Citation
Ian Parker Collection of East African Wildlife Conservation: The Ivory Trade

Material Information

Title:
Ian Parker Collection of East African Wildlife Conservation: The Ivory Trade
Creator:
Parker, Ian.S.C.
Publication Date:
Language:
English
Physical Description:
t.ypescript report plastic spiral bound

Subjects

Subjects / Keywords:
Ivory Trade
Africia wildlife

Notes

Abstract:
"A consultancy undertaken for Dr. Iain Douglas-Hamilton on behalf of the united states fish and wildlife service of the department of the interior, and the international union for the conservation of nature and natural resources, Morge, Switzerland."
General Note:
The Ivory Trade which consists of the commerce in ivory, biological aspects, discussions and recommendations and tables.
General Note:
Ian Parker Collection Re: East African Wildlife Conservation.
General Note:
Box 17: Galana Game Management Scheme: 21pp. green leather bound transcript report with gold embossed title.

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



-/S' C 6/ -.














































































-7



















(G












GALANA GAME MANAGEMENT SCHEME


S -JUNE 1963


APRIL 1960













LIST OF CONTENTS


1. INTRODUCTION P. 1

2. THE SCHEME AREA P. 2
3. FINANCES AND ADMINISTRATION P. 3
4. STAFF Po 4

5. DEVELOPMENT P. 6

6. RAINFALL P. 7
7. TRANSPORT P. 8

8. GAME CROPPING P. 8
9. GAME COUNTS P.15
10. LICENSED HUNTING P.16
11. ILLEGAL HUNTING P.16

12. REVENUE AND EXPENDITURE P.18
13. CONCLUSION P.19

1i. APPENDICES

























INTRODUCE I ON



The Galana Game Management Scheme commenced in April 1960.

Its objective was to establish game management as an economic

and feasible land use competitive with stock-rearing and

agriculture in certain areas.

An additional, though important aim, was to provide legal

employment for the Waliangulu and other local people who

previously lived by illegal hunting.
This report covers the Scheme's first three years, April 1960
- June 1963.







2.


THE SCHEME AREA


1) GEOGRAPHICALLY.

The Scheme comprises some 3000 square miles in the

hinterland of Kilifi and Tana River Districts of the Coast Region.

Its exact location and boundaries are given in Fig. 1.

With the exception of a few small sandstone hills, the land

lies below an altitude of 1,100 feet above sea level. The western

half of the Scheme is very flat, whilst the eastern portion is

undulating country. A number of seasonal water courses traverse
the latter area from west to east (Figo 2.) Permanent water is

S provided by the Galana (synonymous with Sabaki) River and by

Koromi and Dera waterholes (Fig* 2.)

Rainfall ranges from average annual falls of over 30 inches

in the extreme east to about 10 inches in the west, Correspondingly

there is considerable variety in the vegetation (Fig, 3.) Close

to the coast thick bush and forest occur. Common trees in the
latter are Afzelia quanzensis, Bombax rhodognaphalon and

Brachystegia spiciformiso West of this belt the vegetation merges

into light Diospyros mespiliformis forest and parkland in which

there are extensive tracts of fire-induced grassland. In turn

this merges into thick Acacia Commiphora bush with large numbers
of Euphorbia robecchii, Again in this area there are considerable

fire grasslands. In the lowest rainfall zones the larger acacia

species are replaced by Commiphorae and to a lesser extent by

Delonix elata, Euphorbia robecchii also becomes less abundant.

Dense patches of Sansevieria spp. characterise all areas.

A preliminaary checklist of the grasses occurring is given as


appendix no. 4


2) POLITICALLY.

Human distribution in relation to the Scheme is given in

Fig, l. Legally there is no such land status as a' Game

Management Scheme.' Hence the present Scheme sanctity has had



























































0


GALANA SCHEME


TSAVO NATIONAL
PARK


16


32


48


M I L E S
O-G aORMA GALLA PASTORALISTS
PC = POKOMO CULTIVATORS
CG = GIRIAMA CULTIVATORS
CT= TAITA CULTIVATORS


It= CHARCOAL BURNERS
GS = GIRIAMA STOCKMEN
DS = DURUMA STOCKMEN
TS = TAITA STOCKMEN


RIVERS SEASONAL RIVERS ROADS ... RAILWAYS- i4-


FIG I.


B:I;::
m

















































SEASONAL
DERA WATERHOLE
KOROMI WATERHOLE
SCHEME BOUNDARY


SCALE
I INCH= 16 MI


~r, 4.~jz


SCALE: I INCH.16MILES
--w-- SCHEME BOUNDAP
i OPEN FIRE*
/ -INDUCED
V GRASSLAND
COMM I PHORA
S BUSH


FIG 3.


- m








30.


to be enforced through the Crown Lands Ordinance, Through this

the Administration agreed to preclude human activities from the

area, other than those instigated by the Scheme. Exception to
this were the Orma Galla who continued grazing the Triva River

valley.

Apart from a handful of cultivators on the Galana and the

sporadic presence of the Galla along the northern boundary, no

humans inhabit the Scheme north of the Galana.

However, south of the river charcoal burners ravage the

country in pursuit of their trade. Their technique consists of

setting fire' to as large a tract of land as possible to kill the

trees0 A year later these are made into charcoal and a new area

burnt. This practice is inimical to good land use.

Representations were made to the Administration to preclude

charcoal burning within the Scheme boundaries. Though agreed in

principle, they would not evict the burners until Scheme borders

were clearly demarcated. Finance wvas not available for this in
the three years covered so the problem remains acute.



FINANCES AND ADMINISTRATION

Within the structure of normal government accounting, the

Scheme has been run as a business concern. Money granted and

earned was credited to a Fund created for the purpose and all

expenditure met from ito Amounts in this Fund do not revert to

General Revenue at the end of Financial Years as do normal

departmental monies. Ivory and rhino horn are sold by Government

auction. The revenue thus earned is returned to the Scheme Fund


in the form of a "grant" by the Government in the following

financial year. Initial capital was generously provided through

a 10,000 grant from the Nuffield Foundation.

The Scheme is administered by the Kenya Game Department,

though the salaries of officers seconded for the purpose, are met

from the Scheme Fund.








4+


STAFF

Fig. 4 gives details of all people employed by the Scheme

between April 1960 and June 1963, including Wardens and Drivers

seconded from Government. For 14 of the 39 months covered by this
report there was one warden instead of two. This deficiency had

adverse effects on progress.
The 4aliangulu and other local hunters whom the Scheme was to

afford permanent employment are referred to as participants.
Initially they were paid a very low basic wage with rations, and

scope to earn a bonus on all tasks done- It was hoped that this
would increase incentive to work. However as the system proved

cumbersome and was not appreciated by the men, it was abandoned in

favour of a somewhat higher basic wage with no bonus. Their terms

are: a 6 month probationary period at Shs 40/- per month on the

satisfactory completion of which a man would be awarded an
increment of 5/-, 10/- or 151/- depending on the qualities he had

displayed. Every man received an annual increment of 1/-, and

could be awarded additional increments at any time for increased

responsibility. All participants and labourers received a ration

of maizemeal, milk, tea, sugar and salt worth 20/- per month, and

in addition obtained free meat and fat from animals killed which
was valued at approximately 12/- per month. All were entitled to

free medical treatment,

Originally fear was felt that the 4aliangulu, being nomadic

hunters, would not take to regular employment. However, the

following figures do not show this to be the case. Total
Waliangulu employed between April 1960 and June 1963 = 133. Of


these 5'8 were still employed on 1st July 1963 with an average

length of service of 21.8 months. Five men had completed more

than 36 months' service and 16 more than 24 months. Some 75 men

had been employed but left the Scheme before June 30th 1963 with

an average employment period of 8.3 months. The average for the

133 participants who were employed is 14.2 months. It has










been previously worked out that the average Waliangulu man has
2.2 dependants. This being the case the Scheme has supported an
average of 230 people per month,
Labourers and artisans were engaged on temporary terms when
development projects such as building or road cutting were more
than the participants could handle. These labourers were drawn
mainly from the Giriama and Waliangulu tribes.
During the period covered by this report, some 89571 was
paid out to Participants and Labourers in salaries and rations.


Month Wardens Clerk
Apr 60 1
"i 61 1 1
" 62 2 1
" 63 2 1
May 60 1 1
" 61 1 1
" 62 2 1
" 63 2 1
Jun 60 1 1
" 61 1 1
, 62 2 1
" 63 2 1
Jly 60 1 1
" 61 1 1
" 62 2 1
Aug 60 1 1
" 61 1 1
"t 62 2 1
Sep 60 2 1
tp 61 1 i
"I 62 2 1
Oct 60 2 1
" 61 1
" 62 2 1
Nov 60 2 1
" 61 1- 1
"t 62 2 1
Dec 60 2 1
" 61 1-- 1
" 62 2 1
Jan 61 2 1
" 62 2 1
" 63 2 1
Feb 61 2 1
" 62 2 1
" 63 2 1
Mar 61 2 1
" 62 2 1
" 63 2 1


FIG. 4____
Drivers Artisans Participants LabourersTotal
15 16
2 2 35 17 58
2 1 66 1 73
2 o62 25 92
20 22
2 1 35 17 57
2 64 1 70
1 61 65
S23 44 69
2 1 35 7 47
2 64 1 70
1 59 -63
1 19 31 53
2 1 46 7 58
2 -63 21 89
1 2 23 33 61
2 1 44 24 73
2 61 19 85
1 2 25 52 83
2 1 46 10 61
2 61 33 99
1 2 40 50 96
2 1 46 3 54
1 63 30 97
1 2 38 49 93
2 1 48 3 56
2 66 28 99
2 2 38 26 71
2 1 48 2 55
2 64 22 91
2 2 40 17 64
2 1 59 12 77
2 66 31 102
2 2 39 18 64
2 1 60 11 77
2 67 31 103
2 2 41 18 66
2 1 68 1 75
2 63 28 96


Average monthly employment 71.8 72 people.


S


.-WAM


50







6.


DEVELOPiMENT

The Scheme headquarters were situated on the north bank of

the Galana River against the Tsavo Park boundary (Fig. 5.)

Reasons for this were:

i) that some 2,700 of the Scheme's 30000 square miles lie to

the north of the river. Thus if communications were cut

by floods personnel would be isolated inj rather than from
the greater area;

ii) there is a potential causeway or bridge site at this

point*

iii) at the outset it was not clear whether the land south of

the river would be included in the Scheme.

Only temporary buildings were constructed in case experience

should later indicate a more suitable site. Buildings erected

were: 2 Wardens' bandas with servants' quarters

12 4 man huts for participants

9 6 man "

1 Office-cum-store

2 large sheds for preparing trophies and drying meat

1 Garage

1 Fuel store

1 Concrete armoury

5 pens for elephant calves0

Roofs were of makuti or grass thatch, walls of split sisal poles

and the floors of office, armoury, meat sheds and wardens' houses
of concrete. (Total cost of materials for all buildings was

under 600.) Tents were provided for employees when out in the


field.

Two airstrips were constructed and an old one at Dakadima Hill

made usable (Fig. 5.)

In April 1960 only one very rough road and some even rougher
Land Rover tracks existed in the area (Fig. 5.) Since then,

somewhat over 400 miles of mnotorable tracks have been cut, 227 of























































~I00 *


HADU


IMALINDI


/


FIG. No.5


m -, SCHEME GRADED ROAD O SCHEME H.Q.

SSCHEME CUT TRACK =.--SCHEME BOUNDARY

g ROADS OTHER THAN SCHEME X =AIRSTRIPS

........... = ROUGH TRACK MADE PRIOR TO SCHEME.


*
*
*
*
*
*
*


*
*
*
*
*
*
*
*





70


which have been graded. Of the latter figure 50 miles from

Mackinnon Road to the Galana were made by the National Park

authorities with 1,000 given them by the Scheme for the purpose.

(Apart from serving as a major route of access to the Scheme,
this road also demarcates the Park boundary between the Galana

and Mackinnon Road,)
The Scheme purchased an International B450 tractor combined

with a Trans Atlas grader in 1961, Road making costs worked out

per mile as follows&
Rough handcut track 7

Handout, then graded 1+
Floods in the latter half of 1961 did such damage to roads

and tracks made prior to this that the majority had to be
completely remade. Had this flooding not occurred, the mileage
of roads and tracks cut would have been far greater. The necessity

for, and the amount of time required to make roads, were greatly

underestimated at the outset. More time was spent on cutting

tracks than on any other activity.


RAINFALL

Rainfall records were kept as from August 1960, Results are

shown in Fig, 6. These are felt to be representative of the

extreme west of the Scheme only. No records are available from
other areas but with the kind help of Dr. P. Glover of the

Veterinary Department, an attempt has been made to gauge rainfall

through the types of vegetation (Fig, 3.)

FIG. 6


Yr Jan Feb Mar Apr May Jun Jly Aug Sep Oct Nov Dec
60 *.05 .06 4o34 1.03 1.81
61 .20 1.87 1.13 1l.8 0 .12 1,15 .2918.12 3.*0 8,13 2.46
62 .30 .37 1.31 1.76 0 .05 0 .75' .11 .08 2.05 1.50
63 .33 .02 2.71 1.29 2*31 .91

Unusually heavy rain fell late in 1961 causing extensive

flooding. This also occurred throughout the catchment area of the















FIG NO: 7
Most Scheme
Buildings
were damaged
by the floods

















FIG NO: 8
The River rose
more than
30 feet above
normal level
at this point















FIG NO: 9
The River at
normal level
photographed
from the same
place as
Fig. 8.






8.


Galana, which reached an unprecedented level and destroyed a

great deal of the riverine vegetation (Figs. 7, 8 & 9.)


TRANSPORT

The Scheme transport complement comprised I long wheel-base

Land Rover and 2 5 ton Bedford diesel lorries. Overall running

costs for the three vehicles have worked out at Shs 2/40 per mile

even though much of the driving has been through bush away from

roads. As a result of this bush running, tyre costs were high

and amounted to somewhat over 1,000 in the three years.


GAME CROPPING

At the outset no information was available on the abundance

of animals apart from elephant. Thus exploitation in the three

years was confined to this species.

aigephant.
Though no game counts had been carried out in the area prior

to the advent of the Scheme, it was known that more than 3,000

elephant inhabited the adjacent Tsavo National Park. It was also

known that there was some movement between the two areas and for

lack of better information, 3,000 was taken as the figure on which
to base Scheme cropping. Simpson and Kinloch suggest that an

elephant population can replace an annual loss of lO% through

natural increase. (Appendix 1. Annual Report Game & Fisheries

Department, Jganda Prot.5 1953.) This was used as a guide but

the annual Scheme quota was set as 200 until further data became

available. Until the ratio between sexes was known, males and.


females were cropped equally.

Hunting commenced in July 1960. At the start, lack of roads

restricted accessible areas and kills were a matter of luck. After

rain, when water was available everywhere, elephant moved

constantly and were hard to keep up with. Both these difficulties

decreased as tracks were cut until, in the last year, it was rare







9.


for a hunting party not to come up with elephant, det weather

occasionally made roads impassable though this obstacle was not

as serious as predicted. The floods in 1961 were an exception
to this, and restricted movement for nearly 3 months.

During the first year no attempt was made to kill more than
one elephant at a time. The meat ivory, ears and feet were

removed from the carcass where the animal was shot and taken back

to a base camp. The flesh was then cut into thick strips and

hung on racks in any available shade (Fig. 10) ; it was then cut
into thin. ribbons and left to dry (Fig. ii.) In December 1960,

this procedure was changed. One lorry was equipped with a winch

and ramps enabling the whole elephant, minus the viscera, to be
*
loaded in six pieces. The advantage in this was that fewer men

were required in the field and the animal arrived in base camp

in a shorter time. However, in the second year the technique was,

again modified. By this time every effort was made to kill two
or three elephant on a hunt. Together with hide and bone, it was

not possible to load 2 elephant onto a 5 ton truck. As no good
price was obtained for hide or bone, it was decided to leave these

where the kill was made and remove only the meat, tusks, ears and

feet as done originally. Raw meat was carried on the lorry in two

metal tanks 4' x 3' x 31'.
At the outset meat was dried without salt. However once it

became possible to handle 2 or more elephant at a time, salt was

added to the meat as it was put into the tanks, This extended the
time it could be kept in the field by several hours, but resulted

in uneven salting. This eventually led to salting in the field

being abandoned. Latterly the meat was lightly salted after being


cut into the final .thin strips. Also in the last six months, all

drying was done in a thatched shed as much had been lost through

rain when hanging in the open,
A gang of 35 men could handle two elephant a day with ease

and on several occasions dealt successfully with four at one time.

Approximate costs of a hunting operation work out as follows:























FIG NO. 10
The meat is
cut into thick
strips and
hung on racks
(Photo
E.DeA. Hill)



























FIG NO: 11
It is then
cut into thin
ribbons and
left to dry.
(Photo
F.W. vkoodley)


I 1 1111 it 11 1101mom- .







10.


35' men for 1 day at an average monthly wage
of Shs 80/- per man 93.45 Shs 100
Cost of ammunition at Shs 5'/- per round,
+ rounds per elephant, 2 elephant 40
SCost of transport 100 miles at Shs 2/40 per
mile 240
Cost of supervising officer 60

Total Shs 440
or 22
Income from 2 elephant at an average price
of 63.3 each 126.6 127
Profit therefore =i0' per hunting trip if two elephant can

be shot, and they are shot within 50 miles of the base camp.

81,738 lbs of biltong were produced in the 3 year period.
S 70,000 lbs were sold at an average price of 89 cents per lb.
2,000 lbs were given to neighboring cultivators as compensation

for damage done to crops, and 8,290 lbs were issued as rations to

employees or destroyed by Dermestes. The flesh of a number of
elephant decomposed before it could be dried. In a carefully
controlled experiment with 20 elephant an average of 5'00 lbs per

beast was. obtained* However, the overall average for the period
covered is much lower through wastage, decomposition and pilfering.
The quality of biltong produced was poor. Though it was

possible to sell it, sales were slow and in April 1963 it was

decided to turn out a higher quality product. Insufficient time
elapsed for sales of this to get under way in the period covered.

(Though out of place in this report it is perhaps worth recording

that this better quality product was selling at Shs 4/- per ib in

limited quantities by July 1963.)

Ears and feet were skinned and salted. A considerable demand


exists for the former, prices being Shs 40/- (large) and Shs 30/-
(small) per ear. Feet did not sell well, though in June 1963 a

buyer expressed interest in purchasing all at Shs 40/- each. There

was also a moderate demand for elephant hair bracelets,

Carcass use was on the whole inefficient. Bone, sinew,
viscera, skin and an inevitable small quantity of meat left on the

carcass were wasted. However, without capital to develop







11.


facilities for handling these products, little could be done.

A total of 274 elephant were shot producing 13,78- ibs of

ivory at an average of 50.3 lbs per beast. Of animals killed,

136 were males and 138 were females. (Four were killed for

research or during experiments with inmmobilising drugs.) Scheme

finances were at a low ebb after the 1961 floods and as a hasty

measure to balance accounts, 25 elephant (included in the total

of 274) were shot for ivory only, in June 1962.

Zoos afford a profitable market for the sale of a limited

number of elephant calves each year. However, it would have been

unfair competition against professional trappers had the Scheme,

being a Government concern, competed for the sale of live animals

on the open market. As a result, those animals caught had to be

sold at greatly reduced prices : 250 per female and 125' per

male, to local trappers0 Males are unpopular in zoos as they

have a reputation for being dangerous. A total of 18 calves were
caught by various methods (app. Io.) Of these 5 were sold and

the remainder died0

Four of those caught were found to be afflicted by a Babesia

(Brocklesby D.v. & Campbell H.,E E.A* ild Life Journal, Vol. 1,

p.119o) Of these four, only two lived0 The two that survived were

cured with a series of phenamidine isethionate injections. It is

felt that this Babesia may have been responsible for the other

deaths, but this is surmise as most carcasses were not carefully

examined. Most of the calves were caught after extensive bush

fires had destroyed much of their food source. They were thus in

poor condition.
Though it was strict policy not to shoot cows with unweaned


calves, 4 were shot and attempts were made to rear the calves.

Only one survived.

Aerial observations were made as frequently as possible to
count elephant and observe their movements0 It was believed that

Scheme and Park beasts were closely linked if not part of a







12.


homogenous population. This work was therefore carried out in

close co-operation with Park authorities0 Apart from numerous

local reconnaissances three extensive flights were made over the

combined areas in October 1961, February and June 1962. In

September 1962 a detailed count revealed 15,603 elephant

(Glover J., E.A. Wild Life Journal, Volo 19 po30.) Of these,

3,282 were in the Game iM4anagement Scheme, and the majority were

in very thick bush and forest and most difficult to count. It

was felt that the 3,282 represented a maximum of two thirds of

the elephant in the area, making the probable total nearer 5,000O

Since September 1962, ground observation of the extent to which

vegetation is browsed in comparison with the Tsavo National Park

raises the possibility of an even larger population.

Despite the data obtained from aerial flights it is still

insufficient for absolute certainty on the size of the elephant

population or the ratios between males females, adults and

juveniles. However, from what has been learned there is sufficient

information to put the culling programlne on a sounder basis. From
the study of aerial photographs of large herds of elephant, there

is a strong suggestion that the ratio between the sexes in adults

S is equal (Glover Jo., E.A. Wild Life Journal, Volo 19 p.37o) This

suggestion is supported by Perry's statement "oo.it is obvious
that cows do not in fact greatly outnumber bulls in the present

population..." referring to elephant in Jganda. (Perry J.S.,

Phil, Trans. Royal Soco Lond, Vol. 237, "The Reproduction of the

African Elephant9 Loxodonta africana.") From these observations

it is assumed that the sex ratio at birth is also approximately

equal. (Scheme data from an incomplete series of foetuses examined


suggest that there may be a preponderance of males at birth, but

as yet the sample is too small for consideration.)

Perry (p.108) found "...it is evident that the female East

African elephant in the wild state generally breeds at about 10

years of age9 sometimes earlier.0.", and ".oothat the capacity to

produce young is, in some cases at least, maintained until







13.


extreme old age."11 These observations are borne out by field

observations in the Scheme, A.M.D. Seth Smith while a Game

Warden with the Scheme, observed a female elephant of about 7%6"

at the shoulder by itself with a newly born calf. Comparison of

this shoulder height with "Fatouma" and "Sampson", both tame

elephant which have grown up under natural conditions at the Park

Headquarters in Voi, and whose ages are known, suggests that the

cow observed by Seth Smith was under 10 years of age. There have

been other observations of cows of similar size with calves, but
the fact that they were in herds raised some doubt as to whether

the calves were actually theirs. Suffice it to say that it appears

almost certain that all females will be breeding at 1i years of

age. Numerous observations have been made of apparently old

animals with small calves at foot bearing out Perry's finding that

breeding continues into old age. Perry (p.108) is also of the
opinion that there is an average calving interval of 4 years, and
that this figure is maximal0 Again with the known growth pattern

of the tame elephant at Voi, observations in the field tend to

support this.

The following Figure for a culling programme is drawn up on

these premises:-

i) The sex ratio for old and young is equal
I0
ii) That there are 8 young for every~adult females(Glover J.,

E.A. Wild Life Journal Vol. 1, p.37.)

iii) That a female produces a calf once every + years from the

age of l onwards to 60 years. (The latter conjecture based
on the fact that they consistently breed well into Told age')

iv) That there is a 5O% mortality rateAin the young born each


year (Conjecture to introduce conservatism.)

v) That the Scheme population is comprised of 59000 animals*

vi) That the calving rate is regular every year; thus as cows

calve once every 4 years only one quarter of the adult cows

calve and one quarter need to be served in any one year.

vii) That if one quarter of the cows need serving at a rate of











one cow to one bull per year, three quarters of the adult

bulls are redundant for breeding purposes.
viii)That there will be no mortality through old age as these

beasts will be removed through culling, J&c&.1 'i tj L1k

ix)That as it takes 14 years to reach breeding age, only one

14th of the total young available will be recruited into the

adult groups in a year.

x)That the cows shot will be those verging on senility and that

as the annual age groups are proportional to the rate of

recruitment the senile animals will constitute l/+6th of the

adult females, there being a 46 year period between the

recruitment age (14) and senescence (60.)

In the mathematical calculations of the table only whole

numbers are used. Where a fraction has been involved it has been

ignored; thus 1.8 would appear as 1 and not 2. This has been done
to introduce a further measure of conservatism. It will be seen

that by culling 120 males annually and only the senile females the

adult males decrease while the overall population increases. By
the tenth year though much reduced, the adult males are still more

than half of the adult females. However, from this point onwards,

females and juveniles have increased to the point where the

pressure can be lifted from the males and distributed evenly over

both adult groups9 When this is done and the total kill

maintained at the same level, the adult male population halts its

decline and commences to increase. The other sectors of the

community also maintain an increase though at a reduced rate. The

population can be held static at any time or depressed by


increasing the females killed to equal or exceed the recruitment

rate into the adult female class.






15.


FIG. 11
YEA A DUL T i MATURE C U L L E D
_MALES FEMALES MALES FEMALES MALES FEMALES TOTAL
1 1785 1785 719 71 4 120 38 158
2 1716 1798 775 775 120 39 159
3 1651 1814 833 833 120 39 159
1590 1834 888 888 120 39 159
1 33 1858 941 941 120 40o 160
6 180 1885 991 991 120 40 160
7 1430 1915 1040 1040 120 41 161
8 1384 1948 1087 1087 120 42 162
9 1341 1983 1133 1133 120 43 163
10 1301 2020 1179 1179 120 43 163
11 1265 2061 1223 1223 80 80 160
12 1 1272 2068 1265 126 80 80 160
13 1282 2078 1304 130 80 8o0 160
14 1295 2091 1341 1341 80 80 16o
15 *| 1310 2106 j 1377 1377 80 80 16o

It is suggested that this plan forms the basis of Scheme
culling in the future, pending further more complete data. It is
based on a resident Scheme population of 5,000 and takes no account
of the periodic forays of elephant from the Tsavo Park into the
Scheme. Due to the destruction of habitat in the Park it has been
deemed desirable to kill as many of these as possible when they
come out of the Park. However, this source of elephant must be
regarded as separate and can only be exploited irregularly.

Crocodile.
In 1961 and 1962 the crocodile on the Galana were exploited
to a limited degree. Fifty two were shot and their skins sold for
approximately 200. Details of length, weight, sex and stomach

contents were kept (see app. 2.)

Leo pard.

Four leopard were shot and their skins sold for 200. These

animals appear fairly numerous throughout the area though no


attempt was made to assess their numbers,


GAME COUNTS

In May and June 1963 limited road strip counts were carried
out in the centre and western part of the Scheme north of the
Galanao The results were not conclusive as insufficient mileage







16.


was completed in each area. However, the preliminary results

suggest that through the contribution of elephant the area may

support as large a live weight of animals as some of the better
plains game areas such as southern Masailand (see app. 3.)


LICENSED HUNTING

Though the Scheme was regarded as a normal controlled area

hunting block in which licensed hunters could, and did, hunt as

any other game areas, no revenue was received by the Scheme from

this source. However, as the Scheme is an experiment into the

overall potential of wild life in the area, the results of sport

O hunting must be taken into consideration.
Forty hunting parties conducted by professional hunters

visited the Scheme, Their stays varied from a week to a month.

Taking it that the average safari costs the sportsman approximately

1000 per month, and that the average visitor spent 10 days in the

area, it may be argued that some 13,000 was generated through the

sport hunting potential of the Scheme.
Forty one elephant were killed by licensed hunters and an

absolute minimum of 5,000 worth of ivory removed from the area by

them in this way. Combining this figure with the previous safari
costs, some 18,000 resulted from sport hunting over the three

year period.
Needless to say, the fact that the Government refuses any

income from sport hunting to be credited to Scheme funds, raises

considerable doubt in the participants' minds about the professed

aims of the Scheme toward them.


ILLEGAL HUNTING

Poaching was serious and appeared to be getting worse

latterly. The immediate cause for this being the Game Departmentts

inability to cope with the situation. Though money was available

from Scheme funds for the formaation of an antipoaching unit, the







17.


legal powers necessary for such work were refused to Scheme

personnel and very reluctantly the project had to be dropped.

Anti-poaching work was carried out sporadically when Game Scouts

could be borrowed from neighboring Wardens. Where roads were

cut and hunting took place, poaching diminished or ceased

altogether. Scheme development therefore had the effect, not of

greatly reducing the number of poachers, but of compressing

poaching into the eastern part of the Scheme, which as yet is

inaccessible.

During the period covered, ivory representing 233 elephant

was handed in at the Scheme Headquarters' Rewaras were paid on

the basis of Shs 2/- per pound for good ivory, 1/- per pound for

defective ivory, -/50 cts per pound for rotten ivory and 6/- per

pound for rhino horn. 6194.4o0 was paid out in this way. Though

most ivory was found while roads were being cut and on elephant

hunts, a considerable quantity was brought in by people not

employed by the Scheme. Of the 233 elephant represented, 82 were

males, 113 females and 38 unsexed. The latter figure includes

16 calves. An arbitrary estimate attributes l15 of these elephant

to poaching.

Between April 1960 and June 1963, 72 poaching cases were

brought before the Malindi courts. Of these 21 were brought by
the Scheme and the remainder by Game Department, Police and

Administration. In the 72 cases, there were 6 acquittals, 3

withdrawn and 63 convictions averaging sentences of 13.l1 months.

This latter figure is very satisfactory. A further ii cases were
/
taken by the Scheme to the magistrate in Voi and successfully

concluded. With the exception of one acquittal, all Scheme cases


resulted in convictions.

Rewards totalling 208.6 were paid out for information

resulting in conviction of poachers and the recovery of 601+ lbs of
ivory representing 18 elephant.

Details obtained from Malindi court records of cases other

than those brought by the Scheme, and from informers whose






18.


information one had every reason to believe, (but could not act on
through lack of personnel) indicate a further 103 elephant killed
illegally in the period covered by this report. This figure,
together with the 15 arrived at through found ivory and the 18
known to have been killed by poachers that the Scheme arrested,
reveals a total of 275 elephant poached in the 3 years. A further
79 deaths were attributed to natural mortality.


REVENUE AND EXPENDITURE


Yearly income to and expenditure from the Game Management
Scheme Fund are given below.


APRIL 1960 JUNE 1961


CREDIT
Nuffield Foundation Grant 8000.00.00
Kenya Government Grants 7000.00.00
Meat Sold 530.00.00
Trophies Sold 165.00.00
Interest 59.15.00


DEBIT
All Salaries
Vehicle purchase,
Running expenses
Firearm purchased
Ammunition
Tools, stores et(
Buildings
Track cutting
Rewards
Aerial work


1575o5.19.68
Credit Balance 1773.01.41


4310. 0o4.19
3 4173,o4.28
1261.0O6.96
3 339.19.00
4-.02.60



8912.18. 29
383*12.64
13982.18.29


2.


CREDIT


JULY 1961 JUNE 1962


Nuffield Foundation Grant 2000.00.00
Kenya Government Grant 6000.00.00
Credit balance forward 1773.01.41l
Meat Sold 981.00.00
Trophies Sold 172.00.00
Interest 61,06.82


10987.08.23


All Salaries 4307.09.01l
Running expenses 2187.00.49
Ammunition 74.1000
Tools, stores etc 497.09.98
Rewards 363.05*.25
Aerial work 432.01.81
Track cutting 365o07.97
Purchase of Grader 2158.16.00
10386.00.5 1


Credit Balance 601.07.72


JULY 1962 JUNE 1963


DEBIT


Kenya Government Grant
Credit balance forward
Meat Sold
Trophies Sold
Elephant calves Sold


6000.00.00
601.07.72
1605.08.00
842.00.00
780.00.00


All Salaries
Running expenses
Ammunition
Tools, stores etc
Rewards
Aerial Work
Track cutting


9829.06.92
Credit Balance 309.02.71


1493814.37
2521.10.54
79.05'.30
523.06.*8
374.169.65
339.13.16
742.17.61
95o20. 0o.21


1.


DEBIT


3.


CREDIT







19.


Revenue from Scheme ivory sales is credited directly to

Government and returned to the Scheme as "grants" at a later date.

However, as it is a basic product its value must be taken into

consideration in assessing results. During the 3 years,

Government received 15,132 from ivory sold In addition there

were a further 1,800 lbs on hand at the Scheme Headquarters on the
30th June 1963. Valued at Shs 151- per lb., this brings the

overall value of ivory produced to 16,482.

Combining ivory with other produce total earned income

amounted to 21,557.08.99, some 2,557 more than Government have
put into the Scheme in grants (19,000). However it would be

incorrect to term this real profit as the Scheme has not recovered

the full 10,000 'investment' of the Nuffield grant.


CONCLUSION

In the course of the first three years, a variety of

obstacles were encountered. Most serious of these were poacherst

and charcoal burners' activities, Unless these two pastimes are

effectively controlled further time and money will be wasted on

wild life management in the area. Future plans must include

measures to eradicate these activities with priority over other

considerations.
Through Scheme and licensed hunting, the capture of calves,
poaching and natural mortality, 687 elephant deaths are accounted

for. poaching is responsible for more than a third of this total

and its seriousness is apparent.
Exploitation has been at a relatively primitive level.


Efficient techniques for carcass use and disposal are obvious and

urgent requirements. However their development will require

considerable capital. Though ivory was the mainstay of the

economy, the 36.5- tons of dried meat produced for human

consumption are equivalent to an offtake of some 2,000 scrub

cattle,






20.


The two districts occupied by the Scheme benefited in a

variety of ways. Through exploitation of the area's wild life, a

hitherto unproductive region has been opened up and a valuable

source of employment created. Hitherto men from the hinterland

areas had to go long distances to Mombasa or the large coastal

estates for work. The monthly average of 230 people dependant on

the Scheme received some 8,500 in salaries and food. All

employees were up to date in tax payments, the majority for the

first time in their lives,

In the short span of three years it has not been possible to

fully demonstrate the economics of wild life management as a land

use. However, despite limited capital, the lack of any previous

example for guidance and having to recuperate from severe floods,

the potential has been made much more obvious. When Scheme

income is combined with the tourist (sport-hunting) revenue, it

is seen o,0000 was earned through the area's wild life. It

should be borne in mind that capital spent has been entirely on

means for exploiting the existing potential Nothing has been

spent on developing the basic asset of the land and its wild life.

If progress is to be made beyond the present subsistence level

such development must be undertaken,

To the degree possible within Government, the Scheme has been

run as a commercial venture. As Government is not designed to

indulge in business this has resulted in numerous inefficiencies

and retarding of progress. Reluctance to compete with private

enterprise, which in itself is correct, has restricted development

in several spheres. This in turn has its unwanted repercussions

on employment. It is also doubtful whether Government, with its


massive commitments in other fields, is able to contribute large

sums foT capital development at this stage. Now the Scheme is

established, and three years of Government pioneering have revealed

the various possibilities in the area, it seems correct that

private enterprise should be invited to take over development and

exploitation. Returns from this type of land can only be obtained









21.


through heavy investment. The ares potential is sufficient to

warrant this

It is therefore recommended that, providing the Government's

responsibility to create employment for the local people can be
satisfactorily transferred, serious consideration should be given

to future development being carried out through private enterprise.

I would like to express my sincere thanks to Dr. P. Glover,

Mrs. A. Rickards and Mr. Mo Franks of the Veterinary Department

for their assistance in the preparation of this report.





I.S.C. Parker
Game Warden


14th September, 1963.
ISCP/ACP







APPENDIX 1.


ATTEMPTS TO IMMOBILISE AND CAPTURE ELEPHANT


In 1962 and 63 a number of attempts were made to immobilise

both adult and immature elephant. Equipment used was either

generously supplied by the Palmer Chemical & Equipment Co. Inc.,

of Atlanta, U.SoA, or locally made variations of the Palmer designs.

Immobilisation drugs were conveyed to the elephant and

injected by standard Palmer "Cap-Chur" syringes. Five different

types of weapon were used to project the syringes:-

1. The Crossbow. A crossbow was supplied by the Game

Department. The particular weapon had several disadvantages,

the most outstanding of which was the high trajectory. In a

flight of 80 yards the dart rose as much as 11 feet above the

straight line between weapon and point of aim. In thick bush,

though quarry might be clearly visible with no obstructions

along a straight line from the hunter, twigs and branches

above one's sight line obstructed the dart's flight. Again

in dense vegetation it was difficult to move fast balancing

the dart on crossbow, and was reminiscent of an egg and spoon

race. Verdict on the crossbow: only of use in open country

where the high trajectory is not a hindrance.

2. The "CIiP-CHUR" Pistol. This weapon had too limited a
range with the 10 and l5 cc darts used, to be of use on

elephant.

3. The "Harthoorn" Powder Gun, Dro Ao Harthoorn kindly

loaned the Scheme his 'powder' gun. This comprised of a

sleeve the same calibre as the dart, fitted into a 12 bore

shotgun, The breach end of the sleeve was chambered to take


a cut down 10 cartridge. This weapon had a fairly flat

trajectory up to 50 yards, and proved successful.

4 The "Palmer 32 Gauge". This was a rifle by Beretta made

to take standard calibre syringes. This weapon had an

exceptionally flat trajectory and could be used up to

distances of 100 yards,







II.


5o The "Palmer /8 Gauge". This was a rifled 20 bore which

took a correspondingly larger dart. Again with this rifle,

long ranges were possible but the weight of the darts at high

velocities was sufficient to inflict a severe wound,

The "Harthoorn", 32 gauge and 5/8 gauge trajectories and

velocities could all be altered by increasing or decreasing the

charges in the propelling cartridges.

The darts were altogether too weak to withstand the velocities

set up by all the weapons used except the "Cap-Chur" pistol. Care

had to be taken that ranges were in excess of 35 yards. Under this

range the darts would almost certainly break on impact, while

above it, there was a chance of them remaining intact, Commonest

dart failures were snapping needles. As a result they were

locally reinforced with steel ribs which improved performance

The aluminium syringe casing often bent jamming the plunger, or

the threads by which it was screwed onto the needle-base,stripped.

Initially actual injection of the drug was obtained by the

reaction of an acid on a 'Cap-Churt tablet. The gas thus

generated drove the plunger down the syringe forcing the drug

through the hypodermic nooeedle. Prior to firing, the tablet which

* was placed in a recess at the back of the plunger, was sealed off

from the acid by a brass plugo On the initial acceleration the

plug fell out allowing the acid to contact and react with the

tablet, (See Fig.lo) This had disadvantages as the reaction, and

thus injection, started before reaching the target. This was ove

overcome by placing the tablet in another plunger in the tail of

the dart with the recess facing forward, This way the brass plug

did not come out until deceleration on impact with the target.


It was also found that the bubbles of gas from the reaction

of acid on the tablet formed a coat around the tablet and slowed

down the reaction considerably. This was overcome by powdering
the tablet beforehand*







III.


FIG. 10 Direction of Flight

Drug .cid I __it'l Tail Tuft

Plungeor7 Brass Plug ----
'Cap-Churl Tablet


__ Direction of Flight

.- ..Drug I Acid I Tail Tuft
a 'Plungers IC'apChur^tablet
Brass Plug


Latterly injection was achieved by the explosion of a

detonator in the recess behind the plunger on impact with the

target. Some of the detonators were found to have insufficient

power to drive the plunger fully home, This resulted in partial

injection only A small measure of black powder added to the

detonator overcame this drawback. Injection by this method was

almost instantaneous and when several ccs were injected, created

penetrating wounds. Penetration was eliminated by blocking the

end of the needle and making several origices in the side of the

needle. This limited the injection woubd to the surface though

even this was undesirable, A further advantage in having a solid

pointed needle with holes in the side is that the possibility of

a core of skin plugging the bore is obviatedo
A number of detonators exploded when the cartridge ignited,

and ejected the drug while still in the gun barrel0 This was
eliminated by placing a cushion of paper between the detonator and

screw-in-tail of the dart.

A total of 30 elephant were hit by darts and of these only


7 were immobilisedo Damage to the darts on impact and the
subsequent non-injection account for the majority of failures.

Though the principles are sound, the darts are not robust

enough for the task they were intended to perform.







IV.


The first elephant darted was "Lali" a bull of approximately

1,300 1500 lbs weight. Darted with a mixture of:

Phenicyclidine 300 mgms
Themelon 1.5 gms
Scopolamine 100 mgms
Hyalase 1 tablet
The dart injected into the abdominal muscles on the right flank.

15' minutes after injection, Lali collapsed on his brisket. While

in this state he was loaded onto a Land Rover and driven back

to Scheme Headquarters and placed in a pen,

3 hours 35' minutes after injection, he attempted to raise his

forequarters,

3 hours 50 minutes, a similar but more determined effort.
* 4 hours 10 minutes stood up trembling violently.

4 hours 35 minutes collapsed again.

Hours 10 minutes, again got to his feet

10 hours later fully recovered. Soldo

The second elephant was "Bisambala" a cow of approximately

1,500 Ibs weight. Darted with a mixture of

Phenicyclidine 400 mgms
Scopolamine 100 mgms
Hyalase 1 tablet

S Injected into the ribs behind the left shoulder

15' minutes later collapsed.

1 hour 15 minutes, tried to rise. Vehicle not on hand so given.

250 mgms Largactylo Lay quiet until

2 hours 40 minutes when she again tried to rise. Night fell and

it became impossible to move the elephant until the next

morning During the night a further


Phenicyclidine 320 mgms
Scopolamine 100 mgms
Largactyl 250 mgms

were given. Put in pen 19 hours after initial injection.

24 hours seemed alert, eating although unable to raise itself

on its forelegso Continued feeding well but 3 days later

still unable to use its forelimbs so destroyed.







V.


The third elephant was "Diwayui' a bull of between 19500 and

2,000 lbs weight, Darted in the left shoulder with:

Phenicyclidine 400 mgms,

16 Minutes had travelled about half a mile, then collapsed,

Salivating copiously, 5 0 mgms Scopolamine injected intra-

muscularly, and salivation stopped.

2 hours 35 minutes, bracing legs and trying to rise,

Hours five minutes back in pen, conscious, drinking, but

unable to use forelegs. Rear legs apparently all right and

able to support weight.

24 hours later, alert, eating and drinking but unable to use

-forelimbs. Three days later no improvement so destroyed,

The fourth elephant was "Honey", a cow of about 2,000 lbs

weight. Darted into right hind quarter:

Phenicyclidine 200 mgms.

20 minutes later still standing but unable to co-ordinate

movements, was approached and roped, As soon as she was

roped a 500 mgms Themelon injected intramuscularly. Fell

down several times but immediately regained her feet.

1 hour 30 minutes after initial injection co-ordination improved

considerably, a further 100 mgms Phenicyclidine injected.

Loaded into crate still standing. Salivating copiously
given 50 mgms of Scopolamine and 250 mgms of Largactylo Rested

weight against side of crate.

8 hours 30 minutes fully recovered in pen. Soldo
No. 5 was a bull of 8,000 ibs estimated weight. Darted

behind the left shoulder with

Phenicyclidine 800 mgms


Morphine 2 gms
Hyocine 400 mgms

30 minutes approximately after darting, found standing, trembling

violently and incapable of any co-ordinated movement,

1 hour after injection it fell on its side.

1 hour 45 minutes, given lethidrone to counteract morphine, but

no reaction observed.







VI.


2 hours 30 minutes later, the beast died Attempts were made to

keep the animal cool. Its rectal temperature taken before

death was 106F. On dissection it was found that the

injection, effected by an explosive detonator, had scored

the animal's heart.
No. 6 was a very large bull estimated at more than 10,000 Ibs

weight Darted in the rump with

Phenicyclidine 1 gm
Morphine 2 gms
Hyocine 400 mgms

This animal travelled for 10 minutes after being darted before

becoming incapable of further movement0 It stood for a further

10 minutes before collapsing. Lethidrone was given intravenously

without effect0 The animal died approximately 2 hours after the

initial injection0 Every effort was made to prevent it overheating

and it was shielded from the suno

No, 7 was a bull of approximately 4000 lbs weight. It was

darted in the rump with

Phenicyclidine 250 mgms
Morphine 250 mgms
Hyocine 100 mgms

10 minutes later it collapsed. Lethidrone (1 dozen phials) was

injected intravenously with no immediate effect.

4 hours later the animal stood up. As it was still very 'drunk'
it was walked onto a lorryo

7 hours after initial injection, it became very aggressive and

threatened to demolish the lorry0 A further 100 mgms of

Phenicyclidine were injected and it lay down shortly after0

12 hours after initial injection it was again able to stand up.

It was too big to sell, so when in the following evening it


demolished its pen and walked away, no attempt was made to

recapture it. It was fully recovered, feeding and drinking by then

Summing up, immobilisation was not carried out successfully0

Equipment failure was the major reason0 Three of the young
elephant captured, numbers 1, 4 and 7 were immobilised and

recovered successfully0 Two lost the use of their forelimbs, and







VII.


though fully recovered in every other sense had to be destroyed.

The reason for this disability is obscure, Two large elephant

darted died. Again the reasons are unknown, but it appears that

large elephant cannot spend more than a limited amount of time

on their sides.

Four calves were caught through pursuit with a vehicle. A

suitable calf is selected and cut out of the herd. Once well

away from the other elephant the calf is seized, or if too large,

roped. This technique is effective but has the disadvantage of

creating considerable disturbance among herds and severely stresses

the animals caught. Three of the animals chased and caught this

way were afflicted by a Babesia (Brocklesby DoW. & Campbell H.,

E.A. Wild Life Journal Vol. 1, p.19.) Of these three two died.

N. Steyn Esq., of Arusha caught all four Of those obtained by the

technique outlined in this paragraph.

Apart from those darted, caught from a vehicle or when their

mothers were shot, several calves were found alone, pursued on foot

and caught. All these were in poor condition and died.


I.S.C. Parker







APPENDIX 2.


NOTES ON CROCODILE (C.niloticus)
IN THE GALANA RIVER


INTRODUCTION,

During the Game Management Scheme's first three years,

52 crocodile were killed and examined. The information thus

gained forms the basis of this appendix Where possible data is

compared with Cott's results (Cott H,.B., Trans Zoo. Soc. Lond.,

Vol. 29, Part 4, April 1961.)


DESCRIPTION OF AREA.

Crocodile occur throughout the Galana's 25 mile stretch

through the Scheme. In this section the average width of the

river is between 180 and 270 metres. Though subject to

spectacular flooding during the rains, for the greater part of the

year most of the water is less than a metre in depth, and very few

pools exceed lo.5 metres. For most of its course, the river is slow

flowing and its bed comprised of shifting white sand, except where

it crosses the occasional rock seam. As a fish habitat it seems

poor. This seems borne out by the stunted growth of Tilapia

mossambica in this particular stretch. Normally this fish, an

algae feeder, attains weights in excess of 1 ibo However 670

netted in the Galana averaged 0.96 oz., though many were actually

breeding Other fish occurring in this area, are a Eutropius

nr depressirostris, Synodontis mossambicus, Clarotes laticeps,

Clarias mossambicus, Labeo gregorii and several small species of

Barbuso Few of these attain weights in excess of 2 ibs.

The river banks are lined by a thin belt of large trees. In


this Acacia elatior, Hyphne corcea and Poulus ilicifolia are

prominent The latter grows nearer the water's edge than the

others and also occurs on the larger sandbanks with dense reed beds

(Phr agmites s p.) Many mammals ranging in size from warthog to

elephant water on the river and in the dry seasons, so do large

numbers of birds (Particularly Ploceidae who are associated with







IIo


the reed beds and Columbidae and Pteroclididie who come in vast

numbers to drink from the sandbankSo)


SIZE, SEX R..TIO & BREEDING CONDITION,
The 52 specimens obtained were either baited and shot at

night with the aid of a spot-light, or stalked while on sandbanks

in daylight0 Crocodile hunting was carried out as a sideline to

other Scheme activities and was somewhat sporadic0 Kills were as

follows: 1961 1962

June 2 April 1
July 12 June 18
August 4 July
October 10

All but 3 specimens were sexed and all had their weight and length

recorded0 The smallest collected was a male o0.44 metres long,

weighing 0.23 kg; the largest, a female, was 2o55 metres and

weighed 62.2 kg. Crocodile longer than 2.5 metres are uncommon

on the Scheme section of the Galana,

Of the 49 sexed,' 25 were males and 24 females. This is a

ratio of 51.02 to 48,98 which compares well with Cott's ratio of
49.4 to 50O.6 for 411 specimens under 2.5 metres in length (p.251.)

Gonads of the 49 were examined and seven males and seven

females were found to be in breeding condition. All were in excess
of 1.8 metre in length. Breeding condition was determined in the

male by the presence of enlarged testes and in the female by those

with ova larger than 2 mm in diameter, the presence of oviduct eggs,
or by distended oviducts and flaccid discoloured ovaries,

indicating recent laying. A further 3 specimens in excess of this

length were not in breeding condition and included the largest


specimen obtained., (The other 2 were a male of 2.11 m, 27.73 kg,

and a female of 1.99 m, 20,0 kg.) Fig 1 gives details of weight

and length for those in breeding condition. It will be seen that
there is little difference in the size at which the sexes start

breeding, or in the relation of length to weight between sexes.









+ = MALE
o FEM4LE


60t


FIG. NOS: 1

A comparison of.length
and weight between
breeding males and
females over 1.8 metres
in length.


W
E
I 5o-
G
H
T

I
I
N

K 40-
I
L
0
G
R
A
M
S
30


20J
Io


Minimum
M. Kg.
Males 1.o89 25.0
Females 1.96 26.4


Maximum
Ma Kg.
2.48 54.5
2.43 52.7


Average
M. Kg.
2.13 35o0
2.17 35.4


0 /


+


/


0


p


+


/


0


0


d


109


2oU


2.1
LENGTH


2.2
IN METRES


2.3 24 2


7Z


".


J


!












-----r--- -= Scheme data, each dot representing i croco
.-.-.. = Females as depicted in Cott's Fig.19, p.25
200+.. -- -. = Males ,,
4 W !
2 0.
E
G
H
T


N
FIG. NO: 2
K
The relation of I /
length to weight of L
0
Scheme crocodile G 100 /
together with Cott's R 00
data on breeding /
crocodiles from theM/ /
Kafue FlatsLuangwa
Valley and Upper
Zambesi as depicted
in his Fig.19,p.256o.
-'
1. 00
0 0
7*:
e ~~ ae
o *^ ^ "

0 0 0
05 1 00
L T N
-., .,0'-,.... 0LENGTH IN MTRES
0.5" 1.0 io 2.0 2. 3.C

LENGTH IN METRES


dile
6
/ /
/
I

/ /1

/ /
/ /
/
/
/ /
/
/ /
/
/
/
/
/
/















~3.~5







III.


Fig. 2 gives the weight-length curve produced from the

measurements of all 52 specimens obtained. In addition Cott's

curves for crocodile in breeding condition from the Luangwa Valley,

Kafue Flats and Upper Zambesi are included for comparison. (Taken

from his Figol9, p.256.) It will be seen that Galana weight

length ratios form a remarkably tight pattern up to the maximum

lengths obtained. Cott found a marked difference in male-female

length-weight ratios in breeding crocodiles as depicted in the

figure


FOOD.

The stomach contents of 24 specimens were examined

Occurrences of the different types of prey are as follows:

Insecta Araneida Crustacea Mollusca Pisces Amphibia Reptilia
13 1 7 0 3 2 0
Ayes Mammalia Carrion Vegetable matter
2 4 11 2

Coleopotera formed the bulk of insect remains found, but Hemiptjqr

and Orthoptera were also identified

All the Crustaceans eaten were fresh water prawns(Caridina spp

and M- crobrachiun spp.) which abound in the river. No mulluscs at

all were recorded0 Three types of fish were eaten- Tilapia,

Eutro ius and a Goby. However, on several occasions I have found

scales of Labeo ge iii on sandbanks with crocodile tracks about

them. It may be presumed that they do feed on this fish to some

extent0

The frogs caught were all of the same species of Rana No

reptiles were taken as food0 Turtles occur frequently in the

Galana and have been observed feeding off carrion at the same time


as several crocodile of about 2 metres in length No attempt was

made by the latter to take the turtles, who evinced no fear.

QUele and doves (Streptopeli sp.) were the only birds

represented in the sample and Archer once killed a specimen in the

same area which had taken 7 doves0 I have personally witnessed

a crocodile taking sandgrouse (Eremialector decoratuso)







IV.


The only two mammals it was possible to identify were a small

rodent taken by the smallest crocodile collected (O.44 m) and a

warthog taken by a specimen of 2.43 m. Two other specimens

contained mammal hair

Crocodile came readily to baits of rotten meat placed in the

river, and tended to concentrate near the Scheme Headquarters

when elephant hunting was going on and offal was being dumped in

the river. This must have replaced other normal foods to a marked

degree. On several occasions they were known to leave the water

and remove biltong from racks on which it was drying some 20 yards
from the river0 When dealing within a bait, their technique was to

seize a portion in their jaws and roll over several times until

the meat broke away from the carcass. It was then swallowed. This

method is also described by Cott po303. It was noticed that when

a buffalo was used as bait the crocodile were unable to break into

the carcass until the skin had decomposed considerably. Cott's

statement p.303 "...crocodiles are well able to feed from a large

fresh carcass" must apply to larger crocodile than occur on the

Galanao

The most surprising stomach contents found were 2 wild figs

(fruit of Ficus sp.) in the largest crocodile shot0 One other

specimen had 2 small pieces of stick which may well have been
accidentally ingested-


STOMACH STONESo

All 24 crocodiles whose stomachs were examined contained

stones0 These stones were weighed and the results shown in Table].

TABLE 1.
Length group iNo stomachs Stone weight as a percentage of total
in metres examined body weight
jMaximum __ Minimum Mean
0o3 0.5 1 Pre-sent but too slight to weigh
05 1.0 I 2 0o64 0.)42 .053
1.0 1.5 12 1.64 0.16 0.92
1.5- 2.0o 3 o1.27 0.35 0.78
2.0 2.5 4 i 1.16 0.64 0.87
2 3.0 1 j 1.27







V.


One crocodile of 1.99 metres excluded from Table 1, had

stones comprising 2,66% of its total weight. This animal had the

latter half of its tail missing. All stones found in stomachs

were water worn quartze pebbles or gravel.


GENERAL .BEHAVI OUR.

Small crocodile less than 0.5 metres in length were found to

be common at night in very shallow water around reed beds. In

daylight they concealed themselves in thick vegetation at the

water's edge, though on occasion some were seen basking on rocks

away from cover0 Several very small specimens were seen some

six feet up Phragmites stems overhanging water0

There seemed little difference in diurnal habit between

crocodile of O.5 and 2.5 metres, except that as a general rule the

smaller the animal the shallower the water it was found ino

Animals in this size range would emerge to bask openly on

sandbanks between 7.3.0 and 8.30 a.mo*, returning to the water about

10.30 to 11 a.mo., and again between 3 and 3.30 pom., to 6 p.mo

They were rarely found out of water after dark except when feeding

on a bait or offal placed on the river bank0 After the 1961 floods

a crocodile of approximately 2 metres was seen in a water pan some

7 miles from the river,


DISCUSSION.

From Cott's Fig.35, p.281, it will be seen that approximately

65% of a 2.e metre crocodile's diet is comprised of fish, reptiles,

birds and mammals. For crocodile in excess of this length the

percentage reliance on these four sources becomes progressively


greater A crocodile's technique for the capture of reptiles,

birds and mammals is "stealth, surprise, and a sudden final burst

of speed" (Cott, po302.) Thus there must be good concealment until

the final stage. In the Galana the fish population is poor. A

2.5 metre crocodile would therefore have to place greater reliance

on the capture of reptiles, birds and mammals However the Scheme








VI.


Galana's overall shallowness makes for poor concealment (for

protection as well as hunting.) Also it would be extremely

difficult to drown a mammal of any size as there are very few

spots which it could not keep its feet, and drowning seems the

only method whereby a crocodile kills large mammals. It thus

appears that the section of the river through the Scheme is an

unsuitable habitat for crocodile 2.5 metres and over This fact

must either restrict growth to about this length, or on attaining

it, the crocodile emigrate to more suitable environs. In this

particular case that latter seems feasible. In the lower reaches

of the river near the coast and the highest reaches of the Athi

S (the Galana before being joined by the Tsavo) crocodile in excess

of 3 metres occurred before skin hunters eliminated them The

main tributary, the Tsavo, runs through the Tsavo National Park,

is safe from hunters, and still contains large crocodile., Both

the upper and lower reaches of the Galana and the Tsavo have deep

pools in which crocodile can drown large prey. Records of frequent

man-eating in the former two and of an attempt to secure a buffalo

in the Tsavo indicate that large mammals are tackled in these

waters.

Cott shows that for breeding crocodile in Rhodesia there is

a difference in the relation of length to weight between males and

females which is not apparent with Galana data, Mcllhenny as
!
quoted by Cott p.2 7, shows a similar disparity between breeding

males and females of the American alligator (Alligator

mississippiensis) but also shows that for the first five years

there is no disparity between the sexes. Through lack of evidence

Cott assumes a similar pattern to be the case with C. niloticus,


and this assumption is strongly supported by Galana data where no

difference between sexes is evident and specimens were much
smaller than those from which Cott obtained his evidence. If the

Galana data is representative of that part of the growth curve

before a difference in males and females is obvious, it may be







VII.


deduced from Cott's Figo.13, po246 (Age-length) that the divergence

in growth rates between sexes becomes apparent after 11 years

(2.4 m) though Cott's data on Fig.19, p.256 suggests it is

somewhat earlier0 Growth in the Co niloticus being slower than

A. mississippiensis (Cott Fig.13, po246, C. niloticus takes

approximately 22 years to reach 2.6+m7 Fig,20, p.257 (males only)

A mississippiensis takes approximately 10 years) it is to be

expected that the divergence would occur later.
The smallest Galana breeding female was 0.21 m smaller than

the smallest female on Uganda breeding grounds from a sample of

between 1500 and 2000 crocodile. (Cott quoting Pitman p.o25.) The

smallest male was 1.89 metres. Cott believed sexual maturity in

males is achieved at between 2.9 and 3.3 m (po253.) The Galana'

female figure does not decrease the minimum recorded female

breeding figure by more than a negligible amount. However5, the

male figure does. This may be a result of the lack of larger

males. Inter-male fighting is recorded (Cott, p.267) and in some

instances attributed to the acquisition of mates. The difference

in weight between 2 and 3 metre crocodile is so great, and the

latter so much more massive, that combat between the two would be

pointless for the smaller size. The presence of larger males may

delay the onset of breeding conditions in smaller males,
Cott records that in populations in which specimens occurred
well in excess of 3 metres, the sex ratio for specimens under

2o5 metres (young animals) was approximately equal though this was

not the case in specimens over this size. The fact that Galana

specimens were under 2.5 m and produced an equal sex ratios the

fact that there was no disparity in weight for crocodile of both


sexes of equal length and the fact that females in other areas can

breed at about the same length as those in the Galana, all suggest

the crocodile population is composed of young animals. This

favours the belief that larger specimens emigrated If the
population does contain old crocodiles, which remain static but are







VIII.


restricted in growth through habitat, it would be logical to

expect some evidence of similar though stunted patterns of growth

rate differential between sexes and some variation in sex ratios

in the larger size groups as occur in the adults of Cott's

populations However, emigration can only be conclusively proved

by marking individuals and following their movements over a long

period. A worthwhile comparison would be with crocodile from a

locality such as Baringo in which they do not exceed 2.5 metres

and have no opportunity to move elsewhere on attaining this length

as in the Galana.

That the stomachs examined all contained stones suggests

they must have a specific function. The Galana specimens contained

stones more constantly in greater quantity than Cott's at a similar

size, Though there was variation up to lo67% and down to 0.16%

the majority were much closer to the average of 0.85% of the

animal's total weight. Cott's conclusion that the purpose of stone
is the acquisition of a negative bouyancy seems sound. One

particular specimen not included in the figures of variation and

average already given, was minus half its tailor It had lost this

a considerable time before collection as the would showed well

healed scar tissue0 The tail must constitute one of the least

bouyant parts of a crocodile's anatomy0 This loss must have

increased the specimen's positive bouyancy considerably0 It was

therefore remarkable that this animal contained stones comprising

2.66% of its total weight0 One cannot but deduce that this was

a measure to counteract its loss, and gives even greater credence

to Cott's reasoning0
The only variance in general behaviour to Cott's findings


is the fact that crocodile between the lengths of 0.5o metres and

1.5 metres were frequently seen during daylight. Elsewhere,

according to Cott (p.278) this size range is conspicuous by its

absence during daylight and cannibalism is given as the reason for

this group's secretiveness0 Reasons for the situation on the








IX.


Galana would again appear to be the absence of large specimens

and the difficulty of stalking through shallow water.


C ONCLUSIONSo

Crocodile in the Scheme section of the Galana seldom exceed

2.5 metres in length as the habitat seems unsuitable for larger

specimens, Those that occur breed at shorter lengths than

recorded by Cott. There are grounds for believing that on

exceeding 205 metres crocodile emigrate either up or down stream.

Types of prey taken are similar to Cott's findings, Stomach

stones occur consistently and support Cott's deduction that their

*rpose is the acquisition of negative bouyancy. A minor

difference in general habit to Cott's findings in the 0.5 1.5 m

group length is attributed to peculiarities of the environment,


I.S.C. Parker







APPENDIX 3

ROAD STRIP COUNTS


Expense and the density of vegetation over much of the Scheme

is such that it is impracticable to count animals other than

elephant and giraffe from the air. Cheaper alternative methods
must therefore be used to assess numbers and distribution. In

Southern Rhodesian bush country a 'road strip', counting technique

has been used with some success. (Dasmann & Mossman, Journal of
Wild LifeManagement, Vol. 26 No. 1, p.101 and No, 3, p.262.) This

procedure was applied to some areas in the Scheme during April and

May 1963 and the results of these initial counts are given below.
The strip count is based on the principle that a strip visible

from a road is representative of the area traversed, If the area

of the visible strip can be assessed and the animals in it

accurately counted, the results applied to the whole area would

give an estimate of the total animals in it. Dasmann & Mossman's

procedure adopted in the Scheme counts was as follows: Observers

drove slowly through the area to be counted. Each animal seen was

recorded and its distance at right angles to the centre of the road

measured in yards. This was done initially by pacing each

observation, but latterly by judgment of eye with occasional paced
checks to ensure consistency. In the case of herds observed the

distance was to the animal furthest from the road. Animals on the

road counted as one yard off. Mileage was recorded. The data

acquired was subjected to the following formula:
a) the number of animals counted was divided into the total


miles covered to give the average number of animals per mile.

b) the distance of each observation from the centre of the

road was totalled and divided by the number of observations to
give the average distance of each observation from the road.

c) the average distance from the road represents the average

width of the strip visible to one side of the road. Multiplied by

two this gave the average width of the representative strip

traversed by the road.






II.


d) the average width of the representative strip was then

divided into 1760 yards to give the average number of strips per

mile.

e) the number of strips per mile was then multiplied by the

average number of animals per mile to give the average number of

animals per square mile.

f) the compacted formula appears thus:-

number of animals seen 1760 yards
total miles covered average sight distance x 2
= the average number of animals per square mile.

Dasmann and Mossman recommended that at least fifty miles be

covered before tabulating data. As a check a further twenty five

miles must be covered and the results added to the original figures.

If when worked out, there is no significant difference in the

answers the count is valid, but if there is significant variation

then further data must be collected until results are constant.

Difference in size and habit among the various species counted

resulted in varying degrees of visibility. A giraffe would be

plainly visible at 200 yards in bush that concealed rhino, oryx

or eland at a lesser distance. To include such an observation

with those of smaller animals would lead to an erroneously wide

representative strip where the latter were concerned. Generally
if a Lesser Kudu was visible at a given point, Gerenuk and Grant's

Gazelle would also be seen at that point if there. Therefore the

data from the Lesser Kudu observation is valid in computing the

representative strip for the two other species and vice versa.

In the analysis of results animals have been divided into

Ivisibilityt classes. These classes are:


I Giraffe
II Dik-dik and Warthog
III Lesser Kudu, Grant and Gerenuk
IV Waterbuck, Oryx, Eland, Rhino, Zebra, Buffalo and Ostrich.
The width of the representative strip for a species has been

arrived at by combining the distances from the road of all animals

within the class, and dividing by a total of the class observations.







III.


There were 8 separate count areas determined by the

vegetation and soils occurring (for location of the count areas

see map.) Brief descriptions of the areas, A to H, are as follows:

AREA A = A minimum of 27 square miles around the base of

Lali Hills. Coarse red soil, stony in places. Vegetation open

Commiphora woodland with Terminalia spinosa and Delonix elata

abundant. Much bare earth and dense clumps of Sansevieria spp.

mainly around the base of trees. Grasses and low vegetation

too sparse to burn well and area generally unaffected by fires.

Mileage covered- 4, in 6 separate counts. All the animals

listed on the tables have at one time or another been seen in

Area A.
AREA B = A minimum of 172 square miles. Mainly grey

'black cotton' type soils. Very open Commiphora parkland with

Dobera glabra abundant and extensive tracts of grass. Severely

damaged by fire in 1962 and also burnt extensively in previous

years. All the animals listed on the tables occur in Area Be

Mileage covered = 99, in 9 separate counts,

AREA C = A minimum of 640 square miles. Very friable red

soil, Open Commiphora woodland with frequent Euphorbia robecchii

Sand occasional Delonix elata. Acacias mainly small species

under 10 ft. Grasses sparse, the most noticeable being

Aristida adscensionis. Dense clumps of Sansevieria spp. around
the base of most Commiphorae and Delonix. There is much bare

earth and fire can only travel after an exceptionally good rainy

season. The extensive fires of 1962 did not greatly affect the

area. Mileage covered = lii- in + separate counts. All the

animals listed on the tables have been recorded in Area C, with


the exception of Waterbuck.

AREA D = A minimum of 96 square miles. Mainly grey

'black cotton' type soils, in places very light and friable.

Acacia- Commiphora parkland with some dense areas of

Sansevieria spp. Grass cover good. Severe fire in 1962 killed

many trees, especially Acacias. Mileage covered =40 in 2






IV.


separate counts. All animals listed on the tables have been

recorded in Area D, with the exception of Waterbuck.

AREA E = A minimum of 60 square miles. Mainly grey soils.

Poor fire-induced grassland. Only trees occasional Euphorbia

robecchii. Severely burnt in 1962 and in previous years. Buffalo,

Giraffe, Dik-dik, Gerenuk, Lesser Kudu and Waterbuck have not

been recorded in Area E. Mileage covered = 16, in 3 separate

counts.

AREA F = A minimum of 123 square miles. Grey soils with

occasional large areas of reddish loam. Very open Acacia -

Commiphora parkland with abundant Euphorbia robecchii. Very

good grass cover. Severely damaged by fire in 1962. All animals

listed on the tables have been seen in Area F, with the exception

of Waterbucko. Mileage covered = 49, in 3 separate counts.

AREA G = A minimum of 12 square miles. Red sandy soil. Thick

Commiphora woodland with Delonix elata, Acacias and an occasional

large Erythrina sp. Not greatly affected by 1962 fire. Grant's

Gazelle, Ostrich and Waterbuck have not been recorded in Area G.

Mileage covered = 18, in 2 counts.

AREA H = A minimum of 12 square miles. Grey soil of

'black cotton' type. Very dense bush. Trees : Acacia bussei,

Diospyros mespiliformis., Mimusops densiflora, Terminalia parvula.

Very little open ground or grass0 Not touched by fire. Area H,

according to records so far obtained, contains no Grant's Gazelle,
Gerenuk, Ostrich~or Waterbuck. Mileage covered- 20, in 2 counts.

Data obtained is shown in the tables 1 8. Each table

contains specific figures as well as class figures from which the

results were calculated. Table 9 gives the square mileage of each


area and the estimated total number of animals.
The eight count areas are adjoining and form a block of

1142 square miles in the centre and west of the Scheme north of the

river. None have permanent water though areas A and B are both

within reach of the Galana. At the time the counts were made, rain

had fallen and pans in every block held water. Unfortunately,





















1 ~


1'


F
jPakadakacha ,';
Dakadima 0 E


c G

S, H
C / D
\ ^- i
\ *^ \\ /v '.'" ..
D/

TSAVO \ B BB DDalal
\ "*, a/'~


L al i .. .
***. 'ky-- -

PARK / /



/ i
L______


-\fl
a
>n
i*a

\
i i ,

Si
\
i i.

i


/
/
/
/
buko


\
\
/
/


, !./ /


\ i\


/
//
/


/
-.-_ ---I


A diagram to show the approximate positions
of the count. areas within the Scheme.






TABLE 1_ AREA A_


Species Miles


Giraffe
Dik-dik
Warthog
L. Kudu
Grant
Gerenuk
Waterbuck
Oryx
SEland
Rhino


44

44


S446


44
44
1^.
wb


Total seen


11
70
7


2
7
1


S P -E C IC J, A SCL SS Answer per sq. m i Corrected to
No. Obs Total Yds Avo. Yds No. bs Total Yds Av. Yds at m at 44 m nearest animal
1 ] oo.40 40 .5 6


2
*,3


; 1 j
!i


615
42

21
173
50


13 67
18
14l


47
47


21
4325
50


S651

63


223


13.86
' 13o86
S15o75


10
1


44.6


1001
.0 i
7.62


0.64
o j6


101
10
8


1


Zebra j i
' Buf f alo | : | j I i |
Ostrich ; I \ I i


TABLE 2 -. AREA B

SSpecies Miles Total seen S P E C I F I C... __CLAS Sj Ans
SNo. Obs Total Yds Av. Yds Noo Obs Total Yds. Av Yds at


Giraffe ___________ _______________ __ ____
Dik-dik 99 36 2 491 l964 12
, Warthog__ '.i _
L. Kudu 99 8 4 349 87c25 29 1859 6li 1
Grant 29 1859( 6
Grant 99: 7 3 216 72 | 29 189 6.1 i
Gerenuk 99 I 22 1294 + 58.82 i 29 1859 64.1 4
Viaterbuck I
Oryx 99 33 4 807 2018 8 1307 1 163.4
SEland 99 8 2 275 137.5 8 1307 634
Rhino i
Zebra I i
Buffalo i I "
Ostrich 99 i 4 1 150 15 0 8 1307. 63.4
------------------------------------3--01-------


wer per sqo m i Corrected to
53 m at 99 m Inearest animal


o8 o.29 9 1l

o 7 2 1 ,0 l1
.23 0.97 i i '
,92 4.8 .

.01 1.80 2
0.44.I


0.76 |


I


i: ;
- -


i


r-


___63___~~. ___






TABLE 3 AREA C
.. . . ..- -- .- .. ...- -
'-- ~~S P E C I F I C C LASS.... ... As
Species s Miles Total seenASS As
Species Miles Total seen No. bs Total Yds Av. Yds No bs Total Yds Av Yds at
Giraffe I 4 : 1 6 65 10.16 7 ------- 1
Dik-dik 114 130 91 i1032 11.3 93 109 11.77 108
Warthog 114 ,i 6 2 5. 6 283109i77
L. Kudu I 16 11 303- 27.o55 19 523 27.52, + 5
Grant
Gerenuk 114 11i 8 220 27.50 19.._23.272
, Waterbuck.
* Oryx i 114 9 2 136 68.0 5 143 28.6 + 2
SEland :
Rhino 114 4 2 6 3.0 5 143 28.6
SZebra 11 4 8 1 1 1 5 1 43 28.6 '
Buffalo ;
Ostrich I
- -+ -
Osrich-------------- --------------------------



TABLE 4 AREA D

Species Miles Total seen SPECIFIC CLASS Ans
________No. Obs Total Yds Av. Yds No. Obs Total Yds Av. Yds at
Giraffe 0.40 .9 1 .. 200 200
Dik-dik 40 23 I 214 13.44 17 215 I0. 05
lWarthop 4o 10 111 17 215 10.05
L. Kudu 404 5 261 65.25 ;
Grant :
Gerenuk ,
waterbuck
Oryx 40o 2 2 110 55 3 : 175 58.33
Eland 40 2 1 65 65 3 175 58.33
Rhino .
Zebra
Buffalo :
Ostrich ---


wer per sq. m; Corrected to
70 m at14 m nearest animal
o56 loT06O. 1 ___
0o 85*84 86
_____ o3 273 ____ 3 ___
.912 4.488 4

.113 3.086
.37 2.4.29 2

1.079 1
2.159 2


at 79 miles.




wer per sq. m Corrected to
m at 40 m nearest animal
S0.99 1"
50.36 50
S21.90 22
S3.035 3


S0.754 1
S0.754 1




------- TABLE 5 AREA E

Species Miles Total seen- O S P E C IF CC L A SS C AnsL
No. Obsi ;Total Yds AvY Yds No. Obs Total Yds Av. Yds at
GiraffeJ
Dik-dik
Warthog i i
L. Kudu
Grant 16 25 4 535 133.75 10
Gerenuk ;
Waterbuck
Oryx
Eland .16 1 1 50 50 3 46o 15.33
Rhino, : ;
Zebra 16 8 1 300 300 3 460 15.33 2
Buffalo
Ostrich 16 2 1 110 110 3 460 15.33 0



TABLE 6 AREA F

Species iMiles Total seen. S PCLA'SSAnS
j ___ No. Obs Total Yds Av. Yds No. Obs Total Yds Av. Yds at
Giraffe 49 11 351 70.2 2
SDik-dik 49 33 21 127 605 22 128 .82 0
Warthog 149 1 1 1 1 22 128 '.82 3
SL. Kudu 149 32 17 789 216.41 224 1136 47.34 12
' Grant 49 8 4 15 2 38.0 24 1136 ,7.34 3
Gerenuk. 49 24 3 -195' 65.0 24 1136 47.34 2
Vi Waterbuck
I Oryx 49 21 2 370 185.0 5' 615 129 2
Eland
Rhino 49 4 3 275 91.67 5 615 129 0
Zebra
Buffalo :
Ostrich


wer per sq. m Corrected to
16 m at m nearest animal



.28 10


3588
.871 3

.787 1





wer per sq. m Corrected to
49 m at m nearest animal

.18 10
.086 _L
.9 13
.035 3
.577 3

.923 3
.55681 I







TABLE 7 AREA G
~S P E C I F I C C L ASSiAs
Species Miles Total seen .....
Spce _________ No. Obs Total Yds Av, Yds No. Obs Total Yds Av. Yds at
Giraffe ______ _________________
Dik-dik 18 25 16 1T 9.63 17 179 10.53 l1<
SWarthog... 18 ...1. 1i 25' 22. 0 17 179 iO_1
L. Kudu
Grant i
SGerenuk 18 2 82 26'0130 "
Waterbuck
Oryx i
Eland '
Rhino :
Zebra
Buffalo
Ostrich .



TABLE 8 AREA H

Species Miles Total seenS P E C I F I C C LA S S Ansi
Species 14Miles s Total seen .........
No. Obs Total Yds Av Yds- No. Obs Total Yds Av. Yds at
Giraffe S .
SDik-dik 20 16 12 22 1.833
WarthoI
L. Kudu 20 2 2 37 18-5 i
SGrant
Gerenuk j 1
Waterbuck,
Oryx
Eland
Rhino
Zebra
Buffalo
Ostrich ; '


wer per sq. m Corrected to
18 m at .m nearest animal

E^ li^" "
I. ,: *6
6 4



"Z2 i 8
78












wer per sq. m Corrected to
20 m at m| nearest animal I

401 ;l 38+




i :

j!
L-F' 7T






-- -









TABLE 9'
'-A B 3C D E F DG H
SPECIES 27 sq.m 172 sq.m 64o0 sq.m 96 sq.m 60 sq. 123 sq.m 12 sq.m 12 sq.mr
9 5OI 9


Giraffe

Dik-dik
i Warthog

L. Kudu
Grant
SGerenuk

i Waterbuck

i Oryx

Eland
Rhino
i
SZebra
Buffalo
Ostrich
i ostrich.


162

2727
270


2


2236


.16 172

S 172
81 860


27


344


64o

55040
1920

2560


96
4+800
2112

288


1920


1280 I


96
96


640
1280


172


!














180


60


369
1230

369

1599
369

369


369


123

1
,


1392
60


ll
TOTALS
-I 04, r7
i------


4608 72,033

4 731

60 4,895
1,1411

3,278


2,116
96

763
1,460


232


-Now





V.


there was insufficient time to motor a full 50 miles in any blocks

other than B and C.

Results were worked out at 53 and 99 miles in B, and 70 or 79

and ll4 in C. Variances in results are as follows:

Area B Area C
Per sq. mile Per sq. mile
Dik-dik 3.49 22.16
Giraffe Nil 0 06
Lesser Kudu 0.613 1.424
Gerenuk 0.063 0.027
Grant 0.259 Nil
Oryx 0.786 0.059
Other species Insufficient data for comparison.

In both tables B and C, Gerenuk results show consistency.

Dik-dik show such variation that a great deal more data is

required. Giraffe, kudu, grant and oryx show variance but are

* probably close to the truth.

Waterbuck only occur in parts of A and B and though data from

observations of them is used in computing their class

representative strips, no attempt has been made to assess their

numbers until more information on their limited range is available.

Elephant are not included in the road strip counts as more

conclusive work has been carried out from the air.

It would appear that in addition to a minimum mileage there

S must also be a minimum number of observations for a species before

worthwhile results can be achieved. In area D the figure of

2,112 warthog is achieved through one observation of a sounder of

10 standing in the road. This seems erroneously high even though

it is "diluted" with 16 Dik-dik observations (being in the same

class.) Experience has not suggested that the Warthog population

is any higher in D and in F or A. Similarly the figure of a rhino


per square mile in C is taken from 2 rhino observations and is

most definitely incorrect. This also applies to Zebra in the same

block.
That the effectiveness of the technique varies between

species is borne out by the fact that buffalo were present in all

areas at the time of the count (Except E) but were not recorded

,once. Also rhino, eland and zebra were present though not seen in

all blocks.






VI.


Movement in and out of the area definitely takes place with

buffalo, eland and zebra and appears related to the availability

of water. The animals able to go long periods without drinking,

Lesser Kudu, Grant, Gerenuk, Dik-dik, Oryx and Giraffe are thought

to be fairly static. The latter two species do travel considerable

distances, but seem to be present in the count areas over most of

the year.

Dik-dik are few in areas severely burnt and seem more

immediately affected by this hazard than any other species. After
the serious fires in 1962, a number were found burnt to death in

Area B. Grant are best represented in areas opened up by fire,

Despite the lack of a complete series of counts, even at this

S premature stage it is of interest to make a rough estimate of the

live weight of animals supported in the better counted blocks

B, C and F. In the following assessment, elephant have been

included. From aerial counts the annual density of elephant is

arbitrarily assessed at one per square mile over the whole area.

Rhinoceros and Zebra in Block C are assessed at one quarter

of the number shown in Table 3 as the figures given are believed

to be gross overestimates. Figures on which the calculations are

based are estimated minimum adult weights and are as follows:

Giraffe 1760 lbs
Dik-dik 1 "
Warthog +0 "
Lesser Kudu 150 "
Grant (petersi) 85 "
Gerenuk 70 t"
Oryx (callotis) 320 "
Eland 800 "
Rhino 2000 "
Zebra 500 "
Elephant 6000 "


Ostrich weights are not considered. Average weights have been

deliberately underestimated to be conservative Buffalo, Bland

and Zebra are poorly represented though at times they frequent the

areas in some numbers. This will also tend to bias the results

towards an underestimate.






VII.


Results are given in Table 10 and equated to cattle carrying
capacity based on 20 acres per beast and 5'00 lbs per cow.
~~_____ ______TABLE 10________
Block iS Miles Animal lbs Equiv. No of Carrying capacity acres per
s il per sq.ml cattle p.s.m acres per cow game anml


B 172 7,298 f 48
C 64o 10,1424 20.85
F 123 16,815 33.63


43.89 24.6
30.69 6.35
19.03 16.0


Bearing in mind that these results have been reached through
one incomplete count series, it is nevertheless interesting to
compare them with Stewart & Zaphiro's average biomasses for the
Athi-Kapiti, Mara-Loita, Leroghi, Baragoi, Wamba and Isiolo areas.
(Stewart D.RM, & Zqphiro D.R.P.. "Biomass & Density of Wild
Herbivores in Different East African Habitats" to be published..)
Comparisons are presented in Table 11.

TABLE 11
GALANA SCHEME ST~vART & ZAPHIRO
AREA B C i F I A LMOGHI BARAGOI WAMBA ISIOLO
IIAPITI'LOITAI___ __
LBS S ------ I---------------
per I 7289 10484116815; 11080 16822 6810 1781 5101 13610
sq.m..j.______________________
There is thus a suggestion that through the contribution of

elephant, the biomass of some parts of the Scheme is nearly as
Sigh as that of the Mara-Loita area.

CONCLUSION:

The road trip count technique shows promising results with
some species occurring in the Scheme, but seems ineffectual with
others, notably buffalo. It appears that a minimum of 5'0 miles

motoring in the variety of areas sampled is too low for good
results. Also a statistically acceptable minimum number of
observations for a species must be established. Insufficient data
was obtained so no firm conclusions can be reached. However, there
is a hint that the coast hinterland bush country may carry as heavy
biomass as the Acacia Themeda areas in southern and south
western Kenya, though using different species to achieve this.


I.S.C. Parker






APPENDIX 4.

A LIST OF GRASSES OCCURRING WITHIN
THE GALANA GAMlE MANAGEMENT SCHEME.


Miss Do Napper of the E.A. Herbarium very kindly identified

the grasses collected in the Game Management Scheme, and her work

is greatly appreciated.


PAPPOPHOREAE
Enneapogon elegans
It cenchroides
Schmidtia bulbosa

ERAGROSTEAE
Eragrostis aethiopica
it aspera
S caespitosa
cilianensis
t ciliaris
t horizontalis
"t perbella
"t rigidior
superba
Pogonarthria squarrosa
Cypholepis yemenica
Leptochloa chinensis
t obtusiflora
Dinebra retroflexa
Eleusine africana
Dactyloctenium aegyptium
it giganteum
if scindicum
Drake-Brockmania somalensis

CHLORIDEAE
Enteropogon macrostachyus
it somalensis
Tetrapogon bidentatus
it cenchriformis
it tenellus
Eustachys paspaloides
Chloris barbata
S mossambicensis
S r oxburghiana
S virgata
Schoenefeldia transiens
Cynodon dactylon
tt plectostachyus


1.
2.
3,


4.
5..
6.
7,
8.
9,
10.
11.
12.
13.
14o
15.
16.
170
18.
19,
20o
21.
22.


23,
24.
25.
26.
27.
28.
29,
30,
31.
32.
33,
34,
35.


36.
37.
38.
,39.
4o.
41.


bogdanii
helvolus
marginatus
pyramidalis
robustus
spicatus


AVENEAE
42. Lintonia nutans
STIPEAE
43, Aristida adscensionis
440 i mutabilis


45.,
46.
47.
48o
49.
50.
51.


52.
5.3.
55.
56.
57.
58.
590
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73,
74.
75.,
76.
77.
78.
79.
80.
81.
82.
83,
84.
85.
86.
87.
88.


89.
90.
91.
92.
93.
94.


95.


ZOIZIEAE
Tragus biflorms
1 berteronianus
Latipes senegalensis
Dignathia gracilis
"t hirtella
it nov.?
Perotis hildebrantii
PANICEAE
Panicum coloratum
.chusqueoides ?
deustum
"t infestum
maximum
meyerianum
subalbidum
Sacciolepis curvata
Setaria holstii
"t verticillata
tt nov.?
Echinochloa colonum
it spo
Urochloa gorinii
trichopus
Paspalidium geminatum
Brachiaria deflexa
"I leersioides
t nigropedata
it ramosa
"t serrifolia
"t eruciformis
t pubifolia
glauca ?
H cencaerantha
Eriochloa nubica
t partispiculata
Digitaria aridicola
t macroblephara
t pennata
it rivae
"t setivalva
velutina
Rhynchelytrum repens
Tricholaena eichingeri
Cenchrus ciliaris
prieurii

ANDROPoGoNEAE
Sorghum versicolor
verticilliflorum
Chrysopogon aucheri
Bothriochloa radicans
Heteropogon contortus


Themeda triandra
Ischaemum afrum


SPOROBOLEI
Sporobolus
if
it
it
ft
it



























































































































































































































(



















































































































I







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 EL0TMCCW0_CICTLG INGEST_TIME 2014-05-30T21:17:05Z PACKAGE AA00020117_00024
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 2

/. 5. e. '/->/-1 RKr-R

PAGE 3

./

PAGE 5

,/ GALANA MANAGEMENT SCHEME APRIL 1960 JUNE 1963 J

PAGE 6

LIST OF CONTENTS 1. IN'1'RODUC'1' I ON P. 1 2. THE SCHEME AREA P. 2 3. FINANCES AND ADMINISTRATION P. 3 .I 4. STAFF P. 4 5. DEVELOPMENT P. 6 6. RAINFALL P. 7 7. TRANSPORT P. 8 8. GAl'1E CROPPING P. 8 9. GAME COUNTS P.15 10. LICENSED HUNTING p.16 11. ILLEGAL HUNTING p.16 12. REVENUE Al"JD EXPENDITURE p.18 13. CONCLUSION P.19 14. APPENDICES I

PAGE 7

I N T ROD U C T ION The Galana Game Management Scheme commenced in April 1960. Its objective was to establish game management as an economic and feasible land use competitive with stock-rearing and agriculture in certain areas. An additional, though important aim, was to provide legal employment for the waliangulu and other local people who previously lived by illegal hunting. This report covers the Scheme's first three years, April 1960 -June 1963.

PAGE 8

2. THE SCHEME AREA 1) GEOGRAPHICAL1:(. The Scheme comprises some 3000 square miles in the hinterland of Kilifi and Tana River Districts of the Coast Region. Its exact location and boundaries are given in Fig. 1. lili th tho exception of a few small sandstone hills, the land lies below an altitude of 1,100 feet above sea level. The western half of the Scheme is very flat, whilst the eastern portion is undulating cou'ntry. A number of seasonal "l
PAGE 9

OG .. .. ... ... .... ...... . CT CT FIG I. . .. ... . ... . . ..... ...... ...... . . . .. ..... ... : :;;&-. :. .. 0\ .. .......... . . . . . . . . . . ...... .. . . . . o GALANA SCHEME TSAVO NATIONAL PARK 16 32 48 I MIL E S o-G .. ORMA GALLA PASTORAlISTS PC = POKOMO CULTIVATORS CC = GIRIAMA CULTIVATORS C1= TAITA CULTIVATORS SEASONAL RIVERS "'ROADS N \ I N o I A N -t = CHARCOAL BURNERS GS = GIRIAMA STOCKMEN D S = DURUMA STOCKMEN TS = TAITA STOCKMEN RA IL WA Y S --1"-""--''""'"''''--'1 SA

PAGE 10

F rG ---z-: \ __ 0 _0 o DAKADAKACHA -B e HILL -DAKADIMA--HILL ,----. \ SEASONAL DERA WATERHOLE KOROMI WATERHOLE SCHEME BOUNDARY SCALE I INCH = 16 MILES ANNUAL RAINFALL e ----. ---,IVA .-'" 'V 'f 't---D E L ON I X-.--,;:::::t \ N t SCALE: I INCH .. 1 6 MI LES _.SCHEME BOUNDAR 't' OPEN FIRE -INDUCED '+' '" GRASSLAN D COMMIPHORA-BUSH

PAGE 11

e 3. to be enforced through the Crown Lands Ordinance. Through this the Administration agreed to preclude human activities from the area, other than those instigated by the Scheme. Exception to this were the Orma Galla who continued grazing the Tiva River valley. Apart from a handful of cultivators on the Galana and the sporadic presence of the Galla along the northern boundary, no humans inhabit the Scheme north of the Galana. However, south of the river charcoal burners ravage the country in pursuit of their trade. Their technique consists of setting fire to as large a tract of land as possible to kill the trees. A year later these are made into charcoal and anew area burnt. This practice is inimical to good land use. Representations were made to the Administration to preclude charcoal burning within the Schem$ boundaries. Though agreed in principle, they would not evict the burners until Scheme borders were clearly demarcated. Finance not available for this in the three years covered so the problem remains acute. FINAl'IJCES AND ADdINISTRATION Within the structure of normal government accounting, the Scheme has been run as a business concern. Money granted and earned was credited to a Fund created for the purpose and all expenditure met from it. ilinounts in this Fund do not revert to General Revenue at the end of Financial Years as do normal departmental monies. Ivory and rhino horn are sold by Government auction. The revenue thus earned i s returned to the Scheme Fund in the form of a "grant!! by the Government in the following financial yearo Initial capital was generously provided through a ,000 grant from the Nuffield Foundation. The Scheme is administered by the Kenya Game Department, though the salaries of officers seconded for the purpose, are met from the Scheme Fund.

PAGE 12

4. STAFF Fig. 4 gives details of all people employed by the Scheme between April 1960 and June 1963, including vJardens and Drivers seconded from Government. For 14 of the 39 months covered by report there was one warden instead of two. deficiency had adverse effects on progress. The and other loca l hunters whom the Scheme was to afford permanent employment are referred to as participants. Initially they were paid a very lovv basic wage with rations, and scope to earn a bonus on all tasks done. It was hoped that this woul d increase incentive to work. However a s the system proved cumbersome and was not appreciated by the men, it was abandoned in favour of a somewhat higher basic wage with no bonus. Their terms are: a 6 month probationary period at Shs per mont h on the satisfactory completion of which a man would be a warded an increment of 5/-, 10/or 15/-de pending on the qualities he had displayed. Ev ery man received an annual increment of 5/-, and could be awarded additional increments at any time for increased responsibility. All participants and labourers received a ration of maizemeal, milk, tea, sugar and salt worth 20/-per month, and in addition obtained free meat and fat from animals killed which w a s valued at approximately 12/-per month. All were entitled to free medical treatment. Originally fear was that the Waliangulu, being nomadic hunters, would not take to regular employment. However, the following figures do not show this to be the case. Total Waliangulu employed betl,'18en April 1960 and June 1963 = 133. Of these 58 were still employed on 1st J -uly 1963 with an average length of service of 21.8 months. Five men had completed more than 36 months' service and 16 more than 24 months. Some 75 men had been employed but left the Scheme before June 30th 1963 with an average employment period of 8.3 months. The average for the 133 participants who were employed is 14.2 months. It has

PAGE 13

been previously work e d out that t h e average W -aliangulu man h a s 2.2 dependants 0 i 'his being the case the Scheme has s upp o rted an average of 230 p e r Labourers and artisans were e ngag e d on tempora r y terms whe n development projects such as building or roa d cutting were more than the participants co u l d h andle. These labourers were drawn mainly from the Giriama and wa1iangulu tribes. During tbe p eriod. covered by this report, some ,571 was paid out to Participants and Labourers in salarie s and r ations. Apr 60 11 61 II 62 II 63 May 60 61 62 II 63 Jun 60 61 II 62 63 Jly 60 61 62 Aug 60 61 1/ 62 Sep 60 61 62 Oct 60 61 f1 62 Nov 60 f1 61 62 Dec 60 II 61 If 62 Jan 61 62 63 Feb 61 62 11 63 Mar 61 62 63 1 1 2 2 1 1 2 2 1 1 2 2 1 1 2 1 1 2 2 J: 2 2 2 2 12 2 1 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 1 2 2 1 1 2 2 1 2 2 1 2 2 1 2 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 FIG. 4 2 1 1 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 15 35 66 62 20 35 64 61 23 ia 59 19 46 63 23 44 61 25 46 61 40 46 63 38 48 66 38 48 64 40 59 66 39 60 67 41 68 63 Average monthly employment = 71.8 = 72 people. 17 1 25 17 1 44 7 1 31 7 21 33 24 19 52 10 33 50 3 30 49 3 28 26 2 22 17 12 31 18 11 31 18 1 28 16 58 73 92 22 57 70 65 69 47 70 63 53 58 89 61 73 85 83 61 99 96 54 97 93 56 99 71 55 91 64 77 102 64 77 103 66 75 96

PAGE 14

DEVELOPiVIENT The Scheme headquarters were situated on the north bank of the Galana River against the 'l'savo Park boundary (Fig. 50) Reasons for this were: i) that some 2,700 of the Scheme's 3,000 square miles lie to the north of the river. Thus if co@nunications were cut by floods personnel would be isolated in, rather than from the greater area; ii) there is a potential causew ay or bridge sit e at this point; iii) at the outset it \'vas not clear ",rhether the l and south of the river would be included in the Scheme. Only temporary buildings were constructed in case experience should later indicate a more suitable site. Buildings erected were: 2 12 9 1 2 1 1 Wardens' bandas with servants' quarters 4 man huts for participants 6 man It II It Office-cwn-s tore large sheds for preparing trophie s and drying meat Garage Fuel store 1 Concrete armoury 5 pens for elephant calves. Roofs were of makuti or grass thatch, walls of split sisal poles and the floors of office, armoury, meat sheds and w ardens' houses of concrete. ( Total cost of materi a l s for all building s was under 600.) Tents were provided for employees when out in the field. Two airstrips were constructed and an old one at Dakadima Hill made usable (Fig 5.) In April 1960 only one very rough road and some even rpugher Land Rover tracks existed in the area (Fig 50) Since then, somewhat over 400 miles of motorable tracks h a ve been cut, 227 of

PAGE 15

. . . . .... X'....... DAKADAKACHA ... . .... ........... :,:'" ....... --...... .-". ...... ." .-.;, .... ;:.,'" ..... \ '-;..-J) ," -o"_"_",,-.. -" DAKADI MA \ \ . ......... _:1 ".... -. \ : I DAKABUKO" ( MACKI NNON RD. 5"' MILES :: SCHEME GRADED ROAD .:: SCHEME H.Q. :: SCHEME CUT TRACt< _. = SCHEME BOUNDARY FIG. No.5 ... _-_ .... -.,-ROADS OTHER THAN SCHEME X = AIR5TRIPS = ROUGH TRACK MADE PRIOR TO SCHEME.

PAGE 16

which have been graded. Of the latter figure 50 mile s from Mackinnon Road to the Galana were ; made by the National Park authorities with ,000 given them by the Scheme for the purpose. (Apart from serving as a major route of access to the Scheme, this road also'demarcates the Park boundary between the Galana and Mackinnon Road.) The Scheme purchased an International B450 tractor combined with a Trans Atlas grader in 1961. Roa d making costs worked out per mile as Rough handcut track Handcut, then graded 7 14 Floods in the latter half of 1961 did such damage to roads and tracks made prior to this that the majority had to be completely remade. Had this flooding not occurred, the mileage of roads and tracks cut would have been f a r greater. The for, and the amount of time required to make roads, were greatly underestimated at the outset. Hore time was spent on cutting tracks than on any other activity. RAINFALL Rainfall records were kept as from August 1960. Res ults are in Fig. 6. These are felt to be representative of the extreme west of the Scheme only. N o records are available from other areas but with the kind help of Dr. P. Glover of the Veterinary Department, an attempt has been made to gauge rainfall through the types of ve getation (Fig. 3.) FIG. 6 .. _-.-A-p-r. n;c:1 60 05 0 06 4 .. 34 1. 03 1. 81 611020 1.8 7 1.13 1.48 0 .12 1 .. 15 .29/8.12 3.40 8.13 2.46 I 62103 0 .37 1.31 1.76 0 .05 0 075 .11 .08 2.05 1.50 I 631033 .02 2.71 1.29 2.31 .91 -L. __________ ___ _ ___ .. _. __ _ .... heavy rain fell l ate in 1961 causing extensive flooding. This also occurred throughout the catcmnent area of the "

PAGE 17

FIGNOg 7 Most Scheme Buildings were damaged by the floods FIG NO: J1 The River rose more than 30 feet above normal level at this point. FIG NO: 9 The River at normal level photographed from the same place as Fig. 8.

PAGE 18

Gal ana which reached an unprecedented level and destroyed a great deal of the riverine vegetation (Figs. 75 8 & 9.) TR.A,.r-.JSPORT 8. The Scheme transport complement compr ised 1 long wheel-base Land Rover and 2 5 ton Bedford diesel lorries. Overall running costs for the three vehicles have worked out at Shs 2/40 per mile even though much of the driving has been through bush away from roads. As a result of this bush running, tyre costs were high and amounted to somewhat over ,000 in the three ye ars. GAME CROPPING At the outset no information was available on the abundance of animals a part from elephant. Thus exploitation in the three years w a s confined to this species. Elephant. Though no game counts had been carried out in the area prior to the advent of the Scheme, it 1;las known that more than 3,000 elephant inhabited the adjacent Tsavo National Park. I t was also known tha t there was some movement between the two .areas and for l ack of bette r information, 3,000 was taken as the figure on which to base Scheme cropping Simpson and Kinloch suggest that an elephant population can replace an annual loss of 10% through natural increase. (Appe ndix I. Annual Report Game & Fisheries D e partment, Uganda Prot., 1953.) This was used a s a guide but the annual Schem e quota was set as 200 until further data became available. Until the ratio between sexes was known, males and fem a l e s were cropped equally. I Hunting commenced in July 1960. At the start, lack of roads restricted accessible areaS and kills were a matter of luck. After rain, when water 'las available everywhere, elephant moved constantly and were hard to keep up with. Both these difficulties decreased as tracks were cut m1.til, in the las t year, it was rare

PAGE 19

for a hunting party not to come up with elephant. viet weather occasionally ma6e roads imp assable, though this obstacle was not as serious as predicted. The floods in 1961 were an exception to this, and restricted movement for nearly 3 months. During the first year no attempt waS made to kill more than one elephant at a time. The meat, ivory, ears and feet were remov e d from the carcass where the an imal V.JaS shot and taken back to a base camp. The flesh was then cut into thick strips and hung on r acks in any available shade (Fige 10) ; it was then cut into thin ribbons and left to dry (Figo 11.) Ip Dec ember 1960, this procedure was changed. One lorry I,vas equipped wi th a vlinch and ramps -enabling the whol e elephant, minus t he viscera, to be \ loaded in six pieces 0 The advantage i n this Idas that few-er men wer e required in the field and the anim a l arrived in base c amp i n a shorter t ime 0 However, i n the second year the technique was again modified. By this t ime every effort was made t o kill two or three elepha.;."lt on a hunt. Togethe r ",rith hide and b one1 it was not possible to load 2 elep h ant onto a 5 ton truck. As no good price w a s obtained for hide or bone it was decided to leave these I,vhere -che kill was made and remove o nly the meat, tusks, ear s and fe-et as done originally. R a w meat was carrie d o n the lorry in two metal tanks 4 x 3 x 3f'. At the outset meat was dried without salt. However once it becam e possible to handle 2 or more elep h ant at a t ime salt was added to the meat a s it was put into the t an.1{s. This extended the time it could be k ept irithe field by s everal hours, but resulted in uneven salting. This eventually led to salting in the field being abandoned. Latterly the meat was lightly salted after being cut into the final thin scrips. Al s o in the last six months, all drying w a s done in a thatched shed a s much had been lost through rain when hanging in t h e open. A gang of 35 m e n co u.1d handl e two elephant a day with ease and on s evera l occasions dealt successfully with four at one time. Approximate costs of a hunting op e r ation work out as follows:

PAGE 20

. FIG 1 0 The meat is cut into thick strips and hung on racks (Photo E D .A. Hill) FIG NOJ _ 11 It is then cut into thin ribbons and l eft to dry. (Photo F W \iJoodley)

PAGE 21

10. 35 men for 1 day at an average monthly wage of Shs 80/-per man = 93.45 Shs 100 Cost of armnunition 8."C Shs 5/-per round 4 rounds per elephant, 2 elephant Cost of transport 100 miles at Shs 2/40 per mile Cost of supervising officer 40 240 60 Total Shs 440 or 22 Income from 2 elephan t at an average price of .3 each = .6 Profit therefore = per hQnting trip if two elephant can b e shot, and they are shot within 50 miles of the base camp. 81,738 lbs of biltong were produced in the 3 year period. 70,000 Ibs were sold at an average price of 89 cents per lb. 2,000 lbs were given to neighbouring cultivators as compensation for damage done to crops, and 8,290 lbs were issued a s rations to employees or des troyed by Dermes tes 0 1 1he flesh of a nu;.'11ber of elephant decomposed before it could be dried. I n a c8.refully controlled experiment with 20 elephant an averag e of 500 lbs per beast was. obtained However, the overall average for the period covered is much lower through wastage, decomposition and pilfering. The quality of biltong produced was poor. Though it w a s possible to sell it, sales vrere slovi and in April 1963 it was decided to turn out a higher quality product. Insufficient time elapsed for sale s of this to get under way in the period covered. (Though out of place in this report it is perhaps worth recording that this better quality product vias selling at Shs 4/-per Ib in limited quantities by July 1963.) Ears and feet were skinned and salted. A considerable demand exists for the former, prices being Shs 40/(large) and Shs 30/(small) per ear. Feet did not sell vrell, though in June 1963 a buyer expressed interest in purchasing all at Shs 40/each. There was also a moderate demand for elephant hair bracelets. Carcass use was on the >."hole inefficient. Bon e sinew, viscera, skin and an inevitable small quantity of meat left on the carcass were wasted. However,' without capital to develop

PAGE 22

ll. facilities for handling these products, lic,tle could be done. A total of 274 elephant were shot producing 13,784 lbs of ivory a t a n average of 50.3 lbs p e r beast. Of animals killed, 136 were males and 138 were fem a les. (Four were killed for research or during experiments vTi th im mobilising drugs.) Scheme finances vIere a t a low. ebb after the 19 6 1 and as a hasty measur e to balance accounts, 25 elephant (included in the total of 274) were shot for ivory only in J u n e 1962. Zoos afford a profitable m arket for the sale of a limited number of e l ephant calves each year. However, it wou l d have been unfair competition age.inst professional trapper s had the Scheme, being a Governmen t concern, C0i11pe for the sale of live animals on the open market. A s a result, those animals caught h ad to be sold at greatl y reduced prices : per fem ale and per male, to loca l t rappers. Males are unpopular in zoos a s they have a reputation for being dangerous. A total of 1 8 c alves were c aught b y v ariou s met h o ds Cappo 1.) Of these 5 were sold and the remainder died. Fo u r of those c aught were found to be afflicted by a Babesia ( Brocklesby D W & Campbell H 0, E. A. I vi l d Life Journal, Vol. 1, p.119.) Of these only t w o lived The two tha t survived were cured with a series of phenamidine isethiona.te injections. It is felt that this Babesia may have bee n responsible for the other ) but this i s surmise as most carcasse s were not c arefully examined. Host of t h e c alves were caught after extensive bush fires had destroyed much of their food source. They were thus in poor condition. Thoug h it vras strict policy not to shoot cm-IS wi th unweaned calves, 4 were shot and a t tempts Vlere made to rear the calves. Only one survived. Aeria l observations were made a s frequently as possible to count elephant and observe their movements. It was believed that Scheme and Park beasts were closely linked if no t part of a

PAGE 23

12. homogenous population. This work was therefore carried out i n close co-operation with Park authorities. Apart from numerous loca l reconnais sa..r:tces, three extensive flights were : uade over the combined areas in October 1961, F ebruary and June 1962. I n September 1962 a detailea count revealed 15,603 e lephant (Glover J., E.A. wild Life Journal, Vol 1, p.30o) Of these, 3,282 wer e in the GaIi1e f llanagement Scheme, and the majori ty were in very thick bush and forest and most difficult to count. It was felt the. t the 3,282 represented a maximum of b,vo third s of the elephant in the area, making the probable total nearer 5,000. Since September 1962, ground observation of the extent to which vegetation is browsed in compar ison with the T'savo National Park raise s the possibility of an even larger population. Despi t e the data obtained from aerial flights it is still insufficient for absolute certainty on the size of the elephant population or the ratios between males, females, adults and Juveniles. However from what has been learned there is suffici ent information to put the culling programme on a sounder basis. From ,the'studyof aerial photographs of large herds of elephant there i s a strong suggestion that the r atio between the sexes in adults is equal (Glover J., E .A. Wild Life Journal, Vol. 1, p.37.) This suggestion is supported by Perry' s statement 1I it is obvious that cows do not in fact greatly outnllifiber bulls in the present population . II referring to elephant in Uganda (Perry J S. Phil. 'Trans. 'Royal Soc. Lond., Vol 237, liThe Reproduction of the African Elephant, Loxodonta africana.tI) From these observations it is assumed that the sex r atio a t birth is also approximately equal. (Scheme data from an incomplete series of foetuses exwmined suggest that there may be a preponderance of males at birth, but as yet the sam p l e is too small for considerati on.) Perry (p.lOS) found it is evident that the female East Africa n in the wild state generall y breeds at about 10 years of age, sometimes ear lier II, and II that the capaci ty to produce young i s in some Cases at least, mainLained until

PAGE 24

extreme old age.1I These observations are borne out by field observations in the Scheme. A.M.D. Seth Smith while a Game Warden with the Scheme, observed a female elephant of about 7'6" 13. at the shoulder by itself with a newly born calf. Comparison of this shoulder height with "Fatoum.a'l and "Sampson", both tame elephant which have grown up under natural conditions at the Park Headquarters in VOi, and whose ages are known, suggests that the cow observed by Seth Smith was under 10 years of age. There have been other observations of cows of similar size with calves, but the fact that they were in herds raised some doubt as to whether the calves were actually theirs. Suffice it to say that it appears almost certain that all fem ales will be breeding at 14 years of age. Numerous observations have been made of apparently old animals with small calves at foot bearing out Perry's finding that breeding continues into old age. Perry (p.l08) is also of the opinion that there is an average calving interva l of 4 years, and. that this figure is maximal. Again with the known growth pattern of the tame elephant at VOi, observations in the field tend to support this. The following Figure for a culling programme is drawn up on these premises:-i) The sex ratio for old and young is equal ii) That there are 8 young for 10 t everY A aCiul fem aleS (Glover Jo E.A. Life Journal Vol. 1, p.37.) iii) That a female produces a calf once every 4 years from the age of 14 onwards to 60 years. ( The latter conjecture based on the fact that they consistently breed well into 'old age') at \ri1.tI\' iv) That there is a 50% mortality rateAin the young born each year (Conjecture to introduce conservatism.) v) That the Scheme population is comprised of 5,000 animals. vi) That the calving rate is regular every year; thus as cows calve once every 4 years only one quarter of the adult cows calve and one quarter need to be served in anyone year. vii) That if one quarter of the cows need serving at rate of

PAGE 25

one cow to one bull per year, three quarters of the adult bulls are redundant for breeding purposes. viii)That there will be no mortality through old age as these beasts will be removed through .. J-,,",i.lr'1" '" ix)That as it takes 14 years to reach breeding age, only one 14th of the total young available will be recruited into the adult groups in a year. x)That the cows shot will be those verging on senility and that as the annual age groups are proportional to the rate of recruitment the senile animals will constitute 1/46th of the adult females, there being a 46 year period between the recruitment age (14) and senescence ( 60.) In the mathematical calculations of the' table only whole numbers are used. Where a fraction has been involved it has been ignored; thus 1.8 would appear as 1 and not 2. This has been done to introduce a further measure of conservatism. It will be seen that by culling 120 males annually and only the senile females, the adult males decrease while the overall population increases. By the tenth year though much reduced, the adult males are still more than half of the adult females. However, from this point onwards, females and juveniles h a v e increased to the point where the pressure can be lifted from the males and distributed evenly over both adult groups. when -this is done and the total kill maintained at the same level, the adult male population halts its decline and commences to increase. The other sectors of the community also maintain an increase though at a reduced rate. The population can be held static at any time or depressed by increasing the females killed to equal or exceed the recruitment rate into the adult female class.

PAGE 26

15 FIG. 11 r YEARS! .f\-DlILT IMMATURE I C U L L E D I HALES FEMALES MALES FEMALES MALES FEMALES TOTAL I 1 I 1785 1785 714 714 I 120 38 158 2 1716 1798 775 775 120 39 159 1 651 181)+ 833 833 120 39 159 1590 1834 888 888 120 159 5 1433 1858 941 941 120 160 6 I 1 8 0 1885 991 991 120 40 1 6 0 I 1430 1915 1040 1040 120 41 161 1384 1948 1087 1087 ],20 42 162 9 1341 1983 1133 1133 120 43 163 10 i 1301 2020 1179 1179 120 43 163 i 11 I 1265 2061 1223 1223 I 80 80 160 12 i 1272 20 68 1265 126 4 I 80 80 160 I 13 1 1282 2 0 78 1304 130 I 80 8 0 160 I 14 I 1295 2091 1341 1341 I 80 80 160 I 15 1310 2106 1377 1377 I 80 80 160 J It is suggested that this plan forms the basis of Scheme in the future, pending further more complete data. It is based on a resident Scheme population of 5,000 and takes no account of the periodic forays of e lephant from the Tsavo Park into the Scheme. Due to the destruction of habitat in the Park it has been deemed desirable to kill as many of these as possible when they come out of the Park. However, this source of elephant must te regarded as separate and can only be exploited irregularly. In 196 1 and 196 2 the crocodile on the Galana vJere exploited to a limited de gree. Fifty two were shot and their skins sold for approximately Details of l ength, weight, sex and stomach contents were kept (see app. 2.) Four leopard were shot and their skins sold for These animals appear f airly numerous throughout the area though no attempt \vas made to assess their numbers. GAME COUNTS In May and June 1963 limited road strip counts were carried out in the centre and western part of the Scheme north of the Galanao The results were not conclusive as insufficient mileage

PAGE 27

16. was completed in each area. However, the preliminary results suggest that through the contribution of elephant the area may support as large a live weight of animals as some of the better plains game areas such as southern Hasailand (see appo 3.) LICENSSD HUNTING Though the.Scheme was regarded as a normal controlled area hunting block in which licensed hunters could, and did, hunt as any other game areas, no revenue was received by the Scheme from this source. However, a s the Scheme is an experiment into the overall potential of wild life in the area, the results of sport hunting must be taken into consideration. Forty hunting parties conducted by professiona l hunters' visited the Scheme. Their stays varie d from a week to a month. Taking it that the a v e rage safari costs the approximately per month, and that the average visitor spent 10 days in the area, it may be argued that some ,000 was generated through the sport hunting potential of the Scheme. Forty one elephant w ere killed by licensed hunters and an absolute minimum of ,000 worth of ivory removed from the area by them in this way. Combining this figure with the previous safari costs, some ,000 resulted from sport hunting over the three year period. Needless to say, the fact that the Government refuse s any i ncome from sport hunting to be credited to Schem e funds, raises considerable doubt in t h e participants' minds about the professed aims of the Scheme toward them. ILLSGAL HUNTIN'G Poaching was serious and appeared to be getting worse latterly. The immediate cause for this being the Game Departmentfs inability to cope with the situation. Though money was available from Scheme funds for the formation of an antipoaching unit, the

PAGE 28

17. legal powers necessary for such work were refused to Scheme personnel and very reluctantly the project had to be dropped. work was carrie d out sporadically when Game Scouts could be borrowed from neighbouring ivardens 0 Where roads were cut and hunting took place, poaching diminished or ceased altogether. Scheme development therefore had the effect, not of greatly reducing the numbe r of poachers, but of compressing poaching into the eastern part of the Scheme, which as yet is inaccessible. During .the period covered, ivory representing 233 elephant was handed in at the Scheme Headquarters. Rewaras were paid on the basis of Shs 2/-per pound for good ivory, 1/-p e r pound for defective ivory, -/50 cts per pound for rotten ivory and 6/-per pound for rhino horn. .4.40 was paid out in this way. Though most ivory waS found while road s were being cut and on elephant hunts, a considerable quantity w a s brought in by people not employed by the Scheme. Of the 233 elep hant represented, 82 were mal es, 113 fem a les and 38 unsexed. The latte r figure includes 16 calves. An arbitrary estimate attributes 154 of these elephant to poaching. Between April 1960 and June 1963, 72 poaching cases were brought before the Malindi courts. Of these 21 were brought by the Schem e and the remainder by Game D e p artment, Police and Administration. In the 72 cases, there were 6 acquittals, 3 withdrawn and 63 convictions averaging sentences of 13.14 months. This latte r figure is very satisfactory. A further 11 cases were taken by the Scheme to the magistrate in Voi and successfully concluded. With the exception of on e acquittal, all Scheme cases resulted in convictions. Rewards totalling .6 were paid out for information resulting in conviction of poachers and the recovery of 604 Ibs of ivory repre s enting 18 e l ephant. Details obtained from Malindi court r ecords of cases other than those brought by the Scheme, and from informers whose

PAGE 29

18. information one had every reason to believe, (but could not act on through lack of personnel) indicate a further 103 elephant killed illegally in the period covered by this report. This figure, together with the 154 arrived at through found ivory and the 18 known to h ave been kille d by poachers that the Scheme arrested, reveals a total of 275 elephant poached in the 3 years. A further 79 deaths were attributed to natural mortality. REVENUE M1D EXPENDITURE Yearly income to and expenditure from the Game Ma.nagement Scheme Fund are given below. 1.. APRIL 1960 JUNE 196 1 CREDIT DEBIT Nuffield Foundation Grant Kenya Governm ent Grants Meat Sold Trophies Sold Interest .00.00 7000.00.00 530.00.00 165.00.00 59.15.50 .19.68 All Salaries Vehicle purchases Running expenses Firearm purchases Ammunition Tools, stores etc Buildings Track cutting Rewards Aerial work .04.19 4173.04028 1261006.96 339.19.00 48.02.60 913.10.54 446.10.54 2014.16.04 91.11.50 383.12.64 .18.29 Credit Balance .01.41 2. JULY 1961 -JUNE 1962 CREDIT Nuffield Found ation Grant Kenya Government Grant Credit balance forward Meat Sold Trophies Sold Interest .00.00 6000.00.00 1773.01.41 981.00.00 172.00.00 61.06.82 .08.23 DEBIT All Salaries .09.01 Running expenses 2187.00.49 Ammunition 74.10.00 Tools, stores etc 497.09.98 Rewards 363.05.25 Aeria l work 432.01.81 Track cutting 365.07.97 Purchase of Grader 2158.16000 .00.51 Credit Balance 601.07.72 CREDIT Kenya Government Grant Credit balance forward Meat Sold Trophies Sold Elephant calves Sold JULY 1962 JUNE 1963 .00.00 601.07.72 1605.08.00 842.00.00 780.00.00 .06.92 DEBIT All Salaries Running expenses Ammunition Tools, stores etc Rewards Aerial Work Track cutting Credit Balance 3090 02071 .37 2521.10.54 79005.30 523.06.58 374.16.65 339.13.16 742.17.61 .04.21

PAGE 30

19. Revenue from Scheme ivory sales is credited directly to Government and returned to the Scheme a s "grants" at a later date. However, as it is a basic product its value must be taken into conside ration in assessing resultso During the 3 years, Government received ,132 from ivory sOld. In addition there were a 1,800 Ibs on hand a t the Scheme Headquarters on the 30th June 1963. Valued at Shs 151per lb., this brings the overall value of ivory produced to ,482. Combinin g ivory with other produce total earned income amounted to ,557008.99, some ,557 more than Government have put into the Schem e in grants (19,000). However it would be incorrect to term this real profit as the Scheme has not recovered the full ,000 'investment! of the Nuffield grant. CONCLUSION In the course of the first three years, a Variety of obstacle s were encounteredo Most serious of these were poachers' and charcoal burners' activities. Unless these two pastimes are effectively controlled further time and money will be wasted on wild life management in the area. Future plans must include measures to eradicate these activities with priority over other considerations. Through Scheme and licensed hunting, the c apture of calves, poaching and natural mortality, 6 87 elephant deaths are 'accounted for. foaching is responsible for more than a third of this total and its seriousness is apparento Exploitation has been a t a relatively primitive levelo Efficient techniques for carcass use and disposal are obvious and urgent requirements. However their development will require conside r ab l e capital. 'rhough ivory w a s the mainstay of the economy, the 36.5 tons of dried meat produced for human consumption are equivalent to an offtake of some 2,000 scrub cattle 0

PAGE 31

20. The two districts occupied by the Scheme benefited in a variety of ways 0 Through exploitation of the area's wild life, a hitherto unproductive region has been opened up and a valuable source of employment created. Hitherto men from the hinterland areas had to go long distances to Mombasa or the large coastal estates for work. The monthly average of 230 people dependant on the Scheme received some ,500 in salaries and food. All employees were up to date in tax payments, the majority for the first time in their lives. In the short span of three years it has not been possible to fully demonstrate the economics of wild life management as a land use. However, despite limited capital, the lack of any previous example for gUidance and having to recuperate from severe floods, the potential has been made much more obvious. -When Scheme income is combined with the tourist (sport-hunting) revenue, it is seen ,000 was earned through the area's wild life. It should be borne in mind that capital spent has been entirely on means for exploiting the existing potential. 'Nothing has been spent on developing the basic asse, t of the land and its wild lifeo If progress is to be made beyond the present SUbsistence level such development must be undertaken. To the degree possible within Governm ent, the Scheme has been run as a commercial venture. As is not designed to indulge in business this has resulted in numerous inefficiencies and retarding of progress. Reluctance to compete with private enterprise, which in itself is correct, has restricted development in several spheres. This in turn has its unwanted repercussions on employmel1t. It is also doubtful whether Government, with its massive in other fields, is able to contribute large sums for capital development at this stage. Now the Scheme is established, and three years of Government pioneering have revealed the various possibilities in the area, it seems correct that private enterprise should be invited to take over development and \ Returns from this type of land can only be obtained

PAGE 32

21. through heavy investment. warrant this. I The areas potentia l is sufficient to It is therefore recommend e d providing the Government s responsibility to create employment for the local people can be satisfactorily transferred, serious consideration should" be g iven to future development being carried out through private enterprise. I would like to express my sincere thanks to Dr. P. Glover, Mrs. A. Rickards and Mr. 1'1. Franks of the Veterinary Department for their assistance in the preparation of this report. 14th September, 1963. ISCP/A C P 1. S C Parker Game Warden

PAGE 33

APPENDIX 1 ATTEMPTS TO IlVIl'10BILISE ELEPHi iNT ====================================== ===--In 1962 and 63 a number of attempts were made to immob ilise both adult and immature elephant. Equ ipm ent used was either generously supplied by the Palme r Chemical & Equipment Co. Inc., of Ltlanta U .S.A., ot locally made variations of the Palmer designs. Immobilisation drugs were conveyed to the elephant and injected by standard Palmer I1Cap-Churll syringes. Five different types of weapon were used to project the syringes:-1. Th e 1-1. crossbow was supplied by the Game Department. The particular weapon had several disadvantages, the most outstanding of which was the high trajectory. In a flight of 80 yards the dart rose as much a s 11 feet above the straight line betwee n weapon and point of aim. In thick bush, though quarry might be clearly visible with no obstructions along a straight line from the hunter, twigs and branches above one's sight line obstructed the dart' s flight. in dense vegetation it was difficult to move f ast balancing the dart on crossbow, and w a s reminiscent of an egg and spoon race. Verdict on the crossbowg only of use in open country where the high trajectory is not a hindrance. 2 Pist()l. This weapon had too limited a range with the 10 and 15 cc darts used, to be of use on elephant. 3 The IlHa.r.ih.Q.Q.r.n:' Powder GuYJ,o Dr. ,"0 Harthoorn kindly loaned the Scheme his 'powder' gun. Th i s comprised of a sleev e the sam e calibre as the dart, fitted into a 12 bore shotgun. The breach end of the sleeve was chamb ered to take a cut down 410 cartridge. This weapon had a f a irly flat trajectory up to 50 yards, and proved successful. 4 The 32 Gaugel1 This was a rifle by Beretta made to take standard calibre syringos. This weapon had an exceptionally flat trajectory and could be used up to distances of 100 yards.

PAGE 34

II. 5. The "Palmer 5/8 Gauge". This was a rifled 20 bore which took a correspondingly l arge r dart. with this rifle l ong ranges were possible but the w eight of the darts at high velocitie s was sufficient to inflict a severe wound The "Harthoorn", 32 gauge and 5/8 gauge trajectories and velocitie s could all b e altere d by increasing or decreasing the charges in the propelling c artridges. Tho darts were altogethe r too weak to withstand the velocities set up by all the weapons used oxcept the "Cap-Chur" pistol. Care had to be taken that ranges were in excess of 35 yards. Unde r this r ange the darts would almost certainly break on impact, while above it, there Ivas a chance of them remaining intact. COInmonest dart failure s were snapping needle s As i result they were locally r einforced with stee l ribs which improved performance. The aluminium syringe casing often bent j amming the plunger, o r the threads by which it was scre wed onto the Initially actual injection of the drug was obtained by the reaction of an acid on a Cap Chur' tablet. The gas thus genera t e d drove the plunger down the syringe forcing the drug through the hypodermic needle. Prior to firing, the tablet which was place d i n a recess at the back of the plunger, was sealed off from the acid by a brass plug On the initia l acceleration the plug fell out allowing the acid to contact and react with the tablet. (See Fig.l.) This had disadvantages as the reaction, and thus injection, starte d before reaching the t a rget. This was ove overcome by placing the tablet in anothe r plunger in the tail of the dart with the recess f acing forward. This way the brass plug did not come out until deceleration on impact with the t argets It was also found that the bubble s of gas from the reaction of a c i d on the t ablet formed a coat around the tablet and slowed down the reaction considerably. This was overcome by powdering the tablet beforehand.

PAGE 35

III. Tail Tuft _/ ___ Direction of Fligbt "-, ----_______ ._. ___ J Dru;-T l T ail Tuft It\ J i l l --._ -" P 1 ung e r bl e t Brass Plug Latterly injection w a s acbieved by tbe explosion of a detonator in the recess behind the plunger on impact with the target. Some of the detonators were found to have insuffici ent pow e r to drive the plunger fully home o This resulted in p arti a l injection only. A small measure of black powder added to the detonator overcame this drawback. Injection by this method was almost instantaneous and when several ccs were injected, c reated penetrating wounds. Penetration was eliminated by blocking the end of the needle and making several origices in the side of the needle. This limi ted tbe inj ection ,,!Oulltd to the surface though even this was undesirable, A furthe r advantage in having a soli d pointed needle with holes in the side is that the possibility o f a core of skin plugging the bore is obviated. A number of detonators exploded when the cartridge ignited, and ejected the drug while still in the gun barrel. Th i s was eliminated by placing a cushion of paper between the detonator and screw-in-tail of the dart. A total of 30 elephant were hit by darts and of these only 7 were immobilised. Damage to the darts on impact and the subsequent non-injection account for the majority of failures. Though the principles are sound, the darts are not robust enough for the task they were intended to perform.

PAGE 36

IV. The first elephant darted was "Lali'! a bull of approximately 1,300 1,500 lbs weight. D arted with a mixture of: Phenicyclidine Themelon Scopolamine Hyalase 300 mgms 1.5 gms 100 mgms 1 tablet The d&rt injected i nto the abdominal muscles on the r ight flank. 1 5 minutes after injection, Lali collapsed on h i s brisket. Whil e in this state he was loaded onto a Land Rover and driven back to Scheme Headquarters and placed in a pen. 3 hours 35 minutes after injection, he attempted to raise his forequarters. 3 hours 50 minutes, a simila r but more determined effort. 4 hours 10 minutes stood up trembling violentl y 4 hours 35 minutes .cOllapsed again. 5 hours 10 minutes, again got to his feet. 10 hours later fully recovered. Sold The second elephant was "Bisamb ala" a cow of approximately 1 500 lbs weight. Darted with a mixture of Phenicyclidine Scopolamine Hyalase 400 mgms 100 mgms 1 tablet Injected into the ribs behind the l eft shoulder. 1 5 minutes later collapsed. 1 hour 1 5 minutes, tried to rise. Vehicle not on hand s o given 250 mgms Largactyl. Lay quiet until 2 hours 40 minutes when she again tried to rise. N ight fell and it became imposs"ible to move the elephant until the next morning. During the night a fUrthe r Phenicyclidine Scopolamine Largactyl 320 mgms 100 mgms 250 mgms were given. Put i n pen 1 9 hours after i nitial injection 24 hours -seemed alert, eating although unable to raise itself on its fore legs. Continued feeding well but 3 days later still unable to use its forelimbs so destroyed

PAGE 37

The third elephant was Diwayu1 1 a bull of between 1,500 and 2,000 Ibs weight. Darted in the left shoulder with: Phenicyclidine 400 mgms. 1 6 Minutes -had travelled about half a mile then collapsed. Salivating copiously, 50 mgms Scopolamine injected intra muscularly, and salivation stopped. 2 hours 35 minutes, bracing legs and trying to rise. 5 hours five minutes back in pen, conscious, drinking 5 but unable to use forelegs. Rear legs apparently all right and able to support weight. 24 hours later, alert, eating and drinki ng but unahle to use -forelimbs. Three days late r no improvement so destroyed. The fourth elephant was llHoneyll, a cow of about 2 000 Ibs weight. Darted into right hind Phenicyclidine 200 mgms. 20 minutes later still standing but unable to movements, was approached and roped. As soon as she was roped a 500 mgms Themelon injected intramuscularly. Fell down several time s but immed iately regained her feet. 1 hour 30 minutes after initial injection co-ordination improved considerably, a further 100 mgms Phenicyclidine injected. Loaded into crate still standing. Salivating copiously? given 50 mgms of Scopolamine and 250 mgms of Largactyl. Rested weight against side of crate. 8 hours 30 minutes fully recovered i n pen. Soldo No. 5 was a bull of 8,000 Ibs estimated weight. Darted behind the left shoulder with Phenicyclidine Morphine Hyocine 800 mgms 2 gms 400 mgms 30 minutes approximately, after darting found standing, trembling violently and incapable of any co-ordinated movement 1 hour after injection, it fell on its side. 1 hour 45 minutes, Biven lethidrone to counteract morphine, but no reaction observed.

PAGE 38

VI. 2 hours 30 minutes the beast died. Attempts were made to keep the animal cool. Its rectal temperature taken before death was 106F. On dissection it vias found that the effected by an explosive detonator, h a d scored . the animal I 5 heart. No.6 w ns a very large bull estimated a t more than 10,000 Ibs weight. Darted in the rump with Phenicyclidine Morph ine Hyocine 1 gm 2 gms 400 mgms This animal travelled for 1 0 minutes after being darted before becoming incapable of f urthe r movement I t stood for a further 10 minutes before collapsing. Lethidrone was given intravenousl y without effect. animal d ied approximately 2 hours after the initial injection. Every effort was made to prevent it overheating and it was shielded from the sun. No. 7 was a bull of approximatel y 4000 Ibsweight. It was darted in the rump with Phenicyclidine Morphine Hyocine 10 minutes late r it collapsed. 250 mgms 250 mgms 1 00 mgms L ethidrone (1 dozen phials) was i n jected intravenously with no immediate effect. 4 hours l ater the animal stood up. As it was still very I drunk I it was w alked onto a lorry. 7 hours after init ial i t became very aggressive and threatened to demolish the lorry. A further 100 mgms of Phenicyclidine were inj ec ted and i t lay down s110rtly after. 1 2 hours after initial injection it was again abl e to stand up. It was too big to sell, so when in the follm.ving evening it demolished its pen and w alked away no a ttempt was made to recaptur e it. It was fully recovered, feeding and drinking by then. Summing up;, immobilisation w a s not carried out successfully. Equipment failure was the major reason. Three of the young elephant captured, numbers 1, 4 and 7 were immobilised and recovered s uccessfully. Two lost the use of their forelimbs, and

PAGE 39

VII. though fully recovered in every other sense had to be destroyed. The reason for this disability is obscure. Two large e lephant darted died. Again the reasons are but it appears that large e lephant cannot spend more than a limited amount of t ime on their sides. Four calves were caught through pursuit with a vehicle. A suitable calf is selected and cut out of the herd. Once well away from the other elephant the calf is seized, or if too l a rge, roped. 1his technique is effective but has the disadvantage o f creating considerable disturbance among herds and s e verely stresses the animals caught. Three of the animals chased and caught this way were afflicted by a Babesia (Brocklesby D .W. & Campbell H., E.A. Wild Life Journal Vol. 1, p.119. ) Of these three two died. N Steyn Esq., of Arusha caught all four of those obta ined by the technique outlined in this paragraph. Apart from those darted, caught from a vehicle or when their mothers were shot, several calves were found alone, pursued on foot and caught. All these were in poor condition and died. I.S.C. Parker

PAGE 40

APPENDI X 2 INTRODUCTION. NOTES ON CROCODILE (C.niloticus) I N THE G.cl.LANA RIVER ========= === ============ ====== = = During the Game Management Scheme' s first three years, 52 crocodile were killed and examined. Th e information thus gained forms the basis of this appendix. Where possible data is compared with Cott' s r e s ults (Cott H.B Trans Zoo. Soc Lond Vol 29, Part 4 April 19 6 1 ) DESCl3.IP'l'ION OF 1ffiE;;'. Crocodile occur throughout the Galana's 25 mile stretch e through the Scheme. I n this section the. average width of the rive r is between 180 and 270 metres. Though subject to spectacula r flooding during t h e rains, for the greater part of the year most of the w ater is less than a metre in depth, and v ery f e w pools e xceed 1.5 metres. For most of i t s c ourse, the rive r is slow flowing and its bed c omprised o f shifting white sand, e xcept wher e it crosse s the occa sional rock s eam As a fish habitat it seems poor. This s eems borne out by the stunted growth of Tilapia moss ambica in this p articular stretch. Normally this fish, an algae feeder, attains w eights in excess of l lb. However 670 netted in the Galana averaged 0 .96 oz though many were actually breeding. Oth e r fish occurring in this area, are a Eutropius nr depressirostris, Synodontis mossrunbicus Clarotes laticeps, Clarias mossambicus, Labeo gregorii and severa l small species of Barbus. F e w of these attai n weights in excess of 2 Ibs. The rive r b anks are lined by a thin b elt of large trees. In i this e l atior, HYPQ.a..Qve and Populus are prominent. The l atte r grows nearer the water's edge than the others and also occurs on t h e large r sandbanks with dens e reed beds (Phri2.gmites j2 ) Hany mammals r a n ging in s i ze from warthog to elephant w ater on the rive r and in the dry seasons, so do large nwnbers of birds. ( P articul arly Plocejdae who are associated with

PAGE 41

II. the reed beds and Columbidae and who come in v ast numbers to drink from the sandbanks.) SIZ.2.;, SEX & BRErI:DING CONDITIOll. The 52 specimens obtained were either b aited and shot at night wi t h the aid of a spot-light, or stalked \,vhile on sandbanks in daylight. Crocodile hunting was carried out a s a sideline to othe r S cheme activities and w a s som ewha t sporadic. Kills were a s follows: l.2..6.l 1962 June 2 April 1 July 12 June 18 August 4 July 5 October 10 A l l but 3 specimens were sexe d and all had their weight and length recorded. Th e smallest was a m a l e 0 .44 metres long weighing 0023 kg; the largest, a female, was 2 55 metre s and w e ighed 62. 2 k g Crocodile longer than 2.5 metres are uncommon on the Scheme section of the Galana. Of the 49 s e xed, 25 were m a l e s and 24 f e m a les. This is a ratio of 51.02 to 48. 98 which compares well with Cott' s ratio of 49. 4 to 50 6 for 411 specimens under 2 5 metres in length (p.25l. ) Gonads of the 49 were examined and s even males and s e ven fema les were found to be in breedin g condition. All were in excess of 1.8 metre in l ength. Breeding condition w a s determine d in the male by the presence of enlarged testes and in the female by those with Ova l a rger than 2 mm in diame t er, the presence of oviduct eggs, or by distended oviducts and flaccid discoloured ov aries, indicating recent l aying. A furthe r 3 speci men s in excess of this length were not in breeding 'condition and included the largest specime n obtained. ( Th e othe r 2 w e r e a m a l e of 2.11 m, 27073 kg, and a fem a l e of 1.99 m, 20 0 kg. ) Fig 1 gives details of weight and length for. those in breeding condition. It will b e seen tha t there is litt l e difference in the size a t which t h e sexes start breeding, or in the r e lation of length to weight betwee n sexes.

PAGE 42

FIG. 1 A comp arison ofl e n gth and weight b etween breeding m ales and f ema l e s over 1 8 metres in length. vi E I G H T I I N K I L 0 G R A 1'1 S + = MIlliE o = F E1'111L E 60t I I I i 50i i 40 30 I + I -1 20 I IT o M i n imum M Kg. Males 1 .89 25. 0 Fema les 1 96 26 4 o 2 0 2.1 Maximum M Kg. 2.48 54-. 5 2 .43 52.7 + .h.verag e 1'1. Kg. 2 .1 3 35. 0 2.17 35. 4 /,/' / o / + / / LENGTH I N lvrnTRES ./ ; ./ ./ --.r' + ___ L. 2.5

PAGE 43

FIG. NO: 2 VJ E I G H T I N K The relation of I length to weight of L Scheme crocodile 8 together with Cott's R data on breeding A crocodiles from the 1 S Kafue Flats,Luangwa Valley and Upper Zambesi as depicted in his Fig.19,p.256 /., 0.5 .0. ----0'7---= Scheme data, each dot representing 1 crocodile = Female s as depicted in Cott's Fig.19, p.256 = Males II II "II il II ,. / / / i / I I / I I / / / / / / / I / / / I / I I / / I I / / i I / ._------,--2.0 2.5 3.0 IN METRES /

PAGE 44

1110 Fig. 2 gives the w eight-length curve produced from the measurements of all 52 specimens obtained. In addition Catt' s curves for crocodile in breeding condition from the Luangwa Valley, Kafue Flats 8.nd Uppe r Zambesi are included for comparison. (Take n from his Fig.19, p.256. ) It will b e seen tha t Galana weight length ratios form a remarkably tight p attern up to the maximum lengths obtained. Cott found a m arke d difference in male-female length-weight r atios in breeding crocodiles a s depicted in the figure. FOOD. The stomach contents of 24 spe cim ens were examined. Occurrences of t h e different types of prey are as follows: Insecta 1 3 Araneida 1 Crustacea Mollusca Pisces Amphibia Reptilia 7 o 3 2 0 ii-ves 2 1'1ammalia 4 Carrion 11 Vegetable m atte r 2 ColeQQ..ter a formed the bulk of insect remains found, but H el1LtMJ.:..a and were also identified. All the Cr ustaceans eaten w e r e fresh water spp and Mac.roQ..raQ....QJun spp. ) which abound in the ri ver. No mulluscs at a l l w ere recorded. ,Three type s of fish were eaten and a Goby. Howe ver, on several occasions I have found scales of L a beQ regorii on sandbanl{s wi th crocodile tracks about the m I t may be presumed tha t they do feed on this fis h to some extent. The frogs caught were all of the sam e specie s of Rana. No reptiles w e r e take n as food. Turtle s occur frequently in the Galana and have been observe d feeding off c arrion a t the same time as s everal crocodile of about 2 m etres in length. No attempt was made by the latter to take the turtles, who evinced no fear. ..Q.uelea and doves ( S tr e12 t cm...tli.q, s p 0 ) wer e the only birds represented in the sampl e and Arche r once k illed a specimen in the same area which had t a k e n 7 doves. I have p ersonally witnesse d a crocodile taking sandgrouse ( E remialector decoratus.)

PAGE 45

IV. The only t w o mammal s i t w a s p ossi b1 8 to identify were a small rodent take n by the smalle s t c rocodil e c o l l ecte d ( 0 .44 m ) and a warthog taken by a sp ecimen of 2 4 3 m Two oth e r s pocimens con t aine d m amma l h air. C r o c o dile c a m e readily t o baits of rotten m eat placed i n the river, a n d t ende d t o conce n t r ate n ear t h e Scheme Hea dqu a rters whe n e l ep h ant hunting was g oing o n and offal w a s bein g dumpe d in the r i v er. This m ust have replaced o t h e r no r m a l food s t o a m arke d deg ree. On s e v e r a l occasion s they were known t o leave t h e water and remov e bil tong f rom racks o n w hich i t !'va s dry i ng some 20 yards f r o m the river. W hen dealing with a b a it, their technique w a s t o s eize a portio n in t h eir j aws and rollove r s e v eral t imes until the m Gat b r ok e awa y from t h e c a r cass. I t w a s t h e n swallowed. This m ethod is also des cribe d b y Cott p .303 I t was noticed t h a t whe n a buffalo w a s u s e d as the crocodile w e r e u n ab l e to b reak int o the carcass u ntil the skin h a d decompos e d con s i de r ab l y Cottl s state m ent p .303 I t c r o codll e s a r e w e l l ab l e t o f e e d from a l a rge fresh c a rcasslt mus t apply t o l a r g G r crocodile than occur on the Galana Th e most surprising stomach c ontents f ou n d w e r e 2 wild figs (frui t of Ficus sp. ) i n t h e l a rgest crocodil e shot. One o t h e r spe c i m o n had 2 small p iect:;s of stick whi c h may w e l l h a v e been accidenta l l y i ngested. A l l 24 c rocodile s w h ose stomachs w e r e exa m ined co n t aine d stones. Th e s e sto n e s w e r e w e i g hed and the res ults show n in Table 1. Tll.BLE 10 ..---. . "': L ength g r ou p i N o s tomachs I Ston e w e ight a s a p ercentage o f tota l : I i n metres J i e xamine d : body wei ght . I I o. 3 0.-5 t b u t 0 5 L O i 2 I 0 6 4 0 .42 0 53 i 1.0 1.5 1 2 1.6 4 0 1 6 0 9 2 11.5 2 0 3 1 27 0 .35 0 78 i 2 0 2 5 4 1.16 0 6 4 008 7 2. 5 3 0 0 1 1 2 7

PAGE 46

V One crocodile of 1 99 metres excluded from Table 1, had stones comprising 2 66% of its total w eight. This animal had the latte r h alf of its t ail m issing. stones f ound in stomachs w e r e water worn quartze pebbles or g ravel. G EN ERi:;.L VI OUR Small crocodile l ess than 0 5 metres in length wer e found to be common at night in very shallow water around reed beds. In daylight they conceale d the mselve s i n thick vegetation at the wate r s edge, though on occasi on some w e r e seen basking on rocks away from cover. Several v ery small specimens wer e seen some six feet up Phragmites stems ov erhangi ng water. There seemed little difference i n diurnal habit betwee n crocodile of 0 5 and 2.5 metres, except tha t a s a general rule the smaller the anim a l the shallower the water it was found in. Anim a l s in this size I'ang e would emerge to bask openly on sandbanks betvleen 7 .30 and 8 .30 a .m., returning to the water about 10. 30 to 11 a .m., and again between 3 and 3 .30 p.m. t o 6 p m They were rarel y found out of w ater afte r dark except when fosding on a bait or offa l placed on the rivG r bank. After the 196 1 floods a crocodile of appr oxim ately 2 metres was seen in a water pan some 7 mile s from the river. DISCUSSIO N From Cott' s Fig.35, p.281, it will be seen that approx imately 65% of a 2.5 metre crocodile' s d iet is comprise d of fish, reptil e s birds and m amma ls. For crocodile in excess of this length the p ercentage reliance on these four source s b e come s progressively greater. crocodile' s technique for the captur e o f r eptile s birds and mammal s is "steal th, surprise, and a sudden final burst of speed" (Cott p .302. ) Thus there must be good concea lment until the f i n a l stage. I n the Galana the fish population is poor. 2 5 m etre crocodile woul d therefore h a ve to place g reater relianc e on the capture of r eptiles, birds and m a mmals However the Scheme

PAGE 47

VI. Galana's overall shallowness makes for poor concealment (for protection as w ell as hunting.) Also it would be extremely difficult to drown a mammal of any size a s there are very f e w spots which it could not keep its feet, and drowning SOC3ms the only method whereby a crocodile kills large mammals. It thus appears that the section of the river through the Schem e is an unsuitable habitat for crocodile 2.5 metres and over. This f act must either restrict growth to about this length, or on attaining it, the crocodile emigrate to more suitable environs. In this particular case that latte r seems feasible In the lower reaches of the rive r near the coast and the highest reache s of the Athi tt (the G alana before being joined by the Tsavo) crocodile in excess of 3 metres occurred b efore skin hunters eliminated them. The main tributary, the Tsavo, runs through the Tsavo National Park, is safe from hunters, and still contains l arge crocodile. Both the upper and lower reaches of the Galana and the Tsavo have deep pools in which crocodile can drown large prey. Records of frequent man-eating in the forill e r two and of an attempt to s ecure a buffalo in the Tsavo indicate that large m3J'11li1als are tackled in these waters. Cott shows that for breeding crocodile in Rhodesia there is a difference in the relation of length to weight between mal e s and females which is not apparent with Galana data. McIlhenny as quoted by Cott p .257, shows a similar disparity between breeding males and females of the American alligator (Alligator mississippiensis) but also shows that for the first five years there is no disparity between the sexes. Through lack of evidence Cott assum e s a simila r pattern to be the case with C. niloticus, and this assumption is strongly supported by Galana data where no difference between sexes is evident and specimens wer e much smaller than those from which Cott obtained his evidence. If the Galana data is representative of that part of the growth curve before a difference in mal e s and female s is obviOUS, it may be

PAGE 48

VII. deduced from Cott' s Fig.13, p .246 ( Age-length) that the divergence in growth r ates b etween sexes become s apparent afte r 11 years ( 2 4 m) though Cott' s data on Fig.19, p 25 6 suggests it is somewh a t earlier. Growth in the C n iloticus being slower than A mississipp:i,ensis (Cott Fig.13, p .246 C niloticus takes appro xim ately 22 years to reach 2 6 + m Fig. 20 p 257 ( m a l e s only) A mississippiensi s t a kes approximately 1 0 y ears) it i s to be expected that the divergence would occur l a t er. Th e smallest Galana breoding fem::tl e was 0 21 m smaller tha n the smallGst female on Uganda breeding grounds from a sample o f between 1500 and 2000 crocodile. (Cott quoting Pitma n p 255 ) The smallest male w a s 1 0 89 m etre s Cott belie v e d sexual maturity in males is achieved a t b etween 2 9 a n d 3 3 m ( p 0 253 ) The Galana' fem a l e figure doe s not decrease the minimum recorded fem a l e breeding figure by mor e than a negligible amount. However} the mal e figure does. Th i s may be a result of the l ack of larger males. Inter-male fighting is recorde d (Cott, p 267) and in some instances attribute d to the acquisition of mates. The difference i n weight between 2 and 3 metre crocodil e i s so great, and the latter so much more massive, tha t comba t between the two wou l d be tt pointless for the sma l l e r size. The pre s ence o f large r mal e s may delay the onset of breeding conditions in smaller m a l e s Cott r ecords tha t in populations in which specimens occurred wel l in excess of 3 metres, the sex r atio for specimens under 2 5 metres (young animals) w a s approximately equal thoug h thi s was not the case in specimens ov e r this size. The fact tha t Galana speci mens were under 2 5 m and produced an equa l sex r atio; the fact that there w a s no disparity in weight for crocodile of both sexes of equal leng t h and the f act tha t females in othe r areas can b reed at about the s a m e l ength as those in the Galana, all suggest the crocodile p .opulation i s compos e d of young anim als. Th i s f avours the b elie f tha t l arger specim ens emigra t e If the popul ation do e s contain o l d crocodiles, whic h rem ain sta t i c but are

PAGE 49

'f.. v VIII. restricted in growth through habitat, it would b e logical to expoct sum e evidence of though stunted patterns of growth .i rate dIfferential between s e x e s and some v ariation in sex r atios in the l arger size groups as occur in the adults of Cott' s popul ations. However, emigration can only be conclusively by marking individuals and following their movements ov e r a long period. A worthwhile comparison would be vii th crocodile from a locality such as Baringo in which they do not exce e d 2 5 metres and have no opportunity to move e l s e\vhere on ,attaining this length as in the Gal ana. That the stomachs examined all contained stone s suggests they must have a specific function. The Galana specimens containe d stones more constantly in greater quantity than Cott' s a t a similar size. Though the r e w a s v ariation up to 1 6 7 % and down to the m ajority w e r e much closer to the a v e rage of 0 085% of the animal's tota l w 2ight. Cott's con61usion that the purpose of stone is the acquisition of a n e g ative bouyancy s eems sound. One p articula r specimen not included in the figur e s of v ariation and average already given, was minus h alf its t ail. It had los t this a considerable time b efore collection as the would s how e d w ell hea l e d scar tissue The t ail must constitute on e of the least bouyant parts of a crocodile' s anatomy. This loss must have increased the specimen' s positive bouyancy considerably. It was therefore remarkable tha t this animal contained stone s comprising 2 066% of its tota l w e i ght. One c annot but deduce tha t this w a s a measure to counteract its loss and give s even g reat e r credence to Cott s reasoning The only v ariance in gener a l behaviour to Cott' s findings is the fact tha t crocodj_le betwee n the lengths of 0.5 metres and i 5 metre-s were frequently seen during daylight. Elsewhe r e according to Cott C p .278) this size range is conspicuous by its absence during daylight and cannibalism is give n as the reason for this group' s secretiveness. Reasons for the situation on the

PAGE 50

IX. Galana would again appear to bo the absence of l arge specimens and the difficulty of stalking through shallow water .9 ONCLYSI ONS. o Crocodile in the Scheme section of the G alana s e ldom exceed 2.5 metr e s in l ength as the habitat seems unsuitable for l arge r specimens. Those tha t occur bree d at short e r l engths than recorded by Cotto There are grounds for b elieving that on exceeding 2.5 metr e s crocodile emigrate eithe r up or down stream. Typ e s of pre y takon a r e simila r to Cott's Stomach stones occur consistently and support Cott's deduction that their -.rpose is the acquisition of negative bouyancy.o A minor difference in general h abit to Cott's findings in the 0.5 1.5 m group length is attributed to p eculiaritie s of the environment. loS.C. Parker

PAGE 51

APPENDIX 30 ROAD STRIP COUNTS ================= Expense and the density of vegetation over much of the Scheme is such that it is impracticable to count animals other than and giraffe from the air. Cheaper alternative methods must therefore be used to assess numbers and distribution. In Southern Rhodesian bush country a 'road strip' counting technique has been used with some success. (Dasmann & Mossman, J-ournal of Wild Life Management, Vol. 26 NOG 1, p.10l and No.3, p.262.) This procedure was applied to some areas in the Scheme during April and May 1963 and the results of these initial counts are given below. The strip count is based on the principle that a strip visible from a road is representative of the area traversed. If the area of the visible strip can be assessed and the animals in it accurately counted, the results applied to the whole area would give an estimate of the total animals in it. Dasmann & Mossman's procedure adopted in the Scheme counts was as follows: Observers drove slowly through the area to be counted. Each animal seen was recorded and its distance at right angles to the centre of the road measured in This was done initially by pacing each 4t observation, but latterly by judgment of eye with occasional paced checks to ensure consistency. In the case of herds observed the distance was to the animal furthest from the road. Animals on the road counted as one yard off. Mileage was recordedo The data acquired was subjected to the following formula: a) the number of animals counted was divided into the total miles covered to give the average number of animals per mile. b) the distance of each observation from the centre of the road was totalled and divided by the number of observations to give the average distance of each observation from the road. c) the average distance from the road represents the average width of the strip visible to one side of the road. Multiplied by two this gave the average width of the representative strip traversed by the road.

PAGE 52

II. d) the average width of the representative strip was then divided into 1760 yards to give the average number of strips per mile. e) the number of strips per mile was then multiplied by the average number of animals per mile to give the average number of animals per square mile. f) the compacted formula appears thus:-1760 yards number of animals seen total miles covered x average sight distance x 2 = the average number of animals per square mile. Dasmann and Mossman recommended that at least fifty miles be covered before tabulating data. As a check a further twenty five miles must be covered and the results added to the original figures. If when worked out, there is no significant difference in the answers the count is valid, but if there is significant variation then further data must be collected until results are constant. Difference in size and habit among the various species counted resulted in varying degrees of A giraffe would be plainly visible at 200 yards in bush that concealed rhino, oryx .. or eland at a lesser distance. To include such an observation 4t with those of smaller animals would lead to an erroneously wide representative strip where the latter were concerned. Generally if a Lesser Kudu was visible at a given point, Gerenuk and Grant's Gazelle would also be seen at that point if there. Therefore the data from the Lesser Kudu observation is valid in computing the representative strip for the two other species and vice versa. In the analysis of results animals have been divided into 'visibility' classes. These classes are: I Giraffe II Dik-dik and Warthog III Lesser Kudu, Grant and Gerenuk IV Waterbuck, Oryx, Eland, Rhino, Zebra, Buffalo and Ostrich. The width of the representative strip for a species has been arrived at by compining the distances from the road of all animals within the class, and dividing by a total of the class observations.

PAGE 53

III. There were 8 separate count areas determined by the vegetation and soils occurring (for loc'ation of the count areas see map.) Brief descriptions of the areas, A to H, are a s follows: AREA A A minimum of 27 square miles around the base of Lali Hills. Coarse red soil, stony in places. Vegetation open Commiphora woodland with Terminalia spinosa and Delonix elata abundant. Much bare earth and dense clumps of Sansevieria spp. mainly around the b ase of trees. Grasses and low vegetation too sparse to burn well and area generally unaffected by fires. Mileage covered 44, in 6 separate counts. All the animals listed on the tables have at one time or another been seen in Area A. AREA B A minimum of 172 square miles. Mainly grey tblack cotton' type soils. Very open Commiphora parkland with Dobera glabra abundant and extensive tracts of grass. Severely damaged by fire in 1962 and also burnt extensively in previous years. All the animals listed on the tables occur in Area Be Mileage covered 99, in 9 separate counts. AREA C A minimum of 640 square miles. Very friable red soil. Open Commiphora woodland with frequent Euphorbia robecchii and occasional Delonix elata. Acacias mainly small specie s under 10 ft. Grasses sparse, the most noticeable being Aristida adscensionis. Dense clumps of Sansevieria spp. around the base of most Commiphorae and Delonix. There is much bare earth and fire can only travel after an exceptionally good rainy season. The extensive fires of 1962 did not greatly affect the area. Mileage covered 114, in 4 separate All the animals listed on the tables have been recorded in Area C, with the exception of Waterbuck. AREA D A minimum of 96 square miles. Mainly grey cotton' type soils, in places very light and friable Acacia Commiphora parkland with some dense areas of Sansevieria spp. Grass cover good. Severe fire in 1962 killed I many trees, especially Acacias. Mileage covered 40, in 2

PAGE 54

. _--IV. separate counts. All animals listed on the tables have been recorded in Area D, with the exception of Waterbuck. AREA E= A minimum of 60 square miles. Mainly grey soils. Poor fire-induced grassland. Only trees occasional Euphorbia robecchii. Severely burnt in 1962 and in previous years. Buffalo, Giraffe, Dik-dik, Gerenuk, Lesser Kudu and Waterbuck h a ve not been recorded in Area E. Mileage covered = 16, in 3 separate counts. AREA F = A minimum of 123 square miles. Grey soils with occasional large areas of reddish loam. Very open Acacia Commiphora parkland with abundant Euphorbia robecchii. Very good grass cover. Severely damaged by fire in 1962. All animals listed on the tables have been seen in Area F, with the exception r of. Waterbuck. Mileage covered = 49, in 3 separate counts. AREA G = A minimum of 12 square miles. Red sandy soil. Thick Commiphora woodland with D elonix elata, Acacias and an occasional large Erythrina sp. Not greatly affected by 1962 fire. Grant's Gazelle, ,Ostrich and Waterbuck have not been recorded in Area G. Mileage covered = 18, in 2 counts. AREA H = A minimum of 12 square miles. Grey soil of 'black cotton! type. Very dense bush. Trees Acacia bussei, Diospyros mespiliformis, Mimusops densiflora, Terminalia parvula. Very little open ground or grass. Not touched by fire. Area H, according to records so far obtained, contains no Grant's Gazelle, Ger enuk, Ostrich,or Waterbuck. Mileage covered = 20, in 2 counts. Data obtained is shown in the tables 1 -8. Each table contains specific figures as well as class figures from which the results were calculated. Table 9 gives the square mileage of each area and the estimated total number of anim als. The eight count areas are adjoining and form a block of 1142 square in the centre and west of the Scheme north of the river. None have permanent water though areas A and B are both within reach of the Galana. At the time the counts were made, rain had fallen and pans in every block held water. Unfortunately,

PAGE 55

/ I . -.. . --'" ,Dakadakacha /) ,', (. : Dakadima \ \ \ \ \ \ TSAVO c B c 1 \ Lali j .:".-' / >I.A / I I ; I I L ________ c -, ...-;; "-'-" .. . "'---, ./ .... '" ---.... ........ : \ ..... . !.)VJ'achu H \ i \ \ \ T \ a \ \ \ R. i \ \ \ I \, I i -".,! j 1 \ : i I I / 1 I ': I -i. / I / / / (::::\ Dak'abuko \ ...... '.... *:r' '\. / I \ \ I I I I J I I j / 1 1 -.. I / I / I / """" .. --... _ ..... / ( A diagram to show the approximate positions of the count areas within the Scheme.

PAGE 56

TABLE 1 ABEA A Species Miles j Total seen : S P E C I F I C C LAS S : Answer per sq. m : Corrected t o i I Noo Obs Total Yds i Avo Yds No o Obs j Total Yds ; Avo Yds : at m at 44 m : nearest animal t __ .. ......l. I ___ .. ___ L. Gir aff e 44 11 1 40 i 40 -.L 5 5 6 Dik-dik 44 70 L;5 --;r5 13.67 47 651 --'---13 086 101.0 101 :..Warthog 44 Z 2 56 18 47 : 651 13.86 J. 10.1 1 0 ____ Lo Kudu 44 6 3 42 1 4 It 63 15.75 7.628 Grant 4t 1 6 3 _lie Ii. L ----5'-.4,---+ __ 3 O ryx 4 4 1 1 50 50 5223 44.6 0.64 1 E land Rhino Zebra Buffalo Ostrich i TABLE 2 AREA B Species Miles Total seen: S P E C I F I C ; C L 1). S S .-J Answer p e r sq. m 1 Corrected to : Giraffe Dik-dik \ Warthog L Kudu Grant Gerenuk vJaterbuck Oryx Eland Rhino 99 99 99 99 99 99 No Obs : Total Yds J No. Obs i Total Yds Av. Yds at 53 m at 99 l : i i + L 19064 1 : 12.8 i 1 6 .29 I 16 I i j 1 i 8 4 --t --3It9 87.25 29 1 859 64-.1 : 1.72 7 3 216 72 I 29 1 859 6401 i 1.23 0 97 1 35 22 1294 2'8. 82 29 1 859 64.1 4.92 4 85 -----<5'---__ ! 33 8 4 2 80 7 275 201.8 1 37 5 8 8 1 307 1307 163.4! 1.0 1 1.80 t 163.4 i 0.44", I 2 i Zebra i Buffalo i Ostrich 99 14 1 1 5 0 150 i 8 1307 163.4 0.76 1

PAGE 57

TABLE 3 AREA C = ==== 1 Species i Gi raffe Dik-dik : Warthog I L. Kudu Grant ; Gerenuk ; : Waterbuck ; Oryx i Eland Rhino i Zebra : Buffalo Ostrich I Hiles: Total seen: 114 15 114 130 114 5 114 16 114 11 114 9 114 4 114 8 8 2 2 1 220 136 6 1 270 50 6800 300 1 5 5 5 30113 2.429 10079 20159 : 3 2 1 2 ====================================== = =================================================================================== + = at 79 mileso TABLE 4 -AREA D ; Species Miles i Total seen S P E C I F ICC LAS S Answer per sq. m Corrected to I Noo Obs Total Yds ; Av. Yds, No Obs : Total 1 200 200 16 214 13.44 17 215 i Giraffe 40 9 : i Warthog 40 10 L 0 Kudu 40 5 : Grant Gerenuk Waterbuck; Oryx : Eland Rhino Zebra : Buffalo Ostrich 40 40 2 2 1 4 2 1 1 261 110 65 1 65025 : 55 65 17 3 3 215 175 175 Yds Yds 10.05 10005 at m at 40 m nearest animal' o 1 0.754 0.754 1 1

PAGE 58

TABLE 5 -AREA Ee Species Giraffe ; Miles: Total seen S PEC 1 F 1 C C LAS S Answer per sq. m Corrected to ; : No Obs! Total Yds Avo Yds. No. Obs Total Yds Avo Yds at 16 m at m nearest animal. Dik-dik Warthog L. Kudu Grant 1 Gerenuk Waterbuck ; Oryx : Eland ; Rhino j Zebra : Buffalo : Ostrich 16 1 6 1 6 25 1 8 2 1 1 1 535 50 300 110 50 300 110 3 3 3 4 6 0 460 460 10.28 10 15.33 .3588 15033 2.871 3 15.33 0.787 1 ( TABLE 6 -AREA F ===================================================================================================================== = ==== I ; Species I Miles Total seen i I Giraffe i Dik-dik Warthog L. Kudu Grant Gerenuk ; liJaterbuck ; i Oryx Eland Rhinb : Zebra ; Buffalo Ostrich Lt9 11 Lt9 33 49 1 Lt9 34 49 8 49 4 49 21 49 4No. Obs 5 21 1 17 4 3 2 3 s P E C 1 F 1 C Total Yds : Av. Yd s Noo 351 70.2 127 6.05 1 1 789 Lt6. 41 152 38.0 192 6200 370 185.0 275 91.67 : C LAS S Answer per sq. m Corrected to Obs Total Yds Avo Yds at 49 m at m nearest animal 2.81Lt 22 ----r21f -5.82 10.18 10 22 128 2.82 3.086 3 24 113 6 Lt7.3Lt 12.9 13 24 1136 47034 30035 3 24 11]6 47.34 2 3 5 645 129 2.923 3 5 645 129 0.55 6 8 : 1 .

PAGE 59

TABLE 7 AREA G Species Hiles per sq. m Corrected to Total seen ;.! No. Obs : Total Yds I Avo Yds Noo Obs Total Yds I Avo Yds: at 18 m at m nearest a n imal Giraffe Dik-dik Warthog Lo Kudu Grant Gerenuk Waterbuck Oryx Eland Rhino Zebra Buffalo Ostrich 18 18 18 25 1 6 -1 1 2 2 15 4 --9063 10 '053--11601 ---' :TI) .------. 17 179 11 22 22.0 12 122 10023 4064 L_ 5: i 26 13.0 : __ L22 8 -_-__ -_--_-_-_===============d=================d==== = ================ _ -TABLE 8 AREA H Specie s Giraffe Dik-dik Warthog . __ _________ .:::.C---=:.L_.A S S Answer per sq. m Correcte d to : Mile s Total seen:--20 20 2 No. Obs Total Yds : Avo Yds No. Obs Total Yds Av o Yds at 20 m a t m : near est animal i 12 22 1.83 2 37 384 0 1 -4076 'rr:--' .---1 38'+ ; 5 L. Kudu Grant Gerenuk Waterbuck Oryx Eland Rhino Zebra Buffalo Ostrich;.. i!,

PAGE 60

TABLE 9 I SPECIES A B : 640 C D E F G H 27 sqoffi 172 Sq.ffi Sqoffi 96 sqoffi : 60 Sqoffi 123 Sqoffi : 12 Sqoffi 12 Sqoffi TOTALS i I 162 640 96 369 I Giraffe 1,267 i I f I Dik-dik 2727 2236 55040 4800 1230 1392 4608 7 2 ,033 r I Warthog 270 1920 2112 369 60 4,731 I t 216 172 i Lo Kudu I 2560 288 1599 60 4,895 I Grant 600 369 1,141 i 172 i i I Gerenuk 81 860 1920 369 48 3,278 I i Waterbuck I 344 1280 96 369 2,116 Oryx 27 I I Eland 96 96 I Rhino 640 123 763 I i Zebra 1280 180 1,460 BUffalo Ostrich 172 60 232 ,

PAGE 61

v. there was insufficient time to motor a full 50 miles in any blocks other than Band C. Results were worked out at 53 and 99 miles in B, and 70 or 79 and 114 in C. Variances in results are as follows: Dik-dik Giraffe Lesser Kudu Gerenuk Grant Oryx Other species Area B Per sq. mile 3.4-9 Nil 0.613 0.063 0.259 0.786 Insufficient Area C Per sq, mile 22.16 0.506 1.424 00027 Nil 0.059 data for comparison. In both tables Band C, Gerenuk results show consistency. Dik-dik show such variation that a great deal more data is required. Giraffe, kudu, grant and oryx show v ariance but are probably close to the truth. Waterbuck only occur in parts of A and B and though data from observations of them is used in computing their class representative strips, no attempt has been made to assess their numbers until more information on their limited range is available. Elephant are not included in the road strip counts as more conclusive work has been carried out f rom the air. It would appear that in addition to a minimum mileage there must also b e a minimLLl1l number of observations for a speci e s before worthwhile results can be achieved. In area D the figure of 2,112 warthog is achieved through one observation of a sounder of 10 standing .in the road. This seems erroneously high even though it is "diluted" with 16 Dik-dik observations (being in the same class.) Experience has not suggested that the Wartho g population is any higher in D and in F or A. Similarly the figure of a rhino per square mile in C is taken from 2 rhino observations and is most definitely incorrect. This also applies to Zebra in the same block. That the effectiveness of the technique varies between species is borne out by the fact tha t buffalo were present in all areas at the time of the count (Except E) but were not recorded once. Also rhino, eland and zebra were present though not seen in all blocks 0

PAGE 62

VI. Movement in and out of the area definitely take s place wi t h buffalo, eland a nd zebra and a pp ears related to the a v ailability of w ater. The animals able to go long periods without drinking, Lesser Kudu, Grant, G e r enuk, Dik-dik, Oryx and Giraff e are thought to be fairly static. The l atte r two specie s do tfBvel considerable distances, but seem to be present in the count areas over most of the year. Dik-dik are few in areas severely burnt and seem more immediately affected by this hazard than any other species. After the serious fires in 1962, a number were found burnt to death in Area B. Grant are best represented in areas opened up by fire. Despite the lack of a complete serie s of counts, even at this premature stage it is of interest to make a rough estimate of the live weight of animals sup ported in the better counted blocks B, C and F. In the following assessment, elephant have been included. From aerial counts the annual density of elephant is arbitrarily assessed at one per square mile over the whole area. Rhinoceros and Zebra in Block C are assessed a t on e qu arter of the number shown in Table 3 as the figures given are believed to be gross overestimates. Figures on whicll the calculations are based are estimB.ted minimllill adult weights and are a s follows: Giraffe 1760 lbs Dik-dik 4 II Warthog 40 Lesser Kudu 150 II Grant (petersi) 85 II Gerenuk 70 II Oryx (callotis) 320 II Eland 800 Rhino 2000 II Zebra 500 Elephant 6000 II Ostrich weights are not considered. Average w eights have been deliberately underestimated to be conservative. Buffalo, Eland and Zebra are poorly represented thoug h a t times they frequent the areas in some nwnbers. This will also tend to bias the results tOidards an underestimate.

PAGE 63

VII. Results are given in Table 10 and equated to cattle carrying capacity based on 20 acres per beast and 500 Ibs per cow. TABLE 10 Miles lbs Equiv. No of Carrying c apacity acres per BlocklSq. i cattle p. s.m acres p e r cow g_?ille -l per sq .ml B I 2406 I 7,298 14058 43.89 I 172 C 640 10,424 20.85 30.69 I 6.35 I F I 123 16,815 33.63 19.03 I 16.0 I I Bearing in mind that these results have been reached through one incomplete count series, it is nevertheless interesting to compare them Stewart & Zaphiro's average biomasses for the Athi-Kapiti, Hara-Loita, Leroghi, Baragoi, Wamba and Isiolo areas (Stewart & Z qphiro D.R.P., "Biomass & Density of Wild in Different East African Habitats" to be published. ) are presented in Table 11. AREA! 1 LBS I I per sq.m; TABLE 11 G ALAl"JA SCHEME H STE\tJART & ZAPHIRO I B : e l F I BARAGOI IWAMBA 1810LO 7289 104-84-! 16815 ii 11 080 i 16822 i 681 0 5101 1364-0 I There is thus a suggestion that through the contribution of elephant, the biomass of some parts of the Scheme is nearly as 4igh as that of the Mara-Loita a r ea. g 0N.Q1lLJOli : The roa d count technique shows promising results with some species occurring in the Scheme, but seems ineffectual with others, notably buffalo. It appears that a minimum of 50 miles motoring in the variety of areas sampled is too low for good I results. Also a statistically acceptable minimum number of observations for a sp ecie s must be established. Insufficient 1atR. was obtained so no firm conclusions can be reached. However, there is a hint that the coast hinterland bush country may carry as heavy c.i. '.:Jiomass 8 C;: thl? AC8cia Themeda areas in southern and south western Kenya, though using different species to achieve this. I.S.C. Parker

PAGE 64

APPENDIX 4. A LIST OF GRASSES OCCURRING WITHIN THE GALANA G i-U1.8 MANAGEl'1ENT SCHEl'1E. ================================== Miss D. Napper of the E.A. Herbarium very kindly identified the grasses collected in the Game Management Scheme, and her work is greatly appreciated. 1. Enneapogon elegans 2. cenchroides 3. Schmidtia bulbosa ERAGROST EAE 4. Eragrostis aethiopica 5. aspera 6. caespitosa 7. cilianensi s 8. ciliaris 9. horizontalis 10. perbella 11. rigidior 12. superba 13. Pogonarthria squarrosa 14. Cypholepis yemenica 15. Leptochloa chinensis 16. II obtusiflora 17. Dinebra retroflexa 18. Eleusine africana 19. Dactyloctenium aegyptium 20. giganteum 21. scindicum 22. Drake-Brockmania somalensis CHLORIDEAE 23. Enteropogon macrostachyus 24. s omalensis 25. Tetrapogon bidentatus '26. cenchriformis 27. tenellus 28. Eustachys paspaloides 29. Chloris barbata 30. mossambicensis 31. roxburghiana 32. virgata 33. Schoenefeldia transiens 34. Cynodon dactylon 35. plectostachyus 36. 37. 38. 39. 40. 41. SPOROBOLEAE Sporobolus " tI II AVENEAE; bogdanii he1volus marginatus pyramidalis robustus spicatus 42. Lintonia nutans STIPEAE 43. Aristida adscensionis 44. mutabilis 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. ZOIZIEAE Tragus biflorus tI berteronianus Latipes senegalensis Dignathia gracilis hirtella novo? Perotis hildebrantii PANICEAE Panicum tI " coloratum chusqueoides ? deustum infestum maximum meyerianum subalbidum Sacciolepis curvata Setaria holstii verticillata nov.? Echinochloa colonum tI sp. Urochloa trichopus Paspalidium geminatum Brachiaria deflexa II leersioides II nigropedata II ramosa It serrifolia II eruciformis pubifolia II glauca ? II cencaerantha Eriochloa nubica 11 partispicula ta Digitaria aridicola II macroblephara pennata 11 rivae setivalva velutina Rhynchelytrum repens Tricholaena eichingeri Cenchrus ciliaris II prieurii ANDROPOGONEAE 89. Sorghum versicolor 90. II verticilliflorum 91. Chrysopogon aucheri 92. Bothriochloa radicans 93. Heteropogon contortus 94. Themeda triandra 95. Ischaemum afrum

PAGE 65

'I e