botswana hunters is that they have to rely on their own
resources and limited time away from work. The third group
- the U.S. sportsmen do not seek ivory to sell, but as trophies.
They want the biggest tusks available, have more time than the
other two classes and, more important, through their professional
guides are able to deploy more people searching for elephants.
Expense is less limiting to them. Consequently their average
tusk weight is 22.5 k4 (period 1968-1978); taking substantially
larger and older animals than the country's residents.
The third example of weights is from cropping operations in
the Kruger National Park of South Africa (1972 1978). An
annual cull is made to keep the population at a desired level
(Smuts 1975). The data are presented in Figures 59-62 and in
Tables 183-189 and 210. The assumed sex ratio of the overall
population was 44.6% male: 55.4% female, and it was on this
basis that the under 7.5 kg segment was estimated.




FIGURE 58.

BOTSWANA. MALE IVORY FROM THREE HUNTING
CLASSES EXPRESSED AS CUMULATIVE FREQUENCY
CURVES AND HISTOGRAMS RELATED TO AGE
CLASSES.
BATSWANA HUNTERS n = 477
RESIDENT EXPATRIATE HUNTERS n = 235
U.S. TOURIST HUNTERS n = 65




>55

46-50

I .t I I
0-5 6-10 11-15 16-20 21-25
AGE

" i I I i N 1
0-5 6-10 11-15 16-20 21-25 26-30 31-35. 36-40 41-45 45-50 51-55 >55
AGE C L AS S E S

P 70
60
50
o

'/!
f--U.S. Tourist
/ hunters
//

Resident
.expatria
hunters

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES

50.
P
e40,
r
30
c
e20
n
t
10

BATSWANA HUNTERS

46-50 51-55

11-15 16-20 21-25 26-30 31-35 36-40 41-45

AGE

CLASSES

50-
P
40-
e
r
30-
c
e 20 -
n
t10 -

RESIDENT EXPATRIATE
HUNTERS

26-30 31-35 36-40 41-45 46-50 51-55
CLAS S ES

>55
>55

50 -
P
40-
e
r
30-
c
e20-
n
t 10-

U.S. TOURIST
HUNTERS




0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES

0-5 6-10 11-15 16-20- 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES

I I-- I i I I I 1 | i I I i
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES

50
P
e 40 -
r
r 30-
e 20
n
t 10--

1972
(n = 742)

50
40

S1973
(n = 177)

c
e 20-
n
t 10 -

P
e 40 -
r
30 -
c
e 20
n-
t l0

1974
(n = 434)
, j
26-30 31-35 36-40 41-45 46-50 51-55 >55
CLASSES

0-5 1 6-10 1 11-15 1 6-20 1.21-25
AGE

50 -
P
40 -
e
r
30-
c
e 20
n
t 10

1975
(n = 686).

FIG. 59 AGE STRUCTURE OF MALE ELEPHANT POPULATION IN KRUGER
NATIONAL PARK DERIVED FROM CROPPED MATERIAL AND AN
OVERALL SEX RATIO OF 44.6% MALE : 55.4% FEMALE.
1972 1975.




_ ILT

I......

50
P40-
e
r
30
c
20-
e
n
t10-

1976
(n = 434)

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55
AGE CLASSES

>5I I
>55-

50-
P 40-
e
r
30-
c
20-
e
n
t 10
50-
P 4
40-
e
r
30
c
e 20
n
t 10

1977
(n = 204)

263 i 1I I i
26-30 31-35 36-40 41-45 45-50 51-55 >55

0-5 6-10 1 11-15 16-20 1 21-25
AGE

CLASSES

1978
(n = 484)

I 1 I1 1
46-50 51-55 >55

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45
AGE CLASSES

1972-78 Average
(n = 3161)

0- I i 1 1 2 3 1 1 1
0-5 6-10 11-15 16-20 21-25. 26-30 3i-35 36-40 41-45 46-50 51-55

>55

C LA S'S E S

AGE .

FIG. 60 AGE STRUCTURE OF MALE ELEPHANT POPULATION IN KRUGER
NATIONAL PARK DERIVED FROM CROPPED MATERIAL AND AN
OVERALL SEX RATIO OF 44.6% MALE.: 55.4% FEMALE.
1976 1978, AND AVERAGE 1972 1978.

50
P
S40-
e
r
30
c
e 20
n
t 10




-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 > 50
AGE CLASSES

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >50
AGE CLAS SES

I I i I .

0oo-
90-
80-
70
60 -o
S40 -
30-
20
10
" o

100
90
80
70
S60
L)05
/ so
40
30
20
S972
10
I I I I I I I I

I-4
O -
*
H
0"
*I-
MT >
a)
ci
En#
L: -3
C
J
or
C]

1973
I I I I I I I I I I I
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >50
AGE CLAS S E S

100
too
90
so
70
60
S50
40
30
10

100
90
80
70
*
60
30
2U
10

1974

1975

r r t i I I I

I I

I I

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 50
AGE CLASSES




1976
-I
I 1 I I I I I I 1 I I
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >50
AGE CLASSES

- -

-

100
90
80
70
o 60
040
5
30
0
10
90
80
70
e
60
o
-4o
4J 50
rc
E40
30
20
10-

- 4
H
r
f-3
Z.0
HQ
t
>
r
o
0
F-
bi
VI

1971

1 I I I I I I 1 1 I I
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >SC
AGE CLASSES

1978

I I I 1 I I1 I I I

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >50
AGE C L A S S E S




185

As the Kruger population is believed to-be young and
fertile (Joubert pers. comm.) the pattern of the male index
should be in the range between the MKE and combined East African
,data. The data from 197 2 meet this expectancy and the
population resembles MKE. Subsequent years produce rather more
adult males and one is left with the impression that in 1972
mature males were not sampled at a level commensurate with their
true position in the population.
A striking feature of the histograms for Kruger 1973-1978
(Figures 59 and 60) is the consistent depression in the 16-20
year old age class. The cumulative percentage curves (Figures
61 and 62) also emphasise its position persistently below a
curve which joins the other pivotal points. This is largely an
artefact of the process of determining the number of males
below 7.5,kg tusk weight. The division 44.6% male : 55.4%
female is made arbitrarily across all tusk size/age classes for
convenience. In reality it is unlikely that the male 16-20 year
class would be in this relationship with the 31 >40 female
age classes it corresponds to in tusk weight.
The artefact notwithstanding, its application is uniform
from year to year, and the differences in.male population
structure must represent some variance in cropping focus or in
population structure, or both. This is particularly noticeable
in the first two age classes : 0-5 and 6-10. There also seems
to be a variation of approach toward males in the year classes
21-25, 26-30 and 31-35, over the period covered. With further
time at my disposal satisfactory explanations would undoubtedly
have been obtained. The issue of consequence here is not so
much what is happening in Kruger, but in the demonstration that
the .-ivory crop reveals the form of offtake and that it has not
been even from one year to the next.
Earlier it was pointed out that males in their 20s are
marauders around the edge of stable elephant sanctuaries. The
Kruger National Park is one such area and male elephants




186

persistentlybreak out into surrounding lands (Joubert pers.
comm.) In consequence they are shot. In the Transvaal,
responsibility for this is with the Nature Conservation
Authorities. Figure 63 and Tables 190 and 211 provide evidence
of the age of bulls involved.
Two sets of data on tusk weights from southern Angola and
northernmost Namibia were obtained; one at the border post of
Rundu, the other from a trader in Johannesburg. They are
completely separate, and presented in Figures 64 and 65, and
Tables 191, 192, 212 and 213. The overall proportion of males
in the sample was assumed to be 62.95% male. The two samples
are similar in form and indicate generalised hunting with
selection for males in the 20-30 year classes. In origin this
ivory is comparable with the South-Western circumference data.
The circumferences suggest greater emphasis on younger males
which would be the case if larger tusks had been extracted on
arrival in Hong Kong as suspected. The average weights (males
combined with females) are in accordance with this supposition,
being 8.89 and 8.84 kgs vs. the circumference derived 5.32 kg.
The data in Figures 64 and 65 illustrate not only the selection
of larger males but the corollary : that such animals exist.
Mozambique on the other side of southern Africa to Angola,
provided data which is nearly contemporaneous and presented in
Figure 66 and Tables 193 and 214. Using the same sex ratio in
estimation of the male component of the sample, it is clear
that a different situation existed in Mozambique. The type of
hunting general for meat and ivory was similar. However, the
results indicate a far younger kill with considerable emphasis on
the males still in family units. The most likely explanation for
this is that available mature males formed a smaller proportion
of the elephant population than occurred in Angola/Namibia : in
other words they were insufficient to meet hunt demand. The
average tusk weight of the whole sample was 6.95 kg.




0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55

I

go0-

20 -

AG E

CL AS SE S

P 40-
e
r

0-5 6-10 11-15

26-30 31-35 36-40 41-45 46-50 51-55 >55
C L A S S E-S

16-20 21-25
AG E

FIG. 63

TRANSVAAL NATURE CONSERVATION AUTHORITIES.. MALE
IVORY EXPRESSED A CUMULATIVE FREQUENCY CURVE AND
HISTOGRAM RELATED TO AGE CLASSES., (n = *125)




100 -
90
80
70 -
do 60 -
o-I
AGE CLASSES
50-
e 40-
r
30-
S20
t10-
0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES
FIG. 64 ANGOLA/NAMIBIA. MALE IVORY EXPRESSED AS CUMULATIVE
FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE
CLASSES. (n = 729)




100 -
90-
80 -
70 -
S60 -
2 0 -
50
r
40
e
r
30-
c
e 20
n
t 10
50-
P40-
e
r
30-
S20-
n
t l-

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55
AGE CLASSES

0-5 6-10 11-15 AGE CLASSES16-20 21-25 26-30 51-55
A GE C L S E

FIG. 65

ANGOLA/NAMIBIA BORDER. MALE IVORY EXPRESSED AS
CUMULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED
TO AGE CLASSES. (n = 269)




0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55

1 -1 1-1 1-2 2-2 2-M 13 3513640141-45 46-50 > 1551 ,55

100

I f

90
80

dP60
40

I I

AGE

CLASSES

50-
P 40-
e
r
30-
c
e 20-
n
t"

0-5

AGE

CLASSES

FIG. 66

MOZAMBIQUE. MALE IVORY EXPRESSED AS CUMULATIVE
FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE
CLASSES. (n = 535)




187

In Rhodesia elephants are very largely confined to national
parks or reserves. The major source of ivory is from the
regulation of populations within these areas cropping. This
is done in similar manner to that described from East Africa by
Laws et aZ. 1975. The cropped ivory is augmented by that from
control work -about the periphery of the elephant areas, and
limited sport hunting. This picture is borne out by the data
-from 3, 604 tusks illustrated in Figure 67 and Tables 194 and
215. The data are interpreted on a cropping ratio of 44.6% males
to 55.4% females. The preponderance of very young, and decline
in subsequent age classes is in accordance with East African
cropping data. The artefact of a lowered 16-20 age class is
apparent as in the Kruger data. The exaggerated 21-25 year class
is likely to be the product of peripheral control shooting.
As in South Africa and Rhodesia, Malawi's elephants occur
in national parks, game or forest reserves (Bell -1979). There
is as yet, no case for reducing or stabilising numbers in any
of these sanctuaries. However, as is to be expected, elephants
sally forth from their preserves and damage crops. For this
they are shot. As usual, the majority are males in the 21-30
year classes. Malawi's data are given in Figures 68 and 69
and Tables 195-198, 216 and 217. The estimates of males less
than 7.5 kg tusk weight have been based on an assumed population
sex ratio of 44.6% males : 55.4% females (except in north Malawi
where the whole sample has been treated as male).
In Figure 68 two collections of data were made; one being
tusks registered under a Kasungu National Park code, the other
registered-unaer a Lilongwe code. The sources are, however,
essentially the same, namely. ivory predominantly from the Kasungu
National Park and its environs and, to a much lesser degree, the
Nkotakota Game Reserve. The patterns illustrate the ef f ect of
control shooting on raiding males superimposed upon younger
classes from natural mortality in the national Dark.




1 I 1 1 1 1 1 1 11I1 1I I

0-5 16-10 11-15 16'-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55I

20E-

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55

AGE

CLASSES

50
r
30
c
e2
n
t 1-

AGE

CLASSES

FIG. 67

RHODESIAN AUCTION IVORY. MALE IVORY EXPRESSED
AS CUMULATIVE FREQUENCY CURVE AND HISTOGRAM
RELATED TO AGE CLASSES. (n = 3604)




o C0 0 0D 0 C0 0>
r-4 % OATIPlIfno

r- %D LA -T c
% aAflqpjnwno

o Co
'm -4

Ln
00
E-4 -n
zc '
z
%D
-4
IA
g~I~ I Ir4
Ln
10
LA
04
C7
10
Ln
zm
00
('D
tn
Ln -V rt 0
04 W 0 ) r- 4

(N -4

MALAWI LILONGWE AND KASUNGU NATIONAL PARK. MALE
IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND
HISTOGRAMS RELATED TO AGE CLASSES. Lilongwe n =322;
Kasungu National Park n = 235.

IFIG. 68




188

In Fig. 69 two further collections are illustrated. The
f irst again is evidence of male mortality about the edge of a
sanctuary the Vwaza Marshes. The second stems f ram the
Liwonde National Park, the Mangochi and other southern f orest
reserves. It presents an entirely different picture of heavy
juvenile mortality. The reasons for so different a situation
around the southern sanctuaries is not clear. Nevertheless it
provides a good example of how an ivory shadow can draw attention
to anomalies. The observation has been relayed to the Malawi
authorities for such attention as they may think necessary.
A final analysis of African tusk weights broken down to a
male population index concerns data from Tanzania. These were
obtained by Douglas-Hamilton and Davitz from the Tanzania Ivory
Room in Dar es Salaam and are the individual weights of 44,074
tusks recorded between 1971 and 1977. They have been analysed
by Davitz (1978) and are of some notoriety, for it was the
average weight (4.8 kg) of this large collection of tusks that
led to the belief that each tonne of ivory exported from Africa
represented the deaths of 100 elephants.
The Tanzania Ivory Room tusks (illustrated in Figure 70
and given in Table-218) derive mainly from elephant control,
most of which takes place in the south-east of the country
(Rodgers et al. 1978). This is so intensive that it must approach
random sampling at least in all but the very smallest elephants
and, perhaps, largest males. On this basis the separation of
males with tusks under 7.5 kg from females has been made on an
assumed sex ratio of 44.6% males : 55.4% females. 7he resulting
male population kill illustrates concentration on voting elephants.
However it' fails to show the slope one expects in n al
populations falling progressively with increased age. This
could be the consequence of such selection for size as may occur,
counteracting the natural age distributions. What is rather
more dif f icult to accept is the small representation of males
over 25 years in a sample of 19,658. It is not as though males
of this age are absent from the control areas. Along the eastern




0-5 6-10. 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60

0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 55-60

AGE

CLASSES

P
e 40--
r
30-
c
e 2-
n
t

.AGE

CLASSES

FIG. 70 TANZANIA. MALE IVORY EXPRESSED AS CUMULATIVE FREQUENCY
CURVE AND HISTOGRAN RELATED TO AGE CLASSES. n = 19,658.




189

borders of the Selous Game Reserve many elephants are shot on
control (Rodgers et at. 1978) and in this area there are many
males older than 25. For this the Selous Reserve is much
esteemed by sportsmen as a hunting ground. The prospect is thus
raised that the Ivory Room data represent 'filtered' tusks, the
larger specimens having gone elsewhere. Such a possibility has
some support from the evidence of inconsistency in Tanzania's
internal ivory records (e.g. the manner in which national park
ivory is sold; or the discrepancies between Ivory Room data and
Customs figures see Chapter 4 Volume 1). The trade also
recognises quite substantial amounts of Tanzania ivory coming
out clandestinely through Burundi. Finally the Tanzania data in
Table 168 also suggest a higher average tusk weight for the
country.
The diverse data in this chapter provide a variety of
insights into elephant/human interactions, and the potential
that ivory has for revealing information on conservation practice
or non-practice. However one which calls for particular
consideration concerns natural mortality. The Hong Kong data
indicate that 21% of tusks arriving there come from natural
mortality. Examination of this is the basis of the next chapter.







190

VOLUME 2 (5) NATURAL MORTALITY
The only certainty in an elephant's life is that it will
die. If that happens after 2 years of age, it will leave tusks.
Irrespective of all other factors, natural mortality is a
substantial source of ivory.
How much of a population dies each year will vary widely
according to local conditions. However as a general index to
use on a continental level, I believe 7% to be sufficient for
the purpose of discussion. Laws 1969a has demonstrated the
mode of mortality and I have used his approach to construct my
own mortality Table (Table 219). From this I have postulated
that 42.5% of annual deaths will'be under 2 years of age and
therefore tuskless. The balance will leave tusks varying from
0.23 kg to c.50 kg in weight. The distribution of these is also
included in Table 219. From the age range of deaths and combining
both sexes, the average tusk weight should be 8.06 kg.
Douglas-Hamilton (IUCN Elephant Provisional Estimate 1978)
has estimated that the minimum elephant population of Africa is
1.3 million. If this be true, and a 7% annual death rate
prevails, then 91,000 elephants will die every year on average.
If, of these 42.5% are tuskless, then 52,325 will leave tusks.
Not all elephant carry two tusks. Some females are tuskless and
others are congenitally single tusked or become so through
accident. In view of this it would be incorrect to multiply the
number of tusked deaths by 2 to obtain the number potentially
available. I shall thus follow Rodgers et at. (1978) and use
a factor of 1.88. The 52,325 deaths should therefore produce
98,371 tusks which, at an, average of 8.06 kg each would weigh
792,870 kg.
The foregoing figures may not be precise, but they indicate
the order of what is potentially available through natural
mortality. Were all the tusks collected, they alone would come
close to meeting world ivory demand as it was in 1978.




191

The crux of the matter is finding and collecting the tusks
before they, too, decompose and disappear. The only methodical
attempt to analyse this problem of which I am aware, was that
of Graham and Laws (1971). They describe searching for ivory
from natural mortality in the Murchison (now Kabalega) Falls
National Park, Uganda. They ascertained that only 27.6% of
carcasses from elephants dying annually were being found. As
will.be seen later the recovery rates in Uganda were exceptionally
high. A further unspecified amount was believed to have been
found illicitly and the balance decomposed.
Found ivory is a source of revenue to many national parks
in Africa. Since one of the major constraints upon their
development is shortage of funds, tusks are collected as and
where possible. Rewards are offered for those handed in by
staff, and few if any rangers would leave tusks if they found
them. Thus the proportion of ivory recovered from natural
mortality is an indicator of the efficiency with which they cover
the area under their control. In addition, mortality being a
reciprocal of survival, efficient recovery of ivory would provide
an accurate measure of the age structure of the extant population.
Some ivory recovery rates are presented in Table 220 for
a sample of 9 parks and reserves. Other than in Tanzania's
Manyara National Park, and Kenya's Marsabit National Reserve,
recovery rates are low : below 6% of what becomes available
annually. The two with higher recovery rates are also exceptions
in being of very small size Manyara 80 km 2 and that part of
Marsabit frequented by elephant (other than in wet periods) being
less than 80 km 2. Two of the 9 are montane forest zones -
Mt. Elgon and the Aberdares/Mt. Kenya the rest are 'savanna
woodlands'. That the montane forest recovery rates are similar
to the savanna's is somewhat surprising, as in these at least,
the dense vegetation could be expected to conceal carcasses far
more effectively than in the more open areas.




192

An individual case of considerable detail is available from
the Tsavo East National Park's record of found ivory from
1950-1978 (Table 221). At the outset of this period, through to
c.1970, I believe that Tsavo East contained at least 12,000
elephants a f igure not out of keeping with the estimates that
have been made between 1968 and 1978 (e.g. Laws 1969a to quote
the earliest in this period). If there were 12,000 elephants
and they had an annual mortality rate of c.7%, 57.5% of which
bore ivory, 869 tusks should have become available annually
for collection. The data in Table 221 show that until 1970,
substantially less than this number were recovered. From 1970
onwards overcrowding, drought and subsequently poaching, brought
about a population crash and a complete departure from the
previous mortality patterns.
In Table 222 1 have re-presented tusks found, against an
assumed constant annual output of 869 becoming available for
the "stable" period of 1950-1970 in Tsavo'East. 'On the assumption
that a tusk remains recoverable albeit somewhat weathered for
3 years, I have presented the uncollected surpluses annually and
as running 3 year accumulations. At any one time ivory available
for collection would have been annual production from natural
mortality plus the running cumulative surplus. Thus the success
of recovery is not measurement of a simple proportion of what
becomes available in any one year, but must also take into
account the accumulated component. On this basis I estimated
Tsavols annual success in recovering ivory from natural mortality
at an average of 7.72% of what lay on the ground.
The approach is of course somewhat academic. It takes no
account of the possibility that some tusks may have been recovered
by unauthorised people. However between 1957 when a successful
anti-poaching programme was launched and 1970, there was very
little evidence of poaching in Tsavo East. It is also for
convenience of demonstration that I have assumed both a constant
population and a constant mortality. The population was
definitely not stable during this period. It "acquired" new




193

components of between 2,000 and 4,000 elephants from across the
southern and eastern borders south of the Sabaki river during
my tenure as a Game Warden in the area between 1957 and 1964.
(Advancing human settlement drove them back mainly by
monopolising very local and limited water resources.) In
addition the middle of the period c.1960-1961 was assailed by a
severe drought in which it would be ecologically unlikely that
there was no rise in mortality. Further the period 1957-1970
was characterised by a progressive depletion, of woodland
(described by among others, Glover 1963, Laws 1969a) which is
likely to have had influence on elephant survival, even if it
was not immediately apparent.
The Tsavo data are ground for believing that the recovery
rates in Table 220 are optimistic. The collections from what
becomes available and stays on offer for 3 years or more reduces
actual recoveries to a still lower index of success. It is
worth comparing the structure in tusk size classes of what should
become available from natural mortality, with what was found, for
it may convey reasons for the poor recovery rates.
Using Table 219 as a base, the number of tusks per 1,000 males
and 1,000 females dying are synthesised into a histogram in
Figure 71 and into a Table 223 to illustrate the distribution of
tusk size classes from my postulated natural mortality. The
resulting pattern provides a comparison for the distribution of
found tusk weights in Kenya's Aberdares and Mt. Kenya Park,
Marsabit National Reserve, Tanzania's Manyara Park and finally
in more detail Kenya's Tsavo-- East Park. A loose overlay of the
outline of 'Figure 71 is provided to facilitate comparison with
the other patterns.
Figure 72 and Table 224 present the data from the mountain
parks of Kenya Aberdares and Mt. Kenya for 1959 to 1978. The
pattern deviates from the expected in a dearth in the first tusk
size class and an unexpectedly high proportion in the intermediate
tusk sizes. The classes 7-14.99 kg and 15-29.99 kg are smaller




50-
P
40-
e
r
30
0-
e 20
n
t 10.
0-0.99 1-1.99 2-2.9) 3-3.99 4-4.99 5-5.99 6-6.99 7-14.9915-29.9 >29.99
T u s k we ig h t c 1 a s s e s
FIG. 71 THE EXPECTED PATTERN OF TUSK SIZES AVAILABLE
FROM NATURAL MORTALITY. (AVERAGE WEIGHT
OVERALL 8.06 KG).




ABERDrAE AND Mr. KENYA
P 40- (n = 173)
r
30-
C
e 20-
n
t1
0-0.99 1-1.99 2-2.99 3-3.99 4-4.99 5-5.99 6-6.99 7-14.99 15-29.9 >29.99
Tusk weight classes
50-
MARSABIT
P 40- (n = 54)
e
r
20-
e
n
t 10-
0-0.99 1-1.99 2-2.99 3-3.99 4-4.99 5-5.99 6-6.997-14.9 15-29.9 >29.99
Tusk weight classes
50-
MANYARA
P 40- (n = 341)
e
r
30-
e 20-
n
t 10
0-0.99 1-1.99 2-2.99 3-3.99 4-4.99 5-5.99 -6.99 -14.99 1529.9 >29.99
Tusk .weight classes
FIG. 72 FOUND TUSKS BY WEIGHT IN KGS FROM ABERDARE AND
MT. KENYA (1959-1978); MARSABIT (1962-1969);
MANYARA (1972-1978).




194
than expected probably because of the earlier intermediate
deaths. The average tusk weight of 7.11 is in keeping with early
mortality and a younger than projected population (i.e. it is
less than 8.06 kg.)
Figure 72 and Table 225 present Kenya's Marsabit found ivory,
1962 1969. The pattern in this case is anomalous with a very
high proportion in the large tusk classes, none at all in the
first weight category and erratic results in the intermediate
ranges. The sample size -is small, but nonetheless represents
a high recovery rate 33.5% of the expected (Table 220). The
average tusk weight is 2.29 times higher than the expected. The
pattern for tusks shot on 'control' (Figure 73) over the same
period produces the same picture a population of old animals -
predominantly males. It was this community which produced the
publicised "Mohamed", "Ahmed" and "Abdul" and which has been
known as the haunt of very large (old) tuskers for the past 40
years or so. It would seem to have been a relict population, but
quite how much of an anomaly or oddity has not been apparent until
the evidence of these ivory 'shadows'
Figure 72 and Table 226 present Tanzania's Manyara elephant
- among the best known populations in Africa (Douglas-Hamilton
1972). Once more the absence of the smallest weight class stands
out. The larger than expected intermediate classes represents
a major die off of the population which will receive more
detailed description from Douglas-Hamilton. The absence of old
males is in accordance with what is known of the population, i.e.
they are absent.
Figure 74 and Tables 227 234 present the Tsavo East story
in 8 segments. The first is in the early years of the park when
development was impeded through Kenya's turmoil in the Mau Mau
rebellion. The mortality shadow has a form fairly similar to the
predicted. However the smallest tusk class does not reach
expectations, and the 7-14.99 kg unit is larger than expected.
The average tusk weight is 9.68 kg which is high. There was




50-
P
e 40-
r
30-
C
e 20-
n
t 10
0-0.99 1-1.99 2-2.99 33.99 4-4.99 5-5.99 6-6.99 7-14.99 15-2929>29.99
Weight classes Kg
FIG. 73 MARSABIT NATIONAL RESERVE KENYA.
FROM A SAMPLE OF 49 ELEPHANTS SHOT 'ON CONTROL'
BETWEEN 1962 AND 1966. AVERAGE TUSK WEIGHT
20.99 KG.




p
e 40-
r
30 -
e 20-

10-6.99 1 1-1. 991 2-2.9913 3.9144.99 15-5.99 166.99 17-14.99115-29.991 > 29.991
SWeight classes Kg

0o
un1
i)

Z04 1- 9 2-2.9913-3~.994 9 6.991.9915-2999 >29.991
W e ig ht cl1aa se K g

1964 1969
AV. W. 8.67 KS
(n = 1,769)

1950 1955
AV?. TIr. 9. 68 XG
(n = 469)

1956 1957
AV. Wr'. 10. 50 XG
(n 499)

p
e 20-I
30
p
e 40-
r
30-
C
e 20
S101

1961 1963
AV. WT'. 8.73 IG
(n 821)

S0-0.99'1-1.99'2-2.99'3-3.99'4-4.99 '5-5i99'6-6.99 7-14.9915-299>N-N
SW e igh t c1a s ae a K g

P 1970 1973
e 40- AV, Wr'. 6.86 MG
r (n -8, 572)
30-
c
e 20
io

'0-0.99 1-1.9Y' 2-2.9!?3-3.99'4-4.99 15-5.9916-6.9917-14.
Weight classes Kg

5-29.99 >29.991

0-0.99'1-1.99'2-2.99'3-3.9914-4.99'5-5.996-6.99 '7-14.99'15-29,99'>29.99'
Weight, classes Kg

1974 1976
AV. Wr'. 6.71 KG
(n -3,945)

p
e 40 -
r
30-
c
0 20-
S10-

10-0.99 11-1.99 2-2.9913-3.991 4-4.99 -.9'-.9-4.9 99'2.9

II~ 1 a. g ai e al s K g

1977 1978
AV. Wr'. 4.17 X13
(n = 195)

T I I I I I Ii
0-0.99 1-1.99 2-2.99 3-3.99 4-4.99 5-5.99 6-6.99 7-14.9915-2199)29.99
Weight classes Kg

)9 122.99 -3399 '4-4.99 I 5-*5.9966.99 I7-14.99'15-Z99 >29.99'
W e ig ht clIa9se s K g

50-
e 40-
r

1958 1960
AV. W'. 9.85 IG
(n 225)

304
20d

so-
p
e 40-
r
30-
C
o 20-
S10-

.W e i g h t




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EODE1ETRI_DUUIT4 INGEST_TIME 2014-06-02T20:01:54Z PACKAGE AA00020117_00002
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

THE I o '79 I Parker

PAGE 3

THE I V 0 R Y T R A 0 E Volume 2 Biological Aspects J U N E '79 Parker

PAGE 4

147 VOLUME 2 (1) IVORY -THE POPULATION INDEX The following comments on tusks start with a synopsis of a paper in preparation by ,Laws and Parker. As this will be published shortly with data and detail, I shall forego repetition here. Suffice it that the material was obtained from the tusks of 2,900 elephant culled in Uganda, Kenya and Tanzania during the years 1965-1969. Information on this culling and analysis of other relevant data obtained have been published (e.g. Laws 1966, 1967a,b, 1969a,b, Laws and Parker 1968, Laws, Parker and Johnstone 1975, Parker and Archer 1970) and further papers will be forthcoming over the next decade (e.g. Laws -Development, growth and wear of mandibular teeth in the African elephant Loxodonta africana -is already written). Elephant tusks are modified incisors which, in the African species, are present in both males and females. A 'milk' tusk is present in the late"'":term foetus and at birth which seems to be'resorbed internally and never features as an externally visible tooth. The permanent tusk which replaces it erupts at an average age of 16 months, but may not be visible beyond the lip until several months later. The age at tusk is variable, in 'some cases it may not take place until the animal's third year. Tusks grow continuously throughout life, albeit at declining rates from the middle of the 3rd decade, in the form of a logarithmic spiral or helix. Both tusks and molars exhibit pronounced 'layering' in their structure, which correspond to seasonal changes in the environment, and which (presumably) are related to nutrition. These act as an indicator of growth rates. In the main they confirm the validity of Laws' (1966) earlier age criteria. Aspects of tusk growth are illustrated in Figures 32-35 in which total tusk length, length of that portion lodged in the head, circumference at the lip and weight are respectively plotted against age. Male and female tusks show markedly different growth patterns, though with both, size increases

PAGE 5

250L E N G T 150...; H I N 100C M S 50-. 1 .. 1 1 : ; J t I I I . '. .' : .' : o. I 0: t 0:- I I . .: .. : ; 'J' r '. '.' t I '.. I oS :.: ::.: h ; J ; J :; .. ". : . I I ; . : . . '., =.r: : ." l' !. '. L i j .. ... -. T s ... ..., 1- o. L _:.;1 ":" 1'''. '; : IJ' ... ." "..-. I .. I ... ':' 1 r' $ . : .... L.1:: ..... t :Ir ;f.\-i ' .. : l'l ." : l' 'S' j ; ::. 1 .. :t.. I I J '., :. N. '. . -' . l.. 'or '; I ii' 1 ii-$ 10 20 30 40 AGE IN YEARS I Black = Male Orange = Female 50 60 FIG.32 SeA TTERS RELA T I N G M ALE A ND FEMA L E TUSK LENGTH T O AGE (n ; mal es 705; f e ma les 8 2 7)

PAGE 6

90-' 80-70-60L E N G T50H I N 40-C M S 30-20-10o : I : : I .. . . . . 0 !. ...... . t ... :. .. ... ... :.".: ... ... .. I 20 '. ... .. Black = Male Orange = Female I i 40 50 AGE IN YEARS FIG. 33 Scatters relating length of male and female tusks in the atreolus (n = male 147; female 170) 60

PAGE 7

11 lD W +>-.--... OJ (j) C ::J ,...,. 0 =:OJ I u ,...,. :3 ,...,. R OJ ,...,. ro 0 J C ro U1 OJ J U --.J lD 0 ro ro M OJ ,...,. F ...., ::J ro E :3 lD OJ :3 R ro OJ E OJ ro N N OJ +>-::J C 0.. ...., E ro :3 OJ I ro ,...,. N c U1 A 0 C -. J M 0 c S :3 ...., ro J ro ::J 0 ro 55 50-40-30 20-10o .. I: .. .. .;. . .. . L : -:. -; .. ;'.i. I' .j: t u'i .: l' L" ".: .. : ........ .. . ; .-': -' *' . ............ :. j :. i .. '.' .. : .. : + .. ,." .... I ...... .. :. "......... : +_ ... 1. ... -!--:-.c':' .. .... ",..... ......... ........... ... J....... .... .. 1 J:, .. II'.: .,' f+;f'j,:L .. j f : ': -: i t '!.' 't ,. ..1 .. :: C-L .... ';''' M .. .. ... .. .... .. J.'.,..', ., -i J ,-i .. .. '!' .. .... -:+t&', ..). ,':' .-.............. ....-+;'-oIl .. -:.,:. ... .. ,t""4 : : 1"" -ii, I : I .. .. Black =: Male Orange = Female 1 10 20 30 40 50 AGE IN YEARS 1 60

PAGE 8

W E I G H T I N K G S 7 5 20 10 o Black = Male Orange = Female . : .: -!-' ... ... 30 AGE IN YEARS r t' 40 Fig. 35 Scatters r elating male and female tusk weights to age (N = male 1862, female 2733)

PAGE 9

148 with age until the last decade of life when -in some individuals at least -wear and breakage exceeds replacement through growth. The incremental layers have permitted calculation of potential cumulative increment in length and weight. These are 555 cm (c.18 oft) and 80.8 kg (178 lbs) and 521 cm (c.17 ft) and 21.7 kg (47.8 lbs) at 60 years in males and females respectively. Actual lengths and weights at 60 years average 250 cm (c.8 ft) and 61.0 kg (134 lbs) in the male and 160 cm (c.5 ft) and 7.$ kg (165lbs) in the female. Thus male elephant in East Africa attain only 75.5% of their tusk weight potential, and females only 34.60%. The loss is caused by breakage and wear throughout the elephants' lives. The differential loss between the sexes is almost certainly the consequence of the female tusk being a smaller tool, relative to the size of the animal. The potential weight of the male tusk (80.8 kg) is 1.5% of the male's potential total body weight at 60 (5314 kg -Laws et al. 1975), whereas the potential of the female tusk is 0.7% of the female potential total body weight (2986 kg -Laws et al. 0 01975). While the mechanics can no doubt be described elegantly in engineering terms, the female can at all times exert a proportionately greater stress (body weight : potential tusk weight) than males upon tusks of similar lengths (greater weight on a thinner lever). The differences between tusks of the sexes that are so marked in Figs. 32-35 are easily discernible at a glance, for the tusks also differ in overall shape. Almost all male tusks within the normal range of female tusk weights are pronouncedly conical from the base. Even if the tooth beyond the lip is broken, the taper is still noticeable in that section of the tusk that was within the elephant's head (the alveolus). Female tusks appear cylindrical -their taper in the first half of the tusk being too gradual to perceive easily. Male tusks only become cylindrical when well beyond the limits of female growth. The relationship of circumference, length, weight and shape permit male tusks to be separated from females' with ease:

PAGE 10

a fact of which both hunter and trader have been aware for centuries. The dynamics of elephant tusk growth are such that given tusks' weights alone, it is possible to determine those which 149 are unlikely 'to be female, i.e. all over kg. Of those over 1.5 kg it is possible to ascribe ages to the bearing elephants at the time of death. If, in addition to the weight, tusks can be seen, their shape and weight at any size permit both sexing and aging of the elephants which bore them. Similarly, on their own, total lengths, circumferences and or the length of tusk in the alveolus, are all useful indices of age. With shape or in combination with anyone other of the parameters illustrated, they will determine sex and age. Precision increases if combinations are made with two or more of the other parameters. A fifth measurement on which Laws and Parker will publish data in due course is the volume of the tooth pulp (bollow' to the trade and 'nerve' to the hunter). Up to 35 years of'age this increases in volume up to 5.85 litres in the male, and 0.82 litres in the female tusk. After this it regresses down to 0.3 litres in very old females, being replaced with ivory. The degree of male regression is not yet confirmed in view of limited records. The replacement of pulp with ivory must partially compensate for any slowing down in incremental length growth, and for external wear and breakage. The salient point is that tusk measurement can provide an objective record of human/elephant interactions. If data on weight, lengths and circumferences and can be collected they will impart a great deal about the manner in 'which hunting is affecting populations, and also considerable detail on elephant status. The data from which our descriptions derive are exclusively East African. It has yet to be shown that they are valid for other African regions. Subjectively, I can recognise 4 types of ivory: East African, South-Western, Sahelian and Cyclotiform.

PAGE 11

150 Throughout this survey, when seeing ivory in Europe or Hong Kong, I estimated.its origin before asking the owner from whence he obtained it. In all cases I ascribed it to the correct region. Experienced traders are also able to make similar distinctions as to ivory's origin, particularly when it is in large lots so that common features stand out. East African ivory occurs the whole length of the Indian Ocean seaboard and inland to include Somalia, southern Ethiopia, southern Sudan east of the Nile, Uganda east of the western rift, Kenya, Tanzania, eastern Zambia, Malawi (with reservations), eastern Rhodesia, Mozambique and the Kruger Park of South Africa. In terms this region produces East African "Soft" or "white" ivory. In older animals it is long, and the tusks generally bow and curve in several planes: it is this ivory which grows in a noticeably logarithmic helix. The twist on a large male tusk is very obvious. South-Western ivory is from southern Angola, western Zambia, Namibia, Botswana and western Rhodesia. In trade terms this is South African "Soft" or "white" (andnot so highly prized as East African white). Subjectively I believe these elephants to be substantially bigger than the East African form. However their tusks seem much shorter, with a bigger proportion in the alveolus, and seldom attain the same weight. It is far more heavily "shalced" than most East African ivory. Their bluntness and lack of length suggest very much higher breakage rates. South-Western tusks seldom attain sufficient length to show curvature in more than one. plane which, however, is very pronounced. The absence of an obvious twist in the grain suggests that these tusks grow more in a logarithmic spiral within one plane and less in helical form. As the name implies, Sahelian ivory derives from the sub Saharan zone from Senegal to.the Nile. It is more similar to the South-Western, than East African, but it does get much larger that the former. Again there appears to be a very much higher

PAGE 12

151 breakage rate than in East Africa -particularly in the female. A common phenomena in Sahelian ivory is a very fine 'herringbone' shake, which is so dense as to impart a grey appearance to the tusk. This does occur in South-Western tusks, but not to the same degree. The East African, South-Western and Sahelian elephants are all "bush" or "savanna" types, i.e. Loxodonta a. africana. The fourth is the forest elephant of "the equatorial forests in the Zaire basin the West Africa: Loxodonta a. cycZotis. It is the least known of all the African elephant forms." Suffice it that it is much smaller than any of the savanna elephants. The tusks are equally distinct, being slender (particularly in the female), very straight with no twist; if there is curvature it is slight and in one plane only. The proportion of a cycZotis tusk in the alveolus is far less than an africana tusk of similar weight (Fig. 36). In both sexes the tusks point downwards rather than forwards as in the case of africana. Further, the male tusks show far less taper than in the savanna forms, although it is still quite discernible. CycZotis tusks have very few "shakes", are often black or deep mahogany in colour (presumably from.vegetable staining) and where use removes the staining, show a characteristic yellow patina. To the trade they are "hard" or "yellow" ivory. No clear-cut "division separates the small forest cycZotis from savanna forms of elephant. Rather, all around the periphery of the rain forests there are clines in which elephants are intermediate, showing features of both forest and bush types. It is in this "hybrid" zone that large individuals occur, carrying cyclotiform ivory in"all aspects but its large size. Many of the biggest tusks I have seen, have been straight, without shakes, and with the characteristic black colouration of the forests, or forest edges. Traders refer to it as "semi-hard" or "semi-soft".

PAGE 13

FIG. 36 A B TRACINGS FROM A PHOTOGRAPH TO ILLUSTRATE THE DIFFERENCE IN SHAPE AND DEPTH OF TUSK IN THE ALVEOLUS BETv1EEN' TV!O 20 KG MALE TUSKS A. FROM oxodonda africana africana AND B. FROM L. a. cycZotis. THE UNSHADED AREA IS THAT WITHIN THE ALVEOLUS.

PAGE 14

152 Figure 37 illustrates the distribution of the four 'gross forms of ivory. It must be stressed that they are based on broad generalisation. The characteristics are more apparent when a number of tusks from one region are assembled in proximity to a group from another. The individual tusks can show wide variation. An example of intra-family similarities and inter-family differences is given in Figure 38 from two herds taken within the same East African population. Objective data to support the contention of four ivory regions are few. Elder (1970) examined some aspects of tusk morphometry from south-Western, East African and Cyclotiform ivory. He found no difference in the specific gravity of the three forms either between sexes or between regions. He did comment on the greater depth of tusk in the alveolus of the South-Western elephant which suggests a bigger animal. He mentioned a greater circumference at lip than in either East African or Cyclotiform tusks, but this was not allied to age. If his Botswana sample derived from older elephants, the apparently greater circumference would be explained. I have seen a great deal of Botswana ivory and, albeit subjectively, have never gained the impression.that it was noticeably thicker than East African tusks. Extensive data have been collected on western Rhodesia ivory which were not available to this study and in due course their analysis will settle the matter. The records of William Finaughty who hunted the SouthWestern elephants extensively between 1864 and 1875, indicate the heaviest tusks he got were two which weighed 56.7 kg each. While these are big, they are not outstanding by East African standards, but they were obviously.exceptional to him. Stevenson-Hamilton (1947), an accurate observer, wrote "I am inclined to think that South African ivory can never have equalled that carried by elephants in equatorial Africa3 where even to-day tusks 9f well over 100 lbs (45 kg) are not uncommon."

PAGE 15

... m TON OF THE DISTRIBUTION MMATIC FIG. 37 DIAGRIL TVPES'IN OF FOUR IVORY ...

PAGE 16

FIG. 38 TUSKS FROM TWO FAMILY UNITS IN THE SAME EAST AFRICAN POPULATION TO ILLUSTRATE (i) INTRA-FAMILY SIMILARITY AND (ii) INTER-FAMILY DIFFERENCES. (Photos : R.M. Laws)

PAGE 17

On such as this and my own I am inclined to that weight in the South-Western elephant tusk is 153 lower at age than in East Africa, and that if we apply criteria developed from the Laws and Parker material, we would underrather than over-age these elephants. Sherry (1978) indicated asymptotic weights of 5.6 kg and 6.9 kg for female tusks at 60 years of age in two south-east Rhodesian elephant populations. These are substantially less than the 7.5 kg we have postulated from East African data. However, we have evidence of differential breakage and wear between our 5 East African populations, and these low Rhodesian estimates may be rather more extreme expressions of the same phenomenon. I have many reservations about applying East African tusk criteria on a pan-African scale. However they are the only yardsticks available to this survey. In the following chapters I shall apply them to material from the ivory trade with the caution that the results.may need some re-adjustment (particularly those relating to Cyclotiform ivory) in the light of future, more precise data.

PAGE 19

154 VOLUME 2 (2) IVORY -THE STANDING CROP An elephant population's standing crop of ivory (the total weight of ivory carried) is obviously the product of its age and sex structure. Such structures vary widely, and appreciation of this is necessary before ivory entering the trade can be analysed. The elephant culling recorded by Laws et aZ. 1975 took random samples that were representative of population age and sex structure. Five populations -2 in Uganda, 1 in Kenya and 2 in Tanzania -were sampled in the manner described. Below I present 3 examples: 1) from Uganda (MFPS) is representative of an 'old' population, 2) from Tanzania (MKE) representative of a 'young' unprotected population and 3) being the sum of 1 and 2 and the 3 other populations intermediate between them to give a 'generalised' population. The age structures of the MFPS population are presented as 5 year classes in Figure 39 and Table 162, males and females separately and combined. The sex ratio of the sample was 42:58 males to females. In the same figure and table are the distributions of ivory weight in 5 year age classes to produce 'ivory shadows' of male, female and combined sectors of the population. Of the gross ivory crop, 64.8% was held by males and 35.2% by females -an exaggerated reversal of the sex ratio. The average male tusk was 10.08 kg and 68.7% of male ivory wa s in animals of 26 years or older. The average female tusk weight was 3.94 kg and 72.7% of female ivory was in elephants of 26 years or older. The .population's average tusk weight was 6 51 kg and 70% of its ivory held by animals of 26 years or older. The MFPS population had a very low recruitment rate (insufficient to maintain. the population) with correspondingly few young animals. Its range at the time of the culling had been a sanctuary for 52 years which reduced human inflicted mortality and permitted many individuals to approach their

PAGE 20

50 o c::: .-140 rU .j.J o .j.J .j.J 3 .-I rU .j.J 0 .j.J .... 0 .j.J c::: 0 0 I-< (l) 050 c::: .-I rU u 0 .j.J ,,-, 0 .j.J c::: (l) 0 I-< (l) f!< 050 C A .-I C III 40 .j.J 0 .j.J .... 30 0 .j.J 20 c olD 1.1 QJ Po AGE C LAS S E S 50 B .-I D :l 40 0 .j.J .... 0 .j.J c::: (l) 0 ,.., I' 114 o-s o-s AGE 50 .-I F rU .j.J 0 .j.J .... 0 .j.J c::: (l) 010 (l) 114 o-s AGE FIG. 39 AN ILLUSTRATI O N OF THE DISTRIBUTION O F AGES IN 5 YR CLASSES AND THEIR CORRESPONDING PR O P O RTIONS O F THE STAND ING IVORY CRO P IN AN OLD ELEPHANT POPULATION, MFPS. A = MALES. B = MALE IVORY. C = FEMALES. D = FEMALE IVOR Y E = SE XES COMBINED. F = IVORY SE XES C OMBINED

PAGE 21

maximum potential longevity. These factors produced a very 'old' population that is unlikely to have many (if any) counterparts .in Africa. 155 The age structures of the MKE population are presented in Fig. 40 and 163, males and females both separately and combined. The sex ratio of the sample was 49:51 males to females. The corresponding ivory shadows for both sexes separately and combined are also illustrated. Of the gross standing crop of ivory, 59.9% was held by males and 40.1% by females. The average male tusk weight was 4.63 kg and 25.7% of male ivory was held in animals of 26 years or over. The average female tusk weight was 2.99 kg and 47.2% of female ivory held in elephants of 26 years or more . The population's average tusk weight was 3.81 kg and only 34.4% of the population's ivory was held in animals of 26 years or over. The MKE population was characterised by very few old elephants -none at all in the last decade of life and only 2.6% over 41 years (cf. 11.8% in MFPS). Constant hunting and harassment were the cause of this (Parker and Archer 1970) and mortality was such that only 5.4% of males were 26 or older and none reached 41. Recruitment was high, producing a correspondingly large se.gment of immatures. As such MKE was a particularly 'young' population and it is doubtful whether there would be younger groups elsewhere, other than in artificial circumstances (e.g. as in the Akagera National Park of Rwanda where the whole population is of translocated immatures). The combined age structures and ivory shadows of all 5 East African samples are presented in Figure 41 and Table 164. The sex ratio in this sample is 45:55 males to females. Of the standing crop of ivory 63.4% was held by males and 36.6% by females. The average male tusk weighed 7.26 kg and 55.1% of male ivory was in elephants of 26 years or older. The average female tusk weighed 3.36 kg with 65.4% being held in animals of 26 years or more. The population tusk averaged 5.09 kg and

PAGE 22

, 0 50 c: ..... III ao .j.J 0 .j.J 30 .... 0 .j.J 20 c: QI U 10 QI a.. .j.J :so ..... III a D 0 J.J 30 0 .j.J c: 20 IV U 10 QI p, g 50 ..... Z 110 o .j.J .... 30 o 20 OJ u 1-l1O OJ a.. A B 0-5 AGE E 0 50 c: .-I III aD .j.J 0 .j.J 30 .... 0 .j.J c: QI u I-l QI a.. 50 .-! a D Cd .j.J 0 .j.J .... 0 .j.J c: OJ u I-< OJ Il< .J 50 ) ';;lao .j.J 0 .j.J 3 0 .... o .j.J20 c: OJ u 10 I-l OJ a.. C r eo .. D 0-5 C LAS S E S F FIG. 40 AN ILLUSTRATION OF THE DISTRIBUTION O F AGES IN 5 YEAR CLASSES AND THEIR CORRESPONDING PROPORTIONS OF T H E STANDING IVORY CROP IN A YOUNG ELEPHANT POPULATION MKE. A = MALES. B = MALE IVOR Y C = FEMALES. D = IVOR Y E = SEXES COMBINE D. F = IVORY SEXES

PAGE 23

50 sa 0 !::: ao It! .j.J .3 JO .. --..,. 50 0 C C 40 It! -g 30 .j.J -g 20 Q) u 10 Q) p.. AGE C LAS S E S sO B .j.J D ao It! .j.J o 30 *' -g2Q Q) U 10 Q) p.. o-s AGE 50 E .j.J F .... It! -g 30 *' 'l:l 20 Q) u 10 Q) p.. o-s C LAS S E S FIG. 41 AN ILLUSTRATION OF THE DISTRIBUTION OF AGES IN 5 YEAR CLASSES AND THEIR CORRESPONDING PROPORTIONS OF THE STANDING IVORY CROP IN A GENE PAL POPULATION COMBINED EAST AFRICAN DATA FROM 5 POPULATTONS. A = MALES. B = MALE IVORY. C = FEMALES. D = FEMALE IVORY. E = SEXES Cm-1BINED. F = IVORY SEXES Cm-1BINED.

PAGE 24

58.9% of its ivory is held in animals of 26 or older. The distribution of ivory within the population is illustrated in Figure 42 and Table i65. 156 The pattern presented in Figure 41 is the standard against which I shall compare data obtained from the ivory trade in subsequent chapters.

PAGE 25

56-60 51-55 46-50 til 41-45 QJ til 36-40 til ro 31-35 .--l u 26-30 21-25 QJ tJl 16-20 o::t: 11-15 6-10 0-5 10% MAL E S 63.3% 0 10% F E MAL E S 36.7% 20% FIG. 42 DISTRIBUTION OF ALL IVORY IN AGE CLASSES FOR 5 EAST AFRICAN POPULATIONS COMBINED.

PAGE 27

157 VOLUME 2 (3) HUNTING EFFECTS Elephants are killed for a variety of reasons which involve different techniques and have different influences upon populations. Perhaps the most widespread reason for hunting elephants is still for meat. Even if tuskless they.would still be taken. Thus in Malawi (Bell 1979), Zambia, Mozambique, Zaire and parts of the Sudan, there are usually meat smoking racks and other evidence of their use as food, in close proximity to carcasses of hunted elephants. Even in Angola where ivory buys ammunition, elephants are equally important for the meat that they provide for guerilla troops (N. Steyn, pers. comm.). In most subsistence situations there is insufficient manpower available to make optimum use of all the meat an elephant provides. Cutting it up for smoking and drying is time consuming and laborious. Even an adolescent specimen provides a surplus, and only animals killed relatively close to large settlements can be fully utilised. Thus with subsistence hunting for meat there is little incentive to select particularly big elephants. If there is bias in what is hunted, it probably arises more from the elephant's social organisation than any other factor. Females and young in family units show considerable cohesion and are more dangerous to the lone hunter with a primitive weapon. Lone males, in contrast, are among the easiest to approach of.all game. However while this bias will possibly result in more adult males being selected and hunted when using spears, bows and arrows, muskets and rifles, it would not apply where the more economic and less dangerous techniques of trapping are employed. Indeed here the bias would work the other way, for it will be the weaker, less experienced young individuals who would be more likely to succumb to a drop spear wound or less able to break snares. Overall, the effect of subsistence hunting is likely

PAGE 28

158 to affect all ages and sexes of the population. To my knowledge it has not been seriously suggested as a cause of elephant extermination. At an elementary level it is difficult to separate subsistence from commercial hunting. For centuries the Watta of eastern Kenya have taken elephants for both meat and ivory. Ivory was the key to their social bond with the Galla pastoralists, who provided protection in a somewhat overlord/ serf relationship. At this level they were in truth subsistence ivory traders and in having a special use for ivory, were appreciative of big tusks. A great deal of their hu'nting, however, was opportunist: lying in wait at a waterhole, they would ambush any elephants which came to drink. The volley of arrows loosed would be aimed at the nearest to the bowmen and this technique would probably have swamped the better hunters' efforts to be selective of large male elephant. As the 20th century unfolded, so pressures to enter the monetary economy will have influenced the traditional 'subsistence' ivory hunter. Killing for meat will have receded in importance and for ivory have made corresponding gains. With the decline in the value of ivory relative to rising commodity prices (Volume 1 Chapter 7) the need to kill more for the same return would have been augmented by human population growth and further demand upon elephants. Selectivity will almost certainly have declined. Overall the commercial hunting will have reduced life expectancies of elephants, the degree depending on hunting The direction will have been toward an MKE population of low age and high fecundity. The most extreme expression will have been that of the' Somalis in Kenya in the present decade. 'Control' shooting is a widely used misnomer applied to the destruction of elephants in defense of human interests. At its more simple level it entails shooting those which venture out of sanctuaries and damage crops (usually males in their late teens and twenties). This is the most common form of control where

PAGE 29

there is a stable -in area -sanctuary surrounded by human settlement. 159 A different complexion is taken on by control work where humans occur as islands in a sea of elephants. In view of human population increase, politics and economics the boundaries of these islands are seldom constant. In this century they have, for the most part, been expanding, but on occasion have contracted -e.g. with the 'villagisation' of peasants in Tanzania (Rodgers, Lobo and Mupunda, 1978). However, whether expanding or contracting, there is almost constant elephant "attrition about their edges. Thus in south-eastern Tanzania over 100,000 elephants have been shot since 1930, and c.2,500 are still shot annually (Rodgers et 1978). Solutions have ranged from a limited policy of shooting one or two members of a herd in the vicinity of cultivation to 'teach' the survivors to leave it alone, to grandiose quasi-military strategles to drive the elephants 'back' and to establish frontiers which they should not cross. "In other words a system of defensive fronts has been instituted to ward off the encroachments of e (Anon 1925). This involved shooting virtually every elephant on the wrong side of the 'frontiers'. At the same time there was a persistent desire to leave big tuskers for sportsmen. were to spare with tusks" (Anon 1926). The records of the Uganda Game Department from 1925-1959 provide the most cohesive documentation of the effect intense and prolonged human activity had upon elephants. A synopsis of the relevant data is presented in Table 166. It includes the yearly number of elephants shot on 'control' (males and females); their average tusk weight; the number of" elephants shot on licence and their average tusk weight; the number of tusks found -both in the battle grounds and elsewhere, with their average weights; the total weight of ivory produced and overall Uganda average tusk weight; the contribution tusks of less than 4.5 kg made to

PAGE 30

160 the overall number of tusks; the weight of ivory claimed by the Game Department to have been produced compared with Customs records of export; and lastly a comparison of the number of rifle permits and game licences issued over a 9 year period. To gain'confidence in the record let us first compare Game Department stated production with Customs records of export for the 19 years 1939-1957. They should marry approximately as Uganda had no internal ivory uses of consequence. The Game Department 'produced' 434.5 tons when Customs claimed exports of 434.9 tons: an insignificant difference of 0.09%1 With this backing it is reasonable to accept Game Department'volumes of ivory as accurate. The first set of columns in Table 166 concerns elephants killed on control. These rose from an average of 705 a year 1925-1929, to a peak of 1351 a year 1935-1939. From this high point the 5 year average declined to 667 annually between 1950 and 1954. From then it rose steeply back past 1027 a year in 1955-1959 and, though there is a gap in the record as the Game Department stopped publishing reliable data after 1959, the trend proceeded to the large-scale cropping described by Laws et aZ. 1975 and culminated in the wholesale slaughter of survivors under the dictatorship of Idi Amin. It was all very well to decree from behind a desk that control shooting should concentrate on 'unwarrantable bulls'. It was quite another matter putting it into practice. Shooting elephants in dense Pennisetum c.4 metres (12 ft) high was unhealthy for the hunter. Grass this tall hides elephants as effectively as a stand of wheat conceals rats. Invariably they remain invisible until at very close quarters. Much shooting was also done in thick forest -which was equally uncomfortable. It was extremely difficult to determine the sharp from the blunt end of an elephant, let alone whether it was an 'unwarrantable male'. The de facto system was shoot what you can. Thus control kills were virtually random -except that small immatures tended

PAGE 31

to be under-represented. Jhey were more concealed by the vegetation and bodies of other elephants. 161 In the circumstances, the offtake should have been similar to a cross section as illustrated in the preceding chapter -minus most'animals under c.5 years of age. Consequently the expected tusk weight should have been slightly above the mean population weight of 5.09 kg (Table 164). At first glance this is indeed the case -average control tusk weight remained remarkably constant between 5.14 and 7.61 kg over the 35 years. However this constancy is at complete variance with the sex ratios given. The average tusk weights are reasonable given an expected 45:55 male to female sex ratio, minus the youngest classes of tusk bearers. They are not reasonable if the sex ratio is reversed and exaggerated to an average of 65:35 males to females as in the case in Table 166. Further examination shows that the claimed sex ratios varied very widely -between 52:48 and 84:16 -always in favour of the male. Such variation clashes with the constancy of average tusk weights. Either the sex ratios or the average tusk weights are wrong. The manner of control hunting, the known range of sex ratios in elephant populations, and the independent agreement with Customs statistics indicate that the weights are right and the sex ratios wrong. From my own experience I know that many game scouts and not a few game wardens cannot sex an elephant accurately. The myth of the dominant "herd bull" dies hard and family unit matriarchs, which are often substantially bigger than other members are claimed as "lords" of the herds. This misidentification of sex has also been reported from Tanzania (Nicholson pers. comm. and Rodgers et al. 1978). Added to these ge,nuine misidentifications, was a desire to claim males shot. In those days it was the epitomy of bad game-keeping to shoot females! The constancy of average weights in animals shot on control run somewhat contrary to expectations. Theoretically continuous harrying should have reduced life expectancy towards an MKE level (Table 163); 35 years of control should surely have

PAGE 32

162 produced a declining tusk weight? In fact the reverse is true, for it only rose to over 7 kg at the end of the period. The elephant shot on licence bear out the expectation of a declining tusk size. This was the subject of concern in the Game Department shortly after the second World War (Pitman pers. cornrn.). The records were analysed by Brooks and Buss (1962) who documented that there had been a decline in tusk weight. This was inversely related to the number of elephants taken on licence: the more elephants taken, the lower the average weight of tusk. The assumption is that the hunting demand exceeded the population's supply of 'warrantable' bulls. (Note: at the time it was not appreciated that control shooting the 'unwarrantable' males today, removed the P!ospect of 'warrantable' males tomorrow.) The assumption of over-hunting was tackled in 1952 by raising the price of elephant licences. This produced a sharp drop in the number bought and, as expected, a rise in the average weight of tusk taken on licence. However, even though the number of male elephant taken on licence dropped to below the 1938, 39, 40 and 41 figures, the average tusk weight taken did not return to the previous level. From this one might deduce that the consequence of all forms of hunting had so reduced male life expectancy that the average tusk weights of the 1930s were no longer attainable: the population had become 'younger'. The average weights of 'found' tusks were higher than those obtained on control. However they show no obvious trends. The combined Uganda average tusk weights are, from the contribution of sport hunters, higher than those of either control or found tusks. In view of this contribution, and the constancy of control weights it would have seemed reasonable if the overall tusk average had declined in sympathy with the fall in sportsmen's tusk size. In fact this didn't happen. Overall average weights increased. This could have corne about if the contribution of the lightest tusk classes declined at a greater rate than the contribution of sportsmen's results. This in fact occurred. In the mid 1930s tusks of less than 4.5 kg constituted

PAGE 33

163 over 40% (over 51% in 1934) of all tusks recorded. After 1948 the proportion never rose above 31%. A logical assumption would be that the control officers became more selective. The truth is more likely that there were fewer small elephants to shoot, and I offer the following rationale. Figure 43 presents graphs of 4 indices: 1) average tusk weight taken on licence, 2) number of elephants killed on licence, 3) average tusk weight from all sources and 4) the contribution that all tusks under 4.54 kg made to the total number of tusks. The year 1939 is taken as 100 for all 4 indices (see also Table 167) These illustrate graphically what was stated in the preceding paragraphs. Tusk weights from licensed hunting declined as the number of elephants taken went up. The big drop in such elephants killed after 1952 produced a rise in average tusk weight, but it never rose above an index level of 98. In the 1930s the index had reached 122. The overall average tusk weight rose more or less continuously from 1933 to.1957 in a reverse trend to the decline in licensed tusk weight. The cause of this is clear in the diminishing contribution made by tusks of less than 4.54 kg. The decline in licensed tusk size is easy to understand: hunting exceeded the rate of replacement i n the o lder age classes. However, added to the control killing this might have brought compensating mechanisms into play. Natural mortality should have declined, and faster growth, increased fecundity and earlier puberty should have tilted Uganda's elephant toward a MKE situation. If this happened, the contribution of under 4.54 kg tusks should have increased. It didn't because throughout the period 1925-1959, the elephants were losing not only numbers but range. Brooks and Buss (1962) documented the fact that at the outset of this span, elephants occurred over 70% of Uganda's land area. By 1959 they occurred in only 11%. At the same time as

PAGE 34

23 220 21 20 19 18 17 160 15 140 130 120 110 100 90 80 70 60 50 40 30 20 10 o FIG. 43 TRENDS IN 4 ELEPHANT POPULATION PARAMETERS IN UGANDA 1926-1957. 1939 = 100 .OF Average .tusk weight of elephant o shot on licence. elephant shot on licence. Average tusk weight all sources. Tusks under .4.54 kg as proportion of annual total number of tusks. \ \ \ \ : A ,.I \ \ \ I L_/'" 1930 I I .I I 1940 ,.... I I I \ I v ..... 1950 ;"" I \ I' ,_oJ I .

PAGE 35

164 the shooting, elephants were being displaced by expanding humans. The situation changed from one in which humans were islands in a sea of elephants, to its converse. As the humans increased and their islands coalesced, elephant distribution became discontinuous and fragmented -islands constantly diminishing in size. It is 'here that Laws et aLe (1975) work on MFPS elephants is of particular relevance for it took place in one of the declining islands. They described the progressive retraction of the North Bunyoro elephant range from 1946-1968 at the rate of a mile inwards a year (Laws et aLe Ope cit. p.18). They summarised the situation thus (p.260) "ModeLs for the MFPS popuLation have been constructed for 1946 (or earLier)3 19663 and 1971. The aduLt mortaLity rates appear not to have aLtered significantLY3 but caLf mortaLity has apparentLy increased by more than 50 per cent (perhaps as much as 100 per cent) since 1946. Together with deferred maturity and reduced fecundity (through overcrowding)3 this has Led to a massive decline in recruitment. It is estimated that relative nataLity may have fallen to 34 per cent of the 1946 level and relative recruitment at four years to 25 per cent of the earlier leveL. PopuLation size is estimated to have decreased from 22 000 in 1946 to 9400 in 1966 and to 7900 in 19713 that is by 64 per cent. Less than half of this decrease was due to control shooting and sport hunting. Natural reguLatory mechanisms alone were apparentLy responsibLe for a decrease in the residual population of about 51 per cent up to 1971." I suggest that this is what happened wherever elephants became diminishing islands in Uganda. The severity of effect would of course have been related to the pace at which range was depleted. A further factor worthy of note is that a declining area in Which to live would have caused an apparently constant hunting effort to become progressively more severe in effect. The rate of range loss was c.2% per annum. In the years 1950-54 an annual average of 1,110 elephants were being taken, as against an average of 1,540 p.a. in 1930-34. However while the later elephant crop'was 72% of the earlier, it was being taken on only 60% of the former range. The actual hunting intensity had risen -not declined.

PAGE 36

165 The figures in Table 166 are of course minimal, foi they do not take note of the tusks poached successfully, nor account of all the deaths by wounding which were considerable (Laws et ale 1975). The discrepancy between the number of game licences and firearms permits issued (Table 166) may be an indication of the extent to which unlawful hunting took place. The Game Department knew that a minimum of 45,583 elephants died in the 34 years and the real figure may well be double this amount. The Game record is no more than an index of what happened, yet it illustrates several important points. The first is that the tusk weights of elephant taken on licence are useful indices of declines in average age -for as long as the licensees seek the biggest elephants. The second is that an increase in overall tusk weight may herald a worsening situation for the elephants rather than. an improvement. This is likely to be the case if the elephants occupy a rapidly contracting range. One Uganda experience deserves some prominence. In the Budongo Forest the removal of elephants was decreed in the interests of forestry (Laws et ale 1975). It was never achieved despite intense effort and illustrates that in thick vegetation elephants are difficult to exterminate. In contrast to Uganda, the position reported by Rogers et ale (1978) in Tanzania accords with the expected situation of a stable or slightly declining tusk size. Here the elephants have not been deprived of range; indeed they may have temporarily gained ground. The attrition around the peripheries of the human islands has produced an MKE effect and not an MFPS situation. The large scale control shooting reduces numbers in the vicinity of human settlement; biological compensating mechanisms corne into play, more food is available to the individual through reduced density; there is lowered natural mortality, faster growth, earlier puberty and increased fecundity. In short young, highly productive populations have been induced.

PAGE 37

VOLUME 2 (4) INFORMATION FROM TUSKS IN TRADE (A) There have been a number of competent African elephant studies in the past 3 decades -e.g. Caughley 1973, DouglasHamilton 1972, Laws 1969a, Perry"1953, Sherry 166 and Smuts 1975. They have produced a broad comprehension of the species' biology. However through expense, the time necessary to undertake them and a wide variety of human factors, studies of this quality are of necessity local. They do not reveal much about the continental status of elephant. What they have shown is that, short of disciplined study, reliance on local lore and guesses about the abundance of elephants is generally not worthwhile. Efforts to overcome this through aerial counting have been made, but on the basis of expanse and consequent expense, the technique has limited application. Added to this it is notoriously crude. To census elephants in the 7,000,000 km2 (2.7 million square miles) Elephant Study Interim Report) of Africa where they stili occur is, on the basis of logistics alone, an impossible task. In theory, tusk parameters should provide a relatively cheap, practical and precise index of overall population trends. The idea was approached by Quick (1963) and the principle determining changes in age structure from characteristics of trophies secured by hunting (e.g. the number of tines on deer antlers etc.) is not new. The question at the outset of this survey was merely whether access to large samples of tusks could be obtained. Ideally, examination should be at the points of the trade's greatest constriction -the marts of Hong Kong and Belgium. Of the two, Hong Kong was the obvious choice as it receives ivory from a far wider base than Antwerp (refer Fig. 5 Volume 1). The Hong Kong traders were approached and, with one exception, gave me access to their stocks of tusks. The only stipulation was that I maintained normal business confidences -a courtesy I was happy to comply with. A sample of 18,453

PAGE 38

167 tusks weighing gross 185,942.81 kg, averaging 10.08 kg and related to countries of origin (Table 168) was obtained through examination of invoices. A separate and probably at least partially overlapping sample of 22,260 tusk weights, grossing 214,829.79 kg and averaging 9.65 kg was obtained from official import documents covering the period June-November 1978. Examination of all or part of the stocks of 10 traders selected at random provided data on 5,132 tusks. The following information from each tusk was recorded: circumference at lip, sex, ivory-type region or, if possible, country of origin and an estimate of whether it was from elephants killed by man 'or which had died from natural cause. The sheer volume of ivory to be examined and the disruption of business entailed in moving it about, limited me to measuring one parameter. I chose circumference at lip because it was a) easy, b) uninfluenced by individual and regional patterns of wear and c) uninfluenced by breakage in trade -which often affects the first part of the tusk's hollow section. Judgment as to whether a tusk had come from man-induced or natural mortality derived from personal experience in collecting ivory from a wide variety of sources. My decisions were not entirely subjective however. An elephant tusk is fixed in the alveolus by a mass of fibrous connective tissue. To remove it from a freshly killed elephant, the surrounding maxillary bone has to be chopped away with an axe. It is virtually impossible to do this without leaving some slight evidence on the alveolar section of the tusk. After the bone has been removed and the tooth pulp extracted, it is normal to try and remove such tissue as is connected to the tusk. This adheres strongly and can only be removed by scraping with a knife. Again, this leaves minute but tell-tale marks on the tusk, which are easy to perceive. The alveolar section of a tusk is longitudinally fluted and very thin layers of tissue remain in the grooves of the fluting, which dries quickly, remains tightly bonded to the tusk and is easily seen.

PAGE 39

168 It is of course possible to leave the tusks in situ until decomposition has broken down the connective tissue which holds them. Complete dissolution of the tissues takes at least ten days and often longer. Usually they have broken down sufficiently within 4-7 days for the tusk to be loose. If it is wrenched at and turned from side to side, the remaining fibres break and the ivory can be drawn. However there is usually still some tissue adhering and which dries onto the tusk if left. It is rather noisome and commonly the hunter will endeavour to scrape away the offensive material, again leaving tell-tale marks. Whether a hunter waits till decomposition allows him to draw tusks or not depends on a number of factors -the most important being whether he can afford the 4 day wait. Throughout this time he is at some risk of losing his ivory to rivals or, to a lesser degree, to officials. However much also depends on how far from his base the kill was made -the greater the distance, the less likely he is to want to wait around. This is countered if the elephant was killed for meat, which often takes a week or more to smoke. The point is that whether tusks are cut out or drawn after 4-7 days, the evidence of tissue still remains. A third technique for extraction is to build a fire under the fresh elephant head and literally cook the tusks out. I have only heard of this being done in parts of Zaire. Again the process seldom removes all tissue and the tusksoften show signs of over-heating -characteristic lateral cracking and flaking. When an elephant dies and is left to decompose without human interference,-all tissue ultimately disappears. Tusks then fallout of their sockets when a scavenger (or other elephants) move the skull about. Such tusks have a distinctive greenishkhaki stain on the alveolar section -the product of advanced decomposition. Rain and weathering will remove this, but replace it with very fine hair-line shakes -which never occur in a drawn or hacked out tusk. Still later in time the tusk becomes cracked, chalky and the surface flakes away -seasonal fires, rain and sunlight all combine in the process 6f weathering. Another

PAGE 40

169 indication that tusks have lain unowned are the gnawings of porcupines (Hystrix spp.). This occurs allover Africa but is particularly common in the Cyclotiform ivory region. Moreover porcupines don't just nibble ivory, they eat it. I have seen many tusks of over 40 cms in circumference gnawed clean through. The objective criteria for man-induced/natural mortality with tusks were thus : Killed by man The presence 'of 1) Any axe, knife or scrape mark -irrespective of how trifling; 2) Any tissue no matter how little; 3) Any blood stain; 4) Any sign of burning without other form of weathering. Natural Mortality The absence of 1 ) AXe, knife or scrape marks; 2) All tissue; 3) Blood stains. The presence of : 4) The characteristic stain of putrescence; 5) Hair line shakes in the alveolar section; 6) Weathering; 7) Burns and porcupine gnawings in the absence of 1), 2) and 3) in the opposite column. Tusks drawn and buried in moist earth would lose tissue through bacterial action and, if they had not been scraped, would appear to have been from natural mortality. Conversely, tusks from natural mortality which were found before the carcass had advanced in decay would appear to have come from killed elephants. Despite such obvious sources of misidentification I feel that the criteria provide a workable basis for separating killed from found ivory. Most of the 5,132 tusks examined and measured had been acquired in the preceding 18 months. A minority were much older dating back to the previous decade. Where possible my estimates

PAGE 41

170 for region of orisin were checked with the tusks' owners. In no case was an estimate of region incorrect and in several cases country of origin was established. Thus all Sahelian ivory examined was tied either to the Sudan or to the Central African Empire. Identifications were aided in a number of cases by characteristic markings made upon tusks by African Governments: e.g. much ivory from Angola bore a South African stamp with 'Rundu: on it (a post on the Angola/Namibia border). The Sudanese either brand tusks with a hot iron or stamp them with metal punches in a characteristic manner. The raw data on circumference, sex, origin and mortality are presented in Tables 169-178. Where individual weights of tusks were available en masse, they were collected and tabulated. The following data were acquired : 1. A sample of Botswana Government 'found' ivory. 2. A sample of tusks taken on licence by Botswana hunters. 3. A sample of tusks taken on licence by expatriate residents in Botswana. t 4. A sample of tusks taken on licence by tourists from the U.S.A. 5. The number of tusks and their weights from the Kruger National Park, 1972. 6. The number of tusks and their weights from the Kruger National Park, 1973. 7. The number of tusks and their weights from the Kruger National Park, 1974. 8. The number of tusks and their weights from the Kruger National Park, 1975. 9. The number of tusks and their weights from the Kruger National Park, 1976. 10. The number of tusks and their weights from the Kruger National Park, 1977. 11. The number of tusks and their weights from the Kruger National Park, 1978. *12. A sample of tusks and weights from the Transvaal Nature Conservation Authority. 13. A sample of tusks and weights from the Angolan/Namibian border. (Table 179) (Table 180) (Table 181) (Table 182) (Table 183) (Table 184) (Table 185) (Table 186) (Table 187) (Table 188) (Table 189) (Table 190) (Table 191) '-

PAGE 42

171 *14. A sample of tusks and weights from Angola/Namibia recorded in Johannesburg. (Table 192) *15. A sample of tusks and weights from Mozambique recorded in Johannesburg. (Table 193) *16. A sample of tusks and weights auctioned in Rhodesia 1977-78. (Table 194) 17. A sample of tusks and weights from Kasungu National Park, Malawi. (Table 195) 18. A sample of tusks and weights from Lilongwe, Malawi. (Table 196) 19. A sample of tusks and weights from northern Malawi. (Table 197) 20. A sample of tusks and weights from southern Malawi. (Table 198) Indicates data supplied from company records; t Provided by hunters; The rest are supplied by Government authorities. Both tusk circumferences and weights can be presented as ivory 'shadows' of the elephants they came from. Even if the sex of the elephants is unknown they can be compared to a combined male + female distribution of tusk weights from known populations to give some idea of the sectors which they represent. When it is possible to determine sex as well as circumference or weight, the picture becomes much clearer. Conversion of tusk circumference and weight distributions into sex and age distributions are based entirely on data obtained from the elephant cropped in Uganda, Kenya and Tanzania. In the first instance age was determined in the original collections by the method described by Laws (1966). Various mathematical expressions were examined with the aid of a Wang 2200 computer and the following were selected as best fits for male tusks. Circumference age(y) = 13.0439 + 0.81268 x (tusk circumference) which explained 90% of variance between ages 0-49 (Figure 44).

PAGE 43

4 44 42 4 38 3 U) 32 IOl o 3 2 s:: -0-1 26 (1) 2 0 (1) 22 H (1). 20 4-l IOl 18 o 16 H -0-1 14 () 12 10 8 6, 4 2 I Age (y) = a + bx FIG. 44 MALE TUSK CIRCUMFERENCE RELATED TO .AGE FROM 5 EAST AFRICAN POPULATIONS COMBINED. (n = 701) Age in Years

PAGE 44

172 Weight : log age(y) = log -2.5312 + 1.5193 log x (tusk weight) for ages 1-23 which explained 91% of variance (Figure 45)# then age(y) = -16.5286 + 1.1318 x (tusk weight) for ages 24-52 which explained 63% of variance (Figure 46). Data for males in their last decade of potential life (50-60 years) are too few to make estimates of worth on either tusk circumference or weight. However I would expect them to call for different treatment to cater for senescent slow-down. Within the limits of time and available funds it was not possible to establish a satisfactory mathematical expression relating either female tusk circumference or weight to age and growth curves were drawn by eye to relate age from dentition to mean tusk circumference and weight (Figures 47 and 48). This produced adequate results for the purposes of this report, but the defection will be rectified in the pending Laws and Parker paper on tusk growth. Attribution of each tusk circumference or weight to an age class seems adequate for present requirements -viz. 0-5, 6-10, 11-15 etc. Accordingly the pivotal circumferences and weights for each age class were calculated from the male regressions or read by eye for females, both of which are presented in Table'199. These pivotal measurements are rounded to the nearest 0.5 cm for circumferences and 1.0 kg for weights. If these relationships have retained their value as aging criteria through the lumping into 5 year classes and rounding into 0.5 cm or 1.0 kg classes, an elephant should still be assigned to the same age class using dentition, tusk circumference or weight. Further, if any two of these parameters are plotted against each other, the slope should be 1.0. Samples are given in Figures 49 a-c. The scatter along the diagonal is small enough to have confidence in aging from either tusk circumference or weight and that the treatment they have been subjected to has

PAGE 45

10 7 til tJ' -.-I cO Q) EO: 4 P.! :::l 0 )...j t9 3 log age (y) = log a + b(log x) 2 1 Age in years FIG. 45 MALE TUSK WEIGHT RELATED TO AGE (1-23 YRS) FROM 5 EAST AFRICAN POPULATIONS COMBINED. (n=1862)

PAGE 46

44 42 40 38 36 34 32 30 28 Y 26 s:: 24 22 s:: ro Q) 20 S 18 P.! ::1 0 16 t9 14 12 10 6 4 2 5 10 15 20 Age 25 in age 30 35 Years (y) = a + bx 40 45 50 FIG. 46 MALE TUSK WEIGHT RELATED TO AGE (24-52 YRS) FROM 5 EAST AFRICAN POPULATIONS COMBINED. (n=527) (See Fig: 45 for Ages 1-23 Yrs)

PAGE 47

3 29 28 ) 27 - 2 6 C 2 i r 2 c u 23 m f 22 e r e 21 n c 20 e 19 . i 18 .. n 17 C m 16 '" s 15 14 13 1 2 11 1 0 2 4 6 8 1 0 12 14 1 6 18 2 0 22 2 4 26 28 30 32 3 4 36 38 4 0 42 4 4 46 48 5 0 5 2 54 5 6 A G E I N Y E A R s FIG. 47 A CURVE RELA TING FEMALE TUSK CIRCUMFERENCE TO A GE, DRAWN B Y E YE, B ASED O N r1EANS AT AGE FROM A SM1PLE OF 824.

PAGE 48

10 ,. 9 8 ------------------------------------------------------------ ,. - .. T u 7 s . k . .. 6 I w I I e ", I i I I g ,. h 5 --------------------------. t I , ,. I I i 4 -----------------------I n I I K I 3 9 I I 2 ------------I I I 1 I I I l. I I I I I I 2 4 6 8 10 12 14 16 18 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 A G E I N Y E A R s FIG. 48 A CURVE RELATING FEMALE TUSK WEIGHT TO AGE, DRAWN BY EYE, BASED ON MEANS AT AGE IFROt-1 A SAMPLE OF 2734.

PAGE 49

40 35 30 E-t ::c C,!) I H 25 fil 20 Ul E-t 15 fil C,!) 10 5 40 35 30 20 15 5 AGE . " . "" ... (D E N TIT I 0 NJ " E N TIT 0 N) x .. 1( I 40 35 30 fil U Z fil 25 p:: fil r.. : 20 u p:: H 15 u fil 10 C,!) 5 40 35 30 til u Z fil 25 p:: H 15 u "l 10 e" FIG. 49a AGES DERIVED FROM TUSK WEIGHT, DENTITION AND TUSK CIRCUMFERENCE PLOTTED AGAINST ONE ANOTHER TO SEE IF THEY FALL ABOUT A SLOPE OF 1.0 AS A CHECK THAT LUMPING INTO 5 YR CLASSES HAS NOT DISTORTED THEIR VALUES. A = MFPS FEMALES (Tusk wt/Dent n=141; Circ./dent n=137). B = MKE 'FEMALES (Tusk wt/Dent n=212; Circ./dent n=203).

PAGE 50

55 50 4 5 40 E-< 35 :>:: (!) H 3 0 ;.:; tI) 2 5 ::> E-< 2 0 (!) 1 5 AGE ... ........ -/_ ...... J "I. .;t ........ X' x "" .. DEN TIT ION )< : . .... '" ,....x,)(. x .,.. FIG. 49b AGES DERIVED FROM TUSK WEIGHT, DENTITION AND TUSK CIRCUMFERENCE PLOTTED AGAINST ONE ANOTHER TO SEE IF THEY FALL ABOUT A SLOPE OF 1 0 AS A CHECK THAT LUMPING INTO 5 YR CLASSES HAS NOT DISTORTED THEIR VALUES. M FPS MALES. 55 5 0 4 5 40 U 3 5 z a: 3 0 ::;: ::> 2 5 u a: H U 20 (!) IS 10

PAGE 51

50 45 40 Eo< :c 35 C) H w 30 (J) :J 25 Eo< 20 w t:J -=: 15 10 5 . :)( ')I. . .,.. ." .. } ..... ,. .. ,... if.-y.. y.. "; :-.: -'1:N"". )( x-x y.. '-';ro ;: A G E N T I T FIG. 49c AGES DERIVED FROM TUSK WEIGHT, DENTITION AND TUSK CIRCUMDERENCE PLOTTED AGAINST ONE ANOTHER TO SEE IF THEY FALL ABOUT A SLOPE OF 1.0 AS A CHECK THAT LUMPING INTO 5 YEAR CLASSES HAS NOT DISTORTED THEIR VALUES. MKE MALES .I 45 40 w U 35 z w p:: W 3C tt. ::E: :J U 25 p:: H U 20 w C) -=: 15 10 5

PAGE 52

not seriously distorted the original direct relationship be,tween age by dentition/circumference or weight at age. 173 As male tusks were separated from females' at the time circumference data were collected in Hong Kong, conversion to appropriate age distributions is merely a matter of referring to Table 199. Table 200 presents average female tusk weight for each year class, sample size and the size of that class as a percentage of the total sample (2,734), average tusk weight being read from the graph in Figure 48. These data are also presented in Table 200 as 5 year running averages for female tusk weight to smooth out fluctuations in simple year to year averages due to small samples etc. In Figure 50 the running averages are presented as cumulative percentages of the population, which illustrate that ./ tusks of 7.5 kg have a 94% chance of being male, a proportion which rises steeply to 100% thereafter. For practical purposes I take this weight the arbitrary point above which all tusks will be classified as male. If we know, or can assume a sex ratio in our ivory populations we can estimate what the male component of the under 7.5 kg tusks might be through the formula in the following example : All tusks over 7.5 kg are male. The assumed sex ratio is 40% male 60% female. The sample is 1,000 tusks. m 1 = 100 tusks kg and therefore male m = male tusks <7.5 kg f = total female tusks, all <7.5 kg

PAGE 53

:, ---7 6 J I>( 5 f-o :c I-' 4 H 3 :.; III ::> 2 E-< 2 4 6 I) ----------------_ 1 0 12 14 16 100 9 0 8 0 .. 5 0 .... -:;; 4 0 ] 30 2 0 lQ 1 8 20 A 22 G E 24 2 6 .. 2 28 I .. 3 )0 N 32 34 4 Y 36 38 E A 40 42 44 46 48 R S r "" .. ...... I - I I I I I I B I I 5 6 7" T U S K WEI G H T K G. 50 52 54 8 FIG. 50 FIVE YEAR RUNNING AVERAGES OF MEAN FEMALE TUSK" villIGHTS RELATED TO AGE (A), AND SAME DATA E X PRESSED AS CUM ULATIVE PERCENTAGE TO ESTABLISH 7.5 KG AS CUT-OFF POINT BEYOND W HICH ALL TUSKS HAVE BEEN ARBITRARILY CONSIDERED MALE (B) (n = 2734 ) A 56 5 8 60

PAGE 54

then (m1 + m) = 400 and therefore 100 + m = 400 f = 600 m = 300, which can be presented as a single 174 <7.5 kg class or broken down into 1 kg classes in proportion to the distribution of tusk weight <7.5 kg. The structure of the female segment of the population could be calculated by yet further assumption, which in this report will not be attempted. In the analyses which follow, the actual (>7.5 kg) + hypothetical .5 kg) males will be combined to provide comparative population indices. To provide pause, analysis of the circumference and weight data presented in tables of this chapter, form the basis of the following section.

PAGE 56

175 VOLUME 2 (4) INFORMATION FROM THE TUSKS IN TRADE (B) In this Chapter tusk circumference data obtained in Hong Hong (summed in Table 201) and weights collected in southern Africa, have been transformed from 'ivory populations' to elephant populations. The pictures derived are presented as cumulative percentage curves and -as an indulgence -as histograms, for some find one easier to follow than the other. Presentation of these population pictures is more to demonstrate what could be achieved with the type of information the ivory trade could provide, than to give definitive statements on the status of any elephant population. The data could be subjected to more penetrating analysis, but as they stand should not be stretched beyond their worth. This caution notwithstanding, the information provided by ivory in the trade can give a more objective insight into general trends in Africa's elephants, than any other presented to date. The circumference data have been transcribed into male and female population structures of 5 year age classes," for maninduced and natu+al mortality. The patterns from the man-induced deaths should illustrate what humans in a particular region are doing to their elephants. A cropping programme as that already described in Chapter 1 of this volume, should represent all age classes in the proportions in which they occur in a population; the highest number being in the categories, diminishing progressively with age. Hunting for meat will produce a broad spectrum of ivory from across the population, with only irnrnatures being under-represented. Hunting for ivory will show selection of older classes of both sexes. Trophy hunting will provide the most extreme expression of this selectivity -concentrating on old males. Defence of property will tend to emulate sporthunting when it is largely concerned with destroying male venturers from elephant islands (national parks) in the midst of humans, though somewhat younger classes of males would" be taken. In situations such as occur in southern Tanzania, where humans

PAGE 57

form islands in an ocean of elephants, killing will be unselective and approach a random cross-section of the population -minus the youngest age classes. The picture from natural mortality in a population not subjected to human interference should approximate to the structure of the living population -mortality curves being 176 more or less reciprocals of survival curves. Laws 1969a and Laws e t ale 1975 have described the pattern of natural mortality in elephants. In brief it is one of a high death rate in the first year of life which declines steeply to a low and constant rate by adolescence. In the female this trend of low mortality persists through adulthood until the onset of senescence late in the fifth decade, when it accelerates .to achieve 100% at the end of the sixth decade. In males the pattern is similar to the females' until the mid-20s when there is a pronounced increase in deaths. This appears to slow down slightly in the fourth and early fifth decade, then accelerate to 100% by the end of the sixth decade. The differential mortality between the sexes in their third decade is such that it reduces the chances of any significant number of males reaching the potential of 60 years. Indeed very few ever pass 50. The tusk data should reflect the expected mortality curve with a high proportion of young animals, a descending but even distribution of adult female tusks over the age span 15-45, then an upswing with rather more old females in the last decade of life. The males should show a marked bulge in the 20-30 year range, a slackening in the 30-40 class and then a secondary bulge at the end of life, 45-55 years. In areas of heavy hunting with muzzle-loading guns, bows and arrows, cable-snares and.wholesale 'control' work, wounding rates are usually very high (Laws et ale 1975, Rodgers et ale 1978). It is unfortunate that the provision of heavy calibre rifles does not as a rule, reduce wounding rates. In many cases it increases them, for the timorous hunter can shoot at greater distance but correspondingly lower accuracy. Elephants dying

PAGE 58

177 of bullet wounds may take months to succumb and when they do, there will be no way to differentiate their tusks from those available from genuine 'natural' mortality. Such deaths will cause the two mortality sources to produce similar pictures. Tusk circumference Sahelian ivory from the Central African Empire The data are illustrated in Fig. 51 and Tables 169, 170 and 202. The sex ratio of 69.3% males and 30.7% females is a strong reversal of natural sex ratios and evidence of deliberate selection for males. The proportion of males taken in each age class between ll-15 and 31-35 is similar. In the natural population the 11-15 class would be far more abundant than 31-35 year olds. This too is evidence of a strong selection of larger males, for. they form a greater proportion of the kill than they are of the population. The female picture is based on only 130 tusks, but it suggests positive selection for animals over the age of 21. The data from natural mortality are too few for speculative analysis, other than to observe that they are in sympathy with the hunting kills and lack the expected proportion of many young elephants. The average male tusk circumference of elephants killed is 32.28 cm (Table 169) which would convert to an age of c.23 years and a tusk weight of 9.3 kg. The average female tusk circumference was 19.5 cm which converts to an age of c.19.5 years and weight of 2.75 kg. With the sex ratio of 69.3% and 30.7% the average tusk weight of the hunted elephants would be 7.29 kg.

PAGE 59

100 90 80 70 60 II U 50 ,. II 40 II< 30 20 10 0 .... c: II U ,. II II< .... c: II u ,.. II II< FIG. 51 SAHELIAN IVORY FROM CENTRAL AFRICAN EMPIRE. MAN-INDUCED MORTALITY AND NATURAL MORTALI T Y AS CUMULA .. IVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIV "ED HONG K O N G CIR CmIF ERENCES. 100 Man-induced mortality Natural mortality 90 80 70 '/ ,@ .... 60 s:: IV u I 50 ... IV 40 I II< I 30 20 Female 10 0 AGE CLASSES AGE CLASSES M A L E F E M A L E Man-induced mortality 50 Man-induced (n-294) (n=130) .... 40 c: IV u 30 ... 20 II II< 10 0 AGE CLASSES AGE CLASSES Naturel mortality 50 Natural mortality (n-67) In=7) 40 .... c: 30 ,.. 20 II II< 10 0 0-5 CLASSES AGE CLASSES

PAGE 60

178 This is substantially less than the average of 17.75 from 1,050 tusks from the Central African Empire shown in Table 168. However the two sources are not directly comparable as the data in Table 168 will have included both Sahelian and Cyclotiform ivory, both of which are exported from that country. The small samples from male and female natural mortality (Table 170) suggest older animals: a male average circumference of 34.76 cm = 27 years = 14 kg tusk weight. The female average circumference was 22 cm = 26 years = 4.3 kg tusk weight. Combining these with the killed sample raises the average circumference only slightly: males = 32.74 cm = 24 years = 9.9 kg, while females stay the same. The overall sex ratios to 72.5% male: 27.5% female which, with the slight increase in male tusk size raises the average tusk weight to 7.93 kg. Sahelian ivory from the Sudan The data are illustrated in Figure 52 and presented in Tables 171, 172 and 203. The sex ratios of 47.6% male: 52.4% female in the hunted elephants and 48.5% male: 51.5% female in natural mortality are both close to what might be expected in the population, and suggest only slight bias in selecting males. In all sets of data, both male and female, man-induced and natural mortality show proportionately far more immatures than in the C.A.E. sample. The inference is that the hunting is relatively indiscriminate and that tusks from natural mortality are recovered with greater efficiency than in the C.A.E. The average male tusk circumference in those taken by man is 27.22 cm = 18 years = 6.4 kg. The corresponding female tusk circumference was 19.37 cm = 19 years = 2.3 kg. The average tusk weight of both sexes combined is 4.25 kg.

PAGE 61

100 90 80 70 .. c: 60 u 50 ... .. 40 II< 30 20 10 0 5 4 .. c: 03 "'2 .. II< 1 0-5 FIG. 52. SAHELIAN IVORY FROM SUDAN. MAN"'INDUCED MORTALITY AND MORTALITY AS CUMULATIVE FREQUEtlCY GRAPHS AND HISTOGRAMS, FOR EACH SEX. DERIVED FROM HONG KONG CIRCUMFEREN CES. Man-induced mortality 0/ / / / /0 / 6 Female 6-10 11-15 16-20 21-25 26-30 31-35 36-40 AGE CLASSES MAL E Man-induced mortality (n-348) AGE CLASSES Natural mortality (n-94) AGE CLASSES >40 100 Natural mortality 90 80 70 .... 60 c: Q) u 50 40 II< 30 20 0/ / 10 cf Cl I .. c: 5 4 3 5 4 ... 2 Q) II< 0-5 6-10 / o Female / o / 11-15 16-20 21-25 26-30 31-35 36-40 AGE CLASSES F E MAL E Man-induced mortality (n-383) A 1; E CLASSES Natural mortality (n-l00) AGE CLASSES >40

PAGE 62

179 average male tusk circumference from natural mortality (Table 172) was 25.84 cm = 16 years = 5.37 kg. The corresponding female circumference was 19.45 cm = 19.5 years = 2.75 kg. The combined tusk weight from natural mortality would average 4.02 kg. Again this and the average from hunted elephants is substantially' lower than the average of 11.17 from 3,405 tusks in Table 168. Cyclotiform ivory The data are presented in Figure 53 and Tables 173, 174 and 204. The most striking feature of the illustrations in the cumulative percentage curves and histograms in Figure 53 are the markedly different patterns between the sexes. Both hunted and natural mortality show similar patterns of selection for mature males. Both female illustrations give a much broader spectrum of age classes with far greater weight being placed upon young animals. There may be some valid reason for differential hunting of the sexes, but one would not expect it to be reflected in the natural mortality. A more plausible explanation is that in Loxodonta a. cyclotis female tusks are relatively slender at age (which subjectively I believe to be the case) than in L. a. africana and that my circumference interpretations are inappropriate in female cyclotiform ivory. If this is so, it would explain the apparent abundance of younger female classes and believing this to be likely, I am unable to use the material for general comparisons. The male data suggest marked selection for the large adults: a that is borne out by the sex ratio of 77.5% males: 22.5% females in hunter kills. That the pattern is reflected closely in the natural mortality suggests a very high wounding rate producing deaths to mask the population's intrinsic rates. As in the two preceding cases the expected high proportions of immatures are absent.

PAGE 63

100 90 80 70 .., c 60 III u 50 '" III 4Q II< 30 20 10 0 .., c III U '" :. 0 .., c Q) u '" ., II< 0 FIG. 53 CYCLOTIFORM IVORY FROH CENTRAL AFRICAN EMPIRE, CONGO, CAHEROUN, SUDAN AND ZAIRE. MAN-INDUCED MORTALITY AND NATURAL MORTALITY AS CUHULATIVE FREQUENCY GRAPHS A N D HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMFERENCES. 100 Man-induced mortality Natural mortality 0 ...... '" / 0 I I 0 / 0/ / I cf I 0-5 6-10 11-15 16-20 AGE CLASSES M A L E Man-induced (n-701) 0-5 AGE CLASSES ,Natural mortality (n=267) 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40 A G E CLASSES 90 80 70 .., C 60 Q) u 50 '" Q) 40 '" 30 I / 20 I 0 10 0 0-5 .., C II U '" tJ II< 5 4 .., c Q) 3 u '" 2 Q) '" ,.. if '" / Female 0 / I 0 I 6-10 11-15 16-20 21-25 26-30 31-35 AGE CLASSES F E MAL E Man-induced mortality (n-224) AGE CLASSES Natural mortality (nz147) A G E CLASSES 36-40 >40 36-40 >40

PAGE 64

The average male tusk circumference from hunted animals (Table 173) is 31.99 cm = 23 years = 9.3 kg of ivory. The average from male natural mortality (Table 174) is 30.58 cm 180 = 22 years = 8.7 kg. Obviously the inclusion of female ivory would drag these down yet further and widen the difference between the average of 10.83 from 6,276 tusks shown in Table 168. East African ivory The circumference data are presented ,in 'Figure 54 and Tables 175, 176 and 205. The patterns of man-induced mortality show larger proportions of animals under 20 years than in any of the preceding figures, suggestive of more intense and general hunting. This gains support from the hunting sex ratios of 53.4% males: 46.6% females which indicates relatively slight male selectivity. For the first time in the natural mortality patterns, we have a pattern approaching the expected high proportion of immatures. The suggestion is that they are sought more diligently and successfully. At this point it is perhaps apposite to comment on the more or less consistent secondary (or in East Africa's case, primary) peak in male mortality that occurs in the 11-15 year old class. It is in this age group that young males reach a height that renders them among the bigger animals in a family unit. This relative size makes them a target for anyone hunting animals from family units. At this age they are expelled from family units (Douglas-Hamilton, 1972), the process of which is protracted. During this period they tend to hang about the outskirts oJ the herds, frequently tagging along some distance behind, making them the most vulnerable members of, family units and many are shot. Once the break with the maternal unit has been made, and the adolescent male is in the company of other males, his (then) relatively smaller size switches selectivity in favour of the larger males. At the same time the young male is no longer such a herd-fringe animal and consequently less vulnerable.

PAGE 65

4.1 c: ... c: II 0 II II U 100 90 80 1 50 40 30 20 Po 10 0 50 40 .... c: II 30 u II Po 0-5 .-.:FIG. 54 EAST AFRICAN (TANZANIA AND KI;:NYA) IVORY. MAN-INDUCED MORTALITY MID NI\TURAL liS CUMULATIVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM HONG KONG CIRCUMf'I::RENCI::S. Man-induced mortality Male 'I 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40 AGE CLASSES MAL E Man-induced mortality (n=641 ) 100 90 60 70 .... 60 c II U 50 40 Po 30 2 0 10 0 5 4 4.1 3 u II Po 2 0-5 Natural mortality 6-10 11-15 16-20 21-25 26-30 31-35 36-40 >40 AGE CLASSES F E MAL E Man-induced mortality (n=559) 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 40 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 AGE CLASSES Natural mortality (n-148) AGE CLASSES .... C II U II Po 4 AGE CLASS""ES Natural mortality (n%110) AGE CLASSES

PAGE 66

181 carryover of the 11-15 year age class in the natural record is likely to be an indicator of fatal woundings which are not recovered by hunters. The average male East African tusk circumference from human kills (Table 175) is 28.66 cm = 19 years = 7 kg tusk weight. The average hunted female tusk circumference is 19.07 cm = 18.5 years = 2.35 kg. The sexes combined in the ratio of their occurrence in the sample would give an average tusk weight of 4.83 kg. The corresponding data from natural mortality '(Table 176) are: male tusk circumference 24.80 cm = 14 years = 4.34 kg; female tusk circumference 16.81 cm = 13.2 years = 2.0 kg. The average tusk weight of the sexes combined in the ratio of their occurrence (male 57.4% : female 42.6%) is 3.34 kg. A sample of 574 sexed and weighed tusks was obtained in June1979 in Kenya. This is worth inclusion here for comparison with the sexed circumference data. Details are presented in Figure 55 and in Table 206. The average male tusk weight was 7.79 kg = 20 years = 29.30 cm circumference; the average female tusk weight was 3.19 kg = 21.5 years = 20.5 cm circumference. The sex ratio was 42.9% male, 57.1% female and this gave a combined sample tusk weight of 5.16 kg. The sample was predominantly. from poached elephants and the hunting was indiscriminate (Somali). It illustrates a preponderance of young animals. range of average tusk weights derived from circumferences are close to, but still lower than the weights from the weighed and sexed sample or the average of 5.36kg for Kenya or 7.08 kg for Tanzania in Table 168. south-Western ivory Tusk data are presented in Figure 56 and Tables 177, 178 amd 207. The kills illustrate that both family

PAGE 67

50 40 30 O M 20 rJ) g 10 50 t:J1 40 ';g 30 t:J1 !OM 20 rJ) g 10 MALES 0-5 AGE C LAS S E S FEMALES 0-5 11-15 .16-20 21-25 26-30 31-35 36-40 41-45. AGE CLASSES FIG. 55 DISTRIBurION OF AGES OF 246 MALES AND 328 FEMALES F.RCM A SAMPLE OF 574 TUSKS OBTAINED IN KENYA IN 1979, MAINLY :rn:::M POACHERS.

PAGE 68

100 90 80 70 .., 60 t: G) 0 SO ... 40 G) '" 30 20 10 0 50 40 .., 30 o 20 ... G) '" 10 .., t: o so 40 30 ... 20 G) '" 10 o !Hi FIG. 56.1 S O UTH-WESTERN IVORY FROM ANGOLA AND SOUTH AFRICA. MAN-INDUCED MORTALITY AND NATURAL MORTALITY I AS CUMULATIVE FREQUENCY GRAPHS AND HISTOGRAMS, FOR EACH SEX, DERIVED FROM BONG KONG CI RCUMFER E NCES mortality 0/ / 0 / g / / / CLASSES M ALE Man-induced mortality (n-466) AGE CLASSES Natural mortality (n=87) AGE CLASSE S 100 Natural mortality 90 80 70 60 Gl 0 SO ... .f 40 30 20 if 10 0 SO ..,3 t: Gl 02 ... QJ '" 1 / / o Female ./ 6-10 11-15 16-20 2 1 -15 26-30 31-35 AGE CLASSES F E MAL E Man-induced" mortality (n=306 AGE CLhSSES Natural mortality ( n =53) AGE CLASSES --36-4 0 >40

PAGE 69

182 units and independent males are hunted, selection in both cases being for the bigger animals in these social components of the population. The sex ratio of 60.4% males: 39.6% females bespeaks a positive selection of males.' The data on natural mortality (Table 178) are few, but suggest a moderately effective recovery of what should be available, overlain with the consequence of a high fatal wounding rate. The average hunted male tusk circQmference (Table 177) is 28.08 cm = 19 years = 7.0 kg tusk weight. The average female circumference is 19.5 cm = 19.5 years = 2.75 kg tusk weight. In the sex ratio of the sample, the average tusk weight from this South-Western ivory would be 5.32 kg. This has no in Table 168 as the Botswana ivory though South-Western, is not comparable (see later). Before proceeding to examination of the un sexed weight data, let me summarise the salient points from the circumference material. That there are differences in hunting effect is plain. In the Sahelian C.A.E. and Cyclotiform ivory there is heavy selectivity of males. In East and South-Western Africa, selectivity for males persists but at a lower level. The hunters of the Sudan are the least discriminating in what they take. A major inconsistency is apparent between the derived average weights from circumference and those obtained from the bigger samples in Table 168. The Hong Kong dealers were asked why this should be. They pointed out that as a general rule tusks of 14 kg or more were not held in Hong Kong. The majority were extracted from each consignment shortly after arrival and re-exported to Japan. The tusks I examined were influenced by this selection process. Recourse to official export documents confirmed this : a sample of 38 export shipments to Japan grossed 19,641.41 kg consisting of 1,357 tusks. The average exported tusk weighed 14.47 kg. It did include a number of less than 14.0 kgs, but I accept the contention that stocks held in Hong Kong will have been biased by the extraction of large ivory. In consequence the male : female ratio will have been

PAGE 70

183 biased toward an over-representation of females, and the patterns of male mortality.in particular will have been biased towards a far younger structure than was the actual case. Tusk weight The tusk weight data which follow were obtained inside Africa at concentration points in the ivory flow, i.e. the records of general ivory buyers and official conservation authorities. In view of the several assumptions which would have to be made I have not tried to extrapolate female population structures from the data. Instead male populations are presented as comparative indicesjtheir under 7.5 kg structure being devised by the formula outlined .in the preceding Chapter, and in which one of three sex ratios was applied. (1) 44.6% males when there were grounds to believe that a general cross-section of the population was likely to be represented in the ivory (e.g. when it derived from cropping in national parks or generalised 'control' work); this male proportion being derived from the East Afr.ican cropping material of Laws et a Z. (1975) combined with Smuts' (1975) South African records from Kruger culling; (2) 62.95% males when tusk weights were from general huntingj the proportion being derived from a combination of sex ratios from the examined Hong Kong circumference-measured tusks; (3) 100% males when the tusks were from sportsmen or 'peripheral' control shooting. The first example is of Botswana found ivory -that is tusks handed in to Government from natural mortality. The data are presented in Figure 57 and Tables 179 and 208. The sample size is small (150) but the pattern is the converse of the expected in natural mortality, i.e. numerous immatures, few matures and rather more senescent animals.

PAGE 71

100 90 80 70 dP Ql 60 :> -,-i +l 50 Ct\ E 40 u 30 20 10 50 P 40 e r 30 c e 20 n t 10 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46 51-55 >55 AGE C LAS S E S 0-5 6-10 AGE CLASSES FIG. 57 BOTSWANA. MALE IVORY FROM NATURAL MORTALITY EXPRESSED AS FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES. (n = 150) I

PAGE 72

The second example is of 3 Botswana hunting results (i) from Batswana hunters, (ii) from expatriates resident in Botswana, and (iii) from visiting U.S. sportsmen. 184 In these cases the weights were treated as 100% male, although the Batswana 'hunters do take a few females. The results are presented in Figure :58 and Tables 180 182 and 209. They clearly illustrate three levels of selective hunting for male elephants. The Batswana shoot elephant to make money from both tusks and hide. They select the largest anLmals they can in available time, but having a modest standard of living do not seek such high rewards as the next group white expatriate residents. Consequently they accept slightly smaller trophies and their average tusk weight is c.14.4 kg (period 1974-1978). The white expatriate residents also shoot elephants to make money, but being much keener to get the most out of ivory, they are more selective and take an average tusk weight of 16.5 kg (period 1974-1978). The constraints upon the two local groups of Botswana hunters is that they have to rely on their own resources and limited time away from work. The third group the U.S. sportsmen do not seek ivory to sell, but as trophies. They want the biggest tusks available, have more time than the other two classes and, more important, through their guides are able to deploy more people searching for elephants. Expense is less limiting to them. Consequently their average tusk weight is 22.5 kg (period 1968-1978); taking substantially larger and older animals than the country's residents. The third example of weights is from cropping operations in the Kruger 'National Park of South Africa (1972 1978). An annual cull is made to keep the population a t a desired level (Smuts 1975). The data are presented in Figures 59-62 and in Tables 183-189 and 210. The assumed sex ratio of the overall population was 44.6% male: 55.4% female, and it was on this basis that the under 7.5 kg segment was estimated.

PAGE 73

FIGURE 58. BOTSWANA. MALE IVORY FROM THREE HUNTING CLASSES EXPRESSED AS CUMULATIVE FREQUENCY CURVES AND HISTOGRAMS RELATED TO AGE CLASSES. BATS\A7ANA HUNTERS n = 477 RESIDENT EXPATRIATE HUNTERS n = 235 U.S. TOURIST HUNTERS n = 65

PAGE 74

90 80 dP 70 Q) ::> I :8 60 co .-i 50 :I U 40 30 20 10 50 p e 40 r 30 c e20 n t-10 50 p e 40 r 30 c e 20 n t 10 '" . 50 P e 40 r 30 C e 20 n t 10 Batswana / / .... / ..... / X / / / I \ I<-U.S. Tourist Resident / / X "" 0-5 6-10 11-15 16-20 21-25 A G E BATSvlANA HUNTERS 0-5 6-10 A G E RESIDENT EXPATRIATE HUNTERS 0-5 6-10 11-15 16-20 21-25 A G E U.S; TOURIST HUNTERS I hunters I I / I I I I I I I 26-30 31-35 C LAS C L A S 26-30 31-3S C L A S .-:. ..: 36-40 S E S S E S 36-40 S E S 0-5 21-25 i 26-30 I 31-35 136-40 A G E C L A S S E S 41-45 41 .... 45 46-50 51-55 >55 46-50 51-55 >55 .: t. i

PAGE 75

50 p e 40 r 30 C e 20 n t 10 50 P e 40 r 30 c e 20 n t 10 50 p 40 e r 30 c e 20 n t 10 AGE C L ASS E S ,1973 (n=I77) AGE C LAS S E S 1974 (n = 434) 'l AGE C LAS S E S 1975 (n = 686) 0-5 6-10 11-15 '16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 AGE' C LAS S E S FIG. 59 AGE STRUCTURE OF MALE POPULATION IN NATIO N;;' PARK DERIVED FROM CROPPED A N D AN SEX RATIO OF 44.6% 55.4% FEMALE. 1972 -1975.

PAGE 76

50 P 40 e r 3 C 20 e n t 10 ; 5 p 4 e r 3 c e 20 n t 10 50 p 40 e r c e 2 .-A G E C LAS S E S .. 1977 (n = 204) 0-5 6-10 16-20 I 26-30 31-35 36-40 AGE C L A S S E S 1978 (n = 484) i 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-48 41-45 46-50 51-55 >55 A G E C L A S S E S 1972-78_Average (n = -... I 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 AGE rIG. 60 AGE STRUCTURE OF MALE ELEPHANT POPULATION IN KRUGER NATIONAL PARK DERIVED FRml CROPPED lATERTAJ. AND l>.!': OVERALL SEX RATIO OF 44.6% 55.4% 1976 -1978, AND 1972 -1978. 3161)

PAGE 77

100 100 90 90 t'lj 80 80 H G) 70 70 0'1 .. ...... 60 60 w Q) > > :;; 50 ..... 50 .&J .. .-.. en::o 40 e 40 HC =' .. :J ZG) u u G)tr::! ,30 30 ::0 en 20 20 ]972 1 97 3 tr::!1-3 10 10 H 00 0-5 6-10 JI-15 1 6-20 21-25 0-5 6-10 11-15 16-20 21-25 26-30 31-35 3 6-40 41-45 :x:.oZ 1-3:x:.o l\ G E C LAS S E S AGE C LA, S S E S :x:.ot-l :x:.o"O en:x:.o Z I-Ij Hn C t-I 100 lJ1:x:.o \01-3 H 90 :x:.oo:::: Ztr::! 80 0 "0 0'Itr::! 70 0::0 n .,. 60 tr::! Q) Z > 1-3 ...; 50 .&J :x:.o '" .... G) 40 tr::! n 30 C ::0 2U 1974 1975 to 10 (f) 0-5 1\ G E C LAS t; E 5 AGE C LAS S E S -----------------

PAGE 78

100 90 80 70 "" 60 :0 5 0 '1l 10 9 0 80 70 "" 60 Q) ; 'M +J 5 0 ttl ...... =' E 40 =' (J 3 0 20 10 0-5 0-5 1976 6-10 11-15 26-30 31-35 36-40 >50 AGE C LAS S E S 1978 1 1 I 16-20 21-25 26-30 31-35 36-40 41-45 46-50 >50 '1\ G E C L 1\ SSE S 100 90 80 70 dP 60 Q) > :;j 50 III ...... =' 40 3 (J 30 20 10 0-5 6-10.11-15 16-20 21-25 26-30 36-40 4 6-50 AGE C LAS 5 E 5 '. \ -

PAGE 79

185 As the Kruger population is believed to be young and fertile (Joubert pers. comm.) the pattern of the male index should be in the range between the MKE and combined East African data. The data from 1972 meet this expectancy and the population resembles MKE. Subsequent years produce rather more adult males and one is left with the impression that in 1972 mature males were not sampled at a level commensurate with their true position in the population. A striking feature of the histograms for Kruger 1973-1978 (Figures 59 and 60) is the consistent depression in the 16-20 year old age class. The cumulative percentage curves (Figures 61 and 62) also emphasise its position persistently below a curve which joins the other pivotal points. This is largely an artefact of the process of determining the of males below 7.5 kg tusk weight. The 44.6% male: 55.4% female is made arbitrarily across all tusk size/age classes for convenience. In reality it is unlikely that the male 16-20 year class would be in this relationship with the 31 >40 female age classes it corresponds to in tusk weight. The artefact notwithstanding, its application is uniform from year to year, and the differences in.male population structure must represent some variance in cropping focus or in population structure, or both. This is particularly noticeable in the first two age classes : 0-5 and 6-10. There also seems to be a variation of approach toward males in the year classes 21-25, 26-30 and 31-35, over the period covered. With further time at my disposal satisfactory explanations would undoubtedly have been 6btained. The issue of consequence here is not so much what is happening in Kruger, but in the demonstration that the .i vory crop reveals the form of offtake and that it has not been even from one year to the next. Earlier it was pointed out that males in their 20s are marauders around the edge of stable elephant sanctuaries. The Kruger National Park is one such area and male elephants

PAGE 80

186 persistently .break out into surrounding lands (Joubert pers. cornm.) In consequence they are shot. In the Transvaal, responsibility for this is with the Nature Conservation Authorities. Figure 63 and Tables 190 and 211 provide evidence of the age of bulls involved. Two sets of data on tusk weights from southern Angola and northernmost Namibia were obtainedj one at the border post of Rundu, the other from a trader in Johannesburg. They are completely separate, and presented in Figures 64 and 65, and Tables 191, 192, 212 and 213. The overall proportion of males in the sample was to be 62.95% male. The two samples are similar in form and indicate generalised hunting with selection for males in the 20-30 year classes. In origin this ivory is comparable with the South-Western circumference data. The circumferences suggest greater emphasis on younger males which would be the case if larger tusks had been extracted on arrival in Hong Kong as suspected. The average weights (males combined with females) are in accordance with this supposition, being 8.89 and 8.84 kgs vs. the circumference derived 5.32 kg. The data in Figures 64 and 65 illustrate not only the selection of larger males, but the corollary : that such animals exist. Mozambique on the other side of southern Africa to Angola, provided data which is nearly contemporaneous and presented in Figure 66 and Tables 193 and 214. Using the same sex ratio in estimation of the male component of the sample, it is clear that a different situation existed in Mozambique. The type of hunting -general for meat and ivory was similar. However, the results indicate a far younger kill with -considerable emphasis on the males still in family units. The most likely explanation for this is that available mature males formed a smaller proportion of the elephant population than occurred in Angola/Namibia : in other words they were insufficient to meet hunt demand. The average tusk weight of the whole sample was 6.95 kg.

PAGE 81

100 90 80 70 dP 60 Q) :> "j 50 co .-i 40 u e r c e n t 30 20 10 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 AGE C LAS S E S i I 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 F:IG. 63 AGE CLASSE-S TRANSVAAL NATURE CONSERVATION AUTHORITIES. MALE IVORY EXPRESSED A CUHULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES.. (n = '125) 1

PAGE 82

100 90 80 70 I / I', ... d(l 60 .... 50 ,J co rl ::l 40 3 u 3 0 20 10 6-10. 11-15 16-20 21-25 26-30 31-35 36-40. 41-45 46-50 51-55 >55 A G E C LAS S E S \ 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 '. GEe LAS S E S FIG. 64 IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES. (n = 729)

PAGE 83

100 90 -80 70 . dP 60 " iv :> 50 ro --i ::l 40 E ::l U 30 20 10 0 5 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 AGE C LAS S E S ,0-5 6-10 AGE C LAS S E S ,FIG. 65 ANGOLA/NAMIBIA BORDER. MALE IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES. (n = 269)

PAGE 84

100 90 70 :j 50 III .-; 4 0 :::J u 30 -70 10 \; I J ,\ I ', I '. {; \ \ " 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 AGE C L SSE S '.:, -'. 0-5 AGE C LAS S E S FIG. 66 MOZAMBIQUE. MALE IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES. (n = 535)

PAGE 85

187 In Rhodesia elephants are very largely confined to national parks or reserves. The major source of ivory is from the regulation of populations within these areas : cropping. 1his is done in similar manner to that described from East Africa by Laws et aZ. 1975. The cropped ivory is augmented by that from control work -about the periphery of the elephant areas, and limited sport hunting. This picture is borne out by the data from 3,604 tusks illustrated in Figure 67 and Tables 194 and 215. The data are interpreted on a cropping ratio of 44.6% males to 55.4% females. The 'preponderance of very young, and decline in subsequent age classes is in accordance with East African cropping data. The artefact of a lowered 16-20 age class is apparent as in the Kruger data. The exaggerated 21-25 year class is likely to be the product of peripheral control shooting. As in South Africa and Rhodesia, Malawi's elephants ooarr in national parks, game or forest reserves (Bell .1979). There is as yet, no case for reducing or stabilising numbers in aD! of these sanctuaries. However, as is to be expected, elephants sally forth from their preserves and damage crops. For this they are shot. As usual, the majority are males in the 21-]0 year classes. Malawi's data are given in Figures 68 and 69 and Tables 195-198, 216 and 217. The estimates of males less than 7.5 kg tusk weight have been based on an assumed sex ratio of 44.6% males: 55.4% females (except in north Malawi where the whole sample has been treated as male). In Figure 68 two collections of data were made; one beDq tusks registered under a Kasungu National Park code, the other registered 'under a Lilongwe code. The sources are, however. essentially the same, namely ivory predominantly from the National Park and its environs and, to a much lesser degree. the Nkotakota Game Reserve. The patterns illustrate the effect o f control shooting on raiding males superimposed upon younger classes from natural mortality in the national park.

PAGE 86

100 90 80 70 "" 60 c:> :> .,..j 50 .v r.; .-t ::l 4C E ::l (oJ 30 2 0 10 0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 >55 AGE C LAS S E S f AGE CLASSES FIG. 67 RHODESIAN AUCTION IVORY. MALE IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND HISTOGRAM RELATED TO AGE CLASSES. (n = 3604)

PAGE 87

!-rJ H GJ 0'\ ex> PlH<::t>' UlOOOt"i l.Q GJ H XI Z 1'0 rt t%jH f-J, 00 t'" Ot%jooO !:It'''t%jZ PI:t>'OGJ t%j:t>'t%j 1'0000 PI :t>' GJt"':t>' 1It%j::t:>'OO 00 OO!-rJZ 'IOH CO t"'t%jZ f-JZ ::t:>' I-'(')t"' o !:l 1'0 (') ::t:>' NOt%j N 1(0 9 7
PAGE 88

188 In Fig. 69 two further are illustrated. The first again is evidence of male mortality about the edge of a sanctuary -the Vwaza Marshes. The second stems from the Liwonde National Park, the Mangochi and other southern forest reserves. It presents an entirely different picture of heavy juvenile mortality. The reasons for so different a situation around the southern sanctuaries is not clear. Nevertheless it provides a good example of how an ivory shadow can draw attention to anomalies. The observation has been relayed to the Malawi authorities for such attention as they may think necessary. A final analysis of African tusk weights broken down to a male population index concerns data from Tanzania. These were obtained by Douglas-Hamilton and Davitz from the Tanzania Ivory Room in Dar es Salaam and are the individual weights of 44,074 tusks recorced between 1971 and 1977. They have been analysed by Davitz (1978) and are of some notoriety, for it vas the average weight (4.8 kg) of this large collection of tusks that led to the belief that each tonne of ivory exported from Africa represented the deaths of 100 elephants. The Tanzania Ivory Room tusks (illustrated in Figure 70 and given in Table-218) derive mainly from elephant control, most of which takes place in the south-east of the country (Rodgers et aZ. 1978). This is so intensive that i t must approach random sampling -at least in all but the very smallest elephants and, perhaps, largest males. On this basis the separation of males with tusks under 7.5 kg from females has been made on an assumed sex ratio of 44.6% males: 55.4% The resulting male population kill illustrates concentration on young elephants. However itfails to show the slope one expects in n ormal populations -falling progressively with increased age. This could be the consequence of such selection for size as may occur, counteracting the natural age distributions. lVhat is rather more difficult to accept is the small representation of males over 25 years in a sample of 19,658. It is not as though males of this age are absent from the control areas. AloD9 the e astern

PAGE 89

dP Q) :> .,...j +l cO ,.....j ::I E; ::I U P e r c e n t 10 9 8 70 60 50 40 30 20 10 ,;-.. -0-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 55-60 AGE AGE C LAS S E S 31-35 36-40 41-45 46-50 51-55 56-60 C LAS S E S FIG. 70 TANZANIA. MALE IVORY EXPRESSED AS CUMULATIVE FREQUENCY CURVE AND HISTOGID\M RELATED TO AGE CLASSES. n = 19,658.

PAGE 90

189 borders of the Selous Game Reserve many elephants are shot on cqntrol (Rodgers et al. 1978) and in this area there are many males older than 25. For this the Selous Reserve is much esteemed by sportsmen as a hunting ground. The prospect is thus raised that the Ivory Room data represent 'filtered' tusks, the larger specimens having gone elsewhere. Such a possibility has some support from the evidence of inconsistency in Tanzania's internal ivory records (e.g. the manner in which national park ivory is sold; or the discrepancies between Ivory Room data and Customs figures -see Chapter 4 Volume 1). The trade also recognises quite substantial amounts of Tanzania ivory coming out clandestinely through Burundi. Finally the Tanzania data in Table 168 also suggest a higher average tusk weight for the country. The diverse data in this chapter provide. a variety of insights into elephant/human interactions, and the potential that ivory has for revealing information on conservation practice or non-practice. However one which calls for particular consideration concerns natural mortality. The Hong Kong data indicate that 21% of tusks arriving there come from natural mortality. Examination of this is the basis of the next chapter.

PAGE 92

190 VOLUME 2 (5) NATURAL MORTALITY The only certainty in an elephant's life is that it will die. If that happens after 2 years of age, it will leave tusks. Irrespective of all other factors, natural mortality is a substantial source of ivory. How much of a population dies each year will vary widely according to local conditions. However as a general index to use on a continental level, I believe 7% to be sufficient for the purpose of discussion. Laws 1969a has demonstrated the mode of mortality and I have used his approach to construct my own mortality Table (Table 219). From this I have postulated that 42.5% of annual deaths willobe under 2 years of age and therefore tuskless. The balance will leave tusks varying from 0.23 kg to c.50 kg in weight. The distribution of these is also included in Table 219. From the age range of deaths and combining both sexes, the average tusk weight should be 8.06 kg. Douglas-Hamilton (IUCN Elephant Provisional Estimate 1978) has estimated that the minimum elephant population of Africa is 1.3 million. If this be true, and a 7% annual death rate prevails, then 91,000 elephants will die every year on average. If, of these 42.5% are tuskless, then 52,325 will leave tusks. Not all elephant carry two tusks. Some females are tuskless and others are congenitally single tusked or become so through accident. In view of this it would be incorrect to multiply the number of tusked deaths by 2 to obtain the number potentially available. I shail thus follow Rodgers et al. (1978) and use a factor of 1.88. The 52,325 deaths should therefore produce 98,371 tusks which, at ano average of 8.06 kg each would weigh 792,870 kg. The foregoing figures may not be precise, but they indicate the order of what is potentially available through natural mortality. Were all the tusks collected, they alone would come close to meeting world ivory demand as it was in 1978.

PAGE 93

191 The crux of the matter is finding and collecting the tusks before they, too,.decompose and disappear. The only methodical attempt to analyse this problem of which I am aware, was that of Graham and Laws (1971). They describe searching for ivory from natural mortality in the Murchison (now Kabalega) Falls National Park, Uganda. They ascertained that only 27.6% of carcasses from elephants dying annually were being found. As will .be seen later the recovery rates in Uganda were exceptionally high. A further unspecified amount was believed to have been found illicitly and the balance decomposed. Found ivory is a source of revenue to many national parks in Africa. Since one of the major constraints upon their development is shortage of funds, tusks are collected as and where possible. Rewards are offered for those handed in by staff, and few if any rangers would leave tusks if they found them. Thus the proportion of ivory recovered from natural mortality is an indicator of the efficiency with which they cover the area under their control. In addition, mortality being a reciprocal of survival, efficient recovery of ivory would provide an accurate measure of the age structure of the extant population. Some ivory recovery rates are presented in Table 220 for a sample of 9 parks and reserves. Other than in Tanzania's Manyara National Park, and Kenya's Marsabit National Reserve, recovery rates are low : below 6% of what becomes available annually. The two with higher recovery rates are also exceptions in being of very small size -Manyara 80 km2 and that part of Marsabit frequented by elephant (other than in wet periods) being less than 80 km2 Two of the 9 are montane forest zones -Mt. Elgon and the Aberdares/Mt. Kenya -the rest are 'savanna woodlands'. That the montane forest recovery rates are similar to the savanna's is somewhat surprising, as in these at least, the dense vegetation could be expected to conceal carcasses far more effectively than in the more open areas.

PAGE 94

192 An individual case of considerable detail is available from the Tsavo East National Park's record of found ivory from 1950-1978 (Table 221). At the outset of this period, through to c.1970, I believe that Tsavo East contained at least 12,000 elephants -a figure not out of keeping with the estimates that have been made between 1968 and 1978 (e.g. Laws 1969a to quote the earliest in this period). If there were 12,000 elephants and they had an annual mortality rate of c.7%, 57.5% of which bore ivory, 869 tusks should have become available annually for collection. The data in Table 221 show that until 1970, substantially less than this number were recovered. From 1970 onwards overcrowding, drought and subsequently poaching, brought about a population crash and a complete departure from the previous mortality patterns. In Table 222 I have re-presented tusks found, against an assumed constant annual output of 869 becoming available for the "stable" period of 1950-1970 in TsavoEast. On the assumption that a tusk remains recoverable -albeit weathered -for 3 years, I have presented the uncollected surpluses annually and as running 3 year accumulations. At anyone time ivory available for collection would have been annual production from natural mortality plus the running cumulative surplus. Thus the success of recovery is not measurement of' a simple proportion of what becomes available in anyone year, but must also take into account the accumulated component. On this basis I estimated Tsavo's annual success in recovering ivory from natural mortality at an average of 7.72% of what lay on the ground. The approach is of course somewhat academic. It takes no account of the possibility that some tusks may have been recovered by unauthorised people. However between 1957 when a successful anti-poaching programme was launched and 1970, there was very little evidence of poaching in Tsavo East. It is also for convenience of demonstration that I have assumed both a constant population and a constant mortality. The population was definitely not stable during this period. It "acquired" new

PAGE 95

193 components of between 2,000 and 4,000 elephants from across the southern and eastern borders south of the Sabaki river during my tenure 'as a Game Warden in the area between 1957 and 1964. (Advancing human settlement drove them back -mainly by monopolising very local and limited water resources.) In addition the middle of the period c.1960-1961 was assailed by a severe drought in which it would be ecologically unlikely that there was no rise in mortality. Further the period 1957-1970 was characterised by a progressive depletion of woodland (described by among others, Glover 1963, Laws 1969a) which is likely to have had influence on elephant survival, even if it was not immediately apparent. The Tsavo data 'are ground for believing, that the recovery rates in Table 220 are optimistic. The collections from what becomes available and stays on offer for 3 years or more reduces actual recoveries to a still lower index of success. It is worth comparing the structure in tusk size classes of what should become available from natural mortality, with what was found, for it may convey reasons for the poor recovery rates. Using Table 219 as a base, the number of tusks per 1,000 males and 1,000 females dying are synthesised into a histogram in Figure 71 and into a Table 223 to illustrate the distribution of tusk size classes from my postulated natural mortality. The resulting pattern provides a comparison for the distribution of found tusk weights in Kenya's Aberdares and Mt. Kenya Park, Marsabit National Reserve, Tanzania's Manyara Park and finally in more detail Kenya's Tsavo-East Park. A overlay of the outline of Figure 71 is provided to facilitate comparison with ; the other patterns. Figure 72 and Table 224 present the data from the mountain parks of Kenya -Aberdares and Mt. Kenya -for 1959 to 1978. The pattern deviates from the expected in a dearth in the first tusk size class and an unexpectedly high proportion in the intermediate tusk sizes. The classes 7-14.99 kg and 15-29.99 kg are smaller

PAGE 96

p e r c e n t 50 40 30 20 10 T u s k wei g h t cIa sse s FIG. 71 THE EXPECTED PATTERN OF TUSK SIZES AVAILABLE FROM NATURAL MORTALITY. (AVERAGE WEIGHT OVERALL 8.06 KG).

PAGE 97

1 p e 4 r 3 c e 2 n t 1 ABERDA..'RE AND MI'. I
PAGE 98

194 than expected -probably because of the earlier intermediate deaths. The average tusk weight of 7.11 is in keeping with early mortality and a younger than projected population (i.e. it is less than 8.06 kg.) Figure 72 and Table 225 present Kenya's Marsabit found ivory, 1962 -1969. The pattern in this case is anomalous with a very high proportion in the large tusk classes, none at all in the first weight category and erratic results in the intermediate ranges. The sample size ,is small, but nonetheless represents a high recovery rate -33.5% of the expected (Table 220). The average tusk weight is 2.29 times higher than the expected. The pattern for tusks shot on 'control' (Figure 73) over the same period produces the same picture -a population of old animals -predominantly males. It was this community which produced the publicised "Mohamed", "Ahmed" and "Abdul" and which has been known as the haunt of very large (old) tuskers for the past 40 years or so. It would seem to have been a relict population, but quite how much of an anomaly or oddity has not been apparent until the evidence of these ivory 'shadows'. Figure 72 and Table 226 present Tanzania's,Manyara elephant -among the best known populations in Africa (Douglas-Hamilton 1972). Once more the absence of the smallest weight class stands out. The larger than expected intermediate classes represents a major die off of the population which will receive more detailed description from Douglas-Hamilton. The absence of old males is in accordance with what is known of the population, i.e. they are absent. Figure 74 and Tal;>les 227 -234 present the Tsavo East story in 8 segments.' The first is in the early years of the park when development was impeded through Kenya's turmoil in the Mau Mau rebellion. The mortality shadow has a form fairly similar to the predicted. However the smallest tusk class does not reach expectations, and the 7-14.99 kg unit is larger than expected. The average tusk weight is 9.68 kg which is high. There was

PAGE 99

5 p e 4 r 30 c e 20 n t 10 Weight c 1 ass e s Kg FIG. 73 HARSABIT NATIONAL RESERVE KENYA. FROM A OF 49 ELEPHANTS SHOT 'ON CONTROL' BETltJEEN 1962 AND 1966. AVERAGE TUSK ltJEIGHT 20.99 KG.

PAGE 100

1950 1!>55 195fi 1957 AlT. 1'7I'. 9.68 J(G e 40 AV. wr. 10.50 w.; r (n '" 469) r (n .. 499) 30 c c e 20 n t 10 GJ -...J weight c 1 ass e s Weight c 1 ass e s Kg 5 50 p 1958 1960 P 1961 -1963 I e 4 AV. wr. 9.85 J(G e 40 AV. ,wr. 6.73 J(G 8 r en = 225) II r (n :: 821) 3 30 c c e 2 e 20 n n t '0+ . t 10 0:;:99 1-1.99 (/')a weight cia B s e s Kg Weight c 1 ass e s 50 SO P 1964 1969 P 1970 -1973 e AV .wr. 8.67 J(G e 40 AV, wr. 6.86 J(G r '(n'" 1,769) r (n = 8,572) 30 c c e e 20 n n t t 10 -...J ?'!2! weight c 1 ass e s Kg Wei q h t c 1 a s I!I e )(9 50 p P 1977 1978 1974 1976 e 40 c AV. wr. 6.71 J(G AV. wr. 4.17 m r r (n '" .3,945) (n :: 195) 30 c c c e 20 n t 10 0-0. .99 6-6.99 7-14.9915-2g,99 )29.99 ,Weight c 1 ass e 8 Kg c 1 ass e 8 Kg

PAGE 101

195 substantial poaching at this time (1950-1955) by watta and Kamba bowmen. The histogram's slightly thicker 'waist' than expected may well be a reflection of their influence. In 1956 and 1957 a vigorous anti-poaching campaign was waged and in the process substantial quantities of ivory were recovered. This raised the average .tusk weight to 10.5 kg and the picture is not one of true natural mortality. Concentration upon the poachers produced the histogram's greatly thickened waist where mortality should be least. From 1958 until 1960, there was virtually no poaching in Tsavo East. The number of tusks recovered fell back to pre campaign levels. Immatures in the lowest age class are totally absent, the waist of the-histogram is inexplicably thickest where it should be thinnest, the peak of 7-14.99 kg class is twice as tall as anticipated. The average tusk weight remained .up -no doubt due to the higher than expected contribution of the intermediate classes and the absence of the smallest class. No ready explanation arises from the circumstances of the time. The period 1960 -1963 was characteriseq by a drought of unusual severity, unprecedented heavy rain and floods and finally extensive fires. In this period widespread rhino mortality occurred and elephant removal of woodland became general. In these a rise in elephant mortality would not have been unexpected -but was not reported. The mortality pattern approaches the expected with a return of the smallest size class, though still not up to expectation. The 7-14.99 kg class 'stack' is perceptibly shorter than in preceding years, the mortality represented by the thicker preceding 'waists' perhaps beginning to tell on older classes? Their recruitment is failing. Mortality in the intermediate classes is still slightly higher than it should be. The drop in average tusk weight (8.73 kg) attests to a decline in average age.

PAGE 102

196 Between 1964 and 1969 there is little change in pattern. The average tusk weight ramains similar -8.67 kg. A period of reasonable rains retard but do not stop the elephant clearing away woodlands. In 1970 there is a sharp rise in elephant mortality, which is followed by a spectacular population crash in 1971 with the onset of a prolonged drought (which was also the trigger to Somali problems, see Appendix No.7). The sample from which the pattern for this period derives comprised 8,572 tusks. For the last time, the pattern resembles the expected to some degree. As usual juveniles with tusks 0-0.99 kg are under represented. The shortening of the 7-14.99 kg stack first observed a decade earlier has moved forward and is apparent in the diminished last two classes. The average tusk weight has fallen nearly 2 kg to 6.86. The population has become very much 'younger'. Hard on the heels of the drought in 1974 a wave of poaching develops. This soon rises to unprecedented proportions with the arrival of the Somalis. The pattern of mortality is no longer 'natural'. The smallest class has again receded, and the intermediate waist has a pronounced bulge -similar to, but exceeding that in the 1956-57 anti-poaching drive; this time the histogram has greatly receded older age classes. The eighth and final shadow has the characters of an accentuated die off as happened in Manyara (Figure 72). Recruitment into the smallest class appears to have failed and the oldest class has gone entirely. The average tusk weight has fallen to a meagre 4.17 kg. The foregoing description is deliberately anecdotal and I have avoided all attempts to be technical. The evidence largely speaks for itself. With the wisdom of hindsight the ivory record provided early warning of the population trends. They are summarised in Figure 75 and Table 235 as 2 indices (i) the average annual weight of found ivory (8 kg = 100), and

PAGE 103

200 190 -----INDEX OF TUSK vlElGiT 8 Kg. = 100 180 INDEX OF FOUND TUSKS 1,700 p.a. = 100 170 160 150 1\ 140 I \ \ 130 I \ I 120 I 110 I I 1 /, .' j I '\ /\ I, ....-\ 1 \ \ I \ ....\ 1\ ,\ ,.. \ I "' \ 1/ \ / \ I \ / \ \/ \ 1 \ / \ I \ I \ '/ \t \ I \ / \ \, \ / \ \,' \ I \ I \ V \ \ 7 90 80 70 60 50 40 I f .,1 30 20 10 i 1950 1955 1960 1965 \ / v 1970 r-..../ / / 1 1975 \,

PAGE 104

197 (ii) the number of tusks re90vered (1700 = 100) each year. From the anti-poaching days of c.1958 until 1970, the manpower available to find ivory remained of the same order(c.50 men). Yet there was a progressive increase in the amount of ivory recovered. The suggestion is that mortality was rising. An aircraft was added to the 'ivory-finding equipment' (not deliberately as such) in c.1964, but does not seem to have changed the rate of finding beyond the trend already apparent. The rise in ivory finding rates was matched by a counter trend in the weight of tusk. Again with hindsight and a 29 year span of data to dwell on, the trend is obvious. However, in 1962 it would hot have been so apparent -even with mathematical insights. With the evidence now before us, there can be no doubt that careful recorded ivory data is a simple and sensitive tool for analysing elephant population trends. It has, however, certain weaknesses. Tusks in the youngest age classes are almost invariably under-represented. They are so small and light that even at today's prices they are hardly worth the trouble of collecting. They are easily broken up by scavengers such as hyaenas, and having so little body, they must decompose and disintegrate rapidly -though no rates have yet been established. The corollary to this is that large tusks must take longer to rot and disappear and therefore be on offer for collection for much longer than small tusks. The argument has logic and while I am certain that it does apply, the question is to what degree? From the absence of large tusk weight classes outside the expected range it would appear that this is not of such consequence as might be expected. Vultures are a prime locator of dead elephants and indiscriminate where tusk sizes are concerned. Thus fresh carcasses to which vultures draw the ivory seeker should be a random sample of the available. Where they are absent -such as in montane forest -one would expect a large tusk's greater durability to bias finds strongly toward their recovery. The

PAGE 105

198 Aberdares and Mt. Kenya records do not reflect this expectation. Indeed they suggest that random chance may be the most important element in ivory recovery. That this is indeed the case is borne out by Laws' 1966 and 1969a calculations on survival and mortality, which were based on 'found' jaws (located only in the search for ivory). Had Laws' samples been biased toward large (old) animals his postulated mortality rates would not have been as accurate as they have been demonstrated to be by extant population structures. However these are points of tangential consequence for this report. The issue of greatest pertinence from found ivory in the national parks of Africa is how little is recovered. It is strong support for the contention that staffing in these reserves is inadequate. That on average much less than 6% of what is on offer at any time is collected is clear indication that the ground is covered inefficiently. The ivory remains as an attraction to others to find it -and enter the areas illegally -or is lost. Outside the parks, where men may wander and search unhindered, recovery of ivory from natural mortality may be more efficient. That this is almost certainly true -at least locally -is borne out by the high proportion of found ivory, 21% in the Hong Kong samples I measured. suffice it that the issue needs to be researched more thoroughly from the conservation and economic points of view, but that the contention of under-staffing stands out yet more clearly.

PAGE 107

199 VOLUME 2 (6) IVORY WEIGHT Being the standard measure of ivory, weight is the most widely available statistic on elephant populations. Changes in average weights, regionally and generally, are useful indicators of either population trends or the manner of human/elephant interactions (or both). In this chapter I shall assemble and comment on a miscellany of 'average' tusk weights. The Uganda data (Volume 2 Chapter 3) illustrate that a decline in male elephant life expectancy is apparent at an early stage from the size of sportsmen's trophies. The Botswana data (Volume 2 Chapter 5) show that this can be refined by categorising sportsmen : the more demanding they are, the more precise their selection and more accurate the information will be. The Uganda material also indicate that an overall average tusk weight can rise while a population declines in size, though this is only likely on an extended scale when elephants are faced with progressive displacement and loss of range. Another illustration from the same set of data is that one should know of which 'average' one speaks -for there may be several within a population, i.e. from sport-hunting, control shooting, poaching and natural mortality. Further, as shown by Uganda's falling weight of sportsmen's tusks and rising overall average weight, these different components can run contrary to one another. These cautions notwithstanding, downward trends in tusk size can be taken as general indicators of lowered life expectancy -usually from hunting at rates which exceed replacement into given age classes. Obviously there is a threshold below which either reproduction is impossible (sexually liamature) or the species is not viable socially. In view of the complex social organisation of elephants (Douglas-Hamilton 1972) it is not unreasonable to assume that a viable population has a threshold ratio of immatures to adults, below which it may survive. However this has yet to be demonstrated.

PAGE 108

200 Unfortunately I have been unable to locate any other records which are either as detailed or sound as the Uganda data in Table 166. The nearest to it is that compiled by Rodgers et aZ: 1978 -a synopsis of which is presented with other miscellaneous data in Table 236. These cover a span 1920-1977 (58 years). Between 1920 and 1970 the data show no obvious trends, other than that the average weight of control tusk may have been declining. The average weights of tusk on licence, found and overall showed no particular trends. Between 1971 and 1977 control tusks, those taken on licence and found, all dipped by between c.19-25%. Only Rodgers' et aZ. (1978) overall tusk weight remained within the previous 51 year range. This overall tusk average weight should be what is detectable on arrival in Hong Kong. From Table 168 the 1978 average for 2,248 tusks arriving in Hong Kong from Tanzania was 7.08 kg. If this is ,substituted for Rodgers' et aZ. (1978) overall average weight post-1970, the drop is within the range recorded under the other categories. I disregard the U.s. hunters' apparent increase in tusk weight, for this is largely the product of specialised hunting in the Selous Reserve -formerly a sanctuary -and concerns very few elephant. The average tusk data from Tanzania 1971-1977 indicate a general decline in tusk size. This complements the picture in Figure 7 and Table 40 of greatly expanded ivory exports in this period. Information on Kenya's average tusk weights are given in Table 237. The record is incomplete. Nevertheless trends are apparent. The limited U.s. sportsmen's data illustrates gradual decline from 1952-1975 (it also illustrates that some hunting was condoned after 1974 when all sport-hunting was "banned"!) Returns from elephants taken on licence from 1927-1973 also show a decline in average weight. Control data are inadequate for any conclusion. only represented are that of 1957 Poached ivory shows consistent decline, although by 4 figures. The two of particular consequence 9.04 kg -and that of 1979 -5.16 kg. Both stem from anti-poaching work in the same region of Kenya (coastal hinterland -Tsavo) largely carried out by the same man -F.W. Woodley.

PAGE 109

201 General found ivory data is too sparse for speculation, the local Tsavo data already referred to in the preceding chapter is in keeping with overall decline in tusk size. A sample of auction records from 1959 to 1976 again establish a pattern of decline which would appear to have accelerated post 1970. The one figure from Kenya's exports to the Ivory Coast of 3.16 kg should be treated with caution as that market is selective of small ivory. Nonetheless the low average weight apparent in the Hong Kong record (Table 168) is in keeping with the overall pattern. The trade data in Figure 9 and Table 43 which illustrate steep rises in exports post 1970 are consistent with a decline in tusk weights. Information on Zambia's average tusk weights were obtained for 1968-1978 only. These were provided by the Zambian National Parks and Wildlife Service. Their data which incorporate all sources, show an overwhelmingly adult male offtake. The only exception in the material presented in Table 238 is the average of 6.76 kg for 1971. This included much immature and female ivory from the Luangwa Valley elephant cropping. The other contrasting figure is the 1978 Hong Kong record of 8.04 kg (see Table 168) which -as a source ex-trade as in apposition to sport hunting -is from a small sample of 130 tusks. The U.S. hunters' data (which, due to its selectivity may be the most sensitive to the onset of trends) suggest the start of a downward trend in male tusk size. They have some support from the last figure in the Zambian safari clients' returns (1976), of which they must be part. However, overall, the average tusk weights from Zambia do not give clear indication of any upward or downward trend. Botswana's average tusk weights are presented in Table 239. They are obviously predominantly adult male and -of note -are larger.on average today than 110 years ago when William Finaughty hunted elephants there for a living.

PAGE 110

202 Data from Zaire are difficult to obtain and often inaccurate (Volume 1 Chapter 4) As ivory production is disorganised and much leaves through neighbouring states, it is to the latter one must look for evidence Zaire tusk weights. From 1936-1959, with omissions of 5 years, the Uganda Game Department kept records of the number and weight of tusks shipped from Zaire through Uganda, to the Kenya coast. Their average weights are presented in Table 240, together with the tonnes they derived from A single 1963 Kenya figure complements the set of data. Their outstanding feature is that for such large consignments (average 42.7 tonnes per annum) they present a high average tusk weight. Further, over this span 1936-1963 (28 years) there is -if anything -an increase rather than) a decrease in tusk size. What has happened since 1963 is difficult to establish. Odd intra-African records such as that for c.l00 tusks passing through Zambia 1973, or 227 entering the Ivory Coast in 1977-78 do little to help understand the present position. The Hong Kong data indicate that in 1978 the average weight of tusk from Zaire was 10.83 kg, which is substantially lower than the 1936-63 figure. However this could well be a reflection of the larger tusks having been exported through neighbouring states. That substantial quantities of Zairoise ivory leaves Africa from the Central African Empire is widely known in the trade, and I have seen invoice evidence of Zaire ivory moving on C.A.E. permits. The process has been confirmed independently by Dr. Douglas-Hamilton (pers. corom.) Table 241 illustrates that the Central African Empire does indeed have a high average tusk weight : currently the highest in Africa. Further these weights derive from large samples (Table 61). If the C.A.E. and Zaire records are combined, their average tusk weight is 14.29 -and within the range of the earlier 1936-63 data from Zaire. The salient feature is that while the data are too few to determine trends, there' is evidence that the average tusk weights from Zaire/C.A.E. are heavily biased toward large males and on these grounds alone, the situation is not critical.

PAGE 111

203 Miscellaneous average tusk weights from some other countries are given in Table 242. For interpretation these call for some knowledge of their background and in particular whether they have been subject to any selective process. This is likely, but not confirmed, for all imports recorded from the Ivory Coast which has a tendency to buy smaller -less favoured -sizes than the extra-continental buyers. Figures from the Ivory Coast sources should be considered as minima -where they are sufficient in number to be considered representative. Thus data from Liberia and Nigeria serve only to show that bhey produce some ivory. The Angolan/Namibian data establish points from which future trends can be judged. On their own they only indicate there are elephants old enough to produce this average weight of tusk. The prospect that South-Western ivory is genetically smaller at age than East African would substantially raise the "awarded" age of animals concerned. During this survey I have seen or obtained weights of 1910 southern Angolan/Namibian tusks, representative of 1016 elephants (1910 f 1.88). In the IUCN provisional estimates for 1978 it is believed that southern Angola has only c.2,400 elephants and adjacent Namibia outside of the Etosha Park (where poaching is not serious) has 1640. The data upon which these estimates were made were obtained prior to my seeing or measuring the 1910 tusks from the region. The suggestion thus arises that since the estimate of numbers was made, 1016 (25%) of the 4040 southern Angolan/Namibian elephants have died. At the same time as estimating that there are only 4040 elephants in the region, the IUCN estimates a range of 404,042 km2 (156,000 square miles) in which the 3024 survivors roam; i.e. 136 km2 per elephant -or if in herds of 10, 1360 km2 per herd. At these densities they would be too rare for UNITA guerillas to hunt for meat or ivory; nor would they support the flow of tusks still coming out of the region. While the tusks leaving the area do not yet allow us to say how many elephants there may be, they are strong ground for believing that elephants are

PAGE 112

204 substantially more abundant than represented in the preliminary estimates. The data from Gabon -386 tusks averaging 10.43 kg exported to the Ivory Coast -allow little discussion. The Malawi information shows a national average of 10.07 kg per tusk, with variation between the different regional sources. Mozambique's average from 1975-76 of 6.95 kg may be somewhat dated. The low Rhodesian and South African averages are explained by the elephant management policies of those countries. The Sudanese average of 11.17 kgs from a fairly large sample of 3,405 tusks is indicative of male selection and the reputation the country has among sportsmen as being a good place for 'big' elephants bespeaks some abundance of older males. However, even allowing for extraction of large tusks, the circumference data inject a contrasting sense. Further more complete data are needed to interpret the situation. The data in Tables 236 -242 show heterogeneous situations across the continent as one would Comparison with pre-1914 tusk weights in Table 243 show a correspondingly wide selection of weights in the trade of the past. Selecting one group of averages for extrapolation beyond that group's base is unlikely to give a worthwhile estimate of trade at large. Thus taking the average weight of Bombay's imports of ivory between 1848 and 1855 as evidence of some biological function (e.g. reduced elephant ages) would be to chase a hare. The reason was no more than reflection of American and European entry into Zanzibar's mart, and deflection of large ivory direct to their countries. Similarly, .taking such as one of the modern Tanzanian control estimates of 4.8 kg as a basis for Africa-wide application would be wrong.

PAGE 113

205 The only figures which will provide a reasonable basis for average tusk weight must come from the point at which all aspects of the trade are represented in proportion to their true abundance. Hong Kong is the premier location and it is from here that I accept 9.65 kg from 22,260 tusks as closest to the real average ex-Africa figure. I use it with the caution that it may in fact be higher if direct> 14kg tusk imports by Japan are taken into account. Using this average we can now contemplate the number of elephant represented by the total ex-Africa flow of ivory. The estimates of trade were 1976 991 tonnes, 1977 827 tonnes and 1978 766 tonnes. In turn these devolve into 102,694, 85,699 and 79,378 tusks respectively. At an arbitrary 1.88 tusks per animal, the estimates for the number of elephants represented over the three years are 54,624, 45,585 and 42,222. Of these a proportion died naturally. The only guide we have to go on stems from my Hong Kong analysis of 5,132 tusks of which 21% were judged from natural deaths. This reduces the estimates of direct elephant kills to 43,153 in 1976, 36,012 in 1977 and 33,355 in 1978. As outlined in the previous chapter, 98,371 tusks should become available through natural mortality each year. If their life 'on offer' is 3 years the total available should be of the order of 295,113. If this be so, 21% of ivory recovered from this source over the past 3 years (21,556, 17,996 and 16,669 tusks respectively) are 7.3%, 6.1% and 5.6% of what is available. These are basically higher rates than in the national parks, but are not unreasonable given the far greater manpower and incentive to find them outside the parks. The postulated kills of 43,153, 36,012 and 33,355 elephants are 3.3%, 2.8% and 2.6% of the estimated 1,300,000 elephants extant (IUCN Provisional Estimate 1978). Even if all tusks entering trade are credited as kills, and natural mortality is discounted, elephant deaths accounted for in the ivory trade are

PAGE 114

206 4.2%, 3.5% and 3.2% of extant elephant. Throughout this report I have advised against taking the data presented as precise. I reiterate this here. The data and findings are indicators of the order of things and will be exact only by coincidence. As pointed out in Volume 1 Chapter 4 there are elements of the trade -imports by minor ivory countries -which have not been covered. All these factors notwithstanding, the biological evidence from the ivory trade does not support the contention that the trade as such, is responsible for the decline of Africa's elephants. The sum of the off takes is well within what most biologists would accept as a "sustainable yield". This is, of course, a continental overview. At a regional level -specifically East Africa -there is evidence of accelerated population decline in the 1970s. The declines are not new. There was evidence of them in Kenya and Uganda decades ago. Their change of speed has undoubtedly a strong connection with the returned value of ivory; but quite what this is and whether it was a primary or secondary stimulus is a matter for discussion.

PAGE 116

.'(1