Front Cover
 Title Page
 Table of Contents
 Society notes
 Some aspects of the history of...
 Abamalaki in Buganda, 1914-191...
 Cotton and the Uganda economy,...
 The cairns of Koki, Buganda
 The life of Yakobo Adoko of Lango...
 The fishes of Uganda - IV
 Index to Volume 21 (1957)
 Back Cover

The Uganda journal
External Link ( Journal issues, 1999-2002 )
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00080855/00042
 Material Information
Title: The Uganda journal
Abbreviated Title: Uganda j.
Physical Description: v. : ill. ; 22-24 cm.
Language: English
Creator: Uganda Society
Publisher: Uganda Society etc.
Uganda Society etc.
Place of Publication: Kampala etc
Frequency: irregular[1976-<1995>]
semiannual[ former 1934-]
annual[ former <1948>-1973]
completely irregular
Subjects / Keywords: Periodicals -- Uganda   ( lcsh )
Genre: serial   ( sobekcm )
periodical   ( marcgt )
Spatial Coverage: Uganda
Dates or Sequential Designation: v. 1- 1934-
Numbering Peculiarities: Suspended May 1942-Mar. 1946; replaced by the Society's Bulletin.
Numbering Peculiarities: One issue published every few years, v. 38 (1976)-<v. 42 (1995)>
Issuing Body: Vols. 11-13 published in London.
General Note: Journal of The Uganda Society (formerly The Uganda Literary and Scientific Society).
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - ABS8002
oclc - 01644239
alephbibnum - 000301510
issn - 0041-574X
lccn - 52026895
System ID: UF00080855:00042
 Related Items
Related Items: Uganda journal

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
    Society notes
        Page v
        Page vi
    Some aspects of the history of Western Uganda
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
    Abamalaki in Buganda, 1914-1919
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
    Cotton and the Uganda economy, 1903-1909
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
    The cairns of Koki, Buganda
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
    The life of Yakobo Adoko of Lango district
        Page 184
        Page 184a
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
    The fishes of Uganda - IV
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 220a
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
    Index to Volume 21 (1957)
        Page 245
        Page 246
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text
MNF -4
h" U ur,7.
dr ii ;o a'Int



pcA&pmts of- the Hik6ry of W tem Uganda K, fNGwm 13
_Abxtbafakii4n, BugaLda_ 1914-1919 W
ttonalod.tfte UgandaFconom 1903,1009,
Buganda a
ajrn of Kok 176
Y'kobo Ad Ak*,
oko of La6g'ODiswO1 I 'G. Himt)m,
The'Fis4es-of Uganda-IV P.Ji. GkrEKWOOJ

Roddy O'llivan's Grave L. SwNNW_ 22D'`
A- T, MATsoN 221,
YPro sals fi Federal Ug4nd, K., INQUW
Auptida"Tw(wty-five YearsAgo W.' J., BAwbk ZV,
PbpUlatibn in Kasilang Erony, Tcs J. Mil mm R N. WiLsON, 229
lh *Ilsc).ns of Early Uganda RL'B. Tuonks 24
Atur'LSOCjety (by A W. Southall) P. P. 'HOV'i 234
'Ti A filican Yaeidnary Handbook (by P. Z. Mackinzicand R. K
D, ff, Rotsi
Vi N,
Ofilpm A 1. Richai 7 R. McADAm 241`-!I,"
,,rownsineniwtbo Making (by A. W-. Southall and Gutkind)
M. N. MANi 24Z,
RecoNcd, 243A
jndex to -volimic 21 of'The. Tganda.,16urnal 2

-publi'l by
Priee, Shs-. J5 (15s.)


!i 0

Hi xeln StFeeikCafrKCMGOBE _
Mr .BrigtnWr r H .Mri


Uganda Journal



No. 2


(assisted in England by H. B. THOMAS, O.B.E.)

Published by



Some Aspects of the History of Western Uganda
Abamalaki in Buganda, 1914-1919 REV.
Cotton and the Uganda Economy, 1903-1909 -
The Cairns of Koki, Buganda -
The Life of Yakobo Adoko of Lango District -
The Fishes of Uganda-IV P

Roddy Owen's Grave -
The Game of Mweso -
Eugen Wolf, 1850-1912
Early Proposals for a Federal Uganda
Ruanda-Urundi -

Uganda Twenty-five Years Ago -
Population in Kasilang Erony, Teso J. McFE
The Wilsons of Early Uganda




Alur Society (by A. W. Southall) P. P. HOWELL
The African Veterinary Handbook (by P. Z. Mackenzie and R. M.
Simpson) D. H. L. ROLLINSON
Annales d'Ethiopie P. L. SHINNIE
The Geology of Part of South-East Uganda (by K. A. Davies)
Chisungu (by A. I. Richards) R. MCADAM
Townsmen in the Making (by A. W. Southall and P. C. W. Gutkind)
Other Books Received -
Index to Volume 21 of The Uganda Journal -








At the Uganda Society's Annual General Meeting on 15 May 1957, it was
decided to increase all subscriptions, with the exception of that for associated
membership, by Shs. 10 a year from the 1 January 1958. The single subscription
will thus become Shs. 30 and the double subscription Shs. 40 a year. As a
corollary the price of The Uganda Journal, which is sent free twice yearly to
all members other than associated members, has been raised from Shs. 10 to
Shs. 15 a copy. This action was necessitated by the increasing costs with which
the Society is faced. Part of this additional expenditure is due to the rise in
printing costs of the Journal and part to the increased activities of the Society.
For the last two years, the Society's balance sheet has shown a substantial
deficit and it is clearly impossible to continue to use up reserves in this way.
The present subscription has remained unaltered since 1949.
During the last year an unusually full programme of lectures was arranged
on a wide variety of subjects, as the following list of titles will show:
Heredity and human nature. Professor David Rife, B.Sc.(Agric.), M.A., Ph.D.
German Literature. Mrs. M. Welter.
*Negro Universities in the State. Miss Jean Fairfax.
The problem of disease in the tropics and the remedy. Professor T. H. Davey,
O.B.E., M.D., D.T.M.
Relation of the Soul with the Supreme. Dr. Radhakrishnan, Vice-President
of India. (In co-operation with Makerere College.)
Indonesia, some economic problems. Professor Newmark, M.A.
*Negro Education in the United States. Professor R. L. Lucas, M.A.
British expansion in tropical Africa. Dr. R. E. Robinson.
Some problems of Bible translation. The Rev. W. J. Bradnock.
*Social problems in Sierra Leone and the Gold Coast. G. K. T. Roberts and
J. Ridy-Williams.
Taking things for granted. His Worship the Mayor of Kampala, Councillor
Clifford Lewis.
World Guiding. Lady Baden-Powell, G.B.E.
Forestry for the future of Uganda. Mr. G. Webster.
Uganda at the Industrial Conference in London. Mr. D. S. Kigozi.
Prospects of archaeological research in Uganda. Mr. P. L. Shinnie; and The
archaeology of musical instruments, with special reference to Africa. Dr.
K. P. Wachsmann.
The Y.M.C.A., as a world movement. The Hon. Isobel Catto.
Some social and economic problems of Ruanda-Urundi. His Excellency
Monsieur Jean-Paul Harroy, Vice-Governor-General of the Belgian Congo,
Governor of Ruanda-Urundi.
*Stories behind Buganda names. Mr. M. B. Nsimbi.
*The history of Western Uganda. Prof. The Hon. K. Ingham, Dept. of History,
Makerere College.
*Geographical background of Western Uganda. Prof. S. J. K. Baker, Dept. of
Geography, Makerere College.

*The peoples of Toro. The Hon. J. K. Babiiha.
*The Ankole traditions of Kingship. Mr. H. F. Morris.
*The peoples of Bunyoro. Miss S. Nyendwoha.
*The peoples of Ankole. The Hon. Z. C. K. Mungonya, C.B.E.
During the year, negotiations were carried out with the Interim Trustees
of the Uganda National Cultural Centre as a result of which the Uganda Society
became a founder constituent member of the Cultural Centre. In return for a
contribution representing a quarter of the capital cost involved, the Society
will be offered accommodation on nominal terms in the new Cultural Centre
building. The Society will have the exclusive use of this accommodation which
is adequate to house its library and to provide necessary office space and a
meeting room for lectures. The Society will also be represented on the Advisory
Council of the Cultural Centre.

In co-operation with the Extra-Mural Department, Makerere College.

K. Ingham. Senior Lecturer and Head of the History Department at Makerere
College; Member of the Legislative Council of Uganda; author of Europe and
Africa and Reformers in India.
The Rev. F. B. Welbourn came to Uganda in 1946 as Chaplain and Lecturer in
Physics at Makerere College. In his spare time he is working on the separatist churches
in Uganda and certain related movements in Kikuyu and North Nyanza.
P. H. Greenwood. Senior Scientific Officer of the East African Fisheries Research
Organization, has, since 1951, studied the fishes of Lake Victoria and Uganda.
Before joining the E.A.F.R.O. he held a Colonial Office Fisheries Research Student-
ship, and studied at the British Museum (Nat. Hist.) for a year.
C. Ehrlich. Lecturer in Economics at Makerere College since 1952; author of The
Uganda Company Limited The First Fifty Years.
E. C. Lanning joined the Uganda Administration in 1948. While stationed at
Masaka and Mubende, he was able to develop a long-standing interest in archaeology
by investigating and recording many of the sites in which these districts are so rich.
J. G. Huddle joined the Uganda Administration in 1950. After a two years' second-
ment to Malaya he served as a district officer in Lango District from 1954 to 1956.



THE absence of written records for any but the most recent years and the
lack of any significant archaeological remains to throw light upon the
history of western Uganda has tended to discourage academic historians from
trying to penetrate into a field in which the accepted methods of research are
inapplicable. At the same time, the limited but intriguing legends of the
western kingdoms have been in danger of becoming inflated by tribal pride or,
even more, by Europeans with an enthusiasm for the unusual. In fact, if the
earlier history of western Uganda were better known it might well provide a
story which in outline is not seriously dissimilar from the history of England
between the departure of the Romans and the Norman Conquest. Just as Angles
and Saxons swept across England and set up their kingdoms to be followed
some time later by the greater power of the Danes, so too Hamitic invaders
poured into western Uganda setting up their kingdoms to be followed in their
turn by the more widespread authority of the kingdom of Kitara. And although
the rule established by the Hamitic and Nilo-Hamitic dynasties was by modern
standards primitive, the customs of the western kingdoms were complex and
society lived according to a rigid set of rules.
The place of origin of the Hamitic invaders is uncertain and the route they
followed to Uganda still unknown. While the generally accepted view is that
they came from the north, possibly from Ethiopia, there has been considerable
sympathy for the suggestion that they ultimately derived from Egypt and among
the inhabitants of Mpororo and Ruanda there is a strong tradition to the effect
that the invaders entered Uganda from the west although originating far to the
north.2 Even the date of the migration cannot be fixed with any degree of
certainty. All that is known is that the rolling grass-lands between Lake Albert
and Lake Victoria attracted the pastoral Hamites who established there a
number of kingdoms. Among those kingdoms were Ankole, Mpororo, Buhwezu
and Igara. The invaders set themselves up as a ruling race who mixed but little
with the indigenous Bantu population. They had little taste for the duties of
government, desiring only to find suitable pastures for their cattle. From the
anthropologist, Roscoe, we get a picture of a mildly autocratic society in which
law, as in the heroic age of ancient Greece, consisted of a series of individual
judgements given by the senior chief. He would be sitting in the shade of a
euphorbia candelabra surrounded by his pages and private guard and
attended by his subordinate chiefs. Or perhaps the judicial power was delegated
to lesser chiefs, but in either case the problems posed were limited in scope and
1 Two lectures arranged by the Uganda Society and the Extra-Mural Department,
Makerere College.
2 F. Lukyn Williams: Nuwa Mbaguta, Nganzi of Ankole. Uganda J., 10 (1946), 124-5.

were mainly concerned with cattle or women.3 The Bantu agriculturalists acted
as serfs, but were usually at liberty to cultivate any land they wished provided
they fulfilled the duties required of them by their overlords.
These kingdoms were frequently at war with one another, although in peace-
time trade was carried on between them with a considerable degree of freedom.
There were cases of intermarriage, one of which resulted in the subordination
of Mpororo to Ankole. Ankole gradually emerged as the strongest state,
although the others paid tribute only if the ruler of Ankole was strong enough
to collect it.
Possibly about four hundred years ago there appeared in the area which
includes Bugangadzi, Bwera, Mawogola and Kabula a group of people whose
standard of civilization was such that it left a lasting impression with succeed-
ing generations of Hamites and whose existence is remembered in the verbal
history of the later kingdom of Kitara. These people, remembered with awe as
the Bachwezi, may have been responsible for the earthworks on the Katonga
River at Bigo. Imposing legends have grown up around them but, as PKre Gorju
argues at some length, there is no reason to suppose that these men were
different in origin from the other Hamites. It is quite possible that they were
the most recent wave of Hamitic invaders and were, therefore, able to bring
with them the fruits of contact with Europeans further north in Ethiopia. They
were certainly pastoralists.4 Why the Bachwezi disappeared we do not know,
although once again Gorju introduces a realistic note when he suggests that
there is no reason why internecine strife, sickness or cattle disease should not
have brought about their downfall and disintegration as similar mundane
causes have destroyed far greater empires.5 From the traditions of Kitara, how-
ever, their passing seems to have coincided with the rise to power of the Babito
dynasty which established the empire of Kitara, later to be known as Bunyoro,
with its centre in Bugangadzi.
That the Babito dynasty was of Nilo-Hamitic origin seems fairly clear.
Probably the Nilotic migration which populated Acholiland in the sixteenth or
seventeenth century caught up with and became partly absorbed into the closing
stages of the slow southward advance of the Hamites producing a mixture of
the two races which came to occupy both ethnographically and geographically
a central position between the Hamites of south-western Uganda and the Nilotes
of the Northern Province.6 It may well have been due to the infusion of indus-
trious Nilotic blood that the rulers of Kitara came to look beyond the distant
horizons of the cattle prairies and extended their influence to the east of the
Nile and south-westward along the shores of Lake Victoria instead of remaining
simply lords of a petty kingdom of herdsmen like the Hamitic chiefs to the
south. It may, too, have been the same enterprising spirit that added an econo-
mic bulwark to their empire by developing the salt trade of Lake Katwe and
the iron deposits further north.
The very wideness of Kitara's boundaries, however, while sharpening the

3 J. Roscoe: The Banyankole. 1923. pp. 13-4.
4 J. Gorju: Entre le Victoria l'Albert et IEdouard. Rennes. 1920. pp. 50-3.
5 Ibid. pp. 54-5.
6 A. C. A. Wright: Lwoo Migrations. A Review. Uganda J., 16 (1952), 82-8.

military instincts of the Bakitara, militated against the development of more
systematic government. The empire, moreover, depended for its existence upon
the loyalty of the colonists who established themselves in Busoga and Bugerere,
in Koki and Kiziba, and upon the power of the ruling Mukama's armies rather
than upon any superior civilization or upon a rule of law. Energies which might
have been more usefully employed in evolving a sound administrative system
were dissipated in the effort to maintain the Mukama's overlordship. More
fundamentally, the Babito remained at heart a cattle people and as such had
little incentive to investigate deeply the needs of a static society. The Abakama
might ravage Ankole and carry off cattle and even range as far afield as Ruanda.
But, like all pastoralists, their joy lay in raiding rather than in ruling and when
the constant struggles for the succession had weakened the country internally
they fell victim to the more compact power of Buganda to the east. Early in the
nineteenth century, also, the eldest son of Mukama Kyebambe III, despairing
of ever succeeding his long-lived father, carved out for himself a domain in the
south-western portion of Kitara and so founded the kingdom of Toro which
remained independent until it was reconquered by a more warlike Mukama in
the latter half of the century.
It was in the late eighteenth and early nineteenth centuries that several of the
outlying districts of Kitara fell into the hands of Buganda by right of conquest.
In addition a number of tributary states including even Koki, known as the
little Kitara in the south-east, transferred their allegiance to the rising star of
Buganda. But Kitara was not a completely spent force. The future might have
witnessed a renewal of its former influence if it had not been for the intrusion
of new, non-African influences. For, after the usual struggles for the throne,
there succeeded in 1869 a warrior whose prowess was to rival that of the greatest
among his ancestors. Of Kabarega as a ruler little is known. Of his aggressive-
ness in war there is no doubt. In a very short time he had reconquered Toro
and a number of the districts which under more peaceful rulers had been sur-
rendered to Buganda. The young heir to the throne of Toro fled to take refuge
with the Mugabe of Ankole and then, since he was still not safe from Kabarega's
wrath, moved on to seek sanctuary in Buganda. Meanwhile, Kabarega entered
the very borders of Buganda itself as champion of a renascent African empire
at a critical moment in Buganda's history. It was the moment when Buganda
was beginning to get in touch with the outer world.
The western kingdoms missed the first impact of the Arab traders who came
to Buganda from the south in the middle of the nineteenth century and who
travelled close to the shore of Lake Victoria. The European explorers, Speke
and Grant, who followed in 1862, also by-passed the Hamitic kingdoms but
made their way through the northern marches of Bunyoro where they encoun-
tered Kamurasi, Kabarega's predecessor.7 Bunyoro, as Kitara now seems to
have been called, had more immediately serious troubles than could be aroused
by the enquiries of two wandering Europeans. In the late 1860s the kingdom was
rapidly falling prey to the ivory raiders and slave traders of mixed race who
had pushed through the swamps of the southern Sudan to find new hunting
grounds further south. Indeed, Kabarega himself had seized the throne in the
7 J. H. Speke: Journal of the Discovery of the Source of the Nile. 1863. p. 526.

face of opposition from his more popular brother mainly through the interven-
tion of the slavers. But the latter were dangerous friends and Bunyoro was
severely harassed by them for twenty years. Fortunately Bunyoro was the
furthest limit to which they penetrated so that their activities did not prove so
completely devastating as they were in Acholi.
A still greater threat to Bunyoro's independence arose from Egypt's attempt
to extend her empire southward to the very heart of the Lake Region in the
1870s. Her professed object enabled Egypt to enlist the aid of both Sir Samuel
Baker and Colonel Charles Gordon in carrying the flag up the Nile. Baker
even attempted to proclaim Bunyoro a part of the Egyptian empire in 1872 but
was forced to withdraw as a result of Kabarega's hostility. Gordon, who suc-
ceeded him, did not give up hope of establishing Egyptian influence through
tropical Africa to the east coast until the opposition of the African rulers of
the Lake Region and the protests of the British agent in Zanzibar had clearly
shown him the inadvisability of his project. The final blow to Egypt's imperial
ambitions was, however, struck by the Mahdist rising in the 1880s which severed
the Egyptian Equatorial Province from the Mother Country. Simultaneously
the rising cut off the slave-trade at its roots, and Bunyoro and the other king-
doms of the west were safe from further threats from the north.
The western kingdoms were given little respite. Within a decade they were
faced with a new problem arising out of European intervention in tropical
Africa. Already in the 1870s and again in 1889 the explorer, Henry Stanley,
had traversed western Uganda. But he had made no attempt to stay there. Even
the redoubtable Emin Pasha having transferred to the service of Germany after
quitting his Egyptian charge in Equatoria, was forestalled by the Anglo-German
Agreement of 1890 in his design of making western Uganda a German depen-
dency and, like Stanley, had simply passed through that area on his way to the
west. But events in Buganda and further afield round the conference tables of
Europe drew the western kingdoms inevitably into the sphere of British influ-
ence. It is true that Britain was not yet anxious to become involved in administra-
tive obligations in so remote and unpromising an area. But strategic reasons,
the desire to control the head waters of the Nile and thereby to dominate the
Egyptian route to the Far Eastern trading grounds, made her wish to prevent
any other power from acquiring influence north of Lake Victoria. By the Anglo-
German Agreement of 1890 this object was achieved on paper. It then remained
to see that the paper agreement was made effective.
The Imperial British East Africa Company which undertook in 1890 to
administer the country north of Lake Victoria on behalf of the British Govern-
ment began its task cautiously. Its resources were limited and it could not
afford to assume heavy commitments. If it was to pay its way, however, trade
was essential and the most promising source of income appeared to be the
ivory trade from western Uganda and the Congo Free State. It was the view of
the Company's Directors, therefore, that the first step to be taken after securing
a firm base in Buganda, about which Christian missionaries had already con-
veyed considerable information to Britain, was to open up a connection between
Lake Victoria and Lake Albert. This, it was hoped, would produce a number
of important results. First, if a post could be built in Ankole the trade of

western Uganda might be diverted from its southern outlet through German
East Africa' into British channels. Second, the post would prevent the illegal
importation of arms and gunpowder from German East Africa into Buganda
and possibly into Bunyoro. Third, it would open access to the salt deposits of
Lake Katwe which would thus provide an alternative source of income for
the Company. In addition to these considerations, Captain F. D. Lugard, the
Company's representative in Uganda, was extremely conscious of his weak
position in a Buganda torn by politico-religious rivalries and was therefore
anxious to enlist the support of a body of Sudanese troops reported to have
been left to the west of Lake Albert by Emin Pasha when he withdrew from
Equatoria. With these various objectives before him Lugard started westward
from Buganda in 1891 taking with him Kasagama, heir to the throne of Toro,
who had appealed for the Company's aid in re-establishing himself in his
The route which Lugard chose lay through Ankole where he hoped to make
a treaty with its ruler, the Mugabe Ntare. He was anxious, too, to avoid as
much as possible the territory of the warlike Kabarega. Ntare, however, was
reluctant to meet Lugard, but he sent a representative to make blood brother-
hood with the Englishman. In return Lugard made a treaty with Ntare's envoys
placing Ankole under the Company's protection. This, however, was a tenuous
arrangement since Lugard was at that time hard pressed to maintain the
Company's position even in Buganda. His treaty, however, gave Britons the
right to pass through Ankole or to build and settle there, while Ntare was to do
whatever he could to prevent arms from passing through his territory to Bun-
yoro or to the Baganda Muhammadans who had threatened the stability of
Buganda.8 In Lugard's estimation this arrangement implied the annexation of
Ankole to the Company's territory. It is unlikely that Ntare appreciated this,
more particularly since Lugard had to abandon his idea of building a station
in Ankole owing to the lack of food to maintain a garrison.
As Lugard continued his advance towards Katwe Kabarega's Banyoro made
a token show of resistance but from Katwe onwards their opposition became
stronger if equally unavailing. On entering the former kingdom of Toro Lugard
therefore built Fort Edward and established Kasagama in the vicinity with a
garrison to protect him. The Batoro came forward to welcome their Mukama
although they were extremely fearful of the consequences of Lugard's ultimate
withdrawal. Lugard, however, made a treaty with the young ruler by the terms
of which Kasagama agreed to obey the Company's Resident in Buganda in all
matters and to prevent the importation of arms and gunpowder in return for
the Company's protection.9
From Fort Edward Lugard continued his northward march until he made
contact with the remnant of Emin's troops at Kavalli's, south-west of Lake
Albert. Then he retraced his steps, setting up a line of stockades, garrisoned by
detachments of Sudanese troops, along the eastern border of Toro to protect
the frontier against Kabarega's forces. Thus, as a result of one determined
march, Kabarega's forces had been driven back, the kingdom of Toro had been

8 F. D. Lugard: The Rise of our East African Empire. 1893. ii, 160.

9 Ibid. p. 188.

re-established and both Toro and Ankole had entered into treaty relationship
with the Imperial British East Africa Company.
Almost immediately that relationship appeared to be in danger of crumbling
away. On his return to Kampala, the Company's headquarters in Buganda, in
December 1891, Lugard was greeted with an order from his Directors to with-
draw from Uganda since the Company could no longer afford to maintain an
administration there. Lugard was reluctant to surrender the results of his work
and, suspecting that the C.M.S. missionaries would feel that their association
with the Company had jeopardized their position in an independent Buganda,
planned that they should found a British settlement in Toro. By this means the
missionaries would be able to retain a foothold in the Lake Region and at the
same time Britain's obligations to Kasagama and his people would be honoured.
Salt from Lake Katwe and ivory would pay the cost of the ammunition required
to defend the new state and contact would be maintained with the coast through
Ankole and round the southern end of Lake Victoria.'0
Due to the financial assistance given to the Company by supporters of the
Church Missionary Society and subsequently to the decision of the British
Government in 1894 to take over the Company's commitments in Uganda it
was never necessary to put Lugard's plan into effect. But the position in Toro
and Ankole remained uncertain during the interim period while Sir Gerald
Portal on behalf of the British Government investigated the situation in
Buganda in 1893 and made recommendations upon the future relations be-
tween Britain and Uganda. Toro in particular suffered serious depredations at
the hands of the Sudanese troops left by Lugard to protect the country. Having
received complaints from Bishop Tucker, the Anglican Bishop of Eastern
Equatorial Africa, Portal sent Major Roddy Owen to withdraw the Sudanese
garrisons. But Owen quickly realized that to carry out his orders would be to
leave Kasagama at the mercy of the ever watchful Kabarega and Owen's own
lively temperament favoured an all-out attack upon the aggressive ruler of
Bunyoro. This was scarcely possible in view of Britain's uncertainty regarding
her future obligations in Uganda. Furthermore Portal, suffering from ill-health
and stricken by the news of the death of his younger brother, Captain Raymond
Portal, who had accompanied him to Uganda, had begun his return journey
to the coast, in May 1893, leaving Major J. R. L. Macdonald temporarily in
charge in Uganda with instructions to refrain from intervening in native affairs
except where absolutely necessary.
The need for action along the lines which so appealed to Major Owen came
more quickly than anyone had anticipated. A rising of the Muhammadans in
Kampala was quickly checked by Macdonald, but the defeated rebels retreated
westward cutting Owen's communications with Kampala. Owen, however, was
fully occupied with affairs in the western kingdoms. It was swiftly becoming
clear that the possibility of the withdrawal of the Sudanese garrisons was
arousing Kabarega's rapacious instincts, while the capture of a caravan of
ivory and slaves despatched by Kabarega to the south revealed that the barrier
which it had been hoped to erect between Kabarega and the outside world
had not been securely established. There were also indications that in spite of
10 Ibid. p. 290.

his treaty with Lugard, Ntare was exacting levies from illegal caravans passing
through Ankole instead of trying to stop them." In these circumstances Owen
felt it vital to maintain at least one fort garrisoned if British prestige and influ-
ence were to be preserved in central Africa. Fort George was his choice. Situated
in a dominating position on the narrow neck of land between Lake Edward
and the Katwe Salt Lake and commanding an excellent route into the Congo
the fort was of the greatest tactical importance. But now Owen could no longer
ignore the Muhammadans in his rear. With his usual courage he turned to face
an overwhelming force and inflicted a severe defeat upon it. No sooner had this
danger been averted, however, than his plans to protect the west received a
check in the form of a suggestion from Macdonald that Kasagama should
withdraw from Toro and take up an area of land in Singo offered to him by the
Baganda.12 Owen immediately countered with an enquiry as to what would
become of Kasagama's subjects if the plan were put into effect. His own attitude
remained unchanged. Now that the Muhammadans were defeated he was con-
fident that he should and could hold on to the forts in Toro.
Kabarega's increasing aggressions at last convinced Macdonald that with
the Mukama in Bunyoro in such a mood the future of Uganda was likely to
be jeopardized. He agreed, therefore, that a full-scale campaign should be
launched as soon as possible. In the meantime he authorized Owen to try to
win over some of Kabarega's chiefs.
Before the main campaign could be launched Colonel Henry Colvile arrived
in Uganda as British Commissioner. Colvile had instructions to check a
rumoured Belgian advance from the Congo Free State and to protect British
interests in the Nile Basin. In these circumstances Macdonald's proposed cam-
paign suited his plans completely. In November he instructed Owen to com-
mence operations from the south-west and in December the main attacking
force consisting of a small body of Colvile's own troops and a host of Baganda
spearmen, advanced from the south-east. The campaign was a memorable one
for it led to the ultimate defeat of Kabarega and the dismemberment of
Colvile's advance was well-timed, for Kabarega had just succeeded in over-
running Toro and driving Kasagama to seek refuge in the foothills of the
Ruwenzori Mountains. News of the threat to his rear induced Kabarega to
hasten northwards, where, after two months campaigning, he was conclusively
defeated. A line of forts was built across the centre of Bunyoro to prevent the
Mukama from making contact with the southern portion of his kingdom. For
some time, however, he continued to threaten the line of communications
between Kampala and Bunyoro, but after a number of setbacks he withdrew
northward across the Nile towards the end of 1894. Meanwhile Colvile's expedi-
tion had returned to Buganda leaving a small force in Bunyoro under the
command of Captain A. B. Thruston who had instructions to try to make a
treaty with Kabarega. The Mukama, however, was in no mood for treaty-
The effects of the campaign upon Bunyoro were extremely grave. A pros-
11 Entebbe MSS. Inward 1893. Owen to Kampala. Fort Rosebery. 19 June 1893.
12 Entebbe MSS. Outward 1893. Macdonald to Owen. Kampala. 19 and 26 June 1893.

perous countryside was reduced to desolation, and villages were left unin-
habited. Nor could the Banyoro look to the British occupation for the sort of
benefits which a civilized administration might normally be expected to bring
to a defeated country. For Britain was still undecided about its attitude to
Bunyoro. In August, 1894, Colonel Colvile had been able to announce a British
Protectorate over Buganda but there for the time being official British commit-
ments ended. The British occupation of Bunyoro, therefore, amounted to little
more than a frontier garrison to protect Buganda against a resurgence of
Kabarega's power. In addition, as a reward for their participation in the
campaign, the Baganda gained large areas of land which had formerly belonged
to Bunyoro. Among these was the territory now included in Mubende District
which had for centuries been the heart of the kingdom of Bunyoro and which
contained the tombs of many former Abakama. No official record of this trans-
action was made at the time, however, so that for more than a year the situation
remained clouded. In 1896 the Commissioner, Mr. Ernest Berkeley, sought to
enforce the claims of the Baganda and later the transfer took effect. In
the 1900 Agreement between Britain and Buganda the disputed area was at
length formally included within the borders of the Buganda kingdom. The
Banyoro had indeed suffered heavily for Kabarega's rapaciousness and for his
determination to maintain the independence of his kingdom.
If the position of Bunyoro was uncertain so too was the relationship between
Britain and the more southerly kingdoms of Toro and Ankole. In March, 1894,
after restoring Kasagama to the throne of Toro, Major Owen made a treaty
with the young Mukama, ostensibly on behalf of the British Government, the
terms of which were similar to those of the earlier agreement made by Lugard.
But Owen's treaty could not be regarded as official unless the British Govern-
ment decided to honour it. In the same way and equally unofficially a treaty
was made by Major Cunningham on behalf of Her Majesty's Government with
Ankole in August, 1894, after Nuwa Mbaguta, the Nganzi or chief minister,
had asked for British protection. But the strange situation was regularized in
1896 when Britain agreed to extend her protection to Bunyoro, Toro, Ankole
and Busoga. Even now, however, the relationship was not wholly clear. The
term 'Protectorate' had an elastic connotation. Basically it implied nothing
more than an agreement whereby a strong power was prepared to protect a
weaker from external threats and possibly from internal rebellion while in
return the protected power would refrain from any relations with other powers
without the consent of the protector. It was in this simple form that protection
was extended, for example, to Ankole in 1896 although the situation changed
appreciably in 1898 and again in 1901. But beyond this minimum arrangement
there were many degrees of closer relationship ranging from participation by
representatives of the protecting power in the administration of a country which
seemed to lack sufficient men of ability and integrity for its own government to
the complete control of a country's affairs by the protecting power. It was due
to the existence in Uganda of a number of protected districts, each requiring
a different degree of supervision, which led to the establishment of an overall
Protectorate Administration distinct from the governments of the individual
protected areas.

The declaration of a Protectorate in fact did little to protect Ankole from her
enemies. The Banyaruanda were raiding within the very borders of Ntare's
kingdom, and soon after driving them back. Ntare himself fell victim to small-
pox. A troubled period followed during which a number of rivals claimed the
right to succeed him and again it was due to the intervention of Mbaguta that
decisive action was taken. After appealing for help to Kabaka Mwanga, the
ruler of Buganda, Mbaguta succeeded in placing the youthful Kahaya, Ntare's
only son, upon the throne of Ankole. This did not put an end to Ankole's
difficulties. Whatever his inclinations Kahaya was in a poor position to act
effectively since he himself was far from secure. His succession had not been
accepted without opposition and the tributary kingdoms which had acknow-
ledged Ntare's overlordship were not equally prepared to admit subordination
to his young successor. Mbaguta, now Kahaya's chief adviser, recognized that
if he wished to stabilize the situation in Ankole he must ask for the active
intervention of the Protectorate Administration. In 1898, therefore, he appealed
to the British to make their protection effective and an Administrative Officer,
Mr. R. J. D. Macallister, was sent to take charge of Ankole. Kahaya then made
the strategic decision to set up his headquarters alongside the Protectorate
The Protectorate in Bunyoro took on a more severe form. Trevor Ternan,
who became Acting Commissioner in Uganda in 1899, maintained that Bunyoro
as a conquered country might be administered in any manner that the Protec-
torate Administration thought best. The powers of the Mukama were declared
to be in abeyance during the minority of Kitahimbwa, one of Kabarega's sons
who had been proclaimed Mukama in place of his fugitive father during the
previous year. All authority was now to reside in a Sub-Commissioner.
Shortly before this Apolo Kagwa, Stanislas Mugwanya and a number of
other prominent Baganda chiefs had made the surprising claim that Buganda
had formerly exercised sovereignty over Bunyoro, Toro, Ankole and Busoga
and had asked that Buganda's rights should be restored.'1 George Wilson, the
British official to whom this claim was first made, investigated it with charac-
teristic thoroughness and sympathy. He concluded beyond doubt that although
Busoga might formerly have been a tributary of Buganda the claim to any sort
of sovereignty over the western kingdoms had no foundation. He did, however,
suggest that a compromise might be achieved if a central council were set up
consisting of delegates from all the kingdoms south-west of the Nile.14 The
idea had some appeal for Berkeley who, however, left Uganda before any action
could be taken to investigate the proposal in detail. And with the arrival in
1899 of Sir Harry Johnston as Special Commissioner to make recommendations
on Uganda's future Wilson's scheme was relegated to the background.
After negotiating a detailed agreement with Buganda in 1900 Johnston turned
his attention to the western kingdoms. Although Kabarega had been captured
and exiled in 1899 Bunyoro was not thought to be ripe for an agreement. Ankole,
also, was in Johnston's eyes too disunited to make the signing of a treaty a
13 Buganda Residency MSS. Apolo Kagwa, Stanislas Mugwanya and others to the
Sub-Commissioner, Uganda District, 8 November 1898.
14 Buganda Residency MSS. Wilson to Berkeley, 8 February 1899.

sound proposition. He did, however, recognize Kahaya as principal chief and
he suggested that an agreement might be considered at a later date, possibly
in two years' time. In Toro, however, the situation was different. After his
second restoration Kasagama had quickly become the prot6g6 of the Church
Missionary Society. In 1894 two qualified Baganda teachers had been invited
to Toro and some of the leading chiefs had begun to build churches. There was
a brief set-back in 1895 when Kasagama was charged with trading in gun-
powder and with attempting to bribe the Government interpreter. He fled to
Mengo where the charge against him was thoroughly investigated and he was
acquitted. He took advantage of his stay in Buganda, however, to receive
further instruction in the Christian faith and on 15 March 1896, was baptized
by the Rev. E. Millar. Returning to Toro he encouraged the expansion of
missionary activity and by 1900 the first European women missionaries had
arrived in his kingdom in response to his appeal on behalf of the women of
Toro. Kasagama, therefore, appeared to Johnston as an enlightened young ruler
and a fitting signatory to a treaty. An agreement was accordingly made in 1900
which, however, was soon to prove unsatisfactory. It is possible that Johnston,
who was unable to spend as much time in preparatory work as he had done
in Buganda, failed to understand the position which Kasagama occupied. He
recognized that Toro as re-created by Lugard was a larger kingdom than it
had formerly been. He therefore agreed that Kasagama should function as
Saza chief of the old Toro, described by Johnston as Toro Proper, while at the
same time he would be accepted as supreme chief of Lugard's Toro. This was
an unworkable arrangement. Toro Proper consisted of four Sazas each of
which required the attention of one Saza chief. In addition, Kasagama's position
as at once the overlord and equal of the other Saza chiefs of the new state was
an invidious one. Fortunately the Saza chiefs of Toro Proper who had been
displaced by Johnston continued to serve Kasagama loyally until, in 1904, the
difficulty was brought to the attention of the Protectorate Administration by
Mr. H. E. Maddox of the C.M.S. After some enquiry the problem was taken up
by Mr. George Wilson, and in 1906 Sir Hesketh Bell obtained the Colonial
Secretary's approval for redividing Toro Proper into four Sazas and for the
recognition of Kasagama as Mukama of the whole of Toro.
In one important aspect the Toro Agreement differed fundamentally from
the Agreement with Buganda. With the exception of a limited number of grants
of land to Kasagama and to a few leading chiefs no allotments of land in free-
hold were made in Toro.
In spite of Johnston's scepticism Ankole was anxious not to be left out of
the treaty-making. In April, 1901, Kahaya visited Entebbe to ask for an agree-
ment similar to that granted to Toro. The Acting Commissioner, Mr. F. J.
Jackson, replied that no treaty could be made until the other chiefs of the area
recognized Kahaya's sovereignty.15 Like Lugard, Jackson probably mistook
the tributary relationship of the neighboring chieftains for absolute subordina-
tion to the ruler of Ankole. It was, in fact, only with the assistance of Mr. R. R.
Racey, who succeeded Macallister as Administrative Officer, and after some
fighting that Mbaguta was able to induce the neighboring chiefs to recognize
15 F.O.2/464. F. J. Jackson to Lansdowne. Entebbe, 31 October 1901.

Kahaya's overlordship. The chief of Igara committed suicide rather than ack-
knowledge allegiance to Ankole. Nonetheless, in August, 1901, George Wilson
visited Ankole and came to the conclusion that the country was ready for an
agreement. Acting on Jackson's instructions he drafted a document similar in
terms to the Toro Agreement. This he explained with great care to the chiefs
and people of Ankole and the Agreement was signed on 7 August 1901. It was
later withdrawn by the Protectorate Administration in 1905 in consequence of
the murder of the Sub-Commissioner in Ankole, Mr. H. St.G. Gait. This evil
deed was the work of certain chiefs who superstitiously attributed the difficulties
which Ankole was undergoing to British intervention. The chiefs were banished
and a communal fine was paid by the people of Ankole. In 1912 the Agreement
was restored.
The immediate results of British intervention in the western kingdoms were,
therefore, of the greatest importance. The power of Bunyoro was permanently
destroyed. The young kingdom of Toro was firmly re-established and enlarged.
Most of the Hamitic kingdoms of the south-west were consolidated under the
overlordship of the ruler of Ankole. Compared with these events the peaceful
introduction of a modern form of administration and the attempt to set the
western kingdoms on a prosperous footing which were the main tasks of the
twentieth century seemed relatively unexciting. Nevertheless, these latter
developments were rendered difficult by a variety of causes, some provided by
nature and some by man himself. Of the first sort the plague of sleeping sickness
which swept the areas along the shores of Lake Albert and Edward and the
banks of the Nile in the first decade of the century was perhaps the most severe.
It became necessary to clear those areas of all their human inhabitants and
re-settle the latter in new country. It was a task which challenged the courage
of the boldest administrator and fortunately such a man was available in the
person of Sir Hesketh Bell, Commissioner and Governor of Uganda between
1905 and 1909.
The two most persistent problems which faced western Uganda, however,
were the need for money to assist development and for suitable personnel. The
latter problem was one which assumed serious proportions owing to the shortage
of British administrators due to the Imperial Government's desire to economize
as much as possible in the administration of Uganda. The indigenous authori-
ties in western Uganda, however willing and able they may have been, and they
were not always either willing or able, were untrained in the type of administra-
tion which was now being introduced. The ability to command a body of
Kabarega's warriors on a raiding expedition was not necessarily the quality
required of a new generation of leaders. To ease the situation, therefore, a
number of Baganda were introduced into the western kingdoms as chiefs and
held office alongside the Banyoro, Batoro and Banyankole chiefs. Kigezi pre-
sented a separate problem. The District in its present form is the product of the
Anglo-German-Belgian Boundary Commission of 1911. Before this time the
Uganda-Congo boundary had been undefined although the line of longitude
300 East had been agreed upon as the dividing line between the two territories.
This line, however, took no account of geographical features and was not
demarcated on the ground. The result was that on a number of occasions

Belgian expeditions tried to lay claim to land regarded as lying within the
British sphere. These unfortunate occurrences came to an end after 1911 and
relations on the border thenceforward were of a friendly character. The new
boundary, however, meant that an area of land was added to the south-west
of Uganda and this, together with a part of the former kingdom of Mpororo,
became Kigezi District. The dominant tribe in the area were the Bakiga, a
sturdy agricultural people who recognized clan obligations but in general had
no social organization of a more complex or widespread character. Under
British administration these people were permitted to select their own chiefs
who were then educated in their duties by Baganda agents who lived with them
and gave their advice.16 For the most part these agents acquitted themselves
well. When the last of them was withdrawn in 1929 they had established a
tradition of honest and efficient administration fit to act as a model for the
whole Protectorate.17
The main duties of the chiefs were to assist in maintaining order, to
administer justice, to collect taxes and to help their people to develop the
prosperity of the District. Possibly because the Protectorate Administration
laid great emphasis upon the importance of collecting revenue and possibly
because their income in the early years was derived from court fines and from a
rebate of a proportion of the poll-tax, tax collecting and the administration of
justice were carried out by most chiefs with a fair degree of efficiency. It was as
leaders of their peoples' progress that many of them displayed their short-
comings. To this there were exceptions. In spite of recurrent outbreaks of witch-
craft which troubled the District right up to the 1920s, in spite too of their
tribal associations with trouble makers over the Belgian and German borders,
the chiefs of Kigezi displayed both enthusiasm and aptitude for their work. In
Ankole, after an indifferent beginning overshadowed by the murder of the
Sub-Commissioner the chiefs almost without exception began to carry out their
tasks with fair competence. In this they were no doubt inspired by the leader-
ship of Nuwa Mbaguta who dominated the Ankole scene for forty years. In
Toro and Bunyoro, however, the quality and ability of the chiefs varied greatly.
While some served their people well for thirty years or more the majority
sought only to exploit their authority at the expense of their subjects. In Toro.
too, the leaders were disgruntled at the delay in implementing the land clauses
of the 1900 Agreement. Each District had, however, a Mukama who had been
greatly influenced by missionary teaching and each in his different way played
an important part in the affairs of his kingdom. Kasagama of Toro, although
frequently unpredictable in his behaviour, displayed his enlightenment in very
gloomy surroundings. In Bunyoro, Andereya Duhaga, who quickly succeeded
his brother, Kitahimbwa, was a quieter personality than Kasagama but for
nearly a quarter of a century his calm influence was felt both within the circle
of the royal household and also in the peace and good order of his formerly
turbulent kingdom.
In both Toro and Bunyoro the Church Missionary Society and White Fathers
were active in promoting both Christianity and education. Both missions
16 Western Province Annual Report, 1912-13. p. 16.
17 Report on the Western Province for the half year ending 30 June 1929. p. 15.

regarded it as being of the greatest importance to enable their better students,
the future leaders of their people, to escape from their uninspiring background
to Buganda where they found places in the upper schools at Budo and Rubaga.
In Bunyoro in particular there was a ready response to the educational oppor-
tunities that were offered and in 1914 the Governor laid the foundation stone
of the Kabarega School near Masindi which was to be built by the Mukama and
chiefs in memory of their former ruler."1 The results of these educational
activities were not always what had been expected. Wider education brought
with it a decline in the respect shown to the old-fashioned chiefs. This situation
had its advantages in providing a check to corruption and oppression, but it
had its drawbacks because it created a spirit of irresponsibility among some of
those who should have been the chief's natural successors.
One of the main disruptive forces in both Bunyoro and Toro was the delay
in introducing a permanent land settlement and even of implementing those
aspects of the 1900 Toro Agreement which referred to grants of land to some
of the chiefs. There was a similar delay in Ankole but the actual possession of
land in freehold was not such an important matter to a pastoral aristocracy
which looked only for grazing grounds for its herds of cattle. In Toro, however,
the delay produced a state of chronic insecurity among the peasants who lived
in constant fear lest the land they had carefully tended should be claimed by a
chief by right of the Toro Agreement. Tenants were frequently evicted on these
grounds and there was little encouragement to progressive agriculture. In
Bunyoro, meanwhile, the Government's inaction over the land issue led to fears
that areas needed by the Banyoro might be allotted to European planters. It
was not in fact till the end of 1915 that the alienation of land in freehold to
non-Africans was forbidden in Uganda, and for some years this prohibition
was still widely believed to be only a temporary one. In these circumstances
and in view of the comparative popularity of Toro and Bunyoro among Euro-
peans the fears of the inhabitants cannot be regarded as entirely groundless. In
fact, a committee had been appointed in 1907 to consider the land question but
it had confined its activities almost entirely to the position in Buganda. It did,
however, without any detailed consideration of the subject recommend that the
Kiganda system should be introduced into the western kingdoms and Busoga.
No action was taken on this recommendation but a new committee was set up
in 1911 under the chairmanship of the Chief Justice, Mr. (later Sir) William
Morris Carter. Working on the assumption, now clearly untenable, that all land
in Uganda had formerly belonged to the chiefs the committee repeated the
view that the Kiganda pattern should be adopted throughout the Protectorate.
The proposal was accepted by the Governor, Sir Frederick Jackson, but was
rejected in 1916 by the Secretary of State for the Colonies, on the ground that
it was contrary to the interests of the peasants and to the development of the
country. In 1921, therefore, a more limited scheme was introduced to define
the chiefs' rights in land in the light of the now almost historical Toro and
Ankole Agreements. In Toro this belated action did not satisfy the chiefs when
they learned that they must confine their choice of land to areas which they

18 Northern Province Annual Report, 1914-15. p. 74.

had developed themselves or through an agent.19 Some also found that the land
to which they were entitled was considerably less than the area which they had
claimed for the past twenty years, while a further cause of discontent arose
from the fact that the peasants tended to vacate the chiefs' land as soon as it
was clearly defined so as to avoid the oppression of a landlord. In addition
many of the chiefs were unable to find the money for their certificates of owner-
ship. In 1926 the Mukama of Toro himself was called to Entebbe and repri-
manded for his outspoken criticism of the land settlement. Meanwhile, the
peasants were still uncertain as to the terms upon which they occupied their
plots. Early in the 1930s, therefore, this problem was tackled in western Uganda
by the issue of certificates of occupancy which it was hoped would give the culti-
vators the security of tenure upon which they could build a stable agricultural
society without making them the owners of land. In Bunyoro a similar arrange-
ment was introduced a little later as a bye-product of the Bunyoro Agreement
which was signed in 1933.
Important though the land issue was it was only one of a number of kindred
problems. Another was the inadequacy of communications. At the beginning
of the twentieth century traders flocked to Toro lured there by the ivory trade
over the Congo border. By 1907 there were already forty-five traders in the
District, three Europeans, thirty-one Indians and eleven Arabs.20 Ankole, too,
attracted a reasonable number of traders, among them an agent of that prince
of East African merchants, Allidina Visram. In Ankole hides and skins were
the chief product and Americani was the main import. Bunyoro, also, drew
profit from the ivory trade in the early years of the century until the transfer of
the Lado Enclave to the Sudan in 1909 diverted the ivory trade of that area to
the north.21 Owing to the absence of good roads the only regular outlet for the
produce of western Uganda was on the heads of porters. There was little scope,
therefore, for attempting to produce a large-scale economic crop which would
make impossible demands upon the supply of porters. Nevertheless, tentative
efforts were made in the very early years of the century to develop some econo-
mic crop suited to the climate and agricultural habits of each of the peoples of
the western kingdoms. Cotton was tried with poor results for many years. In
Ankole the population showed little interest in cotton growing and persistent
endeavours on the part of members of the Department of Agriculture only
served to prove that the greater part of the District was unsuitable for cotton
growing. In certain areas of Toro where climatic and soil conditions seemed
propitious the lack of any outlet for the crop soon discouraged the peasants.22
Similarly, in Bunyoro, in 1910 attempts to arouse enthusiasm for cotton were
frustrated by bad weather, by the poor prices offered by unscrupulous dealers
and by the demands made upon the population for a supply of porters.23 Wheat
which grew in parts of Toro found a market over the Congo border for a number

19 Western Province Annual Report, 1921. p. 9.
20 Western Province Annual Report. 1906-7. Toro District, Trade.
21 The Lado Enclave on the west bank of the Nile was granted to Leopold II of the
Belgians, ruler of the Congo Free State, for his lifetime. Before his death in 1909 the
natural outlet for the ivory trade of the area had been through Bunyoro to the east coast.
22 Western Province Annual Report, 1912-13. pp. 4-5.
23 Busindi District Annual Report, 1910-11. p. 5.

of years. But its quality was poor and it could scarcely have been sold else-
where even if means of transport had been available. During the 1914-18 War
the demand for Toro flour for the Belgian forces gave a temporary boost to the
trade but the Mukama and chiefs who invested in the Toro Mills Company
were disappointed in their hopes of a good return. This was partly due to
competition from two other flour mills, but even more to the apathy of the
chiefs in failing to encourage the cultivation of wheat. As a result, when the
wartime demand came to an end the importance of wheat growing declined
Save for the need to raise money by taxation the importance of developing
peasant agriculture was not always appreciated by all the administrative officers
in the Protectorate in the early years of the century. Chief Justice Carter's
Committee, for example, assumed that the future prosperity of the country
would be based upon plantation crops and this point of view had already
received some approval in England. For a variety of reasons history proved
this prophesy to be wrong. As a result, although Toro and Bunyoro had greater
attractions for planters than other parts of the Protectorate, plantation agricul-
ture never became a completely dominant feature of the economic development
of the two kingdoms. The planters who settled in the west at the beginning of
the second decade of the century were, in fact, faced with the same problem as
the peasants, that of transport for their produce. After the first apprehensions
of the Africans had been allayed the planters met with no opposition from the
people of the west and there was usually an adequate labour supply to make
their work possible.
Although the labour supply appeared to be adequate there was, however,
considerable concern among administrative officers as early as 1910 due to
the steady flow of labour from the western kingdoms to Buganda. The move-
ment not only deprived the area of an important section of its manpower but
also reduced the poll-tax collected in the districts. In addition, as the years went
by the disruption of society occasioned by these migrations became a serious
problem. Undoubtedly the higher wages paid in Buganda were a great attrac-
tion to people whose own plots of land produced little that was marketable.
Nevertheless it is also fairly clear that many of the young men and women who
trekked eastward did so in search of adventure as much as for a purely econo-
mic motive. The independent spirit of these migrants was reflected in the fact
that the blandishments of labour recruiting agents made little impression upon
them and they preferred to go to work on estates where they or their friends
had received good treatment in previous years. Returning as they did from
time to time to their own districts they were quickly reabsorbed into tribal
society. Detribalization was a temporary condition applicable only to the
migrants during their absence from their home country. Nonetheless their wider
experience proved of considerable importance in introducing a new spirit and
wider outlook into the western districts.
The 1914-18 War brought a variety of difficulties to planters and peasants
alike. The danger to Ankole and Kigezi was immediately apparent and for a
brief period it seemed as if the close tribal links across the Anglo-German
border in the extreme south-west might make it necessary for the British

Administration to evacuate Kigezi District. The banks of the Kagera River,
separating Ankole from German East Africa, soon became the scene of active
patrolling and skirmishes between the opposing forces. Heavy demands for
porters were made upon all the western districts and as the forces were built
up for the invasion of German East Africa in 1916 Ankole became both the
British military headquarters and also an important source of supplies for the
troops, the provision of food fully occupying the peasantry of the district. The
planters' crops also, coffee and rubber (cocoa had quickly proved itself of little
use) were just beginning to bear when restrictions upon trade imposed 'by the
submarine campaign were added to the difficulties of internal communications
to make the marketing of the crops almost impossible.
The end of the war brought with it a temporary boom in trade. There was a
heavy demand for all the goods of which the country had been starved for four
years. The planters celebrated the declaration of peace with a bumper coffee
harvest which they were only able to harvest with Government assistance and
by the importation of labour from West Nile into Bunyoro.24 Robusta coffee
grown by Africans in Toro also found an outlet for the first time. By 1920,
however, the effects of the world-wide economic depression reached Uganda
and the price of coffee dropped so sharply that it did not pay the growers to
harvest it. Cotton, too, felt the deadening impact of the crisis just as severely.
Soon afterwards there came a marked revival and in Bunyoro the cotton crop
began for the first time to show promise. Indeed, the 1920s witnessed an
awakening among the Banyoro which must in part be ascribed to the activities
of the new Mukama, Tito Winyi, who succeeded Andereya Duhaga on the
latter's death in 1924. The new Mukama was a man of energy and ability and
his firm guidance did much to strengthen a country which tended to look back-
wards to its former greatness rather than forward to the possibility of a pros-
perous future. Tito Winyi's accession marked the beginning of a new age in a
number of ways. The previous year Kabarega himself had returned from exile
to Uganda and had died almost immediately. He was buried at Mparo, near
Hoima, the site of his last capital. The new ruler now inherited the strength
of character of the old warrior and added to it the foresight necessary to lead
his people into a new age. In the 1930s therefore, when the planters of Bunyoro
saw the price of rubber drop to such an extent that it was not worthwhile to
tap their trees and when even the coffee growers were suffering heavy losses,
the peasant cotton crop expanded steadily and to it was added a new source of
wealth in the form of tobacco. This latter particularly appealed to the peasants
because it could be grown to produce two crops a year. In the 1930s peasant
agriculture in Toro also began to show signs of improvement. A start was made
at growing Arabica coffee under the direction of the Department of Agriculture.
The Batoro were not the best of cultivators but some profit accrued. In Kigezi
experiments were made with coffee and wheat and although many difficulties
were encountered the industrious population contrived to prosper. In parts of
Ankole, also, coffee proved to be a useful crop.
With the exception of the 1933 Bunyoro Agreement there were no important

24 Northern Province Annual Report, 1919-20. p. 7.

constitutional developments in the western Districts in this period. Just before
the outbreak of the second World War, however, the introduction of a policy
of decentralizing certain administrative functions resulted in the amalgamation
of the Northern and Western Provinces. Hitherto, except for a brief period
during the first decade of the century when Bunyoro had been attached to the
Western Province, the northern kingdom had formed part of the Northern
Province. From 1939 onwards, however, Bunyoro remained in the Western
Province even though the two provinces were divided once again after the war.
Bunyoro's position of isolation in regard to the other western kingdoms was a
matter which from time to time gave rise to discussion some of which centred
round the possible return of Mubende District to Bunyoro. Apart from Bun-
yoro's own claim to the 'Lost Counties', Mubende, projecting as it did between
Bunyoro and Toro, denied the natural link between the western kingdoms. At
various times suggestions were made that the adjustment of the south-western
boundary between Buganda and German East Africa might by a consequent
addition of land to Buganda encourage that kingdom to look favourably upon
the return of Mubende to Bunyoro. But as the years went by the reiterated
request for the transfer of Mubende to Bunyoro became more and more a
forlorn hope.
The outbreak of the second World War once again imposed a heavy strain
upon the man-power of the Protectorate and some 20,000 men joined the forces
from the western districts. Bunyoro's main contribution was in supplying food
for the forces. The post-war period brought with it many revolutionary changes
in the west. The policy of the Protectorate Government after 1945 was based
upon economic recovery and expansion and every effort was to be made to
encourage the development of the country's resources. The prospects of mineral
wealth in the west had proved illusory. The discovery of tin in Ankole in the
mid 1920s had given rise to excessive hopes which quickly proved to be without
foundation. Some gold was mined in Kigezi but for most of the prospectors to
whom licences were issued the results were disappointing. The Ankole tin mines
had at their peak employed several thousand Banyaruanda labourers, the Ban-
yankole themselves finding the work too heavy for them. After the war a bolder
policy was adopted. Investigations indicated that the prospects for mining
copper and cobalt at Kilembe in the foothills of the Ruwenzori Mountains were
extremely good. Towards the end of 1953 Messrs. Frobisher Ltd., reached an
agreement with the Colonial Development Corporation and the Uganda
Development Corporation for providing the necessary finance to bring the
Kilembe mine into production at an estimated cost of 6,500,000.
Development of this kind required a substantial improvement in the country's
system of communications and one of the main results of the Kilembe mines
project was the westward extension of the Uganda Railway to Kasese. Almost
fifty years earlier Sir Hesketh Bell had dreamed of a railway stretching to the
Congo border and in the 1930s district officers had taken heart at the news
that a line was being surveyed to open up the west. It was not, however, until
the 1950s that these hopes were at last fulfilled. A committee appointed to
investigate the possibility of a westward extension submitted its report in 1951
under the title of The Way to the West. The Legislative Council agreed to

advance 3,250,000 to the East African Railways and Harbours in support of
the scheme and the line was officially opened in November 1956.
Kigezi had its own problem arising not from industrial development but from
the agricultural needs of its rapidly expanding population. The year 1946 wit-
nessed the beginnings of a bold scheme of voluntary resettlement whereby the
people of the overcrowded areas close to Kabale were moved away to an empty
part of the district. Meanwhile some 1,200 land-hungry people moved of their
own accord and unassisted from Kigezi into Ankole. Some, it is true, found
their way back to their old homes. In addition, the population of Kigezi was
increasing rapidly and there was little room within the district for further move-
ment, even if it had been easy to convince the people of the urgency of the
situation and to persuade them to move. Nevertheless a start had been made
with an experiment of the first importance and its results were vital to the con-
tinued progress of the District.
Politically, too, the post-war period was one of great advance. In both the
central government and also in the local governments changes of the greatest
importance took place. During the war, proposals had been introduced for
consolidating the lower councils and broadening the basis of the District Coun-
cils with a view to making them more representative of public opinion. These
proposals were put into effect immediately the war was over. In 1946, too, the
first three African members were appointed to the Legislative Council, among
them being Mr. P. Nyangabyaki, the Katikiro of Bunyoro. Four years later, in
1950, the representation of the Western Province in the Legislative Council
was increased to two members and in 1953 the District Councils were called
upon to perform a function for which they had not been chosen but which
carried with it great responsibility, namely the election of a representative from
each district to sit in the Legislative Council, which was now enlarged to in-
clude fourteen African representative members in addition to a number of
other Africans nominated by the Governor, of whom one member also came
from the Western Province. In 1955 the western kingdoms made yet a further
contribution to constitutional development in the Protectorate when, on the
introduction of a ministerial system of government, Mr. Z. C. K. Mungonya,
Nganzi of Ankole, became Minister of Land Tenure.
The District Councils themselves were being subjected once again to review
as a result of an inquiry carried out by Mr. C. A. G. Wallis into the system of
local government in Uganda. His report stressed the importance of avoiding
the impression that the object of the District Councils was to develop into
governments of petty states. This tendency which in the west had been
emphasized by the existence of kingdoms and by the example of Buganda was
one which it was difficult to eradicate. In 1954 Mr. Wallis's recommendations
were subjected to detailed consideration before a bill was passed by the
Legislative Council empowering the Government to introduce the new system
of district administration where he thought fit.
More important, however, in the eyes of the Banyoro at least, were the con-
stitutional changes accomplished by the new Agreement of 1955. The revision
of the 1933 agreement was brought about by a constitutional committee with
the assistance of Sir Keith Hancock and by its provisions the Mukama became

a constitutional ruler while the main responsibility for the administration of
the District was conferred upon the Katikiro and District Council, or Rukurato.
A similar attempt to revise the Toro Agreement failed when the Toro Rukurato
refused to accept the new District Councils Ordinance as the basis for a new
Throughout the twentieth century the main feature of life in the western
districts has been the pacific nature of the developments which have taken place.
The contrast between the warring days of the late nineteenth century and the
years which succeeded them is most marked. The war against the tsetse fly in
Ankole is perhaps the most vicious campaign which has been carried on under
the new administration. The passivity of the people has not always been a
good sign, however, for in some areas it has been the result of public apathy
which, starting with the chiefs, has permeated the life of certain districts. In
contrast the great advance from an extremely primitive society in Kigezi to an
advanced agricultural district is of particular significance. Nevertheless, the
quiet competence with which Bunyoro made its political advances in 1933 and
1955 was a reassuring sign after the despair of the 1890s and the lack of con-
fidence in the District's economic prospects before the first world war.



HIS article covers the history of the Bamalaki in Buganda from 1914 to
1919, the historically formative period' and the richest in documentary
material. It is by no means the whole story. I wish to express my thanks to
Mr. J. Kamulegeya for information about his father and much other help; to
Mr. P. Nkambo Mugerwa, to whose energy I owe much of the present material
which derives from interviews; to the Bishop of Uganda and to officers of
Government, who have given me ready access to files; to many private indi-
viduals who have given information; and to undergraduates of Makerere
College who have acted as interpreters and agents.
On 20 June 1891, the Rev. G. K. Baskerville records in his journal2 a visit
with George Pilkington to Joswa Kate Mugema, with the primary object of
persuading him to accept medical treatment for ulcers of the leg.
"Him we found clothed only in bark cloth and looking anything but one
of the greatest chiefs of the country being one of the largest landed proprietors
in the country. He is a Protestant and seems a most pleasant man We
induced the Chief to concede so far as to have his sores bathed ... he has
promised to regularly bathe them and if in five days they are no better, he
will come over to our place for treatment."
"We found the old man quite crippled and the ulcer very foul being just
covered with a piece of banana leaf ... He was a man of considerable power
and influence and it was natural that his household and dependents should
adopt his views."
There is no record either of his recovery or of his seeking medical treatment.
Indeed Baskerville records,3
"From then till now I believe he has never taken medicine."
Joswa Kate was born in 1850 and was clearly a man of strong character and
deep convictions, amounting at times to obsessions. He is described as "a quiet
thoughtful man, impossible to move from his convictions".
"We have regarded him as just a crank of years ago being otherwise an
excellent man of sound education. Bishop Tucker gave him a license as a
lay reader which empowered him to hold services and teach his people."4

1 From the Uganda Government Essay Competition, 1956.
2 The documentary sources used in this paper are a manuscript letter from Archdeacon
G. K. Baskerville to the Provincial Commissioner, Buganda, probably dated 27 October
1914; Baskerville's manuscript journal and typescript History of the C.M.S., Uganda
Mission (both held by the Bishop of Uganda); and archives of the Secretariat, Entebbe,
the Resident, Buganda, and the Provincial Commissioners, Eastern and Western Provinces.
Mugema's own Enjawukana is at present in process of publication and should yield
further material of value.
3 See also Mugema's own testimony in archives of the Resident, Buganda. He refused
medicine even in his last fatal illness.
4 This took place on Trinity Sunday, 28 May 1893. He was one of ten lay-evangelists
then "set apart and licensed" by Bishop Tucker. See Church Missionary Intelligencer,
October 1893, p. 760.

In 1898 he was responsible for sending the second unsuccessful party which
attempted to evangelize Ankole, Sir Apolo Kagwa having sent the first. In
startling opposition to Kiganda custom, he would not receive gifts nor eat with
those whom he visited, preferring to travel with his own cook. Even the Kabaka's
food he refused and thus earned the nickname Wambwa (Mr. Dog). He refused to
accept a medal awarded by Queen Victoria and became Semusota.5 In 1904, in
order to prevent the traditional ill-treatment of relatives of dead men with a
view to obtaining their share of the inheritance, he distributed his estate in
rupees and live-stock and declared that he had finished his funeral rites. No
doubt due to peculiarities of this kind, many other stories of doubtful veracity
are still in circulation. It is commonly said that those who fed with him had to
eat each food in turn and separately, ending with the salt.
"He refused to wear a headcovering for God had given him hair. He also
had a square bed made because he did not see why he should be forced to
lie in any given direction."
His son states that his refusal of medicine was based on the deep impression
made on him by the story of the burning fiery furnace.6 If God could save men
from fire, surely no human aid was needed in disease. In 1919 he was ready to
lose his chieftainship and his cattle rather than submit the latter to inoculation.
In 1920 the Principal Medical Officer writes: "He is now in my opinion mad;"
and the acting Provincial Commissioner records, "The attitude of Mugema was
so offensive that any discussion was quite impossible and I was therefore com-
pelled to tell him to leave the office."
Nevertheless, he was a man of great generosity, regularly distributing food to
people in the markets and to all who came to his headquarters. He was known
as Asasira abanaku (Merciful to the poor); and, for the next few years, he was
actively supporting the land claims of the newly-formed Bataka Society. His
son says that he had a secretariat at Lugala in Busiro, through which he for-
warded to the appropriate quarter any grievances which he thought legitimate.
The title Mugema is that of the head of the Nkima (Monkey) Clan, who as
ritual father (Nakazadde) of the Kabaka, instals a new kabaka and is one of the
few men entitled to greet the Kabaka standing. He was 'Katikiro of the dead'
and, as such, governed Busiro, where are most of the tombs of the kings.7 Joswa
Kate, who became Mugema in 1889, was thus not only a great landlord (he
received 26 square miles in the mailo allotment); he also had considerable
hereditary rights, both as an administrative chief and as head of his clan, and
a ritual importance attaching to his traditional relationship with the Kabaka-
an importance increased by the fact that he had personally installed the reign-
ing Kabaka. It is hardly surprising that, once he gave his support to a new move-
ment, it should flourish; and that he was the power behind the Bamalaki in
Buganda, there can be no doubt.
It is surely also significant that in the Eastern Province, near Mbale, another

5 Semusota guli mu ntamu. You cannot drive a snake from the cooking pot without
breaking the pot. The Kabaka could not try Mugema for contempt of the Queen without
defying their traditional relationship (see infra).
6 Daniel iii.
7 See a note on the title of the Saza chief of Busiro at the end of this article.

once-powerful and now discontented Muganda, Semei Lwakirenzi Kakungulu,5
of the same generation and experience, should be giving his support to a
similar movement. He was a friend of Mugema and the two are said to have
paid frequent visits to one another within Buganda. His first explicit connection
with the group appears to have been at Jinja, when on his way to take part in
the coming of age ceremonies of the Kabaka Daudi Chwa in August 1914. He
and his followers were informed that, before entering Buganda, they must be
inoculated against the plague. He refused and returned to Mbale. But there is a
strong suggestion that his formal defection from the Anglican Church was also-
due to a desire for divorce.
The first record of the formation of a new sect in Buganda is in a letter, dated
20 May 1914, from Mugema to the Assistant District Commissioner, Entebbe
(H. A. Mackenzie), in whose area Busiro lay; and, in view of later judgments,
it is important to notice the complete objectivity and sanity of this communica-
tion. He confirms the visit of Baskerville and Pilkington in 1891, says he has
not since then taken medicine and records discussions with the Rev. W.
Chadwick in 1911, Bishop Willis in 1912 and the Rev. F. Rowling in 1913.9 A
group, he says, has now separated from the Church Missionary Society and
built its own church at Kitala.
But there had been considerable preparation. According to his son, a small
group had for some time met in Mugema's house to discuss points of difference
from the C.M.S. Baskerville writes, "For many years we heard of no spread
of these opinions but somewhat (sic) the Mugema and others (3 or 4) began
writing letters to the Diocesan Council-to the Bishop and Archdeacon-say-
ing that they objected to taking medicine and that doing so was contrary to the
Bible and that doctors were emissaries of Satan." In 1912 Mugema had written
to the Diocesan Synod, protesting against prayer for doctors. But this small
body, though showing "a rigid refusal to have any dealings with doctors or
medicine", was still, until 1914, within the Church.10
The other members of the group to achieve any prominence in the movement
were Malaki Musajjakawa, James Biriko and Reuben Musoke. Malaki was a
local landowner of some importance, and a certified church teacher. On separa-
tion he was asked to return his certificate; but "this he refused to do saying that
he was no longer governed by the Diocesan Council but by the Book of God
alone". It seems probable that his original point of difference from the Church
Missionary Society was over the question of baptism. It is the practice of the
Mission to insist on at least six months' instruction prior to baptism. But
Malaki argued that John the Baptist had baptized freely without instruction,
that the Apostles had done the same and that the dominical command is to
"make disciples of all nations, baptizing them";" and the Provincial Commis-
sioner, writing to the Chief Secretary on 4 November 1914, says that Malaki
has mixed his interpretation with regard to baptism with Mugema's anti-doctor
8 The main facts about Kakungulu are given by Thomas, H. B., Uganda J., 6 (1938-9),
9 All three were members of the C.M.S.
10 Uganda Notes, November 1914; Cook, A. R., Uganda Memories, Uganda Soc., 1945,
p. 324.
11 Matthew iii.; Acts ii. 41, etc.; Matthew xxviii. 19.

teaching. The conclusions of the Diocesan Council, meeting on the same date,
were that, while Mugema and Kakungulu were strongly opposed to doctors, and
those coming for baptism were urged to have nothing to do with them, the great
majority were baptized without accepting the teaching. Attendance at Mengo
hospital had actually increased and Malaki's two children were regular patients.
In an interview with Mackenzie, Malaki asserted that he asked candidates for
baptism only if they believed in God and the resurrection of Christ. If they said
"yes", they were baptized and exhorted to go out and preach the gospel and to
follow the teachings of Christ and the example of his life. He did not ask people
to give up medicine but told them that he did not use it or believe in its efficacy.
He told Rowling that "the Germans are coming soon and they (sc. the Baganda)
ought to come and be baptized first".
There was, indeed, a strong rumour that, while the new movement was
officially opposed by the Buganda Government, members of that Government
were, privately, in favour of widespread baptism. Malaki quickly became the
main agent of the movement, partly at least because Mugema felt the responsi-
bilities of his office under government and is reported to have baptized few, if
any, with his own hands. The popular name Abamalaki (followers of Malaki-
anglicized as Malakites) was quickly given; and on 28 October 1914, the Pro-
vincial Commissioner recorded the view of the Buganda ministers that Malaki
was the only teacher of the new doctrine. He had originally been a follower ot
Mugema but had now branched off purely as a dissenter against reading for
baptism. Very few of his followers (as opposed to Mugema's) objected, the
ministers said, to doctors. But there is every evidence that Mugema and
Malaki, even if they started from different points of disagreement with orthodox
teaching, were firmly united in their separation from it. Indeed it is possible
that the publicly expressed view of the Buganda ministers was a deliberate
attempt to drive a wedge into an all too powerful partnership.
James Biriko was a catechist under Chadwick at Kitala. He is called
"Mugema's High Priest" when, in July 1915, he and Mugema had an incon-
clusive meeting with the Rev. E. Millar of the Church Missionary Society in
the Provincial Commissioner's office. In 1929 he is referred to as Mugema's
Musigire (Deputy); and he was co-signatory of many early letters, where he is
sometimes described as Omukulu w'Ekibina (Elder of the Society). After the
deportation of Mugema and Malaki in 1929,12 he emerged quite clearly as
religious head of the group, a position which, after some vicissitudes, he still
fills from his headquarters at Kazi in Bugerere. It is significant that Mugema
left the trusteeship of the group's funds to his son, Mr. Joswa Kamulegeya, a
loyal Anglican who still treats his trusteeship extremely seriously.
Reuben Musoke was a mailo holder and interpreter in the District Com-
missioner's office at Entebbe and appears later as a religious pamphleteer."
On 22 May 1914, Mackenzie acknowledged Mugema's letter of 20 May, say-
ing that Government does not interfere with religious beliefs; but, if the law
says a man must go to a doctor, the law must be obeyed. On 16 July Mugema
responded at length, quoting many references from the Old Testament, though
12 See infra, p. 158. A full account of the incident is outside the scope of this article.
13 See infra, p. 160. Reuben Musoke's main activities are outside the scope of this article.

the bearing of most of these on the point at issue is obscure. The main force of
the argument derived from Deuteronomy xviii. 11, where one of the people
banned is the charmer, translated in the Luganda version as omusawo, the
word normally used for members of the medical profession. We must obey the
word of God rather than of man. But there is also the insistence that we must
trust in God alone; and it is from this insistence that there derives the official
title of the group, Ekibina kya Katonda Omu Ayinza Byona14 (The Society of
the One Almighty God-usually abbreviated to K.O.A.B.). There is no sug-
gestion, although it was imputed at the time by Wallis and later by Roscoe and
Stock, of the Christian Science doctrine that illness is not real-rather,
"If God wishes to bring sickness or ill-health to human beings or animals,
no one can stop it; there are two facts in the Old Testament-obedience and
disobedience-and two in the New-belief and disbelief".
This is a very remarkable faith; and the conviction with which it was held
by the leaders must have accounted in large part for the success of the move-
ment. Mackenzie again replied to Mugema on 20 July, pointing out the two
meanings of omusawo and entering into theological argument. And there the
matter seems to have rested until the end of October, when the Provincial
Commissioner began to take an interest and reported to the Chief Secretary.
Baskerville, writing on 27 October, says that the Church Missionary Society
first heard of the new developments "almost a month ago" and that Malaki
had, at that time, baptized sixteen people at Mugema's County Headquarters
at Kitala.
The mass movement seems to have begun near Malaki's home at Kasenge,
where he baptized "at his home and in the swamp between Budo and
Kasenge".s1 On 9 November two African police officers arrived at Kitala at
about 9 a.m. and later reported:
"There were about 1,000 natives there and one Malaki was enrolling them
in Mugema's new religion. Each person was taken by the hand and if he
promised as follows he was enrolled: 'I will not drink the European's medi-
cine16 or go to them for any; when ill I will drink water only.' In the afternoon
they were to be baptized with water and receive a new name. Each man's
name is written on a separate slip of paper and all the slips are put in a
basket. When a teacher is required the basket is shaken up and each man is
allowed to draw a slip. Should he draw a slip with his own name he becomes
the chosen teacher. In this way Malaki was chosen. Mugema returned from
Budo about 2 p.m. and went to his house ... Protestants only are admitted
to the new religion; but if Catholics say they will go no more to a Catholic
Church they also are admitted."

14 Mugema et al. to P.C., Buganda. "The proper title of the sect is Aba Katonda
Omuinza Webintu Byona" (Those of the God who can do all things). But the printed
literature, and most of the correspondence, regularly refers to K.O.A.B. Eugene Stock
(The History of the Church Missionary Society, iv, p. 101, 1916) says, "They then began
to call themselves 'The Church which does not drink medicine' "; but there is no other
evidence of this name having been used.
15 Archdeacon Y. K. Bina, private communication.
16 Although some members of K.O.A.B. take Kiganda medicine in secret, the official
ban is on medicine of all sorts and not simply on European medicine.

Mackenzie's letter, enclosing this report, adds, in a passage which is
reminiscent of The Acts of the Apostles:
"Up to the end of October, 9,530 were baptized into this new sect. These
consist of all tribes Baganda, Banyoro, Banyankole, Nubis, and Swahilis and
sects Mohamedans, Catholics and Protestants and Pagans. The Counties
chiefly affected are Busiro, Kyadondo, Mawokota, Kyagwe, and Bulemezi
and the Sesse Gombololas. On Monday the 9 October I learn that Malaki
was going to Kakindo in Mawokota to baptize 700 people."
Mackenzie continued:
"My information differs somewhat from that of the Police report.
"1. Malaki merely asks a man if he believes in God. If the person says yes,
he is baptized.
"2. Roman Catholics and Mohamedans discard their beads and crosses
and place them in bowls separately. Malaki keeps these. Roman Catholics
are not baptized again but they are given a new name. Mahomedans are
"3. After baptism these people are considered as Protestants and can go to
C.M.S. Churches if they like.
"4. Many Bakope continue to go to the Doctor after they have joined the
new sect."
Baskerville17 has a different account again:
"Last Sunday Malaki baptized 4,247 people being engaged from early
morning when he began in a private house and finding the crowds too great
he then went outside with a vessel in his hands and baptized people up to
8 p.m. He asked them, 'Do you believe in God the Father Maker of Heaven
and Earth?' On an affirmative answer being given he poured water on them
saying 'I baptize thee in the name of the Father and of the Son and of the
Holy Ghost'."
It is not surprising that such a movement caused concern.
"The Katikiro thinks so seriously of it that he told us at the Bishop's house
two days ago that unless it is checked it might issue in, if not civil war, at
least in violent contact between these people and the orthodox members of
our church or with members of the Roman Church."
Although he had previously reported that Mugema had never refused to carry
out orders to send men to a doctor, Mackenzie suggested to the Provincial
Commissioner that action should be taken against Malaki for breach of the
laws regarding Infectious Diseases; and on the following day the Provincial
Commissioner wrote to the Chief Secretary, suggesting action under the
Venereal Diseases Ordinance, since "he has two small boys living with him
who have V.D.". On 13 November the Acting Governor (H. R. Wallis) ruled,
with characteristic moderation, that no legal action should be taken without
the Governor's special permission, unless the law was seriously broken-in
which case the Chief Justice was to be consulted. The Anglican Diocesan Coun-
cil decided, in agreement with the Roman Catholics, that baptism by Malaki
could not be accepted and that conditional baptism must be imposed on those
later wishing to join one of the mission churches; and Bishop Willis wrote to.
17 Baskerville is remarkable for his lack of punctuation.

Saulo M. Sebugwawo,"1 saying that nobody should be called by the name given
by Malaki. But a minute by Wallis on 21 November (which Sir Frederick
Jackson endorsed, 'I concur') says,
"The less notice that is taken of the movement, the sooner it will die out.
Persecution will only fan it into flame and should not be adopted unless there
is grave reason to believe the public health is endangered, which does not
appear to be the case at present".
And the Principal Medical Officer writes,
"These people at present seem careful not to break these laws and regula-
tions, as only one case has come to my notice and this was dealt with drasti-
cally by the Lukiko, with excellent results".
By the beginning of December Malaki had started preaching in Kyagwe,
where the Sekibobo (the late Hamu Mukasa) reacted violently, sending him
before the Lukiko which instructed him to return home to Kasenge. He refused;
and in January 1915 he and others were again expelled by the Sekibobo, since
they had not the Kabaka's authority as itinerant preachers. On 8 February the
Kabaka wrote to all saza chiefs, authorizing the Bamalaki to preach, providing
they obeyed the laws. This letter was conveniently interpreted by Mugema and
Malaki as a direct order from the Kabaka to itinerate and preach; it was
printed and circulated widely and the Rev. H. Bowers of Mityana reported: "for
the past few weeks, natives armed with a letter of authority signed by the
Kabaka ... At the end of April, Mugema refused an instruction from the
Kabaka to return the letters. A new instruction was issued to chiefs, saying
that they should not interfere with meetings of the Bamalaki, provided they
were orderly. It appears that the meetings were, in fact, interfering with the
authority of the chiefs and indirectly obstructing the supply of labour; and at
one point Mugema promised to stop such assemblies.
Meanwhile, on 2 January 1915, Bishop Willis had returned to the attack.
Writing to the Governor, he quotes the resolution of the Diocesan Council that
'Malakite' baptism cannot be accepted. It is granted without instruction and
motivated only by a desire for social status.19 The recognition of these names
by Gombolola chiefs is bringing the name of Christianity into contempt; and
he asks that they should be instructed to demand a baptism certificate before
registering a Christian name. The Chief Secretary comments,
"The main fact that there is a large section of the Baganda striving, no
matter how feebly, after something better, and desiring to call themselves
Christians is to me a tribute to the Church's work and influence, and I con-
sider the matter is one which should be dealt with sympathetically with the
Churches and by the Churches".
Even Uganda Notes (1914) comments:
"The movement is not without its good features. At the basis of it lies the

18 Presumably Sebugwawo was a Gombolola Chief. The Provincial Commissioner
rightly comments that this is a purely diocesan affair and that he cannot make chiefs
change their records. This is an extremely interesting example of the Church's attempts
to interfere in secular administration.
19 A very common motive in 'orthodox', as well as separatist, missions. Mr. A. H.
Cox (private communication) says that the K.O.A.B. was known by some of its critics as
Dini ya laisi (religion on the cheap), since baptism could be obtained simply for the asking.

determination to place implicit reliance on God alone. The widespread
demand for Christian baptism testifies like the Ethiopian movement in South
Africa20 to a desire for a church, indigenous and independent, and not exotic,
dependent upon foreign support".
The Governor replied in a manuscript letter to the Bishop, regretting that
Government could not intervene.
"It appears to me that so long as you issue cards proving baptism into the
Native Anglican Church the mere assumption of a Christian name will not
raise them in the social scale as they hope, but will only stamp them as
followers of silly, misguided, obstinate old men like Mugema, Kakunguru
and Malaki who have practically reverted to paganism;21 and when they come
to realize this the movement will cease, and the free use of ridicule will go a
long way to hasten its end."
Before the year closed "numbers of those so recklessly baptized were putting
themselves under proper instruction". But the sect grew apace, both in numbers
and new teachings. Polygamy, as in so many separatist movements, had been
an issue from the beginning. Did not Abraham and the patriarchs have more
than one wife? Wallis thought that natives were concerned by investigations
into their private lives by members of the C.M.S. and White Fathers' Mission.
The present secession was due to drastic discipline by the churches. Kakungulu
took a second wife in 1915. But Bishop Willis could say in November 1914
that chiefs who had joined the movement were "conspicuously moral". He knew
no case of a polygamous chief having joined. There seems no reason to suppose
that, though the Bamalaki were, and are, polygamous, polygamy has ever been
a fundamental part of their teaching.
More important was their teaching about the Sabbath. They "searched the
scriptures" voraciously and interpreted almost any text to their advantage; and
it did not take them long to discover Exodus xx. 8. On 10 May 1915 Mackenzie
notes,22 "This sect now calls Saturday its 'sabbath' ". On 7 March 1916 Mugema
wrote to Mackenzie, asking official permission for the K.O.A.B. to keep Satur-
day as the sabbath. This meant, of course, that they would not be available for
government work; and the reply was that, since Muslims work on Fridays,
Bamalaki could do so on Saturdays. Mugema (whom the Acting Provincial
Commissioner now thought "to be suffering from religious mania in an acute
form") answered at length, likening himself to Moses sent by God to Pharaoh.

20 Sundkler, B. G. M., Bantu Prophets in South Africa, Lutterworth, 1948, has a full
description and bibliography of the 'Ethiopian Movement'.
21 This charge is surely unjust.
22 The Seventh-day Adventists reached Zanzibar in 1903 and the Nyanza Province in
1909 (Olsen, M. E., A History of the origin and progress of Seventh-day Adventists,
Washington, D.C., 1925, pp. 510, ff). Joseph Booth, an Australian missionary with S. D. A.
leanings, visited Uganda between 1904 and 1906. In Nyasaland he had practised baptism
without probation and was responsible there for many separatist movements. But Booth
claimed that he first came into contact with sabbatarianism among Africans who had
apparently reached it through independent study of the Old Testament; and, in regard
to indiscriminate baptism, it is doubtful whether he did more than stimulate what was
already a widespread tendency in African separatism (Shepperson, G., Africa, 24, 233-46,
and private communications). It is tempting, but illegitimate to link the K.O.A.B. directly
with these missions.

Sunday is the day of the Lord (that is, of the resurrection) but not the sabbath.
Among the usual string of texts is the remarkable juxtaposition of Deuteronomy
x. 22 ("the Lord your God made you as the stars of heaven for multitude") and
Exodus xxiii. 2 ("you shall not follow a multitude to do evil"). Apparently a
suggestion was made that, as a special concession, in view of his long service,
Mugema himself might rest on Saturdays; but this he refused for himself alone.
Later, referring to the setting aside of Sunday, 6 January 1918, as a day of
prayer for victory, Malaki wrote to King George V, saying that this is not the
proper sabbath appointed by God. It is, in any case, useless to pray for victory
while there are still doctors in the land; but the Provincial Commissioner is
asked to cable the king for permission for the K.O.A.B. to hold their services
on 5 January. No objection was raised. There is no doubt that, to employers of
labour, this doctrine was a considerable nuisance.
Further reading suggested a ban on the eating of pork, a doctrine which led
later to the formation of a splinter group known as Balyambizi (pig-eaters).
But the ROOT of the new religion remained the objection to medicine for man
and beast. Bowers reported in May 1915 that, on the Singo Estates' plantation,
porters were threatening to run away rather than be vaccinated, and attributed
this to Malakite propaganda, since at Mityana there were more applicants for
vaccination than could be coped with. Mugema told the Mengo Lukiko that
the Vaccination Ordinance was contrary to God's law; and the Provincial
Commissioner thinking that the public health was endangered, considered
whether he should recommend his dismissal. Despite doubts in Entebbe, the
evidence grew. On 6 June three Miruka chiefs were listed as refusing permission
to vaccinate and, in view of the danger of the smallpox epidemic to the troops,
action was taken against them under the Vaccination Ordinance. The Principal
Medical Officer (annotated by the Governor, "I agree") even suggested the
possibility of appealing to martial law. In December 1917 Mugema was refusing
to assist in measures against rinderpest and in 1918 his sebalija (head cow-herd),
at Mugema's instigation, failed to report an outbreak of rinderpest involving
several hundred head of cattle. For continual refusal to co-operate he was dis-
missed from his sazaship on 24 June 1919. In 1921 Malaki renounced all his
cattle and was imprisoned for the same convictions.
It is surprising that, in view of this continued evidence of activity, Govern-
ment apparently persisted in regarding the movement as harmless. In May 1917
Mugema reported to the Provincial Commissioner a call to baptize natives in
the Bukoba District of Tanganyika. Yet in November 1918 the Chief Secretary
told the Provincial Commissioner, Western Province, that "under a treatment
of placid indifference the movement is quickly evaporating in Buganda". The
extent of this evaporation is evidenced by the fact that by 1921 there were
91,740 Bamalaki in Buganda alone23 and, under the influence of a "very active
evangelist", one policeman had been converted at Kitgum. In 1929 it became
necessary to deport both Mugema and Malaki, and the K.O.A.B. was giving

23 Uganda Protectorate, Census Returns, 1921, Government Printer, Entebbe. No
figures are available for the rest of Uganda till 1930, when the movement was clearly
on the wane and, against 56,952 in Buganda, there were 11,300 in the rest of the

ABAMALAK1 IN BUGANDA, 1914-1919 159
trouble as late as 1948.24 Groups of them are still to be found in Buganda,
Busoga, Bukedi and the Western Province; wealthy farmers in Bugerere and
Buddu pay their tithes (in some cases as much as 100) to the society; but their
numbers are few and they appear to have little social significance or influence
over their own young people.
Is it possible to account for the extraordinary success of this movement in a
country where western medicine was apparently so popular?25 Norman Leys26
refers to the unpopularity of doctors in Uganda due to depopulation made
necessary by sleeping sickness and the restriction of movement for bubonic
plague. It was recognized by Government that there had been friction over
venereal treatment at the newly-opened hospital at Mulago. But it seems much
more likely that the anti-doctor element was due primarily to the personality
of Mugema. Sir Gerald Portal had written,27 "the religion of the peasant is that
of his immediate superior, that is of the man who has most power to cause him
constant inconvenience"; Sir Frederick Jackson28 wrote of "winning over a
chief first, who in turn brings over with him a host of converts on the principle
of follow-my-leader". A man of Mugema's importance, force of character and
known popularity would readily get followers on these terms; and the same
must have been true of Kakungulu in the Eastern Province. Add to this the
demand for baptism on easy terms, itself the product of social unrest, and there
are all the makings of a popular movement. What is more difficult to explain
is its viability, despite constant defections. (In the absence of special divine
intervention, to which even the K.O.A.B. makes no claim, it would be reason-
able to expect a higher death-rate than in the rest of the community.) It is
difficult not to believe that, at the back of it, lay not only the vast disturbance
of society caused by the coming of Islam, of the Christian missions and of
British commerce and administration but, far more importantly, if for many
Baganda at that time unconsciously, the fundamentally religious question of
land, which the Lukiko had, in fact, failed to settle in 1900.29
"Johnston was able to write, 'I think I may say that nothing has tended
to bring about friendlier relations between the European Administration and
the native population than this adjustment of the land question'. For the
Baganda with whom he had had the greatest contact were primarily those

24 In 1936 2,133 Bamalaki in Bulemezi refused vaccination and rinderpest quarantine.
The Provincial Commissioner suggested proceedings against the ringleaders. In 1948 the
Colonial Insecticide Research Unit had to agree not to spray houses of Bamalaki. There
is a letter of complaint to the Resident, Buganda, from the K.O.A.B. at Kitala as late as
25 Cook, A. R., op cit., p. 93, speaks of 100-150 outpatients daily at Mengo Hospital
and of crowds up to 400 demanding medical attention outside Kampala in 1897-98.
26 Leys, N., Manchester Guardian, 8 February 1930.
27 Portal, G., The British Mission to Uganda in 1893, London, 1894, pp. 199 f. (quoted
Gale, H. P., Uganda J., 20 (1956) 75).
28 Jackson, F. J., to Sub-Collector, Jinja, 23 October 1901, quoted Gale, H. P., op. cit.,
p. 75.
29 Wilson, G., Letter dated 2 January 1903, bound with Lukiko Allotment Lists, says
that the difficulty encountered by the Lukiko in connection with the Bataka had been
dealt with by the Lukiko and had better be left alone by the Protectorate Government.
It involved about 2,000 people.

who had profited most from the 1900 Agreement... it is extremely unlikely
that the peasantry ever really understood what had happened."30
Thomas and Scott say, "From 1921-25 the Butaka controversy occupied a
prominent place in native politics." They add that it was laid to rest by the
Buganda Government Busulu and Envujo Law of 1927. How wrong they were
in thinking that it had been laid to rest is proved by the riots of 1945 and 1949,
in which the land issue, if not dominant, was still at least prominent.31 And the
high emotions which it can still arouse were shown by the recent public reaction
to the Lukiko's allocation of the remaining 154 square miles to private indi-
As early as 1900 Mugema had taken the initiative in going to the Special
Commissioner to call his attention to shortcomings in the original allocation
of land under the 1900 Agreement. Exhibits for the successful prosecution of
James Miti, Reuben Spartas and others for their part in the Disturbances of
194932 include evidence of gatherings of the Bataka Society at Mugema's house
in 1920, a letter from Mugema to the Governor in 1928 on behalf of the Bataka
and a letter signed by Reuben Musoke, the former interpreter in the District
Commissioner's Office at Entebbe, as Secretary of the Bataka Society. Other
connections through Mugema's office at Lugala have already been mentioned.
Spartas, although opposed to the Bamalaki, is leader of the African Greek
Orthodox Church, the only other genuinely separatist movement in the Uganda
Church; all his early letters are headed, "Africa for the Africans"; and, however
respectable he and his church may now have become, he himself would not now
question the nationalist sentiments inspiring its early days. There are, therefore,
clear links, if not explicitly between organizations, at least between the leaders
of religious separatism in Buganda and the outstanding political issue of their
time. There is a further link, at least through Spartas, between the early desire
to reject European leadership while remaining Christian in something more
than name-and the later development of a self-conscious political nationalism
with a total rejection of European leadership. It is inconceivable that the land
was not an important issue during the formative years of the K.O.A.B., how-
ever difficult it may have been, and still is, to discover overt evidence of con-
nection between the two movements in those years. Indeed, perhaps the most
important lesson which emerges from a study of Bamalaki is the realization of
how little is known of what is going on under the surface of life in Buganda;
and, of what is known, how little is understood.
Land; friction over compulsory medical and veterinary measures; people
"touched, at times deeply, by the power of the new religion"33-a religion which,

30 Ingham, K., Uganda J., 20 (1956), 7 f. The land question in Buganda is discussed
by Thomas, H. B., and Spencer, A. E., A History of Uganda Land and Surveys, Govern-
ment Press, Entebbe, 1938, Chap. XIII; by Buell, R. L., The Native Problem in Africa,
Macmillan (N.Y.), 1928, vol. i, pp. 594 ff; and by Thomas, H.B., and Scott, R., Uganda,
O.U.P., 1935, pp. 100 ff. See also, for comparative material, Kilson, M. L., The Land and
the Kikuyu, J. Negro Hist., 40, 2 (April 1955).
31 Report of the Commission of Inquiry into the Disturbances which occurred in
Uganda during January 1945 and Report of the Commission of Inquiry into the Civil Dis-
turbances in Uganda during April 1949. Government Printer, Entebbe, 1945 and 1950.
32 Principal Court of Mengo, Criminal Case No. 642 of 1949.
33 Ingham, K., op. cit., p. 8.

with an open Bible, is so readily subject to private interpretation; the demands
for baptism not simply as a religious but as a political symbol; a general unde-
finable unrest, involving at least some irritation with the occupying power; the
impact of a powerful personality of traditional public importance and strong
private convictions. Perhaps these are in themselves sufficient to explain the
growth of this peculiar movement. But it seems likely that its one lasting contri-
bution to Kiganda culture is the name malaki still given to suede shoes, the
shoes which need no polish.34

The title Mugema belongs properly to the head of the Nkima clan and was
used by Joswa Kate till his death, when it was inherited by his son, Mr. Joswa
Kainulegeya (recently deposed from that position). But, traditionally, Mugema
was also Saza Chief, Busiro, and, when Joswa Kate was deposed from that
position (but not from the headship of the clan), his successor, Tefero Kisoson-
kole,35 the Deputy Mugema, was usually referred to as Mugema and Joswa
Kate (wrongly) as ex-Mugema. In 1927 Samwiri Wamala, not a member of the
Nkima clan, was gazetted as 'Mugema, Ow'esaza of Busiro' in succession to
Kisosonkole. But objection was taken and the official title of the Saza Chief,
Busiro, was changed to Sebwana in 1932.37 Nsimbi38 is wrong in saying that
this was the first use of the title. It appears frequently in the mailo lists, some-
times attached to Mugema, sometimes standing alone, but always referring to
Joswa Kate, who, on occasions, used it as part of his signature. A provisional
hypothesis might be that, though the clan and administrative functions of the
Mugema were at that time indivisible, Mugema referred primarily to the former
and Sebwana, whose pedigree is given at length by Nsimbi,39 to the latter.

34 Tanywa dagala. The word edagala is used to mean medicine, shoe polish and a
number of other commodities.
35 Official Gazette, 1919, p. 254.
36 O.G., 1927, p. 104.
37 O.G., 1932, p. 29.
38 Nsimbi, M.B., Amannya Amaganda, Uganda Society, p. 79.
39 ibid., see index.

IN 1903 the products of Uganda were limited in quantity and value. Their
range can be seen in Fig. 1. The economy was essentially a 'natural' one, cash
playing only a peripheral role on the fringes of the economy and hardly even
that outside Buganda. The revenue of Government was inelastic and insufficient
even to pay for the basic needs of administration. A grant-in-aid from the British
Treasury paid for some 84 per cent of Government's expenditure. Ten years
later the value of Uganda's exports had increased twelvefold, the grant-in-aid,
which was shortly to disappear completely, accounted for only some 22 per cent
of expenditure and cash transactions were firmly entrenched almost throughout
the Protectorate. The prime mover in this striking change was, of course, cotton.
Professor Frankel has suggested that the discovery or propagation of a viable
export is perhaps "the most important single factor in the economic development
of backward areas".1 The economic history of Uganda in the decade before the
first Great War offers a striking demonstration of this thesis. Moreover these
fundamental changes in the structure and output of the Uganda economy are
particularly impressive when contrasted with the previous decade during which,
as I have attempted to show in another paper (The Economy of Buganda,
1893-1903. Uganda J., 20 (1956), 17), very little change occurred in economic
How are we to account for this improvement? There was throughout our
period a buoyant market for primary products, particularly for cotton, but to
explain Uganda's response to this market, we must, I suggest, look for the
initiators of economic change. Who were the people who, in Schumpeter's
classic definition of an entrepreneur, were 'getting things done' in Uganda in
1903..I think there were two active groups: Government and traders, most of
whom were Indians. I exclude planters or settlers and perhaps I should digress
at this point to explain why.
By 1902 it was already becoming apparent that Uganda was not to be a
white man's country. Thomas and Scott claim that "Johnston and his immediate
successors fostered the view that plantation development, primarily under
European management, of such products as rubber, coffee and cocoa, offered
the most promising economic future".2 But I think this is misleading. Certainly
Government did little to encourage settlement, as a cursory glance through the
files of that period reveals. For example, Jackson to the Foreign Office,
February 1901:
"Were land to be let at much lower rates than at present, I have little doubt
but that many Indians and others would settle permanently in the country ...
I am not, however, in favour of reducing the present rates in order to encourage
settlers in the Kingdom of Uganda, as we are not yet in a position (and as far as
I can see we cannot be for some two years to come) to define the boundaries of
I Quarterly Journal of Economics. May 1953.
2 Thomas and Scott. Uganda, 1935, p. 41. [But cf. p. 145 ante. ED.]

(I) Unginned cotton (2.) Coffee

FIo. 1

the estates of the Chiefs, which estates have now become private property by
the terms of the Agreement. The question of the definition of the boundaries
will have to be approached with great caution, or it will lead to widespread
disaffection; and I therefore think it would be unwise to encourage settlers
until such time as our Survey Department has had time to survey the estates,
and thereby prevent any chance of encroachment."3
In February 1902 Uganda Notes commented:
"Our opinion is that Uganda proper offers practically no advantages to
settlers of any kind, and anyone thinking that he will get special concessions
from Government will be grievously disappointed. More than one prospector
has already gone home quite satisfied that he can get no fortune here, and
several poor fellows are struggling to get a bare existence ... The object of the
Government is to do all they can to enable the natives to reap the benefit of
whatever riches their country possesses, and the latter are not merely 'sitting
under a tree and smoking a pipe', but are making a great effort to learn how
they can avail themselves of the advantages the railway offers, to get their
products into a good market."4
By 'Uganda proper', of course, is meant only Buganda, but it was Buganda
which had the communications essential for commercial production.
Admittedly Hayes Sadler's instructions on his appointment as Commissioner
in 1901 were that 'every encouragement is to be offered to capitalists and settlers',
but in practice the fundamental incentive, security of tenure, could not be
promised. Uganda Notes complained in April 1902 about the "impossibility of
obtaining land on any reasonable terms". Government, they said, had not been
anxious to attract settlers until the railway was finished; but now it was and
conditions had not changed. Government must introduce "more favourable
land laws otherwise it is not to be expected that settlers will come to an
untried country like Uganda, when land can be obtained on far more favourable
terms in known and proven countries such as Ceylon, Australia, New Zealand,
South Africa, North and South America".5
An applicant from Salisbury, Rhodesia, asked Entebbe what inducements
were offered to settlers and prospectors; he was referred to the East Africa
Protectorate Government. A similar brush off was handed to a self-styled
'energetic young man' from Transvaal. The archives at Entebbe are full of
cases like these. For the history of land policy during this period I need only
quote from an (unpublished) paper by Christopher Wrigley. By 1908 "The
authorities were in the absurd position that, though in theory they had some
8,000 square miles of Buganda at their disposal, there was practically none
which they could actually offer to planters". The alternative to securing grants
of Crown Land was to purchase it from its native owners, but for this the
consent of both the Lukiko and the Governor were necessary. The consent of
the former was difficult to obtain because "certain of the bigger and more far-
sighted chiefs were busily amassing land themselves and naturally wanted to
keep non-natives out of the market as far as possible". Outside Buganda land
3 Foreign Office Correspondence. 1902. No. 24. 14/12/01.
4 Uganda Notes. February 1902, p. 9.
5 Uganda Notes. April 1902, p. 29.

was more easily acquired though the process tended to be very slow and
inadequate communications were a considerable disincentive. For reasons such
as these the role of planters in the emerging Uganda economy was never more
than that of a small, though exceedingly articulate, minority. We can now turn
to Government and to traders.
For Government the immediate pressing need was the establishment of a
regular flow of revenue. In a letter to Bishop Tucker, refusing the C.M.S.
exemption from hut tax, Johnston complained:
"the Imperial Treasury cannot go on many years longer supporting the
Protectorate at so vast a cost to the British taxpayer." "When we have something
like a revenue the Government might be able to subsidize education."
"Pending this development you cannot render a greater service to your converts
than to teach them English and. useful industries by which they may earn
good wages ... I cannot tell you how earnestly I look forward to a fair return
from native taxation during the current year. More than anything else this will
encourage the British Treasury to keep up its heart and to go on for some time
supplying funds for the development of Uganda."6
That was in 1900. The institution of hut tax in that year did little to solve the
problem. The reason of course was lack of cash. The tax of Rs. 3 was often paid
in kind or in labour. Government then had to store a curious range of articles
and persuade merchants to buy them before it could realize useful revenue.
Mortality among livestock collected, the destruction of foodstuffs by damp and
insects, and the sudden flooding of markets with Government stocks meant
that the actual flow of cash was a barely significant trickle. Further complica-
tions in the first years arose from the demonetization of the cowrie. Millions of
shells were purchased by Government and then burned. Abundant supplies of
quicklime did little to compensate Government for this loss.7 Even more serious
was the loss of the old Eastern Province to the East Africa Protectorate in
1902. This meant a substantial loss of revenue from a densely populated area.
Then followed the sleeping sickness epidemic which "not only demanded
direct expenditure on control and research measures but resulted in the death
of tens of thousands of native tax-payers".8
For all these reasons Government was constantly aware of the need to get
cash into the people's hands so that some of it could flow back as revenue.
Moreover the desire for cash was strongly felt, particularly in Buganda, by the
mass of the people who had long experienced a desire for the products of the
industrialized world. In his Preliminary Report on the Protectorate of 1900,
Johnston had contrasted the "more civilized Bantu-speaking populations
occupying the old Empire of Uganda" where "calicoes are greatly in demand"
with other tribes who "go absolutely naked".9 In 1904 a perceptive missionary,
writing about the education of the Basoga, said "we may look for commercial
improvement owing to a desire on the part of the people to possess the same

6 Entebbe Archives. A7/6/Misc. Johnston to Bishop Tucker. 13/5/1900.
7 Foreign Office Correspondence. 14/12/01.
8 Thomas and Scott. Uganda, 1935, p. 223.
9 Foreign Office Correspondence. Enclosure in No. 101 Uganda. 27/4/1900.

comforts and conveniences of life that they see the peoples of Uganda
The arrival of the railway at Kisumu in December 1901 had accentuated the
need for exports. The numerous requirements of the people could now be
brought in at greatly reduced rates but how was their demand to be made
effective? What was the railway to carry back to the coast? As can be seen in
Fig. 1 ivory, hides and skins were by far the most important exports, but the
total value of exports in 1903 was only some 53,000 against 123,000 of
imports. Moreover ivory was a fast wasting asset. It had been exploited through-
out the pre-railway age because it was practically the only commodity which
could stand the cost of head transport to the coast.
The emergence of cotton as the overwhelmingly important export crop can
be seen in Fig. 1 which shows the changing relative importance of the principal
exports. It does not, of course, indicate the actual increase in cotton production;
for this we must turn to Fig. 2. Let us now examine, in some detail, the pattern
of this development.
First it must be stressed that there was nothing premeditated or inevitable
about the emergence of cotton as the dominating crop. Many others were tried.
To the administration the problem was simply to find something-anything-
to replace slaves and ivory, those well-matched bedmates of the primitive
economy. The search for viable exports is to some extent illustrated in Fig. 1,
but many other commodities were tried which never became prominent in the
export lists. Some aroused great hope and enthusiasm; some were cranky in the
extreme. The ramie episode is a good illustration of this. Some interest had
been aroused in this fibre, particularly after the great cotton scare which
re-awakened Lancashire's fears about her raw cotton supplies. A Mr. Edwards
Radclyffe published an article in the British Trades Review in May 1905 with
the ebullient title 'Ramie, the Textile of the Future'. Inevitably this article
found its way to Entebbe and was sent on to the Scientific and Forestry Depart-
ment with instructions to "study this paper very carefully and bring to the Com-
missioner's notice any points he should know". Mr. Edwards Radclyffe was
eloquent in his praise of the fibre's many qualities: "Wealthy Chinamen prefer
it to all other textiles." It had the lustre of silk; it was not destroyed by washing,
did not rot, was stronger than jute, flax or hemp; its staple varied from 3 in. to
18 in.! It grew where cotton would not and was much harder wearing. French
banknotes were made of it. For 5s. 6d. Mr. Edwards Radclyffe was prepared to
supply a quantity of seed and to test and report on any product free of charge.
There was a snag of course: no one had yet succeeded in producing an efficient
decorticator. But Mr. Edwards Radclyffe was hopeful if vague: "I am prepared
to produce a decorticator, when the planter has a sufficient acreage to use one ...
if any of our Colonial Governments will take up Ramie, I am prepared to offer
my services, on a no cure no pay principle."" Ramie came to nothing in Uganda
but Government's willingness to investigate it seriously is some indication of its
anxiety to promote marketable exports. Rubber was another hope of these

10 Uganda Notes. April 1904.
11 Entebbe Archives. A6/15.23/5/05.

early years. As late as 1908 Hattersley only echoed popular opinion when he
suggested that it had a greater future than cotton.12
Cotton's debut therefore resembled the tentative first appearance of an

1904-05 56 6-7 7-8 8-9 9-10 10.11 11-12 12-13 13-14 14-15 15-16 16-17 17-18
FIG. 2
Cotton exports (logarithmic scale).
unassuming ingenue rather than the tumultuous entrance of a prima donna.
The fact that cotton was merely one among several experimental crops is worth
labouring if only because it helps to explain the apparently amateurish way in
12 C. W. Hattersley. The Baganda at Home. 1908, p. 70.

which it was introduced, and the nearly fatal mistakes in policy which led to
the crisis of 1907. The plant had long been known in Uganda. It was noticed by
Speke, by Baker, by Emin and by Lugard between 1862 and 1892. But before
1903 it was of negligible importance as a commercial crop. In that year the
Government imported 1- tons of three different types of seed from the Khedivial
Agricultural Society of Egypt and distributed it "for trial cultivation by
peasants in all likely and accessible parts of the Protectorate" during April and
May 1904.'1 At about the same time Mr. Borup, on behalf of the Uganda
Company imported 21 tons of five different types of seed from the British Cotton
Growing Association, which were distributed among 27 chiefs in eight districts
of Buganda.14 Considerable interest was aroused, particularly among Baganda
chiefs, so the Government doubled its original order for seed before the results
of the first sowing were known. Three tons of seed were distributed in Septem-
ber 1904 mainly in Buganda, Bunyoro and Busoga. At the first harvest it was
clear that the American seed was by far the most successful of the Uganda
Company's mixed lot, so the Government procured another ton of this type
through the British Cotton Growing Association and distributed it in April
1905 in Buganda, Bunyoro, Busoga and Ankole.
The reason for experimenting with different types of seed was explained by
the head of the Scientific and Forestry Department in August 1904: "Like many
other countries Uganda has locations which differ considerably respecting
climatic conditions and soil. It is therefore unlikely that one particular variety
of cotton may be adapted to every part."15 It was not then realized how
disastrous this mixture of seed would prove. Meanwhile three experimental
stations, on a very modest scale, were set up, at Masaka, Kakumiro and near
Jinja, to investigate the best varieties and most suitable soil. Trials were also
made with indigenous varieties. It was already obvious that "under native
cultivation the introduced varieties would soon degenerate" and it therefore
seemed "most important that the Experimental Farms should make a careful
selection of seed from the most prolific varieties year by year so that supplies
for planting (would) be available and deterioration (would) be prevented from
becoming general".16 But unfortunately the administration underestimated the
natural tendency of cultivators to use their own seed. Further confusion arose
from the hand-gin fiasco which I shall describe below. Instructions to cultivators
were rudimentary and optimistic. In December 1903 a pamphlet was issued "as
considerable interest is now being taken by the Baganda in the cultivation and
preparation of commercial products" so that they "may be informed as to the
best way to get to work in order to get the best possible results". It contained
notes on ramie, rubber, groundnuts, chillies, simsim and cotton. For cultiva-
tion of the latter, growers were merely advised that the seed should always be
selected from plants which are bearing very heavily. "By this means only the
best strains and the most prolific varieties will be perpetuated." Arrangements
for marketing and ginning were similarly rudimentary. The Uganda Company

13 Entebbe Archives. File 759/09. 5/7/08.
14 Prospectus of The Uganda Company Ltd., 20 April 1906.
15 Entebbe Archives. File A7/9/04-06. 27/8/04.
16 Entebbe Archives. File A7/9/04-06. 27/8/04.

had installed a 'hand-power factory' in 190417 and employed agents to purchase
cotton from those chiefs to whom they had distributed seed. The Government's
unfortunate contribution was to sell 62 very primitive hand gins at 1 each to
chiefs and to give away another 43. Pamphlets explaining the use of the
machines were issued and demonstrations were arranged.18 This was a well
intentioned attempt to encourage the spread of cultivation and above all to
avoid transport difficulties. The ginning process removes about two-thirds of
the weight of cotton. It was therefore most desirable that cotton should be
ginned as near as possible to the place of cultivation. Moreover Uganda cotton
is rain-grown and rains are likely to come while picking is still taking place. In
contrast, for example, to the irrigated Sudan crop, Uganda cotton runs great
risks of staining if it is left lying around.
Unfortunately the effects of hand-gins within a few seasons were disastrous.
Unaccustomed to even the most primitive machinery and the exacting demands
of world markets, native ginners produced cotton which was practically un-
saleable. Moreover, long hauls of loosely baled lint led to bad staining before
it ever reached Kampala. Worst of all Government lost all control over the
issue of seed and hopelessly inferior and mixed seed was planted.
By 1907 there was no questioning the fact that cotton was a success. In that
season cotton accounted for over 35 per cent of total exports. But it was becom-
ing increasingly apparent that the economy was heading for a first-class crisis.
In 1907 adverse reports on the quality of Uganda cotton began to come in from
the British Cotton Growing Association. Not merely was it badly stained, but
several varieties of cotton were mixed in the same bale. For example one bale
contained Egyptian, American and a roughish Peruvian type. Baled separately
these cottons would have sold respectively at 10d. and 7d. a pound. Mixed
together they would fetch 6d. at most. Spinners would begin to distrust Uganda
cotton and might even grow to regard it with the same loathing as the notorious
'surats'.19 The British Cotton Growing Association worried the Colonial Office
and saw Hesketh Bell personally when he was in England. They also wrote to
the manager of the British East Africa Corporation, their agents, who had
started taking unginned cotton across the lake and ginning it at Kisumu. Con-
siderable pressure was put upon the Uganda Government to 'take strong steps'.
At this stage it is important that we clear our minds on the subject of quality.
This is not merely of historical interest. The quality problem is still contro-
versial as can be seen by glancing at recent reports of the Lint Marketing Board,
or a recent Legislative Council debate on coffee in which diametrically opposed
views were stated with no apparent attempt at reconciliation. There are, I sug-
gest, two fundamentally different attitudes towards quality-the technical and
the commercial. To the technician, in this case the agriculturalist, good quality
is intrinsically desirable. To the businessman 'quality' has little meaning except
in relation to costs and to the ultimate market for the product. It is true that
certain minimum standards of quality and particularly of homogeneity are
essential if world markets are to be penetrated and held. But it is arguable that
17 Prospectus of the Uganda Company Ltd.
18 Entebbe Archives. File SMP 1141/1906.21/8/07.
19 Entebbe Archives. File 1677/1907. Letter from J. A. Hutton to Major E. H. Leggett

profit-seeking buyers of produce can be relied upon to refuse 'inferior' qualities
if they are not marketable at a profit, without Government having to intervene,
In a recent discussion, mainly of West African marketing, Bauer and Yamey
have made this point with some force.20 Although I accept this argument I
think it lacks historical perspective. It does not take sufficient account of the
difficulties of establishing the reputation of products from a 'new' country
unknown to the commercial world. Once this reputation has been established,
then it is arguable that Governments should not allow themselves to become
obsessed with a search for quality standards which may be unrelated to real
market conditions and the abilities of peasant producers. The cotton policy of
the Uganda Government could perhaps be criticized along these lines but that
would take us outside the scope of this paper.
The interest and importance of the 1908 crisis is that it led to the first detailed
discussion of the quality problem in Uganda out of which emerged Govern-
ment's first drastic interference and attempt at some control over the industry.
It was clear by then that cotton was not to be a plantation crop; the uncertain
conditions of land tenure already referred to offered small scope for large scale
production under expert management. A few entrepreneurs endeavoured to get
Government's support for plantation schemes but the Uganda Company's
decision to concentrate on ginning and marketing was representative of the way
in which most foreign business men saw their role in the industry. Cotton was
therefore destined to remain a peasant crop and the Uganda Government had
for the first time to face a set of problems which have beset it ever since. Much
of the economic history of Uganda can therefore be written in terms of the
demands which successful production of a world crop makes upon the adminis-
tering authorities and the pattern of the economy's response to those demands.
That story really begins in 1908.
For most of the details of general administration the Uganda Government,
like other British colonial governments of the period, was largely dependent
upon Indian experience. Whole sections of the Indian legal code were taken over
bodily and constant reference was made to Indian precedent. It was therefore
unfortunate that in the sphere of cotton Indian experience was singularly un-
helpful. Transport conditions in India were immeasurably superior, commercial
enterprise was abundant, peasant skills were far more highly developed, and the
type of cotton grown there was frowned on in Lancashire. The Uganda Govern-
ment made some attempt to borrow from Indian experience. Details of the
Berar marketing system were studied, for example.21 But the general tendency
was to avoid Indian methods. Government had therefore to learn by trial and
error. It could borrow, to a small extent, from other parts of the Empire; exam-
ining, for example, agricultural techniques and field costs in St. Kitts.22 But with
ludicrously meagre resources it had, on the whole, to find its own way. More-
over it must be remembered that Empire cotton growing, with the exception of
India, was in its infancy. The British Cotton Growing Association was only

20 Bauer and Yamey. The Economics of Marketing Reform. Journal of Political
Economy. June 1954
21 Entebbe Archives. File 1677/1907. Record of Cotton Conference.
22 Entebbe Archives. File SMP 106/1907. Letter from H.H.B. 22/1/07.

just beginning to acquire that fund of information which was to prove so use-
ful in the future. It was still at a stage when it could give seriously misleading
advice simply because of the absence of one man from its head office, advising
the Uganda Government to obtain further supplies of seed locally and thus
prolonging the existence of mixed seed which was responsible for most of the
industry's problems.2
When we list therefore the mistakes made by Government in the initiation
of cotton growing we must realize that they did not arise entirely from insularity
and ignorance. It is true that the responsible administrators were little more
than enthusiastic amateurs but this was, after all, the heyday of the all-rounder
and unspecialized colonial servant.
At the first cotton conference called by Government in December 1907 these
mistakes were laid bare.24 It was unanimously agreed that only one type of seed
should be grown and Government agreed to establish a seed farm for this pur-
pose. This meant a complete reversal of official policy, for as late as June 1907
Government was telling native growers to "select good seed".25 But this decision
implied much more. The first Cotton Ordinance of 190826 gave the Governor
extraordinarily wide powers over the whole range of growing and marketing.
The first rules made under the Ordinance27 laid down that only seed distributed
by Government could be sown and all cotton plants were to be destroyed after
the first crop was picked. The immediate impact of this policy which was ruth-
lessly pursued in a country which had barely become accustomed to the crop
is described in a letter from the manager of the Uganda Company to his directors
in London:28
"Fortunately the Government Inspectors have not yet been able to cover all
the ground and consequently there is a great deal of cotton coming on in dis-
tricts which have not yet been visited by them. The great danger is not so much
what they actually do to the crop as what impression they make on the native
mind. In seeking to improve the cotton there is no small danger that they may
improve it altogether away the primal consideration is that the Baganda
should plant cotton and then they can be induced gradually to improve its
The effects of this policy on the following season's crop can be clearly seen
in Figs. 1 and 2. As Sir Hesketh Bell admitted: "Regulations of such a drastic
nature could hardly have been applied to any territory which did not enjoy the
somewhat unusual conditions that prevail in Uganda. The authority exercised
by the Native Government over the peasantry is so great that the bare orders
of the chiefs were expected to suffice to ensure effective obedience of the Rules
framed under the Ordinance. The Regents and Chiefs of Buganda readily
appreciated the necessity for maintaining a high standard in the quality of

23 Entebbe Archives. File SMP 106/1907.
24 Entebbe Archives. File 1677/1907.
25 Cotton and its cultivation in Uganda published in Luganda, Lunyoro, Lusoga, and
Lunyankole languages for the instruction of native growers. Issued by the Botanical and
Scientific Department, June 1907.
26 The Uganda Cotton Ordinance, 1908. No. 5 of 1908.
27 Rules. The Uganda Cotton Ordinance. 19/3/08.
28 Uganda Company Archives. Letters Out. 27/10/08 and 2/11/08.

cotton exported from the Protectorate, and quite understood the evil effect
which the mixture of varieties would have on prices. They entered loyally into
our views and agreed to act as distributing agents. They promised that all their
tenants and the peasants in general would destroy all cotton plants that were
not of the 'American Upland' variety, and see that no seed, save that which
was given out by the Government, should be planted."29
Unenthusiastic chiefs could be encouraged by the Protectorate Government-
for example in 1909, this letter from the Provincial Commissioner to the Dis-
trict Commissioner, Masaka: "I regret to hear that the chiefs in your district
show want of interest in this most important industry. There is no excuse what-
ever for Mugema, who as Saza chief must be made to understand that he is
responsible to us for the progress of his country. If he had set an example as he
should have done, instead of showing reprehensible lack of interest, his people
would most certainly have exerted themselves more as they have done in other
sazas. Please make the sub-chiefs responsible clearly understand that they will
forfeit their positions unless they show more attention to their duties and send
Mugema to me on my return to Kampala."30
The extraordinary fact is that recovery was so quick. By 1909 exports were
forging ahead again. This says a great deal for the fertility of the soil, the desire
for cash and the persuasive powers of the chiefs.
Another necessary step in Government's attempt to control seed was to recall
and destroy hand-gins. This was done in July 1908; those which had been pur-
chased were bought back at the original price. But this left unsolved the problem
of extending cotton production to outlying areas. The ultimate answer was that
suggested by the British Cotton Growing Association: "It is the duty of Govern-
ment to provide transport and we will do all that is necessary where transport
is provided and only where transport is provided."31 But this solution was not
to come for many years mainly because funds were insufficient but partly per-
haps because Government never fully appreciated the dynamic role of transport
in economic development until very recently.
Meanwhile the administration's unhappy experience with hand-gins made
them suspicious of anyone who proposed introducing simple machinery. In
December 1909 an Indian32 applied for permission to install hand-gins in
"cotton producing districts such as Mubende, Masaka and Masindi" where he
claimed he would be able to offer higher prices than were paid in Kampala and
would thus "benefit the natives as well as stimulating cotton growing". "You
are perhaps aware," he continued, "that a 'ring' at present exists at Kampala to
keep the prices of unginned cotton low and with this end in view it is shortly
to be proposed to H.E. the Governor that hand-ginning be prohibited." He
asked whether Government could assist him by reducing import duty on his gins
and by granting a small plot of land where he could install his workshop. The
29 Sir Hesketh Bell. Report on the Introduction of The Cotton Industry in Uganda.
Cd. 4910.1909.
30 Entebbe Archives. File 1635/09. Enclosure 4. 7/10/09.
31 Entebbe Archives. File 1677/1907. Letter from J. A. Hutton to Major Leggett,
32 Entebbe Archives. File SMP 1155/1908. Letter from L. R. Ramsingh. Kampala.

type of machine he proposed installing was the more substantial hand-gin,
costing about 15 in Kampala and producing about five pounds of lint per
hour. There were at that time four such gins working in Kampala where they
could easily be supervised and Government's seed regulations were observed.
But Government did not want to encourage their use elsewhere. If "small
capitalists attracted by present dazzling prices" were allowed to go into remote
areas it would lead to a resurgence of all the previous difficulties and anyway
these small capitalists must be saved from inevitable disappointment when
prices fell. Moreover it would seem unfair if Government allowed Indian
ownership of hand gins having prohibited African ownership. When the industry
became more firmly established of course, the spread of hand-gins under proper
supervision should be "in every way encouraged, particularly to minimize
difficulties of transport". So argued the newly formed Cotton Department33
anticipating many facets of future Government policy: the distrust of the small
man, the fear of bankruptcies, the putting of quality before quantity, the promise
of better things when the industry is on a "firmer basis". The Indian applicant
was politely rejected.
Another question discussed at the 1907 conference was destined to come up
again and again in the future and was to lead to the extraordinary logic
of the 1929 Commission which finally entrenched the monopolistic structure of
the industry today. I refer, of course, to the bogey of competition. In 1906 the
Uganda Company were practically the only buyers of raw cotton. But by the
following season there were two new ginneries in Kampala34 and a large number
of African and Indian buyers appeared, many acting as agents for the British
East Africa Corporation who, it will be remembered, exported unginned cotton
across the lake, a trade which flourished during our period (Fig. 2). Before the
quality scare Government would have welcomed this as a stimulus to the
industry. Earlier in 1907 it was feared that there might be insufficient demand
for the crop and growers would be discouraged. A Government officer was
instructed: "Should you see any sign of the cotton not being bought you should
act on H.E. Mr. Commissioner Bell's authority to buy it from Government's
Fund. At the present juncture we cannot allow any feeling to exist that cotton
will not be readily bought."35
Yet only a few months later Government was deploring the effect upon
quality of competing buyers: "The native could dispose of anything he could
produce irrespective of quality."36
Nothing was done about it at that stage but the discussion continued in
Government circles. A letter37 from the Cotton Department in 1909 argued
that "Uganda is not benefiting in the least by the boom in cotton the real
crux is that the quantity of cotton at present grown in Uganda does not justify
the number of ginneries at present in Kampala and the amount of European
33 Entebbe Archives. File SMP 1155/1908. Cotton Department. Report on Hand-gins
in Uganda. 9/12/09.
34 P. H. Lamb. Report on Cotton in Uganda. Agricultural Department. Kampala.
35 Entebbe Archives. File SMP 106/1907. Undated.
36 Entebbe Archives. File 1677/1907. 17/12/07.
37 Entebbe Archives. File 1635/1909. Letter from the Cotton Department. 25/11/09.

supervision thereby entailed". It goes on to hint at collusion between ginners
to keep prices down-an accusation which was made, it will be remembered,
by the Indian hand-ginner. Indeed only if this were so could the Department's
argument make sense. It then recommends that prices should be fixed by
Government. The Governor's reactions to this were very different from those
of most of his successors. A note in the margin of the letter reads: "Afraid of
formal agreement. Price might be fixed by law but I know of no precedent.
One always has to face the danger of killing healthy competition. The same
difficulties must have arisen in other places. Should we write to the B.C.G.A.
and ask if they know of fixing prices by any other means than in the open market.
I could not agree with fixing prices by the Government. It would be interfering
too much with the liberty of the subject. Competition invariably sets things
right and I see ho reason why it should not do so here."
In fact it did. To some extent there appears to be an inherent tendency
towards monopsony (monopoly of buying) in cotton ginning but I have seen no
evidence of the continuation of a price agreement among ginners beyond one
season. Entry to the industry was too easy for any pool to enjoy a monopoly
profit for long. Only Government could provide pools with absolute security
but this only happened in a later period when the cult of 'rationalization'
dominated the official mind.
In our period the fundamental problem remained that of providing sufficient
ginneries to stimulate cotton production throughout the Protectorate. In the
absence of hand gins only two answers were possible. Greatly improved trans-
port so that cotton could be carried cheaply to a few central ginneries, or the
building of ginneries in remote areas. By 1913 there were 20 ginneries in the
country but 19 of these were in a few centres in Buganda and Busoga. The
Annual Report for that year claimed that "the future expansion of the industry
depends mainly upon the solution of the transport difficulty"38 but in fact the
real stimulus was to come from the great ginning boom of the twenties. Through-
out our period many areas shared the plight of Masindi: "Cotton has been
brought in but there is no demand for it owing to the poor communications with
Kampala. The Lake Kyoga route is the only one the freight for which allows any
profit and the boat is only able to clear 21 tons per fortnightly trip. Mr. Manley
refrained from buying 15 tons on this account."39
I have left the trader until the end of this paper, but I certainly do not wish
to minimize his role. It is easy to underrate him, as he has left little documen-
tary evidence and he rarely enters Government's files except where he has
offended against official policy. As I see it, he performed two vital services
through his willingness to buy practically anything for cash and to sell goods
which acted as an incentive to peasant producers. He was usually an Indian
who demanded few home comforts and was prepared to go anywhere. This can
be interpreted as an insidious oriental lowering of standards or as intrepid
enterprise. I prefer the latter view. A missionary, for example, writing of his
arduous journey to Nimule in 1904 when Acholi was not yet fully explored,
exclaimed "I was directed by the collector to pitch my tent in the market place
38 Uganda Annual Report. (Cd. 7622/22) 1913-14, p. 11.
39 Entebbe Archives. File SMP 835/1909. Masindi Station Report. February 1909.

and was told that any assistance I wanted I could obtain from the Indian
trader! "40
Unquestionably the most outstanding trader in East Africa at this time was
Alidina Visram. "At present" wrote Jackson to the Foreign Office in 1901, "one
merchant, Alidina Visram, supplies almost all the small traders with trade
goods, etc., and they repay him by monthly instalments while trading under
his name."41 In 1902 he was prepared "to buy up as much as the natives like to
cultivate of sugar cane and simsim".42 By 1904 he had 30 branches throughout
East Africa.42 Nor were his trading activities always a merely 'adaptive response'
in Schumpeter's phrase. When in 1903 a bed of alum was discovered near Jinja
and samples were submitted to the Crown Agents it was thought that the pro-
duct could not bear the cost of freight to London and still fetch a profit. It was
Alidina Visram who sent an experimental consignment of 10 tons to London
and to Bombay to find out definitely what it would fetch. This venture was not
a success but it is typical of Visram's 'creative response' to the risks and possi-
ble profits of a backward economy.
Writing in 1893 Lugard43 had stressed East Africa's desperate need of capital
goods which "render no immediate return". To some extent Government and
to a much lesser extent private enterprise had supplied some of this long term
investment by the end of our period. But the pattern of consumers' demand
may sometimes have as significant an impact on productivity as investment.44
The first few years of cotton production may have depended primarily on the
administration's hold over the chiefs and the chiefs' power over the people.
But there can be little doubt that the eventual success of the crop and the
growth of the economy depended essentially on the peasant's desire for cash
beyond the immediate demands of taxation. It was the role of the trader to
promote and stimulate that demand.

40 Uganda Notes. February 1904.
41 Foreign Office Correspondence. 1902. No. 24, Jackson to Foreign Office. 14/1/01.
42 Entebbe Archives. File A7/9 Misc. 30/8/04.
43 Lugard. Rise of Our East African Empire. 1893, Vol. i, p. 440.
44 Cf. Kuznets in Survey of Contemporary Economics, ed. Haley. 1952, p. 180.



THE cairns that lie in the remote hills of Koki County in Buganda Kingdom
were first recorded by Mr. D. L. Baines when Collector at Masaka in 1908.
Those that are most accessible are situated near Kibanda, some twenty-five
miles from the western shores of Lake Victoria, and less than ten miles north
of the Tanganyika border. Usually resting in a saddle between two grass-covered
cols, amidst the numerous green summits of surrounding larger hills, these
heaps of earth and stone are unmistakably the work of man. They average about
4 feet in height and some 7 to 15 feet in length. They are known to occur in
groups of from two to as many as ten or more at certain places amidst the
undulating hill tops of the County.
Five groups of these mounds or cairns are known to have been identified or
reported. These are situated high in the hills, which have an average altitude
of 4,700 feet reaching to heights of just over 5,000 feet. The two groups nearest
to one another-Tegabisunsa and Kiswere-are about seven miles apart. The
remotest within the general area, is that lying 25 miles north-west of the Tega-
bisunsa mounds, near the Gombolola Headquarters of Sabadu at Lwamagwa.
There are possibly more cairns elsewhere, particularly further south in the still
unpopulated folds of the rolling hills which extend into Tanganyika and Ankole.
Six mounds are in fact spoken of by local peasants as occurring in the hills of
Budebutagya just across the Tanganyika border from Kiswere. In each group
the mounds are strung out along the centre of a small saddle connecting two
minor hills.
The object that the builder or builders had in mind and the purposes that
the mounds were meant to fulfil are hard to determine. Local tradition is not
altogether helpful. At least, it can be established that the cairns mean little to
those Bakoki families which have been in that area for a century and a half.
The history of this one-time minor kingdom is comparatively recent. It goes
back to the first half of the 18th century, when Bwohe, a Munyoro prince,
set up his authority over the thin populated area which bordered with the lands
of the people of Kiziba and of Ankole1 in the south and south-west and with
the kingdom of Bunyoro-Kitara in the north. The cairns have no place in the
known oral accounts of the nine rulers of Koki up to 1896, when the kingdom
was merged with Buganda. Yet most elders believe that the cairns have their
origin in the days before the advent of the Munyoro prince. The mounds are,
in fact, traditionally linked with a name, that of Sekeija, meaning in Lukoki,
'a person who has come' or 'stranger'. Perhaps through oral tradition over the
centuries or through mere observation that more work than usual must have
been required in the erection of these mounds, it is said that Sekeija caused
much hardship to those who laboured for him on this work. There is extant an
1 At that time known as Nkore and comprising the eastern part only of the present
Ankole district.

expression used by the older men if there is reason to complain to one of higher
authority about a set task or job of work-"do not make me work as Sekeija
made his people work".
It has been suggested that this person might be identified with that other
stranger, Mugenyi, the Muchwezi, traditionally identified with the Katonga
earthworks. I can find nothing to substantiate this. Local oral sources are
adamant that these persons were not one and the same. Amongst the elders of
Koki of today, Sekeija is not associated with the Bachwezi. The general belief
is that some such person came to the country from the north at a time during
the period of the Babito dynasty of Bunyoro-Kitara. After a short sojourn he
disappeared in the direction of modern Ankole. His 'name' has been left to the
numerous mounds which are generally known as 'Amabale Gasekeija'-the
stones of Sekeija.
Another local belief is that the cairns were erected as landmarks centuries
ago by an unknown people who, having come from the north, moved southwards
across the present borders of Tanganyika. (Baines 1908.)
With little light thrown on their origin, the first query is, whether or not these
mounds contain or conceal anything? Their purpose might well be divulged
if excavated, especially if they should turn out to be graves or barrows. Some
such work has already been done. The records available are scant, and of the
few objects said to have been unearthed none is now traceable. Despite these
sporadic diggings it has not been possible for deductions to be made as to the
purpose or dating of the mounds.
In order to obtain the fullest picture it is worth noting what the early investi-
gators found.
(i) KISWERE GROUP. Gombolola Musale (Kiganda), Muluka Sabadu, near to
Kiswere village. Lat. 0 56' 00" S. Long. 31 22' 30" E.
There are 11 mounds situated along the length of a pass known as 'Mukakiro
Kenturege' (Zebra Pass).2 The seventh mound from the southern end was
opened up during or before 1908. Nothing was found in it. Rock was met with
at four feet. (Baines 1908.) According to locals at some date during the First
World War a part or whole excavation of one of these cairns-possibly No. 3-
uncovered a hoe and a small long-necked pot (nsumbi). The present where-
abouts of these articles are unknown.
(ii) NSESE MOUND. Gombolola Sabawali (Byakabanda), at Nsese, near to
Nabunga Hill.3 (Unidentified.)
It is said that there is only one mound to be seen. In the White Fathers'
Mission paper Munno of May 1924, mention is made of cairns at Kiswere as
well as some on Nabunga Hill. Reference is made to an attempt to dig into
three mounds in 1913, but it is not stated at which location. I have been unable
to trace any mounds myself on or near Nabunga Hill, although the term
'Amabale Gasekeija' is well known locally. The excavators, it is reported, found
many 'things', in particular a large stone which was hollow in the centre and

2 The porters' road to German East Africa passed this way, alongside the mounds.
(Baines 1908.)
3 Masaka District Map 1:250.000, A 950 (1944).

had evidently been worked by man. Unfortunately it has not been possible to
trace any member of the party concerned, nor the stone.
(iii) TEGABISUNSA GROUP. Gombolola Musale (Kibanda), Muluka Sabagabo,
near to Tegabisunsa Hill. Lat. 0 53' 45" S. Long. 31 20' 00" E.4
There are seven mounds situated in a small pass. A lone mound is to be
found near the next hill about a quarter of a mile to the north. Of the seven
mounds in the group, Nos. 3, 5, 6 and 7 from the southern end, have been
disturbed at some time or other. The last three were excavated by Comdr. O. R.
Sitwell (District Commissioner, Masaka) and Mr. E. L. Walsh (Survey Officer)
in 1928. Nothing of note was found. However, it has been recorded that in one
cairn, No. 6, 'below ground level, three narrow flat pieces of stone were found,
set vertically in the soil close to each other and forming the arc of a circle'.5 I
am informed by Comdr. Sitwell that these stones were about 6 inches high by
11 inches wide and perhaps I to I inch thick.6 According to a reliable local
informant the third cairn from the south was opened up in 1916. There are no
records available of this excavation. Some local inhabitants assert that what
was to them an unidentifiable object was found, namely a thin piece of iron,
an object which was neither the haft of a spear nor a needle.
(iv) KYAKABAJU MOUND. Gombolola Sabadu (Lwamagwa), on Kyakabaju
Hill. Lat. 00 33' 30" S. Long. 310 14' 00" E.
There is only one mound to be seen, on the side of a track which leads to
Lake Kachira. I have been informed that originally there were two other similar
mounds here-one on either side of the surviving mound-but that both were
levelled to the ground when the track was widened in 1953. No one can recall
having noticed anything in the earth and stones that were removed.7
(v) Cairns have been reported in Gombolola Musale in the Muluka of
Mumyuka. I have been unable to trace any, nor have I been able to find any
written records relating to them. There is no reason to believe, however, that
they do not exist.8
With so much attention having been paid to the cairns in the past it is disap-
pointing that, of the articles said to have been recovered from excavations,
none received expert examination; also, as has already been mentioned, none
appears to have found its way to any place of safekeeping with the result that
none is available today.
In view of this lack of physical data it was proposed in 1956 to open one of
the mounds. After careful consideration had been given to all known groups
of cairns it was decided, mainly because of accessibility, to excavate at the site
in the vicinity of Kiswere village.9 Results were relatively negative. The excava-
tion may be summarized thus:

4 Original mailo sheet South A 36 B III 77. Scale 1:10,000.
5 Masaka District Tour Book, 1928-29. Entry dated June 1929.
6 Sitwell, O. R. (1956). Personal communication.
7 Local information (1956).
8 Local information (1956).
9 This is probably the 'Kisuere' mentioned by Speke, where he camped when en route
to the Pokino's headquarters at Masaka and thence to Kabaka Mutesa's court at Banda
near Kampala. (Map of Eastern Equatorial Africa by J. H. Speke (1863), accompanying
Journal of the Discovery of the Source of the Nile.)

The group consists of eleven cairns strung out from the top of a saddle which
carries a track leading southwards from Kiswere village towards Tanganyika.
The distance from the first to the last cairn-the difference in level being about
forty feet-is some 460 feet. The general line of the cairns is in a NNE-SSW
direction and roughly follows the path. Two round cairns are distinctly offset
to the west of the line. Of the others two, which are large symmetrical oblong
mounds, lie at right angles to the pass, whereas five of the remainder have been
constructed so as to follow the direction of the pass. The remaining two are
double the size of the others, each being composed of two distinct but con-
joined mounds. (Fig. 1.) Termites have been busy building on some of the
cairns, in one case almost obscuring the heap. Small trees have grown from
the centres of two cairns, namely Nos. 2 and 11.
The cairn chosen for excavation was No. 8a (numbering from the south end).
This heap formed the southern portion of one of the two conjoined mounds,
being approximately 14 feet long, 11 feet wide and 4 feet high. The surface,
slightly overgrown, consisted of small stones-sandstone and quartzite. The
structure of the mound was formed of sandstone slabs of varying sizes bound
together with a sandy soil. From 20 to 42 inches large sandstone slabs (the
largest 24 x 12 x 2 inches were encountered. Two of these, at the south-west
quadrant, were standing vertically at the perimeter of the mound. At 32 inches,
at a point perpendicularly below the apex of the cairn, a flat almost triangular
slab of sandstone (17 x 9 x 1 inches) was uncovered. This rested horizontally.
From 38 to 48 inches smaller stones were found. Excavation ceased at 55 inches,
where a sandstone outcrop from the hillside runs at this depth into and below
the cairn.
There were no artifacts. No borrow pit was found. What appeared to be a
shallow ditch was seen to run along the south side of the line of cairns but this
may well have been a natural formation. It was not excavated. According to
the opinions of peasants the stones used in the formation of the cairns 'were not
born there'. That may be so, but the rock from the exterior and interior of the
cairn corresponds to the outcrop on the hillside, identified by Mr. Du Bois of
the Geological Survey, as a shaley facies of the Karagwe-Ankolean system.
Mrs. Vowles of the Uganda Museum has observed that the two vertical slabs
mentioned above do not appear to have formed a revetment to the mound.
They might perhaps explain Comdr. Sitwell's description of 'three narrow flat
pieces of stone found at cairn No. 6 of the Tegabisunsa Group', though those
were much smaller and were found below ground level. The position of the
triangular slab, found dead centre at 32 inches, suggested that it might have
been deliberately placed in that position. Of course, it is equally possible that
the slab might have been thrown into that position.
Though these mounds have puzzled many since the beginning of the century
we have, it is true, made little progress in probing their true significance. That
some of the Uganda cairns have proved to be barren whilst others have yielded
odd articles is not without precedent. In the Northern Frontier Province of
Kenya at Wajir no contents were found in cairns after many had been broken
up (Watson 1927), whereas at the same place in subsequent years other investi-



50 4 3 Z I 0 50

Tree 11 (

10 /

100 FEET / .

Intact 8b (I '?

Jew excavation 8a

Excavated 7 0 I


'I (

b /i
//6a /
/) /



Disturbed 3

Tree in Cairn 2

l 4 Ant Hill Cairn I




FIG. 1



gators of mounds have found human bones together with pottery and some
copper. (Parkinson 1935.)
Wayland (1934) has already pointed out that the importance of cairns may
easily be overrated, for heaps of stones do occur throughout Africa, generally
connected in one way or another with ancestor worship. Cairns are plentiful in
southern Ethiopia where they are as high as ten feet or more, larger than the
Galla type found in Somaliland. (Azais et Chambard 1932.) In East Africa they
are numerous enough to be frequently referred to as 'Travellers' Cairns'.
(Gillman 1944.) In Kenya they are found in certain parts of the Northern
Frontier Province, in the country of the Suk in the Rift Valley Province, and in
the lands of the Masai. In the two latter areas stones are known to be heaped
over shallow graves. In Suk, cairns are believed by the inhabitants to mark
spots where lightning or meteorites have fallen. (Huntingford 1953.) Stone
mounds also occur in northern and southern Tanganyika. (Baines 1908, Gillman
1945.) Grant (1864) remarked when passing through Usui in the Uthungu
Valley in south Karagwe, that sometimes he observed by the pathway cairns
of stones, 'such as are found all over the world'.10 His leading porters 'generally
threw their mite on the heap'.
In Uganda cairns have been noted above Kibiro on Lake Albert in Bunyoro,
also in the north-east of the Protectorate, and in Buganda and in Ankole.
(Wayland 1920/1931/1934.) In Kyagwe in Buganda on a ridge called Dwenyoni
Hill, south-west of Grant Bay, Wayland has recorded fifteen cairn-like struc-
tures, also others offset to one side or the other. One cairn having a diameter
of 23 feet and a height of four feet was opened up. Had it yielded human remains
there would have been a strong suggestion that the cairn was associated with
some burial rite; but none were found. Wayland has observed that these par-
ticular cairns have no place in native tradition; nor is their purpose known to
the local inhabitants. It would therefore seem that they are of some antiquity.1
Certainly in Koki some legend has emerged associated with the origin of
those cairns situated in that region. Could the Koki cairns have some signifi-
cance of their own? The known groups extend from east to west over some
thirty miles. No evidence has yet come to light that any of the mounds are
graves. Owing to their position in narrow passes high up in the hills, it cannot
be suggested that any of the recorded cairns have resulted from clearance of the
land for cultivation. Had this been the case, many more cairns could be expected
to have survived amongst the more gentle undulations of the Koki hills. Be-
cause of their position in groups it is unlikely that any of these cairns represent
landmarks. If they were landmarks, they would probably have been sited so
as to blaze a trail, whereas the known Koki groups point not only in varying
directions, but are, if anything parallel to one another. Their construction does
not represent an outstandingly arduous task, particularly if done collectively.
There can be no comparison, for instance, with the work entailed in the digging
of the great earthworks further north (Lanning 1953). But if many more mounds
originally existed throughout the hills then numbers alone might suggest a
task of greater magnitude.
10 This would be Biharamulo, in the Lake Province of Tanganyika Territory.
11 Wayland, E. J. (1956). Personal communication.

A possible explanation of their origin and purpose can be found in a custom
attendant on certain heaps of stone which occur in Karamoja. In Labwor
(Wayland 1931) small cairns, composed of stones or even twigs and grass, are
situated in passes. These mounds are said to be treated with the greatest respect
by locals. Occasionally a cairn may be dedicated to some departed hero and it
is the practice of the local inhabitants to place a stone, a twig or some grass on
such a heap on the occasion of departing on some special journey, so as to
ensure good fortune.12
It is possible that on a well-used route, as a result of the frequent attention
of passers-by, time would see the steady rise of a mound, the subsequent com-
mencement of another mound as the first reached an unpractical height, and
then another and another. A group like any of these in Koki, could result from
their being raised one by one along the path or way following the floor of a
pass. Different articles, such as a pot, an unwanted piece of iron, or an odd-
shaped stone might easily have been thrown on the heap instead of the more
usually handy slab of stone. This argument could be applied particularly to
those mounds at Kiswere. That group evidently borders a route that has been
in use for many generations. The presence of the articles recovered from
different cairns by earlier investigators could be explained in this way.
Nevertheless, strong as this argument might at first appear, it is barely sub-
stantiated, inasmuch as, with the exception of those mounds which appear to
have been dug into during the past fifty years, there is no great variation in the
dimensions of the cairns to be seen. There is no one cairn which could obviously
have been the last to have been started or in use. At Kiswere the difference of
position of the mounds themselves is striking; two lie in the line at right angles
to the pass, another two are offset, standing away from the line of direction of
the others, whilst another two are double cairns, the only two of their kind.
Finally, and perhaps most significant of all there is no local knowledge of any
cult wherein these cairns have ever received the physical attention of people
in the same manner as has been recorded elsewhere, particularly amongst the
Karamojong of Uganda. However, in Koki the custom might have fallen into
disuse centuries ago, well beyond the reach of living memory. As it is, we are
left with the unsatisfactory notion that the true purpose of these cairns is
still unexplained.
Whether the Koki cairns have a different meaning from those elsewhere in
the Protectorate or in other parts of Africa is still open to question. In all
probability the solution will be forthcoming only when extensive and systematic
excavations have been undertaken at each site.

Azais et Chambard (1932). Cinq anndes de recherches archdologiques en Ethiopie.
Baines, D. L. (1908). The Uganda Official Gazette. 1 November 1908.
12 A similar custom prevails amongst the Masai. Anybody passing by will throw a
stone on to the cairn covering the grave of an important person. (Huntingford 1953.)

Gillman, C. (1944). An Annotated List of Ancient and Modern Indigenous Stone
Structures in Eastern Africa. Tanganyika N. & R.
No. 17, 54.
(1945). Supplementary Note on Stone Structures in East Africa.
Tanganyika N. &R. No. 19, 66.
Grant, J. A. (1864). A Walk Across Africa, p. 133.
Huntingford, G. W. B. (1953). The Southern Nilo-Hamites. pp. 90, 120. International
African Institute.
Lanning, E. C. (1953). Ancient Earthworks in Western Uganda. Uganda J., 17,
Parkinson, J. (1935). The Stone Cairns of Northern Kenya. Man, 163.
Speke, J. H. (1863). Journal of the Discovery of the Source of the Nile. p. 267.
Watson, C. B. G. (1927). Wells, Cairns and Rainpools in Kenya Colony. Man, 30.
Wayland, E. J. (1920). Some Facts and Theories relating to the Geology of
Uganda. Geo. Surv. Uganda. Pamphlet No. 1. p. 43.
(1931). Preliminary Studies of the Tribes of Karamoja. J.R.A.I.,
61, 207, 208, 221.
(1934-5). Notes on the Biggo bya Mugenyi; Stone Cairns and Pit
Dwellings and other Occupation Sites. Uganda J.,
2, 27.


As told to J. G. HUDDLE
AS far as possible this life story has been set down in the idiom, order and
emphasis with which Yakobo Adoko told it to me in the course of two
long evening meetings. I was helped by Yakobo's son, Mr. Benjamin Benefit
Kindly Odur, who acted as intermediary and interpreter.
Yakobo Adoko succeeded his father in 1916 as a sub-county chief and rose
in the course of forty years of service to the highest position in the Local
Government of Lango District. For thirty-five years as a Divisional (Sub-
County) or County Chief he exercised an authority similar in some respects to
that of his father before him, but bent to the new tasks of the British administra-
tion with the ever increasing opportunities and resources available to him.
Throughout this period, his work earned repeated tributes from a long line of
District Commissioners who recognized his vigour, authority and efficiency,
although his strong character and suspected involvement in internal jostlings
for power sometimes caused anxiety.
The climax of Yakobo Adoko's career as senior official (Rwot Adwong) of
the Lango African Local Government was not attained without difficulty. His
election in 1945 by the Lango Native Council (as the District Council was then
called) to the post of Secretary-General led to a clash between his supporters,
who wanted to elevate his position to one analogous to that of a ruler in one of
the Agreement districts, and the Protectorate Government, which firmly set
its face against this proposal, considering it repugnant to the tradition of the
Lango tribe and to the path of sound development of local government, along
which the people of Lango had already started to progress.
In 1949, however, when in addition to his duties as County Chief, Yakobo
Adoko was elected Chairman of the Council, the Won Nyaci agitation (named
after the title which it was sought to confer upon him) had died out. It says
much for the good sense both of the Government officials concerned and of
Yakobo Adoko himself that this critical period was followed by one of con-
fident and fruitful co-operation between them. J.G.H.

I, Yakobo Adoko, was born at Akokoro at the time when the Kabaka Mwanga
passed through there from Bunyoro on his way to Kangai in Lango, where he
met Kabarega. Soon afterwards they were both captured by the British Govern-
My father, Akaki, was a chief before the British Government came to Lango.
His title was Rwot and the area over which he ruled stretched from Akokoro
to Cawente, and from Mayugi to Apac. After the British Government came to
Lango District my father carried on as a chief, but simply as Jago of Akokoro,

1 In 1897, Kabarega, the Mukama of Bunyoro, and Mwanga, Kabaka of Buganda,
took refuge in Lango. They were captured in 1899.

FIG. 1

Yakobo Adoko, Rwot Adwong of Lango, 1956.

[face p. i84

and not as Rwot of the area he used to rule. The British moved the Rwot's
headquarters from Akokoro to Ibuje, and installed another person as the Rwot.
My father was a fierce man, and did not find it easy to get on with the British
A chief's work before the British came was much like that of the work of a
chief today. He would listen to the complaints of the people and try to stop
troubles. A chief had a way of influencing the selection of his successor. He saw
to it that the son whom he had chosen as his heir was shown at the right times
to his people. When fighting broke out nearby, he would send this son with a
spear to warn the people to be on their guard, and when important guests came
to his compound he would charge this son with the arrangements for their well-
being. In this way the people would know who it was that should succeed the
My father, Akaki, was the head of the Jo Oyima clan. It is likely that I shall
now become the head of this clan; a meeting will have to be called to decide
this. Many years ago, before I was born, tle Jo Oyima clan lived in Amaic.
Most of them left later for Maruzi. Nowadays the Jo Oyima live all over the
District in groups of fifty to a hundred, though the majority of them are still in
Maruzi. In those days the Lango people lived together in clans. Their young
people would marry with members of neighboring clans. When I was young,
the Langi did not travel on long journeys, and if they had to pass through strange
areas they did so at night.
Many Rwodi ruled over various groups of the Lango tribe in my father's day
and there was much fighting between the people of the different chiefs. All the
tribe, however, called themselves the Lango, although the people who lived in
different areas were distinguished by classifying names such as Langi me tung
too for the people of Maruzi County, and Jo Kide2 for the people of Moroto.
I can remember well the way we lived when I was a child. There were many
big differences from present day life in Lango District. The homesteads used
to be very far apart. In compounds like that of my father there were special
huts for the askaris who guarded the home. The houses had very low narrow
doors which you entered lying down. Young boys would sleep four in a hut,
the youths two in a hut, and a man about to marry would be alone. All the
young unmarried girls in a homestead would sleep in the same hut for protection.
Goats were better looked after in those days than they are now, and were
taken for grazing in the swamps. The fields of simsim and finger millet were
cultivated above the swamps. There was much more meat for the Langi to
eat in those days than there is now. There was an abundance of all kinds of
game for the young men to hunt, antelope, bushbuck, kongoni, warthog, buffalo,
leopard, rhino, lion and elephant. The elephant was hunted by ring-firing.
Elephant meat was highly prized, and the ivory was sold to Arabs and to traders
from Bunyoro. The young Lango men of those days looked forward more than
anything else to hunting. Their hunts provided occasions for dancing, with
drumming and singing. Another pleasure of hunting was the honey, which was
so plentiful in the bush.

2 These names merely mean 'Langi of the western part' and 'people of the east'.

I remember too about marriage. It was always a matter which had to be dis-
cussed between the young man and the girl beforehand. When the girl had
agreed, the bridegroom with his friends would visit her house and drag her
away, after a mock struggle. Sometimes indeed, this was done without the
girl's agreement. There were very few cases indeed of adultery or fornication
in the days of my childhood. The penalty for adultery with another man's wife
was forty goats, but as often as not the aggrieved husband would visit the home-
stead of the culprit and destroy his granaries and other possessions, and some-
times kill him. Sexual intercourse with unmarried girls was not regarded so
seriously, but my father was strongly opposed to it, and none of his daughters
was ever violated.
Theft was the crime which was held to be the next most serious. Thieves
came from a long way away; they were never people of our own clan. If a man
was caught stealing cattle he was killed. I remember seeing a man called Asigwa
killed for stealing a goat. In small matters the punishment was a beating with a
stick. t
There were a few native doctors, mostly women, who were trusted by the
people. Sometimes a man would become ill after taking medicine from them,
and he would complain to a chief, who would call together the people to discuss
and settle the matter.
The first District Commissioner in Lango District lived at Ibuje. There were
K.A.R. and Police in his compound. Later, another District Commissioner
from Bululu went to Nabieso, and in 1914 to Lira, the present District head-
quarters, and thereafter Ibuje was without a District Commissioner.3 There
was no fighting here in Lango when the British took over. Baganda chiefs came
with them and were put in charge of gombololas to work alongside Lango
chiefs. There was often quarrelling between these Baganda chiefs and us Langi,
but during the time of Mr. Driberg, who helped the Langi so much, the number
of Baganda started to decrease.
I learnt reading and writing from the Government clerks at the first Boma in
Ibuje, and kept at my studies there until the mission started teaching people
at the C.M.S. church in Ibuje in 1914. Mr. Lees was the European missionary
who used to come to Ibuje from Gulu. After him, the C.M.S. missionary whom
I remember most clearly was Daudi Vitatuli, a Munyoro. He lived at Ibuje in
the early days, but was later transferred to work as a chief in Lango District.
He converted many people, and taught them to read and write.
In 1911 I became an askari under Daudi Odora. Odora wanted to become the
Kabaka of Lango, and claimed authority over all the tribe. He moved from
place to place, but was finally left in charge of the present counties of Kyoga
and Kwania, following trouble between him and the Baganda chiefs. There
were 46 askaris, and we had 15 guns between us and our spears. Part of our
work used to be to guide Government officers on tour and to guard their camps.
In 1916 I left the work of an askari and was appointed by the Protectorate

3 The first Government station for Lango was sited at Bululu (now in Teso District) in
1907. A second station was established at Ibuje in 1909. In 1911 the district administra-
tion was united in one headquarters at Nabieso, which was moved to Lira, the present
district headquarters, in 1914.

Government to take over as Jago of Akokoro from my father, who was becom-
ing blind. It was my father who proposed me for this appointment. I remained
Jago of Akokoro until 1930. This was at the time when Mr. Driberg was
District Commissioner of Lango.4 He was a fine man. I remember well the talks
that I used to have with him. He would come on long foot-safaris holding spears
like one of us, and he could throw them as well as any Lango.
Akokoro in those days was a very bad area. Many people were killed and the
corpses used to be hidden in big holes. I had to try very hard to put a stop to
this. A big difficulty for chiefs during the First Great War was the demands
made upon them for drafts of young men to join the K.A.R. In fact, they had
to compel men to join, for there was trouble if the required number was not
When the cultivation of cotton was introduced, the people at first disliked
it, for they found the work too difficult. We chiefs had to force them to go
ahead with it, or else we would get dismissed. The building of tracks was
another big work for the chiefs. The first motor vehicle in the District was the
motor cycle of Mr. Tomblings, a District Commissioner, in 1921.5 After that,
many motor vehicles followed.
From 1930 until 1951, I was County Chief of three counties in succession-
Maruzi, Moroto and Oyam. My biggest work as a Rwot was to get the people
to co-operate together and to be friendly to one another. The different chiefs
had to learn to meet each other in peace. The progress of sports in Lango has
been very important in this respect. The Langi are now known as the best
athletes in the whole of Uganda, and their prowess in this field has had a long
history. The first district sports in Lango were held in 1927 under an Assistant
District Commissioner, Mr. Mitchell, whom we nicknamed Akakara, by which
we really meant 'the man who judged cases quickly without hearing witnesses'.6
Of all the counties in which I served, I think I liked Moroto County best.
Things went more easily there. The people were wealthier than in the other
counties, and paid their debts more quickly.
In the early 1930's there was a District Commissioner who caused the people
to bring many cases against us chiefs for 'eating' their livestock, beating them
and other alleged offences. This made things very difficult for the chiefs.7 In
1935, however, Mr. Rogers came as District Commissioner,8 with Mr. Cox (who
became District Commissioner himself in later years) as his Assistant District
Commissioner.9 This was a good period for Lango. Mr. Rogers was very fair
with all the people. It was he too who started the chiefs on building their houses
with corrugated iron roofs. Before this there were no roofs in Lango that were
not made with grass. Mr. Cox was most energetic. He was clever and liked only

4 J. H. Driberg, Assistant District Commissioner and later District Commissioner,
Lango, 1912 to 1918.
5 D. G. Tomblings, District Commissioner, Lango, 1920 to 1922.
6 G. K. Mitchell, Assistant District Commissioner, Lango, 1925 to 1927.
7 J. E. T. Philipps, District Commissioner, Lango, 1933 to 1934. (See Uganda J., 19
(1955), 166-8.)
8 F. H. Rogers, District Commissioner, Lango, 1935 to 1940.
9 T. R. F. Cox, Assistant District Commissioner, Lango, 1933 to 1935. District Com-
missioner, 1940 to 1944.

clever people to work under him. Mr. Cox first introduced dams into Lango.
Now of course, there are many dams and everywhere requests for new ones.
Whilst at first the people agreed, later they raised strong objections to these
dams. People were afraid of them because of the rule against cutting trees near
to dams, but this rule was later cancelled, and the opposition has now cleared
When I was County Chief of Oyam there was often trouble with the Acholi
tribe regarding the inter-district boundary. The present boundaries of Acholi
District marching with the Jaber division of Oyam County contain an area over
40 miles in depth which was formerly regarded as Lango hunting country. Many
Langi had to move from the rivers in the area now in Acholi to land near
Minakulu and Kamdini.
In 1938 the first council for all Lango District was started in Lira. It had a
membership of ten Rwodi (county chiefs), ten clan representatives and ten
Awobi ('young men', i.e. unofficial members, but chosen in those early days by
the chiefs). This council did not meet very frequently. When it did it discussed
little but marriage customs and passed laws to restrict the amount of dowry
which could be paid. The council kept on passing laws like this, but people
never obeyed them. It was not until after the Second World War that the lower
councils throughout the District were established. These have spoilt the work
and life of the chiefs. In the early 1950s too the Uganda National Congress first
became active in Lango District.
It was in 1945 that the Lango Native Council chose me to be Won Nyaci.
This means something like Kyabazinga.10 Literally Won Nyaci means 'owner
of all'. The Government, however, did not agree to this title, and decided that
the post should be called Rwot Adwong (i.e. Senior Chief). I think this is mis-
leading, however, as the Rwodi of the various counties could each be called
Rwot Adwong. However, this was Government's decision. The arguments
between us and the Government ended, and in 1949 I was elected to be Chair-
man of the Lango District Council, an office which I held at the same time as
the office of County Chief of Oyam. There was not a great deal of work for the
Council Chairman at that time, mainly because the Uganda National Congress
had not yet emerged in the District. In 1951 I was elected Angol Kop (i.e. Chief
Native Court Judge). I was the first Angol Kop. After being responsible for
only one county, I found my duties covering the whole District to be very heavy.
The Langi love to take many cases. They do not like a chief to settle a dispute
on the spot; they prefer the case to be heard through as many courts as possible
before they are really satisfied. Another difficulty of a judge's work is the
untruthfulness of the people in court. Long ago the Langi told the truth, but
now many tell lies, though you can see this when they are speaking. Even so, I
found the work of judging to be straight forward.
After one and a half years as Angol Kop, I was elected in July, 1952, to be
the second Rwot Adwong. The period of more than three years which I spent
in this work I regard as the most important of my Government career. I had to
10 i.e. the Kyabazinga of Busoga District. The title Won Nyaci is comparable with the
Kyabazinga and not with the Kabaka of Buganda, as the line of Kabakas must follow
the same clan, whereas the choice for a Kyabazinga is unrestricted.

spend a great deal of time on safari in urging the people to follow the right
methods of agriculture, in persuading them to leave their tribal doctors for
treatment at Government dispensaries, and in considering the resolutions passed
by their councils. There was also much work too for the Rwot Adwong to do in
Lira. There were far too many papers to read, many meetings and visitors to
attend to. Hospitality for councillors and committee members was also a
burden. The actual control of business at council meetings, however, was not
really difficult. These meetings did take up a long time, and every question and
speech had to be listened to at length, but if you did this, affairs could be
managed as you wanted. The work of committees was more difficult for a
chairman; but generally if a matter was thoroughly explained, agreement could
be reached.
My time as Rwot Adwong saw many big changes in Lango. There was great
progress in education, with expansion of junior secondary schools in the Dis-
trict, and the prospect offered of Lango obtaining its own senior secondary
school. Education tax and graduated Local Government tax were instituted.
Strong new bridges and dispensaries were built. The first Lango students
departed to study overseas. Government agreed to the restoration of many
divisions which had been closed down. The chiefs and senior staff of the Local
Government now possessed cars with which to go about their duties. Better
control of the cotton buying centres reduced the amount of cheating of African
sellers during the buying season. Gramophone records were made in England
of many old Lango songs and dances, and were placed on sale to Africans in
the District.
In 1954 when Her Majesty the Queen came to Uganda I went with other
senior Lango chiefs to see her. She conferred upon me the Chiefs' Medal in
Silver Gilt, and I was one of the few whom she personally invested. I remember
when she had shaken hands with me she said 'I congratulate you', and I
answered, thanking her in my own tongue. We had seen her face often in
pictures of her Coronation, and had heard her voice too on the wireless, so that
when she stepped from her plane at Entebbe we recognized her at once.
My term of office as Rwot Adwong came to an end in February 1956.
Though I am now quite settled in private life, I was not at first pleased to be
away from Government work. I asked whether I could be given other work for
three more years, but there was nothing suitable which could be offered to me.
I know that my ignorance of English is a disadvantage here. Now however, I
am free from official duties and content. I am at present preparing a house at
Akokoro, my real home where I was born. My old house was burnt down
many years ago. Actually, I have three homes, with one wife at Akokoro, one
wife near Lira, and a third wife at Iceme.
In conclusion, if I were to describe the biggest differences between Lango
District when I started work in it, and Lango District today, I would give four
things. Firstly I would say that whereas then there was much truth, now there is
little. Secondly, those councils which the Government has set up, bring great
trouble to the chiefs. Unofficial councillors want to get the work into their own
hands, and they have no respect for anyone. A man must be very clever now to
be a chief. Thirdly, there is the great spread of education. So many people now

read and write. Lastly, in those days the Langi used to drink very little beer in
order to keep themselves ready for fighting and hunting, but now they drink
very much.

Body and head covered by scales. Mouth protractile and directed obliquely
upwards. Lateral line absent. Dorsal, anal and pelvic fins without spines. No
circum-oral barbels.
The Cyprinodontidae are a family of small fishes, often brightly coloured and
greatly favoured by aquarists. The African members of the family are oviparous
and carnivorous.
The identification of cyprinodonts is not simple, and, as in many other
families of fishes occurring in Uganda, there is still much to be learned of their
ecology and taxonomy.

First dorsal fin ray situated above or very slightly in advance of the first
anal ray Nothobranchius
First dorsal fin ray situated well behind the first anal ray 1
1 Body slender, not deepening markedly in the abdominal region
(depth of body contained 31-5- times in the standard length)
Body deep (depth contained less than 31 times in the standard length),
abdomen compressed (at least in males) A
A Teeth in several irregular rows, those of the outer rows in both jaws
enlarged and more widely spaced than the teeth of the inner series.
Anal fin with 19-21 rays Hypsopanchax
Teeth in several irregular rows, but some teeth in the outer rows of
each jaw greatly enlarged and canine-like. Anal fin with 15 or 16 rays
Nothobranchius Peters 1868
Origin of the dorsal fin in the same vertical line as the origin of the anal fin,
or slightly anterior to it.
This genus is widely distributed in Africa; its range extends from Somaliland
to Natal and from the Great Lakes to Northern Nigeria, with species in the
Seychelles and Zanzibar. One species occurs in Uganda.
Nothobranchius taeniopygus (Hilgendorf) 1888. (Fig. 51)
Native name: Unknown.
Description: Small fishes not exceeding 5 cm. in length. Depth of body
contained 31-33 times in the standard length. Body 'chubby'. Upper surface
of head curved from between the eyes posteriorly, but somewhat flattened

anteriorly. Snout short and broad. Mouth directed upwards; the space between
the eye and the lip narrow. Dorsal fin with 16 or 17 rays, its origin slightly in
advance of the first anal ray. Anal fin with 16-18 rays. Caudal fin rounded.
30 or 31 scales in a longitudinal series (lateral-line scales absent, but indicated
by an interrupted line of pits). 22-24 scales around the body immediately in
front of the pelvic fins.
Coloration: Males brightly coloured, the body and fins reddish, flecked with
turquoise or green; a bright orange band, bordered ventrally with black, on
the anal fin. Females rather drab, bluish-grey.
Ecology: Since this species is rarely encountered in Uganda, nothing is known
of its ecology in the Protectorate. Some notes on N. taeniopygus in Tanganyika

I cm
FIG. 51
Nothobranchius taeniopygus. (Drawn by Barbara Williams.)

have been published (Vanderplank, 1941). In Tanganyika the species lives in
temporary streams; breeding takes place during the rainy season. Drought
resistant eggs are deposited in the mud, where they remain throughout the dry
season; when the streams are filled again, the embryos resume and complete
their development. According to Vanderplank, young N. taeniopygus feed on
Protozoa and the adults on crustaceans and insect larvae.
Further studies on this species would be of particular interest since few
African fishes are known to produce drought-resistant eggs. Nothobranchius
taeniopygus is tolerant of a wide temperature range (Vanderplank, op. cit.)
and lives well in aquaria.
Distribution: In Uganda, the species has been found in streams near Lake
Victoria and in the Soroti district (E.A.F.R.O.). Elsewhere, it is recorded from
streams, rivers and small lakes in Tanganyika Territory.

Aplocheilichthys Bleeker 1863
Body relatively slender, its depth contained 31-53 times in the standard
length. Origin of the dorsal fin behind the first anal ray. Dorsal surface of the

head flattened. Teeth in several rows, those of the outermost row somewhat
The genus is widely distributed in tropical Africa, extending as far south
as Zululand and Angola. Four species, and possibly a fifth, are recorded from

24-30 scales in a longitudinal series 1
36-39 scales in a longitudinal series A. pelagicus
1 Dorsal fin with 7-8 rays A. kassenjiensis
Dorsal fin with 9-11 rays (i)
(i) Caudal fin shorter than the head A. pumilus
Caudal fin longer than the head A. eduardensis

Aplocheilichthys pelagicus (Worthington) 1932
Description: Body markedly elongate, its depth contained 5-5a times in the
standard length. Dorsal fin with 9-11 rays, anal fin with 14-16 rays. Caudal fin
somewhat truncated. Scales relatively small, 36-39 in a longitudinal series;
22-24 scales around the body in front of the pelvic fins.
Coloration: Bright yellow, a dark mid-lateral streak; caudal fin edged with
Size: Not exceeding 5 cm. standard length.
Ecology: Aplocheilichthys pelagicus is unique amongst the east African
members of the genus in that it inhabits the surface layers of the deep, open
waters of a lake (Worthington, 1932a). Other species are generally confined to
shallow, inshore water and to rivers.
The food of A. pelagicus consists of planktonic crustaceans and some insect
larvae (Worthington, 1932b).
Worthington (op. cit.) is of the opinion that A. pelagicus spawns in the open
Distribution: Known only from Lake Edward.

Aplocheilichthys kassenjiensis (Ahl) 1924
Description: Depth of body contained 3-4 times in the standard length.
Dorsal fin with 7 or 8 rays, anal fin with 15 rays. 24-26 scales in a longitudinal
series, 18-20 around the body in front of the pelvic fins.
Coloration: Yellow, with a thin, black mid-lateral stripe.
Size: Not known to exceed 32 mm. standard length.
Ecology: No data are available.
Distribution: Known only from Lake Albert.
Note: Poll (1952b) is of the opinion that this species may be identical with
Aplocheilichthys loati (Blgr.), which occurs in the White Nile.

Aplocheilichthys pumilus (Boulenger) 1906. (Fig. 52)
Description: Depth of body contained 33-41 times in the standard length.
Dorsal fin with 10 or 11 (rarely 9) rays, anal fin with 13-16 rays. 25-29 scales

in a longitudinal series, 16-18 around the body in front of the pelvic fins.
Caudal fin shorter than the head (rarely almost as long as the head).
Coloration: Orange-yellow above, turquoise below. Fins faint orange-yellow.
An iridescent blue spot on the operculum. When seen from above, the eyes are
bright silver.
Size: Not exceeding 5 cm.
Habitat: Lake Victoria: Marginal vegetation, including papyrus swamps.
The species has also been found in temporary streams flowing into the lake.
Food: Insect larvae and small crustaceans (E.A.F.R.O.).
Distribution: Uganda: Lakes Victoria, Nakivali and George; the affluent
streams of Lake Edward (including those in the Belgian Congo); the Victoria
Nile. Elsewhere: Lake Tanganyika and its affluent streams.
Note: Poll (1952b) draws attention to the great similarity between A. pumilus
and A. mahagiensis (David and Poll), a species known only from Lake Albert.
A. mahagiensis differs from A. pumilus mainly in its slightly deeper body
(31-3 times in standard length) and longer caudal fin (as long as the head). More
specimens of A. mahagiensis are required before the relationship of the two
species can be determined satisfactorily.

FIG. 52
Aplocheilichthys pumilus. (Drawn by Barbara Williams.)

Aplocheilichthys eduardensis (David and Poll) 1937
Description: Depth of body contained 31-41 times in the standard length.
Dorsal fin with 9 or 10 (rarely 8 or 11) rays, anal fin with 14-16 rays. 28-30
scales in a longitudinal series, 20-26 scales around the body in front of the pelvic
fins. Caudal fin longer than the head.
Coloration: Olivaceous-yellow, the scales with dark borders; a fine black
line mid-laterally.
Size: Not exceeding 5 cm. standard length.
Ecology: No data available.
Distribution: Lake Edward, its affluent rivers (including those in the Belgian
Congo), and the Semliki River. Percy and Ridley (1955) consider that some
cyprinodont fry collected by them at the edge of the Niansimbi Hot Swamp
(Bwamba) should be referred to this species. A single specimen was also
obtained by the same party in the Mongiro River, near the Mbuga Hot Spring.

Three Aplocheilichthys, which agree with A. eduardensis in all characters,
were caught recently in Lake Victoria at Bugungu near Jinja (E.A.F.R.O.);
further specimens are awaited with great interest. The shoals from which these
specimens were obtained were smaller than the shoals of A. pumilus in the same
habitat, and were not seen to swim near the surface, as is typical of A. pumilus.

Hypsopanchax Myers 1924
Resembles Aplocheilichthys, but with a markedly deeper body especially in
the abdominal region, which is compressed in males. Teeth in several irregular
rows, those of the outer rows in both jaws somewhat enlarged and more widely
spaced than the teeth of the inner series. In Uganda, at least for the present, this
genus may be recognized easily by the high number (19-21) of anal rays.
One species occurs in Uganda.

Hypsopanchax deprimozi (Pellegrin) 1928
Description: Depth of body, in adult males, contained 21-3 times in the
standard length, and, in females 33-4 times. Dorsal fin with 12 or 13 rays, anal
fin with 19-21 rays. 30-33 scales in a longitudinal series, 22-26 around the body
in front of the pelvic fins.
Coloration: Pale yellow, the edges of the scales darker in males; fins pale.
Ecology: The species is apparently confined to rivers (Poll, 1939; Worthing-
ton, 1932b).
Distribution: Affluent rivers of Lake Edward (including those in the Belgian
Congo) and the Semliki River.
Note: Haplochilichthys analis, Worthington 1932, is considered to be a
synonym of H. deprimozi (Poll, 1939).

Cynopanchax Ahl 1928
Resembles Hypsopanchax and Aplocheilichthys; the teeth are in three series
in each jaw, with those of the outermost row greatly enlarged, curved and
A single, apparently rare species is known, and is recorded only from Bukoba
(Tanganyika Territory), Lake Victoria.

Cynopanchax bukobanus (Ahl) 1928
Description: Body moderately deep, strongly compressed. Dorsal fin with
10 or 11 rays, its origin above the middle of the anal fin, which comprises 15
or 16 rays.
The only specimens of C. bukobanus are those from which the species was
described; no others have been collected subsequently.

(Only those publications not included in the bibliography for Chapter I are
listed here.)
Vanderplank, F. L. (1941). Nothobranchius and Barbus species: indigenous anti-
malarial fish in East Africa. E.A. Med. J., 17, 431-6.

Distribution of species of the families Cyprinodontidae, Anabantidae and Mastacem-
belidae in Uganda.

LAKES Victoria
Nothobranchius taeniopygus..
Aplocheilichthys pelagicus ..
Aplocheilichthys kassenjiensis
Aplocheilichthys pumilus .. P

Aplocheilichthys eduardensis ?P

Hypsopanchax deprimozi

Cynopanchax bukobanus .. P

Ctenopoma murei .. P
Ctenopoma damasi

Mastacembelus victoria .. P

Kyoga Edward Albert






Streams near Lakes
Victoria and Kyoga

Lakes George and
Nabugabo; streams
flowing to L. Edward.
Semliki and Mon-
giro rivers. Affluent
streams and rivers of
L. Edward
Semliki river; affluent
streams and rivers of
L. Edward

Affluent streams of
these lakes
Affluent streams of
L. Edward

Victoria Nile

I Kyoga I Edward I Albert

Family: Centropomidae
Head and body covered by scales. Mouth protractile, the maxilla large and
not completely covered by the pre-orbital bone. Two nostrils on each side.
Lateral line continuous.
Most members of the Centropomidae are marine fishes; one genus, Lates,
is found in the fresh-waters of Africa.

Lates Cuvier and Valenciennes 1828
Body deep, somewhat compressed; scales small and ctenoid. Small villiform
teeth in the jaws and on the vomer, palatines and ectopterygoids (bones forming
part of the roof of the mouth). Pre-orbital and pre-opercular bones armed with
spines; a large spine on the free edge of the operculum. Dorsal fin almost com-
pletely divided into two parts by a deep notch; the anterior part comprises
7 or 8 spines and the posterior part one spine and 10-14 branched rays.
The Nile Perch Lates niloticus (Linnd) has long excited the interests of man.
Over two thousand five hundred years ago, L. niloticus was worshipped by
the Egyptians. A special Lates cult was centred at Esneh, on the Upper Nile,
where numerous mummified Lates have been recovered from a burial ground
near the town (Gudger, 1947). Lates were also recovered from an animal-
cemetery at Gurbo, about sixty miles south of Cairo. Although other animals
were found in this burial ground, it is clear that preferential burial was given to
the Lates, thus suggesting that these fishes were of more than usual significance
(Gudger, op. cit.). A brief account of Lates in the religious life of early Egypt
is given by Gudger in the paper referred to above.
At present, interest in Lates is mainly confined to anglers and fishermen. As
might be expected, there are decided schools of thought on the angling qualities
of the fish (see Kinloch, 1956; Anderson, 1956 and McCormick, 1949), but there
is no disagreement regarding the high palatability of its flesh.
The student of African fresh-water fishes too, has a special interest in Lates
because of its distribution in the Great LakQs, and of the effect it may have
or have had, on the other fishes in these lakes. Again, opinions are divided (see
Worthington, 1954a; Iles and Fryer, 1955).
Lates are found in the Nile from Lake Albert to the Delta region, but no
Lates occur in present-day Lakes Edward, George, Victoria or Kyoga. The
genus also occurs in Lake Tanganyika and in the Senegal, Niger and Congo
river systems, but is absent from the Zambesi and other rivers of southern
Lates were, however, present in the Lake Victoria basin during the Miocene
period (about 25,000,000 years ago) and persisted until comparatively recent
times in Lake Edward. It is not yet possible to determine what factors were

responsible for the extinction of Lates in these lakes; that the genus was pre-
vented from recolonizing Lakes Edward and Victoria is undoubtedly due to the
barriers offered by the Semliki rapids and the Murchison falls.
At one time, two species of Lates, L. macrophthalmus and L. albertianus
were believed to occur in Uganda (Worthington, 1929a and b). Lates macroph-
thalmus, a small species, was thought to inhabit open waters, 60-120 feet deep,
whilst the larger L. albertianus -apparently occurred inshore and in shallow
Later (1940), Worthington expressed the opinion that the two 'species' were
probably only sub-specifically distinct. This view was adopted by David
and Poll (1937) and will be followed here. There is, however, a distinct
possibility that the relationship of the Lake Albert Lates with the species of
the Nile (L. n. niloticus) might be better expressed if, in Lake Albert, only one
sub-species was recognized. This view is supported by the fact that the sup-
posed correlation of morphological with ecological characters in the two Alber-
tine sub-species breaks down when large numbers of specimens are examined.

Lates niloticus albertianus Worthington 1929. (Fig. 53)
Native names: Mputa (Lunyoro and in general use); Gurr (Alud and Jonam).
English: Albert Perch.
Description: The generic description, together with the figure, should enable

-7 -- \ V


FIG. 53
Lates niloticus albertianus. (Drawn by Barbara Williams.)

the species to be identified immediately. The sub-species differs from L. n.
niloticus, and the other sub-species of Lake Albert, in having the caudal
peduncle more slender than in L. n. niloticus but stouter than in L. n. macroph-
thalmus. (Length/depth of caudal peduncle 1-1 in L. n. niloticus; 11-11 in
L. n. albertianus and 11-2 in L. n. macrophthalmus). Also, as compared with
L. n. macrophthalmus, the eye appears to be somewhat smaller, but this

character cannot be relied upon as eye-size is variable in adult fishes and is
proportionately greater in small fishes.
Coloration: Silvery, darker above and almost white below. Young fishes,
particularly those caught amongst reeds, are mottled with irregular dark-brown
Size: The largest recorded specimen weighed 360 lb. and was caught in a
seine-net; the heaviest fish caught by angling weighed 226 lb. and was six feet
long (Kinloch, 1956).
Economic importance: Lates n. albertianus is of considerable importance in
the fisheries of Lake Albert, particularly in the catches made by nine-inch mesh
gill-nets and in seine nets.
Ecology: Despite the great importance of this fish, very little is known of its
ecology in Lake Albert. Breeding habits are unknown and no detailed analysis
of its food and feeding habits has been made. Indeed, Worthington's original
study (1929a) remains the most comprehensive published account. According
to Worthington (op. cit.) young Lates feed mainly on invertebrates, particularly
prawns (Caridina nilotica) and insect larvae. Larger individuals (75-170 cm.
long) feed almost exclusively on other fishes, but it is not known whether any
particular genus, or group of genera, forms the main item in the diet.
Worthington (op. cit.) considered that L. n. albertianus is restricted to the
shallow, inshore waters of Lake Albert and to the Murchison Nile, but that it
occasionally moved into deeper waters. The species is abundant in sheltered
Sport: Lates n. albertianus can certainly be ranked amongst the premier
sporting fishes of Uganda. Full accounts of tackle, baits and good fishing areas
are given by Kinloch (1956) and Anderson (1956). Further accounts of angling
for 'Perch', and some statistics of catches, are given by Margach (1948) and
McCormick (1949).
Distribution: Known only from Lake Albert, the Albert and Murchison Niles.
The closely related Lates n. niloticus is widely distributed in Africa, occurring
in the Nile system, the Congo, Niger, Volta and Senegal, and in Lake Chad.
Recently, the Uganda Game and Fisheries Department have introduced L. n.
albertianus into the Victoria Nile (below the Owen Falls dam) and Lake Kyoga.
The results of this experiment are not yet available.

Lates niloticus macrophthalmus Worthington 1929
Native name: Apparently no distinction is made between this fish and
L. n. albertianus; consequently the term 'Mputa' is used to cover both sub-
Description: Very similar to L. n. albertianus, but probably distinguished
by its more slender caudal peduncle (length/depth of caudal peduncle 11-2 cf.
11-l1) and by the longer third dorsal spine. The principal difference between
the two sub-species was thought to be ecological, namely, that L. n. macroph-
thalmus occurred offshore in deep waters whereas L. n. albertianus was found
inshore and in shallow water. As mentioned above, recent observations seem
to indicate that neither the morphological nor the ecological differences are as
clear-cut as they were first thought to be (see Kinloch op. cit.).

Coloration: Similar to L. n. albertianus. Worthington (1929b) notes that the
young of L. n. macrophthalmus are not mottled or blotched, but this may be a
consequence of the habitat in which they were caught (see above).
Ecology: Apparently confined to water more than 60 feet deep (Worthington,
1929a and b). The food of adult fishes is probably similar to that of large
L. n. albertianus (Worthington, op. cit.). No precise data on breeding habits
are available. Worthington (op. cit.) believes that L. n. macrophthalmus breeds
in deep water. This view was based on the capture of sexually active adults,
and of fry 4-9 cm. long, in the open waters of Lake Albert.
Distribution: Known only from Lake Albert.

Periodic mortality of Lates in Lake Albert
The occurrence of periodic mass mortalities of Lates in Lake Albert has
been known for many years, but there is still no positive explanation for the
phenomenon. A peculiar aspect of these mortalities is the fact that apparently
only Lates are affected and that more large than small fishes are killed.
A recent explanation for these sudden and widespread deaths is based on a
combination of certain physiological peculiarities of the fishes coupled with
certain hydrological changes taking place in the lake (see Fish, 1956 and
Beauchamp, 1956). Lates are intolerant of moderately deoxygenated water and
die of asphyxiation if the oxygen content of the water falls below a level which
would not seriously affect many other species (Fish, op. cit.). At certain times
of the year the main inflow of water to Lake Albert (from the Semliki River) is
cooler than the lake; this relatively cold water, being denser than the warm
lake-water, sinks to, and flows along the bottom of the lake without mixing
with the upper layers of water. The cool lower layer is in contact with the
bottom deposits of the lake, and could therefore become deoxygenated more
rapidly than the upper layers, especially since these are agitated by wind action.
Because their oxygen requirements are greater, Lates living in deep water
would be affected by this deoxygenation more severely than would the other
Other factors, such as heavy parasitization of the gills (to which Lates is
particularly prone) and the general condition and age of the fishes, are un-
doubtedly contributory factors in a mass mortality.

(Only those publications not included in the bibliography for Chapter I are listed
Anderson, A. M. (1956). A fishing trip on Lake Albert. Uganda wild life and sport, 1,
Beauchamp, R. S. A. (1956). The electrical conductivity of the head-waters of the
White Nile. Nature, 178, 616-8.
David, L., and Poll, M. (1937). Contribution a la faune ichthyologique du Congo
Belge. Ann. Mus. Congo, Zool., (1), 3, Fasc. 5, 189-294.
Fish, G. R. (1956). Some aspects of the respiration of six species of fish from Uganda.
J. Exp. Biol., 33, 186-95.

lies, T. D., and Fryer, G. (1955). Predation pressure and evolution in Lake Nyasa.
Nature, 167, 407.
Gudger, E. W. (1947). The giant fresh-water Perch of Africa. Uganda J., 11, 106-9.
Kinloch, B. G. (1956). Fishing in Uganda. Uganda wild life and sport, 1, 13-21.
Margach, T. P. (1948). Nile Perch in Lake Albert. Uganda J., 12, 105-6.
McCormick, L. J. (1949). In Game Fish of the World, London.
Worthington, E. B. (1940). Geographical differentiation in fresh waters with special
reference to fish. In The New Systematics, Oxford.


Family: Cichlidae

Body-form variable; scales ctenoid or cycloid. Head not completely covered
by scales. Mouth protractile; teeth variable in form and number; no teeth on the
vomer, palatines or pterygoids (bones forming part of the roof of the mouth).
Well-developed pharyngeal teeth are carried on a triangular bone. A single
nostril on each side. Lateral line interrupted, usually in two parts.
This family is widely distributed in the fresh-waters of Africa, Central and
South America, and in Syria, Palestine, Madagascar, India and Ceylon. Certain
species inhabit brackish waters.
In Africa, the Cichlidae are the predominant fresh-water percomorph fishes
and have long attracted the attention of biologists, both from the scientific and
economic viewpoints. Although no cichlid species attains the great size of the
Nile Perch (Lates), the family is of great economic importance in most East
African lakes; for example the major fisheries of Lakes Victoria and George are
based on the exploitation of Tilapia species.
Certain east and central African genera, in particular the genus Haplo-
chromis, provide interesting problems in evolution. In each major lake there is
a unique flock of species exhibiting a wide range of adaptive types which has
evolved within the lake basin during geologically recent times (Brooks, 1950;
Greenwood, 1951a, 1956a and b).
The phenomenon of parental care is well developed in the majority of African
cichlids. In most species the female carries the fertilized eggs and, subsequently,
the young larvae in her mouth. Even when the young are free-swimming, the
female guards the shoal and takes the brood back into her mouth at the first
signs of approaching danger. Those species which do not practise mouth brood-
ing guard their eggs and young for some time after spawning.
The identification of most cichlid species is far from simple and much research
remains to be carried out on almost every genus in the family. Since many
species are small and of no great interest, except to the specialist, the account
of the family given below is of necessity somewhat abbreviated.
Cichlidae occurring in Uganda may be divided into two groups, namely, the
Tilapia species and the Haplochromis generic complex. The latter group con-
tains seven genera, of which the genus Haplochromis is the most widespread
and alone has numerous species. The six other genera are each represented by a
single species.
The many species of Haplochromis can only be identified with considerable
difficulty and by reference to large collections of specimens; a simple key to
the species has yet to be devised. Morphological differences between the species
are slight and individual variability within a species is high. Consequently, this
genus and the related monotypic genera will be described here in general terms.

References to papers dealing in detail with the Haplochromis of various lakes
in Uganda are, however, given.
Since the species of Tilapia are fewer in number, are of considerable economic
importance and of more general interest, greater attention will be paid to them
than to the Haplochromis.
Difficulties may be experienced when identifying individual Tilapia; usually
ii is simpler to identify several specimens from one place than to attempt the
identification of a single fish. In case of doubt it can only be recommended that
the specimen be referred to a specialist, rather than hazarding a guess as to its
identity. Much disappointment in the past could have been avoided if an expert
diagnosis had been given before Tilapia were introduced into ponds and dams
where it was hoped that they would flourish.

Scales cycloid Tilapia'
Scales ctenoid 1
1 Anal fin with four or more spines Astatoreochromis
Anal fin with three spines A
A Outer teeth close set, enlarged and with strongly incurved, flattened
crowns with rounded edges Schubotzia
Outer teeth not as above (i)
(i) Teeth in the upper jaw in two or more series anteriorly but in
a single series posteriorly (a)
Teeth in the upper jaw in two to five series both anteriorly and
posteriorly (b)
(a) Anterior outer teeth disproportionately longer than the adjacent
lateral teeth, slender, pointed and procumbent Paralabidochromis
Anterior outer teeth forming a graded size-series with the lateral
teeth, not forwardly directed; crown form variable Haplochromis
(b) Outer teeth enlarged and stout, the crowns obliquely truncated
and pointing inwards Macropleurodus
Outer teeth in both jaws small, their crowns bi-cuspid or uni-
cuspid; inner teeth in broad bands anteriorly and posteriorly (c)
(c) Lower jaw broad and flat; the tooth bands in both jaws of uniform
breadth anteriorly and posteriorly Hoplotilapia
Lower jaw stout, rounded anteriorly; teeth grouped anteriorly
into two pear-shaped patches Platytaeniodus

Tilapia Smith 1840
The majority of generic characters cannot easily be checked without dissec-
tion; the cycloid scales of this genus should, however, prove diagnostic. Small
fishes are characterized by the 'Tilapia-mark', a well defined black blotch or
marbling on the anterior part of the soft dorsal fin.
Since the most readily-determined specific characters are included in the
1 Small Tilapia may be distinguished from other cichlid genera by the presence of a
large dark spot or circumscribed marbling on the soft part of the dorsal fin.

key and figures, the description of each species will usually be confined to an
account of the coloration of live fishes.
There is a considerable body of literature dealing with various aspects of the
biology of Tilapia species in Africa. An additional bibliography dealing
specifically with Tilapia is therefore given at the end of this chapter.
In preparing this account of Tilapia, I have drawn heavily on the experience
and unpublished data of my colleague Mrs. R. H. McConnell, to whom I tender
my sincere thanks.

Note: Only those species occurring naturally in Uganda are included here.
A few other species from various parts of Africa have been introduced into
Uganda, but they are neither widespread nor common (see Lowe, 1955).

8-12 gill-rakers on the lower part of the first gill-arch T. zilli
18-28 gill-rakers on the lower part of the first gill-arch 1
1 Caudal fin with numerous vertical and wavy bars (i)
Caudal fin intensely spotted. (ii)
Caudal fin weakly if at all spotted; body not markedly deep (cf. Figs.
57 and 58 with Fig. 59) (iii)
Caudal fin without spots or bars; body deep (Fig. 59) (iv)
(i) Lateral-line series with 31-33 (usually 31) scales; 22-28 (usually
24-26) gill-rakers on the lower part of the first arch T. nilotica
(ii) Lateral-line series with 28-30 (usually 29) scales; 19-24 (usually
22) gill-rakers. Length/depth ratio of the caudal peduncle
0-5-0-8. Body dark, with numerous lighter spots T. leucosticta
(iii) (a) Lateral-line series with 30-33 (usually 31) scales; 19-23
(usually 20 or 21) gill-rakers. Length/depth ratio of the caudal
peduncle 0-8-0-9. Dorsal fin tipped with orange. A well-
developed bump in front of, and partly above, the eye T. variabilis
(b) Lateral-line series with 32-35 (usually 32 or 33) scales; 18-21
(usually 19) gill-rakers. Length/depth ratio of the caudal
peduncle 0-9-1-1. No marked bump anterior to, and above, the
eye. Dorsal fin sometimes with red tips, but never orange T. esculenta
(iv) Body deep (see Fig. 59). Lateral-line series with 28-31 scales;
21-26 gill-rakers. Length/depth ratio of the caudal peduncle
0-5-0-7 T. galilaea

Tilapia zilli (Gervais) 1848. (Fig. 54)
Native name: Uncertain.
Description: This deep-bodied species (depth contained 2-2j times in the
standard length) is easily recognized by its low gill-raker count (12-18) and the
distinctive 'Tilapia-mark' which persists in adult fishes.
Coloration: Body olivaceous, shot with an iridescent blue sheen; lips bright
green; chest pinkish (black surmounted by an intensely crimson flush in breed-
ing fishes). Six or seven dark vertical bars, of variable intensity, may be visible

on the body and caudal peduncle. Dorsal, caudal and anal fins olivaceous with
yellow spots, the dorsal and anal fins often outlined by a narrow orange band.
The 'Tilapia-mark' is a large, black, nearly circular spot almost completely
outlined in yellow. The coloration of breeding fishes is more intense than that
of non-breeders.
Habitat: Lake Albert: Shallow inshore waters, especially in sheltered bays.
Lake Victoria: T. zilli is a recent introduction in this lake and its habitat has
not yet been determined.
Food: Unlike the majority of Tilapia, which feed principally on certain algae
in the plankton, T. zilli feeds largely on the leaves and stems of rooted aquatic
plants and their associated epiphytic algae (see Fish, 1955). Since suitable
phytoplankton is often lacking in dams but water-weeds are usually present,
the feeding habits of T. zilli have made this species an important fish for pond
culture in East Africa.


FIG. 54
Tilapia zilli. (Drawn by Barbara Williams.)

Breeding: Tilapia zilli belongs to the group of Tilapia which are not mouth-
brooders. Spawning takes place in a prepared nest-a depression in the bottom;
thereafter one of the parents remains on guard, constantly cleaning the eggs of
any detritus which may settle on them and also maintaining a current of water
over the nest.
Newly hatched T. zilli are smaller than the larvae of mouth-brooding species
and are provided with an adhesive organ on the head. By means of an excretion
from this organ the young attach themselves to plants and stones.
Distribution: Uganda: Before the value of T. zilli as a pond and dam fish
was realized, the species was restricted to Lake Albert. Nowadays it is widely
distributed throughout Uganda. Considerable numbers have been introduced
into Lake Victoria at Entebbe and Kisumu (Kenya).

Elsewhere: The Nile basin; Lake Rudolf; Lake Chad; Sierra Leone to the
Niger; Palestine; Lake Galilee.

Tilapia nilotica (Linn6) 1757. (Fig. 55)
Native names: Mahere (Lunyankole); Ngege (Lunyoro and in general use);
Nsogora (Lugungu); Zogoro (Alur); Oro (Jonam).
Coloration: Greyish-brown ground-coloration, darker above; faint traces of
six or seven dark vertical bars on the flanks and caudal peduncle. Dorsal and
anal fins greyish, somewhat irregularly spotted. Caudal fin grey, with numerous
dark-red, wavy vertical bars. In breeding males the ventral surface of the body,
the anal, dorsal and pelvic fins are black, and the head and flanks are flushed
with red.


FIG. 55
Tilapia nilotica. (Drawn by Barbara Williams.)

Size: Tilapia nilotica reaches a large size in certain lakes; for example, in
Lake Rudolf specimens weighing 14 lb. have been recorded. In Uganda,
T. nilotica does not reach such a large size; specimens weighing 11-11 lb. may be
considered as being of an average size.
Habitat: Lake Albert: restricted to inshore waters, particularly in sheltered
bays and in lagoons cut off from the main lake by sand-spits (Worthington,
1929a; E.A.F.R.O., 1951).
Lake Edward: the comparatively shallow inshore waters; it is less common
in the more open waters and is absent from deep water except that occasional
shoals are seen swimming at the surface (Worthington, 1929a; Poll and Damas,
Lake George: Ubiquitous, but less common near papyrus-fringed shore-lines
(Lowe, 1957).

Tilapia nilotica has been introduced into Lakes Victoria and Kyoga, but no
information is available on its habitat in these lakes. One large (32 cm.) speci-
men was caught in experimental gill-nets set near the mouth of the Kagera
Food: Principally phytoplankton (either in suspension or from the bottom
deposits) of which the diatoms provide the main nourishment (Worthington,
1929a; Poll and Damas, op. cit.; Fish, 1955). Insects and crustaceans are also
eaten and digested (Worthington, op. cit.). In Lake Rudolf, T. nilotica feeds on
blue-green algae; in other lakes these algae are not digested by this or other
fishes. Certain bacteria from a substantial part of the diet of T. nilotica living
in Lake Kivu (Fish, op. cit.).
Economic importance: In Lakes Edward and George, T. nilotica is the most
important element in the catches of African fishermen.
Breeding: T. nilotica is a mouth brooder in which the female carries the young.
Distribution: Uganda: probably the most widely distributed species in
Uganda. T. nilotica occurs naturally in Lakes Edward, George and Albert, and
in many crater lakes in western Uganda. In addition, the species has been
introduced into Lakes Victoria, Kyoga, Nakivali, Kachira, Kijanebalola,
Mutanda and Bunyoni (where the experiment was apparently unsuccessful) as
well as into many smaller lakes and dams.
Elsewhere: The Nile; Lakes Rudolf, Tanganyika and Kivu; the Niger, Volta
and Senegal; Lake Chad; Syria and Palestine.
Historical: Like Mormyrus and Lates, Tilapia nilotica attracted the attention
of ancient Egyptian artists. The fish appears on murals and rock-engravings,
and was apparently kept in ornamental ponds (see plate 9 in the King Penguin
Egyptian Painters, in which book there is also a delightful picture (plate 3) of a
cat eating T. nilotica). Both these paintings date from about 1400 B.C.
Tilapia leucosticta Trewavas 1933. (Fig. 56)
Native names: Kinyamuroro (Lunyankole); Nzizi (Lake Albert).
Coloration: Body dark olive-green to slatey-black, clearly marked with
whitish spots; lower lip often bluish-white; eight to eleven dark vertical stripes
sometimes visible on the flanks. Dorsal, anal and caudal fins dark; the soft
dorsal fin, the entire caudal fin and the anal fin with well-defined bluish-white
spots. In breeding males the ground-colour changes to a dark blue-black, whilst
the whitish spots on the body and fins are intensified; the eye is outstanding
with its bright amber iris crossed by a black bar. Indeed, the pugnacious breed-
ing male invites comparison with some mediaeval 'Black Knight' arrayed for
The colour of breeding females does not differ greatly from that of sexually
inactive fishes.
The 'Tilapia-mark' in young individuals is a relatively indistinct marbling
on a yellow background.
Size: The maximum recorded size of T. leucosticta in natural waters is
30 cm. total length and weighing 565 gm., but specimens measuring 32 cm. and
weighing 610 gm. have been caught in one of the Teso dams (Lowe, 1957).
Ecology: A detailed account of the ecology of Tilapia leucosticta has recently

been published (Lowe, 1957). It will suffice here to give only a brief account of
the habitats occupied by this species in various lakes.
Lake Edward: in bays, on exposed shorelines (Poll, 1939) and at the mouths
of rivers (Lowe, op. cit.).
Lake George: ubiquitous, but more common near the papyrus fringed shores
than is T. nilotica, the other major Tilapia in this lake. (Lowe (op. cit.)
Lake Albert: inshore and in lagoons (Lowe, 1957). Lowe (op. cit.) notes
that in these lakes the distribution of T. leucosticta in relation to T. nilotica is
"remarkably similar to that of T. variabilis in relation to T. esculenta in Lake
Tilapia leucosticta has been introduced into Lakes Victoria and Kyoga but
no ecological data are yet available.
Food: Phytoplankton (Poll, 1939) and bottom deposits largely formed by
the 'rain' of phytoplankton from the surface layers (Fish, 1955; Lowe, 1957).

.. ,. .


FIG. 56
Tilapia leucosticta. (Drawn by Barbara Williams.)

Breeding: T. leucosticta is a female mouth-brooder, without well-defined
breeding seasons; sexually active fishes are found throughout the year (Lowe,
op. cit.).
Distribution: Lakes Albert, Edward and George; recently introduced into
Lakes Victoria, Kyoga and Wamala, and into numerous dams. The species is
restricted to these areas.

Tilapia variabilis Boulenger 1906. (Fig. 57)
Native names: Mpongo and Mbiru (Lake Victoria); Wasamaru (Lunyoro);
Nsalia (Lukenyi).

Description: A characteristic feature of breeding male T. variabilis is the
conspicuous genital tassel. This is a much branched structure, bright orange in
colour and often several centimetres long, developed from the genital papilla,
which lies immediately posterior to the anus.
Coloration: Adult but non-breeding fishes are uniformly grey or greyish-
green, the flank scales somewhat darker at the centre than the edge. Fins grey,
the caudal only weakly, if at all, spotted. Dorsal fin tipped with orange through-
out life.
Breeding males are bluish-grey to bluish-green; the caudal fin is outlined
in bright orange. The coloration of breeding females is similar to that of non-
breeding individuals.
An atypical colour variety also occurs; the ground coloration is basically a
piebald black and silver, variably blotched with bright orange. Such aberrantly
coloured fishes are usually females, although a small percentage is males.

FIG. 57
Tilapia variabilis. (Drawn by Barbara Williams.)

Immature Tilapia variabilis are greyish-silver, with eight to ten dark, vertical
stripes on the flanks and caudal peduncle. The dorsal fin is outlined in orange;
the 'Tilapia-mark' is a relatively indistinct black marbling.
Ecology: Lake Victoria: a recent paper (Lowe, 1956b) describes the ecology
of this species in some detail. Tilapia variabilis occurs mainly in water less than
sixty feet deep and particularly on exposed shores; however, the only Tilapia
caught in nets set on the bottom in one hundred feet of water was a T. variabilis.
Some confusion between T. variabilis and T. esculenta is apparent in the
account of these species given by Copley in The Game Fishes of Africa (1952).
Statements supposedly referring to the ecology of T. variabilis actually refer
to T. esculenta and vice versa; a similar error has been made with many of the
morphological data given in the same section.
Lake Kyoga: confined to areas close inshore, especially where there is a
surface covering of water-lily leaves (Worthington, 1929a).

Food: Mainly phytoplankton derived from bottom deposits (Worthington,
op. cit.; Fish, 1955; Lowe, 1956b), although algae are also scraped from rocks
and aquatic plants.
Breeding: A female mouth-brooder, without well-defined breeding seasons
(Lowe, 1956a and b).
Size: Most fishes in natural waters are sexually mature at a length of 22 cm.;
the maximum size attained is about 30 cm. (Lowe, 1956b).
Economic importance: In Lake Victoria, T. variabilis can, at present, be
considered of only moderate economic importance (see Lowe, 1956b). On the
other hand, in Lake Kyoga and the Victoria Nile, the species is of considerable
importance in the catches of gill-nets and traps.
Sport: Although not ranking as a game-fish, T. variabilis can be caught by
angling, using filamentous green-algae as bait.
Distribution: Tilapia variabilis is indigenous only in Lakes Victoria, Kyoga
and Nabugabo, and in the Victoria Nile; but, the species has been widely distri-
buted in dams and ponds throughout Uganda and in neighboring territories.

Tilapia esculenta Graham 1928. (Fig. 58)
Native names: Ngege (in general use); Nsalia (Lukenyi); Anagu (Lango).
The Lunyara and Ludope names (Binage and Mnege) apply to both T. esculenta
and T. variabilis.



....... ..

FIG. 58
Tilapia esculenta. (Drawn by Barbara Williams.)

Coloration: Reddish-brown; fins greyish, weakly if at all spotted; dorsal fin
without coloured outline, or, if coloured, a deep red (never orange). Breeding
males have the ventral surface of the body sooty, and the flanks suffused with

Young fishes are silvery-grey; the 'Tilapia-mark' is a well-defined black
spot outlined in pale yellow.
Ecology: The most recent and comprehensive account of the ecology of
T. esculenta is that by Lowe (1956b). Preliminary studies are contained in
Graham (1929) and Worthington (1929a).
In Lake Victoria, T. esculenta is confined to water less than sixty feet deep
and is most abundant in sheltered gulfs and bays where the lake bottom is
composed of soft algaceous mud (cf. T. variabilis which is predominantly a
species of open and exposed coast lines).
In Lake Kyoga, T. esculenta is apparently confined to open water (Worthing-
ton, 1929a).
Food: The food of this species consists almost entirely of phytoplankton,
from which only the diatoms are digested (Graham, 1929; Worthington, 1929a;
Fish, 1955; Lowe, 1956b). Insect larvae and planktonic crustaceans occur less
frequently although they may contribute to the diet of young fishes (E.A.F.R.O.,
Breeding: T. esculenta is a female mouth-brooder, without well-marked
spawning seasons; greatest breeding activity occurs between September and
May. A detailed account of the breeding biology of this species is given by
Lowe (1956b).
Size: In Lake Victoria, most T. esculenta are sexually mature at a length of
25 to 26 cm.; the modal adult size is from 30-32 cm. total length. Fishes longer
than 36 cm. total length are uncommon, although specimens 40-50 cm. long
have been caught in the Tanganyika waters of Lake Victoria,
Economic importance: There can be little doubt that T. esculenta must be
considered of primary importance to the fisheries of Lake Victoria. In most
parts of the lake, including many areas in Uganda, T. esculenta is the principal
catch of the commercial fishermen. Its flesh is readily accepted by Africans and
Europeans alike.
Distribution: Tilapia esculenta is indigenous in Lakes Victoria, Kyoga and
Nabugabo, and in the Victoria Nile. It has now been introduced into Lakes
Kijanebalola and Bunyoni and into many dams in Uganda, Kenya and

Tilapia galilaea (Art6di) 1757. (Fig. 59)
Native names: Kaishata (Lunyoro).
Description: The deep body (depth contained 1 -21 times in the standard
length) and unspotted caudal fin provide the most obvious diagnostic characters
for this species.
Coloration: Body and fins uniformly silver-grey; three to five vertical stripes,
of variable intensity, may be present on the flanks.
Ecology: Tilapia galilaea occurs in the shallow, inshore waters of Lake
Albert, where it is by no means common. Daget (1954) gives a fairly detailed
account of this species in the Upper Niger.
Food: Predominantly phytoplankton; planktonic crustaceans are also
eaten (Worthington, 1929a).

Breeding: Tilapia galilaea is a female mouth-brooder (Daget, 1954, op. cit.;
Lowe, 1955b); the small eggs are olive-green in colour, that is, similar to the
eggs of T. zilli, but unlike the bright yellow eggs of the other Tilapia species
occurring in Uganda.
Size: Fishes larger than 37 cm. total length are uncommon.
Economic importance: Slight.
Distribution: Uganda Lake Albert. Elsewhere: Nile; Niger; Volta; Senegal;
Gambia: Lake Chad: Palestine and Jordan.

FIG. 59
Tilapia galilaea. (Drawn by Barbara Williams.)

GENERA: Astatoreochromis, Schubotzia, Paralabidochromis Macropleurodus,
Hoplotilapia and Platytaeniodus.
These genera, each represented by a single species, may be considered as part
of the Haplochromis generic complex. All are small fishes, rarely exceeding
15 cm. standard length.

Astatoreochromis alluaudi Pellegrin 1904
Distribution: Indigenous to Lakes Victoria, Edward, Kachira, Nakivali and
Kijanebalola; the Victoria Nile and the Semliki River.
Since the diet of A. alluaudi consists almost entirely of molluscs, this species
may prove of value in controlling the vector snails of such diseases as Bilharzia
and Liver-fluke. If experiments with A. alluaudi are carried out, it can be
expected that the distribution of the species will be extended; it has already
been introduced into Tanganyika Territory (E.A.F.R.O.).

References: Trewavas, 1933; Poll and Damas, 1939; Greenwood, 1954;
E.A.F.R.O., 1951 and 1953.

Schubotzia eduardiana Boulenger 1914
Distribution: Known only from Lake Edward (see Trewavas, 1933).

The following monotypic genera are known only from Lake Victoria and are
discussed at length by Greenwood (1956a): Paralabidochromis victoria
Greenwood 1956 (a very rare species); Macropleurodus bicolor (Boulenger)
1906; Hoplotilapia retrodens Hilgendorf 1888; Platytaeniodus degeni
Boulenger 1906.

Haplochromis Hilgendorf 1888
Native names: Nkeje (Lake Victoria); Nyamororo (Lunyankole); Butoba
(Lugungu); Mpere (Lunyoro, Luruli and Lunyara); Waloko (Lukenyi); Anang
No attempt will be made here to produce a key to the numerous species of
Haplochromis occurring in Uganda. Keys to the Haplochromis species in the
three major lakes have been published (Regan, 1922, for Lake Victoria; Tre-
wavas, 1933, and Poll and Damas, 1939, for Lake Edward) but personal
experience has shown that these keys are difficult to use unless one has an
extensive knowledge of the fishes, and a named collection for comparison with
the new material. Furthermore, a thorough taxonomic revision of the genus,
coupled with ecological studies, is an urgent necessity. A study of the Lake
Victoria Haplochromis has recently been completed, but all the results are not
yet published (Greenwood, 1956a and b, and 1957). This work has again shown
the difficulty of producing a simple yet effective key which can be used by any-
one who has not made a special study of the group.
Some difficulty may at first be experienced in separating young Tilapia from
young and adult Haplochromis. The following characters should prove useful
in this respect.
Haplochromis Tilapia
Scales: Rough posterior border. Smooth posterior border.
Gill rakers: Rarely more than 12. Rarely less than 18 (except in T. zilli).
Coloration: No large spot on the an- A well-defined spot on the soft dorsal fin.
terior part of the soft
dorsal fin.

In no Uganda lake can the Haplochromis be considered of direct economic
importance. On the other hand, these small fishes, which occur in countless
thousands, do play an important role in the bionomics of the lakes. Haplo-
chromis provide the principal food for many economically important species,
such as Bagrus and Clarias, and are probably an important element in the diet
of Lates. By the mere fact of their living and dying, the Haplochromis help
to maintain the fertility of the lakes by constantly returning essential chemicals
to the water, either in their excreta or as the decomposition products of their
dead bodies.

Around Lake Victoria, some use is made of Haplochromis as food. The
fishes are caught in beach-operated seine nets and in traps, and are usually
sun-dried before being eaten. Haplochromis (Nkeje) played an important part
in the village life and customs of Buganda, where these fishes were an essential
dish at all feasts (Nsimbi, 1956).
The number of species in each lake is of interest and is probably correlated
with various ecological changes which took place during the evolution of the
lakes. There are about seventy Haplochromis species in Lake Victoria, about
twenty-three in Lake Edward, five in Lake Albert, four in Lake Nabugabo and
two in the Koki lakes.
The two major lake systems, Victoria and Edward, together with Lake
Nabugabo, are outstanding for their high percentage of endemic Haplochromis
species (see Chapter I). Species occurring in Lakes Edward, Victoria and
Nabugabo, although mostly endemic to each lake, are very closely related to
one another and at least five species are common to the three lakes. The
Haplochromis of the Koki lakes are species which also occur in both Lake
Edward and in Lake Victoria. Geological and zoological evidence (see Green-
wood 1951a for references) indicates that during the past million years all these
lakes were probably interconnected at some period, and that the species flocks
of Lakes Edward and Victoria were derived from the same ancestral stock.
In each lake, species of Haplochromis have occupied almost every available
habitat. This adaptive radiation is seen most clearly in Lake Victoria, although
the species flock of Lake Edward also shows considerable ecological diversity,
but on a reduced scale (Poll and Damas, 1939). Even the small Haplochromis
flock in Lake Albert contains representatives of several feeding habits
(Trewavas, 1938).
In a comparable marine biotope the different habitats and ecological niches
would be occupied by different species belonging to several genera and even
families, and not, as in these lakes, by different species of one genus.
The Lake Victoria Haplochromis species, may, on the basis of their usual
source of food, be divided into the following categories: mollusc eaters, plant
detritus eaters, grazers on epiphytic and epilithic algae, insectivorous species,
predators on the eggs and larvae of other Haplochromis, and finally, predators
on small fishes-especially other Haplochromis, and Engraulicypris. By far
the greatest number of species in Lake Victoria are predators on other fishes.
The categories listed above represent the chief food of the different species,
but most are facultative feeders and include in their diet a number of
different foods besides the main items. Even the vegetarian species are known
to feed on insects if, for example, there has been a heavy hatch of lake-flies.
Like the majority of Tilapia species, the Haplochromis are female mouth-
brooders; there are, however, numerous species whose breeding habits are
still unknown.
Papers dealing with Tilapia species and having special reference to Uganda:
Daget, J. (1956). Mdmoires sur la biologie des poissons du Niger moyen. II. Re-
cherches sur Tilapia zilli (Gerv.). Bull. I.F.A.N., 18, ser. A., 165-223.

Greenwood, P. H. (1953). Feeding mechanism of the cichlid fish, Tilapia esculenta
Graham. Nature, 172, 207.
Lowe, R. H. (1955a). Species of Tilapia in East African dams, with a key for
their identification. E. A. Afric. J., 20, 256-62.
(1955b). The fecundity of Tilapia species. E. A. Agric. J., 21, 45-52.
Lowe (McConnell), R. H. (1956a). The breeding behaviour of Tilapia species
(Pisces-Cichlidae) in natural waters: Observa-
tions on T. karomo Poll and T. variabilis
Boulenger. Behaviour, 9, 140-63.
(1956b). Observations on the biology of Tilapia
(Pisces-Cichlidae) in Lake Victoria, East
Africa. E.A.F.R.O. Supplementary Publica-
tion No. 1. Kampala.
(1957). Observations on the diagnosis and biology of
Tilapia leucosticta Trewavas (Pisces-Cich-
lidae) in East Africa (in the press).
Papers dealing with Haplochromis and related genera, but not listed in the biblio-
graphy for Chapter I:
Greenwood, P. H. (1954). On two cichlid fishes from the Malagarazi river
(Tanganyika) with notes on the pharyngeal apophysis
in species of the Haplochromis group. Ann. Mag.
nat. Hist., (12), 7, 401-14.
(1956a). The monotypic genera of cichlid fishes in Lake
Victoria. Bull. Br. Mus. nat. Hist., Zool., 3, No. 7,
(1956b). A revision of the Lake Victoria Haplochromis species
(Pisces, Cichlidae). Part I. Bull. Br. Mus. nat. Hist.,
Zool., 4, No. 5, 223-44.
(1957). A revision of the Lake Victoria Haplochromis species
(Pisces, Cichlidae). Part II, ibid, 5, No. 4, 76-97.
Nsimbi, N. B. (1956). Village life and customs in Buganda. Uganda J., 20, 27-36.

Distribution of

Tilapia spp. (Cichlidae) in Uganda. I= introduced i.e., not occurring

Victoria Edward
LAKES and and Albert Introduced Elsewhere
Kyoga George
Tilapia zilli .. I (Victoria) P Dams and ponds
Tilapia nilotica.. I P P Koki lakes; Lake Bun- Many crater
yoni; dams, ponds and lakes; Albert and
some crater lakes Murchison Niles
Tilapia leucosticta I P P Lake Wamala; various
dims and ponds, especi-
ally those in the Teso
Tilapia variabilis P Dams and ponds Victoria Nile
Tilapia esculenta P Lakes Kijanebalola and Victoria Nile
Bunyoni. Many dams
and ponds
Tilapia galilaea .. P Dams in the Teso dis- Albert Nile

Family: Anabantidae
Body short, moderately compressed. Scales large and ctenoid, the entire
head covered with scales. Two nostrils on each side. An accessory breathing-
organ is situated above the gills.
The Anabantidae are wide-spread in Africa and Asia; one genus occurs in
Uganda (see Table VIII, p. 196).

Ctenopoma Peter 1844
Lateral line interrupted, forming an upper (anterior) and lower (posterior)
series of lateral-line scales. Anterior nostril in the form of a short tube.
The accessory breathing-organ is developed from the first gill-arch of each
side; it lies in a large chamber above the gills. Each organ consists of a thin
plate thrown into numerous folds and is covered by vascular epithelium through
which respiration takes place. Because of its ability to utilize atmospheric
oxygen in addition to carrying out aquatic respiration, Ctenopoma can tolerate
foul and stagnant waters.
Two species occur in Uganda.
(i) Dorsal fin with 15 or 16 (rarely 14) spines; sub-operculum with
numerous, fine serrations. Two groups of well-defined spines,
separated by a deep notch, at the upper angle of the posterior
margin of the operculum C. murei
(ii) Dorsal fin with 17 or 18 spines. Sub-operculum smooth or with
feebly-developed serrations. Opercular spines weak, separated by
a shallow notch C. damasi

Ctenopoma murei (Boulenger) 1906. (Fig. 60)
Native name: Kizinzi (Lukenyi).
Description: The generic description, together with the characters given in
the key, should enable this species to be identified easily. C. murei is a small
species, rarely exceeding 90 mm. standard length.
Coloration: Silver-grey, the body with numerous dark spots and a well
defined black spot, outlined in yellow, at the base of the caudal fin and partly
extending onto the fin itself.
Habitat: Streams and swamps, both temporary and permanent; also in the
marginal grass and water-lily swamps of Lakes Victoria and Kyoga.
Food: The scanty information available indicates that C. murei feeds on
insect larvae and Crustacea (E.A.F.R.O.).
Breeding: Larval C. murei have been found in a temporary stream connected
with Lake Victoria. The presence of these larvae, together with numerous
sexually active adults, suggests that spawning takes place when such streams
are flooded during the rainy seasons (E.A.F.R.O.).


Icm c- f
FIG. 60
Ctenopoma murei. (Drawn by Barbara Williams.)

Distribution: Uganda. Lakes Victoria, Kyoga and Edward, and their affluent
streams; streams in Uganda. Elsewhere: The Nile; Lake Tanganyika.

Ctenopoma damasi Poll 1939
Native name: Unknown.
Description: In general appearance C. damasi resembles C. murei, but it is
distinguished by the greater number of dorsal fin spines (17 or 18 as compared
with 14-16), the weakly serrated, or even smooth, sub-operculum and by the
poorly-developed opercular spines, which are separated by a shallow notch.
Coloration: Dark brown to blackish; a small black spot without a surround-
ing lighter area, at the base of the caudal fin (Poll and Damas, 1939).
Ecology: There is no published information on the ecology of this species.
From the locality list given by Poll and Damas (op. cit.) C. damasi would appear
to be more common in rivers and in the vicinity of river-mouths than in the lake.
Distribution: Lake Edward and its affluent rivers.

Family: Mastacembelidae
Body elongate and eel-like, covered with minute, cycloid scales. Anterior
nostril in the form of a short tentacle. Snout produced as a fleshy appendage.
Mouth not protractile. Dorsal and anal fins long, confluent with the caudal fin
(at least in the African species). Pelvic fins absent.
The Mastacembelidae are widespread in southern Asia and in tropical Africa,
(where the family is represented by a single genus). Despite their eel-like appear-
ance, the Mastacembelidae are not related to the true eels.
Mastacembelus Scopoli 1777
Little need be added to the description given above. One species occurs in
Uganda (see Table VIII, p. 196).

Mastacembelus victoria Boulenger 1903. (Fig. 61)
Native name: Unknown. English: Spiny-eel.
Description: Characters given in the description of the family should enable
this species to be identified readily. The dorsal fin consists of 32-34 (rarely

FIG. 61
Mastacembelus victoria. (Drawn by Barbara Williams.)

29-31, or 35) separate spines, and a soft fin (about 100 rays) confluent with the
caudal and soft anal fins. The latter consists of about 100 branched rays and
two separate spines (the anterior spine is often deeply embedded in a fold of
Coloration: The colour patterns shown by M. victoria are extremely vari-
able. Observations on fishes kept in aquaria indicate, however, that the pattern
is fixed and does not change with the environment. The ground coloration, on
the other hand, does show some correlation with the tone of the back-ground
on which the fish is living.
Five common colour-patterns of fishes from Lake Victoria and the Victoria
Nile are:
(i) Ground-colour brown; numerous white spots on the body, becoming
reticulations ventrally. Dorsal and anal fins marbled or reticulate. Head with
numerous white, wavy lines.
(ii) Similar to (i) but body spots replaced by marbling.
(iii) Ground-colour of body and fins almost uniform brown, with darker
blotches at the base of the dorsal fin. Anal fin with large, white reticulations.

(iv) Almost uniform brown, but head with white reticulations.
(v) A combination of (iii) and (iv), marbled above, reticulate and spotted
The brown ground-colour varies from a dark chocolate (almost black in
some cases) to a bright, orange-brown.
Size: Boulenger (1916) gives 33 cm. as the longest recorded length; the modal
adult size is about 17 cm.
Ecology: As M. victoria is not often caught, there is very little information
about its ecology. In Lake Victoria, the species is found in marginal swamps and
has been caught in temporary streams. It is particularly common in the Victoria
Nile below the Owen Falls dam. In this habitat, M. victoria lives amongst the
stones forming the river-bed.
Food: Principally insect-larvae and, to a lesser degree, other bottom-living
invertebrates (E.A.F.R.O.).
Breeding: Unknown.
Distribution: Known only from Lake Victoria and the Victoria Nile.

I am greatly indebted to my colleagues of the East African Fisheries Research
Organization who have helped me in innumerable ways; I am particularly
grateful to Mrs. R. H. McConnell and Dr. P. S. Corbet who have allowed me to
make full use of their unpublished data. Thanks are also due to Dr. E. Trewavas
of the British Museum (Natural History) and to the officers of the Lake Victoria
Fisheries Service and the Uganda Game and Fisheries Department. Finally,
1 must express my gratitude to Mrs. Barbara Williams for making the drawings,
and to Mr. H. B. Thomas for his patient and thorough editing of the various



OF recent years there have been several references in this Journal and else-
where to Major E. R. Owen and his work in Uganda and the Sudan.' There
seems however to be no first hand account of the place where he died and is
This place, Ambigol, lies on the track which runs south from Wadi Halfa
along the east bank of the Nile, and is some 60 miles south of that town. The
road here swings away from the river to run through the desolate stretch of
country graphically called in Arabic el batn el hagar, 'the belly of the rocks',
and Ambigol wells, where the fort Owen commanded stands, are about four
miles from the Nile.
Kitchener's advance in his campaign against the Dervishes took this route
which is still the main north-south road and, though now traversed by an
increasing number of lorries, it remains a rough track probably little different
in appearance from that which must have existed from Pharaonic times.
The importance of Ambigol lies in the existence of its wells; the road runs
away from the river for some 40 miles from Sarras to Akasha and along it water
can only be obtained here and at Murat. It thus became a place of some sig-
nificance on the supply route of the army whilst it was moving south to the
action at Firka. A small fort (Fig. 1) was built on top of a hill overlooking the
wells and Owen was placed in command of it in the summer of 1896. The fort
itself consists of the breastwork forming a strong point on the top of the hill
and a larger area enclosed by the wall which runs along the hill top on the left
of Fig. 1. Within the area so enclosed cleared paths can still be seen as well as
circular areas cleared of stones where the bell tents of the troops were placed.
Here Owen died of cholera on 11 July 1896 and was buried about a quarter
of a mile to the south-east of the fort and only a few yards from the road. Fig. 2
shows his grave as it was in January 1955. The red granite slab covering the
grave bears on one side the inscription 'In ever loving memory of "Roddy"
Roderic Owen, D.S.O., Major Lancashire Fusiliers who died of cholera at
Ambigol Wells July 11th 1896, aged 40 years "Under the Shadow of the Sword
is Paradise."' And on the other, '"I have loved thee with an everlasting love
Jer. xxxi. 3."'
The sword carved on the top of the slab has been badly chipped and an
attempt has been made to move the whole slab which has been shifted a few
inches. The whole grave is enclosed by a dry stone wall.

1 The last of these giving references to other material is by H. B. Thomas in Uganda J.,
18 (1954), 186-7. Attention may also be drawn to the poem by Rennell Rodd printed in
Thomas' article, The Mission to Uganda in 1893-In Memory. Uganda J., 17 (1953), 4;
and to Sir William Gowers' letter in Uganda J., 18 (1954), 76-7.

., .-

.. -,* .*- ; ., "
FIG. 1
Ambigol Fort.

", :.( ". A i ,W


FIG. 2
Roddy Owen's Grave.

[facep. 2zo

Sr~ ~fs~