Front Cover
 Title Page
 Table of Contents
 H. B. Thomas: An appreciation
 The diaries of Emin Pasha - Extracts...
 Sudan-Uganda boundary rectification...
 The distribution of traditional...
 Gully erosion in the Queen Elizabeth...
 An Akarimojong-English check list...
 Some aspects of Kiganda Religi...
 Caves and rock shelters of Western...
 The "lost countries" of Bunyor...
 Inter-tribal warfare on Mount Elgon...
 Notes on contributors
 Editorial note
 Index to Volume 26 (1962)
 Maps of Uganda
 Back Cover

xml version 1.0 standalone yes
PageID P108
ErrorID 3

The Uganda journal
External Link ( Journal issues, 1999-2002 )
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00080855/00046
 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: UF00080855:00046
 Related Items
Related Items: Uganda journal

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Title Page 1
        Title Page 2
    Table of Contents
        Table of Contents
    H. B. Thomas: An appreciation
        Page 119
        Page 120
    The diaries of Emin Pasha - Extracts IV
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
    Sudan-Uganda boundary rectification and the Sudanese occupation of Madial, 1914
        Page 140
        Page 141
        Page 142
        Page 142a
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
    The distribution of traditional types of food storage containers in Uganda
        Page 154
        Page 155
        Page 156
        Page 156a
        Page 156b
        Page 157
        Page 158
        Page 159
        Page 160
    Gully erosion in the Queen Elizabeth National Park
        Page 161
        Page 162
        Page 162a
        Page 163
        Page 164
        Page 165
    An Akarimojong-English check list of the trees of Southern Karamoja
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
    Some aspects of Kiganda Religion
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
    Caves and rock shelters of Western Uganda
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 188a
        Page 188b
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
    The "lost countries" of Bunyoro
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
    Inter-tribal warfare on Mount Elgon in the 19th and 20th centuries
        Page 200
        Page 200a
        Page 202
        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 212a
        Page 212b
        Page 212c
        Page 212d
        Page 212e
        Page 212f
        Page 212g
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
    Notes on contributors
        Page 221
    Editorial note
        Page 222
    Index to Volume 26 (1962)
        Page 223
        Page 224
    Maps of Uganda
        Page 225
        Page 226
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text

The Uganda Journal



W. W. BISHOP 161
A History of East Africa (by Kenneth Ingham) BETHWELL A. OCOT 218
Development from Below (by U. K. Hicks) D. GHAI 219
La:ns about Marriage in Uganda (Uganda Council of Women)
INDEX TO VOLUME 26, [March and September 1962] 223

Published by
Price: Shs. 15 (15s.)



His Excellency Sir
Mr. W. S. Kajubi

The President
The Vice-President
The Hon. Secretary
The Hon. Treasurer
The Hon. Editors
The Hon. Librarian
Mr. A. C. Badenoch
Mr. N. Calogeropoulos
Mr. J. L. Dixon
Mr. R. K. K. Gava
Mr. S. C. Grimley
Mr. B. S. Hoyle
Hon. Secretary:
Hon. Treasurer:
Hon. Editors:

Hon. Librarian:

Walter Courts, K.C.tM.., M.B.E.
Mr. P. Bitature
Mr. P. N. Kavuma, O.B.E.
Mr. B. E. R. Kinran, r.B.B.
Mr. C. M. S. Kisosonkole
Mrs. NI. Macpherson
Mr. D. K. Marphatia, M.B.E.
Mr. R. J. Mehta, o.e.B.
Mr. C. N. Mukuye
Mr. K. K. Nganwa
Mr. B. A. Ogot
Mr. A. H. Russell, M.B.B., D,.SC,
Col. C. D. Trimmer, D.8.o.

Mr. P. Marsh
Mrs. J. Bevin
Dr. J. K. Almond
Dr. C. Gertzel
Dr. M. Posnansky
Miss P. Fiddes

Hon. Auditors: Hon. Legal Adviser:
Messrs. Cooper Bros. & Co. Mr. C. L. Holcom
Corresponding Secretary at Mbale: Dr. H. F. Morris
Corresponding Secretary at Tororo: Dr. W. H. R. Lumsden
Hon. Vice-Presidents:
H.H. Frederick Mutesa II, K.B.E., Sir John Milner Gray
Kabaka of Buganda Mr. E. B Haddon
R. A. Sir Tito Winyi Gafabusa IV, Mr. H. B. Thomas, o.n.F.
c.B.E., Omukama of Bunyoro Professor A. W. Willims
Lord Twining of Tanganyika and
Godalming, c.c.M.c., M..B.E.
Past Presidents:

Sir A. R. Cook, C.M.C., O.B.B.
Mr. E. J. Wayland, c.B.E.
Dr. H. H. Hunter, C.B.e., LL.D.
Mr. H. Jowitt, c.M.c.
Sir H. R. Hone, K.B.E., M.C., Q.C.
Mr. J. Sykes, O.B.E.
Mr. N. V. Brasnett
Captain C. R. S. Pitman,
C.B.E., D.S.o., M.C.
Mr. S. W. Kulubya, C.B.E.
Mr. E. A. Temple Perkins
Mr. R. A. Snoxall
Dr. K. A. Da'ies, C.M..., O.B.E.
Mr. G. H. E Hopkins, o..B..
Mrs. K. M. Trowell, M.B.E.
Dr. W. J. Eggeling

1948-50 Dr. G. ap Griffith
1950-51 Rev. Fr. F. B. Gaffney, w.v.
1951-52 Professor A. W. Williams,
1952-53 Sir J. B. Hutchinson,
C.M.G., F.R.S.
1953-54 Mr. J. D. Jameson, o.B.s.
1954-55 Dr. Audrey I. Richards, c.B.E.
1955-56 Rev. Dr. H. C. Trowell, o.B.E.
1956-57 Mr. D. K. Marphatia, M.B.E.
1957-58 Mr. M. Barrington Ward
1958-59 Dr. H. F. Morris
1959-60 Professor A. W. Southall
1960-61 Mr. J C. D. Lawrance
1961-62 Mr. Bi E. R. Kirwan, M.B.B.

Mr. S. W. Kulubya, c.B.E.

Mr. B. K. Mulyanti, o.n.E.
Secretary: Mrs. J. Bevin

Mr. G. P. Saben




Uganda Journal



No. 2



Published by



W. W. BISHOP 161




A History of East Africa (by Kenneth Ingham) BETHWELL A. OGOT 218
Development from Below (by U. K. Hicks) D. GHAI 219
Laws about Marriage in Uganda (Uganda Council of Women)



INDEX TO VOLUME 26, [March and September 1962]

t. .

H. B. THOMAS, O.B.E. Uganda, 1911-1940.
(Photo reproduced by kind permission of the Director of Lands and Surveys, Kampala).


THIRTY years ago a delightful little book called Sunshine and Rain in
Uganda came out, describing the life of the wife of a Field Surveyor. Looking
through its descriptions of safaris in the bush during the Twenties, one cannot
help being seized with a feeling of nostalgia for the days when out on safari
one made one's own way of life for months on end, with the excitement of
a brief return to civilization when one came in to headquarters for a few months
"to taste of the fleshpots of Egypt" once again.
H. B. Thomas came out to Uganda 21 years before this book was published,
but much of his and Mrs. Thomas's life in Uganda must have followed the same
pattern as that described in Sunshine and Rain: the runner returning from the
nearest station with mail and such provisions as might have been procurable-
certainly his askari escort would be carrying the same old Martini-Henry rifle
which was as obsolete in 1911 as are those which one can still see at Ssaza
Headquarters in this year of grace.
The surveyor in the old days had to pick up a knowledge of Luganda fairly
quickly. There was no question of making do with Swahili, as most surveyors
started off on Mailo Land work in Buganda, and all the headmen and chainmen
were Baganda, a team of whom would probably stay with one officer for ten
years or more. One got to know these men and all their family woes and joys,
and obviously in such surroundings during the years before the First Great War,
H. B. Thomas began to acquire his deep knowledge of the Africans and the land
they lived in. He was a very proficient speaker of Luganda and although he never
bothered to take his Higher Standard his oral fluency was well up to that mark.
Because of his great knowledge of Uganda, he was chosen by Sir Bernard
Bourdillon to work with Robert Scott on the 1935 Uganda Handbook. The two
authors produced a readable, scholarly and authoritative work, and it seems a
pity that no one has been found capable of bringing it up to date.
Over the years H. B. Thomas has written more than a score of articles for
the Uganda Journal, involving historical research, but two in particular bear
the stamp of his enquiring attitude and painstaking research for the truth. The
first of the articles is the story of how Stanley's letter from the court of Muteesa
I, appealing for missionaries to come to Buganda, was at one time believed to
have been found in a bloodstained riding boot when its courier, Linant de
Bellefonds, was killed by the Bari. The second debunks a man called Emile
Jonveaux who hoaxed the public with a book of African adventures written
from the comfort of his armchair.
Today, with so much groundwork done by H. B. Thomas and Sir John Gray,
anyone interested in the history of the early days in Uganda and East Africa has
a firm foundation from which to start, but it must have been hard work for
the pioneers who had to dig and delve into records, publications and often the
tenuous memories of their friends to set down the record which we have today.
After nearly 30 years' service in Uganda H. B. Thomas retired in 1940, but
remained a zealous worker for the Uganda Society, not only in arranging for
the printing of the Journal, but as a buyer for the Library. Many valuable books
would not be available to members today if, during the last 20 years, he had not
picked them up at second-hand bookstalls in Britain.
The Church of Uganda owes a great debt to H. B. Thomas, who was always a
strong supporter of St. John's in Entebbe. He still serves on the Church
Missionary Society Committee in London.


Edited by Sir John Gray

[These extracts from Die Tagebiicher von Dr. Emin Pascha, edited Dr. Franz
Stuhlmann, vols. i, ii, iii, iv, and vi (Braunschweig: Watermann, 1916-27), have
been translated and provided with introductory notes and comments by Sir John
Gray. They are planned to appear in The Uganda Journal as a series covering
Emin's first visit to Buganda in 1876, his visit to Bunyoro in 1877 and his
second visit to Buganda in the same year, followed by such portions of his
later diaries as are relevant to Emin's contacts with the Uganda region during
the years spent as Governor of Equatoria until his withdrawal in 1889. The
first two extracts appeared at pages 1 and 149 respectively of Uganda 1., 25
(1961), while the third extract may be found at page 72 of Uganda 1., 26 (1962).


Introductory Note

THE period 1878-1881 covers the first three years during which Emin was in
charge of the Equatorial Province. There had been frequent changes in that
governorship between 1876 and 1878. When Gordon left for England at the
end of 1876, he handed over the charge of the Province to Colonel H. C. Prout,
an American Officer in the service of the Khedive. Prout in his turn was
succeeded in June 1877 by another American officer, Colonel A. M. Mason.
Within two months of his appointment Mason was recalled to proceed on duty
to Ethiopia. Gordon, who had by this time returned as Governor-General of
the Sudan, appointed Ibrahim Fauzi to fill the post. Within a very few months
grave charges were brought against him and Gordon relieved him of his post.
Finally, in June 1878, as the explorer, Wilhelm Junker, tells us, "some one else had
to be found to fill the post of Chief Mudir of the Equatorial Province . .
On Gordon's asking me to suggest some one, I proposed Dr. Emin Effendi.
Gordon certainly raised objections, but in the end agreed with me." (Junker-
Travels in Africa, 1875-1878 p. 513.)
Gordon's doubts as to Emin's suitability for the post would appear to have
been to some extent justified. Both he and Emin had most decided views on
matters concerning the administration of the province and they did not always
see eye to eye. When differences of opinion arose, Emin would be extremely
intransigent and showed himself decidedly disinclined to carry out superior
orders. At the end of 1878 Gordon had decided that the stations at Mruli,
Foweira, Magungo, Kirota, Padibe and Masindi should be evacuated and that
Dufile should be regarded as the most southerly point of Egyptian administration
(Zaghi-Gordon, Gessi e la Riconquista del Sudan p. 424). As Emin informed
Mason in a letter which he wrote on 16 September 1879, "when Gordon
nominated me for here, I wrote him a letter which was hardly ceremonious, and

by reason thereof such strong differences arose between us that he forbade me
to go beyond Dufile-from fear perhaps that I should go off to Kabarega or
elsewhere." (Stuhlmann-Die Tagebiicher von Dr. Emin Pascha ii. 54). It would
even appear that Emin went so far as to declare that Gordon's order was
insensatee" (Zaghi-loc. cit). In any event his constant procrastination showed
that he was not ready to carry out superior orders.
In the circumstances it was not surprising that Gordon decided to send Romolo
Gessi, a veteran of the Crimean War to supersede Emin and see that the
evacuation was carried out, Emin thereafter decided that he had better eat humble
pie and on 20 July 1879, wrote to Gessi asking him to forward two petitions to
Gordon in answer to the charges which Gordon was making against him. "As
for the charges themselves, I have no need to answer them, but I request
Your Excellency (sc. Gessi) to look into them and to take such steps as appear
to you to be necessary. The men here being such as they are and as I am unable
to go any distance from here without fear of fresh intrigues or even a revolt, it
is more than difficult for me to maintain an efficient administration." (Zaghi-
op. cit. pp. 405).
It would appear that it was on Gessi's representations that Gordon decided
to modify his original instructions to Emin. In his letter to Mason of 16 Septem-
ber 1879, Emin wrote. "Judge for yourself. Today, after having reviewed matters
and seen that for a whole year I had achieved the impossible in maintaining order
here, despite our isolation, he (Gordon) has graciously revoked his order and
I am now about to make my way to the lake (Albert), where I will arrange for a
new station to be built at Mahagi." (Stuhlmann loc. cit.)
In 1880 Gordon resigned the post of Governor-General and was succeeded
by Muhammad Rauf Pasha, who gave orders for certain of the previously
evacuated Egyptian posts to be reoccupied and also reinstated Emin as Governor
of the Equatorial Province.
The rights and wrongs of Emin's dispute with Gordon need not be discussed
here at any length. In his letters and his diary Emin sets out the case for himself,
but it is only right to point out that there is another side to the story. The troops
upon which he had to rely were ill-disciplined, poor in quality and for the most
part badly officered. His military posts were thinly manned and widely scattered
with the result that their foothold was precarious. In the circumstances no
prudent military commander could be blamed for drawing his thin and far
flung outposts closer to his main body. On the other hand, Emin was clearly
much to blame for not being ready to comply with the orders of his superior
After his reinstatement Emin spent the little time that was left to him before
the outbreak of the Mahdist revolt in re-establishing military posts in Acholiland
and in establishing new stations to the west of the Albert Nile. Except for the
station on the south bank of the Victoria Nile at Magungo, no attempt was
made to reoccupy the Bunyoro bridgehead. Though he can hardly have realized it
at the time, the post which he established on the bank of the Albert Nile in the
territory of the Alur chief Fishwa, the son of Lai, whom the Sudanese called
Wat el Lai, was to become a very important strategic point for Emin a few
years later.
Emin's correspondence and diaries for the years 1878 to 1881 contain a
number of references to C.M.S. missionaries who, from time to time, crossed
his path. He conceived a great liking for Robert Felkin, who was a professional
brother who later obtained a German as well as an Edinburgh qualification. For

more than ten years after their first meeting the two maintained a long and
friendly correspondence.
Nevertheless Emin, and for that matter Gordon and Gessi, were extremely
critical of the missionaries' attempts to evangelize Mutesa and the people of
Buganda. Their view was that Stanley's picture of a field ripe for the harvest
was painting the lily, and was a wasteful and ridiculous excess, which came as
a rude shock to the early labourers in that field. Emin and the other critics
were undoubtedly right in foretelling that on their arrival on the spot, the
missionaries would quickly learn that the picture painted by Stanley was a great
illusion. But the convert from Christianity to Islam was wrong in disparaging
the missionaries in other respects. No doubt, when they first met Emin at Lado,
they were mere tenderfoots meeting one who, compared with them, was an old
timer. No doubt also, self-confidence at times led to their ignoring the old timer's
advice in regard to matters affecting their material needs. But they had other
ideals and other enthusiasm, which Emin was clearly not able to appreciate.
On Felkin's return to Lado after only a brief sojourn in Buganda, Emin
announced with evident satisfaction that the Mission to Buganda had proved
the failure that he had always prophesied that it would be, but history shows that
he was wrong! Whatever may have been their outward shortcomings, Emin had
underrated the spirit which had carried the labourers into the mission field.
In peril often, in weariness and painfulness, in watching often, those labourers
in course of time won through to make Uganda one of the brightest jewels in
the Mission's crown. Only a very few years later Emin was to discover that the
nearest friend to him in the time of need was a member of the Mission, which
he had once derided and declared to be a failure.


Extracts from Emin's Diaries

October 8, 1878.... (George Litchfieldl, R. W. Felkin2 and C. W. Pearson3,
C.M.S. missionaries, arrived by steamer at Bor. As the next entry shows, their
voyage from Fashoda to Lado had been seriously impeded by the sudd. They left
Fashoda on 23 August, 1878, and, all being well, should have reached Lado in
twelve or thirteen days, but at times they were brought completely to a standstill.
Having heard of their plight, Emin set out by steamer to rescue them, and met
them at Bor. He and the missionaries reached Lado on 9 October-(Wilson and
Felkin-Uganda and the Egyptian Soudan i. 289-299).)
October 18.... (Pearson supplied Emin with a map of the river blockages which
his party had encountered.) In the evening comedy and visit by Mohammed
Effendi el Nickeli. Prudery of the English.4
October 25 and 26.... Purposelessness of the very irresolute English, who want
to go to Kabarega before Mutesa and thereby wish to block their own road.
October 31. ... I do not rightly understand what will become of the Uganda
Mission and I doubt if the gentlemen themselves know what they want. This is
one more piece of English Don Quixotery.
November 2.... My guests have been for a walk.
November 4.... (At Rejaf) Billeted my guests.
November 7 .... 6.45 in the forenoon depart from Rejaf, where the gentlemen
were to stay till Monday (November 11). Pearson is pleased with his own knowledge.
Felkin is very pleasant, but very young. Litchfield is unapproachable by me.

November 8.... So all my trouble has been to no purpose. Kabarega's people
have been attacked from the station at Kisuga and many of them killed; also some
of our soldiers have been killed or taken prisoners. My letters to Kabarega have
been rejected by him and all relations with him have been broken off with the
result that the existence of our stations in the Sudan once more becomes a matter
of doubt. And all this in defiance of my orders. My hands are tied by Gordon's
inappropriate order that I am not to go further south than Dufile.5 Furthermore
Kabarega has proved his good faith, in as much as he has sent the soldiers, who
were in his hands, back to Magungo without their arms instead of killing them,
as Mutesa would probably have done. I am quite superfluous here.
November 11 .... (Emin sent books to Felkin and Pearson.)
November 13.... Letter from Felkin. It would have been better and more
pleasant if he had stayed with me.
December 5 .... (Emin received letters from the missionaries.)
December 22 .... (Report received that C. T. Wilson of the C.M.S. had
proceeded from Buganda to Foweira in order to meet his colleagues.)
December 24.... (At Dufile.) Murjan Aga, an old Mexican officer and now
governor of Magungo6, was here to receive me. The site of the new station in
the Alur country was settled upon. The place is called Mahagi.
December 27.... After a short march we reached the bend of the Khor
Unyama and immediately afterwards the remains of the former zeriba Jeifi. It
was deserted by its inhabitants a short time ago, because, as they told me, the
fields had "grown old", that is to say, become exhausted. In the middle, among
the ruins of the houses, there rose a magnificent tamarind tree, under which I
had passed the night three years before on my first journey to Buganda. In the
course of conversation with the negroes, who had come in from the surrounding
country, many of whom I had previously known, some of them offered to act as
guides for me. I therefore decided to continue my journey to Patiko by the
somewhat longer, but nevertheless the more interesting way past Paloro and Pabo.
December 28.... Our night was disturbed by legions of mice and concerts given
by hyenas. Our porters arrived early in the morning headed by Chua, the chief
of the neighboring village of Panyoro. We marched in a south westerly direction.
Passing through fields from which sesame and the second crop of durrah had been
recently gathered, we reached the zeriba Panyoro, an imposing collection of
houses with a very dense population. Panyoro is not in the place which Speke
fixed as the Dongolese station. We changed some of our porters here. (Continuing
his journey through hilly country which he describes in detail, Emin eventually
reached Paloro, "an imposing village enclosed in a high stockade".)
The chiefs of the place are two brothers. Ever since the occupation of the place
by the Dongolese, they have borne the nicknames of Dabbe (hyena) and Bu
Hussein (fox). Dabbe at once came to welcome me and brought two small
elephant's tusks as a present. They received their presents in return and promised
porters for the following morning. Their mother, an aged lady, who seemed to
be held in great consideration appeared late to ask for some beads and bars of
copper. She brought as a present some beautiful white flour. The present Paloro
is not in the place which Speke fixed by astronomical observation as the Dongolese
station, but lies several more kilometres to the north-east.
Paloro is a large village with many inhabitants, well built houses and numerous
corn stores, which indicate an abundance of cereals. Between the houses there
are small open spaces or tobacco plantations. The houses are either shaped like
mushrooms or else are hemispherical. If there is an outer wall, it is made of wattle


and daub. Sometimes an open colonnade, which is formed by the projecting roof,
surrounds the house. The doors are so low that the people always have to crawl
through them. They consist of mats of bast supported by a strong stake and are
closed when the occupant goes out. A slanting sort of couch, which is made of pieces
of wood and is supported by four legs, occupies a part of every house. The corn
stores consist of baskets made of split bamboos, which are neatly woven together.
They stand on three or four legs and are covered by conical roofs. They are often
strengthened with clay and whitewashed.
Here and there between the houses may be seen votive trees. They have
suspended from them skulls, antlers' horns and teeth of various animals.
The inhabitants of Paloro belong to the Madi tribe. They are a fine race of
men. Most of them are above middle height, of a light chocolate colour and well
developed muscles. They have a projecting under jaw, a thick growth of hair and
large flat feet. The men specially delight in fantastically dressing their hair.
In contrast to the men, the women wear very little clothing. A tail of twisted
cotton threads is attached to a string of glass or iron beads and put round the
waist. A married woman is distinguished from a girl by a covering in front which
is no broader than a band. The girls wear nothing except a string of beads round
the waist. The women are nevertheless not without their ornaments in the shape
of beads. They often pierce under the underlip so as to insert a small brass rod.
Scattered about the houses there are a number of buildings, which are raised
from the ground and are very much like large granaries. They have in front an
oval doorway and are plastered with clay. There is usually a bench constructed of
pieces of wood in front of them, which is intended to make entrance more easy.
The girls sleep in these huts as soon as the signs of puberty arrive. Boys who
have reached puberty have free access to them. If a girl becomes pregnant, the
young man who has been her companion is compelled to marry her and to pay
her father the bride price.
December 29.... We arrived at about noon at Pabo. This Madi village is not
identical with the Dongolese station of the same name visited by Linant7, but is
situated on the other side of the Khor Asi. According to the statement of the
natives, the old settlement lay about half an hour's journey farther to the south-west.
Here also we were very kindly received, but I was somewhat surprised that at this
place, as well as at Paloro, I was asked to request the government to send back the
Dongolese. The zeriba is similar to that at Paloro, but is less densely populated, as
most of the people live in two nearby villages.
December 30.... We set out again at six o'clock in the morning. At about
4.30 p.m. we reached our station, Patiko, formerly Baker's headquarters in the
Acholi district.
December 31. (Emin was visited by two sons of Rwot Ochama, whom he calls
Shuli and Gimoro, who informed him that their father wished to meet him, but
was afraid to come to the station.)
January 2 and 3, 1879. In Baker's time the Acholi chief Rwot Ochama
(Baker's Rot Jama) came to him, and it was mainly thanks to his influence that
the Acholi had such friendly relations with the government. Nevertheless, Taib
Bay, the Commander at Patiko, thought it fit on account of some small difference
to give him a terrible beating and put him in chains. After that Rwot Ochama
avoided our station and was never seen in the presence of a Governor. To our
very great surprise today there came a request to visit him. He had come to a
place about two hours from the zeriba but could not come here, because once his


blood had flowed here. He has grown old, but is still head chief. I promised to
visit him in the early morning.8
January 4.... Come at last to a large khor in a deeply sunk bed. On its further
bank stood the village of Payera, Rwot Ochama's headquarters. A guard of
honour waited for us, consisting of about fifteen well clothed dragomen of the
chief, armed with old guns. He himself was standing on the side in the middle of
some skin-clad negroes, who were freshly painted red and were waiting for me.
I was now asked to wait a moment until two goats, which had been brought by
us, had been slaughtered and the blood sprinkled on our path. Then the chief
came over the blood to greet me with a handshake.9
He then led me for about three minutes out of the village to where an ankareb
had been erected for him under a tree, whilst my stool was placed nearby under
the shade of another tree. Two dragomen stood close to the chief, guns in hand.
A young man, who appeared to be a familiar friend, reclined on the ground in
front of him. About two hundred and fifty negroes were gathered in a circle
amongst the houses. The coloured clothes of the armed dragomen in the midst
of the skin-clad and iron-adorned negroes made quite a stately appearance in the
sun's rays. Here and there a woman could be seen with her offspring.
After the customary greeting I turned to Rwot Ochama. He is already quite
old, but he knows how to hold himself. He is of middle size and has a peculiar
side glance. He was painted entirely red. He wore iron ornaments and an antelope
skin as a covering and was clean shaven. I gave him my presents which very much
pleased him. I asked him to forget the past and thanked him for the assistance
which his people had given to us, and I asked him to put aside all mistrust. He
thanked me for my visit and asked me to repeat it at a village which is six miles
further to the east on this side of the Khor Assa (River Aswa). He assured me
that his negroes would always submit to the wishes of the Government and asked
me to accept a few presents from him.
Next began the official discussion. After that I was asked for a gun. To this I
agreed. I had obtained a second-hand one at Patiko. Then a man complained of
oppression by our people. They had recently demanded a half basket of corn (five
caravans) but had taken a full basket (ten caravans). For the sake of general good
will I was able to provide a remedy in this matter, and I accordingly reduced the
contribution to the old measure.
I also gave the station chief, who was present, the necessary orders and bade him
to keep on good terms with the people; otherwise I would remove him. Whilst
Rwot Ochama was chatting with the native chief and indulging in milk and
merissa (beer), I went for a walk through the village.
As rain was threatening and as business had terminated in the best possible terms
of friendship, I then took my departure. We reached Patiko in pouring rain.
January 6. (Left Patiko on return journey by a different route to Dufile, where
Emin arrived on January 8.)
January 8.... The English missionaries have left Magungo, but their dragoman
is sick and remains with Felkin at Kirota. The gentlemen have been strongly
intimidated because of a good deal of talk about Kabarega.
January 9.... My letters to Kabarega appear always to remain at Magungo.
On my arrival there I will inflict heavy punishment. Quite naturally what I had
been at pains to devise has long ago been upset and destroyed through the
negligence of the officers.
January 12. (Letters received from the C.M.S. missionaries, R. W. Felkin
and C. T. Wilson at Foweira. The Syrian dragoman Nicola has died at Kirota.


Litchfield is suffering from fever.)
(The missionaries had left Dufile by steamer on December 21, 1878, and had
reached Magungo on December 23. Thence they set out on December 28 by
land for Kirota, where they arrived on December 31. On the way to Kirota they
were attacked by Banyoro, who, however, were driven off by their military escort.
Wilson had met the party on January 2, 1879, at Kisuna between Kirota and
Foweira. The party reached Foweira on January 7. Subsequently they made their
way to Mruli and thence to Buganda-Wilson and Felkin op. cit. i. 267-336.)
(To return to Emin himself, he proceeded from Dufile to his headquarters at
Lado (19 January) and would appear to have ceased for the time being from
keeping a diary. On April 13, 1879, he wrote a letter to August Petermann
enclosing some meteorological observations compiled by Felkin, who was at that
time in Buganda. On May 21 Felkin set out from Buganda to return to Khartum.
He reached Lado on July 23 and having been joined by Wilson on 19 August,
they remained there as Emin's guests until September 18. (Wilson and
Felkin-op. cit ii. 26, 85). The next two letters from Emin contain his views on
the affairs in Buganda derived from the information supplied him by Felkin.
July 28, 1879. (Emin to Colonel Gordon.) The day before yesterday (sic) there
arrived Mr. Felkin, a medical man of the English Church Mission in Uganda, and
at this moment I received news that Mr. Wilson of the same Mission has reached
Mruli on his way to England, accompanied by some of Mutesa's men bound for
England and certain others bound for Khartum.
Please inform the Governor-General so that he may inform England.
The Mission is, as it would appear, a failure. I have received letters in Arabic
from my Arab friends at Rubaga, which have caused me much surprise. Well we
shall see how it ends.
(Correspondence... relative to the Slave Trade-Slave Trade No. 1 (1881)
p. 42).
Same date. (Emin to August Petermann.) I have much pleasure in sending a
route map of the road from Rubaga, Mutesa's capital, to Nanfumbambi.10 I have
drawn this from the drawings of my friend, Mr. R. W. Felkin, who is a medical
member of the English Church Missionary Society, Uganda Mission, and who
fortunately arrived here two days ago (sic) from Uganda.
With reference to the said Uganda Mission unfortunately I can tell you nothing
which is very edifying. All that I told the gentlemen at the time of their
arrival here about Uganda appears unfortunately to have happened to them. At
first Mutesa was friendly and obliging, but after he had received his presents he
became cooler. My Arab friends in Uganda inform me that at the very same time
official letters reached Mutesa from Zanzibar, which, as they made out, proved
the missionaries to be liars.11 For a long time Mutesa meditated sending the
gentlemen as prisoners to Zanzibar. At the time when the French Mission reached
Uganda12 the matter had not been settled. Mr. Felkin is now here. I hope
Mr. Wilson will arrive in a few days. Mr. Copplestone and Mr. Stokes13 have
left for the south. The result now is that only Messrs. Pearson and Litchfield are
now in Buganda. The information of these gentlemen, which one day must obtain
full publicity, is the same as what I experienced in Buganda. It all throws a peculiar
light on Stanley's missionary garden of Eden.
Colonel Grant's denunciation of Egypt and his preference for the Zanzibar
route to Uganda14 has received a heavy blow owing to recent events and King
Mirambo's amiability.15
Just now (28th July at 10 o'clock in the morning) I have received the news that

fortunately Mr. Wilson has reached Mruli and proposes to arrive here on his way
to England, Stuhlmann-op. cit. ii, 50).
(The diary then continues.)
August 11. .. (Reports state that Wilson has reached Patiko.)
August 19.... At three o'clock in the afternoon Mr. Wilson arrived with six
Buganda people, of whom one is Kitaka, my former servant in Buganda.16
August 25... A curiosity. The Katikiro wanted to send me a pretty girl by
Mr. Wilson. Then he mistrusted Mr. Wilson and said he would send her later
with Kanyambo.
September 18.... Departure of Messrs Felkin and Wilson for Khartoum.
October 24.... A mail has arrived from Gessi Pasha, who has now come back.
He is to take over this province from me. I do not understand Gordon. Ruyonga
and Anfina will now be slaughtered by Kabarega and Mutesa. That is the reward
for their service. And the missionaries ?
October 24.... Mail arrived from Mruli with no news from the missionaries
(sc. in Buganda).
October 31.... I learn the following as the result of a conversation with my
Banyoro and Acholi people. Rwot Ochama the acknowledged chief of all Acholi,
belongs to the Babito and is related to Kabarega, Mutesa etc. (Speke talks of the
frequent communication between Kamurasi and the Acholi chiefs). Kitwara or
Kitare was the land of the Bacwesi. An immigration came from NNE. When the
immigration reached Acholi land, one of the four brothers, who were the chiefs,
stayed behind and founded the dynasty, to which Rwot Ochama belongs. Three
brothers went on. One took Bunyoro, one Buganda, and the third Butenga, a
land to the east and north of Busoga, where his dynasty rules to this day. Every
chief takes the title of Gabula or Garba. The inhabitants of Busoga and Butenga
like those of Bunyoro and Karagwa extract the lower front teeth-always a tie
of affinity. It is only in Buganda that this does not happen. Do all cattle folk extract
the front teeth? I think they do. The Babito immigrants took over the language
of the conquered people, but introduced foreign elements into the language. The
languages of Busoga, Butenga and Buganda are very near to one another; those
of Bunyoro and Karagwe are likewise near to one another and agree with one
another (especially in the slow drawl). It is now known that the immigration came
from the east coast, as to which the same Babito (Land of Witu) seems to make
clear. It is also equally clear because the closely agreeing Buganda, Busoga and
Butenga language take in the construction and inflection and character of the
southern languages whilst Bunyoro and Karagwe, whither most of the Bacwesi
returned, have retained more of the rough sound of the original language. Perhaps
this makes clear Mr. Wilson's hypothesis that the resemblance of the Baganda
and ancient Egyptians is due to a common descent from the east, though this
appears to me to be more doubtful. At any rate it would be interesting to compare
the numerals and pronouns of the large Kitara group with those of the Galla
and Abyssinians of the east coast.
November 14. (Left Dufile by the Khedive.)
November 16.... (Reached Wadelai at 2.45 p.m.) As the chief's village is
situated at some distance from here, I have sent to the chief or his brother and
prime minister, Gimoro, with whom I wish to negotiate to erect a zeriba in his
land. We had to wait until the messengers had been sent to him. In the meantime
a brisk traffic had sprung up on the bank, wood for the steamer, sweet potatoes,
gourds, bananas and fowls being exchanged for glass beads. We visited the little
villages without hindrance. They consist of compounds containing ten or twelve


small dome-shaped huts with entrance porches, such as are often seen in
Bunyoro. Several miniature huts are dedicated to the spirits, but they stand
empty. In the centre of the village a small roof raised on posts marks a grave
where occasionally the blood of a goat is offered to the dead. A messenger from
the chief came late in the evening with a large tusk of ivory as a present. He
had been instructed to inquire whether my intentions were good or bad. After he
had been enlightened on this point, he returned with presents both for the chief
and for himself.17
November 17.... Gimoro, the brother or foster brother of the chief Wadelai,
arrived to visit me. He was accompanied by three hundred negroes and also brought
me a tusk. He is a strong intelligent looking man. Green creepers were twined
round his shaved head. His arms were covered with pretty iron ornaments. His
dress consisted of goat skins hanging from the shoulders. After he had received
presents of beads, cloth and copper, he told me that Wadelai himself was unable
to come because he was too stout to walk. He said that the kuftan which I had
sent him the previous evening was too tight, because "when he sits, a child can
stand on his paunch". A long palaver led to a satisfactory conclusion. I received
permission to form a station here, after I had promised to keep my soldiers in
Whilst the people agreed to my request to bring wood for the steamer, I took
the opportunity of inspecting them closely. The whole district of Wadelai is called
Koich. The Acholi and Banyoro pronounce this Koshe, which has led to Koshi
appearing on Baker's map. This district forms one of the many subdivisions of the
large country of Alur. The language of the Wadelai, or rather the Koich tribe, is
The people are a handsome race, mostly of middle height, black in colour, with
a reddish-brown tinge. They have fine teeth and small feet. They wear the hides
and skins of oxen and goats, and sometimes those of small antelopes. These are
fastened over the right shoulder. Little attention was paid to dress and many of
the skins were torn, but great care was expended upon ornaments and painting.
There were all kinds of head-dresses-wigs and cowry head-dresses of the eastern
Acholi, the lofty head-dresses of the western Lango, spiral tresses and corkscrew
curls. I saw that many of them had dyed their hair red. The painting was even
more grotesque. One beauty had her legs painted grey with red stripes, and on
each cheek there was dabbed a bright red spot. The lower lips were pierced and
generally contained a long straw stalk.
Considerable intercourse exists between the Acholi on the eastern bank of the
Nile and the people here.
Somewhat after ten a.m. we continued our journey. After a slow voyage of five
hours seventeen minutes, we anchored near the foot of a chain of hills in order
to visit, another chief. Unfortunately, after we had climbed the hills to his village,
which lies behind them, all the people had fled, so a dragoman had to be sent to
them as an ambassador. I found here ten small huts, all of which were divided
by a partition into two rooms. After we had persuaded one of the fugitives to
return, he promised to call his chief. The language, dress and arms of these
people are exactly the same as those in Wadelai's country. This district is called
Paroketto. Koich is said to be six or seven hours distant and Mahagi eight hours.
November 18. ... Chief Roketto sent a refusal to meet us. He was offended
because we did not go direct to him. We continued our journey early. We reached
Magungo, where we are to make a longer stay, 2.40 p.m.
December 4.... The people have come from Kirota. Six hundred caravans of


durrah have on Murjan Aga's orders been burnt. Kabarega's people have taken
possession of the abandoned zeriba and followed the soldiers on the march without
attacking them. As would have been foreseen, Ruyonga's people have all fled or
gone over to Kabarega. Anfina has gone back to his island. I have discovered
great irregularities at Magungo. Murjan Aga has attacked a friendly village with
the result that four women and one man have been shot.
December 7.... A mail has arrived from Lado and also from Buganda (at last).
Also, the Katikiro's people have come with presents and a chief from Mutesa. In
answer to a written request to send me pots containing a few coffee plants, Mutesa
has forwarded a bundle of dry branches four to six feet in length.
December 9 .... In conversation with Kanyambo I have heard something new
about Masaba, a Kingdom N or NNE of Lake Victoria, which I cannot identify.
Wad-el-Mek has brought camels from the east (West Galla) and so proved his
former statements. He also speaks of horses in the same land. According to some
very confused notes the route lies S and then ESE. The following names appear
as rest camps-Padibeck, Labonga, Pajelle, Hellet, Mukaka-Parsheli-Lira
(mountain of same name), Amia, Loren (here Lango land begins, whence three or
four days march takes one to the camel and horse owning people).
December 12.... A violent thunderstorm delayed our departure, but at 6.43
a.m. the steamer Khedive carried us towards the lake. At 10.55 a.m. we
reached our station Muhagi, which is not situated on the point designated by
Colonel Mason as Mahagi, but is three-and-a-half to four hours north thereof.
Mahagi is really the name of a village to the south.
December 13.... Some of the neighboring chiefs arrived at this station to
pay me a visit. Like Kabarega's men, all of them were dressed in soft cowhides,
but they did not carry large sticks as the Banyoro do. They were all of medium
height but powerfully built and very black. Some of them had artistic coiffures.
Others had their heads shorn quite smooth. Their ornaments consisted of brass
and a few copper rings. They made a good impression by their modest behaviour.
They too call their whole country Lur. The names Toro (south-east corner of the
lake) and Busongora are well known to them. It is said that between this country
and Bunyoro there was formerly frequent communication and that a very active
trade was carried on. Even now the chiefs here acknowledge the supremacy of
Kabarega, ruler of Bunyoro. Communication is carried on by means of boats,
which coast along the north shore, enter the river (Nile), and cross it and then
coast along the other coast, Magungo to Kibiro, where salt and iron are bartered
for the skins of colobus monkeys.
The people, who now inhabit Londu in Kabarega's land, were purchased as
slaves from here, or a little more to the south by Kabarega's people or else the
people of his father Kamurasi, and were settled where they now are, and after-
wards gave the place the name of their own country, A-Londu.
December 14. (Visited Mason's Mahagi and then sailed north, entering the
Albert Nile.) We halted for the night at 4.55 p.m. to the south of Wadelai in the
proximity of many villages and durrah fields. Wood was immediately brought by
the natives for sale. The most remarkable, but the most useless, of the natives was
a man who carried about as a trophy a small looking glass hanging from a bent
stick. He obstructed everybody by reflecting the sun into their eyes. I had previously
given this glass to Wadelai's brother.
December 15. (Reached Dufile.)
January 1, 1880.... Illumination on New Year's Eve. Through the carelessness
of an officer, fire broke out in his hut, which was burnt down. Mail arrived from


Lado and Khartum, with letters from Perthes.18 Hartlaub, Felkin (England) and
January 4.... Preparations for departure to Patiko and Padibe. I have selected
soldiers for the new station and have inspected their outfits, weapons etc. and
arranged for new looms.
(In an article by Emin, which appears in Petermann's Mitteilungen for 1880,
Emin says-
"In order to complete some previous itineraries, I took a new route from
Dufile to Patiko and Foweira, across Jebel Labilla, which route I mentioned
elsewhere. Communications have already been opened between Patiko and our
new station Wadelai, and I received letters from there in two-and-a-half
January 15.... Soon after the withdrawal of our people Runyonga made a
raid jointly with the Lango on Kabarega.
January 16. Among the stores from Mruli and Kodj I have found a quantity of
paper and some ink, a godsend. Post came from Wadelai today; road-Patiko,
Fabo, Fagaki and thence to the river.
January 19. People came in from Kodj. (For the next week Emin was waiting
at Patiko for carriers.)
February 10. (Emin reached Lado.)
(The diary contains the following entry in March, 1880.)
March 22.... Gordon has handed in his resignation and gone to England.
Giegler20 is Vakil. In Gordon's place comes Rauf Pasha21 an enemy of all Euro-
peans. This means a complete reversal of the system and naturally the first thing
to follow is our recall from here.
(Once again there are no entries in Emin's diary which relate to Uganda for
several months. The silence is only broken by a letter which Emin wrote to
Professor Behn on August 19, 1880, which contains the following information.)
"Lado, August 19, 1880.... Now that I am again in possession of paper, I will
follow your advice and write my work on Uganda. Stanley, for whom I have the
highest respect, writes at length on Uganda so much that is false that it is time
that the actual facts were set out. It so happens that I have received a mail from
Uganda, wherein I see amongst other things that on 23rd December, 1879, Mutesa
and his chiefs in most solemn session have arrived at the resolution to forbid
teaching by the French and English missionaries and to punish with death those of
the natives who are instructed by them. A simultaneous decree enacts that the
religion of the white man as well as the Mohammedan religion are forbidden and
his subjects must abide in the faith and creed of their forefathers. That decision
was received with much pleasure by the assembled chiefs, whilst the king ordered
his soldiers outside to fire a salute in honour of the event. The assembly was of
the view that no instruction is required, but they want as many flints, powder
and percussion caps as there is grass."22
"The latest letter in reply from Uganda is one of the 1st June and is completely
despondent about the Mission's case. "Mutesa," it says, "is not dealing well with
us and I think we must give up the mission. The four French missionaries are of
the same opinion as the English that nothing can be put to rights here. The king
is as ferocious as can be. He has killed over two hundred poor persons at the
graves of his ancestors and pays no heed to our words."
"Mr. Litchfield (a member of the English Mission at Rubaga) is ill and has
gone to Mpwapwa to Dr. Baxter because the road has been made impossible for
him to come to me. His colleague Pearson is still at Rubaga, but he believes that


no mission work is possible-just as I told the gentlemen when they came here.
All that Stanley said to the contrary is humbug. I personally am on very friendly
terms with Mutesa, despite his lies and his hypocracy. His Katikiro is my especial
friend and the only gentleman in Uganda."
"You know that Gordon Pasha, in his capacity of Governor-General, withdrew
all the Egyptian posts in the south. Rauf Pasha was appointed as his successor
and we all expected our recall. On the contrary, however, not only am I established
here, but my province has been enlarged and will be separated from Gessi's
province. Our boundary is at Shambe on the Nile (about 7V 10' N. latitude) and
on a line running east and west. What lies to the south belongs to me. I have also
received permission to set up new stations, as always appeared to me to be proper.
For that purpose I am going south in a very short time. I think the Somerset River
will make a good boundary, but the Mwutan belongs to my district. Shortly I am
going to Makraka so that I can push out from Wadelai's land (circa 20 40'
latitude on the Nile), but news of the arrival of a steamer has made it necessary
for me to hurry back to Lado."-Petermann's Mitteilungen (1880) p. 472).
(At this date the following entry appears in Emin's diary.)
From August 17th to 22nd (Sunday inclusive)-Heavy mail. I am now made
independent of the Bahr-el-Ghazal and have received permission to establish
stations. Now at last we can make some progress.
(Emin directed his attention first of all to the establishment of stations in the
Latuka country.)
October 7. (At Kerefi in the Latuka country.) I was told of a curious belief,
which was said to be held in a village not far from here. The inhabitants are said
to have the power of turning themselves at night into leopards and of killing and
eating men. I remember hearing of a similar belief in Bunyoro and on the Blue
Nile the legend of hyena men is absolutely believed.
October 16. (Emin reached Agoro in the Acholi country where he had already
established a military post in the previous year.)
October 17.... The march from Agoro to Padibe is accomplished by the natives,
and also by our own men, when they have no baggage, in one day. We spent
exactly eleven hours fifty-five minutes over it by marching rapidly.... After
passing chief Ogwok's village23, we reached the station of Padibe just in time to
escape a downpour of rain.
This station, which formerly was very flourishing, was given up at the time of
the wholesale abandonment of stations last year. It is now firmly held for us by
the Acholi chief Ogwok, who has repeatedly requested that a station occupied by
soldiers should be erected near him and had brought ivory to Patiko as a present.
His wishes have been readily complied with, as the new Governor-General has
no objection to the erection of stations, provided they cover expenses. This new
station is prospering very well and promises to become one of the finest in the
Chief Ogwok has become a perfect Dongolese in clothing and in manners. He
speaks fairly good Arabic. He sits and sleeps on an ankareb and entertains his
guests with coffee. Nevertheless this does not prevent his many wives and children
from appearing in their national costume, that is to say, almost entirely naked,
except for some short cotton aprons.
October 21.... As soon as he heard of my arrival at Padibe, the chief of all the
Acholi sent his son to invite me to visit him, as owing to illness, he was prevented
from visiting me. So our march turned in that direction.
Passing the large hamlet of Kyatangura, we entered the district of Labongo.


We had still to pass through several extensive hamlets before reaching our night's
quarters, which were situated beside a pool containing rain water all the year
round, and which, like the village, is named Diendi. The chief came to greet me
at every hamlet we passed. Generally speaking, he was dressed in a long coloured
shirt and a tarbush and was accompanied by a motley crowd of men and women.
The Acholi are very polite people. They are always ready with greetings and
inquiries after one's health, but as the national custom is to raise the visitor's arms
four or five times above his head, the repetition of the ceremony becomes rather
October 22.... After crossing a range of hills, the highest of which was the
cone-shaped Nyone, we left the district of Labongo and soon arrived at the
Khor Bagger (River Pager). After a short rest on the southern bank the march
was continued. On reaching the village of Biayo, where Rwot Ochama resides,
we found that he had thoughtfully had some good huts erected for us; but the old
gentleman was very indignant when he heard that I intended to cross Khor Assa
(River Aswa) that same day. There was a long debate, in which even his wives
took part, and an appeal was made to me for old acquaintance' sake. At last I
yielded to his importunity. His men were at once ordered to bring wood and water
as well as a goat for myself and several pots of merissa for my men. I made him
a present of cloths, glass beads, copper and a bottle of wine in return for his
kindness, and this increased his good humour. Rwot Ochama has aged, but he is
still the faithful and amiable old friend I had previously known him to be. He
has no power at all over his people. They make fun of his very long winded
orations, which he is fond of making on every possible occasion. His son is very
young and unable to support his authority, but this is certainly done by his first,
old and ugly wife, who can only be distinguished from her inferiors by a rather
longer tail. She was most obliging.
October 23.... Great men do not make great haste. We could not set off until
friend Rwot Ochama had finished his sleep. After much difficulty and trouble he
then persuaded his men to carry our loads. We reached the village of Odiak, where
Chief Rwot Ochama had once more thoughtfully had huts erected for us. This
village lies in sight of Payera, which we had visited before. The chief of this
place kindly had a feast prepared for my men in the evening. Large quantities of
beer and sweet potatoes were supplied by the villagers. They demanded in return
a share of our men's meat ration. Dancing and singing kept the village awake until
the early morning.
October 24. (Emin reached Patiko.)
October 29. (Passed by Koich, which was temporarily deserted by its inhabitants,
and halted for the night at Koro, "which was exceedingly dirty".)
October 30.... We met and overtook a caravan of Ruyonga's people, who had
bought ivory in Acholiland. I counted six pieces. They probably were going to sell
them to Mutesa for women, cattle and brass.
October 31 .... People with mail from Foweira met us and told us that com-
munication with Buganda was closed by Kabarega's people.
November 1.... We soon reached Foweira, which is in process of renovation.
The men told off to reoccupy it have only been there fourteen days. Ruyonga, the
chief of the district, was little altered. Perhaps he was rather more stupid than he
used to be owing to his indulgence in mwenge and spirits.
November 2... The road from here to Koich is much frequented. Hence one
would think it would be possible to reach Kijaja and Anfina's village without
molestation from Kabarega's people, but we would have to leave behind Anfina's


goods. More Lango chiefs have submitted to the government and have voluntarily
agreed to a tax on corn.
November 4 ... (Ruyonga visited Emin, but said he was sick.)
November 5.... Ruyonga, who, as I supposed, despite expressions to the con-
trary, is in communication with Kabarega and Mutesa, promised me the day before
yesterday that he would send people to Mruli and leave some supplies for me
with Kabarega's people there so that I may send an embassy to Kabarega. Today
two men came, called Bateka and Ugalla, the latter of whom squints. After I had
explained to them that the soldiers are not here to make a raid on Kabarega's
land, but only to occupy Mruli, so as to have a point of connection with the ivory
trade, I gave them a present for Kabarega and myself asked them to cause
Kabarega to send people to me, so that I may come to an understanding with him.
At the same time I willingly declared that I would visit him personally, if he so
wished, and asked for an early reply. The people promised this. If the people
really are from Kabarega and this is not a joke on the part of Runyonga, who I
believe to be quite capable of it, I can get an answer in ten or twelve days and
go back with the steamer expedition to Kabarega by way of Magungo and Kibiro,
and make blood brotherhood with him.
November 6.... It appears that I did not make a mistake about Runyonga.
After he had asked for a girl prior to my departure, which request was naturally
refused, he sent me his dragoman to let me know "that the soldiers here only wish,
if Suliman Aga is to be their chief (who has made him a present of half the
magazine and beaten the soldiers to death!) he does not want any soldiers and I
may remove myself". I thought this was one of the consequences of the acts of
my predecessors. At the same time as he has sent people to Khartum to ask for
soldiers, he has sent people to Kabarega with presents and negotiated a definite
friendship with him. Kabarega has agreed to it, and sent people to Ruyonga and
informed him, that he will readily live in friendship with him and forget all the
past, but Ruyonga must bring no soldiers into the land. Now it so happens that,
just as Ruyonga's people came back with the presents and weapons which have
been sent to him by the* Government from Khartum, Kabarega's people were on the
spot and heard that once more the soldiers were coming. They at once set out
to tell Kabarega and Ruyonga felt awkward in his double role and wanted very
much to be rid of the soldiers, to whom he had never given any corn. I have now
sent Suliman Aga to him and proposed to him that the soldiers shall go back from
here to Foweira; that sufficient soldiers shall remain in Foweira with an officer
so as not to be exposed, and the rest of the soldiers shall go away. If he agrees
to this I will at once depart, if not, it will be clear that I must draw the inference,
because I hold him capable of it, that he wants to sell us to Kabarega.
November 7. ... As I expected, my threat to withdraw the soldiers and to
proceed myself to Anfina and Kabarega, so as to come to an understanding with
him about Ruyonga, has had the desired result. Ruyonga's people have come to
me. They came here at midnight and told me Ruyonga had never said or caused
it to be said that he wanted no soldiers, but only that I must change the commander
(a palpable lie!). He wants everything to remain as it was of old, and he will be
friendly with any chief I give him. After I had let them make their requests for
a little time, I agreed to leave the soldiers here for the time being, until I had seen
Anfina and found a suitable place for a new station, when I would take the
majority of the people from here.
(Emin received a visit from Ruyonga's sister Njakatscupe.)24


November 9. (Emin was visited by Ruyonga's son, Kamisoa, who afterwards
succeeded his father.)
November 10. (Emin left Foweira and reached Deang.)
November 11 ... On our arrival at Anfina's (Panyatoli), his men were dressed
in festive attire and drawn up in ranks, waving Egyptian banners. They received
us with a volley of guns. Their commander, Anfina, was dressed in English flannel.
He did me the honour of conducting me into his enclosure. Here a fine large hut,
built in Kiganda style, was assigned to me. The men encamped outside the fence.
Presents were immediately brought to me. They consisted of sweet potatoes, fowls,
eggs, flour, bananas (both ripe and unripe) six elephants' tusks and a goat. I was
at a loss to know how to repay this generosity. Amongst the great chiefs in the
south one cannot get away with glass beads and other gewgaws of this kind.
Nevertheless my present must have pleased Anfina, because a large calabash vessel
full of banana wine was then sent to me.
Anfina is the only negro gentleman I have known, except Mutesa's prime
minister, the Katikiro.
Immediately after my arrival at Anfina's, caravans of porters were sent in
by his subchiefs with presents for the support of his guests. This is a custom of
the Babito. Anfina is the only negro prince whom I have known, not excepting
Mutesa to whom clothing and such other articles as have found their way here
have become indispensable. Moreover, he is the only chief who uses dishes and
plates for the purposes of eating and glasses for drinking. Soon after evening set
in dancing began. One of Ruyonga's men mimicked me in a most amusing fashion
by inquiring as to the names of mountains, rivers, plants, etc. and taking down
notes as well as making observations with the compass. "What is the name of this
village?" chanted the soloist and then himself gave the answer "Kijaja", whereupon
the chorus took up the refrain, repeating the words "Kijaja" ten times. Linant's
adventures with Kabarega's men at the River Kafu were also illustrated in the
dances. This orgy lasted till three o'clock in the morning, but everybody was astir
again at five o'clock.
November 12. Murjan Aga came from Foweira towards evening. I had sent for
him to meet Anfina and to hear in the chief's presence my orders regarding the
purchase of ivory, relations with the negroes, the setting up of new stations etc.
After this my departure could not be thought to be far off, had not Anfina asked
me to stay for a couple of days so that the Lango chiefs might come and he could
thereby obtain greater credit with them by reason of his good relations with me.
November 15 .. Immediately after leaving Panaytoli the descent to the river
began. After a short journey of not quite four hours we reached the river, which
here flows in a loop from ENE to WSW. There are no rapids at this place-
Mutua in the district of Podi. The river is only about five hundred feet broad, but
there is no village here and only two boats are available for the passage.
:November 16.... Leaving our men to ship our few cattle and goats, we took
our baggage and turned a little further to the west, where there were said to be
five boats. On the way we passed the spot where Linant's tent stood when he visited
Anfina. The island of Mukono (the arm), on which Anfina was then living, lies a
little to the west and not very far from this place. It is now uninhabited and a
good deal of it has been washed away. Keeping close to the river, which here is
very pretty, we reached the ferry of Aweri and were soon sending off the
baggage and porters. Of course there were only two boats to be found here and
the Mutongole, who had been instructed to help us, preferred drinking mwenge in
his village and left us to do the work alone. We crossed the river in seven minutes,


for our rowers were really good and the boat was light. The steep northern bank
is well wooded and there are numerous Lango zeribas. We found our men waiting
in the little village of Mukomere, which is inhabited by Jopaluo. We then
continued our journey in a north-easterly direction, intending to reach the road
to Patiko at Ras-el-Fil. The few huts that we saw were deserted by their
occupants. The trees were hung with cylindrical hives of bark.
At noon we reached a rather extensive group of zeribas, one of which was
kindly given up to us at once. The houses of this village of Pachora were very
much like those of the Banyoro, being divided into compartments by cane parti-
tions, but the granaries were quite different, being oven shaped. Towards evening
a large gathering of Lango collected. They had never seen a white man before.
They were specially interested in my leather leggings, which reached up to my
knees and which they called "elephant's legs".
November 17.... We reached Modo, our old night quarters. Here the water,
which was always scanty enough, had been drunk by the elephants and buffaloes.
So we had to go with our thirst unquenched for another march of two-and-a-
quarter hours to Ras-el-Fil. There we found water in a row of holes, which tasted
good after a march of eight hours.
November 18. (Emin proceeded from Ras-el-Fil to Patiko.)
November 28. (At Wadelai.) The small colony of Banyoro, which settled here
after the abandonment of the station at Magungo, appears to be quite flourishing.
(On this same date Emin wrote to Consul Hansal at Khartum as follows:--)
I myself have superintended the establishment of a station at Padibe, with branch
stations at Agoro, Pajuli and Patanga-this last mentioned is not erected yet-as
well as stations in the south. We have also advanced from Wadelai to Okora.
Such is the summary of my work during the present year. I am not in a position
to give you any news about Buganda. The last people who came from Mutesa
with presents and letters for me found Mruli deserted and returned immediately
to Uganda because they were afraid of meeting Kabarega's people. The letters
were then brought to me by Ruyonga's men. I am now daily expecting Kabarega's
people. As soon as they come I shall speedily re-establish my connections with
Buganda. Furthermore, the chief of Toro has offered me a free passage through
his dominions.
January 1, 1881.... When we had just opened Masindi and were cutting down
the forest and breaking up the earth for the purpose of building houses, we lost
about sixteen out of one hundred men from some febrile disease, which is said to
have had the appearance of scurvy.
(During 1881 Emin's activities were confined to work at his headquarters at
Lado and to journeys to the west of the Nile. As he informed Consul Hansal at
Khartum in a letter of May 23, 1881, he set out in April to make a tour of
inspection of the stations, which had recently been established east of the Bahr-el-
Jebel. He proceeded by way of Gondokoro into the Latuka country in the first
instance, arriving at Tarrangole on April 20, 1881. Thence he made his way
southwards into the Acholi country. The first station at which he arrived in
Acholiland was "the little station of Agoro". Emin's letter to Hansal continues as
follows: -)
The high position of this station (3,700 feet), its cold water, and moderate
temperature, render it the healthiest, most pleasant place of abode in this country.
Its surrounding mountains and hills also make it very picturesque.
From Agoro to Padibe was a hard day's march of eleven hours fifty-five minutes.
We passed by the Langia range, then crossed Khor Arenga (River Aringa), with its


beautiful "gallery" wood, and reached Jebek Lamo (Mount Lamwa), or, as it is
usually called Falogga (Paloga). We were now in the country traversed last year,
when I marked my way on the mountains, such as Lamo (Lamwa) Lalak and
Agga. Padibe has grown into a large and beautiful station, and is the centre of
numerous smaller establishments in the Acholi district, namely, Palabek, Parajok,
Pajule, Lira (?Lira Palwo), etc. The excursion to Pajule, the frontier post towards
the Lango country, which had been proposed last year and which had now become
necessary, was now in fact carried out. Ascending gradually through the district of
Labongo, we reached Khor Bagger (River Pager) which we had crossed last year
on the way from Padibe to Patiko and which may be considered the chief tributary
of the Khor Asua (River Aswa).... The station of Pajule is very unfavourably
situated in the midst of a wild, hilly savannah. It has no surroundings worth
speaking of and no outlook, but it forms a centre for the people to the south and
east, who come to barter their ivory for glass beads, copper, brass, etc. This
accounts for me finding here men from Bognia and Lirem. From them I obtained
information about their countries, which lie at a distance of eight to ten days'
journey from here. They wore the usual Lango head-dresses, and spoke Lango,
which is quite different from Acholi. Unfortunately we had to return to Padibe
by the way by which we had come ....
We turned our steps from Padibe through Madi (Madi Opei) to Parajok, on
Khor Limur. After settling some boundary disputes among the Acholi tribes, we
crossed Khor Atappi much further up than last year, and arrived at Obbo (in
Latuka country)-(Emin Pasha in Central Africa pp. 287-297).
(Brief entries in Emin's diary show that he reached Pajule on May 10, 1881, and
was supplied with porters by Chief Ogwok for the purpose of leaving Padibe on
May 17.)
(The following letters which Emin addressed to Professor Behn contain the only
other references by Emin to affairs in Uganda during 1881.)
July 6, 1881.... I will give up writing the history of Uganda, because Wilson
and friend Felkin want to write it and are more competent to do so than I am.
July 8.... I have as good as no news from the south since by Gordon Pasha's
orders all my work has turned out to be fruitless. The stations have been withdrawn
and since the death of Mureko, Mutesa's uncle, Kabarega's people have blocked
the road to Buganda. According to a letter received by me about two months ago
from some Zanzibar Arabs in Bunyoro, Mutesa has attacked Kabarega, killed many
of his people and put them to flight. In consequence many of Kabarega's women
have been made prisoners. It was not known where Kabarega was. Four days ago
an embassy came to me from Mutesa with an invitation to go to him. I cannot go
at present, but will do so later.
Lado, December 25, 1881.... As regards the Uganda work I will consider the
matter. Do not think I have been idle or want to give up the work because of a
whim. Felkin and Wilson were longer in there (Buganda) than I was and so must
know more about it than I do. If there should be any controversy about what I
formerly told these gentlemen at the time of their departure from here, I must be
allowed to have nothing to do with it by allowing myself to be mixed up in a
useless controversy. Book writing is not every man's job, and especially not mine.
(Stuhlmann-Die Tagebiicher pp. 226, 227, 331.)

IGeorge Litchfield (1854-1945) was ordained in 1878. He and his companions had
arrived under the aegis of the Church Missionary Society. Landing at Suakin, they


had arrived in Equatoria by way of Berber and Khartoum. They reached Buganda in
February 1879. Later Litchfield served under the C.M.S. in India and South Africa
and eventually became a parish priest in England.
2Robert William Felkin (1853-1926) qualified in medicine at the universities in
Edinburgh and Marburg, Germany. He returned to England in 1879. Later he kept up
a correspondence with Emin and took a prominent part in the organisation of the
expedition which set out for his relief in 1887. He was the first Englishman to look upon
both Lakes Albert and Victoria. Went to reside in New Zealand in 1914. Joint author
with C. T. Wilson of Uganda and the Egyptian Soudan (1880) an dalso wrote many
articles on Uganda and on tropical diseases.
3Charles William Pearson (1847-1917) was originally an officer in the merchant navy.
He entered the C.M.S. College at Islington in 1876. He returned in 1879 from Buganda
to England by way of Zanzibar. At the time of his death he was vicar of Walton, Bucks.
4In reading Emin's comments on the C.M.S. missionaries it has to be borne in mind
that Emin was a convert from Christianity to Islam.
50n November 6, 1879, Gordon wrote to inform Sir Samuel Baker that he had decided
to abandon all stations to the south of the lake because "they are not worth
keeping". (Murray T. D. and White, A. S. Sir Samuel Baker: a Memoir pp. 243, 253,
61n 1862 Said Pasha, Viceroy of Egypt agreed with Napoleon III to furnish a
Sudanese contingent for service under the Archduke Ferdinand Maximilian of Austria,
who had been declared to be Emperor of Mexico. The contingent was withdrawn in
1868. Murgan Aga had served with this Mexican contingent with the rank of Captain.
In 1872 he served under Sir Samuel Baker in the Sudan. Later he was appointed by
Emin to be commandant of Amadi. He was killed in a sortie against the Mahdists shortly
before that post was evacuated in March, 1885. (H. B. Thomas-Note on the Sudanese
Corps in Mexico U.J. Vol. 8, 1940. pp. 28-32).
7Ernest Linant de Bellefonds passed through Acholiland on his way between
Gondokoro and Buganda in 1875.
8An account of the career of Rwot Ochama is to be found sub. tit. Rwot Ochama of
Payera in U.J. Vol. 12, (1948), pp. 121-128. To this account must be added the following
information, which appears in a letter of Mrs. Callon Moore printed in U.J. Vol 14,
(1950), pp. 223-224:-
"Apparently Baker heard aright. The chief's name was really Labwor (Rwot Labwor
-Lion). "Rwotcamo" was a nickname. It means literally, "The chief does not eat." and
he received it because he had a habit of not eating in public, but taking snacks and
titbits in the kitchens, so that the constant stream of visitors always .asked in surprise
"Rwot camo?" (Does the chief eat?) to which, the reply was, "Rwot camo" (The
chief does eat.) Hence the nickname."
9So great indeed had been the indignity inflicted upon Ochama that public opinion
evidently demanded that he should be deemed to have died what European medieval
jurists would have termed a civil death. As happened a quarter of a century later in the
case of his son Awich, a funeral dance was evidently held in respect of the yet living
man, and in native eyes Ochama had joined the ranks of the dead. Only by the ceremony
of "Gwor Kome" could he again be brought back to life and be re-established as a
living entity in his clan . the essence of the ceremony has been described by Mr. Bere
in his article on "Awich". U.J. 10 (1946) 76-8.
10Nanfumbambi was the title accorded to the Sabadu of Sekibobo, the chief of
Kyagwe (Roscoe-The Baganda p. 251).
11On March 6, 1879, Mutesa received certain letters from Zanzibar. Their exact
contents do not appear to have been communicated to the members of the C.M.S.
Writing on April 18 to Dr. Robb, Agency Surgeon to the British Consulate at Zanzibar,
Alexander Mackay said, "Then came two men from the Consul-at least they said the
Consul sent them-Kacheche was the head man. Kirk's letter then undid all the good
we had done and nearly cost us our lives. He said that we had got no connexion with
the British Government whatever. That may be true enough, but the Arabs and King
interpret this to mean that we are not British subjects and that we are runaways. They
do not of course understand the British Constitution. They came to the conclusion that
Lord Salisbury's letter never came from the Queen. In other words that it was a
forgery. Many troubles followed and we have had a most anxious time, and are
practically prisoners."
"Wilson and the rest are determined to leave, but they cannot .get off. They are now
striving secretly to inform Col. Gordon of our state, that he may send an army to
release them." "... Stokes and Copplestone are of my mind, but the rest are determined


on leaving. I feel such a step precipitate. The Mission here is only a day old practically,
and it is absurd to expect everything smooth for perhaps years to come." (Yule-Mackay
of Uganda-The Missionary Engineer. pp. 127, 128.) As will be seen, eventually only
Felkin and Wilson left.
12Pere Lourdel and Frere Amans of the White Fathers' Mission reached Entebbe on
February 17, 1879, P&res Livinhac, Girault and Barbot arrived there on June 17.
13There are innumerable references to Charles Henry Stokes in contemporary works
on East Africa. A short account of his career, and in particular of his activities in
Buganda, may be found in The Year of the Three Kings of Buganda, U.J. Vol. 14, (1950),
pp. 31 seq. The circumstances in which he met his end can be found in Africa No. 8
(1896)-Papers relating to the Execution of Mr. Stokes in the Congo State.
14Grant had recommended to the C.M.S. that their approach to Buganda should be
by way of Zanzibar from the south, because, "as Colonel Gordon was regarded in Central
Africa as an invader, a Mission approaching Uganda from northward under his auspices
would seem identified with a policy of annexation." (Stock-The History of the Church
Missionary Society, iii.. 96. 97.)
15For the career of Mirambo see R. J. Harvey-Mirambo the Napoleon of Central
Africa T.N.R. No. 28 (1950) pp. 10-28. The word "amiability" is, of course, intended in
its context to be sarcastic.
16"Kitaka, one who had treated me badly (sc. Felkin on his return journey to Mruli)
afterwards joined himself to Mr. Wilson's party and came with us to England. He
subsequently told me how sorry he was for his behaviour and expressed great regret for
it." (Wilson and Felkin-Uganda and the Egyptian Soudan ii. 40.)
17The real name of the chief was Fishwa. He was the son of his predecessor, Lai.
Hence he, and his place of residence, came to be called by the Sudanese Wat el Lai (the
son of Lai)-Moses-A History of Wadelai--U.]. Vol. 17, No. 1 (1953) p. 79.
18Clemens Theodor Perthes was the head of a Gotha publishing house which was
responsible for the issue of Petermann's Mitteilungen.
19Martin Ludwig Hansal was Austrian Consul at Khartoum from 1862 until his death
in 1885. He was killed by the Mahdists when they took Khartoum in the last mentioned
20Giegler (?Ziegler) Pasha was a German by birth who became Director of Telegraphs
at Khartoum in 1875. Gordon appointed him as acting Governor-General of the Sudan
in 1879. In 1881 he relieved Gordon's successor Rauf Pasha as acting Governor-General.
During this latter term of office the Mahdist revolt came to a head, but he underrated
the strength thereof. He left the Sudan on sick leave in 1882.
21Muhammad Rauf Pasha had served in Equatoria under Baker and acted as
Governor-General of the Sudan between Baker's departure and Gordon's arrival. He
succeeded Gordon as Governor-General in 1880. His failure to check the rising tide of
Mahdiism led to his recall in 1882. He died in 1888.
22"Towards the end of 1879 the king called in the mediums of the tribal gods from
Sese to cure him of a disease which he had contracted. Mackay, who tried to prevent
their coming, got into serious trouble over it and relations became so strained that the
English missionaries proposed to leave the country. This however, did not suit Mutesa,
who was afraid that if they went the Egyptians might come; he was determined to keep
them as hostages. Several incidents made it clear to the French priests that they were
in the same position. Fortunately for them, Mukasa and his fellow gods were unsuccess-
full in their efforts to cure the king. The medium of another god brought post-haste
from the Bunyoro border, also failed miserably and decamped in the evening of 31st
December, 1879." (Thoonen-Black Martyrs p. 43.) Mackay was present at the meeting
on December 23, 1879, and it is clear from his report thereof that Mutesa never issued
a final decree on the subject. All that Mutesa said was "that now they would leave both
the Arab's religion and the Muzungu's religion and would go back to the religion of
their fathers." On December 28 Mackay records that according to the information given
to him by his pupils that day "The decision come to in full court last Tuesday was not
final. The king and some chiefs evidently expect us to continue our teaching as before.
My boys being allowed to come unhindered every day certainly does not look like as if
a very strong veto had been put on our continuing our work." (A. M. Mackay . by
his Sister pp. 164-169, 170).
23For some account of the career of Ogwok see Acholi History, 1860-1891, U.J.
Vol. 15, (1951), p. 133, Vol. 16, (1952), pp. 137, 141.
24See U] Vol. 24 (1960), 117 note 1.



WHEN the Sudan Government created the province of Mongalla on 1 January
1906, the object of the administration was to expand the government's control
beyond the precincts of the few riverain stations already established in the
Southern Sudan. Working from these stations the British administrators slowly
extended their influence over the neighboring tribes and won the support and
the confidence of the inhabitants by tactful promises of rewards, appeals to self-
interest, and occasionally the use of armed force. The Bari and the Bor Dinkas
soon acknowledged the government to be followed in 1908 by the Aliab Dinkas
and in 1909 by the Twi. But Mongalla Province grew not only by the spread
of effective administration but also by the acquisition of additional territory.
On 17th December 1909 Leopold II, King of the Belgians and Sovereign of
the Congo Free State, died. By the terms of the Anglo-Congolese Agreement of
9 May 1906, Leopold had agreed, in return for a railway concession from the
Congo to the Nile, to annul the Anglo-Congolese Agreement of 12 May 1894 and,
on his death, to hand over the Lado Enclave to the Sudan Government. On
16 June 1910 the agreement of 1906 was consummated, and the Enclave was
officially transferred to the Anglo-Egyptian authorities and attached by them
to Mongalla Province.1
But the transfer of the Lado Enclave to the Sudan created nearly as many
problems as it solved, for the greater part of the eastern boundary of the Enclave
from 5" north latitude to Mahagi on Lake Albert, did not adjoin Sudan
territory but rather bordered on that of the Uganda Protectorate. Indeed, before
the Enclave reverted to the Sudan, the presence of the Congolese on the west
bank of the Nile and Uganda on the east afforded the tribes the opportunity to
play off one government against the other. Frequently, the bulk of a tribe, such
as the Bari, would live on the Congo side to avoid paying hut taxes to Uganda,
while keeping their cattle on the opposite bank to avoid paying tribute to the
Congolese. The difficulties, if not the impossibilities, of governing those who
could change their rulers by merely crossing a river was apparent to all, but the
transfer of the Enclave from the Congo Free State to the Sudan did nothing
to correct this anomalous situation. Moreover the administration of the southern
Enclave was much more difficult when undertaken from the posts of Lado,
Rajjaf, or Kajo Kaji in the north rather than from Uganda territory further
south across the Bahr al-Jabal. It was much easier for administrators to march
up the tributaries of the Nile, which flowed eastward from the interior of the
southern Enclave, than to traverse from north to south row upon row of preci-
pitous parallel ridges and steep ravines. If the Sudan Government retained the
southern Enclave, its administration would not only be difficult and expensive,
but the close control required to halt the spread of the sleeping sickness would
became practically impossible.
The logical and indeed self-evident solution to this problem was of course for
the Sudan to hand over the southern Enclave to the Uganda Protectorate in

return for the Uganda territory opposite Lado and Rajjaf. Within a year after
receiving the Enclave from the Congo Free State the Sudan Government had
decided to seek just such an exchange of territory with Uganda and Bimbashi
C. H. Stigand was instructed to draw up recommendations to guide the Sudan
authorities in any future negotiations. Stigand, a captain in the British Army,
had served in Burma, Somaliland, and East Africa before joining the Egyptian
Army and the Sudan Government service in 1910 when the latter took over the
Lado Enclave. His knowledge of the southern Enclave, acquired on hunting trips
and official tours, coupled with his experience obtained in East Africa had made
him one of the few English officers in the Sudan service who could give
information about the territories in question based on personal observations.
Acting on the assumption that the principal object of the proposed exchange
of territory was to bring the terminal points of the frontier on either bank of
the river opposite to one another, Stigand made the following recommendations:
The Sudan Government should obtain the right of waterway between Nimule
and Lake Albert, and make suitable arrangements with Uganda for a bi-monthly
service between Nimule and Butiaba. Nimule itself should be retained, in spite
of the fact that the Uganda Government was about to abandon it in favour of
a more healthy site some forty miles inland. The Sudan should also retain the
right to raise the two steamers of Emin Pasha sunk in shallow water near Wadelai
and Lake Albert. Stigand saw no reason to continue to maintain Gondokoro as
an administrative centre, for trade was at that time being diverted through
Rajjaf and would continue to pass through that station since the river approach
to Gondokoro, always treacherous because of shifting sand-bars was now
impossible at low Nile. Stigand further suggested that the Sudan keep Dufile
and the left bank of the Bahr al-Jabal for some ten miles upstream. Retention of
Dufile was necessary, Stigand argued, in order to provide a direct line of com-
munication between Nimule, Kajo Kaji, and the north, and although this would
mean drawing the boundary through the Madi people, the frontier would at least
pass between the northern and the southern Madi who at that time were
unfriendly to one another. The greatest objection to this proposal was recognized
by Stigand himself who pointed out "that the opposite bank of the Nile to these
Sheikhs [in the Sudan] will almost certainly be Uganda territory and the old
difficulty of natives crossing and re-crossing or having homes and fields on both
sides of the river is sure to crop up in the near future. However, just at present
there is no such difficulty as these Sheikhs are not on good terms with the
Uganda Government."2
The remaining parts of the memorandum were concerned with descriptions
of the country, people, and products rather than with specific proposals for the
drawing of a boundary line. Qii imma quam H. Pearson Bey, who had led the
Sudan Commission which had arranged for the transfer of the Lado Enclave,
generally agreed with Stigand's notes although he warned that the Sudan
authorities must "be very careful about establishing more posts than are absolutely
necessary on the frontier. They are likely to be extremely difficult to provision,
and to be of any use they must be of some strength, and therefore expensive to
keep up."3
The following year, 1912, the Sudan and Uganda Governments agreed in
principle to a rectification of the frontier, and on 18 January 1913 a boundary
commission left Nimule to examine the ground in detail and submit recommen-
dations. The Sudan commissioners were Captain H. H. Kelly, R.E., and Captain
H. J. Huddleston (later Governor-General of the Sudan, 1940-7). Captain H. N.

Tufnell and Captain H. A. Lilley (later District Commissioner, Torit) represented
Uganda. They were instructed to draw up a well-defined boundary so:
"1. That the Sudan Government shall take over that portion of the Uganda
protectorate which lies between the present boundary (5 N. Lat.) and a line
commencing at the junction of the Unyama River with the Nile to the south
of Nimule, thence proceeding up the Unyama River to a point on the right bank
10 miles from its mouth; from this point the line runs N.N.E. to the fourth
parallel and thence (roughly) along this parallel to include the Dodinga and
Dabisa districts, finally turning N.E. to a point on the Sanderson Gulf of Lake
Rudolf, approximately where the shore of the gulf is cut by 40 40' N. lat.
"2. That Uganda shall take over that portion of the Lado district which lies
to the south of a line commencing at a point on the left bank of the Nile
opposite the junction of this river with the River Unyama, thence proceeding
N.N.W. along the northern boundary of the Lugware tribe to the Kaia river,
and following the course of the latter river to the (Belgian) Congolese frontier.4
SMoving slowly eastward the commission carefully investigated tribal
boundaries as well as the natural features of the country in an attempt to
correlate the two. In spite of the difficulties encountered in demarcating tribal
limits so as not to divide a tribe, the commission was able to construct
a boundary between the Nile and Madial which, with a few exceptions, remains
the same today. At Madial, which the Commission reached on 14th March, the
commissioners were forced to abandon their joint investigation because of "the
waterless state of the country."5 The Sudan party, equipped however with camel
transport, was able to proceed alone and subsequently continued their march
from Madial in a north-easterly direction to Jabal Mogilla. Beyond the Sudan
commissioners merely assumed the boundary to the east to be a theoretical line
beginning at a point north of Jabal Mogilla and extending to Lake Rudolf, and
then they continued their march northward to explore the Boma Plateau. Captain
Kelly in his report on the work of the commission described in detail the proposed
boundary between the Sudan and Uganda.
"The boundary will follow up the Unyama River from its mouth to a distance
of some 10 miles to J. Ebijo, from the top of this hill a line toward J. Kadomera
(the westmost of the Amoji group of hills) to a point on the Assua river west of
this hill. Thence follow down the Assua River to Lokai. In order that point on
the Bahr el Jebel accessible to steamers may be secured to the Sudan, an
enclave on the east bank above the Unyama, a thousand yards by a thousand
yards should be given to the Sudan.
From Lokai, the line to run approximately north-east through uninhabited
country to J. Matokko (Batogo or Atokko), this hill falling to Uganda.
From Matokko a line running nearly due east to the summit of J. Aggu,
leaving the Farajok District to the Sudan and crossing the Farajok-Lukung
road near a small hill 2,000 yards south-east of the crossing of the K. Laparra.
From Aggu, a line to the highest point of Hala [Goula] and thence to the
southern peak of J. Langia [J. Moddle], leaving to Uganda the Acholi on the
south and south-eastern slope of the range. From thence [the line goes] by the
small peak of J. Tiya [J. Orpir] to the extreme north of J. Tereteinia. Leaving
the latter hill to Uganda, the line runs approximately east to the small hill
called in Sheet I "Uganda" August 1900, Latome, leaving the Loggire hills and
its southern continuation J. Momoi to the Sudan. From Latome, the line runs
south of J. Harago [Lotuke] (the southern limit of the Dodinga tribe) and

'IAN S'j 0 GE

/ J LABU z

-' -1 I z

--, --


BOUNDARY 1914 ---..-..-
BOUNDARY 1961 ----------
SCALE -ni j

thence eastward to the north of J. Mogila (appr. lat. 40 15' N. and long. 34*
30' E.)
From here a theoretical line to the north of Mount Lubur on Lake Rudolf
is assumed, but if the northern portion of the lake proves navigable, a strip of
territory should be reserved to the Sudan, affording a port on the lake."6
The principal feature of the boundary upon which the Sudan commissioners
insisted was the enclave at Nimule. Owing to difficulties of navigation at the
mouth of the Unyama River, steamers from Lake Albert docked south of the
junction of the Unyama and the Bahr-al-Jabal rivers where goods were taken by
lighter to Nimule. An enclave, a thousand yards square, was consequently given
to the Sudan so that steamers could dock within Sudan territory. East of Nimule
the commissioners endeavoured to draw the boundary so as not to separate tribal
units nor leave members of the same tribe on either side of the frontier. But
such a principle, most laudable in theory, was virtually impossible to carry out
in practice. For instance the Acholi of the Farajok District were only one section
of the main tribe which lived further south in the Uganda Protectorate. In
language, type, and origin they were pure Acholi and accordingly "should, there-
fore, properly speaking go to Uganda."7 But the Farajak Acholi remained in
the Sudan. Separated from the Uganda Acholi by uninhabited country and much
closer to Nimule than to the nearest post in Uganda, they could best be ruled by
the Sudan Government. Consequently, in violation of their principle, the
commissioners drew the boundary through the uninhabited territory between the
Farajak Acholi and those further south. Such a division was later to create
considerable difficulties between the Sudan and Uganda Governments. In 1926
the Uganda authorities proposed that the Farajok-Opari Acholi, as well as those
living in the Agoro hills to the east, be placed under their administration. The
Sudan Government however rejected this proposal on the grounds that a true
tribal boundary simply could not be found in the area and that "most of these
people were not Acholi (there are only two Acholi sections here) but Latuka and
Lango who were closely intermarried with Latuka District: and any alteration
would only make matters worse."8 In 1928 and 1930 the proposal was again
put forward by the Uganda authorities, but nothing appears to have come of
their overtures.
Similar difficulties arose in the Tereteinia-Madial area inhabited by the Lango.
Although a Latuka-speaking people with certain cultural affinities with those
tribes to the north, the Lango of Tereteinia were left in Uganda, for Tereteinia
"has been on the main Uganda route eastwards since the [Uganda] administra-
tion has been pushed to the north, and its inclusion in Uganda territory is
recommended so that there may be no break in the continuity of administration."9
For much the same reasons Madial was included in Uganda, and any "existing
difficulties occasioned by the close relations of a portion of them [the inhabitants]
with the Dodinga can be dealt with administratively later.10 Administrative
freedom to correct such discrepancies never proved a great success, and in 1926
the Tereteinia-Madial area was ceded to the Sudan "on the grounds that the
people were of the same tribe ("Lango") as those in the Sudan to the north of
them and that their inclusion in the Sudan would facilitate sleeping sickness
East of Madial the Sudan commissioners, who proceeded alone because of the
scarcity of water, fixed the boundary through an uninhabited zone depopulated
by intertribal warfare and extending from Jabal Lotuke north-east to a point
north of Jabal Mogilla. Although this zone was unlikely to remain permanently

uninhabited, "the territory falling to Uganda is that which has hitherto been
more or less closely administered by that Government, and although this may
appear to place the Sudan Government at a disadvantage, there is the compen-
sating advantage that when administration is commenced, it will be on new
ground, while there will be no change of system in territory already under
control."12 East of Jabal Mogilla the Sudan commissioners merely assumed the
boundary to be a straight line running direct to Lake Rudolf and, abandoning
any attempt to advance eastward to the lake, marched north to explore the Boma
Plateau. Kelly was confident that this theoretical line would "clear all the grazing
grounds formerly occupied" by the Turkana, but later experience proved this
assumption to be quite unfounded.13 The Turkana had always ranged freely
from the Tapeisi hills on the Sudan-Abyssianian border southward to the
Turkwell River and beyond. Indeed in 1918 the Turkana Patrol crossed back
and forth over the frontier between the Sudan and East Africa in their pursuit
of recalcitrant tribesmen, and the subsequent administrative difficulties produced
by this arbitrary boundary were to remain for nearly twenty years before the
erection of administrative posts and the creation of the Ilemi Triangle brought a
semblance of order to this part of the frontier. It was further agreed that the
Sudan would receive an enclave on Lake Rudolf in the event that the Uganda
railway was extended to the lake or that the country northwest of the lake
provided marketable commodities which could be sent down the lake to Kenya
in shallow draught vessels.14 Needless to say, these expectations were never
West of the Bahr al-Jabal the boundary, like that east of Jabal Mogilla, was
delimited without either party visiting the area. Here, as elsewhere along the
frontier, the commission endeavored to follow tribal boundaries, and in spite
of Stigand's advice to retain the northern Madi in the Sudan, the basic premise
on which the commission acted was to separate the Bari speaking tribes from
the Madi and the Lugware. Recognising the impossibility of carrying out such
a principle without first carefully examining the area, the commission simply
proposed a provisional boundary with the strong recommendation that officials
"conversant with actual conditions arrange the exact line" even if "the trans-
planting of a few villages may prove desirable in order to make a thoroughly
satisfactory boundary."15 The general boundary submitted for ratification
"From a point on the bend of the river opposite the Unyama River, the
riverain strip to the foothills of the escarpment as far as the mouth of the K.
Kayu (or Aju) to fall to the Sudan. The boundary to run up the Kayu from
its mouth to the K. Nyaura (?) [Nyawa?] thence up the latter khor to its source.
From here to a point on the R. Kaia to be determined, and up the Kaia to the
Congo watershed."16
The commission's ignorance of the territory west of the Bahr al-Jabal did not
result in a significant boundary, for the Nyawa was neither a tributary of the
Kayu nor did it flow from the source assumed by the commission. In practice
the frontier has always been drawn as a sweeping line from the Khor Kayu to
the Kaia and then, as fixed by the commission, up that river to its source on the
Congo-Nile watershed. In spite of the recommendation of the commission this
boundary was never redefined although the need for sleeping sickness control
as well as for effective administration made it all the more desirable. In 1930 the
Governor of Uganda requested that a proper delimitation of this section of the
frontier be undertaken, but since his proposal was tied to the demands for the

cession to Uganda of the Sudan Acholi, a proposition which the Sudan authorities
were not prepared to accept, any hope of delimitation vanished with the Sudan
Government's rejection of that proposal.17 Furthermore, the Uganda authorities
had wished to do away with the curious arrangement whereby the boundary
immediately west of the Bahr al-Jabal followed the foot of the escarpment,
replacing that line by the river itself. It is not clear why this riverain strip was
given to the Sudan in the first place. Kelly is silent on this matter though it
was later suggested that it "was probably due to possible irrigation needs."18
In 1932 the Uganda Government again raised the question of the frontier west
of the Nile, and although preliminary conversations were held between the
Provincial Commissioner, Northern Province, and the Governor, Mongalla,
nothing appears to have come of these discussions.
The work of the commission cannot, in retrospect, be regarded as very
thorough or very satisfactory. Once the principle of the exchange of territory
had been agreed upon by the respective governments in 1912, it was the task of
the commission to construct a significant boundary. The failure of the commission
to visit all of the areas under consideration obviously prevented this and
diminished the value of their recommendations. There is no doubt' that the
commissioners regarded the frontier, particularly west of the Nile as only a tempo-
rary arrangement, but it takes only a short time for even the most temporary
arrangement to acquire an air of permanency. Since there was no absolute geogra-
phic feature, like the Congo-Nile watershed for instance, upon which the whole of
the boundary could be fixed, tribal boundaries appeared to be the only principle
upon which a more permanent frontier could have been established, but tribal
boundaries, as the commissioners had discovered in the Farajok District, were
frequently confusing and unreliable. To construct an international frontier from
the claims of rival tribes was at best a difficult and trying undertaking when the
commissioners were present to examine the area in person. To attempt to fix such
a boundary without thoroughly investigating the frontier was simply to invite
unwelcomed disputes in the future. The interests of both the Sudan and Uganda
would undoubtedly have been better served if the Commission had traversed
the whole of the proposed frontier and had devoted more care to delimiting,
particularly in the east and west, a more rational boundary.
On 1 January 1914 the boundary submitted by the Commission was recognized
by the Sudan and Uganda Governments, and the proposed exchange of territory
officially consummated. The Uganda post of Gondokoro was abandoned but the
important river port of Nimule was occupied at once by a section of the
Equatorial Battalion until April when a sufficient number of Mongalla Province
police were made available for duty there.19 At first Nimule was attached to the
Kajo Kaji District, but in 1916 the district headquarters were transferred to
Opari situated in the Acholi country east of the Nile.20 The only other part of
the territory newly acquired by the Sudan to be bruoght immediately under
government administration was the Lutaka country located south-east of Gondo-
koro. Tere Torit on the Kineti River was chosen as the district headquarters
in spite of rumours spread, in a fit of pique, by Chief Lokilei of Tarangole when
his village was not selected, that the Kineti River was poisoned. Torit with ample
water, a strategic location on the plains below the Acholi, Imatong, and
Dongotona Mountains, and a magnificent view of these ranges was obviously
preferable to Tarangole where the water supply was simply inadequate to support
a large administrative post.21 On 17 April the 5th Company of the Equatorial
Battalion arrived at Torit to erect the station which remains today as the

administrative centre of the district and the former headquarters of the recently
disbanded Equatorial Corps.22
Although the Sudan authorities had no plans at that time to take over the
rest of their newly acquired territory which stretched eastwards from the Latuka
District to Lake Rudolf, they were soon called upon to advance further south into
Uganda. During the summer of 1914 the European Powers had moved irresistibly
toward the brink of war, and when hostilities did break out in August the King's
African Rifles, who were responsible for the defence of Kenya and Uganda, were
widely scattered throughout British East Africa. Units of the K.A.R. had only
just concluded operations in Jubaland in the spring, and another expedition was
assembling to march against the Turkana near Lake Rudolf when the war began.
At that time there were two companies of the 4 K.A.R. en route from Nairobi to
the Turkana country while three other companies of the same battalion were on
garrison duty at Morongole and Moroto in northern Uganda. In addition
one-and-a-half companies of the 3 K.A.R. were scattered in posts along the
Northern Frontier of Turkanaland.23 Even before the outbreak of hostilities the
redeployment of these troops had already begun in order to concentrate the
K.A.R. along the frontier of German East Africa within reach of the Uganda
railway. The Uganda railway not only formed the principal line of communicat-
tion between Uganda and the world but the economy of Uganda and the High-
lands of Kenya were totally dependent upon it. In fact any contribution which
British East Africa hoped to make to the allied war effort would depend largely
upon the ability of the K.A.R. to protect a long and particularly vulnerable
railroad which at one point approached the German frontier within a distance
of seventy miles. To patrol and defend the Uganda railway all available forces
were rushed to the frontier, and the Northern garrisons of the K.A.R.
were soon denuded of troops, leaving those tribes already administered by the
Uganda and the East African Protectorates open to attack from neighboring
tribes and Abyssinian raiders. The possibilities of such disturbances in the north
while British East African troops faced the Germans in the south would have
unsettled these tribes hitherto loyal to the government and, since the withdrawal
of government protection, would have undoubtedly provoked them to counter-
attack on raids of reprisal. Unless punitive measures remained in the hands of
the government, widespread inter-tribal fighting and warfare would be the
inevitable result of the withdrawal of the K.A.R. from the Northern frontier.
To keep the border peace therefore the British authorities in East Africa urgently
requested in September that Sudanese troops be sent to help the Uganda police
and the few remaining soldiers of the K.A.R.
On 14 September 1914 the Acting Governor of Uganda telegraphed to Sir
Reginald Wingate, the Governor-General of the Sudan, requesting that Sudanese
troops occupy Madial and "protect joint Northern Frontier from raids from
Abyssinia." He added also the ominious warning that the Sudan Government
should have reinforcements ready if needed against the Germans.24 Wingate
immediately transmitted this request to the Sudan Government Agent in Cairo,
G. F. Clayton, and suggested that the Sudan "could send some small assistance via
Port Sudan and Mombasa provided troops from Sudan would not be required in
the event of a Turkish advance against Egypt."25 Although Clayton did not
personally wish to see Sudanese troops used in offensive operations in German
East Africa, the Sudan Agency in Cairo supported Wingate's offer of troops
and sought at once the necessary permission from the Foreign Office in
London.26 In the meantime the Adjutant General in Khartoum had consulted

with R. C. R. Owen, the Governor of Mongalla Province, as to the number of
troops available in the Southern Sudan for use at Madial and on the Northern
Frontier.27 At that time Owen could spare most of the regular troops of the
9th Sudanese as well as a half company of Equatorials.28 Although Wingate
declined to use the Equatorial troops, probably because of their inexperience, he
informed the Acting Governor at Entebbe on the 16th that two companies of
Sudanese troops could be sent at once to Madial.29 On the same day, however,
the Sirdar in a long telegram to Clayton, raised serious questions as to the extent
of support the Sudanese troops in Uganda were expected to give.
"I will make arrangements to send two companies of Sudanese regulars to
Madial, but it would require a large and costly expedition to make good our
new frontier with Uganda as far as the western shore of Lake Rudolf; see
Kelly's reports and reasons against such a course at present. I do not understand
what action Governor, Entebbe, means by asking me 'to protect our joint
Northern Frontier from possibilities of Abyssinian raids' unless he means me to
take over some other posts in the Rudolf Province of which I am not aware. In
July last I told the Secretary of State for the Colonies that I might assist the
intended Uganda expedition against the Turkanas by making a demonstration
from the Nile, and the occupation of Madial would I think have this effect.
There is evidently some misunderstanding as regards Uganda's frontier with
Abyssinia, and it had better be made clear."30
Privately Wingate expressed deep misgivings about the use of Sudanese troops
in Uganda. Contrary to Clayton's view that a small Sudanese expeditionary
force to Mombasa would be undesirable, the Sirdar obviously preferred that to
patrolling the Uganda frontier. He wrote to Clayton that "a few troops [to fight]
against the Germans . would be a popular move," but that mere police work
in northern Uganda was "quite another matter" and should be left to troops
from India.31
On the following day, 17 September, the Acting Governor at Entebbe clarified
his request for assistance, assuring Wingate that the Uganda Government did
not expect a large expedition to control the whole of the Lake Rudolf area but
simply wanted Sudanese troops at Madial "for the purposes of defence locally
in Rudolf and Morongole or elsewhere should the occasion arise."32 He added
however that any reinforcements required would be needed via Nimule, not
Mombasa, and hoped that in addition to the Sudanese troops to be sent to
Madial a striking force could be also garrisoned at Nimule. Reassured on the
use of the troops at Madial, Wingate at once ordered Owen at Mongalla to send
one and a half companies of 9th Sudanese to Madial by 21 September to be
followed later by the remaining half company.33 By the 18 September Clayton
in Cairo*had received the official sanction of the Foreign Office for the Sudan
Government forces to occupy Madial, protect the Northern Frontier, and to
give "such assistance as you [Wingate] may be able to afford."34
During this flurry of telegrams between Entebbe, Khartoum, Cairo, and
London, the military situation on the southern frontier of Uganda had deterio-
rated considerably. Early in September a German force of some 400 men had
marched swiftly across the frontier east of Lake Victoria and occupied the port
of Karungu. The purpose of the German advance was to destroy the railway
bridges east of Kisumu, and so moving in the direction of the railroad the
Germans occupied the station of Kisii, driving off the British District Commis-
sioner and his police. Reacting quickly to this German advance, a company of
the K.A.R. marched at once from Kisumu followed by two additional companies

from Uganda. On 12 September the British force under Captain E. G. M.
Thorneycroft reached Kisii and attacked the Germans. All day both sides
attacked and counter-attacked but made no real gains. By nightfall the British
forces, having exhausted their ammunition, withdrew, and a few days later the
Germans, who had lost half their Europeans and most of their carriers, retired
back across the frontier.35
Although the Battle of Kisii had no decisive effect east of Lake Victoria, west
of the lake along the southern border between Uganda and German East Africa,
the military situation had become critical. Having rushed all available troops to
Kisumu to halt the German advance toward the railway, only a few hundred
police remained to guard southern Uganda. To cover these inferior numbers
however the police decided to take the offensive, and on 14 September crossed
the Kagera River and seized Kyaka Fort from the German forces. Although
the Germans proved slow to react to the loss of Kyaka, the reports of enemy
detachments along the frontier alarmed the authorities in Entebbe. On the 21st
Wingate received a personal plea for immediate support from the Governor of
Uganda, Sir Frederick Jackson:
"Our position is becoming serious and we require further assistance. Also for
defence of southern frontier can you assist by sending four companies your
local troops via Nimule. Secretary of State for Colonies has been asked to make
arrangements accordingly with Foreign Office."36
Wingate replied at once that he was holding four companies of Sudanese
regulars in readiness pending orders from London, but it was the view of the
British authorities in Cairo that the policy of the Sudan sending reinforcements
to Uganda had been approved several days before when the Foreign Office had
consented to the occupation of Madial and the defence of the Rudolf region by
Sudanese troops.37 Indeed Milne Cheetham, the British charge d'affairs in Cairo,
who was acting in Kitchener's place, took this view and answered the Foreign
Office at once that the "Sirdar can send 31 Companies via Nimule. All can now
go forward."38 But no sooner had the way been cleared for the troops to go
forward than they were not needed. It soon became clear to the officials in
Entebbe that their fears had been unwarranted and that additional Sudanese
reinforcements would not be required. The retreat of the German forces from
Karungu had released the reserve companies of the K.A.R. for service along the
frontier west of the lake and the anticipated German threat in that area had
never materialized. Therefore late in the day of the 22nd the Acting Governor
at Entebbe telegraphed to Wingate that the "German East African troops have
fallen back. Would be sufficient to have a small force ready at Nimule to draw on
direct in emergency. This does not effect assistance for Madial and protection
of Northern Frontier which has been approved."39
Although the Sudan authorities were willing to rush reinforcements to Uganda
in an emergency, they were quite unwilling to keep inactive at Nimule regular
troops who could be used to greater advantage elsewhere in the Sudan. The
Uganda officials did not press their point. The military situation was rapidly
improving, and if Sudanese troops could only take over Madial and the Northern
Frontier, there was little point in immobilizing additional Sudanese soldiers for
an emergency which appeared more unlikely with every passing day.40
By the middle of October the first detachment of the 9th Sudanese had reached
Madial and relieved the few remaining men of the K.A.R. who hurried south to
join their battalion on the German frontier. Other Sudanese troops soon followed,
passing beyond Madial to occupy Morongole and Kitgum where the officers and

men found the situation rapidly deteriorating.41 Not only had the Turkana raids
increased in frequency and violence during the summer, but those tribes of the
Northern and Eastern Provinces of Uganda who were actively administered by
the government looked upon the withdrawal of the K.A.R. with fearful mis-
givings. The prompt arrival of Sudanese troops had relieved much of this
apprehension, and the decision to send in the coming dry season a punitive
expedition into the Turkana country, in spite of the demands of the war, had
raised the hopes of British officials that the Turkana raids would at least be
restricted if not eliminated from the Northern Frontier. Until that time the
Turkana raiders were held in check by the Uganda police and the troops of
the 9th Sudanese. On 10 November 1914 some 73 Sudanese under the command
of Bimbashi D. W. Fairbairn in co-operation with 31 Uganda police under
Mr. Waters, made a night attack against a large party of Turkana raiders
camped on the slopes of Mt. Pelegech. Taken completely by surprise the
Turkana offered only token resistance and then fled, leaving behind 19 dead and
many thousand of looted cattle, sheep, and goats.42
When not pursuing Turkana raiders the Sudanese patrolled the border country,
keeping peace between the tribes and punishing belligerent tribesmen who would
raid their neighbours for cattle, women, and other loot. In areas where the
government had established effective administration there were few disturbances
that could not be handled by the local police. In other parts of northern
Uganda however where the influence of the government was little or none, the
Sudanese had frequently to use force to maintain order. For example on 16
December 1914 a Sudanese detachment encamped at Bira was attacked by the
local inhabitants who wounded several troops before being driven off by the
soldiers. On the following day the tribesmen attacked a second time. Again they
were repulsed, but the Sudanese were finally compelled to retire to Madial when
they discovered that the water supply had been contaminated by poisonous trees
and slaughtered dogs thrown into the pools by their assailants.43 A strong patrol
was later sent against this recalcitrant tribe. Peace was at last restored, but the
post at Bira was removed to a friendlier village further south.44
By January 1915 the Turkana punitive expedition which had been sanctioned
by the East African authorities in the autumn of 1914 was at last made ready at
Nairobi. In spite of the necessity to defend the southern frontier and the Uganda
railway from German encroachment, 20 British officers and 428 men of the
K.A.R. and the Kenya Police Service Battalion were ordered to entrain for
Londiani and the Turkana country on the 11 January. There the patrol was
divided into three columns each of which marched to separate base camps
located at Ngiyan, Marich, and Ngabotok respectively. By February each column
was prepared to move forward into the Turkana country, and on the 4th of
that month this phase of the operations embraced the valleys of the Kerio and
the Sugota rivers and westward to the Wei Wei and the Turkwell. Sweeping
northward the columns drove the Hellai section of the Turkana and part of
the Neseto before them, inflicting numerous casualties and capturing a large
number of livestock. The second phase of the operations lasted only for the first
ten days of April 1915 and was carried out in the territory between Kerio
and the Turkwell rivers and beyond up the valley of the Kozibir not already
visited by the patrol. Here too the government forces were successful in routing
the Turkana and capturing much stock, but the resistance was decidedly more
stubborn, and frequently the Turkana attacked at night in the hope of recapturing
their cattle. The third and final phase of the operations extended over the vast

area between Lake Rudolf on the east and the Karamojong Escarpment on the
west and as far north as the Labur Mountains. To patrol this wide and desolate
plain Lt.-Colonel W. F. S. Edwards, commanding the Turkana punitive expedi-
tion, requested the support of the 9th Sudanese in the forthcoming operations.
Edwards' request was strongly supported by Governor Jackson at Entebbe and
Mr. C. W. G. Eden, the Provincial Commissioner of Northern Uganda at
Masindi.45 Wingate had no valid reason to refuse this request and consequently
gave his approval for the co-operation of the 9th Sudanese with the Turkana
Collecting troops from Madial Bimbashi Fairbairn marched to Morongole
where he left behind a small detachment of 30 men and hurried south to join
the patrol which at that time was operating along the Turkwell River.47 From
26 April to 18 May two flying columns of the K.A.R. and the Uganda Police,
reinforced by Fairbairn and his Sudanese, marched through the Turkana country
capturing livestock and pursuing fleeing Turkana tribesmen. In spite of the large
number of Turkana killed and captured, most of the tribe found safety in the
Labur Mountains and in Abyssinian territory from which they could return
upon the withdrawal of the punitive expedition and continue their raiding and
poaching on the settled tribes in Uganda and East Africa.4
The hasty departure of Fairbairn and his Sudanese for Turkana land again
denuded northern Uganda of troops. The District Commissioner at Kitgum,
Mr. Pellew Wright, was able to furnish a few Uganda police to garrison Madial,
but they were insufficient to check intertribal raiding particularly trans-frontier
raids by Didinga from the Sudan. Unwilling to reduce the already inadequate
police forces on the southern frontier for service in the north, the Provincial
Commissioner of Northern Uganda, Mr. Eden, urged Governor Owen to
establish "a more or less permanent post on or near the border in the locality
of Logire or in the Dodinga hills; or at any rate to post sufficient men in that
area to prevent further raids being made by those people."49 Owen immediately
referred this request to Wingate in Khartoum, suggesting that a company (100
Rifles) of Equatorial troops be garrisoned in the Sudan in the Logire country
and supplied with grain from Kitgum.50 Not wishing to use inexperienced
Equatorial troops to patrol the borderland Wingate refused to sanction this
On 23 June the Governor of Uganda, Sir Frederick Jackson, informed the
Sudan authorities that the operations against the Turkana had been completed
and that the Sudanese troops could now be returned. Jackson took this
opportunity however to repeat Eden's request for Sudanese support on the
frontier and particularly for a border post:
"In my telegram I referred to the post to the north of Madial in Sudan
territory which I trust you will be able to establish as otherwise our common
boundary will be in a continual state of unrest from raids by Dodinga and Bira.
It will be quite impossible for this Government to re-establish the military post
at Madial for many months to come, and the Military administration of the
Sudan territory to the north of this point would do much towards relieving
Although the Sirdar was not prepared to risk Equatorial troops on the frontier he
could not very well refuse Jackson's plea for Sudanese troops to keep the border
peace. He therefore instructed Owen to retain one company of the 9th Sudanese
in the Protectorate, and consequently on 14 July Owen ordered No. 5 Company
under Bimbashi H. F. Hobbs to detach themselves from the Sudanese contingent

returning from the Turkana Patrol and remain "in Uganda territory for the
present either at Morongole or Madial whichever is most suitable to the Uganda
The Sudanese troops were again stationed at Madial, Morongole, and Kitgum
from which they patrolled the border and supported the Uganda administrative
officers in carrying on their duties. Usually the mere presence of Sudanese troops
was sufficient to ensure loyalty to the government, but frequently armed force
had to be employed to persuade recalcitrant tribesmen to comply with the orders
of the administration. In September for instance a detachment of Sudanese troops
accompanied Pellew Wright and his assistant, Mr. Ruble, to Rom Mountain
where they had to use force to remove the inhabitants. Early in the year Chief
Leucharo of Rom and his people had moved at the request of the District
Commissioner to a site selected by themselves at Omiya Pachua. Jabal Rom
had become a refuge for runaway tribesmen and desperadoes who terrorized
the surrounding countryside and intimidated the inhabitants of Rom to supply
them with shelter and provisions. The government was virtually powerless to
apprehend these undesirable tribesmen without evicting the people of Rom.
Such extreme measures were at last decided upon, but although the majority of
Chief Leucharo's people settled at Omiya Pachua, many gradually returned to
the mountain where they continued to harbour the desperadoes. Consequently,
Pellew Wright ordered the Sudanese troops to surround the jabal, burn the huts
and crops, and force the inhabitants back to Omiya Pachua where they could
be administered more easily and would no longer provide protection and
hospitality for raiders, poachers, and others who would threaten the security of
the district.54
In November the Khartoum authorities revived the plan to withdraw the
Sudanese troops from Uganda to a border post just across the frontier in the
Sudan. It was hoped that this station might be placed in the Didinga country so
that the garrison could then undertake the pacification and administration of
the Didinga while at the same time maintaining the border peace. Governor Owen
was not in favour of this proposal. He repeated his earlier warning not to place
an isolated military post in the Didinga Mountains without preparing to
administer the territory and supply an inspector and the necessary staff.
Personally he was willing "to start the administration of Didinga in 1916 if the
necessary Civil Staff can be provided," but until that time the Sudanese should
retire to the Logire village of Chief Ikoto situated on the western spur of Jabal
Logire near the Torit-Madial road.55 The site had formerly been occupied by a
company of the K.A.R. and had been the spot suggested by Mr. Eden the
preceding spring as a suitable location on the Sudan side of the border at which
to establish a military post. Later when the Didinga were finally brought under
administration the Ikoto Post could be abandoned, but until that time the
Didinga should be left "severely alone" until proper administration could be
Acting on Owen's advice Wingate deferred the plan for administering the
Didinga and simply ordered No. 5 Company 9th Sudanese to move from Uganda
territory to Ikoto. By mid-December Beaumont Bey, the District Commissioner
at Torit, and Bimbashi H. F. Hobbs had selected a site for the post at Ikoto,
and on 10 January 1916 the first detachment of Sudanese troops arrived from
Kitgum to be followed shortly thereafter by those from Morongole. These troops
were later relieved, however, for all but one detachment of the 9th Sudanese had
been in Uganda for over 14 months during which time they had been


away from their wives and families. Bimbashi Hobbs had already experienced
general trouble over the "women question" at Morongole and Kitgum and
predicted further difficulties at Ikoto if the 9th Sudanese were not given a well-
deserved furlough. The administration was most anxious to avoid any incident
that might alienate Chief Ikoto and his people, and so consequently in April a
detachment of the 12th Sudanese replaced those of the 9th at Ikoto.57
By the end of February 1916 the last Sudanese troops had evacuated Madial
and withdrawn from Uganda territory to Ikoto Post where they continued to
patrol the Sudan-Uganda border. Small detachments of the Uganda police
occupied Morongole, Kitgum, and Madial, and since the German forces never
really threatened Uganda again, these police units were sufficient to support the
government administrators, and Sudanese troops were no longer needed to keep
the peace in Northern Uganda.


ISee: Collins, R. O., "The Transfer of the Lado Enclave to the Anglo-Egyptian Sudan,
1910," Zaire, Vol. XIV, No. 2-3, pp. 193-210.
2"The Proposed Exchange of Territory Between Uganda and the Sudan," Notes by
Bimbashi Stigand, 5 December 1911, Int. II/36/310.
3Pearson to the Assistant Director of Intelligence, 5 December 1911, Int. II/36/310.
4SIR. (Sudan Intelligence Report), 223, February, 1913.
5SIR, 224, March, 1913.
6Extracts from the Report of Captain H. H. Kelly, R. E. on the Sudan-Uganda
Boundary Rectification Commission and the Reconnaissance of the Boma Plateau, SIR,
228, July, 1913, Appendix.
8Nalder, L. F., Mongalla Province Summary, pp. 28-9.
9"Extracts from the Report of Captain H. H. Kelly, R.E....," SIR 228, App.
11Nalder, op. cit. pp. 28-9.
12"Extracts from the Report of Captain H. H. Kelly, R.E....," SIR 228, App.
17Nalder, op. cit. pp. 28-9.
19SIR, 234, January, 1914.
20SIR, 261, April, 1916.
21SIR, 239, June, 1914.
22SIR, 237, April, 1914.
23Lt.-Col. H. Moyse-Bartlett, The King's African Rifles, Aldershot, 1956, pp. 260, 264.
24Acting Governor, Entebbe to Sir Reginald Wingate, Governor-General of the Sudan,
14 September 1914, Unnumbered Telegram, A.D.I./5/136.
25Wingate to G. F. Clayton, Sudan Government Agent, Cairo, 14 September 1914,
Tel. 193, A.D.I./5/136.
26Clayton to Wingate, 15 September 1914, Tel. 279, A.D.I./5/36.
27Adjutant-General to R. S. R. Owen, Governor, Mongalla Province, 15 September
1914, Unnumbered, A.D.I./5/136.
280wen to Adjutant General, Tel. 4602, 15 September and Tel. 4607, 17 September
1914, A.D.K./5/136.
29Wingate to Acting Governor, Entebbe, Tels. 198 & 199, 15 September; and
Adjutant-General to Owen, Tel. 55, 16 September 1914, A.D.I./5/136.
3OWingate to Clayton, Tel. 196, 16 September 1914, A.D.I./5/136.
32Acting Governor, Entebbe to Wingate, 17 September 1914, Unnumbered,
33Owen to Adjutant-General, Tel. 4609, 16 September and Adjutant-General to Owen,
Tel. 56, 17 September 1914, A.D.I./5/136.


34Clayton to Wingate, Tel. 290, 18 September 1914, A.D.I./5/136.
35Moyse-Bartlett, pp. 272-3.
36Sir Frederick Jackson, Governor, Entebbe, to Wingate, 21 September 1914, Tel. 308,
37Wingate to Jackson, 21 September 1914, Tel. 215, A.D.I./5/136.
38Enclosed in Clayton to Wingate, 22 September 1914, Tel. 308, A.D.I./5/136.
39Acting Governor, Entebbe, to Wingate, 22 September 1914, Unnumbered Tel.
40Acting Governor, Entebbe, to Wingate, 23 September 1914, Tel. 345, A.D.I./5/136.
41SIR, 245, December, 1914.
42SIR, 246, January, 1915. See: Collins, R. O., "The Turkana Patrol", The Uganda
Journal, No. 1, 1961.
43Bimbashi D. W. Fairbairn to Owen, 24 January 1915, U/4/15.
44SIR, 248, March, 1915.
45C. W. G. Eden, Provincial Commissioner, Northern Uganda, to Owen, 20 April
1915, Madial Patrol, Mongalla, 1/2/7.
46Owen to Eden, 10 May 1915, Madial Patrol, Mongalla, 1/2/7.
47Fairbairn to Owen, 6 April 1915, Madial Patrol, Mongalla, 1/2/7.
48See: Collins, "The Turkana Patrol," 1918.
49Eden to Owen, 20 April 1915, Madial Patrol, Mongalla, 1/2/7.
500wen to Eden, 10 May 1915,Madial Patrol, Mongalla, 1/2/7.
51Qa'immaqam C. Little, Assistant Adjutant-General to Owen, 13 June 1915, Madial
Patrol, Mongalla, 1/2/7.
52Jackson to Wingate, 20 June 1915, Madial Patrol, Mongalla, I/2/7.
53Owen to Fairbairn, 14 July, Owen to Pellew Wright, 14 July, and Owen to
Adjutant-General, 1 August 1915, Madial Patrol, Mongalla, 1/2/7.
54"Monthly Report of Kitgum District, September 1915," enclosed in Pellew Wright to
Owen, 6 October 1915, Madial Patrol, Mongalla, 1/2/7.
55Owen to Adjutant-General, Unnumbered Tel., 29 November 1915, A.D.I./5/136.
56I wen to Adjutant General, Unnumbered Tel., 29 November 1915, A.D.I./5/136.
57Hobbs to Owen, 19 December 1915, and Owen to Hobbs, 23 January 1916, Madial
Patrol, Mongalla, 1/2/7.



IN almost all agricultural societies the irregular rhythm of seedtime and harvest
prompts the storage and conservation of food. Only the fortunate few can afford
to live, like Aesop's grasshopper, in present plenty, with little thought for hungry
months ahead: for the majority the incidence of cold seasons or dry seasons
imposes the lessons of providence. This essay is concerned to outline traditional
methods by which the needs for the storage of agricultural produce have been
met in different parts of Uganda.
At the outset certain natural circumstances may usefully be considered. Plants
which will produce food fairly steadily over a period of time offer obvious
attractions to the cultivator. To some extent certain crops in favoured climates
will meet these conditions. The perennial banana grown under an equatorial
regime is one example. In the Lake Victoria region of Buganda if a banana garden
is well-established, carefully tended and planned to carry plants of different ages
and varieties, it can bear fruit throughout the greater part of a normal year.
Failing this ideal type of crop, an efficient form of storage may be performed by
the plant itself. Some of the perennial root crops such as the sweet potato, yams,
and colocasia store food effectively in their tubers for various lengths of time.
Most noteworthy in this respect are the bitter varieties of cassava-the tubers of
some of these remaining usable if unlifted for three years or even more after
attaining maturity. The increased use in the past 30 years of cassava and sweet
potatoes as famine reserve crops reflects these advantages. The grain crops and
the pulses must be gathered when ripe and they present greater problems of
storage. Grain storage can be treated in various ways but the broad alternative
lies between storing the grain in the head or threshing and grinding it into
flour before storage. Flour takes less room to store and is immediately available.
In temperate climates a winter season intervenes after harvest to check insect
activities, mould and fermentation, though rodent damage remains a problem.
In a warm tropical country such as Uganda there is much less check on such
hazards and little use of stone or brick in domestic construction. In the small
grains, such as finger millet and sorghum, storage in the head has advantages:
the small seeds are less easily lost and their husks play their natural part in
partially protecting the grain against premature damage from decay or insect
gnawing. The bitter glume of finger millet gives it a particular advantage against
insects. Traditionally, then, the Uganda peasant has usually chosen this latter
method of storage. The problems outlined are old ones for agriculturists and
it is therefore not surprising that traditional methods of meeting them should
be discernible.
On the first map (Fig. 1 at end of book) an attempt is made to relate these general
considerations to Uganda. The staple food crop zones are determined on the
basis of the percentage of the total acreage under food crop production which was
found to be under each crop in 1958. In each zone the crop named occupied over
30 per cent of the total acreage. In the cassava zone there is much reserve planting

to swell the acreage. and bananas form a very important supplementary food crop
in central Bunyoro and western Toro, whilst finger millet was until recent decades
staple in the West Nile plateau. It will be seen that the zone in which the banana
is staple bestrides the equator. Here the favourable rainfall distribution allows
the agriculturists to approach the optimum of year-long growth and fruition.
In south-west Busoga, southern Mengo, much of Masaka and parts of Ankole
over 50 per cent of all crops grown are perennial ones. Northwards and south-
wards from the equator rainfall regimes become modified towards distributions
exhibiting a single season of maximum rainfall and a single prolonged dry season.
The annual crops and particularly the cereals become more important. Thus
roughly to the north of the parallel 2 N. at least 75 per cent of the total acreage
of crops grown in 1958 was tilled in the central six months of the year. In broad
terms, then, the need for artificial methods of food storage is least in the
equatorial lands of the Lake Victoria basin and intensifies north-eastwards and
south-westwards towards the borders of Uganda. The use of granaries is more
general among the Nilotic and Nilo-Hamitic tribes-some of which have migrated
from areas of even greater rainfall concentration to the north of Uganda-than
it is among the Bantu tribes; most of which have been longer established in
regions of more favoured rainfall distribution.
The existing published studies of granaries in Uganda appear to be fragmen-
tary. Various authors treat incidentally of food storage among particular tribes,
e.g. Driberg (1923) and Lawrance (1957). Trowell and Wachsmann (1953) also
give some indications. There are geographical surveys for some neighboring
territories. Annaerts (1960) illustrates granary types in the eastern parts of the
Congo. In 1939 the East African Agricultural Journal began a series on native
methods of food storage, which produced useful general, illustrated accounts
concerning Kenya (1939), Nyasaland (Lawrence, 1940) and Tanganyika (Harris,
1941). The paper on Uganda (1941) was clearly produced under wartime
difficulties. It briefly treats the topic crop by crop for Uganda as a whole and
includes no illustrations.
The present inquiry into methods of food storage was undertaken in 1958-59
as part of a wider scheme of study of the subsistence crop geography of Uganda
which took the writer to most parts of Uganda. Where they are present, granaries
are fairly prominent features of the homesteads and it was thought that observa-
tions of their presence or absence would help to define the boundaries of zones
dependent upon grain crops and planted crops respectively. Only in the course
of such observation did the variety of granary types become evident to the writer.
Thereafter an attempt was made to distinguish the main types in the field and
to note their distribution. Storage within dwellings was not adequately covered
by field study and is not treated in this essay. Sorghum heads and seeds of
many types for later sowing are commonly stored within the huts or kitchens,
where the smoke certainly plays an important part in deterring insects. The
second map generalizes these preliminary observations (Fig. 2 at end of book). A
symbol has been used for each of the main types of granary seen and it is only
entered on the map in localities where granaries of that type were actually
observed or, in a few cases, where they have been reliably reported. The names
given to the types have no absolute significance.1
1They are in fact applied in a way analogous to that used in naming geological strata,
viz: derived from the dominant tribe of the area in which the type was observed and
sketched, much as, for example, the Cambrian derives its name from Cambria or the
Oxford clay is named from the area of its type occurrence and recognition.


The term "granary" is here used in a restrictive sense to denote storage
primarily intended for cereals, though they may of course also be used for other
crops. In Uganda finger millet and sorghum are the main crops thus housed.
Tentatively, eleven typical granary types have been distinguished (Fig. 3). Brief
descriptions of these may be given: -
1. "Lango" (Lango: Dero). Typified in Lango District and described there
by Driberg (1923). Also observed in central Acholi, East and West Madi, Aringa
County, West Nile, and Kibanda County, Bunyoro. A large, upright cylinder
of mud, cow dung and chopped grass plaster on a framework of upright poles,
twigs and reeds. Covered by a movable conical roof of grass.
In Acholi the type is perhaps less regular and "squatter" in shape. In West
Madi it is generally smaller and set up on a platform of poles.
2. "Teso" (Itesot: Edula). Typified in central Teso and described there by
Lawrance (1957). Also observed in northern Bukedi District, in Tororo County,
Bukedi, and Bugabula County, Busoga. A large spherical bowl with flared neck,
built of a mixture of mud, cow dung and fine grass (Loudetia simplex). The roof,
of flattish cone-shaped, is sometimes supported on poles.
Lawrance describes a portable wicker granary (Ekerei) of roughly similar size
and shape. In northern Bukedi the roofing appears usually to have a steeper
rake and to carry a "topknot" (Fig. 3).
3. "Karamojong". Typical of eastern Karamoja and described by Deshler
(1957) A conical, plaited basket mounted on a platform and with a roof of
flounced thatching.
4. "Lugbara". Observed and typified in the northern half of the West Nile
plateau. A cone-shaped granary with a broad base and narrow open neck,
constructed of mud and fine grass. The thatching is deep, and "shaggy" in
5. "Jonam". Observed in the environs of Pakwach. The writer knows little of
this type and its inclusion and naming must at present be regarded as tentative.
6. "Konjo". Observed on the eastern flanks and spurs of the Ruwenzori
mountains. A slightly tapering cylinder of plaited branches or reeds. Sometimes
mud-daubed on the outside. Placed on a low platform.
7. "Nyoro". (Lunyoro: Enguli). Observed in central Bunyoro. A fairly squat
cylinder of wickerwork, plastered with clay or a mixture of cow dung and wood
ashes. Mounted on a platform of poles. The thatched cover is made of elephant
grass, papyrus or banana leaves and stem. Said to hold about five sacks of
unthreshed grain.
This type of granary is thought to be a comparatively recent introduction into
Bunyoro and has apparent affinities with the "Lango" type. Formerly, in
Kabarega's time, there was much storage of grain in specially prepared pits,
called embiso. These were about 15 feet deep and covered over when in use.
8. "Gisu". (Lugisu: Birara). The common granary of Bugisu District. A
large, deep cylinder, rounded at the base. The base is woven and from it an
upright casing of bamboo stems, reeds or sorghum stems is erected. This is
bound together by fibre rings and the whole container is daubed with mud or
clay. The roof is supported on poles.
9. "Gwe". (Esyaki). Typified in Samia-Bugwe County of Bukedi. Observed in
Bukedi District, extending into south-eastern Teso and the lowlands margins of
Bugisu. The size of these granaries varies considerably. They are upright
cylinders mounted on a low platform. The container consists of a casing of reeds,
sticks, or stalks, bound together by fibre rings, and daubed with clay and cow



2 "TESO"

10 r-r-m 1 ,
3 3




B 0 Scal
8 GISU": Section and Scales

10. "HORORO''


6. "KONJO"


11 'KIGA"

FICURE 3. Granary Types. (The scale is approximately constant from one drawing
to another. The drawings are generalized, with some details such as roof
props omitted.)





9 "GWE





c 10 less
from container

to the ground
Molai cobs ,usPendled
from a tree branch g I
Container on a pole

c O-10 feet



Mud seal


c 1 (vii) LANGO "Tua" (vill) TESO Clay bowl

c. 3

(vi) LANGO 'Agoga' ii

(Ix & a) KIGEZI Woven booklet

FIGURE 4. Some supplementary methods of food storage.
(The scale varies from one drawing to another.)

dung on the outside, especially around the base, and the inside, to which a second
layer is sometimes added.
In southern Teso granaries of this type are said to be associated with Bantu
10. "Hororo". (Lunyankore: Ekitara). Noted and described from Kajara
County, Ankole. Also observed in south Rwampara County, Ankole and
Ruzhumbura County, Kigezi. The main body is woven of the leaf stalks of
swamp palms (Phoenix reclinata) or the stems of other plants. A horizontal
crosspiece, to assist access and handling, projects on either side of the cylinder.
The container is generally daubed with cow dung inside and plastered with mud
outside. An average-sized example holds about 800-1,000 lb. of grain.
11. "Kiga". (Lukiga: Ekitara). Noted in central and southern Kigezi District.
A barrel-shaped granary sometimes of considerable size. Constructed of the
flexible stems of various plants and trees; these are bound together and the
inside of the container is smeared with cow dung. The roof is supported on poles
and thatched with grass collected in swamps.
From these descriptions it will be evident that the granaries consist of three
main components: a base or platform, a container, and a roof. The shape and
construction of the container are the most diagnostic features in attempting to
differentiate granary types. For any one type, the structure of the platform can
vary considerably from one granary to another and the roofs may or may not be
separately supported by poles. Size, also, is quite variable. Sometimes granaries
of unusually large proportions denote the importance of their owner. A prelimi-
nary attempt has been made on Fig 2 to suggest the areas within which particular
granaries are typical. Buganda, and most of Busoga, Toro and Ankole are shown
as comprising an area in which granaries are not typical features of the home-
steads. This largely corresponds to the dominance of the banana as a staple-
the fruits mature over quite a long portion of the year but do not keep many
days once the bunches are picked. The rainfall typically exhibits a double
maximum with no lengthy dry season intervening and this often permits two
harvests of annual crops to supplement the produce of the perennials. In the
past times of rural insecurity, use was also made of pit storage for finger millet in
Toro and Ankole. It will be noted that granaries only become frequent towards
the margins of the contiguous areas of Bantu occupation in Uganda. Most
widespread of individual granary types appears to be the "Lango" style, which
is closely associated with the areas of Nilotic occupation. Indeed, the granary
types show close linkages with particular tribes. This could be expected if they
represent traditional elements of their respective material cultures. Concerning
the antiquity of the types, Lawrance holds that in Teso the forms and modes of
construction of granaries have not changed since the coming of Europeans. In
Bugisu, however, there is some evidence for the historical development of
granary forms. La Fontaine (1959) states that Gisu did not in past time have a
knowledge of basketry and learnt the art through trade with, and imitation of,
the Sebei, though she gives no indication of the time scale implied. The relatively
recent adoption of granaries in Bunyoro has already been referred to.
At present the writer does not discern any sure relationship between granary
forms or methods of construction on the one hand and tribal affiliations or
natural environments on the other. There may be some significance in the wide-
spread use of clay, grass and cow dung as materials in the Lake Kyoga basin
and the northern plateau areas. One also has an impression that the average size
of granaries is greatest in the hilly country of Bugisu and Kigezi, but such

observations are most tentative. Neither is it easy to draw correlations between
the size or numbers of granaries observed at individual homesteads and hamlets
and the prosperity and extensiveness of agricultural operations. Different factors
such as the range of cereal crops, the number of wives, the size of the household,
and the possibility that the produce of separate fields is stored separately all help
to preclude general judgments. Until very recently the storage of a proportion
of the grain crop in communal groups of granaries, placed at the local administra-
tive headquarters, was compulsory in the Northern Province and many such
groups are still to be seen.
Enquiry suggests that the average life of granaries is about five to ten years.
Damage is caused by climatic factors, especially wind and hail; by animals,
especially rats and birds; by termites; and by fire. In past times granaries must
have been vulnerable to enemy raids and malicious destruction by rivals, though
their close attachment to the homestead would offer some protection. The use
of pits was obviously related to this hazard. In some areas of Uganda a degree
of specialization in the construction of granaries is evident. In Bunyoro and
south-west Ankole, for example, the main body of the granaries is usually
bought, whilst the hiring of expert services for construction on the spot is
mentioned from Bukedi and Bugisu. There are also interesting variations in the
social control of granaries which deserve further enquiry. In Bugisu or Lango
it would apparently be regarded as most unbecoming for men to use the
granaries on their own account, whilst in Kigezi it is reported that a wife would
be "shamed" were she seen doing so, as it is the proper thing for men or young
children to do it.
Maize cobs are sometimes stored in granaries. A traditional alternative method
is shown (Fig. 4, i) but is said to be little used today, though cobs will often be
seen suspended beneath the eaves of huts or under the roofs in many parts of
Uganda. Other methods of storage are used for three other types of produce: -
(a) Beans, peas, and groundnuts. The pulses will not keep long periods.
Groundnuts in the shell will do better.
(b) Selected seed for the next sowing.
(c) Simsim and other fine seeds.
Some methods of storing these are indicated in the remaining sketches (Fig. 4).
The use of small cone-shaped containers, set on poles and thatched, (ii) was seen
in Aringa and Terego counties of West Nile, but a similar method was described
in Lango by Driberg. Capsules sheathed in banana fibre (iii) and used for beans,
peas, and groundnuts especially, were observed over a wide area including
Bunyoro, southern Teso, northern Busoga, central Acholi and the southern
parts of West Nile District. They are also illustrated from Tanganyika
by Harris (1941) and from Nyasaland by Lawrence (1940). In Uganda
their use is apparently declining. Formerly they must have been convenient
if families moved or households were shifted but they are correspondingly
easy to steal. It seems likely that the better provision for storage within
modem houses and the use of new and effective containers, such as debis,
now makes them less necessary. Variants on this method from southern Uganda
are illustrated (iv & v). Clay pots and woven baskets, suitable for the storage of
seed, simsim, pulses etc., are exemplified from Lango, Teso and Kigezi (vii-x).
Such smaller containers are often placed under the eaves and against the walls
of huts. They are less easily observed than the larger granaries and a full
description of types and their distribution would demand a longer and closer
study than the present essay and the gathering of much more information.

The statement just made is important. This essay is certainly an interim
account. The frequent use in it of qualifying phrases will have been evident. This
is largely a personal reconnaissance covering Uganda as a whole; but many
individual readers will be in a position to supplement and to correct the
impressions presented here from their intimate knowledge of particular areas
and tribes. Such help would be most valuable: confirmation of statements would
be gratifying; but criticism and correction even more useful. A questionnaire
was prepared in April, 1960 and distributed among some of the Uganda students
at Makerere University College. The replies received, although limited in number,
have provided valuable information. Further assistance would be most welcome.
Two notes of advice may be justified. First, with regard to distribution one
should record only localities where types described have been actually seen to
occur. Secondly, with the smaller containers, unless you know the area and tribe
quite intimately, care is necessary to distinguish some food storage containers
from chicken houses, beehives and other similar constructions. It was salutary
to the writer to receive the suggestions that Fig. 4 (ii) might represent a dovecote,
whilst the small storehouse shown in Fig. 4 (vii), though supported by a local
informant, is distinctly like the Otogo, or bachelor huts, pictured and described
by Driberg. Chicken houses also sometimes display attractive adaptations of
local constructional styles!
In conclusion it may be well to indicate certain applications which this study
may have beyond being an inquiry for its own sake. First, the presence or
absence of granaries does appear in Uganda to be related to facts of climate,
agricultural practice and tribal distribution. Granaries are generally evident in
travel through a region and on good aerial photographs and may thus be of
assistance in preliminary regional studies of big areas. Secondly, it has been
suggested that the granary forms are linked to tribal cultures and if this is
correct the recognition of anomalous granary types in a given area or the spread
of types beyond their traditional limits may give useful hints of population
movements which can be investigated more closely by other means. Thus, the
"Kiga" type of granary is apparently spreading to the resettlement areas of North
Kigezi; the overlapping of "Teso" and "Gwe" types in the Eastern Province
reflect tribal factors; and the writer was able to infer the presence of Bantu
settlers in the Namalu irrigation area of southern Karamoja from the types of
granary seen. This was subsequently confirmed, and it must be stressed that
such observations are only hints and need independent confirmation. Thirdly,
efficient food storage is a matter of practical importance. It will be a long time
before peasant methods are superseded by modern alternatives. Worthington in
his development plan for Uganda (1946) estimated that under the storage
arrangements then general as much as one quarter of the food crops produced
in Uganda was lost through the ravages of insect pests and rodents. This may
appear rather a high estimate but the fact remains that all losses subsequent to
harvest represent a regrettable diminution of the hard efforts of tillage, cultivation
and gathering. Michelmore (1955) has given an account of entomological factors
relating to loss of food in storage. It seems unlikely that each type of granary
and container described in this essay is equally well adapted to the environment
and to the biological hazards which are present. In certain cases simple modifica-
tions in design or construction might be advantageous. For example, effective
use is made in the Bukoba District of Tanganyika of metal discs on the vertical
supports of granary platforms. Harris (1941) illustrates an ingenious use of the
necks of gourds for a similar purpose in Tanganyika and Annaerts (1960) shows


the application of inverted clay "collars" to the supports of granaries among the
Madi of the north-eastern Congo. Such devices do much to prevent damage by
rats. Thus a full account of the main variants of granary construction might still
be a useful step towards the better conservation of Uganda's seasonal food

The travel within Uganda upon which this essay is based was made possible by a
generous grant from the Colonial Economic Research Committee for a wider scheme
of research on subsistence agriculture. I am grateful to the following former students of
Makerere University College for their asstitance: Miss Grace B. Watuwa, Messrs.
C. Baryomunsi, E. F. Fagayo, E. S. Katarikawe, S. Kisoro and D. J. Odanga.

Annaerts, J. (1960). Contribution & l'etude g6ographique de l'habitat et de l'habitation
indig&nes en milieu rural dans les provinces oriental et du Kivu. Brussels, Acad6mie
Royale des Sciences d'Outre-Mer.
Deshler, W. W. (1957). The Dodos country: a study of indigenous settlement in a
semi-arid area of Uganda. Unpublished Ph.D. thesis of the University of Maryland.
Driberg, J. H. (1923). The Lango: a Nilotic tribe of Uganda. London, Fisher Unwin.
Harris, W. V. (1941). Native methods of food storage in Tanganyika. (East African
Agricultural Journal, Vol. 6, 135-138.)
La Fontaine, J. S. (1959). The Gisu of Uganda. London, International African Institute.
(Ethnographic Survey of Africa, East Central Africa, Part X.)
Lawrance, J. C. D. (1957). The Iteso: fifty years of change in a Nilo-Hamitic tribe of
Uganda. London, O.U.P.
Lawrence, E. (1940). Native methods of food storage in Nyasaland. (East African Agri-
cultural Journal, Vol. 5, 376-379.)
McMaster, D. N. (1962). A subsistence crop geography of Uganda. Bude, Geographical
Publications Ltd. (International Geographical Union, World Land Use Survey,
Occasional Paper No. 2.)
Michelmore, A. P. G. (1955). Food storage problems in Uganda in relation to insect
pests. (East African Agricultural Journal, Vol. 21, 65-68.)
Native methods of food storage in Kenya. (1939) (East African Agricultural Journal. Vol.
5, 99-103.)
Native methods of food storage in Uganda. (1941) (East African Agricultural Journal,
Vol. 7, 74-76.)
Trowell, M. and K. P. Wachsmann. (1953). Tribal crafts of Uganda. London, O.U.P.
Worthington, E. B. (1946). A development plan for Uganda. Entebbe, Government


by W. W. BisHop1

The park is largely underlain by Pleistocene sediments of various types, within
a structural framework of part of the Western Rift Valley. The exposed strata
consist principally of unconsolidated gravels, sands, silts and clays, plus fine-
grained calcareous volcanic ashes. Rejuvenation has taken place and the main
drainage to lakes George-Edward and the Kazinga Channel is incised from
100 to 300 feet below a general surface developed upon former lake deposits.
There is some variation in the lithology of the sediments but in general the
deposits exposed within the park can be classified under three main headings.
1. Kaiso Series. These occur as flat lying beds or dip gently in an easterly
direction and outcrop in a belt one mile or less in width at the western end of
the Kazinga Channel and for some ten miles south of Kazinga village, in low
cliffs forming the eastern boundary of Lake Edward. They consist of interbedded,
fine-grained, micaceous sands, silts and clays ranging from pale buff to light or
bluish grey in colour. Ferruginous fossiliferous horizons, the mammalian content
of which suggests a lower Pleistocene age for the deposits, occur as discontinuous
lenses up to two feet in thickness. They are usually sandy, sometimes oolitic or
flaggy, and outcrop as ironstone bands.
The silts and sands are only slightly permeable, while the relatively imperme-
able clays become soft when moist and give rise to slumping and the formation
of steep faces along joint planes. During dry seasons the joint cracks open to
depths of up to ten or fifteen feet from the surface and greatly assist undercutting
and erosion at the onset of further rains. The sandy ironstones facilitate the
movement of groundwater and, owing to their greater cohesion, frequently give
rise to benches on spurs. They also form steps in erosion gullies due to under-
cutting and slumping of the overlying clays, silts and fine sands.
The series as a whole is lacustrine in origin, fine-grained and although contain-
ing some permeable sandy bands is more resistant to erosion, partly by virtue of
induration because of greater age, than the succeeding Semliki series deposits.
2. Semliki Series. The Kaiso series dip to the east beneath the Semliki series
which seldom rise more than 100 feet above Lake Edward and possess a very
regular flat upper surface. The deposits date from the middle Pleistocene and
are in marked contrast to the older, Kaiso series from which it is suggested
(Bishop and Posnansky 1960), they are separated in time by a period of renewed
movement of the Rift Valley boundary faults. As a result, the rivers flowing into
the Semliki series lake were much faster moving and carried considerable
quantities of detritus giving rise to coarse, fluvio-lacustrine, false-bedded sands
and gravels. They contain a very small clay portion but lack pronounced clay
and silt horizons, and are largely unconsolidated and thus easily eroded. Although
II am indebted to -the Director, Geological Survey of Uganda for permission to
publish this article and to numerous members of the staff of the Uganda Parks for their
kind co-operation.

capped almost invariably by a thin layer of windblown volcanic ash they underlie
the whole central part of the park east of the Kaiso series ridge and outside the
area of the explosion craters.
3. Upper Pleistocene Deposits: (a) River Gravels. Local patches of coarse, cross-
bedded, dirty, unconsolidated, river gravel often with limonitic staining and
concretions, occur up to 100 feet above Lake Edward as evidence of former high
levels of the Kazinga Channel and other rivers flowing into Lake Edward.
(b) Volcanic Deposits. These range from coarse agglomerates in the immediate
vicinity of explosion craters in both the Katwe and Bunyaruguru areas, to a
variable layer up to 20 feet or more in thickness, of fine windborne ash. This
dates from the last explosive phase of volcanic activity which seems to have
occurred about 10,000 years before the present (de Heinzelin, 1957). The sub-
aerial ash overlies all the other deposits except terrace gravels 30 feet or less
above the present river. It occurs throughout the game park with the exception of
those portions to the north of Lake George and that south of the Nyamweru
River. The deposits are extremely permeable, have only a poor soil development
and near the surface are often transformed into a very resistant calcareous crust
while other limestone bands occur at various depths due to the concentration of
secondary calcium carbonate by ground water action.
(c) Gully Fill. The most recent deposits in the area consist of the filling of
former gullies cut in the Kaiso, Semliki and volcanic series. The fill consists of
a reworked, heterogeneous mixture of all or some of the earlier deposits, depend-
ing upon locality. The age of these old gullies is not certain but they are
undoubtedly later than the volcanic ash (i.e. less than 10,000 years B.P.) and
they are possibly very recent. Exposures of the deposits are well seen as the
modern erosion gullies are frequently aligned approximately on the lines of the
earlier ones.
Thus the situation of these three groups of virtually unconsolidated deposits, in
cliffs rising almost 150 feet above the base-level of the Kazinga Channel and
Lake Edward, yields an ideal geological setting for gully erosion. Indeed this is
the natural process to be expected in such an area and under the present rainfall
regime. The gullying is gradually readjusting the balance between landform and
drainage as the unconsolidated, "perched" lake deposits are dissected by water
flow to Lake Edward. Similar deposits are suffering from identical processes of
degradation in the Murchison Falls Game Park.

The erosion in the Queen Elizabeth Park may be classified as of two principal
(a) Sheet erosion. This consists of removal of soil by the action of direct
surface run-off following rainfall and may take place in areas of very gradual
slope. It is not proposed to discuss in detail erosion of this type. The flat upper
surface of the lake beds, away from established gullies and drainage channels,
together with the virtually universal mantle of porous volcanic ash, makes the
sheet erosion problem less critical than that posed by the gullying. It is worthy
of note however that the permeable nature of the volcanic and lacustrine sedi-
ments, together with the low water-table caused by the incised drainage, results
in the development of thin soils supporting only poor grassland with occasional
xerophitic bushes. Such vegetation cover is likely to be removed rapidly by
grazing animals, particularly from spur tops between river valleys with large
concentrations of hippopotamus.



5. -.

''; FIGURE 1. Commencement of
collapse into under-
ground tunnel some 15
feet from the parent
t gully seen in the back-
ground of the photo-
graph. Kaiso Series,
S 1 M.- i Peninsula
Queen Elizabeth Park.


FIG'HE 2. Gully undermin-
ing bush cover. Kaiso
Series, Mweya Penin-
sula, Queen Elizabeth

-.4 .

,i .; *f* .

(b) Gully Erosion. Rapid gully erosion is taking place in many areas of the
park. As a result of the unconsolidated strata, gullies develop wherever there
are steep slopes down to the present base level of erosion at the Kazinga Channel
or Lake Edward. The gullies occur in both Kaiso and Semliki series beds and
are principally controlled by undercutting from sub-surface water flow. Rainfall
is rapidly absorbed by the volcanic and lacustrine deposits which outcrop over
large areas of the park. Little effect is normally produced by small rainstorms and
the ground-water from these gradually finds its way by percolation into the main
drainage channels. These by virtue of their depth and of being nearer to the
water-table are usually clothed with and locally surrounded by more dense
vegetation. The intermittent streams that flow in these valley bottoms, cause
little erosion as they flow only a few feet above the local base-level.
However, during heavy rainfall the water-table may rise rapidly and when
supply exceeds drainage by a sufficient amount "perched water-tables" build up
above layers of relatively impermeable strata. If the rainfall is sufficiently con-
centrated the surface of the water-table develops a slope towards the lines of
drainage. Such a build up of ground-water results in rapid subterranean flow
into gully heads, undercutting both unconsolidated beds and any surface cover
of vegetation. The gullying proceeds in steps above each relatively impermeable
band and temporary tunnels up to three feet in diameter can be seen leading off
from the heads of active gullies. With the collapse of tunnels gullies may advance
headwards as much as 10 or 15 feet in a single storm. Fig. 1 shows the commence-
ment of collapse into an underground tunnel of considerable dimensions at a
point some 15 feet from the parent gully. The Kaiso deposits are rather worse
than the Semliki beds in this respect because of the more uniformly porous
nature of the latter. Gully erosion is more pronounced along the outcrop of
the Kaiso beds, particularly where steep faces practically surround a small
area of these deposits as on the Mweya promontory.
It seems probable that appreciable gullying only takes place during periods of
heavy concentrated rainfall. An analysis of the available rainfall figures for the
Mweya Peninsula (Table I) shows that in any one year 35 per cent to 60 per
cent of the total rainfall is in the form of heavy storms occurring on only
twelve days per year. Gully erosion is most likely to take place on such days
and the figures suggest that 1955 and 1956 were years of minimum headward
erosion while 1954 and 1958 were years of maximum retreat of gully heads.

Total Rainfall in Number of
Year rainfall storms of days with
/annum more than more than
0.5' 0.5'
1954 .. 29.21 16.97 16
1955 .. 23.55 8.75 7
1956 .. 23.09 8.62 13
1957 .. 25.44 10.83 13
1958 .. 24.95 14.41 13

The majority of gullies have developed along the lines of earlier ones which
have previously been cut down to a depth similar to the present base level. These
became naturally stabilised, were later infilled and are now being re-excavated in

a second erosion cycle. This suggests that the conditions causing the present
cycle need not necessarily be linked with an upset of the natural balance as a
result of the formation of a game park.
Similar erosion gullies to those in the game park can be seen not only in the
vicinity of Paraa in the Murchison Falls Park but also in areas outside the
game parks including the Kaiso flats in Bunyoro, several coastal areas of
Kavirondo and the Kagera valley in Ankole. In every case the gullies are
developed in unconsolidated lacustrine Pleistocene deposits in areas with a steep
descent to the local base level of drainage. All the areas contain some gullies
which are actively developing but exhibit others in all stages of decay, infilling,
and re-excavation.
From my observations it seems that the cycle of gully erosion described above,
arising initially from unbalanced conditions in the geological setting, may be
aggravated by the following:
(a) Game Tracks. Paths used by game are frequently channels down which
rain-water flows and cuts ever deepening gullies. The hippopotamus by regular
use of one or more routes to and from the water undoubtedly aggravates the
gullying. However it must be conceded in his defence that the "hippo" has to
make use of the pre-existing depressions which breach the steep slopes overlooking
the Kazinga Channel and Lake Edward, and along which access to the water is
comparatively simple. The hippopotamus probably only modifies existing valleys
and gullies.
(b) Overgrazing. The full evaluation of the effect of overgrazing on sheet
erosion and run off must await the results of experiments to try to improve the
vegetation cover on certain interfluves within the park. There is no doubt,
however, that the presence or absence of a thick grass or bush cover has little
effect in preventing the formation, or retarding the development, of gullies of
the type that exist at present in the Queen Elizabeth Park. The majority of
the water flow into the gullies is by underground movement and not by surface
run-off and many of the existing gullies can be seen undermining and killing a
thick cover of vegetation as shown in Fig. 2.
In the United States the rehabilitation of grassland and the planting of bushes
has played a large part in preventing erosion and reclaiming eroded areas but
there the problem was to prevent or divert rapid surface run-off which was
cutting the deep gullies. It seems unlikely that such an approach would
completely resolve the problems in the Uganda parks where underground sapping
is so important in the gullying mechanism.
(c) Roads and Drainage Ditches. The building of roads, particularly where
adjoining ditches are required for drainage purposes, may be a fertile source
of new gullies. This is only dangerous where roads cease to follow contours and
have to descend steep slopes. Few examples of this are present in the game park
but a good illustration is the road from Mweya Lodge to the landing place which
already has deep flanking gullies.
In conclusion it appears that the gullying in the Game Parks is caused because
there are areas where erosion is invited by the rainfall regime and the geologi-
cal and physiographic background. It is emphasised that gullying is the natural
process under which the steep slopes are gradually being eliminated and stabilised.
It is difficult to assess at present the part played in accelerating these natural
processes by animal populations and human activities. However, both may be
proved guilty of acting as trigger mechanisms which initiate new cycles in areas
where the natural balance is already rather precarious.


Bishop, W. W. and Posnansky, M. (1960). Pleistocene Environments and Early Man in
Uganda, Uganda Journal, Vol. 24, No. 1, pp. 44-61.
de Heinzelin, J. (1957). Les Fouilles d'Ishango. Inst. des Pares Nationaux du Congo
Beige, Brussels.



DURING an ecological study of the Karimojong tribe carried out from 1955-
1958, Akarimojong tree names were found to be generally known and used by
the Karimojong, and appear in most cases to be as specific as Latin names. Thus,
the following list of Karamojong names for trees can be of considerable use to the
non-botanist working in Karamoja.
In only three cases was one name sometimes used for two closely related
species, but a second name distinguished them. Thus Ziziphus mauritiana and
Z. abyssinica are both sometimes called "Esilang" but Z. mauritiama is also
called "Ekalale". Commiphora africana and C. madagascariense are both called
"Ekadeli", but C. madagascariense is also called "Lokimeta". Combretum guienzii
and C. molle are both called "Ekuworro", but C. guienzii is also called "Ekuyon".
The singular Akarimojong names have been listed in all cases, since they are
easier for people not acquainted with the language to recognize, and are generally
used when a plant name is asked. The plurals all begin with "ng" and vary
greatly in their formation (e.g. s. "Eboborei" pl. "Ngiboborio", s. "Etirai" pl.
"Ngitira", s. "Epiye" pl. "Ngipiyon", s. "Epeduru" pl. "Ngipedurr"). They have
not been listed.
I believe the following list to be fairly complete for southern Karimojong
country. It does not include the trees found on the mountains, Napak, Kadam
and Moroto-areas not used by the Karimojong. All names were recorded from
informants of the Ngipian section in southern Karamoja. The Ngibokora and
Ngimatheniko sections of the Karimojong tribe show slight dialect differences, so
diversity might be encountered in some of the northern Karimojong names of
less common trees.
Latin names are listed as given by the Royal Botanic Gardens, Kew: and
the numbers refer to identified specimens, which have been retained by Kew.
Since several genera have been revised recently, some of these names do not
agree with those listed in Eggeling and Dale (1951), which I have listed as
synonyms. Otherwise, I have not listed synonomy.
The sound written as "s" is pronounced between the English "s" and "th".
"Ng" is a single sound as in "singing". "J" is pronounced almost like "dj".
Occasionally, Karimojong may add an extra syllable to a polysyllabic word
("Engerengeroi-Erengerengeroi"); or transpose syllables "Abolokoch-Akolo-
No. Vernacular Name Latin Name Growth form
477 AARACHUCH .. .. Acalyphafruticosa Forsk .. Shrub.
407 AITARENG .. .. .. Mimusops hummel Hochst. ex DC. Tree.
346 ALOMORU, ELAMORU .. Plectranthus cyaneus Guerke .. Tree.
185 EBELEBELEBWOIT, ELEBELEBWOIT Flaucourtia indica (Burm. f.)
406 Merr. .. .. .. Shrub/tree






204 J


326 J
143 A


211 EKI

165 1




409 EKUMOIT ..











Name Latin Name Growth form
Ficus sycamorus, L. .. .. Tree.
Croton zambesicus Muell. Arg. .. Shrub/tree.
Ficus platyphylla Del. Tree.
Alibizia coriaria Welw. ex Oliv. Tree.
.Kigelia aethiopica Decne. .. Tree.
Acacia albida Del. (syn. A. seyal
Del. var. multijuga) .. Tree.
Maerua angolensis DC. .. Shrub/tree.
Acacia hockii De Wild .. Tree.
Allophylus rubifolius Engl. .. Shrub seldom
tree size.
.Popowia buchananii (Engl.) Engl.
et Diels .. Tree.
Maytenus senegalensis (Lam.)
Exell .. .. .. Shrub/tree.
.Commiphora africana (Am.) Engl.
(syn. C. pilosa (Engl.) .. Shrub/tree.
Commiphora madagascariense Jacq
(syn. C. habessinica (Berg.)
Engl.) .. .. .. Small tree.
A Rhus natalensis Bernh. ex Krauss Bush/tree.
.fruit of Sclerocarya birrea (A.
Rich.) Hochst. (tree called
ENGAIMU) .. Fruit.
Ziziphus mauritiana Lam. .. Shrub/tree.
Grewia bicolor A. Juss. var.
tephrodermis (K.Schum.) Burret Shrub/tree.
Carissa edulis Vahl .. .. Bush/tree.
.Albizia anthelmintica Brongn. .. Bush/tree.
.Acacia nilotica (L.) Willd. ex Del.
probably subsp. subalata
(Vatke) Brenan .. .. Tree.
Vitex doniana Sweet .. .. Tree.
.Craibia laurentii (De Wild.) De
Wild. .. .. .. Shrub/tree.
.Harrisonia abyssinica Oliv. .. Shrub/tree.
.Ficus sonderi Miq. .. .. Tree.
.Combretum ternifolium Engl. et
Diels .. .. .. Tree.
Harrisonia abyssinica Olib. .. Shrub/tree.
.Acacia hockii De Wild .. Tree.
Terminalia mollis Lawson

KOBEI .. Combretum aculeatum Vent. .. Shrub

Teclea nobilis Del. .. Shrub/tree.
'DOKODWOI Acacia senegal (L.) Willd. sensu
lato .. .. Bush/tree.
EETACHIT.. Gardenia jovis- tonantis Welw. ex
Hiern .. .. .. Shrub/tree.
.Balanites aegyptiaca Del. .. Tree.
.Acacia seyal Del. var. seyal .. Tree.
Solanum sp. perhaps S. renschii
Vatke .Stringy Shrub
Terminalia glaucescens Plansch.
ex Benth .Medium Tree.
.Diospyros mespiliformis Hochst.
ex A. DC. .. .. Tree.



No. Vern

186f EKURAU ..
333 EKUWORR ..



139 EMUS

acular Name








247#w" EPAPAI


195 -EPIYE
414 3
215 J


Latin Name

Acacia brevispica Harms

.Combretum molle R. Br. ex G.

.. Combretum gueinzii Sond.
.Syzygium guieense (Willd.) DC.
Dombeya rotundifolia (Hochst.)
Planch .
.. Combretum gueinzii Sond.

.. .. Albizia amara (Roxb.) Boiv.
.Fluggea microcarpa Blume
.Ximenia caffra Sond.
BBELEBWOIT Flacourtia indica (Burm. f.) Merr.
Adenium somalense (Balf. f.)
.Fruit of Vitex doniana Sweet
(tree called EKARUKWAI)
.Plectranthus cyaneus Guerke .

.Lannea fulva Engl..
Vangueria acutiloba Robyns
.Acacia gerrardii Benth. var.
gerrardii ..
.Ficus ingens Miq. .
.Euclea schimperi (DC.) Dandy

Growth form










.Heeria reticulata (Bak. f.) Engl. Tree.

.Sclerocarya birrea (A. Rich.)
Hochst .. .. .. Tree.
GERENGEROI Cassia singueana Del. .. Tree.
.Grewia kakothamnos K. Schum. Shrub.
.Grewia tenax (Forsk) Fiori .. Shrub, some-
.Piliostigma thonningii (Schumach)
Milne-Redh. .. .. Shrub/tree.
trychnos spinosa Lam. (A. Rich.)
E. A. Bruce subsp. lokua .. Shrub/tree.
Tamarindus indica L. .. Tree.

Terminalia brownii Fres.

.Canthium sp. nov.
Grewia villosa Willd

.Euphorbia candelabrum
ex Kotschy
.Hoslundia opposite Vahl

.. Tree.


Small to large

Succulent tree.


Vernacular Name

Latin Name

411 EPUU



424 ERoRI




217 ETOoj


293 EWAPETo, Ew

















Calotropis procera (Ait.) Ait. f. Shrub.
Terminalia spinosa Engl. .. Tree.
... .. Albizia amara (Roxb.) Boiv.
subsp. sericocephala (Benth.)
Brenan .. Shrub/tre
.Acacia mellifera (Vahl.) Benth.
subsp. mellifera .. .. Shrub/tre
GERENGEROI Cassia singueana Del. .. Tree.
Capparis tomentosa Lam. .. Climbing
.. Boscia salicifolia Oliv. .. Tree.
.Ficus ingens Miq. .. Tree.
Ormocarpum trichocarpum (Taub.)
Harms .. .. .. Shrub/trei
.. Erythrina abyssinica Lam. .. Tree.
Ziziphus abyssinica Hochst. ex A.
Rich .. Shrub/tre,
.Ziziphus mauritiana Lam. .Shrub/tree
Tinnea aethiopica Kotschy and
Peyr. .. .. .. Shrub.
.. Strychnos innocua Del. subsp.
et var. innocua Arg. Shrub/trei
Phyllanthus sepialis Muell. Arg. Shrub.
Dichrostachys cinerea (L.)
Wight et Am. .. .. Tree.
.Acacia tortillis (Forssk.) Chris-
tensen .. .. .. Tree.
.Lannea humilis (Oliv.) Engl. .. Shrub/tree
Vangueria apiculata K. Schum Shrub/trei
INAN .. Garcinia livingstonei T. Anders.
vel. aff. .. Shrub/tre,
Terminalia glaucescens Planch.
ex Benth .. .. Tree.
.Acalypha bipartita Muell. Arg. Clinging s
Combretum hereroense Schinz .. Tree.
Fagara chalybea (Engl.) Engl. .. Tree.
.Acacia drepanolobium Harms ex
Sjostedt .. .. .. Shrub/tre<
Opilia sp. .. .. .Viny shru
LNAN Garcinia livingstoneii T. Anders.
vel. aff. .. Shrub/tre
Tarenna graveolens (S. Moore)
Brem. .. . .. Shrub.
Commiphora madagascariense Jacq
(syn. C. habessinica) .. Tree.
Osyris compressa (Berg.) A. DC. Shrub.

IKOBEI .. Combretum aculeatum Vent. .. Shrub.

DOKODWOI Acacia senegal (L.) Willd. sensu
lato .. .. Bush/tree.
Acacia sieberiana DC. var.
sieberiana .. .. Tree.











Growth form



Eggeling, William J., and Ivan R. Dale. (1941). The Indigeneous Trees of the Uganda
Protectorate. 2nd edition. The University Press, Glasgow.
Turrill, W. B., and E. Milne-Redhead. (1953-1960). Flora of Tropical East Africa,
Crown Agents for Oversea Governments and Administrations, London.


1This information was collected during a field research study of Karimojong ecology
lasting from January 1956 to September 1958. I would like to express my gratitude to
the following sources for financing this work: Guggenheim Foundation; American
Academy of Arts and Sciences; American Philosophical Society of Sigma XI; Post-
doctoral Fulbright Fellowship. I am grateful to the American Association of University
Women for the award of a fellowship enabling me to spend 1958-1959 in preliminary
preparation of materials gathered in Karamoja; and the Society of Sigma XI for a
further grant for the preparation for publication of field data.
2"Eterai" was used for Allophylus rubifdlius Engl. by one Karimojong, but "Egogong"
seems the correct name.
3Collected top of Napak only-not on the plain.
4Possibly a Tepeth name.
5A name similar to this (esyapot) was used for several species of small riverain shrubs
e.g. Clerodendrum myricoides (Hochst.) R.Br. ex Vatke.
6Not generally known.



THE published authorities on Kiganda religion are Gorju (1920), Kagwa (1901,
1905 and 1908), Nsimbi (1956) and Roscoe (1911). What follows is an attempt to
piece together their evidence with such information as I have myself been able to
obtain-by occasional incursions rather than serious research-about the survival
of traditional practices and beliefs in contemporary Buganda. Dr. Audrey
Richards and Dr. Martin Southwold have greatly helped by their criticisms of
my original manuscript.
Of recent books on African religion, Lienhardt (1961) is immensely stimulating
in his interpretations. Middleton (1960) and Wilson (1957) deal fully with societies
in which ancestral ghosts are of paramount importance. Evans-Pritchard (1956)
insists that, in the present state of information, it is important to present a large
number of studies of individual societies before it is legitimate to attempt any
general theory; and this essay is presented in the hope that it will stimulate
Baganda, to whom the material is more readily available, to collect it as carefully
as possible. For those who are interested in the comparative aspects of this study
it is relevant that, in contrast with other societies recently studied in detail, the
Baganda have a highly centralized monarchy; and there is a sense in which
bananas play the part claimed in other societies by cattle.
Buganda began as a number of territorially distinct clans, divided into sub-
clans and lineages, each headed by a mutaka. Traditionally it was Kintu who
imposedd a centralized political authority; and all Baganda-whether or not their
clans were in the country before Kintu-are his "grandchildren". All kings of
Buganda (bakabaka) are said to be descended from him through the male line;
and he, with them, is the central symbol of Buganda's unity as a nation. The
centralization involved, in due course, the appointment of direct agents of the
Kabaka, who came to be known as bakungu: Katikkiro (as Chief Minister),
Kimbugwe (whose functions were mainly ritual) and the ssaza chiefs, exercising
authority over particular areas. Southwold (1961) analyses the evidence of the
extent to which these chieftainships were hereditary in particular clans-of how
far the Kabaka tried to rule through the traditional authorities. But by 1877 when
the first missionaries came to Buganda, while Kasujju and Katambala were still
hereditary, Mugema was the only clan head to be also a hereditary ssasa chief;
and Kagwal records an attempt by Kabaka Mutesa I to appoint to that post a
man of another clan. Joswa Kate, who became Mugema in 1889, told his son
that every Kabaka within living memory had used the bataka as a third force
to whom he could appeal against the strength of his bakungu. In 1900 the
Christian bakungu then in power attempted to suppress the bataka, not only as
potential political opponents but also, perhaps, because the clans were the social
matrix giving meaning to the old religious order which would not march with the
new politics.2 Already in 1897, when Kabaka Mwanga made an abortive attempt
*Based on a talk given to the Uganda Society on 11 October, 1961. The material for
this essay has been collected as part of a study of the relationship between religion and
society in Buganda.

to restore an autocratic pagan monarchy, they had tried to rid themselves of the
other threat to their position. In the infant Daudi Chwa they surely saw a
Kabaka who would develop, on the English model, into an obedient, constitu-
tional Christian monarch. Twenty or so years later he himself was to write
bitterly of the attempts of his ministers to keep his wings clipped.3


Lienhardt's very valuable discussion4 suggests that in all the forces which
western man includes in the idea of the individual 'unconscious' have in other
cultures been exteriorised as personal forces acting on man from without. These
may be reflections of family relations (ancestral ghosts), wider social relations
within the tribe or with neighboring tribes (what he calls clan and free
divinities), the forces of nature (lightning, rain etc). and of man's ultimate
relation to the totality of human society5 (the Divinity who, though known by
different names to different peoples, is yet the One from whom all else ultimately
derives and from whom the other forces are no more than modes of action).
These forces Lienhardt calls "ultra-human" rather than "supernatural".6 In
Buganda these are described as mizimu (ancestral ghosts), misambwa and
balubaale; and Nsimbi7 regards them all as men whose exceptional attributes in
life were supposed to carry over into death. Ghosts might be active at the level
of the family only; or they might enter into natural objects and become misa-
mbwa; or they might be tribal figures who become balubaale; and there was
inevitably some overlapping between the classes.
Gorju8 identified the Buganda of Kintu with the legendary Empire of
Magdasor which, in 1307, sought Christian missionaries from Pekin. Buganda
was, he thought, at that time monotheistic and monogamous; and, more recently,9
the bakabaka have been traced fifteen generations before Kintu, to Tonda Katobo,
who entered Buganda from Ethiopia as a Christian. After two declines into
paganism, it was Kintu who, according to this account, restored Christianity.
Gorju, like Tucker,10 found Christian influence also in the fact that a ritual
action of priests of the lubaale Muwanga resembled the sign of the Cross made
in the manner of the Eastern Church.
But of such a monotheistic belief there is no trace in the accounts which came
to be written from 1877 onwards. Kagwa"1 mentions seventy-three balubaale.
Of these, Katonda, who had neither parents nor children, might be translated
"Creator" and was taken to name the Christian God. It is, therefore, extremely
difficult, today, to distinguish early beliefs about him from Christian accretions-
especially the assurance of some Christian Baganda that he was always the
Creator par excellence, of whom the balubaale were no more than satellites.12
But there seems to be agreement that very little was known about him, that he
created all things and that he gave men no trouble. He had three temples, all
in Kyaggwe and all tended by priests of Njovu clan. The sites of these temples
are all called Butonda, which might mean "the place of Katonda". But Butonda is
a perfectly legitimate place-name in its own right; and Katonda might therefore
be no more than "the person of Butonda". Like the great majority of balubaale,
he may originally (however mythically) be of human origin.13
Katonda's position of primacy was challenged (though it is doubtful whether
the conflict was consciously recognized) by Muwanga, of whom Nsimbi14 says
that he was the leader of all the gods and ruler of all things. He must be the
same as Ruhanga15 of Western Uganda, who was much more clearly the Ultimate

Cause; but his position in Buganda is complicated by his being given a human
genealogy as son of Wanga and grandson of Bukulu, mythological men or Ssese
who are considered below.
Finally, there is Ggulu, the sky-god, who had neither priest nor temple, though
Roscoe, writing in 1911, says that "in recent times" a man of Nseenene clan was
possessed by him and a temple built.16 It is at least interesting that a lubaode
could be recognized who had no overt means of expression through priest or
temple; and it is just possible that he belongs to an earlier culture which
recognized one all-pervading God with little in the way of religious paraphernalia.
He gave birth to Kiwanuka, the god of lightning; and, according to one form
of the Kintu myth, recorded by Miti (n.d.), it was he who sent to earth, as first
man and woman, Muntu Beene and Namuntu Bandi, who became equated with
Kintu Kato and Nambi Nantuttululu. According to the more generally accepted
story, Nambi was at the same time daughter of Ggulu and a woman of Ngeye
clan, which Kintu found in the country when he came.17 Ggulu was father also
of Walumbe, who, as a result of Nambi's disobedience to her father's command,
was able to follow her to earth where he introduced disease and death.'1 His
priests are of the same clan as Nambi.
Other bdubaale, who appear to be nature gods in origin, are (i) Kawumpuli
(of plague) and Ndawula (or Kawaali-of smallpox), both reputedly princes, and
(2) Musisi (of earthquakes), his son Wamala (of the lake of that name), Wamala's
nephew Mukasa (of Lake Victoria) and Mukasa's son Musoke (of the rainbow).
Of these, Wamala, Ndawula and Mukasa appear in western Uganda as
Bacwezi,19 members of the legendary early dynasty and gods of Bunyoro-Kitara.
But in Kiganda tradition the importance of the second group lies in the fact
that, with Wanga and Muwanga, they are all descended from Bukulu, who came
to Buganda with Kintu and settled in the Ssese Islands. That there were
continual settlements of Bassese on the mainland tradition leaves no doubt. One
tradition has it that Kintu originally came from Ssese, or at least that he sought
shelter there while preparing to regain his kingdom.20 Bukulu appears in some
traditions as a Kabaka of Buganda some generations before Kintu; his son
Wanga was brought to the mainland, in the reign of Kabaka Jjuuko, to restore
sun and moon to their proper places in the sky. Eight generations earlier, Wanga's
grand nephew Kibuuka (brother of Mukasa) helped Kabaka Nnakibinge against
the Banyoro; his nephews (sons of Mukasa) Nende and Kirabire shared in a
minor way, his functions as god of war. Kaganda of Bukasa claim to be the direct
descendent of that twin of Nnakibinge left as a hostage in Kibuuka's place;
Mbale,21 the site of Kibuuka's chief shrine, is connected etymologically with
lubaale; and that part of Lake Victoria lying between Ssese and the mainland is
called Nnalubaale, the Mother of the gods. The bakabaka claimed to rule Ssese
but appointed no ssaza chief till 1900. Distantly mysterious in the mist, the
Ssese Islands were perhaps a natural home for the gods. But they seem to have
been gods who, whatever their origin in the forces of nature, were assimilated so
far as possible to the legendary history of Buganda. There were many others,
like Kitaka (of the earth)-a purely nature god-and Wekkaanye-Kino,22
identified with Kisolo the Katikkiro of Kintu and founder of Ng'onge clan.
Perhaps the balubaale were complexly compounded of natural forces and dim
memories of Bacwezi together with heroes of specifically Kiganda origin.

They were served by temples, often scattered throughout the country, each
with a medium (who might be of any clan) and a priest, who had charge of the
temple paraphernalia and acted as interpreter between medium and people. The
priesthoods became hereditary in particular clans and the right to appoint
might be given by the Kabaka to a clan head. But priests of the same god might
be found in different clans. They were tribal figures available for consultation
by any Muganda.
Their relationship with the bakabaka was ambiguous. It is interesting that,
while both Kintu and Kisolo disappeared instead of dying, only the latter became
a lubaale in Buganda.23 I suggest that Kintu is more important than the balu-
baale, because he is the central symbol of tribal solidarity, while they are simply
functional to its existence. But, in addition to the possible avoidance of adminis-
trative interference in Ssese, certain hills (including Kintu's capital at Nnono)
were taboo to the Kabaka and his messengers, so that they became places of
sanctuary.24 This may have been a relic of an earlier religion which always
retained its institutional independence of the Kabaka, a continuing symbol of
Buganda's refusal to submit wholly to the new political authority. But when it
is reported that Kabaka Tebandeke became the medium of Mukasa,25 it suggests
either that he was attempting to gain control of the lubaale cult or that the cult
was too strong for him. His nephew Ndawula agreed to succeed to the throne only
on condition that he was not also medium; and the mediumship passed to a
junior branch of the royal family. At a later date Kabaka Mutesa I was advised
by Mukasa that his illness was due to appointing wrongly to the posts of
Kimbugwe and Mugema and to imprisoning the head of Kasimba clan;26 and
in 1879 Mukasa "tied up" the lake for three months, allowing nobody to touch
its waters. Mutesa sent an offering of 100 slaves, 100 women, 100 cows and 100
goats to untie the lake: and then built for Mukasa a temple at Mmengo, where
the priest would be under his direct control.27 There is a clear suggestion in all
this that the priests and the clans often sided against the centralizing tendencies
of the Kabaka. But it is certain that neither Mutesa I, nor his predecessor Suna
II, had much regard for the balubaale. Both would plunder temple estates, and
even kill priests, if the lubaale concerned did not come up to expectations. To
Mutesa they were little more than pawns in his policy of "divide and rule"; and,
when the missionaries arrived in the 1870s, he tried to treat them as represen-
tatives of two new balubaale, whom he might play off against one another and
against the old gods, as he was already playing off all the other factions in
Kiganda society.2


The misambwa are manifest in animals, snakes, swarms of bees, trees, rivers,
stones. Unlike the balubaale, who are known "in the spirit only" (mu mwoyo
gwokka) and are tribal figures, the misambwa require a material object through
which to reveal themselves and are confined in their activities to a lineage group
or a locality. There is confusion between the two classes. Sserwanga, for
instance, which is the birth-name of the lubaale Mukasa, is also a powerful
musambwa in Buddu; and Kikalanga, who causes dwarfism in the Mabira
Forest, is classified sometimes as one and sometimes as another. I have, indeed,
had reported to me a discussion with a diviner (mulaguzi), who claimed to have
been possessed by the musambwa Kikalanga. When it was pointed out to him
that he lived too far from the musambwa for it to be effective, and that in any

case taking possession of people was a function of balubaale, he agreed that
he had used the wrong word and that Kikalanga must be lubaale.
Roscoe, although he regularly gives Kiganda terms, does not use the word
musambwa at all. Gorju gives it only a passing mention. Although Roscoe speaks
of spirits of natural objects he does not mention their special relation to the
clans; nor, so far as I have been able to discover, does Kagwa. But I have
visited near Kalisizo in Buddu the mutaka of a line (olunyiriri) of Nkima clan,
a professing Muslim. From him came a very clear story, confirmed in all essentials
by my guide, himself a mutaka of a line of Lugave clan, a Christian and reluctant
to show me, though willing to describe, similar features on his own land. Near
the house are three trees, each housing a musambwa: Lubowa, which grew from
a spear planted by Ddungu, founder of the line; Lwebembera, a famous diviner
of the line; and Kagolo, second mutaka of the line. All are still available for
consultation. Some distance from the house, approached through thick under-
growth, is a most impressive cave, entered through a short tunnel two feet high
and full of bats, with large windows giving onto the forest behind. It would have
formed an ideal meeting place for witches in medieval Europe. Out of the cave
leads another tunnel, which nobody cares to enter but is said to lead to Bukakata
and to house the musambwa leopard through whom Ddungu speaks. The leopard
has not been seen in living memory; but members of the line (and only they)
bring offerings of goats and banana beer in cases of sterility and sickness. This
is particularly striking in view of the fact that such practices are punishable
in a Ggombolola Court by five years' imprisonment. Perhaps even more striking
was that all on the road, from whom we asked the way, spoke quite freely of the
musambwa and said that, if I really wanted to see misambwa, I should go not to
Buddu but, much nearer at hand for Kampala, to Kyaggwe.29
It emerged quite clearly that the musambwa belongs to the line and is named
after the founder of the line. It is not approachable by the public in general. I
was assured that every line has a musambwa and that the particular animal or
natural object involved is specific to the line itself rather than being generic to
the clan. Dr. Southwold, on the other hand, tells me that the leopard musambwa
Kitaka belongs to the whole Nankere section of Mmamba clan and suggests that
misambwa were attached to each butaka (clan and lineage burial ground)-that
is to say, to the locality rather than to the social group. That they are the medium
of communication of the lineage founder may also conflict with the recognition
by a living mutaka of his mystical unity with all his predecessors-that is, he
will say, "I did such and such" when he means that it was done by the holder
of his office some centuries ago. He would not claim to be active in the musa-
mbwa. But there is no reason why the two beliefs should not exist side by side
and become overt in different contexts.
On the other hand, the attachment of misambwa to localities is well attested.
North-east of Kalisizo there used to be a pool on the top of Mulondo hill. It was
formed by a girl who gave birth to water instead of a child and was thought to
be the source of the rivers Nakitondo, Kyogya and Naludugavu. Its musambwa
has not been active for many years. But, until the grass was cut, there were
many snakes; and, when the Europeans tried to survey the hill, the Musambwa
used to remove their marks overnight.30 Similar stories are told of the origin
of other rivers-for instance, Sezibwa and Mayanja Kato; and the latter had a
priest of Nvuma clan. Another musambwa is that of Kungu rock, where Kiganira
went on his escape from prison early in 1961, and where evidence of a religious
cult has increased very considerably in the last ten years.31

So far as I know, all these misambwa were reputedly once human beings; and
they seem to share with the balubaale the compound of experience drawn from
nature and from human relations. In Ssese and elsewhere there are misambwa
much more clearly comparable with the contemporary folk-tale of Ireland. You
may go to the shore at dusk and see on a rock a beautiful woman; but she is gone
before you reach her. You may hear at night the paddles and the songs of men
in a canoe which never reaches the shore.
In view of the wealth of information which is readily available, it is important
to ask why Roscoe, describing some of the misambwa, does not call them by
name; and why Gorju, knowing their name, gives them so little space. The
Christian chiefs, who in 1889 became the new rulers of Buganda, did their best
to destroy the old gods, together with the power both of the Kabaka and of
bataka. But while it was relatively simple to destroy temples and alienate the
estates of balubaale, it was not so easy to be rid of the misambwa. Bees continue
to swarm and rivers to flow however furiously the politicians may roar. Perhaps
the best thing was to ignore them; to pretend that, as the bataka had been
reduced to insignificance, so had the spirits of which they were the primary
guardians. I am inclined to think that Roscoe's apparent ignorance reflects not
only the ignorance of missionaries and administrators about the clans as an
integral part of Kiganda society but, perhaps at a deeper level, a possibly wilful
depreciation, by their bakungu informants, of the importance of the clans.32

"The immutable substance, common to all and assuredly the most obvious
feature, of the religion of our Africans is their belief in the ghosts of the
departed. No belief is so deeply rooted, or has so much effect on their practical
life.""Possibly the most venerated class of religious objects were the ghosts of
departed relatives. The power of ghosts for good or evil was incalculable." So
Gorju and Roscoe.33 Each had its shrine, where offerings of beer and barkcloth
were made; and each-male or female-insisted on the appointment of an heir.
Even today, in Christian families, a whirl of dust or a gentle breeze in the
banana garden is still "muzimu". Corpses might be carried hundreds of miles to
be buried in the lineage butaka; and the muzimu, after visiting Walumbe at
Ttanda in Ssingo, returned to the burial ground and tended to cling to the
jawbone (luwanga). A troublesome muzimu might be controlled by removing the
jawbone to a great distance. On the other hand, says Roscoe, "there are jawbones
of men, who lived nearly a thousand years ago, preserved to this day by members
of the clan . (and) bring good fortune'. Nsimbi suggests that they were the
focus of prayer at least to the mizimu of clan founders.34 All men, say both
Gorju and Roscoe, looked forward to living in the next state. But they carried
into it both their physical and their moral character. A thief preferred death to
the amputation of a hand-a loss which he would suffer in eternity. The wicked
return to torment the living, the virtuous revisit relations in their dreams. In the
clan-naming ceremony (okwalula abaana) the continuing concern of mizimu for
their lineage is shown by the selection of one of the paternal ancestors to be
special guardian of the child.35 Today, when mizimu are experienced mainly by
women, the paternal aunt (ssenga)-in life a peculiarly authoritarian and oppres-
sive figure-is the most usually troublesome muzimu; but those of the parents,
the paternal grandparents, the maternal grandmother and, sometimes, the paternal
great-grandfather are mentioned as returning to set right the footsteps of one

who has strayed from the path of wisdom or virtue. "Ghosts thus reflect" says
Lienhardt,36 "for those they visit, the relations which the latter have had with
them in life and still have with them in the conscience and memory".
I want to stress the continuing belief in the activity-and,at least sometimes,
the benevolent activity-of mizimu towards their own family, because most of
the younger generation of Baganda insist that they were always evil. Dr. Richards
writes, "ancestor worship is a bit out of the ordinary in Buganda owing to the
strength of the kingship"; and Dr. Southwold, "I do not think that the deceased
are regarded as still members of the clan. They do not receive regular offerings;
in fact, in their capacity as clansmen, I do not think that they receive offerings at
all. Now and again an offering is made to pacify an angry muzimu. But he is
regarded as a different kind of being from a man; and the offering is made
because he is angry and has power to hurt, not because he is a clansman . .
There is no cult of ancestral graves or shrines and no cult of ancestral spirits
as such." Taylor speaks both of shrines and of maleficent misimu; but his
examples appear to refer to individual grievances rather than to action aimed at
the preservation of family solidarity.37 Neither Gorju nor Roscoe provide sufficient
evidence for it to be possible to say, with any conviction, how important was the
cult of the mizimu eighty years ago. But Gorju's later judgment38 suggests that,
despite its fundamental importance, it had, with the centralization of political
authority, been absorbed into that of the balubaale; and there appears to be
prima facie evidence that it is not now as active as that of the balubaale and
Wilson in Nyakyusa in the 1930s and Middleton, much more recently in
Lugbara, found that the dead are senior kin, beneficent in intention (even if
resenting and avenging neglect) and enjoying intercourse, through offerings of
meat and beer, with their living kin. "We give them meat," said a Lugbara. "It is
like saying 'welcome', and they say 'thank you'."39 It is tempting to think that
the same was true of the early clan society of Buganda. In so far as paternal
ancestors are still concerned in okwalula, and misambwa are identified with early
heroes of the clan or lineage, it is hinted at. But available evidence suggests a
loss of intimacy with the ancestors, a tendency to stress their malevolent rather
than their benevolent qualities, their identification-except perhaps at the
family level-as heroic figures rather than as fellow-clansmen.
It may be questioned how far this intimacy was ever felt beyond the range of
the butaka, where the graves were a constant reminder and the mutaka, still
today, knows as a matter of course his paternal ancestors back to the founder
of the clan itself. But the disruption of the clans as territorial units had started
long before the coming of the missionaries; and the large-scale population
movements at the end of last century, together with the alienation of butaka in
1900, carried the disruption to the primary units of clan organization. The idea
of the clan, with its specific names, is still regarded with affection; bataka can
still be difficult in matters of inheritance; clan loyalties may come between
husband and wife in marriage; and okwalula is still regularly practised. But
family graves on private mailo (individual freehold land dating at most from
1900) are increasingly the order of the day; apart from the mutaka, there are
few families which know the names of ancestors beyond the third, or at the
most the fourth, generation; and neighbourhood, rather than kinship, is
increasingly the relationship between families in any given locality. The whole
character of the clan, as a socio-economic unit, is breaking down; and it would
not be surprising if its mystical overtones were also to dissolve. Wilson found in

Nyakyusa40 that the ancestor cult was concerned solely with relationships between
kinsmen and of chiefs with their people, while witchcraft was practised mainly
between fellow villagers. It might be that the considerable resurgence of witch-
craft which is occurring in contemporary Buganda-although Roscoe, in com-
parison with his accent on mizimu, gives it scant mention-is related to a
fundamental change in the structuring of social relations.
In the second place, whatever may in the past have been the effect of political
changes on religious cult, since 1953 the new emphasis on the tombs of the
bakabaka at Kasubi and elsewhere has been very evident. While the sense of
insecurity, produced first by the deportation of the Kabaka and more recently
by the movement for secession, has produced a new interest in ultra-human forces
as a whole, the main emotional emphasis has been on political, rather than
individual or family, concerns and on Buganda's awareness of itself as a nation.
For this the Kabaka, and not even the tribal balubaale is the only adequate
symbol. The mizimu take a very much lower place.
Thirdly, other new influences have been felt in the country: Christian ideas
have spread very widely; effective medical services have established the idea of
empirical causes of death and sickness; and for the last sixty years Buganda has
enjoyed increasing economic prosperity and continued law and order. Christians
have used the word muzimu to translate "demon" in the New Testament41 and
therefore have encouraged the idea that mizimu are always bad. Empirical causes
are regarded as complementary rather than alternative: few Buganda today die
without suspicion of witchcraft or the interference of a muzimu, however
definite the medical diagnosis may be. But Roscoe42 says that "Whenever anything
went smoothly . a man did not trouble to make offerings . to the objects
of his religion". In any society it is common to attribute good fortune to one's
own efforts and only misfortune to the malevolence of others. The evil that men
do lives after them. It is perhaps hardly surprising if the benevolent mizimu
tend to be forgotten and the malevolent noticed only in individual cases of
misfortune, where memories of unhappy relationship in life are still ripe. On the
other hand, distance lends enchantment: and it would be convenient to be able
to demonstrate that, the further removed the ancestors, the more benevolent they
become. For the mizimu as such there is no evidence. In so far as they are
identified with the heroic figures of balubaale and misambwa, they appear to be
more demanding than gracious. But, if further study is possible, perhaps it
should be undertaken in the light of Evans-Pritchard's judgment of the Nuer
spirits of the air.43 "To the European observer they seem to be greedy, capricious,
and hostile, but this is not quite how they appear to the Nuer, for ... they only
cause misfortunes to those who in one way or another are at fault." If the
balubaale of Buganda did not practise the same justice, it may be that justice
from superiors was not the normal experience in life.


Like and yet unlike the common people, the bakabaka had their own cult
associated with the mizimu of their predecessors-a cult which is surely related
to the belief that each mutaka" shares the personality of all earlier holders of the
office. Some of its contemporary manifestations are movingly described by
Taylor.45 The jawbone and umbical cord of each, when he died, were preserved
in a special temple and guarded by Nnaalinnya (the official sister), ex-Kimbugwe
and a number of widows and other officers. Although the relics of a particular

Kabaka became less important with each succeeding generation, certain of their
officers had to be replaced, on death, from the clan which originally supplied
them; the muzimu of any Kabaka might take possession of a person many miles
away; all the jawbone tombs (as well as the body tombs) of the bakabaka back
to Kimera still survive with a few hereditary officers; and the great majority
have been rebuilt since 1955.46 According to Roscoe, the reigning Kabaka visited
the actual jawbone tomb of his predecessor only rarely-a visit which was marked
by a considerable slaughter of people required for the service of the dead
Kabaka. But a shrine was built in Lubiri (the Kabaka's enclosure), where the
medium from the tomb came frequently to pass on advice about affairs of state.
The dead Kabaka was not available, like balubaale, for public consultation,
though he was held in at least as much awe as they.
The importance of the jawbone is found also in Bunyoro, where the possession
of his predecessor's jawbone was proof of the legitimacy of the reigning
Mukama.47 But, at root, there is here nothing more than the belief that any
muzimu is likely to cling to its earthly jawbone. It is difficult to see, in this
particular cult of the dead bakabaka, anything more than an adaptation to the
royal state of what was common to all Buganda. All mizimu took an interest in
the affairs of their surviving families and could possess men to speak with their
voices. It is not surprising if, just as the Kabaka was in life far and away the
most excellent of all Baganda, so his ghost should, in death, have more excellent
honour and power than they. Just as, in his life-time, men were killed to invi-
gorate him, so in death he would require men to accompany him: in addition
to some of his widows, his chief cook, chief brewer, head cowherd and the
guardian of his water well; and, of women, a cook and those who had charge
of his beer, his bedchamber, his water, his clothing and his milk.4
Kabaka Mutesa I tried to do away with all this. He disinterred the bodies of
some of his predecessors and had them buried in the jawbone shrines. He
himself, and his successors, were buried without dissection; and he left instruc-
tions for the uncovering of deceivers who would pretend to be possessed by
his muzimu.9 He seems to have been a cynic, who enjoyed theological discussion
but was interested in religion only as a tool of his own temporal power. But,
as almost any African visitor to the royal tombs can testify, his own explicit
instructions do not prevent him speaking from the grave to contemporary


The various categories of ultra-human being found in Kiganda tradition seem
to symbolize different types of experience. The mystical aspects of natural
phenomena are personified in the gods of sky, thunder, the Lake and so forth;
but almost always there seems to be a drive to socialize them-to give them a
human genealogy; and, in so far as this process succeeds in deriving them from
Bukulu, "who came to Buganda with Kintu", it makes them junior to Kintu
and thus inferior in status to the kabakaship. Gorju may well be at least
mythologically right when he asserts that the balubaale were introduced by
Kabaka Nnakibinge into a previously monotheistic society in order to strengthen
the authority of the bakabaka. At least they represented the growing consciousness
that Buganda was emerging as a nation out of a mere association of independent
clans. But the rituals of lubaale-worship came to be controlled by the clans and
(along with the misambwa, who stood for earlier loyalties to locality and lineage:


along also with the claim that some of the clans were in Buganda before Kintu
and that the present mutaka was himself the clan founder) could on occasion be
used to assert the continuing independence of the clans from central authority.
The mizimu came to be concerned only with members of their own families and
therefore of no great political importance-with the exception of the mizimu
of the bakabaka themselves. Confined as it was to the ruling Kabaka, this cult
gives every impression of being the ultra-human sanction for a kabakaship which
was always, to some extent, in conflict with traditional authorities and,
at a later stage in conflict also with its own directly appointed bakungu. The
conflict may itself be symbolized by the royal origin of the gods of plague and
That the bakungu had any traditional mystical symbols of their own seems
doubtful. But the advent both of Islam and of Christianity, with their assertion
of a single universal God who worked dynamically in history, provided them
with a religious symbol by which they could assert themselves against both the
clans and the kabakaship. However deeply the Gospel may have affected many
private lives, the main overt consequence of missionary endeavour was the
establishment of a new, indigenous, political pattern. From this point of view,
it had two serious weaknesses. In the first place, although its association with
western political and technological power was of the greatest benefit to the
political and material development of Buganda, this same association made it
appear, at a later date, as a tool of imperialism. It is hardly surprising if
Baganda, in their effort after tribal integrity, should begin to look again to
the tribal gods. In the second place, Muslims, Catholics and Protestants, all
proclaiming the same God, nevertheless became organized into three distinct poli-
tical groups. It is not that there are three gods but that the one God has become
segmented into three aspects, entering into social conflict with one another.50
Because of the disagreement between his adherents, it is not possible for him
to symbolize national unity. The only figure able to fill this position is the
Kabaka, who has thus achieved a mystical significance socially far more important
than any gods. A prayer, which has been widely used of late and has the
advantage that "it can be used by men of all religions", asks each blessed (omu-
tuukirivu) lubaale in turn "to have mercy upon us" and concludes: "0 all ye
blessed ones, who fled because you feared the atrocities of the European, Arise
and join with us . ."51 This is no amusing parody, but a deliberate attempt to
turn men away from "the European religion" to the Kabaka, "to whom all
power has been given by the balubaale". It is a phase which may well pass with
the present political uncertainty; but it illustrates that the study of traditional
religion is a matter not merely of academic concern.


Evans-Pritchard, E. E. (1956). Nuer Religion.
Gorju, J. (1920). Entre le Victoria, l'Albert et I'Edouard.
Kagwa, Sir A. (1901). Basekabaka b'e Buganda.
(1905). Empisa z'Abaganda (1952 edition cited).
(1908). Ebika by'Abaganda.
Lienhardt, G. (1961). Divinity and Experience: The Religion of the Dinka.
Low, D. A. (1957). Religion and Society in Buganda.
Middleton, J. (1960). Lugbara Religion.
Miti, J. (n.d.). History of Buganda, typescript.
Nsimbi, M. (1956). Ammanya Amaganda.


Richards, A. I. (1955). Ganda Clan Structure, E.A.I.S.R. Conference,
Roscoe, J. (1911). The Baganda.
Southwold, M. (1961). Bureaucracy and Chiefship in Buganda.
Taylor, J. V. (1958). The Growth of the Church in Buganda.
Wilson, M. (1957). Rituals of Kinship among the Nyakyusa.

IKagwa (1905). p. 147; Nsimbi (1956). p. 88.
2Welbourn, F. B. (1961). East African Rebels, p. 19. Dr. Southwold writes that he
thinks the bataka were unable to accept Christianity because of their involvement in
pagan ritual.
lPrivate unpublished correspondence.
4Lienhardt (1961). Chapter IV. See also Jung, C. G. (1959). Collected Works, Vol. 9,
Part I (The Archetypes and the Collective Unconscious), p. 25, para. 54.
5Lienhardt (1961).
6ibid., p. 28.
7Nsimbi (1956). p. 122.
8Gorju (1920). Chapter X.
9Pastor H. C. M. Guwedeko, unpublished typescript.
10Gale, H. P., Uganda 1., 20, 1, p. 72.
llKagwa (1905). pp. 209-228. There is a vivid description of "lubariism" by O'Flaherty
in Church Missionary Intelligencer (1884), p. 224, Kagwa gives only one-and-a-half
lines to Katonda and five pages to Kibuuka!
12This is a difficulty which befogs all contemporary attempts to describe traditional
Kiganda religion, since even pagans have been affected by Christian and Islamic
concepts. A certain practising diviner (mulaguzi), a Catholic by baptism, distinguishes
the balubaale, who possess him, as "angels" (good) and "demons" (bad).
13Kasujju and Katambala-personal names-are, I think, prior to Busujju and
Butambala, the counties which they govern. On that analogy Katonda should be prior
to Butonda. But is Kaganda, then, prior to Buganda?
14Nsimbi (1956). p. 124.
15Kuhanga means "to create" and, in this sense, is equivalent to the Luganda kutonda.
Kuwanga in Luganda has more the sense of "to put things in the proper relationship to
one another". There could be a fascinating investigation of Kuwanga in relation to its
derivatives, buwangwa, eggwanga, kiwanga, luwanga and the probably associated verb
16Roscoe (1911). p. 317. This is not mentioned by Nsimbi (1956).
17Nobody questions the statement that Kintu was a foreigner. By his marriage to a
woman of the land he is accepted into Kiganda society. By her descent from Ggulu,
future bakabaka are given a divine parentage.
18Cf. Evans-Pritchard (1956) and Lienhardt (1961) for the close association between
the sky and the one God; and Lienhardt, p. 33, (as well as Genesis 3) for the relation
between evil and the disobedience of the first woman. Miti, who spent many years in
Bunyoro, may have been influenced by its mythology, where Kintu is the first man and
Kimera the first Kabaka. In Bugisu, Muntu is the first man and Kintu and Masaba his
19Spelled Wamara, Ndahura, Mugasha. In Unyamwezi the last is Ngasha. Gorju (1920)
specifically denies the identity of Wamala and Wamara.
20Nsimbi (1956). pp. 151-4; Kaganda, D. T. (1952). Entandikwa ya Bakabaka ba
2Near Mpigi, Mawokota.
22Nsimbi (1956). pp. 124, 222. He is also called Muwanga.
23Not far from the reputed grave of Kintu in Busoga, I found a man making a very
good trade out of communication with his ghost.
24Roscoe (1911). p. 319; Nsimbi, M., unpublished typescript. Some traditions allege
that Kintu was in fact buried at Nnono.
25Roscoe (1911). p. 220.
26Miti (n.d.).
27Cunningham, J. F. (1905). Uganda and its Peoples, p. 80.
28Low (1957). p. 4.
29Mr. Simon Musoke, to whom I am grateful for introduction to this enquiry, was
told that only a native of the particular locality in Kyaggwe would be able to obtain
reliable information.


30Mr. E. K. Ssabaganzi, personal communication. Snakes have a close association
with the ghostly world and are in the same noun-class as misambwa and mizimu.
Similar stories of ghostly interference with European activities are told of the house at
Lweza (mile 9, old Entebbe Road) and the Nsamizi Training Centre.
31Early in 1955, during the Kabaka's exile, Kiganira claimed to be possessed by
Kibuuka and was imprisoned after one of the Kabaka's police had been killed.
32But Kagwa and others used, on their notepaper, their clan symbols surrounded by
the motto, Simmuda nnyuma (there's no turning back). Kagwa resisted violently those
who tried to expel him from his clan in 1910 and, in his later years, played an intimate
part in its affairs.
33Gorju (1920). Chapter 14; Roscoe (1911), p. 273.
34Roscoe (1911). p. 282; Nsimbi (1956). pp. 122, 185, 255, 274.
35Roscoe (1911). p. 64; Taylor (1958). pp. 143-7.
36Lienhardt (1961) p. 154. The ghosts mentioned seem to correspond to the enda
(Richards, 1955), though the exact meaning of this word is confused.
37Taylor (1958). pp. 207f.
38Gorju (1920). Chapter 18.
39Middleton (1960). p. 86.
40Wilson (1957). p. 5.
41Taylor (1958). p. 207.
42Roscoe (1911). p. 287.
43Evans-Pritchard (1956). p. 51.
44The Kabaka is Ssaabataka (head of all bataka).
45Taylor (1958). pp. 209ff.
460liver, R. (1959). Uganda J., 23, pp. 124-133, has a complete list of sites.
48Roscoe (1911). pp. 106, 209.
49Kagwa (1905). p. 17, quoted Taylor (1958). p. 207.
5OLienhardt (1961) and Evans-Pritchard (1956) note a similar segmentation of the one
God among Dinka and Nuer
51Ayi abatuukirivu mwenna abaawa nga mutidde ebibambulira by'Abazungu, muzuu-
kuke mwegatte wamu tuzzewo ensi yaffe etereere nga bwe yali.



(Including an account of some of the larger caves in Mubende District)
AN important source of information for widening our knowledge of Uganda's
past still lies in the systematic examination and excavation of the caves and rock
shelters located within the boundaries of the country. Caves and shelters are
most likely to occur in the limestone and lavas of the Eastern Province, the lavas
of the south-west, the laterite of the lakes' shores, as well as in the granites and
dolerites of Buganda and the Western Province. Much is to be learnt from the
evidence left by man during the periods of occupation of such sites since his
days as a maker of stone tools to his progress to working in iron.
At the beginning of this century the Reverend H. B. Lewin recorded a number
of caves in the area of Mubende in Buganda.1 Later, following Protectorate-wide
enquiry and investigation Wayland, with the aid of Administrative officers and
others, compiled a list of 580 caves and shelters.2 Undoubtedly a number of
unrecorded caves have been visited by the curious from time to time, and some
have received cursory examination. By 1960 seven caves and rock shelters-the
majority in Buganda-had been scientifically investigated by part or full excava-
tion.3 Others have also received attention in unpublished reports or have been
referred to in specialist papers.4 There are also the caves in Mubende about
which Ggomotoka has written in his History and Legends of the Rocks of
Although the number of sites examined is small, those investigations that
have been made have yielded considerable evidence of occupation by man during
the various stages of his development. Deposits of stone implements, bones, shells,
pottery, iron, also ochre, even stone missiles and the presence of paintings, tell
of man's habitation of these shelters, permanent or temporary, up to the close
of the last century.
The largest caves are possibly those of Guru-Guru in Acholi6 and others in
Bunyoro, including Omukama 'Chwa's store in the summit of Katuku Hill in
Buruli, which are said to be large enough to shelter between 200 and 500 people.7
Caverns in the cliffs of the two great lakes, usually only approachable by water,
are also large, like Iseke near Kibiro on Lake Albert where the canoes of the
Abakama were kept, and Mizinda Cave on Lake Victoria at the southern tip of
Masaka District.
In recent times, whether in Ankole, Buganda, Kigezi, Toro, Bunyoro, Lango,
Acholi, West Nile, Busoga, Budama, Bugwere, Teso or Bugisu, caves have been
put to a variety of uses. These uses have included: ancestor worship, witchcraft,
rainmaking, food storage, refuge, for dumping belongings including regalia, as
burial places for lepers, and merely as temporary shelters. In Buganda (in Mengo,
Masaka and Mubende), caves have been in frequent use by men until very
recently. They have continued to serve as hiding places, places in which to
practise religious rites and as abodes for witchdoctors.


Many of the recorded sites, and some of those I have visited, especially the
deeper hollows of the water-worn cliffs of Lake Victoria, are enveloped in legend
and factual tales wrapped in fancy and magic. Three such caves are-
(i) Nakyejwe, cave, in Butambala, which takes its name from the daughter of
Kintu. Because of her many misdeeds in life, Nakyejwe was turned to stone
when she died. The cave is in this great stone. Eventually children came to play
here. One day, in an unexplained fit of rage, the spirit of Nakyejwe picked up
a large piece of rock and sealed the cave-mouth, imprisoning about sixty
children inside. This stone and cave belongs to the Engo (leopard) clan. A flat
piece of rock in the cave, supported on four pillars, is said to be Nakyejwe's bed.
A hollow in the rock is her bath. It has long been the practice for babies of the
Engo clan to be immersed in the water in this hollow. Close to the cave there
are three conical stones about two feet high which formed Nakyejwe's kitchen.
These stones are said to have been the three original cooking stones of the
(ii) Namirembe, a large cave in Mawokota, was used by Basesse when fishing
in the vicinity. The fishermen used to entice women into this cavern. As a
punishment a large slab of rock fell from the roof blocking the entrance to the
cave. Those trapped inside died of starvation. Nowadays only a small portion
of the cave remains to be seen. This is still used by Basesse for storing fishing
(iii) Katereke in Busiro, said to be no natural cave but a hole dug on the
order of Kabaka Kalema to a depth of one hundred feet. Likely aspirants to
the throne were hurled inside.
In the lands of the Kingdoms of Bunyoro-Kitara and Buganda the sporadic
raiding and wars of 50 years of Banyoro-Baganda rivalry must have influenced
people into developing the ability to melt away in the face of surprise attacks
or the advance of superior enemy numbers. Safe concealment within the rocks
of their neighbourhood was a regular necessity for people in the daily run of
their unsettled lives. Most accounts about the use of these natural retreats come
from the area covered by Mubende District where, in more recent times, the
Banyoro were much exposed to harassing raids by the Baganda.9
Many of the large rock formations of Mubende, bare and conspicuous, are
honeycombed with fissures and recesses large enough to contain small numbers
of people. The most suitable and plentiful hideouts are those situated in low-
lying rocky areas where a thick camouflage of forest and vegetation conceals the
very rocks themselves. Caves with entrances which, more often than not, are
barely large enough to admit a fully-grown person, which lead into what may
turn out to be large underground rooms, are frequently situated in the ravines
and valleys of this undulating terrain.
The most intimate knowledge is required to locate even the large cave
entrances. Few of these forest caves would be likely to be discovered unless by
betrayal or some tell-tale noise or eddy of smoke. Most caves, whether in
concealed rocks or in the open, could be turned into effective strongholds and,
with the aid of primitive weapons, defended easily. However, as rifles became
available in the latter half of the last century, the strategic value of caves as
strongpoints diminished. With this powerful new weapon the enemy found it
easier to drive his quarry away from the cave mouth into the inner recesses, from
where it was only a matter of time before the unfortunate victims could be
smoked out.
Accounts of the last days of cave warfare (when, for instance, southern

Bunyoro was under constant pressure from the Muganda general Cyprien Muta-
gwanya) are of additional interest.10 The need to retreat into secret places of
safety had, at this date, been part of the every-day life of people for generations.
Generally each group of people had its particular cave or caves into which to
flee. These hideouts were kept in constant readiness for occupation. Dry grass
covered the floors; supplies such as millet, beans, and dry wood were stored inside,
together with household vessels, in readiness for a sudden alarm. Water could
be obtained from the valley streams nearby, or right on the spot where streams
trickled through the lower levels of the shelters.
Usually, ample warning was given of the approach of an enemy. Then families
left their flimsy grass huts, taking with them what food, water and pots they had
time to gather; goats and cattle were taken too. The evacuation would be carried
out quickly, each family making for its habitual hiding place. There were special
places for notables and their servants. In the larger caves, many of which have
two or more entrances, some sections were reserved exclusively for notables.
Normally, because hiding places were well concealed deep in the forests or,
being well guarded were too dangerous to approach, the refugees would be safe
until the departure of the raiding party from the area. Unless a large scale
invasion was afoot an enemy was contented with looting; he seldom burned down
the grass dwellings. Burning was more normally done if a raid was directed
against a particular person or notable.
With the advent of firearms the danger from the invader increased. Traitors
came to the fore and the whereabouts of some caves were revealed. With the
increased use of fire and smoke by the enemy, many unfortunate refugees died of
suffocation or were shot or speared as they chanced a dash to freedom. Evidence
of these tragedies, some as recent as 75 to 80 years ago, can be seen in the
surroundings and recesses of many caves by way of empty cartridge cases,
discarded spears and the remnants of the belongings of those who had once
sought shelter there.
Whilst I have visited a number of the better known and more accessible
caves in the Mubende area, there are many which it has been impossible to
locate owing to the absence of local co-operation. This is probably because the
whereabouts of the family hideouts are no longer of concern to the younger
generations. There is also instinctive suspicion of the stranger, and superstition,
both of which influence those elders who know the location of the hidden
entrance to the family retreat to keep silent. Superstition is mostly inspired by
long-standing legend and by the past dramas resulting in the deaths of those
trapped inside the rocky fastnesses. Because of this aura of superstition and
fear, I have been unable to locate such caves as Kiriluma, Kasana, Kabanoba in
Ssabadu of Mubende, as well as others, which it is said have been the scenes of
Though many of the references to the sizes of caves, their vast chambers,
underground rivers etc.-invariably hearsay-often turn out to be exaggerated
(as are the tales of wild beasts, snakes and various fearsome creatures said to
lurk in almost every corner12) there is no reason to doubt that some large caves
do exist wherein many people have died. In 1919, Wayland was taken in great
secrecy into a cave which was full of human bones, some of which were old in
type and decidedly patinated.13
It cannot be overlooked that there might well be rock engravings or paintings
in or near the various caves. Conditions are not ideal for the preservation of
primitive paint in Western Uganda, although the dryness found in inner rooms

and recesses of certain caves could be conducive to preservation. The ochre and
used haematite found by O'Brien near Kagadi cannot necessarily be taken to
infer their employment for rock decoration. They are just as likely to have been
used for cosmetic purposes or for decorating small objects. At the Iron Age site
of Kibengo (10 miles north-west) and at Semwema (38 miles east) pottery was
found decorated with the colouring from hydrated oxides.14 Rock engravings are
more likely to be found and there is no reason to doubt their existence. Such
marks must be expected to be hidden in darkness, perhaps partly eroded or, if
in the open, overgrown with vegetation.
Sometimes the local people talk of strange marks on the rocks. Invariably, this
is taken to imply "paintings". Careful discussion with informants in the area
always reveals that, in the first place, they really mean indentations in the rock
and not paint smeared or applied to the rock. In the second place, they have
not seen what they are trying to describe, nor are they likely to have visited the
place in question.
One of the most reliable informants of my acquaintance in Northern Mubende,
the late Yakobo Byomere, spoke of strange marks on the walls of two caves,
namely Kasungwa in Buyaga and Semwema in Bugangazzi. He seemed to mean
"paintings" but, in discussion, and with the assistance of visual aids, it became
clear that he meant indentations in the rock. I think it is the misinterpretation
of his remarks made in previous years which gave rise to a "paintings on
Semwema" story.
Omw: Byomere's information has always proved to be basically sound.
Though it has not fallen to me to find any form of rock markings either at
Kasungwa or Semwema, I feel confident that there are some forms of human-
made marks on the rocks at both places. Another likely site is in or near Muzimu
Nowadays it is only hunters with their dogs giving chase to edible rats and
porcupines who visit these places. Whilst disinterest or fear grip most of the
local people it is always rewarding to see how curious onlookers eventually,
if a bit apprehensively, soon follow one's footsteps whatever the size
and darkness of the cave. Sooner or later keen interest is shown in the
surroundings which their forebears must have known intimately; and local
interest with a period soon to be forgotten is worth preserving. If properly
investigated, these sites, with their deposits protected from wind and rain, should
add to our knowledge of a bygone age.
Some of the larger caves of Mubende which I was able to visit between 1952
and 1955 through the generous co-operation of local chiefs and farmers are
described below.15 However, before going on to describe these particular localities
it may be as well to sound a note of warning to present-day investigators: their
work could be confused by evidence left by modern cave-dwellers who, in my
experience particularly in Masaka District, have ranged from worshippers
indulging in pagan rites to distillers of waragi.


Kiraba Hill-a trigonometrical point in the North of Mubende District-lies
close to Kagadi, west of the Hoima-Kyanjojo road.

Half way up the hill there are'some low shelters. From the smoke-stained
interior of the largest it is clearly evident that these shelters, which can only be
entered by crouching, have been used by man for a great length of time.
An unusual feature is the diminutive pillars of rock which reach from floor
to roof. Their average circumference is 46 inches. In the main shelter past
occupiers have evidently chipped and smoothed these pillars. Niches have also
been cut into some of them and into the back wall. Additional head room of
about two feet has been provided at two places where the ceiling has been
hollowed and smoothed into miniature domes. (Fig. 1). A clay fumigator type of
receptacle and some pipes were found here.
There are smaller shelters higher up the hill.
Peasants who dwell at the base of Kiraba have confirmed that this was a
popular place for refuge. Because of the frequent fighting which took place, a
split rock near the half-way group of shelters is known as balyanika-empanga,
referring to the human skulls and bones which lay nearby earlier in the century.
About a mile south of Kiraba Hill there are several huge boulders that had
weathered or possibly fallen into a convenient position for occupational use
which lie near the Rest Camp at Kagadi where the Sunga road joins with the
main Hoima-Kyanjojo road. Here, in 1935, O'Brien recorded an industry of
tools in white quartz of Magosian type. Red ochre and used haematite occurred
at all levels.16

Nyabihuru lies close to the Muzizi River, roughly opposite Nakabimba in
Toro. It lies at the foot of a granite ridge which is best approached via the
Kalaguza-Nakabimba track as far as Lwaga, then past the site of the rock
engravings of Hanfuka at Kisonde, and then towards the river across Butemba
Hill.17 The cave is easily located by a Giant Forest Tree (pterygota sp:) which
grows close to the 215 feet wide entrance. The cave extends for 95 feet into
the depths of the hillside (Fig. 2). People are said to have died here from
suffocation by smoke.
Because of the wide entrance this would have been a difficult place to defend.
However, within- the cave-mouth, four strong positions have been built of piled
stones which bear testimony to a sound plan of defence. The main position or
barricade, which covers a frontage of thirty feet, is at the centre of the cave. It is
well sited on the top of a large granite outcrop. From here the defenders had a
commanding view and distinct advantage over anyone trying to approach. On
the east side there are two smaller stone barricades, both at ground level. The
other flank is covered by a similar barricade. This latter position protects a
separate shelter which varies in height from 3 to 4 feet. The front has been
blocked with stones. Entrance can only be gained from the back of the main cave.
Behind this same barricade there is also a passage which leads away from the
main cave for forty-two feet into a small inner recess, its ceiling blackened with
smoke from indoor fires. A silted-up break in the furthest wall probably
conceals a passage leading into the hillside. The noise of bats from deeper
recesses suggests that this is likely.
A second passage leads from this small room to the large but low shelter close
to the front of the main cave. This place may well have been used for sleeping
quarters. Coloured beads were found here.
A wide cleft in the rear of the east flank of the main cave gives access upwards
to the summit of the ridge. This could have provided a means of escape to the

more agile. Patterned potsheds, blue as well as green beads and the bowl of a
clay pipe have been recovered.18
A hundred yards to the west of Nyabihuru are numerous small rock shelters.
They have the same pillar-like formation found at the Kiraba shelters.

This is a small cave in Gombolola Ssabagabo, west of the Kalaguza-Mugalama
road, near to Kiziko village, east of a hill called Kibanga.
Entrance is by a cleft in the rocks in a wooded valley through which flows the
Nyamagaju river. Low and narrow passages lead down to about twenty feet below
ground level for a distance of a hundred feet into small rooms. From one of
these rooms a low passage, no more than three feet high, descends to an under-
ground stream. From there more passages branch out. Pottery was found by
the stream. In this lower connecting passage a bone bead was found.

This spectacular cave in Gombolola Musale on the Kibale-Kiryanga road is in
the hill Igunga. (This hill must not be confused with Nkunda Hill nearby.)
The approach is thickly overgrown with elephant grass. In front of the wide
entrance there is a low granite arch, having a span of thirty-five feet, but which is
nowhere higher than 4J feet from ground level. Beyond, in the hill face, is a
rock shelter which is about thirty feet deep and sixty feet wide. Only the most
careful inspection of the back of the shelter reveals a small opening which, once
negotiated, leads into a wide corridor. This is really the floor of a large fissure
which extends upwards for 200 feet to the summit of the hill. Ledges make it
easy to find a way up for sixty feet to a wide gallery. This extends in
both directions into uninvestigated depths. A further scramble up convenient
ledges and a second gallery is reached. From here a slight ascent, leading off at
an angle, leads to the bottom of a wide shaft open to broad daylight fifty feet
With the summit within easy reach from half way up the hill-side, a good
escape route is available, a route which can be easily blocked at the small entrance
hidden in the recesses of the shelter below.
There was no surface evidence of long-term occupation.

The site is easily accessible, although not easy to find without a guide. It lies
in Gombolola Ssabagabo near Bwamulamira, a quarter of a mile off the Kibale-
Kagadi road. Because of its relative accessibility it has been visited frequently.
Kachuba is not a single cave but a series of twisting and turning passages at
different levels, leading to at least three inner rooms. The Kachuba stream flows
into and through the toppled rocks down to the valley below where it joins the
river Mbaya.
Thick vegetation covers a wide area of massive boulders. The entrance to
the hideout lies down a four-foot-wide passage open to the sky. From here a
narrow opening gives access to damp passages which descend to levels of ten
and fifteen feet. At the end of one passage the Kachuba cascades from fifteen
feet above in a shower of spray into an underground pool from where it flows
unhindered for a few yards before disappearing down a wide fissure. In the thin
light which filters through the cracks and crannies above one can see complete


FIGURE 1. Kiraba Hill Shelter.


FIGURE 2. Nyabihuru Cave.


'Es -`T


i' E.~ ~d'


FIGt'RE 3. Small pot from Kasungwa Cave (full-scale drawing by George Kakosa).

FIGURE 5. Bigoma Shelter.

and shattered moss-covered earthenware vessels, that still stand a few feet above
the splashing water. Here was the main water supply for those using the hideout.
Two smaller watering points lie at lower levels.
Several shafts which have not yet been investigated appear to lead to lower
rooms. At the two levels explored there are three large rooms. The innermost
room is remarkably dry. Unlike other parts of the cavern the bat guano, normally
moist and sodden, lies in a thick powdery dust on the floor.
A large number of potsherds, some beads, a fragment of bone and the bowls
and stems of pipes have been recovered from various parts of this hideout. In
1953 during a visit by the Curator of the Uganda Museum so many pipes were
unearthed from the outermost room of this cave that the site has come to be
known as "The Smokers' Club".

This is further down the valley from Kachuba Cave but in the same
neighbourhood. A formation of massive rocks is almost totally hidden by
vegetation. At one point a fissure gives access into this granite stronghold. A
steep descent leads into a six-foot-wide corridor which is about 300 feet long.
Streaks of light penetrate a mass of vines and creapers which choke the wide
opening fifty feet above. Two small holes lead out of this corridor. The first
which is a triangular hole just wide enough to admit an adult, leads into a small
room. Loose stones form an uneven floor. Low passages lead off into darker
depths. The second opening from the corridor is smaller and even better con-
cealed. By crawling flat down a descent of five feet, a large chamber about 100
feet long is reached. Patches of sky appear through the covering of vegetation
nearly 100 feet above. Some passages lead off; running water is audible from
a lower level.
Although this must have been an ideal hiding place, only a few potsherds have
been recovered from here.

Great slabs of granite form a massive outcrop which overlooks cultivated land
about 1 miles beyond Kyempisi. No paths lead to the cave entrance. A way
needs to be cut through thick elephant grass to reach the rock face. Here there
are many clefts and cracks. One large cleft in the rock leads upwards and
steeply into a wide corridor. A frenzied squeaking and whirring of wings heralds
the presence of literally thousands of bats; along the corridor the walls seem
alive with the creatures clinging to the rock in throbbing masses, the glow of
their eyes shining in the obscurity like myriads of sparkling beads. Once
disturbed the bats' continuous flying in such numbers can hinder progress.
From this corridor a low passage leads into a large chamber higher up
within the rock formation. This same chamber is more easily entered from the
summit of the outcrop. From the top of the rocks a large naturally-formed shaft
leads down between almost square blocks of granite for over a hundred feet into
the corridor-cum-chamber already mentioned. From the far end of this chamber
a narrow cleft opens to a sheer drop of ten feet from where a passage leads away
into smaller rooms and other passages.
Bats are plentiful here too.19 Bat guano is several feet thick on the cave floors
and must amount to many hundreds of tons. Broken pottery and cowrie shells
have been recovered. A small vessel, which locals believe might have been used
for holding perfume, was found close to the lower entrance.20 (Fig. 3.)

Still within Gombolola Ssabagabo, Muchululu Cave lies near the bottom of
a valley in the neighbourhood of Kalaguza.
Entrance is gained by squeezing through a narrow hole from where two
passages branch into two different underground rooms. One is about 15 feet by
120 feet in size, the other 50 feet by 200 feet. The usual finds of potsherds,
cowrie shells and pipes were enhanced here by the recovery of a decorated ear-
ring or bracelet of copper alloy.21 (Fig. 4.)

This is a shelter and small cave situated about three miles from Namalwa,
close to the county border with Buwekula County. The site is located within a
large granite outcrop near to a massive rock called Mutengensa. Access to the
refuge is through the shelter and along an insignificant but difficult outward-
sloping ledge. The cave, which is really more of a wide passage-way, is
remarkably dry inside. On my only visit in 1953 stalks of dried grass lay on
the dust-laden floor which was quite undisturbed. Some earthenware pots stood
close to one of the walls. Along the other wall brittle forked branches, uprights
and horizontal lengths, were set in place as if to serve as frames for resting
places. There were no bats in the cave, at least at the time of my visit.

FIGURE 4. Ring from Muchululu Cave
(full-scale drawing by George Kakosa.)


Muzimu lies close to Kigulwe trigonometrical point in the neighbourhood of
Kasambya. It is much spoken of and is said to have been used frequently
whenever invaders threatened the area. This cave is not easy to locate owing
to the fact that it is situated in heavily forested undulating country at the foot
of the hills Bisoke, Murahi and Rwanjale. On the death of the keeper of the
cave, Omw: Rukara Omunhyana, some years ago, his successors left the area
and so guides are no longer easily found. As the name implies this is the cave
of the spirit of the dead. Since it is held in considerable awe I was not surprised
when the first local enquiry about its location evoked the written reply-
"Muzimu is very awful and unapproachable place therefore all the inhabitants
of that country can not dare to lead strangers to the place".
Entry is gained along the bed of a stream. This is the Nyabisinja which flows
into the narrow entrance to lose itself in the depths of an inner chamber. This

is the main entrance. Nearby there is another opening in the rock which widens
out and appears to lead off for a considerable distance. On two occasions the
presence of thousands of bats have made it impossible to enter very far.
Investigation of the main chamber shows that it can be divided into three,
basement, ground floor and an upper storey. All these floors have been occupied.
The complete bowl of a pipe was found in the basement as well as a complete
vessel that is believed to be a bowl for sauce, which was on the ground floor.
Potsherds were plentiful too. Two hearths and blackened rock surfaces were
found on the upper storey.
Fissures and small passages lead off for short distances, but it is the one
great vault enclosing the three levels, which seems to comprise Muzimu, with the
addition of the second chamber near the entrance. It has been said there are nine
entrances to Muzimu.22 This seems unlikely, unless reference has been made to
other caves or shelters in the immediate neighbourhood. It is more probable that
loose description has given rise to a misunderstanding. In the event the number
of entrances-to say nothing of size-of caves has invariably turned out to be
fewer than the imagination of informants has led them to believe. There are,
however, according to one local man, nine undemarcated sections or partitions
within the main cave visited. These he named as-
Mugabente Nyakaisiki
Lwentama Nyakabimba
Lwembuzi Kanyinamwiru
Rwenkoko Karukoha

There are numerous caves and shelters in the outcrops and hill-tops of the
area of Kakumiro. Many are easily accessible and almost all have seen occupation
from time to time. In particular there are-
Bwanswa (or Bijogoli) cave which is close to the headquarters of Gombolola
Ssabagabo at Kakumiro, with a large rocking stone nearby. Bwanswa is more of
a rock shelter-two shelters in fact-than a cave; it is large and quite
Bikekete shelters and cave within the Munsa Earthworks.23
The shelters around and beneath Lwensera rock.
The Bigoma shelters, opposite Lwensera rock. (Fig. 5)

Semwema Hill is the most prominent of all the granite hill-tops of the area.
There is no doubt that it has afforded shelter for refugees over the years; but
in no part of it are there any inner rooms or chambers of the same vastness as,
for instance, Muzimu Cave or other caves in Buyaga County. There are a
number of tales of the extent of the Semwema caves and at least one account
tells of "two chambers". These stories are misleading. Invariably they emanate
from well-intentioned informants who, when it comes to double checking, have
never bothered to investigate the places for themselves.
There is a fairly large shelter in the northern section of Semwema Hill which
is known as Kateboha's shelter.25 On the eastern slopes there is one very small
shelter. There are a few minor recesses along the western face. Almost centre
of the western slopes the hillside is honeycombed with narrow passages. The
entrance to these cracks and fissures in the face of Semwema is by one cleft

below a large overhanging rock. With the exception of Kateboha's shelter there
is not a great amount of room for any large number of people to hide within
The area covered by the warren-like passages is extensive. The passages are
narrow, steep in parts and mostly dark. By following them from the foot of
the hill it is possible to find a way up to within a few feet of the summit.
Semwema lies in a well-populated area which, from archaeological evidence,
has been closely settled for hundreds of years. With roots deep in the past
together with affiliations with Kateboha, the legendary builder of the nearby
earthworks of Munsa, exaggeration of the hill's potentialities for concealment
as well as fanciful stories have arisen. The facts are that Semwema's shelters
and passages are relatively small and restricted in number. In addition, claims
that cave paintings are extant on the rocks of the hill have yet to be proven.
One story, not without interest, is told in all sincerity. On a section of smooth
rock near the summit on the western side of the hill in a seemingly inaccessible
position, are what appear to be water-worn indentations. The local farmers call
them kituli. They are believed to be man-made and to resemble the feAtures
of Kateboha. According to the story, after Kateboha's departure from Munsa,
those he had left behind chose this rock on which to fashion his likeness and
so preserve his memory. Unlike my informants I have never been able to see any
likeness in these apparent erosion scars to any form whether animate or inanimate.

The granite in which these shelters are located overlooks the Munsa Earth-
works, Semwema Hill and the Kakumiro area generally. Cultivation spreads to
the very foot of the outcrop and here sherds litter the soil thickly. Two small
shelters are of interest.
At Butozo I, located amidst tumbled rocks, there is plentiful evidence of
occupation. There are potsherds, broken pipes and tell-tale hearths. I have been
told that in addition to being used when necessary as a hiding place, it has also
been a meeting place where beer would be taken in the cool shade of the rocks.
At Butozo II, which is less easily accessible, pottery has been found having
punch marks, and some smeared with paint. Piles of rounded stones were also
found along a ledge commanding a good view over the eastern approach to the
site. The stones were piled as if in preparation for throwing. From the number
of rounded stones lying here it has been assumed these were missiles.26


Caves in Buwekula County-the third county of Mubende District-are no
less spectacular than those in the other counties, even if smaller. Of the various
localities, one in particular that deserves mention is the group of shelters and
small caves of Ibale, 4j miles north of Kasolokamponye on the Madudu-Buta
road, from where a complete bowl was recovered.



ILewin, Rev. H. B. Geographical J., (1909), May, p. 599,
2Wayland, E. J. List of Caves in Uganda, (1920). (Including names, locations and notes
on history and legends). Geol. Survey Dept., Uganda. Unpublished.
3Brachi, R. M. Uganda ]., (1960). 24, pp 62-70.
4Wayland, E. J. Uganda 1., (1934). 2, pp. 28-29 and various unpublished reports.
Lawrance, J. C. D. Uganda J., (1953). 17, pp, 8-13,
Lanning, E. C. Uganda 1., (1955). 19, pp. 179-181, and Man, (1955), 81,
Oilier, C. D. & Harrop, J. F. Uganda 1. (1958). 22, pp. 158-163.
Posnansky, M. U.M. Occ. Paper, 4, (1959). p. 33.
5Ggomotoka, J. T. K. Uganda 1., (1950). 14, pp. 85-91.
6Adimola, A. B. Uganda J., (1954). 18, pp. 174-176.
7Wayland, E. J. op.cit.
SWayland, E. J. op. cit.
9Lewin, Rev. H. B. op. cit.
Ggomotoka, J. T. K. op. cit.
Gray, Sir John. Uganda J., (1961). 15, p. 118.
Wallis, H. R. Handbook of Uganda (1920). 2nd Edn., p. 39.
10 Cyprien Mutagwanya, one-time Luwekula, Mubende (died 1954).
Yakobo Byomere and various Banyoro elders. Late Mugema of Bunyoro (died
2 December 1962). Oral communications 1951-1953.
11Ggomotoka, J. T. K.op. cit. p. 89.
12Bats are plentiful but harmless (however see note 18). Though there is evidence of
rats and porcupines they are seldom seen. As in neglected tomb-huts the greatest care
needs to be taken in caves of hornets which can become swiftly aggressive and temporarily
painful. Fire and smoke are the only remedy.
13Wayland, E.J. Personal communication, 1961.
14Lanning. E. C. Uganda 1., (1960) 24, p. 187 and (1955) op. cit.
15Map. Mubende District, 1:250,000, A 1113. Lands and Surveys Dept., Uganda,
2nd Ed. 1956.
160'Brien, T. P. The Prehistory of the Uganda Protectorate, (1939) Cambridge Univ:
Press. Personal communication, 1961.
17Lanning, E. C. S. Afr. Arch. Bulletin, (1956), 44, pp. 102/3.
18U.M. A 54.19.
19After a visit to this cave in 1953 both my companions, the Kyambalango (County
Chief of Buyaga) and Mr. Tony Henly went down with fever, assumed to be malaria.
This coincidence suggests the greater likelihood of "cave-fever" as described in a note
in the S. Afri. Arch. Bulletin of September 1954-
"Caves in Northern Rhodesia are mostly infested with bats. Two weeks after an
investigation had been completed by Oakley and Desmond Clark, both of them
developed a fever which was at first mistaken for malaria but was almost certainly "cave-
fever" histoplasmosiss), due to a fungoid infection perhaps conveyed by bats. The
disease is at present being investigated by the Medical Research Institute in
20Lanning, E. C. Proceedings of 3rd Pan-African Congress on Pre-History, (1955).
Chatto and Windus. p. 314.
U.M. A 52.178.
21U.M. A 5828/3.
22Ggomotoka, J. T. K. op. cit.
23Ggomotoka, J. T. K. op. cit.
Lanning, E. C. Uganda I., op. cit.
24Lanning, E. C. Uganda J., op. cit.
25Lanning, E. C. Uganda I., op. cit.
25Lanning, E. C. Uganda I., op. cit.
26Lanning, E. C. Man, op. cit.



[In a footnote to his article The British and Bunyoro-Kitara, 1891-1899 (U.J.
1960, p. 239), A. R. Dunbar has referred briefly to the circumstances in which
the "lost counties" of Bunyoro were annexed to Buganda. The following article
attempts to explain how the annexation came about.]
On his way up to Buganda in September or October 1893 Colonel Colvile,
who with three other officers had been despatched by the War Office to meet
Sir Gerald Portal's request for four Arabic-speaking officers, heard that Portal
had in fact already passed him on his return to the coast. Realising that, as senior
officer, he was now Acting Commissioner in Uganda, Colvile opened the des-
patches he had been carrying up for Portal.1 Among them he found orders from
the Foreign Office, dated 10th August, to send emissaries into the Nile basin,
together with forms for them "to negotiate any treaties that may be necessary
for its protection."2 This was, no doubt, the immediate reason for the attack on
Bunyoro which resulted in the annexation of the "lost counties". No safe passage
down Lake Albert and the Nile was possible without British freedom of movement
in at least part of Bunyoro, and this Kabarega was hardly in a mood to concede.
Portal's withdrawal of the Sudanese garrisons stationed by Lugard in Toro had
at once prompted Kabarega, exasperated by the depredations of the Sudanese,
to renew his southward attacks.3 One may surmise that it was this news-received
on his way down to the coast-which caused Portal in his Report4 to revise his
original opinion that there was "no object worth gaining by a forcible subjection
of Unyoro."5 At the time, he could only order the total evacuation of Toro;6
but it was abundantly clear to Colvile, as it had been to Lugard, that, quite
apart from the intelligence requirements of the Foreign Office, British control
over Buganda alone was neither desirable nor possible. If the British were to
maintain their position on the Upper Nile they must control all the country
between the Nile, Lake Albert, and the German and Belgian territories, which
Bunyoro threatened to reconquer.
In December 1893 Colvile therefore launched the first of a new series of
campaigns against Kabarega; and early in 1894 he had driven the Banyoro armies
north of the Kafu and Nkusi rivers. He obtained a footing on Lake Albert as
far north as Kibero, and by controlling the salt mines there deprived Kabarega
of his principal remaining source of wealth;7 he was, moreover, able to send
Major Owen down Lake Albert and the Nile to Wadelai where, by making a
treaty, the latter facilitated safe river communications as far north as Dufile; he
also obtained news of the Mahdi's "dervishes". Colvile then announced, in
public baraza at Kampala on April 9th.,8 that Bunyoro south of the Kafu
and Nkusi was now annexed to Buganda, and he divided this territory into
two parts where Catholics and Protestants could each settle and proselytise with-
out coming into conflict.
It has hitherto been assumed9 that the annexation was a "reward" to the
Baganda for their loyal assistance. But, as we shall see, this was not the reason
why it was eventually confirmed in 1896, and it can only have been a subsidiary


one in 1894. That Colvile was astute enough to give out in public that this was
the reason does not mean that there were no others.
Except for its commander, Kakunguru, Colvile's opinion of the Baganda army
under his control was not exactly flattering (indeed, he was rather more impressed
by what he saw of the Banyoro); and in any case he had been prevented from
going very far beyond the Kafu in pursuit of Kabarega by the fact that the
Baganda were tired of serving in a country which could hardly support them
and wanted to go home.10 "Although I meant business," he wrote, "they [the
Baganda] were happily unaware of the fact," hoping only to "return home, the
richer by a few goats or head of cattle".1 It is unlikely that many of the
Baganda leaders greatly prized this generally devastated and under-populated
territory. From the British point of view, however, southern Bunyoro had
considerable strategic value: if Colvile was to hold a line from Kibero
to Mruli, it would save him much anxiety and expense of man-power
if he knew that so much of the country between this line and Buganda's original
northern frontier was not just "occupied" by native troops but firmly under
traditional Kiganda authority.12 In the east, the annexation meant a partial
insurance against further Banyoro infiltration into Busoga (which would have
made good Kabarega's threat to surround Buganda)13 while in the west it meant
certainty of access to Lake Albert14 through the corner of Buyaga, as well as a
strong barrier against further repetition of Kabarega's attacks to the south as
well as his communications with the Arabs.
When Colvile received news that a British Protectorate had been declared over
Buganda he gathered however that the occupation of Bunyoro was approved
only as a temporary measure of defence.15 He did not, therefore attempt to seek
official confirmation of the annexation, but contented himself with establishing a
line of forts from Kibero to Mruli.16 On the other hand it seems that both the
Baganda and the European missionaries were led to suppose that the annexation
was effective; at any rate, nothing was said to dispel this impression. It was
above all the Catholics who wished to take advantage of it, for though Portal
had improved upon Lugard's settlement of 1892 by giving them, in addition to
Buddu, Mawakota saza and part of Singo (the future Buwekula sasa),17 they
still felt congested enough to welcome the chance of settling in new country.1
The Baganda who joined up at the end of 1893 included some who came from
as far south as Buddu;19 and Pere Achte of the White Fathers came straight from
Buddu after the war with a group of Baganda catechists,20 to establish a
mission station at Bukumi in Bugangadzi. Despite the non-confirmation of
Colvile's annexation (and the Catholic writers cited seem to have been unaware
that it had been suspended), a Muganda Catholic, Cyprian Mutagwanya, known
as Lwekula (and after 1900 saza chief of Buwekula), became "gouverneur de la
province" and thus suzerain over the Munyoro Kikukule, who was chief of
Bugangadzi sasa and was later to desert Kabarega.21
When at the end of 1894 Colvile was invalided home, his successor, Ernest
Berkeley, was left no briefing on the situation in southern Bunyoro. The follow-
ing April, Berkeley, acting on the advice of Captain Pulteney, the military officer
commanding. Bunyoro, appointed a Munyoro-no less a personage than Rwa-
budongo, Kabarega's former katikiro-to be chief over the country north of
the Msisi river. The Catholics protested violently: their bishop, Mgr. Guiller-
main, explained to Berkeley that the Catholics had supposed from Colvile's
religious settlement (of which Berkeley knew nothing) that the country west of
the river Kitabui belonged to them; and that many of them were now leaving

because it was being given to a pagan chief. Anxious to maintain good relations
with the Catholics, Berkeley agreed to transfer Rwabudongo to northern
Bunyoro; and when in December he heard for the first time, about Colvile's
annexation,22 he decided it would solve all problems to have it confirmed. It was
important that the Catholics should feel able to stay in southern Bunyoro as
otherwise they would jealously object to the retention of Protestant Buganda
chiefs on the eastern border around Mruli, who at that time had the useful
effect of permitting the concentration of Sudanese troops in northern Bunyoro.
(Six Baganda posts were set up along the Kafu early in 1896.23) But the Catholics
would only stay if they were ruled by Catholic chiefs, and the only Catholic
chiefs were Baganda. So in May 1896 Berkeley eventually reported the
situation to the Foreign Office and asked permission to declare the annexation
confirmed; he anticipated no real opposition from the Banyoro themselves.24
Before his letter reached London, the Foreign Office, now under Salisbury's
energetic direction and anxious, in face of the Ethiopian victory at Adowa in
February, to consolidate British influence in the Nile valley, had in fact already
decided to extend the Protectorate over Bunyoro, together with Toro, Ankole,
Busoga, and other parts to the eastward.25
Berkeley, on the other hand, wanted the assumption of a Protectorate to
provide for the actual transfer of territory to Buganda. This was not simply
because he needed the authority to set up permanent forts and to introduce civil
administration; it was also because he felt the need to placate the Catholics.
After receiving the approval of the Foreign Office (who were in fact indifferent
to the annexation now that Bunyoro was in any case a British responsibility26)
he therefore called a baraza at Kampala on November 18th and confirmed
Colvile's arrangement of 1894.27
The annexation did not pass unchallenged. In May 1896 when Berkeley
had decided to seek official approval for it, he encountered strong opposition
from Captain Pulteney. According to Pulteney,28 not one chief had made his
submission to British authority (which seems to have implied to him that the
British could not properly claim it as a part of Buganda, which at that time was
as far as the Protectorate writ ran and was bound to Britain by a voluntary act of
treaty on the part of its ruler and leading chiefs). Guessing that at Kampala
Berkeley might have been unduly impressed by the claims of the Catholics,
Pulteney denied that the Catholic Muganda chief Lwekula had any rights beyond
Colvile's advanced outpost at Bukumi; he must have misled Captain Ashburn-
ham, the officer in Toro, if the latter supported his claims; and besides,
Lwekula's raids on Bujiro and Busesa were provoking Banyoro chiefs into
threatening to go over to Kabarega. PNre Achte, said Pulteney, did not claim any
special Catholic rights in Bukumi; Mgr. Guillermain was wrong to say that the
Catholics had penetrated Bunyoro peacefully; and in any case surely a letter
from the Foreign Office (No. 66 of 20th April) had quashed any claims to
religious divisions outside Buganda. If, Pulteney added, southern Bunyoro was,
as seemed likely, handed over to the Catholics, he would be obliged to resign his
civil duties.29 The situation was evidently so critical that a decision had to be
taken quickly: and before the annexation became official, Berkeley accepted
Pulteney's resignation of civil duties and assigned these to Mr. Forster, who was
stationed at Nakabimba.30 In July 1896 Captain Sitwell, on Berkeley's orders,
appointed the Muganda chief Kagero to be chief of the country,31 and in
September Kagero informed Forster that he had been told to "give the Catholic
Baganda their rights". Forster, however, refused to allow either Mgr. Streicher or


Pere Roche to "enslave" the Banyoro;32 and by October the Catholics agreed
to return to Buganda pending a decision on Bunyoro by the Foreign Office.33
Forster was convinced that the Catholics had made themselves unwelcome by
the fact that when going on safari Pere Roche always wanted Forster to provide
him with an armed escort;34 on the other hand, Kikukule, whom Kabarega
sent to Bukumi to seek the White Fathers' assistance in opening negotiations
with the British, "became so friendly with the Fathers that a ceremony of
blood-brotherhood took place between Kikukule and Pere Achte."35
But the Foreign Office letter of approval, dated 8th August, must have reached
Entebbe soon after the Baganda had made their withdrawal, and the annexation
went through despite all protests, with assurances from Berkeley that it need
cause no religious coercion and that it gave the Baganda no right of pillage. The
inhabitants, Berkeley grandly declared, were now Baganda like themselves. No
Banyoro chiefs loyal to the new regime were to be disturbed in their chieftain-
ships.36 For the first few years there was, inevitably, "constant trouble as the
Banyoro chiefs resented having Baganda chiefs over them".37 Late in 1896
Berkeley had to go home through sickness, and Colonel Ternan, who became
Acting Commissioner in his absence, gave the Catholics full rights of possession.
The Baganda used the Banyoro as slave-labour, and Forster, like Pulteney,
resigned in protest. In December the British fort and the French mission were
stormed and destroyed, and the Baganda chiefs were apparently chased back to
their own country.38 Administration became firmer in 1900 when a station was
set up at Kakumiro,37 but many Banyoro emigrated to escape hut tax.39 Between
1900 and 1902 a hundred were reported to have left Buyaga, not to evade tax,
but to escape the rule of Baganda chiefs, "who treat them somewhat as serfs."'4
Nevertheless there do not seem to have been any formal complaints, either from
Banyoro in the "lost counties" or from Bunyoro itself at the time when
Johnston made his Agreement with Buganda in 1900, by which Buganda's
frontiers were so defined as to include these counties (though this does not mean
that Berkeley's announcement of 1896, and the Foreign Office letter of sanction,
were in any way superseded as the legal-or rather, perhaps quasi-legal-basis
of annexation, which was simply confirmed, as it had been in the proclamation
of March 1898 on the appointment of Kitahimbwa to succeed Kabarega).41
Johnston however probably accepted the annexation as so much past history-
though Ternan had admitted that southern Bunyoro had been annexed "contrary
to the wishes of its owners".42 There seems to be no record earlier than 1915 of
official awareness that the royal tombs of Bunyoro lay within the lost counties,43
and in any case no Banyoro claims were likely to be taken seriously as long as
the British spoke, as Johnston did, of "the so-called king of Bunyoro" who should
rather be called "king by courtesy";44 or considered as did George Wilson
(a propos the 1907 chiefs' rebellion) that "Bunyoro must clearly understand that
it is a conquered country, and more than in any other the disposal of authority
lies with His Majesty's Government."45
From this brief study, it should be evident that the annexation of the "lost
counties" of Bunyoro, while due in part to a number of misunderstandings, was
originally undertaken by the British for primarily military reasons, and was
confirmed chiefly in order to avoid provoking barely concealed religious dis-
sensions. Indeed, it may to some extent be regarded as partly an after-effect of
the original religious partition of Buganda and its modification by Portal; though
it was only made possible by the Conservatives replacing the Liberal Government
in 1895. For a period in the 1920s the Protectorate Government became quite


sympathetic to Banyoro claims for the restoration of the lost counties, but there
was no mention of them in the Bunyoro Agreement of 1933, and indeed, by
then, to have returned them to Bunyoro would probably have caused a major
disturbance in Buganda, where they had come to be taken for granted, however
erroneously, as the spoils of war.


(All references, unless otherwise stated, are to files in the Uganda Government
Archives, Entebbe).
IColvile, Land of the Nile Springs, p. 13. This contains the fullest account of the
1893-4 campaign.
2FO to Portal, 10/8/93 (A31/1). Colvile, very properly, did not refer to these orders
in his book (published in 1895), and it is worth remarking that Rosebery, the Foreign
Secretary, sent them without consulting any other member of the Cabinet, who were
furious when they learnt of them nine months later.
(Gardiner, Life of Harcourt, II, p. 315).
SJackson, Early Days in East Africa, p. 264; c.f. J.W. Gregory, The Foundation of
British East Africa, p. 213.
4Portal to Rosebery, 1/11/93 (Africa no. 2, 1894, cmd. 7303).
5Portal to Rosebery, 24/5/93 ibidd).
6Macdonald, Soldiering and Surveying in East Africa, p. 293.
7Colvile to Cracknall (Zanzibar), 5/2/94 (Africa no. 7, 1895).
8Berkeley to Pulteney, 30/5/96 (A5/2/160). According to this letter, Colvile recorded
his declaration of annexation in the "station journal"-presumably at Entebbe-but
this has not been preserved.
9Cf., e.g., Kagwa, Customs of the Baganda (1934 tr.), p. 169; K. Ingham, Some Aspects
of the History of Buganda (UJ 1956, p. 10); Enquiry into Land Tenure and the Kibanja
System in Bunyoro, 1931, p. 6. James Miti, in a typescript history of Uganda in
Makerere College Library (written about 1937) says (p. 466, English version) that on
27th August, 1894 Colvile read out to a large assembly a message from Queen Victoria
thanking Mwanga for his assistance against Kabarega and "as a special gesture of
appreciation" giving to Buganda that part of Bunyoro abutting on its western border.
But this hardly squares with the repeatedly expressed fears of the Liberal Government
that it might be compromised in Uganda by the precipitate action of officers on the spot;
it probably represents a confused recollection of the actual declaration of a British
Protectorate over Buganda, which Colvile issued on this date.
10Colvile, op. cit., passim; Colvile to Cracknall, loc. cit., and cf, Thruston, African
Incidents (1900), p. 140.
11Colvile, op. cit., p. 95.
12In his book, Colvile evidently felt unable to refer to the abortive annexation. But
he made it clear (p. 144) that Bunyoro had to be occupied, and in view of the shortage
of reliable troops he hoped in January to make treaties with the chiefs in southern
Bunyoro. To do this, however, he had to convince them that Kabarega's power was
broken, and though, on the 18th February, he entrusted Owen with the task, among
others, of organising a "confederacy" of friendly Banyoro chiefs (p. 214), no mention
of Owen's failure to do this is made when Colvile later refers to Owen's project of a
Toro confederacy: (p. 229): by the time Owen returned from Toro (17th March), Colvile
had probably decided in favour of annexation.
13Colvile, op. cit., pp. 68-9.
14Macdonald, op cit., p. 306.
15FO to Colvile, 10/6/94 (A31/2/161).
16Cunningham to Jackson, 7/6/95 (Africa no. 1, 1896).
17D. A. Low, in D. A. Low and R. C. Pratt, Buganda and British Overrule (1960),
p. 12.
18(R.P.) Antony Phlipe, Au Coeur de l'Afrique, etc. (1929), p. 121; cf. Colvile to FO,
29/7/94 (A32/2) and FO to Colvile, 18/1/95 (A33/1/4) for further evidence of the
Catholics' understandable desire for lebensraum.
19Jackson to FO, 1/7/95 (Africa no. 1, 1896). Jackson ascribed this simply to popular
enthusiasm to strike a blow against Kabarega, but the hope of new land for settlement
must also have been a powerful incentive.


20G. Leblond, Pere Achte (1928), p. 174.
Archdeacon Walker of the C.M.S. also thought the annexation was effective, cf. his
letter 9/7/94 in the Church Missionary Intelligencer for January 1895.
21Ibid., p. 178. Leblond says disappointingly little about this first missionary
enterprise outside Buganda proper (he thought it was inside Buganda) beyond giving a
blurred picture of peaceful infiltration.
22As A. R. Dunbar has noted (loc. cit.), when Berkeley, in November 1895, sought
Foreign Office approval for including two civilian officers in the 1896-7 estimates, he
explained that they would be very useful between the northern frontier of Buganda and
the Kafu. (Berkeley to FO, 20/11/95; (A34/1/127).
23Berkeley to Ternan, 25/1/96 (A5/2/8).
24Berkeley to FO, 14/5/96 (A34/2/58).
25FO to Berkeley, 17/7/96 (A33/3/98).
26FO to Berkeley, 8/8/96 (Further Correspondence relating to East Africa, XLVI,
p. 134).
27Berkeley to Forster, 19/11/96 (A5/2/194).
28pulteney to Berkeley, 19/5/96 (A4/5/146).
29Ibid., see also Pulteney to Berkeley, 21/6/96 (A4/5/178).
The Foreign Office letter which Pulteney had in mind was probably one of two
letters, dated 18th and 20th April 1895, sent to the Protestant and Catholic missions
respectively, by which the missions were no longer obliged to work in different spheres.
(Further Correspondence, etc., XLIII, pp. 55 and 62).
30Berkeley to Pulteney, 30/5/96 (A5/2/100).
31Forster to Berkeley, 30/7/96 (A4/5/195).
32Forster to Berkeley, 29/9/96 (A4/5/294).
33Berkeley to Forster, 13/10/96 (A5/2/170).
34Forster to Berkeley, 1/11/96 (A4/6/334).
35A. R. Dunbar, loc, cit. p. 234.
36Berkeley to Forster, 19/11/96 (A5/2/194).
37Buganda Annual Report, 1907-8 (SMP 08/1195)..
38J. W. Gregory, op. cit., pp. 217-8.
39Cf. Ormsby to Tomkins, 29/2/04 (A8/4/04/58); Regents to Tomkins, 23/3/05
(A8/6/05/122); Manara to Tomkins, 1/12/05 (A8/7/05/324).
40Bagge to Sadler, 16/4/02 (A12/2/32); cf. Jackson's Comments on Regents to
Jackson, 16/1/00 (SMP C450), quoted in D. A. Low, op. cit. p. 55.
41Enclosure in Wilson to FO, 16/3/98 (Further Correspondence, etc., LIII, pp. 165-7).
42Ternan to Grant, 27/5/99 (A5/5/114).
43In 1915 an arrangement was reached with the Buganda Lukiko whereby the tombs
of the Abakama were to be allotted to the Mukama after being marked out by Govern-
ment surveyors. (Land officer to District Commissioner, Bunyoro, 19/12/15: Western
Province Archives, Fort Portal, no. 443).
44Johnston to Spire, 6/2/00 and 13/4/00 (A5/9/68 and 186).
45Wilson to Eden, 6/5/07, telegram (SMP 07/267). The authors of the Enquiry into
Land Tenure cited above (note 9)-IL B. Thomas and J. G. Rubie-concluded by
remarking that "we think the time is come when reference to Bunyoro and the Banyoro
as a conquered country and people should cease".




MOUNT ELGON was the scene of intensive inter-tribal warfare during the
whole of the 19th and early decades of the 20th century, when the Sebei-speaking
semi-pastoral tribes living on the mountain were involved in conflict with cattle-
raiders from the surrounding tribes. It is proposed to discuss some aspects of
this warfare in this paper. Sociological information about the tribes mentioned,
their methods of warfare and the historical facts laid out in chronological
sequence was collected by the writer during discussions with old people, some
of whom were eye witnesses of or had actually been involved in the fighting. A
list of informants and their background is given as an appendix.1 The dating
of events is, in the minds of the informants, in terms of the age-set cycle; but
it has been possible, by establishing the original and regular system of the age-sets
and then by checking some of these events with known dating by the European
system, to arrive at a reasonably reliable time-scale that reaches back to the
beginning of the 19th century.
The earliest and possibly most formidable enemies of the Sebei were the
Perrko Masai or Burrugo, as the Sebei called them, who were attacking Sebei
in the early decades of the 19th century and probably even earlier. (The Gondjek
of south Elgon say that the Masai raids preceded even the Karamojong). The
Suk are reported to have attacked Sebei for the first time in the 1830s, mainly
against the high hill of Riwa above Greek River on the eastern foot of the
northern slopes. From that date Suk and Karamojong attacked intermittently;
as they still tend to do. In the 1880s the Nandi began to attack, striking right
into the heart of Sebei up to 10,000 feet or more. (This period of Nandi
aggression began after the final defeat of the Uasin Gishu Masai who had
occupied territory between Mount Elgon and the Nandi.) The impression gained
from traditional accounts is that prior to this date the Masopyisiek enjoyed more
peace than the Soyisiek, who had been standing the intermittent hammering
on the lower levels of the mountain for some decades. After about 1910 the
Nandi raiding died down, possibly as a result of the British punitive expedition
in 1905-06 and the establishment of the Reserve.

The Sebei-speaking groups living on Mount Elgon belong to the so-called
Kalenjin peoples of West Kenya, being a pastoral-agricultural people of Nilo-
Hamitic origin. They comprise the Mbaiyek, Soriek, Sabindjek, Gondjek, Pokek
and Somekek, and they inhabit roughly two-thirds of Mount Elgon. In the past
members of these groups grazed their cattle from the plains of Kitale in the
East across the mountains and down beyond the Siroko River in the West.




D Area above
territory of

FIGURE 1. The main waves of incursions.


iF il*
...... I

--lltllll ---- ---

Territory of MBAIYEK




_, __ POKEK ai



SArea above 9000 ft. and
territory of MASOPYISIEK

FIGURE 2. Distribution of the Sebei-speaking groups on Mount Elgon.



They also ranged into the Northern Plain as far out as Tabasiat (Debasien).
These six groups can be taken in order, starting from the Sisii River on northwest
Elgon and following in an easterly and southerly direction, and finally on the
south side of the mountain in a westerly direction as far as the Rivers Terim
and Malakisi. Between the Sisii and Kyebonet Rivers on the northwest slopes of
Elgon is the Mbaiyek group. Beyond them is a small section of country roughly
between the Rivers Kyebonet and Kyeseberr, occupied by the Soriek. The
Sabindjek, who are the true Sebei of Elgon, occupy the largest area of all the
groups, stretching eastwards from Kyeseberr as far as the small river Chepchoina
which flows on the east side of the mountain some five miles south of the River
Suam. Southwards from the River Chepchoina as far as the River Kimilili in
the Elgon Nyanza Province are the Gondjek, who form the second largest group
on Elgon. To the west of this group, roughly between the River Kimilili and the
rivers Terim and Malakisi, are the Pokek and Somekek.
Each of these six groups of mountain people is itself further divided into two
main categories, known in Sebei as the "Masopyisiek" and the "Soyisiek". The
former means "People from the Top" and the latter "People from Below". It is
necessary however to point out here that the Mbaiyek, Soriek, and Sabindjek who
inhabit the north and northeast slopes feel a closer affinity among themselves
than they do with the Gondjek, Pokek, and Somekek who inhabit the south and
southeast slopes. The Masopyisiek inhabit an altitude from about 11,000 feet
(just below the crater rim) down to about 9,000 feet, living mainly above the
forest level, but also in clearings of short grass inside the forest belt. The
Soyisiek do not live much above 6,500 feet (the Forest Reserve stretches between
6,000 and 9,000 feet) and are to be found below that level right down into the
plain. The Masopyisiek tend to identify themselves with the six different groups
already mentioned, depending on the area which they inhabit, e.g. Masopyisiek
of Sabin or Masopyisiek of Mbai or Masopyisiek of Zgoin. They have always
been parasitic on the Soyisiek, making journeys down through the forest to barter
their bamboo baskets and mats and edible bamboo shoots and honey for millet or
maize which they carry up on donkeys to store in their granaries.
Certain main sociological features bind together the Mbaiyek, Soriek and
Sabindjek on the one hand and the Gondjek, Pokek and Somekek on the other.
These are the territorial System of the Bororiosiek, the tribal age-sets and the
clans. The territorial military divisions of the tribe are known as Bororiosiek.
Bororiet, (singular Bororiosiek) is related to a word "Boriet" which implies
"fighting". In the territories of the Mbaiyek, Soriek and Sabindjek there are
twelve of these Bororiosiek; among the Gandjek, Pokek and Somekek there appear
to be only four. These territorial units helped in the past to integrate what was
an apparently loose-knit community. The tribe had no system of chiefs, and
the scattered homesteads within the Bororiet practised a mutual help system
among neighbours and were dependent on an elected counsellor, "Kirwogindet",
who handled cases of crime and misdemeanor. There appear also to have been
one or two leading prophets in the tribe, mysterious figures very active in times
of stress, who inherited their gifts and who were consulted on matters affecting
the tribe as a whole.
The age-sets which run horizontally through the whole tribal life and struc-
ture were the age-sets into which were grouped those circumcised within a
certain fixed number of years. The tribe depended for its defence on the age-set
of warriors, tribesmen in their prime who automatically held power until their
group had passed into the category of elders. The brotherhood of this horizontal

binding force linked men throughout all Bororiosiek who belonged to the same
The Sebei have a system of exogamous clans which is different from that of
the Bantu; this clan system was the third main integrating factor, Among the
Mbaiyek, Soriek and Sabindjek groups there are perhaps from 150 to 200
different clans. In the past the clans tended to inhabit certain parts of the
Bororiosiek with as many as twelve to fifteen in a Bororiet, but now they have
become widely diffused throughout the Elgon area.

Before proceeding with an account of the intertribal raiding on Elgon it would
be helpful here to examine what is known of Sebei methods of warfare. Attacks
on the Sebei were often very sudden, in which case it was every man for himself.
At other times the raiders were reported by lookouts from the hilltops, or pro-
phesied a day or two in advance by the Worrgoyandet (prophet). If time allowed
the prophet would hold a council of war with two or three of the leading warriors
known as Kandoisiek (singular, Kandoindet). People of the neighboring
Bororiosiek were then alerted by the blowing of horns (of Eland, Roan Antelope
or Kudu) from the neighboring ridges and hilltops; also by the blowing of the
"Ariembut", a wooden horn covered with a cow's tailskin. Short, sharp repeated
notes were blown. At the sound of the alarm all warriors would immediately rush
for their weapons and the leader of the Bororiet would tell the boys to hide the
cattle. If asked where they were going they would say "Kyebe Awondo", "We are
going to defend", and in this way the word "Awondo" came to mean a group
for defence. The messengers on the hilltops shouted the name of the meeting
place for all warriors, repeating it again and again and all would go in single file
to the place of assembly. If time had allowed, the Kandoindet would have sent
out spies in the early morning to locate the enemy positions. These were the
Yetik (singular, Yotindet). To distinguish from enemy spies, they were some-
times called "Mororindet". On the return of the spies the leaders might go
forward to see the enemy position. Then the Kandoisiek formed fighting groups
of warriors. Each group was called "Luget". The leader of an enemy group was
called "Kirrgit ap Bunik", the bull of the enemy. Several Bororiosiek might be
called in, depending on the extent of the enemy front. The mobilisation was more
or less automatic. In battle the Sebei wore "Sendet" (red ochre) on their ringlets,
as the Masai do, and smeared it in streaks all over their bodies. The Kandoisiek
usually wore the head-dress "Tombesiet" of Colobus or "Subolyit" and a white
goatskin over the shoulders. The prophet, who himself took no part in the
fighting, wore a leopard or lion skin.
When going in single file to the meeting place the more seasoned warriors
went ahead, discussing the best manner of dealing with the attack, and
encouraging the young inexperienced warriors. When grouping before the attack
the leaders would sometimes say words amounting to "Now is the moment to
be strong; if anyone thinks he cannot fight, let him go home." A coward unable
to face the battle might run away unseen, saying nothing, but if the others
knew it they would spear him immediately, saying "Will you leave it all to us
then?" In the past there were no traitors who went over to the enemy; they would
have gained nothing and would have met certain death. In recent times, however,
Sebei have gone down to the Suk to betray a man or his family for money or
cattle, telling the raiders the time that was favourable to attack. The Sebei often
grouped in a semi-circle to attack; the usual formation was alternate bowman

and spearman. When they had succeeded in spearing or shooting any of the
enemy they shouted in triumph, calling out the name of their clan and the
qualities of their favourite bull, its colour and the shape of its horns. Sometimes
after a long battle only the bravest would be left and perhaps finally two men
would remain to fight it out. If at last the enemy warrior was defeated all
present would yell in triumph, "We have killed the bull".
Just before the fighting the prophet sometimes gave war medicine ("Setanik")
to the leader, Kandoindet. This might be tied to an arrow or to the head of a
club ("Rongut") or placed in a hole probably bored into the head of a club.
This weapon was then thrown towards the enemy. On one occasion before an
expected attack the prophet living among the eastern Sabindjek sent "two brave
men to Riwa to fetch the medicine". In the opinion of certain old men, the
throwing of "Setanik" was only used against the Nandi (possibly because the
Nandi were the only enemy known to possess powerful prophets). But on the
other hand it is known that the prophet Kwoiborrt used it against Masai and
Karamojong as well as Nandi.
The Suk and Karamojong also blew horns to assemble their fighters. The
loud hornblowing of the Sebei at an alert was not considered a disadvantage in
betraying the disposition of their forces. "When once the enemy had launched a
raid they would come in steadily no matter how much the alarm was sounded."
Old Sebei warriors are of the opinion that the Suk were better fighters than the
Sebei or Karamojong. This was said to be due to the Suk methods, which were
to go secretly in small numbers and attack and destroy one family or group
completely. In large battles, on the other hand, the Sebei were often too strong
for them. The Suk only used spears in fighting, but their women had bows and
arrows to protect their homes; for it often happened in long-drawn-out battles
that the Sebei might become weak with hunger and would penetrate into Suk
country to take milk or food from the homesteads, in which case the women
could well defend themselves with bows. It appears that among the Gondjek on
the south side of the mountain it was customary for the women to go to the
hilltops to call the alarm "because their voices carried further." It did sometimes
happen on the north side of Elgon that the Sebei would meet the leaders of the
Karamojong or Suk to discuss conditions for a truce. Such a meeting appears
to have taken place on Riwa some time after 1900, on which occasion the two
representatives of the Sebei were Rongosi and Taskin of Bukwa. Taskin refused
the terms and fighting was resumed, this time resulting in the expulsion of all
Karamojong out of Riwa and even Kacheliba.
After the killing of an enemy, the Sebei warrior had to observe a fairly
elaborate ritual. He would jump four times over the body, each time running
round by the head and on no account by the legs. On arrival home he received
a white goat (white for purification) which he took into the bush and killed.
The man had to remain there alone for fourteen days, during which time he was
allowed no sexual intercourse. After the fourteen days he ate the roots of a
certain plant to rid himself of any worms. He was then able to return home. If
there were a great number of warriors who had killed enemies they all had to
go through this ritual, each man going separately into the bush. A man might
accompany his neighbour but he could only eat his own goat. Among the Gondjek
it was the custom to place a spray of leaves of any tree or bush on the dead
body and then jump over the body (not over the legs). The warrior would then
go for eight days to the "Sigroindit", the hut of the unmarried youths. Each
man killed a goat (which could be of any colour) which he ate in that house.

During that period they drank no milk and had no sexual intercourse although
they could be seen by women. If any man broke this rule it was said that the
woman would not be able to bear afterwards; and that the man would lose his
strength and become ill. Warriors who have killed among the Sebei and Gondjek
make a long scar on the chest on either side (not a raised scar), two and one
half to three inches long, called "Watutek". Sometimes they say jokingly, "I
ate one", meaning "I killed one".

In the early 1830s while the Karamojong and Suk were attacking the Sebei
on and near the hill of Riwa on the northeast slopes of Elgon, the leading
prophet among the Sebei was Kimarr of the clan Kapchai. The centre of the
Kapchai clan was on the southern shoulder of Riwa, where a famous tree of
the clan still stands to this day. There are reports that it was a time of famine
for the Sebei; this would not be unlikely since throughout the 18th century
nearly every decade saw a famine in their country. About the same time, just
after 1834, Nameme, the prophet's son, was circumcised. In those days men
were circumcised often at a later age than nowadays, even as late as 25, although
an average age could be taken as twenty years. Nameme's father Kimarr had
been famous in his youth for having made prophecies before the time of his
circumcision. Kimarr's older brother Kyebet was also a well-known prophet
during those years in Sebei. In about 1849 the Karamojong intensified their
raiding and attacked the Tepes on Mount Debasien (called Tabasiat by the
Sebei), while at the same time they drove hard against the Sebei of Kabroron.
They are known to have attacked as high up as the hills Muzoa (10,951 feet)
and Barrangoidj (over 11,000 feet) on the upper north slopes. Kimarr had moved
to Kabroron area by then where he was living near Kaptim. He probably took
refuge in the cave of that name, which has since been associated with his name.
It was a very hard time for the Sebei and Tepes. Refugees from Tabasiat had
come over and settled near Kaptim and lived in "Kilalmet" (flat-topped) houses
below the cliff of Gurreto and in a cave called Kroetoe but they struck a bad
moment in Sebei for the great famine "Rubet ap Kamarianga" had set in, and
refugees poured westward from the central Bororiosiek of Kono, Kabroron,
Tulel and Mornkatwa towards Kamarianga (from which that famine took its
The Tepes were accustomed to eat honey like the Masopyisiek, and were
accomplished tree climbers, using strong lianas (which the Masopyisiek call
"Berowet ap Kel" to hoist their beehives into trees or rocks or to scale cliffs,
During those dangerous times they stored their beehives in small caverns. At a
later date, however, hearing that conditions had improved in Tabasiat, they
abandoned their temporary refuge on Elgon and returned to their country, leaving
behind certain of their goods which have remained in the caverns ever since,
and which an observer today can distinguish quite clearly. The short sojourn
of the Tepes has long since been forgotten by the majority of the Sebei and today
they regard the occupation of such places with amazement, wondering what
strange people can have ever found access to them.
During the time when the Tepes were living in Kroetoe and famine was
raging near Kabroron, the Nandi began to attack the Gondjek on the southeast
side of Elgon, sweeping across in hordes from the Nandi escarpment. This was
the forerunner of decades of raiding by the Nandi, which did not cease until
early 20th century. At that time, estimated to be between 1849 and 1859, Arap