|Table of Contents|
|Kikukule: Guardlan of southeast...|
|The Lower Miocene fossil site of...|
|Some plant fossils from Bukwa|
|The excavation of an Ankole capital...|
|The origin of Lake Nyabihoko, Kajara...|
|The legend of Lake Nyabihoko|
|The diaries of Emin Pasha - Extracts...|
|Redemption in Ganda traditional...|
|The distribution of certain colonial...|
|Uganda bibliography 1967-1968|
|Index to Volume 32 (1968)|
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
STANDARD VIEW MARC VIEW
|Table of Contents|
Front Cover 1
Front Cover 2
Table of Contents
Table of Contents
Kikukule: Guardlan of southeast Bunyoro
The Lower Miocene fossil site of Bukwa, Sebei
Some plant fossils from Bukwa
The excavation of an Ankole capital site at Bweyorere
The origin of Lake Nyabihoko, Kajara County, Ankole
The legend of Lake Nyabihoko
The diaries of Emin Pasha - Extracts XIV
Redemption in Ganda traditional belief
The distribution of certain colonial weaver bird species in Uganda
Uganda bibliography 1967-1968
Index to Volume 32 (1968)
Back Cover 1
Back Cover 2
ATCEKIKUKULE: GUARDIAN OF SOUTHEAST BUNYORo E .Lmn-1
THE LOWER n1JOCENE FOSSIL SITE OF BUKWMA, SEBeI A. WALKER~ I~~
SoME PLANT FOSSILS FROM BUKWA - A. HAMLLTON
THE EXCAVATION OF AN ANKOLE CAPITAL SrrE
AT B~WEYORERE - - - - NT PosNA~srCy
THE ORIGIN OF LAKE NYABIMOKO, K(AJARA
Country, ANKOLE-- - - - B. H. MP~II~
THE LEGEND OF LARE NYABIHOKO - J. KARIBIA AND,,
E. MI. K(AHZrr
THE DIARIES OF EMITN PASHA--EXTRACTS XIV SIR JOHN GnAY
RIEDEMPTION IN GANDA TRADITIONAL BELIEF A. M. LuomRA
THE DIsTam~rFIoN or CERTAIN COLONIAL. WEAVER
BIRD SPECIES IN UGANDA ------ J. R. HIAx.. 2O
GoaDNows EXPEDITION IN SEARCH OF ,::~
NYAMUYONJO - - - - SIR JOHN GRAY fe
Gonoon's SAH.ING BOATS IN EUTORIA ---H. B. TnOxAs
Af ACDONALD) S MSR THE EVENTS
or 1897-1899 1 .S, --A. T. Mnamso
EARLY CHRISTIAN OR ND
AND THE So *ruL~ - E. B. H~I AND H. B. THousLA 1I~a \
SPINES ON TEEE THE Porr~o -- --A.WLE
OBITUARIES WAY n
Lone TWINING F ANYIKAnK AND GoA--J. Sanms.~
TERENCE PA T IED~I N -- .ONAAt
GEoncE CHAR RNER -,/- .R S. J. K. BAza ..1~
Physiognomie, Struktur Kampalas
(By' K. Vorlaufer) H. CAROL
The RoyalCapitll ofBuganda (By' P. C. WV.Gutkind) F. B. WELBOURN
AIgriculture in the Congo Basin (By M~. P. Miracle) A. R. DLINRAR
Tarms (By D. G.Coursey) - - - MI. POSNANsKYuIS
UGANDA BIBLIOGlu\PHYI 1967-1968
Comrpiled by B. Wt. LANGLANDS
INDEX TO VOLUMlE 32 (1968) -- -- -- -- -- 2
10 THE UGAD SOCIETY
PrceSh, 5/ (gada
I ne uganoa Journart
THE JOURNAL OF THE UGANDA SOCIETY ~~~
Mr. S. G. Grimley, M.B.E
Mr. M. Kaggwa
Mr. W. S. K~ajubi
Dr. G. R. Karongole
Dr. M. S. M. Kiwanuka
Mlr. A. W. R. McCrac
MrS. M. M~acpherson
Dr. J. Mukilih
Mr. J. D. Rubadiri
Dr. WY. A. van Eck
Mr. B. Nabuguzi
Mrs. J. Bevin
Professor B. W. Langlands
Mis. M. Noble
Hon ILegal Adsiser:
Mr. R. A Counihat
& Hon. Life Members:
Mr. E. B. Haddon
Mdr. H. B. Thomas, o.a.E.
Porofamo A. W. Williams c.B.E.
Rlevered Father J. P. Orazzolara
Captain C. R. S. Pitman, ca.B.E- o
The Hon. Secreaory
-" The Hon. Trewaurr
R The Hon. Editor
The H~on. Librarian
Psafesso B. J. K. Boake, o.er.
Dr. W. B. Banage
Mr. J. E. ComptoR
Mr. S. E. Eteari
~Mener Cooper Bmrs. & Co.
Honl. Vice-Presidents c
S~:ii) E~dWard Musesa~, 8.8.n.
ai To Winyi Gafbursa c.es.a.
Gk ~john Miler Grray
Mr~L &W. Kulubya, c.s.n.
Sir A. R. 000, C.M.o., Or.B..
Mr..E,J Wayland, c..ek
i!*:H; Ii. H~unte, cas.n., .La.
liDr: H.~Jewitt, a.m.o.
F'W H;. R. Hoe, n.a.u.o., L.nOa,
:'I 'c, g,.a.
MI(r. J. Sykes, O.Bm.E
) M. V.Brasnett
Ca:qptainr C. R. S. Pitman, c.a.E.,
td. S9. W. Kulubya, c.ar.E
rMdl E. A. Ttmple Perkins
M3r. L A. Snozall
St. & A. Davies, cum.o., O.Bs..
G. H. E. Hopkins, os.a~.
&rdkK M. Trowell, u.B.s.
LDr. ~w, J. ~Eggeing
1948--50 Dr. G. ap Griffith
1956-51 Rev. Fr. F. B. Garey, w.F
1951-52 Profear A. W. Wiliams, a
1952-53 Sir J. N. Hutchinson, c.r.a
1953-54 Mr. J. D. Jameson, o.e.a.
1959-55 Dr. Audrey I. Reichards, 0,
1955-56 Rev. Dr. H. C. Trowell, o.l
1956-57 MrL. D. & Murphatia, as.e]
1957-58 Mr. M. Barrington Ward
1958-59 Dr. H. F. Marria
195!Mi0 Profe~ser A. W. Southall
1968-61 Mr. J. C. D. Lawrnmce, o.e
1961-62 Mr. B. E. R. Kirwan, u.ars
1962-64 Mr. W. S. Kajubi
1964-65 D~r. M. Pomnansky
1965-6 Dr. W. B. Banage
196f 17 Professor 5. J. K. Baker, o.l
167-68 Dr. M. S. M. Kiwanuka
Mr. B. K. Mlulyanti, o.a9.s
Smcrta: Mrs. J. Bevin
Rev. K. I
EFREkcellency the President ofUganda, Dr. A. Milton Obote.
L DinonMr. P. 2
THE UGANDA SOCIETY
Copyright Uganda Society 1968
THE JOURNAL OF THE UGANDA SOCIETY
K;IKUKULE: GUARDIAN OF SOUTHEAST BUNYORo E. C. PLANNING 119
THE LOWER MIOCENE FOSSIL SITE OF BUKWA, SEBEI A/. WALKER 149
SOME PLANT FOSSILS FROM BUKWA - A. HAMILTON 157
THE EXCAVATION OF AN ANKOLE CAPITAL SITE
AT BWEYORERE - - - - M. PoSNANSKY 165
THE ORIGIN OF LAKE NYABIHOKO, KAJARA
COUNTY, ANKOLE- - - - B. H. MoTTRAM 183
THE LEGEND OF LAKE NYABIHOKo - J. KARISA AND
E. M. KAHWTTE 187
THE DIARIES OF EumI PASHA-ExTRACTs XIV SIR JOHN GRAY 189
REDEMPTION IN GANDA TRADITIONAL BELIEF -A. M. LUGIRA 199
THE DISTRIBUTION OF CERTAIN COLONIAL WEAVER
BIRD SPECIES IN UGANDA - - - J. R. HALL 205
GORDON's EXPEDITION IN SEARCH OF
NYAMUYONJO - - - - SIR JOHN GRAY 213
GORDON's SAILING BOATS IN EQLUATORIA - H. B. THOMAS 215
MACDONALD S MANUSCRIPT HISTORY OF THE EVENTS
or 1897-1899 - - - - A. T. MATSON 217
EARLY CHRISTIAN Activrry IN NORTH UGANDA
AND THE Sourn SUDAN - E. B. HADDON AND H. B. THOMAS 220
SPINES ON TEE NECK OF THE POrro - - A. WALKER 221
LORD TWINING OF TANGANYIKA AND GODALMING J. SYKEs 223
TERENCE PATRICK O'BRIEN - - - M. POSNANSKY 224
GEORGE CHARLEWOOD TURNER - - S. J. K.BAKER 225
Physiognomie, Struktur und Funktion--Gross K~ampalas
(By K. Vorlaufer) - - - - H. CAROL 227
The Royal Capital of Buganda (By P. C. W. Gutkind) F. B. WELBOURN 229
Agriculture in the Congo Basin (By M. P. Miracle) A. R. DUNBAR 230
Tams (By D.G. Coursey) - - - M. POSNANSKv 231
UGANDA BIBLIOGRAPHY 1967-1968
Compiled by B. W. LANGLANDs 233
INDEX TO VOLUME 32 (1968) - - - - - 255
THE UGANDA SOCIETY
Price Shs. 15/- (Uganda)
Uganida journal, 32, 2 (1968), 119--147
KIKUKULE: GUARDIAN OF SOUTHEAST BUNYORO
By E. C. PLANNING
The name of Kikukule,' who guarded the southeastern borders of the
Kingdom of Bunyoro-Kitara for his master Omukama Kabarega, had for long
been held in awe by the fighting men of Buganda. His village, which bore his
name, was situated close to the borders of the Singo and Buwekula districts of
Buganda.s There he lived in state with a large entourage of courtiers, pages and
warriors, athwart the main route giving access from the south to the rich lands of
Bunyoro. N~o one could pass through his domains without paying tribute in
ivory or trade goods. Anyone unfortunate enough to incur his ill-will, soon met
his fate. Following the custom, his war parties burned, looted and carried off
women and children from the nearby districts of Buganda and further afield.
It was not until the end of 1892, despite earlier brushes with British encroach-
ment, that Kikukule was temporarily crushed in the first stage of the British
attempt to bring Kabarega to heel. Albeit, he evaded capture on this and other
occasions, invariably to turn up again only to raid and spread more alarm. In
the words of Major Owen, "a hard nut to crack," it was not until 1895 that he
surrendered to the British at Masindi. Later he was conditionally released but,
with the turn of events, he reverted once more to the role of guerilla fighter.
A ruthless and independent character it was not for him to give way easily
to the white invaders whose presence in his country of adoption upset his way
of life. From I886 to 1899 he continued to serve the cause of the fugitive Kagarega,
raiding into Buganda and being very much to the fore in operations in British-
Occupied Bunyoro. His last major success, together with his friend, the chief
Ireta, was the destruction of the fort at Hoima towards the end of 1898. Finally,
with the capture of Kabarega in 1899, all rebel opposition collapsed and, with
other chiefs, he emerged from the safety of the forests to surrender once and for
all. His last days were spent in detention in Kampala where he died towards
the end of 1899, aged perhaps between 55 and 60.3
Bugangaizi, over which Kikukule's authority extended, covered an area
from the Kafu River southwards as far as the Muzizi River, an area nowadays,
approximately equivalent to the Bunyoro counties of Bugangaizi and part of
Buyaga. "Kikakure's," as his village or kyalo was called, stood in Bugangaizi
in southern Bunyoro, on the caravan road which, starting from Fort Raymond
(Mukwenda's-Mityana) in Singo, led deep into Bunyoro-Kitara.4 A neighbour
of his was the powerful Munyoro Ireta, son of Byangombe, chief of Mboga
district which extended into present-day Toro, who dwelt at Nsororo.
Although Kikukule paid allegiance to Omukama Kabarega he was, in
fact, a Muganda of the elephant clan by birth. He was better known as Kikukule
kya Lunego, his home had been in Kyadondo. His father, Kyesereka, had run
foul of the law and so had moved with his family to Kitente in Buwekula, close
to the frontier with Bunyoro. At that time (circa 1850) Ruhonko, the Munyoro
chief of Bugangaizi district of southeast Bunyoro, responsible to Omukama
Kamurasi, had his compound only a few miles further north, close to the site
of the modern township of Kakumiro.
120 UGANDA JOURNAL
Little is known of Kikukule's early life. For some reason, perhaps, because
his father was voluntarily in semi-exile and there was the flexibility of a none
too static border, he became a page in the service of the Munyoro chief, Ruhonko.
As he grew up, it is said he won recognition as a warrior, at first taking part in
Ruhonko's forays coping mainly with the Bunyoro internicine fighting of the
period, rather than with raids on the Baganda. Subsequently his aggressiveness
took him further afield. During these days of apprenticeship to Ruhonko the
first contacts had been made by Europeans with Omukama Kamurasi (Speke
and Grant direct from hostile -Mengo in 1862; the Bakers from the north in
1864). These meetings had left the Banyoro shrouded in suspicions, a distrust
which was to grow a few years later after the return of Baker in 1872, and
following Gordon's erection of stockaded posts in the north of the kingdom in
the attempt to check the slave trade.
Kikukule's reputation was sufficient for Ruhonko to bring him to the
attention of the new Mukama Kabarega who, after the fatricidal strife follow-
ing on the death of Kamurasi in 1869, had assumed the throne of Bunyoro-
Kitara. Furthermore he gained personal favour with the new ruler. He had
won the confidence of one of the contenders to the throne who had found
support for his cause in part of southern Bunyoro and then, at an opportune
moment, had betrayed the princeling. Eventually, after Ruhonko's death,
Kabarega appointed Kikukule in his place as chief of Bugangaizi.s So he took
over Ruhoko's compound, setting up his own huts near to Kasaka Hill where,
today, stands the White Father's Mission, Notre Dame de la Garde, Bukumi.
The rivalry which existed between Bunyoro and Buganda brought Baganda
war parties into Kikukule's sphere of influence and in time he became a leader
to be feared and best avoided. The spasmodic raiding over the years by Baganda
into Bunyoro, and Banyoro into Buganda followed a similar pattern of looting,
destruction and taking of captives, preferably where opposition was weakest.
War parties seldom stayed for long in either territory, at the most a month.
When the alarm was over grass huts could be easily rebuilt and in a short
period crops grew again in the fertile soil.
It may be assumed Kikukule participated in the BanyorojBaganda border
forays of the early 70s. Later (circa 1889) he may well have played his part in
the battle fought in Bugangaizi near his own village when Kabarega's eldest
son, Jasi, defeated an invading force.6 But mention is first made of his activities
when soon after 1873 Kabarega's warriors invaded the disputed region of Toro.
Subsequently, following the death in exile of Nyaika, ruler of Toro, Kikukrule
set forth on the orders of his Omukama on a major raid. This took him with his
men for nearly four months plundering and ravaging the greater part of the
country. Eventually, the Toro chiefs agreed to hand over Nyaika's children to
the Banyoro, and so Kikukule withdrew with his captives.7
Back in his own district to pursue his normal day to day life, collecting
tribute with an occasional raid into Singo or even back into Toro to replenish
stocks of cattle, slaves and women, little of importance occurred. His land
bordered with the sparsely populated Buganda district of Buwekula. If not
openly, he was, however, on good terms with the Muganda district chief, the
Luwekula, Sepiriya (Cyprien) Mutagwanya, who lived only miles away close
to the northeastern flank of the nine mile range of hills, nowadays known as
Relationships between Bunyoro-Kitara and the few Europeans who had
by now entered the kingdom had never been of the best, excepting perhaps in
the case of Emin Pasha.s If anything, the Banyoro through suspicion of the
newcomers, were hostile. No doubt Kikukule was as well informed of current
events as others, particularly the alliance which had been developing in the
kingdom of Buganda with the white newcomers.
Events were taking shape in Buganda. Before the advent of white men,
Arabs had brought Islam to Buganda, and a number of chiefs had turned
Mohammedan. Following the first contacts with European explorers both
Anglican and French Roman Catholic missionaries had arrived at the court
of Kabaka Mutesa at Rubaga. Friction was to follow. After the death of Mutesa
in October 1884, who was succeeded by his son Mwanga, Arab influence
increased to the extent that by 1888 Islam was to the fore throughout that
kingdom. However, the position changed in the following year when Christian
Baganda forced Baganda Mohammedans to flee into neighboring Bunyoro.
There, aided by Kabarega's men, the Mohammedans stood their ground and
inflicted a defeat on the pursuing Christian. Then, in April 1890, the representa-
tive of the Imperial British East Africa Company, Mr. F. J. Jackson, arrived at
Kabaka Mwanga's capital, Mengo. Towards the end of the year he was followed
by Captain F. D. Lugard.
The first major action Lugard took was to lead a force, which in addition
to a nucleus of Sudanese and Zanzibaris, was made up of adherents to the
Christian parties in Buganda, in order to remove the threat of the rebel Moham-
medans still sheltering in Bunyoro. The Mohammedans, reinforced by 1,300
Unyoro guns (probably a number were Kikukule's men), were in position
in Bugangaizi behind the Kyangora River, about seven miles north of
Kikukule's compound. Here, on 7 May 1891, they were defeated by Lugard.
In this battle a son of Kikukule's, Kasaijja, was said have been killed.s
Lugard's next step was to set out in June to contact Emin Pasha's former
Sudanese soldiery who had collected at the south end of Lake Albert. He marched
to the Banyoro-dominated province of Toro by way of Ankole and Katwe,
taking with him the Mutoro prince Kasagama who had been living in Buganda
in exile. After setting up a stockaded post, which he called Fort George by Lake
Edward, he moved into Toro on 8 August 1891.'o The only serious opposition
he met from the Banyoro was at a pass between the lake and the Ruwenzori
mountains (known to the Baganda as Gambalagala) where a force barred his
way. This was not an army sent by Kabarega, but an assemblage of all the chiefs
of southern Bunyoro led by Ireta, who were reported to be seven in unmber.
Rukara, chief of Busongora was no doubt one of them, and probably Kikukule
as well. In the event the wary chiefs and their followers were dispersed. Lugard
wrote, "I advanced with the Soudanese in line and with the porters with their
loads behind. The hills in front were covered with the enemy, who were in
masses like ants. I should suppose there were some 4,000 to 5, 000. They opened
fire on us and I did not reply for some time till my men in the rear should have
time to come up. I then put some shots into them with the Maxim at 800 yards
(with much effect, as we afterwards found). They fired a very great deal of
ammunition, but retired as we advanced, and we captured their camp (with
some food) and camped on a hill beyond. The Soudanese pursued the enemy
some distance. We had fired in all 316 rounds, including 40 from the Maxim,
and had no casualties on our side. The enemy never made any stand whatever."
His next halt became the site for Fort Edward. From then onwards the advance
was easy and Kasagama was installed as Omukama of Toro. Later, after contact-
ing the Sudanese contingent under Selim Bey at Kavalli's, Lugard brought
the large party back into Toro. The Banyoro continued to show their resent-
122 UGANDA JOURNAL
ment at this intrusion, although with no marked effect. In one encounter in
October, Ireta with a well-armed band, was driven off after opening fire on
Lugard's men. In order to deter renewed attempts by the Banyoro to harass Toro,
Lugard next set up a chain of isolated posts for the Sudanese to garrison. These
were completed by the end of December 1891 and placed under the overall
supervision of Fenwick de Winton."'
By 1893 the British were not only fully installed in Mengo but had steadily
extended their influence outwards from the borders of Buganda into Toro and
Busoga.'2 Not to be intimidated, Kabarega planned to raid into both these
regions. He had to act to relieve the pressure which had been building up against
his kingdom and also to curb the regular practice by the Sudanese garrisoned in
Toro of violent raiding and plundering for food over hundreds of square miles
of southern Bunyoro. Such reprisals by the Banyoro were therefore not without
justification.'" The Banyoro chiefs stepped up their attacks on the Toro posts.
One attack in May 1893 was led by Kikukule on a post at the southern end of
Lake Albert, Misozi Mwenge, when the Sudanese officer in charge, Bimbashi
Sukri (Shukni) Effendi, was killed.l4 By mid 1893 the political situation necessi-
tated the withdrawal of the Sudanese garrisons leaving the Toro border posts
abandoned. The Banyoro did not miss the opportunity to storm into the country
once more, the new Mukama Kasagama, having to flee to the relative safety of
the slopes of the Ruwenzoris.'5 No doubt Kikukule played a leading part in this
invasion, for by now his reputation, particularly amongst the Baganda, as a
tough and treacherous fighter, was even more enhanced. To the British he was
obviously the first obstacle to be dealt with if, what seemed to be a continuous
threat by Bunyoro, was to be overcome.
Captain J. R. L. Macdonald, then in command of the military forces in
Uganda, made preparations for an expedition into Bunyoro. He had noted
that ". . .. .the province of Bugengezi, ruled by the powerful chief Chikakure,
was fertile and contained numerous inhabitants who had assisted and encouraged
the Mohammedans. From Chikakure's headquarters two main roads ran into
the interior ofUnyoro. One of these led northwest to Kijambi, where it bifurcated,
the left branch leading to Lake Albert the right hand one to Kabarega's capital."l6
A prerequisite for an attack on Kabarega was the need to have outposts in
the neighboring Buganda district of Singo. Major E. R. Owen was specially
detailed for this task. As the north of Singo suffered continuously from interference
by Kikukule whose compound lay within a few miles of the border, this move
would either bring him to heel or force a clash. Since Kikukule had played a
prominent part in the harassing attacks on the Toro forts, Macdonald authorised
Owen not only to erect outposts but to punish the chief as well." Owen was
well aware of his adversary's strength for on 16 August 1893 he noted in his
diary, "Ciccoculi, the Chief on Kabarega's frontier . .. .", who he anticipated
he might well have to fight, "was reported to possess 2,000 guns."'"
On 6 September Owen, accompanied by the Baganda chiefs Mukwenda
and Luwekula (Mutagwanya),19 went into northern Singo to select sites
for at least two forts. From his talks with the two chiefs Owen expected his
first confrontation with Kikukule to be over the provisions of supplies should
an advance post be sited adjacent to the Bunyoro border. In fact Mutagwanya,
who now openly liaised with Kikukule, gave fair warning that Kikukule might
not prove amenable to any requests for good.
For the next week the party moved along the Singo-Bunyoro border
examining likely sites. With Owen now close at hand Kikukule decided to
parley and play for time. He let it be known he was willing to come to terms but
pleaded that he could do nothing without consulting his master, Kabarega,
first. But it was rumoured he had already sent word to his overlord saying he
had driven off Owen's force and was now preparing to attack Omukama Kasa-
gama in Toro. His messengers preferred friendship and peace and he also sent
Owen a present of a black bull. But these were promises offered only to be
After visiting many possible sites Owen and his party came to Buwulu Hill
(part of the 5,000 feet Butologo plateau) on 14 September; Owen wrote, "Open
hills, fine grazing. This position looks bang on to Unyoro and its population.
The Kitongoro River running at the foot of this mountain is the boundary
between Uganda and Unyoro, and easily crossed. An advanced hill joining the
big one would be just large enough for an unassailable fort holding fifty.
.. .. Many of the people are Wanyoro here. From here Ciccoculi will be at
our mercy when necessary." Later in October, after siting another outpost
further north, Fort Lugard, Owen was back again at Buwulu Hill having, in
the meantime, decided to call this new outpost Fort Grant.20 Work was started
on the new fort on 12 October by marking out its lines on the ground. The
Sudanese troops and porters were kept busy for the next few days completing
ramparts, a store and houses. Now Owen heard that Kikukule had no intention
of supplying him with food. The Luwekula came to the Fort but was unhelpful
over this urgent need for supplies. Owen agreed to receive a messenger from
Kikukule who proved to be unco-operative. There was now abundant evidence
that any Munyoro or Muganda caught supplying the garrison was killed off
by Kikukule's men. Soon his passive play-acting changed to open hostility.21
As the attitude of the Banyoro deteriorated Owen sent messengers to warn
Macdonald. In November he decided to return to Entebbe where he learned
of the arrival, earlier in the month, of Colonel H. Colvile who had come to take
over the command of the forces in Uganda.22 Summing up the situation Colvile
decided to continue with Macdonald's plan to attack Bunyoro in order to
bring Kabarega to terms. Alerted by Owen's reports he noted, "the evidently
hostile attitude of Kabarega, as shown by the conduct of his chief Kikukuri,
who certainly would not have acted without orders, forced me to turn my atten-
tion to questions of defence."23
If Bunyoro was to be quickly invaded it was clear enough that first the
redoubtable Kikukule must either co-operate with the British or be cleared
from southern Bunyoro since the only road which gave easy access to the heart
of the country lay through his district. At a meeting in Kampala with Colvile,
Owen discussed the situation and it was settled that he should return to Singo
as soon as possible and, by making a night march, try to surprise Kikukule at
The moment of reckoning had come. On 19 November, Owen
accompanied by the traveller Lionel Decl6, left Kampala on a five days' march
to the district of the chief Murumma. At Fort Raymond the party was joined
by Captain C. H. Villiers. Owen intended to attack Kikukule on the night of
27 November. Decl6 has given a full account of this attack.24 "If the native
reports were correct we might look forward to a certain resistance as he [Kiku-
kule] was said to have some thousands of men armed with breech-loaders.
To avoid his getting aware of our intentions, we pushed on about 19 miles on
the 25th .. ..As we drew nearer to the frontier of Unyoro the country grew
worse and worse .. .. It rained hard all day, and we reached the village of
Kuruma, whence we were to start the night against Chiccaculi .. .. The
chief of the village seemed to be in a thoroughly bad temper. Major Owen who
124 UGANDA JOURNAL
UNYOto AND UBAND- PioD tut-1AAA
11C O I LE
was very anxious to keep the object of the expedition secret until the last moment,
interrogated him directly about Chiccuculi, with the following result. Our
enemy, it seemed, was one of the most important if not quite the most important,
of Kabarega's subordinate chiefs; he was the guardian of the frontier, and
everyone who wishes to visit Kabarega had to pass through his country. If a
caravan wanted to buy ivory it had to go to Chiccuculi, and he arranged the
business on behalf if his master. He was said to possess over 2,000 guns and his
village was said to be about four hours march from Kuruma. Our chief had
previously expressed to Major Owen his willingness to show him the road if
ever he meant to attack Chiccaculi, so we reckoned to take him as a guide.
Our force consisted of 200 Soudanese armed with Remingtons, with 100 rounds
per man as well as 100 porters with Snider carbines and twenty rounds; also
the Maxim in whose working I had not very great confidence. Villiers was in
charge of the machine and I was to assist him. To complicate matters, at eight
in the evening Villiers was seized with a violent attack of fever.
We started at ten o'clock that evening guided by the chief of Kuruma, who
had promised to take us to Chiccaculi's in four hours. The profoundest silence
was observed. We marched all night through a rolling country without a single halt
we had to wade through no less than seven swamps. The grass was very long, and
we knew nothing of the enemy's movements. Soon after daybreak we came in
sight of a large straggling village, the thatched roofs peeping out from the high
grass and shrubs, with Chiccaculi's kraal standing out in the centre. The natives
opened fire upon us instantly; first spreading out in a line, and then forming up
in groups, sheltered by the tall grass. A good deal of desultory firing followed.
Owing to nature of the ground, the natives were able to rush up to within a few
yards and open fire, and then rush back again. They do not seem to have any
great confidence in their weapons; and I am bound to say that the character of
their shooting was not calculated to inspire it. However, our position was not a
perfectly comfortable one. We were faced, so we afterwards learnt, by an army of
some 3,000 natives whom we could not see." Decl6 then describes a charge upon
Kikukule's in which seventy people including six chiefs and Kikukule's son
So the redoubtable Kikukule, with his following of Banyoro, some Moham-
medan Baganda and Sudanese slaves, had been routed. His losses were probably
considerably more than those noted by Decl6 since owing to the thickness of
the grass it was impossible to form an accurate estimate.25 Owen's casualties
were one killed and two wounded. In seven days Owen had brought his small
force over 110 miles of difficult country to the Bunyoro border. After a hard
night march the enemy had been surprised and scattered; now he was undis-
puted master. On 29 November he led his men back to Fort Grant.
According to the expected pattern of fighting to which both Banyoro and
Baganda were accustomed Kikukule and his followers, in the face of reverse,
had dispersed to lie low beyond the Kafu River until the attackers had passed on.
The effect of this victory had been decisive, and by leaving this large part of
southern Bunyoro almost clear of organised opposition, Colvile's plans were
greatly facilitated. Kabarega's intentions to attack Toro and Busoga were
ruined and so he retreated to his capital at Mparo.
Colvile noted that: "This victory was a geat surprise to the Waganda who,
knowing Kikukule to have a very large force at his disposal, had confidently
expected Owen's defeat. That 200 Sudanese should have defeated so many
thousand Wanyoro was looked upon as a marvellous performance, and added
greatly to their Drestige, and on the strength of this I convened a great meeting of
126 UGANDA JOURNAL
chiefs to announce the declaration of war. The meeting took place on the 4th
of December and having read Owen's report I explained my position with
regard to Uganda and Unyoro, and telling the chiefs that I intended to attack
Kabarega called upon the king to assemble his army to co-operate with me. I
have frequently been told that this co-operation was the last thing he wanted;
but be that as it may, he made no objections and orders were at once issued for
the war-drum to be sounded throughout the land."
One final action was fought on 19 December when a force led by another
of Kikukule's offspring was scattered in the neighbourhood of the destroyed
compound by a party of seventy Sudanese troops commanded by Bimbashi
Faraj Effendi. The Banyoro suffered some thirty killed.26
In the meantime war had been declared by the British against the Kingdom
of Bunyoro-Kitara. Colvile sent a notice to Kabarega in the following terms:
"When we heard that our Soudanese troops in the forts in Toru were oppressing
the natives we sent Europeans to stop this, but your people more than once
attacked us and killed one of our Soudanese officers, a big man (Shukni Effendi).
For this we have received no compensation. We have endured hostility and
insult from your people for many months because we wished to give you a
chance of showing whether you wished to be friendly or not. The time has now
come for a settlement. I have already ordered the punishment of one of your
chiefs (Ciccoculi) who has killed our friends, and have already prepared my
armies to invade Unyoro, and within twenty days from this shall be on your
frontier. So be prepared to receive the punishment which I have in my mercy so
Little time had been lost in finalising preparations for the invasion. Since
November additional British officers had been seconded for service in Uganda.
With a contingent of his Sudanese Owen had returned to Fort Raymond near
Mityana where he met up with the headquarter party of Colvile's force. On
Christmas day 1893 the party of eight Europeans and 221 Sudanese with 408
others, including followers, advanced to Ntuti on the Bunyoro frontier where
the Baganda, with their commander Semei Kakungulu, were already waiting.
This army consisted mainly of spearmen, plus about 5,000 men armed with
guns. Captain A.B. Thruston, who had recently arrived in Uganda wrote, next
day (26 December), "We were well in Unyoro and met the whole of the Waganda
army who were in camp and waiting for us. They numbered from 15,000 to
20,000 men and their camp consisted of about 5,000 bee-hive huts. The country
was devastated for miles around; banana plantations cut down, sweet potato
gardens torn up and houses burnt. The stench of the camp was vile, and the
state of filth indescribable. We tried to find a tolerable clean spot of ground for
our camp, but without success; so a party was told off for the unsavoury task of
cleaning one, and we then camped."'"
Organised opposition having disintegrated since Owen's rout of Kikukule,
the force reached the banks of the Kafu River, a thousand yards wide and choked
with reeds, within four days. Thruston and Owen reconnoitred the banks for
suitable crossings, when Owen came in contact with the rear-guard of a dispirted
Bunyroro force returning from Toro. Eventually a suitable place was found.
A causeway of cut reeds, banana stems, papyrus and grass having been cons-
tructed, the force crossed into northern Bunyoro on 31 December, 1893.
Kabarega's capital at Mparo feel to the invaders on 2 January, 1894. As
adept at evading capture as his lieutenant Kikukule, Kabarega had taken
refuge where he had sheltered before from the Baganda, in the depths of the
Budongo forest. The campaign developed into isolated skirmishes with no
Kikukule had remained for some time in hiding along the banks of the
Kafu River. But with the suspension of active operations against Kabarega in
February 1894~ and the subsequent withdrawal of the Sudanese garrison from
Fort Grant, there was after all nothing to prevent him returning to the south to
reassert his authority oncer more. Although Kabarega remained free, his power
had been broken. The way was now open for British influence to reach into the
domains of Bunyoro and beyond.
Back in Mengo it was agreed that for supporting the British in the campaign,
the Baganda should be awarded the whole of the recently cleared territory
south of the Kafu River. Because of the rift between supporters of the Roman
Catholic and Protestant mission in Buganda, which had turned to civil war two
years previously, steps were to be taken to divide this area between the two
religious parties. Eventually in April an allocation of administrative spheres
was made at Mengo by Kabaka Mwanga, when Bugangaizi and Buyaga were
allocated to the Roman Catholics. Baganda chiefs, supporting the Catholic
party would therefore administer the area. The time now approached for the
Baganda to occupy the territory.
The White Fathers' Mission (the Society of Missionaries of Africa) was
already established in Buddu. The opportunity to enter southeastern Bunyoro,
now free of the determined opposition of Kabarega and his chiefs, attracted the
Mission's attention. It was proposed to make early contact with the inhabitants,
particularly in order to endeavour to prevent a bloodless occupation by the
Father Achte, who was then 32 years old and had been in Buganda for the
past three years, was asked by his superiors to seek out and meet those Banyoro
leaders who retained some authority, before the arrival of any occupying force.'"
But back in Bugangaizi, Kikukule had returned to his burnt-out base where, in
a short space of time, his huts were rebuilt and his authority re-established.
It turned out, therefore, that Father Achte's task would be to establish contact
Early in April 1894so Achte left Bikira Mission (Kyotera, Buddu) going
first to the mission of Villa-Maria (also in Buddu and, at that time, the head-
quarters of the White Fathers). There he was joined by Father Houssin. A few
days later the two Fathers, with some followers, set forth on a journey of about
eighty miles to southern Bunyoro, not only to attempt to pave the way for a
peaceful occupation but also, conditions permitting, to open a mission station
in the newly conquered region.
After fording the Katonga River the party marched through unpopulated
land and on into Buwekula. Roundabout 20 April they arrived at Kawerist at
the foot of Mubende Hill where the Luwekula, Sepiriya Mutagwanya, resided.
From him they learned that Kikukule was in a defiant mood. The belligerent
chief was reported to have said, "If the Baganda want my district then they can
come and take it. I have sound guns and there are impregnable caves." It looked
to the missionaries as if the way into southern Bunyoro would be barred. Un-
daunted, Achte was termined to contact Kikukule before the arrival of the
Baganda in force. He sent messengers ahead offering friendship whilst explain-
ing that Father Houssin and he were coming to Bunyoro as ministers of charity
and peace. To his delight the first messenger was well received. Subsequent
envoys were equally welcome. With these envoys was one Patrisi who, on his
first visit recalls two barkcloths being taken as gifts for the Chief, whom he saw
at the rebuilt headquarters from where Owen had driven him only four and a
half months previously. At this time Patrisi and his party returned with the
present of a hide for Father Achte. On a second occasion they brought back a
gift of an elephant tusk.32
As a result of these exchanges of messages and gifts a meeting was agreed
between Kikukule and Achte on the understanding that the Father would be
accompanied by no more than three other persons. There is no record of the
names of the followers who accompanied Achte on what, in the circumstances,
must have been a most precarious mission. The meeting, he related three
months later in a letter to Mgr. Livinahac, took place one night on a hillock
amidst large rocks. He was confronted by Banyoro carrying guns as well as
blazing torches which lit up the immediate surrounds for the chief and his
lieutenants. Little wonder Achte felt some doubts about the security of the
situation. "A la vue de cet appareil, je me demandai un instant si je n'avais
pas pouss6 la confiance jusqu'd la t~mbrite, et si on n'allait pas me faire un
mauvais parti. Je me recommendai intdrieurement & Dieu et g mon ange
garden, et j'allai m'asseoir en face du chef". The man facing him, though well
advanced in years, was tall and regal in his bearing.
After an initially cool and embarrassing exchange of words, the conver-
sation warmed. Achte boldly explained his presence and what he wanted. The
old chief, surely to the concealed surprised of the Father, acquiesced. A pact
was agreed. On Kikukule's side, he was willing to receive missionaries in his
district, and his people would be free to take instruction in the new religion.
As for Achte, he promised the missionaries would do everything possible to prevent
the occupation following the customary course during raids of burning, pillag-
ing and rape whensoever the Baganda should arrive to take over the region.
The contract was then sealed in the accepted fashion by a ceremony of blood-
brotherhood. With a large knife each made a small incision below the chest. A
coffee berry was smeared on each cut. Kikukule was first to swallow a berry
smeared with the Frenchman's blood; then Achte, in turn swallowed a berry
dipped in the blood of the chief.33 Having reached agreement on the immediate
essential points Achte went further and invited Kikukule to consider surrender-
mng mn person to the British authorities at Mengo.
After spending the rest of the night in the bush Achte retraced his steps
next morning into Buwekula and rejoined Houssin. On 8 May they started out
together for Kikukule's district. Now, with the permission of Kikukule, a camp
was set up on a hill a mile and a half way from the chief's reed-fenced head-
quarters. No time was lost in settling in, and already on 13 May Achte recorded
th~at fifteen young people had come to him asking for religious instruction.
In the time available the Fathers worked hard on Kikukule. Only after
some hesitation did he abide by their counseling agreeing to recognize his acquain-
tance the Luwekula, Mutagwanya, who was a Catholic, as overlord. And then
he finally agreed to inform the authorities at Mengo of his submission. As for
the occupation, Achte had extracted a formal promise from the Baganda leaders
to use all their powers to ensure their warriors spared the women and children.
In this Achte says, they kept their word. So the submission of the proud and
turbulent Kikukule was gained, for the time, being without an angry gesture or
short being fired.
At last a small occupying force of Baganda crossed the border but only for a
short distance. The first reaction was for the nearest Banyoro villages to empty.
As had always been the custom in time of invasion the women and children fled
with their cattle to well-prepared hide-outs in rock shelters and caves."' However,
in due course, with the Baganda heeding the orders of their leaders, people
trickled back and normal village life was soon resumed.
In anticipation of a formal take-over of the whole district Mutagwanya
established a headquarters in Bugangaizi a few miles from Kikakule's which,
like his base at Kaweri, came to be known as Luwekula's.35
The site Kikukule allotted to Achte covered several acres, having three
springs as well as four banana plantations. Here, provided with food by their
host, the Fathers lived under canvas until later, when a hut, a reed chapel and
a store were constructed.36 The Baganda occupation force stayed put a few
miles away. Kikukule remained at his compound, it would seem not altogether
in the best of tempers. Achte has described him at that time, as an astute and
violent man, lacking any scruples. His authority still extended over a wide area.
At his fenced-in compound there were a hundred pages to attend to his needs.
Every night three hundred warriors were on duty within the reed palisade
surrounding his huts. Wherever he went an entourage of advisers, sub-chiefs
and servants accompanied him. There was an exception to this when he visited
the Fathers at their nearby camp. Then he came alone, leaving his arms behind
on the far side of a small stream which had to be crossed.37 Patrisi and another
of Achte's servants called Kyobe, have spoken of him as a powerfully built aging
man, having a hairy chest and hairy outsize calves, which earned him his
nickname N~fundo Byoya by which he is still well remembered by Baganda and
By the end of the month, the Fathers had taken full possession of the site.
The small chapel was given the name of Notre Dame de la Garde, in the verna-
cular Bikira wa Boukoumi. After completing their own reed and grass buildings
they helped out Mutagwanya in erecting huts for his new headquarters
closer to the border.
With the Baganda quiescent and Kikukule playing a peaceful and co-opera-
tive role the Fathers were able to turn their attention to the religious needs of the
local inhabitants. By early June over a hundred young people were receiving
instruction. Achte related an amusing tale at the expense of the Chief. When
some fifteen of his pages had made good progress in learning their prayers,
Achte decided to send a catechist to their quarters within Kikukule's compound,
to take them in morning and evening prayer. The pages were told that on the
first day they would be woken by the beat of a drum when they should assemble
quickly for the serve. Next morning a drum beat loudly and at once the group of
eager teenagers ran to the meeting place. To Kikukule the throb of a drum at this
early hour meant only one thing, an alarm. With the memory of Owen's attack
still fresh in his mind he was quick to seize his gun and call his men to arms. Only
when he saw his pages at prayer was it clear there was no cause for alarm.
Achte reported that as far as Kikukule was concerned he stuck to his pagan
beliefs but kept his word in leaving his people free to approach the missionaries
if they wished.""
The British Government assumed a protectorate over Buganda on 19
June 1894, when the two Bunyoro districts of Buyaga and Bugangaizi were
officially handed to Buganda.
In the north, Kabarega evaded any confrontation with the small British
forces operating in his land. It was known he had castigated Kikukule for being
on such peaceful terms with the invaders in southern Bunyoro,40 whilst elsewhere
the Banyoro remained hostile. In northern Banyoro Thruston, after a number of
skirmishes, had erected a fort at Hoima in February 1894.0L Such was the aggre-
ssiveness of the rebels met with by his patrols, that they had to shoot on sight
anyone seen carrying a gun. In August Kikukule, in order to demonstrate his
130 UGANDA JOURNAL
continued loyalty to his elusive Mukama, (at a time when he had fifty of Kaba-
rega's women in his charge), had sent forty of his men armed with guns to assist
his friend Ireta, as well as Kabarega's son Jasi, the Katikiro Rwabudongo and
the chief Byabachwezi in a strong attack on the fort at Hoima. This hard-
fought engagement showed the Banyoro to be more formidable adversaries
than had been previously supposed. Thruston repulsed the attack for the loss
of eight men, but the attackers' losses were over two hundred.42 The situation
for the occupying forces continued to be hazardous. The monthly supply cara-
vans from Buganda had to fight their way through the enemy, and frequently
Sudanese patrols supported by Baganda and friendly Banyoro were cut off
and killed; then in retaliation villages and plantations near the scene of action
In southern Bunyoro Kikukule played his cards well. Though his wings had
been clipped, he was little worse off than he had been before he had set fire to
his compound in the face of Owen's sudden attack. With the missionaries
settled on his land he was comparatively safe from outside interference; safe in
the knowledge of the restraint which the Fathers exercised on the Baganda
warriors, whilst his long-standing association with Mutagwanya had barely
altered. He still retained power and lived as comfortably as ever with all the
trappings of a great man. The Baganda were loathe to advance. At the same
time Kikukule maintained contact with the Omukama, still counting himself as
his trusted representative in southern Bunyoro. There could be little doubt that
his formal verbal submission sent to the British at Mengo meant little or nothing to
The Baganda at Mengo had started pressing for him and his children as
well as some of his sub-chiefs, to surrender and so demonstrate once and for all
their recognition of Buganda authority over his territory.44 However, from the
Bunyoro point of view no one in his right sense would voluntarily give himself up
to the other side. That peace reigned in this corner of the kingdom during these
early months of the occupations must have been largely due to the wisdom and
patience of Father Achte. As time wore on the Baganda, realising full well the
tough character of the man who still stood in their way, had sent reinforcements
to Mutagwanya whose sole strength lay in steadily increasing the number of
fighting men at his disposal.
Finally matters came to a head at a meeting of both Baganda and Banyoro
parties held on the neutral ground of the mission in October, 1894. Now strongly
reinforced, Mutagwanya felt his presence could be spared whilst he visited
Mengo to report to his Kabaka as well as the British. Since Kikukule had un-
hesitatingly spurned the proposal first put to him by Achte in May to give him-
self up, Mutagwanya now put forward the suggestion for one of the Chief's
sons, together with four sub-chiefs, to come to Mengo with him. Their surrender
to the Baganda and the British authorities would demonstrate the sincerity of
the Banyoro in accepting the peaceful occupation of their country by his force.
The proposal was ill-chosen. Co-operative as he had been since May, Kikukule
had really never wavered from supporting the Mukama who remained free in
the forests of the north. His categorical refusal to have anything to do with the
idea was backed by the threat that he still had plenty of powder and guns at
his disposal. Soon the meeting broke up in disorder, to be followed by the beating
of war drums. From Achte's account the scene at the tiny temporary mission
must, to say the least, have been disturbing. Of the two opposing groups of men,
some three hundred carried guns and another four hundred were armed with
spears, everyone shrieking defiance at one another to the throbbing accompani-
ment of the drums. Grim as the situation must have been, Achte was equal to
the task of trying to restore order. Quickly he told Houssin to evacuate with
their most treasured belongings to take shelter in Buwekula. As arguments
and insults increased in intensity he mingled with Banyoro and Baganda calling
For calm and a return to reason. In the event the night passed without any
incidents, although much to Achte's consternation, everyone stayed near the
The next morning the two parties stirred dubiously. The lone Achte had
barely finished morning Mass in his chapel, when drumming started; then, both
sides refreshed from a few hours sleep, broke into war dances. This continued
throughout the day. In the late afternoon the Baganda, who easily outnumbered
the Banyoro, had at long last worked themselves into a frenzy. Although they
appeared intent on hurling themselves at Kikukule's men, Achte managed to
intervene. However, the climax of the situation had been reached, for seeing the
earnestness of the Baganda and knowing their numbers weighed heavily against
him, Kikukule and his men withdrew into his compound.
Next day, 17 October, Kikukule came on his own to the mission and later
visited the Baganda when he accepted their demands. In the main they wanted
to be free to advance into his district as far as the Kafu river in order to check
the continuous flow of arms to the Banyoro and to Kabarega's guerilla army. Now
Kikukule realized he could not hold out much longer; if he wished to retain
his freedom he would have to take a back seat. He took immediate steps to
evacuate his cattle, his women, his ivory and other loot. Then three days later
he slipped away with his closest followers, leaving a few trusted retainers behind.
They set fire to the deserted huts and soon the compound was ablaze. The
tough Kikukule had gone.
Any relief which might have been felt by Achte and his followers or amongst
the Baganda was short lived, for fortune did not altogether favour the recal-
citrant chief. He had gone north along the Mpongo River and soon started
raiding into Singo.45 But these activities were hampered by proximity to a
stockaded post at Kaduma. It was the height of the rainy season too, so the
movement of his cattle was impeded. Three weeks later, by which time Houssin
had returned to Bukumi, Kikukule reappeared again in the neighbourhood and
started to construct another compound.-
Restless and discontented, he re-opened negotiations with the Baganda who
had in the meantime started to move into the neighboring district of Buyaga.
A special envoy of the Kabaka had been sent to support the Luwekula. This man
was no stranger to Kikukule, so the master of intrigue set to work to convince
his friend of his fidelity and sincerity to the Baganda cause. This, so Achte
recorded, was backed with the gift of ivory and a Remington rifle. The real
enemies of the Baganda, Kikukule subtly explained, were the Catholic priests.
But his cajoling got him nowhere. He went to Achte too, but despite being
advised to accept defeat with dignity and perhaps retire to a minor chiefdom,
he was still unprepared to give way to the new order.
Now that the huts of his small but strong following had risen again, there
was every indication that the two forces must finally clash. Again the war drums
sounded and the Baganda, certain of their numerical superiority, prepared for
battle. There was no sign of Kikukule's men so, contrary to their normal tactics,
the Baganda preferring to wait for the enemy to attack first, they advanced on
the new compound. It was deserted. The old fox, having taunted his enemy to
the utmost, had again slipped away. However, this time, he lost some of his
rear-guard and baggage train to the pursuing warriors.
132 UGANDA JOURNAL
Kikukule had been dislodged. No longer had he a permanent base, so the
occupying force of Baganda ventured deeper into southern Bunyoro. Away from
Achte's influence they quickly reverted to the traditional role of the raider,
killing off the men, seizing the women, and looting generally. As for Kikukule
there were many miles of undulating rocky and forested country in which to
hide, and if needs be, operate from a primitive strong-point. Eventually word
came to the missionaries that he was encamped in the north, but was also opera-
tional. He laid all the blame for the success of the Baganda on the missionaries.
Once rid of them, he was reported to have said, the Baganda denied their support
would be quick to withdraw to their own lands.
All the same, lines of communication between Kikukule and the mission at
Bukumi were open and messengers moved, unmolested, to and fro. Normally
throughout Bunyoro the main villages were linked by roads kept in very good order,
whilst bridges made of trees were maintained across the swamps. On one occasion
Achte's young servant, Kyobe accompanied one of his warrior brothers to take
a message to Kikukule. The great man had little to say other than show his
displeasure at the trend of developments arising from the support given to the
Baganda by his blood-brother, Achte. He uttered the warning that soon he
would have to attack the mission. The messenger and his young brother were
given safe conduct back, and so the missionaries were forwarned of this change of
face." Soon enough the Fathers and the Baganda chiefs were subjected to
continuous ruses. An unsuccessful attempt was made to abduct Achte. Later a
large part of the Baganda force was enticed away from Lwekula's by a story that
only two days distant were a number of head of Kabarega's own herd of cattle.
As the warriors rushed off headlong to regale themselves with the prize, Kikukule
advanced towards Bukumi, but Achte, now well on his guard, was ready for just
such a happening, his followers being armed and prepared to fight. The attack
In the meantime Thruston had been busy in the north. A surprise attack
on Kabarega was almost successful, but the Mukama had slipped away losing
a large amount of ammunition and supplies, cattle and the royal insignia. He
crossed the Nile and took refuge in Lango.
Now, no doubt rueful at the turn of events, Kikukule accompanied by a
handful of sub-chiefs, returned to the mission on 30 November. He told Achte he
would give in if allotted a small area to retire to. Although no-one took this
seriously he was allotted a plot of land a mile away. It was soon detected that
all his old supporters were stealthily moving into villages in the vicinity. If
anything, the two Fathers were now more on their guard than ever before.
Although Kikukule had his spies everywhere, especially amongst the
postulants attending the mission, the Fathers leant through one young boy
whose parents were adherents of the chief, that an attack on the station was
imminent. In the first days of the new year, 1895, hurried steps were taken to
gather their followers and loyal Banyoro at the Mission. A message was then sent
to Kikukule telling him not to worry if drumming and firing took place since
the missionaries expected to engage a strong force of raiders whom, although
unidentified, they confidently expected to disperse. The message got home and
Kikukule, this latest ruse exploded, struck camp and moved off once more.
The respite provided by this departure gave Achte the opportunity to think
carefully about consolidating his own position. Open as the little mission was
to ambush and attack, it was time, he felt, to erect a substantial main building.
So he decided on a mud-brick structure designed for defence. This would be large
enough to shelter non-combattants and strong enough to withstand determined
The hill on which Kikukule's own residence within his compound had
stood was selected for the site. Achte planned the 'fort' along the lines of the
strongly-built Mission of Bukumbi at the southern end of Lake Victoria, near
Mwanza, Usukuma, which had been his first contact with the interior in 1890
at the close of his journey inland from the coast. The plan was straightforward,
four bastions each four metres square, connected to one another by walls forty-
four metres long three metres high, the whole structure pierced by loopholes.
Work was started without delay; the women carried water; the men, their guns
handy, kneaded mud for the bricks which were sun-dried. When the first
thousand bricks were ready the young postulants brought them up the hill
where Achte took on the task of mason. Houssin filled the role of carpenter. On
23 January 1895 the first brick was put in place.47
Work was slow as Achte literally laid each brick in turn. It was not until
the end of March that the building was completed. There had been little time
for relaxation for almost daily Kikukule's men were watching the activities from
a distance. Whenever a group of Banyoro approached the hill, the mission drums
beat the alarm, guns were siezed and the Baganda warriors in the rear stood
ready. These had been tense days for the Fathers but now with their fort com-
pleted they felt there was little more to fear. In fact, confronted with the solid
block-house Kikukule withdrew, crossing the Kafu river to the most northern
limits of his province. For the next three months, free of his menacing tactics,
the whole neighbourhood of Bukumi and Lwekula's, took on a new lease of
life. More Banyoro returned to their villages and new Baganda converts from
Buddu came to settle near the mission.
In northern Bunyoro Thruston had handed over command in January to
Captain G. G. Cunningham. The year did not start too well for the British. An
attempt to surprise Rwabudongo, who was leading a strong band of Banyoro,
was unsuccessful. Later in March Kabarega, who was in a position on the east
bank of the Nile, beat off an attack, when the attacking force of a few British
officers, Sudanese and Baganda suffered severe casualties. In Banyoro a
number of chiefs particularly Rwabudongo, Jasi and Ireta, continued to be
hostile towards the intruders. By April the position of the occupying forces had
been strengthened by a line of forts which traversed Bunyoro from Kibiro on
Lake Albert, through Hoima to Baranwa. Subsequently, with the arrival of
more reinforcements a concerted effort was made to apprehend the guerilla
king, and so the "Unyoro Expedition" under Cunningham left Mruli on 27
When the campaign drew to a close a month later the expedition had
advanced along the east bank of the Nile from Bukedi into Lango and had also
patrolled a large part of northern Bunyoro. Kabarega continued to elude capture
remaining in Lango. His mother, Nyamutahingurwa and others of his family,
including a young ten-year old prince, Kitahimbwa, had been taken prisoners.
A large number of the royal cattle herds were also impounded.'" Rwabudongo,
who had been most active and had had some near escapes, finally met with a
heavy defeat in May near his village at Muzirandura, after which he fled to
Toro.so To consolidate further the position, a fort had been constructed in
Busindi, a district lying northeast of Hoima. From this position, called Masindi
Fort, Major T. B. P. Ternan was kept busy checking the activities of the sur-
viving chiefs still loyal to Kabarega. Amongst these was Kikukule who, after
leaving the southern hunting grounds, was either raiding the homesteads of
Banyoro friendly to the occupying troops or those of Baganda over the border.51
134 UGANDA JOURNAL
Despite the presence of guerilla bands the situation was much improved,
and work had been started on laying a road between the two forts of Hoima and
Masindi. On the other hand southern Bunyoro, as yet unmapped, remained
very much of an unknown quantity to the military. In the southwest Nakabimba
Fort, named after the chief of the area, had been built on the south bank of the
Muzizi River to provide a means to check Arab caravans which were supplying
Banyoro rebels with arms and gunpowder, in exchange for slaves and ivory.
In June, Lieutenant C. F. S. Vandeleur, who had been in Bunyoro since
January, was ordered to proceed south from Hoima Fort to endeavour to impede
the movement of these caravans. The expedition, coming unexpectedly from the
north met with considerable success. Vandeleur's immediate objective was an
Arab trading station, but first, after entering Chikakule's district, he had a
brush with the surprised inhabitants who collected to defend their villages,
but were quickly dispersed. He crossed the Kafu river and then located and
overran the trading station which bore the name of the Munyoro chief of the
area, Mwenda. Leaving a stockaded post containing the captured booty he
continued southwest to the Muzizi River to establish contact with Nakabimba
Fort 52 where he met up with Captain W. P. Pulteney. After further successes
against Arab traders he returned to Hoima on 1 July, having passed through
Mwenda's to collect the small garrison left there, the captured loot, prisoners
and slaves.53 Pulteney accompanied him this far but then returned to the fort
During July, Kikukule recrossed the Kafu into southern Bunyoro, probably
to avoid pressure from increased patrolling. With the Arabs gone and Vandeleur
out of the way, he smartly took over the deserted camp of Mwenda's. From here
he was soon raiding again over a wide radius, even penetrating the neighbourhood
of Bukumi Mission from where, he was reported, to have taken a large number of
women and children. With southern Bunyoro once again under the threat of
its erstwhile chief, the hardworking Fathers at the mission must have grown
despondent. Despite its hazardous start and thanks to the sang-froid of Achte,
at all times staunchly supported by his colleague Houssin, the mission though not
yet two years old, could claim by this time 1,132 converts and about 2,500
In the north the grip of the occupying force, scattered at half a dozen
strategic posts, tightened even more. Nyamutahingurwa, the Queen Mother,
remained in custody at Hoima. Although not the easiest of prisoners to hold,
she had initially responded to the courteous treatment afforded her by co-
operating with her captors in sending messengers to all parts of the country to
induce the people to cease hostilities, and trying to exert her influence to per-
suade Kabarega to submit unconditionally. More rebel chiefs were capitulating.
By the end ofcluly such prominent men as Byabachwezi and Ireta had accepted
the terms offered, especially that no Banyoro chiefs loyal to the new regime
would be disturbed in their cheftainships. 54 All the same resentment against
the new order burned as deeply in the heart of Nyamutahingurwa as in others,
for she soon incurred the displeasure of her captors when it was learned she had
sent word to Kikukule and Mwenda, who had not yet surrendered, to storm
Hoima Fort which was then lightly garrisoned.55 The opportunity to attack
was not taken, and later Mwenda gave himself up. Kikukule still remained at
large. Eventually information that he was encamped at Mwenda's reached
Achte at Bukumi. Word was sent to Ternan at Masindi.56 In October a company
of Sudanese troops left Masindi with orders to find and attack the notorious
chief. The expedition was, in part, successful for it surprised the band, killing
fifty of them. One hundred women and children were recovered and returned
to their villages; likewise fifty-one head of cattle. But again Kikukule made good
his escape. However, in a short while, no doubt shaken by this latest reverse
and having seen his friends give up in turn, he took the best course open to him
under the circumstances and surrendered to Ternan at Masindi Fort on 21
October, 1895. ". . the Chief Chickakuli surrendered; the last of the big
chiefs not actually with Kabarega to come in. I fined the gentleman ten tusks of
of ivory and detained him pending payment."5' The Queen Mother, on her
way to Kampala where she would be beyond the immediate reach of her people,
passed through Masindi, missing by twenty-four hours the arrival of Kikukule.
Kikukule's detention lasted for four months during which period, as well
as Chief Mwenda, he went, or was sent, to Kampala. Before leaving Uganda for
good Vandeleur wrote when in Kampala that, "Chikakule was an old man and
was evidently much impressed with what he saw at Kampala, and must have
wished he had made friends with the white man before.""" In the light of sub-
sequent events the old firebrand was anything but won over, for an independent
spirit simmered sufficiently to need little to rekindle the flames of a rebellious
nature when the opportunity arose to revert to the ways of the past.
Baganda chiefs had assumed control in southern Bunyoro. Not unnaturally,
they played the part of conquerors, victimised those Banyoro peasants who had
not fled and in fact used them as labour. By the end of 1895 the situation had
deteriorated badly. Major Pulteney, having assumed responsibility for the region
as Civil Officer in order to restore administrative order, therefore took the step
of putting Banyoro chiefs in overall charge. First, he obtained the collaboration
of the influential Rwabudongo, as well as another chief, Nakabimba. The
scheme worked, although the Catholic Baganda protested strongly against the
imposition of such a local authority provided not only by Banyoro, but by men
men who still held pagan beliefs. Pulteney went further and, so as to strengthen
the two chiefs' hands, arranged for them to have assistance in the persons of
Mwenda and none other than Kikukule. Both were made available to assume
these duties under the new order during February 1896.59 Certainly, these men
could wield considerable authority, which if controlled and channelled along
peaceful lines could produce a happier and more rewarding effect throughout
Although the region seemed to benefit from this arrangement, Catholic
objection was too strong. As a result Pulteney was forced to resign. Rwabudongo
was moved to northern Bunyoro and his assistants were superseded by Baganda.eo
Taking no chances under the resumption of Baganda authority, Kikukule
likewise went north. He had been sick for some time and returned to Masindi
where he gladly took advantage of the treatment now available at theanedical
station located outside the fort. From this time, in so far as oral and available
written references are concerned, he lapsed into obscurity for about eighteen
The position in Bunyoro had quietened down, although there remained
a number of chiefs of the old order who were reluctant to come to terms, men
who were ever ready to exploit any weakening in the system of control of active
patrolling from the strategically placed forts. The complement of British
officers in the Protectorate had been substantially increased, and the regular
military force reorganised to form the Uganda Rifles.
Perhaps Kikukule's days would have ended peacefully if the temptation
to revert to his old ways had not presented itself once more. His participation in
the attempt to bring order to southern Bunyoro had failed through no fault of
136 UGANDA JOURNAL
his own, but because of religious pressure. It could have been no easy task for men
of the old regime, used to wielding authority by force, suddenly to have to change
their ideas, beliefs, customs and mode of living at the behest of the white new-
comers. There were outstanding examples amongst the younger chiefs, such as
Paulo Byabachwezi, who once won over never waivered in their support in
trying to build a new and better country for future generations of Banyoro.
But many leaders had already died in opposing the encroachment of foreign
influences; others had tried, but failed, to re-orientate their way of life and
methods of governing. Likewise Buganda, the kingdom where Europeans had
been welcome, was beset with a similar problem.
In July 1897 Kabaka Mwanga, straining against the restrictions imposed
on him in conducting the internal affairs of his country, rebelled and, for a
second time, fled into the county of Buddu. Supported by a growing number of
malcontents he was quickly followed by a force of three Sudanese companies
under Ternan, and elements of the Baganda army. After some engagements
he was driven further south to the borders of Koki. Here he suffered final defeat
but escaped across the Kagera River into German administered territory where
he was detained. In September the Sudanese troops who had been engaged in
these operations, had been switched to other exacting activities in Kenya. As
a result they mutinied. Speedy action was taken to try to prevent the mutiny
from spreading. At first this was thwarted in the north by the early disarming
of the Sudanese garrisons of the Bunyoro forts. Then, in December, Mwanga
escaped from German custody. Back in Koki with a considerable following he
soon attracted the support of Mohammedan Baganda. In January 1898 the
revolt came to a head when the uprising was reflected throughout Koki and
Buddu in the flames of burning villages and mission stations. In Bunyoro the
time seemed ripe for renewed efforts to rebel. Kabarega was still free in Lango;
a number of Sudanese mutineers, who had escaped capture, were on the run.
With Mwanga's men on the rampage in Buddu, so Kabarega and his adherents
became the rallying point for all malcontents.
When word reached Bunyoro that Mwanga intended to march north to
link up with his old foe, many Banyoro chose the occasion to rise against their
chiefs who supported the British; and they were incited especially by the old
offender Kikukule.62 The chance to raid, to loot, to re-establish himself in the
position he had originally held in his own lands must have been too great a
temptation. After rallying his supporters he began to harass any Banyoro who
had been actively aiding the British.63 Flushed with success he let it be known
that his main purpose was to oust the missionaries from the foothold he had
given them in southeast Bunyoro. Likewise, such men as Ireta and Jasi were
quick to perceive the opportunity and soon were on the move, intending to
join up with those Sudanese mutineers seeking out Kabarega.
By mid-February Kikukule's presence was being felt in the south as reports
came to the missionaries at Bukumi of a wave of raiding, of Christians being
killed and their children being carried off. In the meantime Mwanga advanced
northwards towards Buwekula. The Luwekula, Mutagwanya, besought every-
one in the neighbourhood to flee; but the four missionaries then at Bukumi,
Mgr. Streicher, Fathers Moullec and Van Wees and Brother Dominique,
stood fast. Father Achte's foresight in building his 'fort' was now to be put to
the test.64 Preparations against attack were completed. On 1 March the main
entrance gate was walled up, only a postern gate being left available to give
access to a nearby banana plantation. All dried grass had been removed from
the walls so as to reduce the chance of fire; every conceivable container inside
the fort was filled with water. News came that the mission at Bujuni, built two
years previously twenty miles to the west, had been burnt down. Even more
serious was the rumour that Kikukule was on the way to fulfil his prediction to
be reinstalled at his old capital within a few days. Refugees, including a group
of Protestants from Singo, and odd groups of Baganda and Ankole warriors,
still found their way through to this haven, although villages and compounds in
the vicinity blazed almost continuously as sparks and flames burst anew in all
The days passed anxiously for the defenders. Isolated attempts by hostile
Banyoro to approach the fort were brushed off, whilst sorties were successful
in scattering groups of the rebels. Some casualties were suffered but those of the
attackers were greater. Mutagwanya's capital at Kaweri in Buwekula fell to a
band of Mwanga's marauders. With other Catholic Baganda chiefs and their
fighting men, Mutagwanya hastened to the one strong-point, the mission at
Bukumi. Once there they did not hesitate to give battle to the increasing numbers
of Banyoro and Mohammedan Baganda who were gathering around.
The beginning of 1898 had seen a further increase in the availability of
officers and men for operations in Bunyoro. Now the arrival of reinforcements
from Mengo relieved the situation. The menace of Mwanga's men, and
of Banyoro inspired by Kikukule, had subsided by 13 of March, when
Captain C. G. H. Sitwell with sixty loyal Sudanese arrived at Bukumi. He was
followed on the next day by Major C. H. U. Price with other British officers and
a contingent of Indian troops (Bombay Light Infantry), as well as 200 Baganda
and Batoro guns. The force pitched tents close to the fort. The last shots near
the mission were fired on the same day and the threat was over.65
The uprising was soon broken. Kikukule and his rebels dispersed in the
face of a drive northwards by Price. Mwanga's men, comprising about 800
guns had split into two parties. One party suffered three quick defeats at the
hands of Price. The other, which had approached Hoima, was also defeated
on two occasions with heavy loss, by Captain Fowler. Ireta, having raised 500
Banyoro guns was intercepted when moving to join up with this second force
and defeated by Captain W. C. Barratt on 16 March.6@ Jasi managed to rejoin
his father, Kabarega. In the south of Bunyoro, the mission at Bukumi, strong in
design and having been resolutely defended, remained the sole haven in what
had become a vast area of devastation.
An important outcome was the setting up of a military post under the
command of Captain H. F. T. Fisher near the eastern flank of Mubende Hill,
overlooking the site of Luwekula's at Kaweri. Not only would this safeguard the
rebuilt headquarters of the Luwekula in the future but, if required, the garrison
would be available to provide aid to Bukumi Mission.67
Eventually, after evading attempts to capture him, Mwanga had reached
Kabarega's side. Joined by a number of Sudanese mutineers, the two fugitive
kings and their followers still presented a serious threat to any attempts to
achieve peace. Kikukule, all pretences to lending his authority towards the
restoration of order cast aside, gathered his scattered men together again.
Though the rebel forces had been split up by the British-led patrols operating
from Bukumi, as well as from the forts at Hoima and Masindi, the surviving
groups of Banyoro and Mohammedan Baganda, though frequently intercepted,
continued to be a constant threat to the country as a whole. During April and
May there were a number of skirmishes over a wide area. By June Ireta, based on
Kerota, had deployed his men between the Budongo Forest and the Murchison
(or Karuma) Falls. Mwanga was reported to be encamped at Kichwati, Jasi
138 UGANDA JOURNAL
had moved with his followers along the Victorial Nile from Foweira to Masindi
Port.ss Kikukule had gone to Kitoro from where he had begun to raid south of
the Masindi-Hoima road, mainly for food.69
Price was now in command in Bunyoro. He planned a major sweep-up of
the rebels. Special operations were started on 17 June. The first move was to
occupy Ireta's base which, in fact, had already been evacuated. On 26 June
three columns, under Captains E. J. Tickell7o and M. L. Carleton from Masindi,
and F. J. Fowler from Hoima, left by different routes to flush out the main rebel
groups in the region up to Lake Albert and the Nile, as far as Fajoa by the
Kikukule was on the move intending to meet up with Ireta when, on 29
June, Tickell and Carleton joined up and after pressing through the Budongo
Forest entered the village of Kitoro. Here they surprised a small outpost of
Kikukule's men who were driven away after a brief exchange of fire. On the
next day, continuing north towards Bulemweri, Tickell again caught up with
the enemy, and also obtained his first sight of Lake Albert to the west. "We had
some awkward places to cross had the enemy been enterprising; one of them was
a swampy river, about 200 yards broad, and about 4ft. 6in. deep, up to my
chin in places. We carried our rifles and ammunition over our heads, and I put
my watch and compass in my hat. As soon as the men had got together, and
were moving off on the far side, we came into and captured a large convoy of
food going to Chekukule, which came in most handily for ourselves."n
Tickell reached Fajao on 3 July where he was joined by the two other
columns led by Carleton and Fowler. Since information was given by a prisoner
of the whereabouts of Mwanga, new dispositions were made. It was left to
Tickell to try to run Kikukule and Ireta to earth. On the receipt of fresh news
about Kikukule, Ticklell marched out of Fajao with his small column of 28
Pathans, 38 Zanzibaris and 70 Baganda on 7 July moving in a southerly direc-
tion towards Igisi Hill. After two days march he reached Bliggi, afewmiles west
of Kerota. Describing the engagement which followed Tickell wrote:
"I had been feeling very seedy, and got through the march with difficulty,
as I was sick and could keep no food down. I think it must have been the effect
of drinking some extra filthy water at Kiswata the night before. At Bliggi we
captured a prisoner, who told me that Chekukule and Ireta had combined
forces, and were encamped in the thick jungle some seven miles to the southeast
of Bliggi, and that he would guide us there in the dark. This news was the best
medicine I have ever taken, for I felt perfectly fit at once, took a dose of brandy,
and ate a good meal. After making all my preparations for a night attack I
turned in for an hour's sleep, and then, at twelve midnight, fell in my attacking
party and marched off. There were three-quarters of the moon, and it rose very
red and stormy-looking. Shortly after starting a heavy thunderstorm came up;
everything was favourable for a night surprise; but, on the other hand, made
night marching through the jungle and dense forest doubly difficult. After
leaving my camp guard my attacking column consisted of 22 Indians, 28 Swahilis,
and 60 Waganda native allies. The night march proved to be a very difficult
one. When we got into the forest it was absolutely pitch dark, a darkness that
could be felt. How my prisoner guide, with his arms tied behind his back so
that his elbows touched, ever found his way through this intricate forest is even
now a marvel tome. The revolver at his head, and spear behind his back, must
have had a wonderfully sharpening effect on his wits. We waded through streams
and bogs, and stumbled at every step over roots, creepers, and fallen trees.
To make a long story short, we only progressed at the rate of one-and-a-half
Sepiriya Mutagwanya, the Luwekula from 1895 to 1920, at his home in
Mubende in 1953.
On his chest is pinned the Silver Star (Uganda Mutiny Star) awarded to
him on the 24th May 1899.
(Photo E. C. Lanning)
1- ii I
Reverend Father Auguste Achte--1892.
(Photo by Ste Missionaires-Afrique,
Pres Blancs-Phototeque, Paris.)
Captain E. J. Tickell--1888, who had several
brushes with Kikukule in 1898.
miles per hour, and at 5 a.m. on the 9th July I was walking with the point of
the advanced guard, having just gone up to check their pace, so as to close up
my column, when we suddenly came on the enemy's outlying sentry, who
fired his rifle and ran off, giving the enemy the alarm. I thought we must be
right on the camp, so gave the order to fix bayonets and double, but the camp
was further off than I had expected, and it took us a good ten minutes' run
before we got three. I must confess that in this dense jungle and, in fact, all over
this part of Africa, it is neverpossible to move two abreast--allmarching has to
be done in single file. When we got close to the camp we were met by a hot fire,
at ranges from ten to fifty yards, from men posted in the thick elephant grass
on our front and both flanks; and my men began firing in every direction and I
was afraid I should not get them to advance; however, a few more paces brought
us suddenly out of the jungle into the camp. Here there was a moment's hesita-
tion. I had only had these native troops a few days under my command, and
naturally, they did not quite know what to make ofa strange officer whom they
knew nothing about, and who did not even speak their language. My twenty-
two Indians, too, found themselves alone, the Swahilis being left far in rear and
our native allies behind them again; and to make matters worse, while we were
being fired at on our front and both flanks, our friends, hearing the firing, aim-
lessly opened fire in our rear, all their bullets however, going over our heads. I
know that if we checked for a moment, or showed even the slightest sign of
retiring, we should be utterly annihilated, for the Swahilis and Waganda would
at once bolt, and my small force of twenty-two Indian rifles (outnumbered more
than twenty to one) could never find their way back through the intricate forest.
Blowing my whistle, I got the men to cease fire, formed line, and advanced at
the double, cheering, and carried the camp from end to end at the point of the
bayonet. The enemy, though over four hundred guns alone, besides spearmen,
not waiting for the cold steel, bolted like hares in every direction into the jungle."
The rebels had retreated and the position was still precarious, for Tickell
was alone in the clearing of the forest with his advance guard of 22 Indians.
The Banyoro soon rallied and started firing on the small force from all sides,
only to be held at bay by carefully aimed volleys fired in turn into the forest
ahead and then into the green vegetation of both flanks. At last, seeing the camp
carried, the Swahilis (Zanzibaris) and the Baganda joined up. With these extra
men Tickell advanced into the forest on both f lanks, clearing it for a short distance.
The first shot had been fired at 5.10 a.m. and the last at 6.50 a.m. In Tickell's
own words, "The enemy had been taken by surprise, never believing it possible
that any troops could find their way through the forest and jungle to the sanc-
tuary they considered safe. Their camp was left standing, cooking pots, sleeping
mats, and worthless native loot of all kinds, which my Waganda allies promptly
looted. I, however, managed to secure Chekukule's war drum.n We picked up
fourteen dead bodies of the enemy in the jungle, but we have since learned from
prisoners that in this and my next two fights they lost over forty killed, including
two chiefs. Our losses were surprisingly small, being only seven casualties all
told. All my plans for surrounding the camp were upset by the guide--treacher-
ously, I believe--leading us right on to the sentry. In the confusion of the fight
the guide escaped, and I never saw him again, Prisoners told us the force we
defeated was 450 guns strong, with two or three times as many spearmen and
followers etc." And so, with the enemy gone the camp of about four hundred
huts was set alight. The force marched back again through the forest to the
camp at Bliggi, in a quarter of the time it had taken to cover the ground in the
140 UGANDA JOURNAL
dark. This clash took place in the Budongo Forest a few miles southeast of
Pabidi Hill. The majority of the prisoners taken eventually escaped, since
only a few men could be spared to act as guards. Tickell's losses in this action
were one killed and four wounded.7
There was little time in which to rest for on 12 July Tickell had a further
encounter nearer to Pabidi, this time with Ireta's men who were hiding in the
forest. Their camp of about 400 huts was burnt and the guerillas pursued for
two hours into the forest--"without any guides, all having bolted when the
fighting began. I fortunately had taken constant compass directions during
the pursuits, so found my way back again all right."74
Another brush with the enemy on the next day showed they had little fight
left in them, and were now on the defensive. Seven guerillas were killed. Evidently
some of the rebels were found to be armed with Martini-Henry rifles. Having
been operating in a foodless area all this time, the column was running short of
rations. The Zanzibari porters and Baganda had been subsisting on the rations
of the Indian troops. Whilst waiting to contact a supply convoy there was yet
another skirmish, the guerilla sentries being regularly posted, whilst others
were in position in rifle pits and behind rocks. After meeting up with Carleton,
Tickell moved by compass, traversing the forest in all directions for the next
eight days, completely scattering the bands of Kikukule and Ireta. During this
period of marching and fighting for fourteen hours on end, working without
guides in an unmapped country, by compass, always wet, with numerous
swamps and at one period being without additional provisions for six days,
there were seventeen cases of fever amongst the Indians, Tickell himself running
temperatures from 102 to 104 degrees for four days. On 21 July he returned to
Although recovered from his about of malaria there was little time in which
to rest for, on 26 July he left Masindi, accompanied by Lieutenant C. U. Price
and 32 Indians on an arduous march of 19 hours to try to surprise Ireta. The
fight was a poor affair, as nearly all the band was away raiding. After a brief
engagement with the few rebels left in the camp, only two enemy dead were
picked up for the cost of three casualties in the attacking force. Wet through
and very hungry the two officers and their men were glad to sit down and eat
the enemy's meal of steamed sweet potatoes which they found cooking. This
was too much for the rebels who counter-attacked, but without making much
impression on the hungry soldiers.'"
By the end of July the power of both Kikukule and Ireta had been consider-
ably reduced, but they were still troublesome enough to necessitate being kept
under pressure. On 5 August Tickell, with Lieutenant M. F. Gage and men of
the Uganda Rifles, once more caught up with Ireta and attacked. "I heard that
some 400 of the enemy were only some ten miles off in the forest in the mountains,
where they had never been tracked to yet, so I started off with Gage and 60 of
my new command for a night attacks, which was completely successful; the
hard work and bad food had been telling on me and I was chilled with the
night wetting, so I sat down by one of the burning huts and had my breakfast
under fire. I had the satisfaction of completely breaking up this hornet's nest
and burned every hut, some 450. The night after my attack I am sorry to say
I had a sharp go of ague and fever, which went to points over 105". It quite
knocked me up, and for the first time since I have been in Africa I was obliged to
be carried through the day's march." However, despite this trouncing Ireta and
most of his following slipped away, deeper into the forests." At the same time
Lieutenant Cape, patrolling from Hoima with a small column of loyal Banyoro
raised by Paulo Byabachwezi, gave chase to Kikukule, but without success.
Later in the month Captain J. Ponsonby and Lieutenant J. A. Hannyngton
came up with Ireta. Though suffering another defeat the chief made good his
escape. The end of organised opposition was in sight but the two tough chiefs of
southern Bunyoro were by no means content to yield too soon to the overwhelming
forces which were now harassing them. Towards the end of August 1898 Major
C. H. U. Price wrote of the situation, "it is a pity it has been so frequently reported
after every skirmish that 'this terminates the affair' for there has been so much
work in connection with the suppression of the mutineers and rebels since Mac-
donald left." He went on to note that Kikukule, with only some sixty followers had
moved to the vicinity of Busanga, about one march south of Fort Kafu on the
west of the Kitumbui River, a former forest refuge from which he had been
driven in March. Ireta, with a body of Mohammedan Baganda was still in
the Budongo Forest in the north, but almost all his Banyoro followers had
deserted him. The hands of rebel Banyoro and Baganda, Price summed up,
were of no more account as resisting powers, being merely disturbing elements
in the country causing, in their raids for food, constant panic amongst the
villagers whose reports, of course, tended to be considerably exaggerated. The
British command, he pointed out, was faced with a very large tract of country
and consequently these rebel bands could take up their abode at considerable
distance from the stations where available troops could be posted, and so move
according to their inclinations.'"
To all intents and purposes the countryside around Hoima and Masindi
had quietened down. In September the greater part of the British forces had
moved north to Fajao to prepare the way for an expedition down the Nile to set
up posts and make treaties with local chiefs as far north as Fashoda. The fort
at Hoima had been evacuated and temporarily entrusted to the charge of
Paulo Byabachwezi, whose own village was located about a mile away. By
October this news had reached Kikukule and Ireta. Despite the recent series
of setbacks they now combined forces in order to attack Hoima. In the north,
on 9 October, a large body of some 260 mutineers and a large group of Moham-
medan Baganda, having separated from Kabarega and Mwanga, had crossed
the Nile from Bukedi. They were quickly intercepted by Hannyngton near
Kisalizi, but broke away. As they moved south they were joined by more rebel
Baganda and Banyoro.7 In the meantime, at Hoima, Byabachwezi had learned
of the threatened attack by Kikukule and Ireta and so withdrew from the station,
which in any case, did not contain any Government property. The guerillas
stormed in and destroyed the fort, returning quickly enough to the security of
the forests. Later, after receiving reinforcements, Byabachwezi reoccupied the
post which he then started to rebuild.ao
Troops had been hastily despatched from Kampala to deal with the muti-
neers and rebels who had broken away from and mauled Hannyngton and his
small force. Though the rebels suffered a severe reverse at Kiweri on 27 October,
they continued to be a menace as, with Captain Fowler on their heels, they
headed southwest intending to move into Anktole. After crossing the Kafu they
went in the direction of Bukumi Mission. It is left to surmise if Kikukule had
joined them, for, from now onwards, there is little mention of him in connection
with the final operations. It would have been typical of him to be to the fore in
yet another attack on the mission, surely still hoping to regain control of the
territory over which he had once held sway. At Achte's fort Father Moullec
and Brother Dominique, both of whom had taken part in the siege earlier in the
year, had been joined by Father Selles.
142 UGANDA JOURNAL
As soon as news of the approach of the rebel band reached them the mis-
sionaries took stock of the situation. Since they could count on the arrival of
forty trained Baganda soldiers from three posts under the command of Captain
Fisher, based at the time at the unfinished fort at Kaweri, Buwekula, and on
Mutagwanya, and other Catholic chiefs of the county, and some from south
Bunyoro, they decided to stand fast. At 10 a.m. on 5 November the rebels swarmed
to the heights of Kasaka Hill overlooking Bukumi. Shots were exchanged. The
second siege in the short life of the mission had begun. The defence (270
guns) was determined. Though the position grew increasingly tenuous the rebels
were held at bay. Torrential rain hampered both besiegers and besieged,
eventually causing the rebels to desist in their efforts. "If the enemy returns
and continues the siege, rain, cold, hunger and disease will powerfillly assist him
in triumphing over us," sums up the feelings of Father Moullec recorded at the
time.s' After a final assault on 7 November the rebels, estimated at 2,000 strong,
withdrew to the commanding position of Fort Grant which Owen had erected
on the Butorogo range three years previously, but which subsequently had
been abandoned. After some anxious days Captain J. A. Meldon arrived at the
mission with a relief force. The danger was over. British-led forces converged
on the rebel group which, by the end of the month, had been driven from the
neighbourhood, the main body seeking refuge afield in the uninhabited areas
between Buddu and Ankole.
The remaining guerillas in the north were given no rest. Being routed on
almost every occasion disintegration set in. By the end of November large
numbers of Banyoro had surrendered but the hard core of guerillas, still includ-
ing some Mohammedan Baganda, hid in the Bugoma Forest. Here on 6 Dec-
ember, Major C. H. U. Price successfully attacked a band of some 400 guns under
Ireta, which was occupying a strong and concealed position.82 A number of
flying columns continued to harass both mutineers and rebels right up to the
end of the year, inflicting casualties and taking an increasing number of pri-
soners. Many were tried and imprisoned in Kampala. Those who had given
themselves up voluntarily received light sentences of nine to twelve months
imprisonment with hard labour."" There were still a number of leaders and
their henchmen who managed to elude capture; for their Mukama, as well as
Kabaka Mwanga, remained free in the lands of Lango. There is no further
mention of Kikukule until his arrival as a prisoner in Kampala. Perhaps,
embittered, even weakened by sickness and he was certainly getting on in years,
he did no more than play second fiddle to the more ebullient Ireta.
The year 1890 opened to a determined attempt to deal with the two fugi-
tive kings. A special operation was mounted under the command of Lt Col.
J. T. Evatt, who, in March with an Indian contingent, two detachments of the
Uganda Rifles and a contingent of Baganda auxiliaries led by Semei Kaku-
nguru and Luwandaga Kimbugwe, proceeded to the north shore of Lake
Kwania. By early April contact had been made with the enemy. After a few
days' pursuit a large party of some six thousand Banyoro, mostly women and
children but including the chief Rwabudongo, and fifty to sixty of Kabarega's
family and servants, surrendered." The end came on 9 April. After fierce: resis-
tance, when No. 4 Company, the Uganda Rifles suffered two killed, and the
Baganda Auxiliaries three killed and twenty wounded,** Kabarega with two
of his sons, Jasi and Duhaga, were taken prisoner. All three were wounded,
Jasi dying three weeks later. Mwanga surrendered later on the same day.ss
A small number of Sudanese mutineers and some of Kabarega's other sons
managed to escape. Broken up, some on the run, some in hiding, the remaining
mutineers, Banyoro rebels and Mohammedan Baganda were now powerless.
On 11 May, Kabarega and Mwanga arrived in custody in Kampala. The news
of this final fight had soon filtered through the forest glades of northern Bunyoro.
In the same month thirty five mutineers, together with seven chiefs, including
Ireta, surrendered at Masindi.87 In all probability Kikukule was with this
Nearly 300 prisoners, including ex-mutineers, were held in prison in
Kampala, a number far in excess of that which could be safely superintended
in the temporary prison sheds. Some of the leading mutineers and rebels were
executed, whilst lesser chiefs received terms of imprisonment. Rebels continued
to give themselves up well into June, but the majority of the peasantry were
allowed to return to their homes unmolested. By the end of the month, Ternan
was able to report that "Unyoro is now peaceably settling down. All the principal
Banyoro chiefs, who have been against us, have surrendered, and the people
are quietly reoccupying their villages and cultivating the land."**
With the other guerilla leaders, Kikukule was by this time back in Kampala.
He was tried and sentenced to term of imprisonment. He was not long in close
confinement, being admitted to hospital suffering from acute fever.
Kabarega's kingdom had collapsed. His conviction that once he let the
white man get established in his country, would herald the end of the indepen-
dence of his kingdom and his own freedom, had been fully justified by the turn of
events."" When Colvile had entered Bunyoro in 1894 it was an important
trading area, with plentiful crops and a considerable population. The prolonged
guerilla warfare which had followed had devastated the most fertile areas, and
soon the trade routes were closed. Shorn of territory in the west and the south,
given by the British to their Baganda allies, the remainder of the kingdom had
been depopulated as a result of the strong resistance to the intruders, whilst the
unceasing destruction and neglect of crops and the fertile soil had brought
famine to large parts of the country. The new administration had perforce to
rely, for most of the upper echlons of native administrators, on chiefs provided
by the hereditary enemy, the Baganda. Kabarega's place had been filled by one
of his youngest sons, Kitahimbwa, who the British had already nominated
Mukama early in 1898. Masindi, athwart the main caravan route to the Nile
stations, had grown into a military centre where, as at Hoima, missionaries of
the Church Missionary Society who had entered northern Bunyoro in 1895,
had begun to settle. Kitahimbwa, who had set up his capital here, had been
baptised early in 1899 and given the name of Yosiya.g0 Other princes and chiefs,
likewise, became Christians.
In July 1899 Father Achte visited Masindi. Here he met a number of
Kikukule's pages and warriors beginning to lead more settled lives, men whom
he had known since his first meeting with the chief in 1894.9' Two months later,
probably unknown to him, his blood-brother Kikukule died in Kampala.92
A tough and sly fighter Kikukule was as two-faced as others of his time,
and no more than those with whom he secretly treated. Directly appointed by
the Mukama as chief of a region, he had power of life and death. Though he
put his interests first, he never faltered in his fight to preserve the 'old order'.
Perhaps because he was more outspoken and more prone than others to prac-
tise threats and guile to attain his ends, he is remembered in Bunyoro by the
"Olala omwa Kikukule ngu ndaire ne ndota Abaganda."
"Should you sleep in Kikukule's house you will surely tell us that you
dreamt of the Buganda."
144 UGANDA JOURNAL
It is said the old chief was laid to rest at Simoni, near the Kampala Sports
Ground. An elder, named Kyampama living at Kibuli (1955) claimed to have
been present at the burial. At one time there was a small African church and a
mvule tree near the spot.93
Whilst the official records state Kikukule died of sickness and fever, there
is extant another version of how he met his death. After a term of imprisonment
he was released. One day in Kampala, so it is said, he quarrelled with a Sudanese
soldier. He was shot dead as he tried to defend himself with an iron bar. He was
buried at Simoni.94
1. There are many variations in the spelling of this name by diarists and writers of the day.
These include Kikakure, Kikoukoule, Chichakule, Ciccoculi and other forms. The form
now generally accepted is Kikukule.
2. This region was originally in Bunyoro until overrun by the Baganda during the reign of
Kabaka Kamanya (c. 1814-c. 1832). The name Buwekula is derived from the title of the
sub-chief Luwekula under Mukwenda, Singo, to whom the new territory was annexed.
(See Note 196).
3. Entebbe Secretariat Archives, Staff & Miscellaneous 1899 IN. Grant to Commissioner,
10 Sept. 1899
4. Kikakure's appears on most maps of Unyoro up to 1900, particularly I.D.W.O. maps nos:
962, 1075 & 1429 (c); also Vandeleur's map of Uganda and Unyoro, 1895.
5. A list of counties approved by Omukama Kabarega during his reign, is given by K.W.,
The Kings of Buyoro-Kitara, Part III, Uganda J>, 5 1937, p. 67.
6. K.W., op.cit., p. 64.
7. (a) Fisher, Mrs. A. B., Twvilight talesofthe black Baganda, 1911i, p. 167, and Nyakatura,J. W.,
Abakama ba Bunyoro Kitara, I 947, p. I 57.
(b) One of these children was not surrendered but was smuggled out of the country into
Ankole. This was Kasagama who later found refuge in Buganda and was subsequently
installed by Captain Lugard (1891) as Mukama of Toro.
8. Beattie, J. Bunyoro, an African kingdom, 1960. pp. 16-24.
9. K.W., op.cit., p. 65.
10. For a detailed description of this post see Uganda J., 17, 1953, p. 20.
11. (a) African No. 4 (1892), C.--555, pp. 121/122.
(b) Though usually referred to as 'forts', these posts were no more than earth-banked
stockades, not even defensible in design, with palisades too high to fire over and without
loopholes. Their main use was as bases for mobile patrols. Moyse-Bartlett, H., ~The King's
African Rifles, 1956, p. 53.
(c) W. Fenwick de Winton died at Kawanga on 28 March, 1892.
12. Sir Gerald Portal had proclaimed a provisional British protectorate at Kampala Fort on
17 March 1893.
13. Ashe, P. R., Chronicles of Uganda, 1894. p. 188.
1.K.W., op.cit., p. 65., also Africa No. 7 (1895), C.--7708, pp. 9/10.
15. Furley, O. W., Kasagama of Toro, Uganda /, 25, 1961, p. 169.
16. Macdonald, J. R. L., Soldiering and surveying in British East Africa, 1897, pp. 203/204.
17. Macdonald, ibid., p. 299.
18. Bovill, M., and Askwith, G. R., Roddy Owen: a memoir, 1897, pp. 117-142.
19. (a) Mukwenda. Title of the County Chief of Singo.
(b) Luwekula. Formerly the title of the sub-chief of Sabawali, Singo. The adjacent area of
Bunyoro occupied by the Buganda (see Note 2) was eventually added to Sabawali. Because
of the increase in size of the area administered by the Luwekula, a separate district distinct
from Singo, was created. Hence Sabawali of Singo, plus the newly occupied territory, came
to be known as Buwekula. The boma, Lwekula's, was located at the foot of Mubende Hill
at Kaweri, nowadays just beyond the boundary of the modern township of Mubende.
20. Thomas, H. B., and Dale, R., Uganda place names Uganda J, z7. 1953, p. 1 12.
21. Africa No. 7, of.cit., pp. 26/27.
22. He was accompanied by Captain J. H. S. Gibb and A. B. Thruston.
23. Colvile, Sir H., The land of the Nile springs, 1895, pp. 48-70.
24. Declt, L., Three years in savage Africa, 1898, pp. 416-428.
25. Africa No. 7, op.cit., pp. 10/11.
26. Africa No. 7, ibid, pp. 19/20.
27. Bovill, and Askwith, of.cit., p. 143.
28. Thruston, A. B., African incidents, 1900, p. 132.
29. Le Blond, G., Le Pire Auguste Achte, 1911, p. 220.
30. Achte to Livinhac; letter dated 16-7-1894, quoted by Le Blond, of.cit., pp.220-229.
31. Chronique, WFM Magazine, April, 1894. (CIPA, Rome).
32. Although my notes of an interview, late in 1954, with this witness are full, the report has
been lost. The only record available to me is that his first name was Patrisi and, at that time,
he was a Muruka Chief at Kawololo in Bugangaizi. He said he had been baptised in Buddu
by Mgr. Streicher.
33. This ceremony of blood-brotherhood between Kikukule and Achte took place at a date
between 2 April and 8 May 1894. The full account is given by Achte in his letter to Mgr.
Livinhac, dated 16 July 1894. (Le Blond, of.cit., pp. 220-229).
34. Lanning, E. C., Caves and rock shelters of western Uganda, Ugandaf 26, 1962, p. 184.
35. This boma was only a few miles distant from Kikukule's original kyalo. The location, locally
known as Kakamera (subsequently Kakumiro), this Lwekula's became the headquarters
for Kiimba, the Muganda County Chief for Bugangaizi. There, in 1901, C. W. G. Eden
opened the first British Collectorate for what was then called Kakumiro District, namely
Buwekula, Buyaga, Bugangaizi and Singo. In 1909 the title was changed to Mubendi
36. Witness Amiri Kyobe. Interviewed at Bukumi Mission, together with S. Mbugeramula,
in October 1954, in the presence of Fathers Grandmaison and Paredis. Born at Mityana.
At the age of about ten, he accompanied his 'brothers' who were going to fight Kikukule as
part of the Buganda occupation force. Arrived at Kikakule's a few days after Fathers Achte
and Houssin had started to settle in. Achte took on Kyobe as a servant. In the absence of
any opposition the rest of the party went on to Nkondo. Kyobe remained with the two
Fathers for some time, but went to Kampala in 1898.
37. According to Patrisi and Kyobe. Both witnesses and other old men with whom I have spoken
identified this first station by the name of Kijaguzo, 'that which rejoices'. They preferred
to identify the whole site, including the location of the modern mission which is on the site,
including the location of the modern mission which is on the site of Achte's fort, as Bukumi.
Kijaguza is now the name of a village northeast of Bukumi. (Also, see Snoxall, R. A., Some
Uganda place-names, Uganda J., lo, 1946, p. 52.
38. (a) According to witnesses Patrisi and Kyobe.
(b) Personal communication, 1955, Mr. M. B. Nsimbi.
39. Le Blond, op.cit., p. 225.
40. Unyoro Field force Diary, 26 May-27 June, 1894. F.O. 2/72.
41. This was a square stockade with side 100 yards long. There were small square bastions at
two of the opposite corners which flanked all four sides. Houses for the troops were built in
streets facing the stockade. Gardens were laid out, irrigation used and rice cultivated.
Within a year the garrison had well over 400 acres under cultivation around the fort. Thrus-
ton, of.cit., p. 154.
42. (a) Unyoro Field Force Diary, 29 Sep.-29 Oct. 1894. F.O. 2/72.
(b) Moyse-Bartlett, of.cit., p. 60.
(c) Thruston, of.cit., p. 128.
43. A Protectorate had already been formally declared over Buganda in June 1894. Bunyoro
was not included until two years later, in June 1896.
44. Nyakatura, of.cit.
45. Africa No. 7, of.cit., pp. 137/138.
46. Personal account by Ameri Kyobe, Oct. 1954. (Note 36 refers.)
47. Le Blond, of.cit., p.250.
49. (a) Africa No. I (1896) C.-7924, p. I.
(b) The herds of the Omukama were estimated to number from 10 to 20,000 head. Mullins,
J. D., The wonderful story of Uganda, I 904, p. I 47.
50. Africa No. 1, of.cit., p. 4.
51, Ternan, T. P. B., Some experiences of an Old Bromrgrovian, 1930, p. 226.
52. More commonly known as Fort Roddy. Built in May 1895 by Mr. F.J. Foaker of the Uganda
Administration. Uganda J, 7, 1939, p. 71, and 17, 1953, pp. 104 and 1 19.
53. (a) Vandeleur, of cit., p. 88.
(b) Many bales of cloth, coloured silks, cowries, 24 tusks of ivory, guns, percussion caps,
33 kegs ofgunpowder; also 20 slaves, 18 prisoners, 35 cows, 3 donkeys, some goats and sheep.
Vandeleur, op.cit., p. 89.
54. Roberts, A. D., The lost counties of Bunyoro, Uganda J., 26, 1962, p. 197.
55. Vandeleur, of. cit., p. 101, also Two years travel in Uganda, Unyoro and on the Upper
Nile, GeographicalJ., 9 1897, pp. 369-377.
146 UGANDA JOURNAL
56. Berkeley to Salisbury, 20 Nov. 1895, F.O. 2/93 (1895).
57. Ternan, of.cit., p. 228.
58. Vandeleur, of.cit., p. 141.
59. Pulteney to Commissioner, 23 March 1896 ESA 1896, Staff & Misc: IN.
60. Gregory, J. W., Foundation ofBritish East Africa, I901, pp. 217-218.
61. Macdonald,J. R. L.,Journey to the north of Uganda, GeographicalJ. 14, 1899, pp. 129-132.
62. Africa No. 7. (1898), C.--8941, p. 40.
63. Dugmore Memorandum on Bunyoro, 7. February 1898, ESA 1898 Staff&8 Mise. IN.
64. Father Achte had left Bukumi Mission on 5 November 1895 in order to open a new mission
in Toro. He did not return to Bukumi until the end of hostilities in Bunyoro, when he visited
the mission for a few days in June 1899. (Le Blond, of. cit., p. 363.)
65. For a full account of this event see Gray, Sir John, The Sieges of Bukumi Ulganda J., n5>
1961, pp. 65-85
66. Africa No. 7, (1898), op. cit., p. 55.
67. Gray, Sir John, o cit., p.68.
68. Dunbar, A. R., Uganda J., 24, 1960, p. 237, and A history of Bunyoro-Kitalra, 1965, p. 64.
69. Africa No. I (1899), C. 9123, p. 38.
70. See biographical note.
71. Africa No. 1, (1899), op. cit., p. 38; also letter dated Masindi 25July 1898 from E.J. Tickell
to his Mother, Louisa Emily Tickell.
72. Normally each chief had his special war drum. The drum was the rallying point for his
men, and its loss regarded as a great disgrace. The list of Lt. Col. Tickell's trophies brought
back from Africa includes 'one drum'. On his death his trophies were either sold or destroyed
and there is now no trace of this particular trophy, if in fact, this was the war drum taken
during the skirmish of 9July 1898.
73. Letter from E. J. Tickell to his Mother, dated Masindi, 25 July 1898 also Africa No. I
(1899), of. cit., pp. 39/40.
74. Letter from E. J. Tickell to his Mother, dated Masindi, 25 July 1898.
76. E. J. Tickell to his brother, R. E. Tickell, in two letters dated Masindi, 25 July 1898 and
Mruli, 29July 1898.
77. Letter from E. J. Tickell to his Mother, dated Fajao, 15-&-98. also Africa No. 1 (1899),
of.cit., pp. 46/47.
78. Letter from Major C. H. U. Price to a member of the Tickell family, dated Kampala,
30 August 1898.
79. Africa No. I (1899), of.cit., pp. 15/16.
80. Africa No. 4 (1899), C. 9232, p. I1.
81. Letter from Father Moullec to Mgr. Livinhac, dated 20 November 1898 quoted by Gray,
Sir John., of.cit., pp. 78-84.
82. F.O. 2 (Africa) 201, Ternan to Lansdowne, 27 December 1898.
83. F.O. 403, 281, (7401), Ternan to Berkeley, 21 January 1898.
84. Gray, Sir, John, Acholi history, 1860-1901, Uganda J., 16, 1952, pp. 45-6.
85. F.O. 403, ofb. cit., Evatt to Berkeley, 2 April 1899.
86. F.O. 2, ofb. cit., Evatt to Ternan, 2 April 1899, lists prisoners and booty taken as:
Kabarega, Jass (Jasi), Duhaga, Mwanga, plus katikiro and members of household, 12
mutineers many Banyoro, 400 head of cattle, 1 Maxim gun.
87. (a) F.O. 403, 282, (7402), Ternan to Salisbury, 18 May 1899.
(b) "The last knot of mutineers was not dispersed until Major Delm6-Radcif fe's Lango
expedition of 1901". Thomas H. B. and Scott, R. Uganda, 1935, p. 39.
88. F.O. 403, 282, of. cit., p. 104.
89. Thruston, of. cit., p. 83.
90. (a) Kitahimbwa was deposed by the Protectorate Government in 1902. He was succeeded
by his brother Andreya B. Duhaga who, with Jasli, had been captured with Kabarega on
9 April 1899.
(b) When I met Kitabimbwa in 1952 he was living in quiet retirement in the north of
91. Le Blond, o cit., p. 367.
92. Grant to Commissioner, 10 September 1899. ESA Staff & Misc. 1899, IN.
93. Personal Communication, 1955, E. W. K. Bulera and M. B. Nsimbi.
94. Personal Communication, 1955, E. W. K. Bulera, who kindly made enquiries in Kampala,
through Erisa Kaliisa, on my behalf.
Paulo Byabackwezi. Died 1912. Mukwenda of County Bugabya, Bunyoro. Resisted British;
put to flight by Capt. Thruston at the storming of Masaja Makura; later came over to British.
Recipient of Mutiny Star for services in assisting to check the Mutiny. Re-appointed Mukwenda,
Bugabya County, where he died. He appears in photographs in Uganda to Khartoum, A. B. Lloyd,
1906, p. 33 and Twilight tales of the black Baganda, Mrs. A. B. Fisher, 1911, p. 195.
Lionel Decle. 1859-1907. Traveller, journalist, author, ethnological collector. French born,
granted British nationality in December 1898. Travelled extensively in Europe, the Far East
and Africa which he visited on three occasions, 1891-1894, 1896-1897, 1899-1901. Addressed
the Royal Colonial Institute in 1906 on his experiences in Uganda. Died in London of pneu-
monia at the early age of 47, 28 February 1907.
Ireta s/o Byangombe. Date of death not known. A Munyankole, he was active in the wars of
succession following the death of Omakama Kamurasi and was captured in Bunyoro by
Kabarega's adherent. Subsequently a loyal supporter of Kabarega, and appointed Chief of
Mboga, Bunyoro. Many skirmishes and actions against British. Surrendered during amnesty
of 1895; reverted to guerilla warfare in 1898; finally gave up in May 1899. A sketch of his
camps near Masindi in northern Buyonro (November 1894) appears in African incidents, A. B.
Thruston, 1900, p. 230.
Jasi s/o Kabarega. Died 1899. Eldest son of Omukama Kabarega. A warrior of considerable
standing. Opposed the British on many occasions. Taken prisoner on 9 April 1899 with his
father and brother Duhaga; died of wounds received in resisting capture.
Sepirsya (Cyprien) Mutagwanya. Died 1954. Had been Luwekula, Buganda for some years
before 1893. Following Buganda Agreement with British Government was confirmed in appoint-
ment of Luwekula on 22 April 1895. Recipient of Mutiny Star awarded for services in assisting
to check the Sudanese Mutiny. Was a witness of the Uganda Agreement, 1900. Retired 1920.
A treasured memento in his possession (1953) was a silver-topped walking stick presented to
him by Sir Hesketh Bell, Governor of Uganda, in 1909.
Rwabudongo. Died 1900. Had been Katikiro to Mukama Kabarega. Opposed British; after
defeat in 1895, fled to Toro. Took advantage of amnesty in same year; later reverted to role of
guerilla in 1898. Finally surrendered in April 1899 and again took up position of authority
under British. His death occurred following a dispute about precedence with his fellow chief,
Ltf. Col. E. ]. Tickell, D.S.O., J.P. 1861-1942. From 1898 to 1901 served in Bunyoro and
then Fashoda, and subsequently in the South African war. For his services in Bunyoro he was
awarded the D.S.O. and the Uganda Medal and clasp. See H. E. Tickell, The Tickllandconnctd
families, 1948, pp. 84-87.
My thanks to Mr. C. S. Busulwa and Mr. A. Tibuhoire whose untiring assistance as inter-
preters in the field has always been invaluable; also to Mr. M.J. Wright for providing additional
data, and to Mr. H. B. Thomas who I have been glad to consult. I am grateful to Major-
General Sir Eustace G. Tickell for placing at my disposal letters and documents, as well as the
photograph of his cousin, the late L~ieut.-Col. E. J. Tickell.
Uganda Journal, 32, 2 (1968), 149-156
THE LOWER MIOCENE FOSSIL SITE OF
B~y A. WALKER
The Miocene fossil site at Lamitina near Bukwa, Sebei was first found in
1965 by Drs. R. Macdonald and R. A. Old of the Uganda Geological Survey.
Macdonald and Old found the hill of sediments that comprises the site when
preparing a geological map of the area; they collected invertebrate and plant
In December 1965 a team led by F. P. Henderson and the writer collected
Dinotherium hobleyi on the south slopes of the hill and that site was designated
Bukwa I. In 1966 the main fossil-bearing stratum was discovered by Dr. W. W.
Bishop and the writer and the gulleys in which it was exposed called Bukwa II.
Surface collection from Bukwa II yielded promising results and in December
1967 a small part of that site was excavated and topographical and local geolog-
ical mapping was carried out in some detail. In June 1968 a second excavation
was made at Bukwa II and at the same time a party of students from the
Uganda Geological Survey under the direction of Dr. P. Brock carried out
more widespread geological mapping and excavated some critical sections.
The site is situated on the northeast slopes of Mt. Elgon (Masaba), 2)
miles east-by-south of Bukwa, Sebei, Uganda (lat. lol7'N., long. 34o47'E).The
fossil site of Bukwa II lies at the foot of a small hill (Kwongori Hill), east of the
village of Lamitina and that of Bukwa I on the south slopes of the same hill.
The base of the main fossiliferous bed is approximately 5,900 feet above sea
Geology (see map 2 and Fig. 1).
The results of mapping in 1965' and of 19682 will be available shortly.
The following general account is based on that work and on the writer's own
A series of subaerial tuffs and intercalated lava flow are seen to fill a base-
ment depression through which the present rivers Chamangeni and Senendet
flow. The oldest sediments to be seen are very coarse boulder beds and are found
in the valley of these two rivers. The boulders are, for the most part, rolled
blocks of lava. These boulder beds appear to be rapid infillings of former river
courses that happen to coincide roughly with the present drainage system. The
succeeding beds are a series of tuffs, conglomerates and basaltic lava flows that
lie either on the boulder beds, where exposed, or on an irregular, lateritized
basement surface. The immediate geology of the fossil site, that is the series of
sediments that comprise the small hill, is as follows.
A low basement hill is seen a few yards to the west of site Bukwa II immedi-
ately west of the track leading to Kabuchai. This basement rise is covered to
the south by a heavily eroded tongue of lava that is seen to extend to the west as
far as Lamitina. Banked against the basement rise and this lava flow are a
series of alternating tuffs and marls that are exposed to a depth of some 22
150 UGANDA JOURNAL
feet. Above these sediments, a boulder bed consisting of spheroidally-weathered
lava boulders dips eastwards from the lava flow, the boulders decreasing in
size but increasing in frequency away from the flow. Succeeding this boulder
bed is the main fossiliferous stratum, a bed of pale green clay that varies in
thickness from a few inches near the flow to 6 or 7 feet at its most easterly exposure.
The top of the green clays is roughly horizontal over all its exposed extent.
The green clay passes upwards into flaggy marls that grade into a series of
tuffs that makes up most of the sediments of the hill. This main series of tuffs
is surmounted by a small cap (14 feet in total thickness) of hard blue lava.
Samples of the capping lava have been radiometrically dated by the
Potassium/Argon method through the kindness of Dr. J. A. Miller of Cambridge
University and Dr. R. L. Armstrong of Yale University. Their results are as
Upper lava Armstrong 20 .1 + 1 .3 x 106 yearS B.P.
Miller 22.0 &t 0.2 x 106 years B.P.
21.9 f 0.2 x IF6 years B.P.
A sample of the lower lava that abuts against the basement, was also dated
by Dr. Armstrong and gave the following result:
Lower lava Armstrong 17.4 f 0.3 x 106 years B.P.
In view of the fact that the two different laboratories both gave closely similar
ages for the capping lava and that there is no evidence for inversion of the
sediments, the date for the lower lava must be regarded as erroneous. A re-
examination of the sample supplied to Dr. Armstrong indicates that there was
a high possibility of Argon leakage and a further sample of a closely associated
but less weathered flow is being sent for further analysis. It is hoped that with
the new sample two dates will be available on the upper and the lower lavas
and hence the age of the enclosed sediments accurately determined. The upper
age limit for the sediments of about 22 million years leaves no doubt that they
and their enclosed fauna and flora are correctly assigned to the Lower Miocene.
The pre-volcanic basement topography was strikingly irregular and in
the Bukwa region had a general slope to the northeast. This slope was dissected
by rivers and steams running to the northeast. Deep weathering surfaces are
found where the basement is not scoured clean. The first evidence of signs of
volcanic activity is the presence of coarse lava boulder beds that are seen to
fill the old pre-volcanic drainage system. Subsequent to this a whole series of
ashes and lavas covered most of the Bukwa area. In some cases there is a great
thickness of ash and in other places the tuffs are seen to abut horizontally against
basement slopes or partly eroded lava flows. In the immediate region of site II
a small western basement rise, together with a slightly more southerly eroded
lava flow, provided the western limits of a small lake. This lake seems to have
been nothing more that a shallow depression in the tuffs, but the presence of a
good aquatic fauna and the fact that a thickness of at least 6 feet of true clays
were laid down indicate that a considerable expanse of water was present. The
base of the fossil bed (i.e. the lake bed) is seen to dip northeastwards and is
covered by rounded pebbles. The basin eventually filled with clays and as the
lake became shallower so the upper marls were laid down. The rest of the
sedimentary sequence seen at Kwongori Hill is a series of subaerial tuffs mostly
horizontally bedded and containing land invertebrates, such as gastropods
and millipedes, as well as several fossil plant horizons including sedge, grass,
flower and leaf beds. There is some evidence to suggest that at certain times
the ash surface was under a shallow sheet of water and that plant debris was
washed in, but the grass and sedge beds have the plants in their positions of
BUKWA FOSSIL SITE
Sketch-map showing location of fossil site Bukwa II with cross section
of simplified geology along the line A-B.
0 10 20 30
Geological sketch-map of the immediate area of the Bukwa II site.
The asterisk marks the site of the main fossiliferous gulley
BUKWA FOSSIL SITE 153
growth, often with the tops of the plants bent and broken by the next ash fall.
It seems most likely that the small capping lava flow, of which only 14 feet
remains on the hill, was not a widespread flow but a narrow tongue. Post-
volcanic erosion has removed nearly all the sedimen, been preserved. The
many of the lava flows in this area. Only by virtue of the protection afforded by
the capping lava have the soft fossil-bearing sediments been preserved. The
present-day drainage system conforms closely to that of the pre-volcanic times.
This could perhaps be accounted for by supposing a rejuvenation of stable
rivers that were outside the control of the more southerly volcanic lavas and
sediments, the pre-volcanic basement mostly being harder than the lavas or
Fauna and Flora.
An extensive fauna and flora has been collected from the Bukwa site and
all specimens are housed in the Uganda Museum. A preliminary account of
the flora is given by Hamilton.3
The faunal list is as follows:
Primates: Limnopithecus legetet Hopwood
Insectivora: Myokyrax oswaldi Andrews
others as yet unidentified
Hyracoidea: Megalohyrax champion Arambourg
Meroehyraxr bataeae Whitworth
Proboscidea: Dinothe~rium hobleyi Andrews
Rhinocerotidae: Chilotheium sp. nov.
Anthracotheriidae: Brachyodus aequatorialis MacInnes
?Hyoboops africanus Andrews
Suidae: Diamantohyus africanus Stromer
?Lystriodon jeanneli Arambourg
Tragulidae: Dorcatherum parvum Whitworth
Dorcatherum figotti Whitworth
large tragulid (not Dorcatherium chappuisi)
Palaeomerycidae: Palaeomeryx sp. (the large species of Whitworth)4
Rodentia: Megapedetes pentadactylus MacInnes
others as yet unidentified
Carnivora: small and medium sized, indeterminate carnivores.
AVEs wading birds
REPTILIA Crocodylia sp.
trionychid and pelomedusid water tortoises
PISCES medium sized and tiny fresh-water fish
Invertebrate fossils include millipedes, potamid crabs, ostracods (species of
Heterocyprisl fresh-water and land gastropods.
To ble a
Species Bukwa Rusinga Karungu Songhor Koru ganu Ombo Napak Moroto Miocene
Limnopithcucs legetet X X X X X
Myokyrax oswaldi X X X ?
fbioni X X X X -- X -
Mereohyrax bataeae X X--- -
Dinotherium hobleyi X ? X 3 3 X
Chilotherium sp.nov. X ? ---
Dicetorhinurs sp. X X X--
torialis X X ? X X X X
Hyoboops africanus ? X X --- X- -
canus X--- -
Lystriodon jeanneli ? X
dactylus X X X X -
Dorcatherium parvum X X X -
Dorcatherium igott X X X X X ?
(large sp.) X X -- -- -- -- -
BUKWA FOSSIL SITE 155
Table I gives a summary of the main elements of the Bukwa fauna coml-
pared with those of other East African Miocene sites. Almost all the mammals
found at Bukwa have also been found on Rusinga Island, but as the Rusinga
sites have been pooled in the most up to date lists it is possible that comparison
with any one Rusinga site would be unfavourable. Nearly all the Rusinga sites
have now been placed in the Hiwegi series and dated at between 17.5 to 18
x 106 years B.P. Site 113 is probably the only one of equivalent age to the
Bukwa deposit The early Miocene site of Karungu compares well with Bukwa
and has been dated at about the same age (22 x 106 years B.P.).
Of especial interest in the Bukwa fauna is the presence of a pig, Diamantohyus
africanus, previously known only from South West Africa and Baluchistan. The
ceratocerhine rhinoceros Chilotherium is known from Africa by an upper third
molar from Loporot, Kenya and by two fragmentary teeth from Rusingas
Plate I is a reconstruction of the left upper premolarjmolar series that have been
recovered from Bukwa. The enormous size of the molars, the great height of the
crowns of all the teeth, the presence of a first upper premolar as indicated by a
facet on the mesial border of the second premolar all serve to place the species
in the genus Chilotherium. The Bukwa teeth and limb bones will certainly enable
a diagnosis to be made.
After excavation with the trowel the fossiliferous green clay was broken
down in water and passed through fine sieves in order to recover the smaller
specimens. Only clay and ostracods passed through the sieve and small teeth
and bones remained. Some of the rodent and insectivore teeth are of such small
size that it is doubtful whether the species from which they come have been
recorded previously from the East African Miocene.
The value of the Bukwa fauna lies in the fact that it is from one of the earliest
Miocene sites in Africa, and that the nature of the sediments allows a repre-
sentative sample of the fauna from the largest to the smallest mammals to be
collected. It is apparent that each of the East African Miocene volcanoes had
a forest fauna that, by analogy with present conditions, should be in some ways
unique to each mountain mass. This, the first mammalian fossil site on Mt.
Elgon, should enable more easy comparison with the faunas of the separated
mountain masses of Napak and Moroto in Uganda and the Kavirondo mountain
(Rangwa) of Kenya, for it is situated midway between the two areas. It is hoped
that future excavation will lead to a greater knowledge of the fauna and flora.
Meanwhile the better specimens are being made available to the relevant
experts for study.
I would like to acknowledge the financial support given by the Ministry
of Culture and Community Development, Kampala; the Trustees of the Boise
Fund, Oxford University; the Makerere University College Research Grants
Committee and Professor and Mrs. D. R. Pilbeam of Yale University. I am
greatly indebted to the following people who have helped at different stages in
the investigation of the Bukwa site: Professor R. L. Armstrong, Dr. W. W.
Bishop, P. Boston, Dr. P. Brock, P. Y. Bulenzi, Mrs. S. Coryndon, Professor and
Mrs. Haines, A. Hamilton, A. Hill, D. Kiyaga, J. Lubega, Y. Lule, Dr. R. Mac-
donald, P. Mbonye, Dr. J. Miller, J. Moyini, A. Minawa, J. Nzabonimpa,
P. Ogwang, W. Opiyo, Professor and Mrs. D. R. Pilbeam, Dr. and Mrs.
M. D. Rose, C. Sekintu, Malkit SinghDiocee, G. Smith, C. Ssali, Professor and
Mrs. E. E. Suckling, A. Suni, Mr. and Mrs. S. Tomkins, Mr. and Mrs. J. White,
R. White, W. White and J. Williams.
156 UGANDA JOURNAL
1. Macdonald. R, and Old, R.A. (In press--Records of the Geological Survey).
2. Brock, P.G.W. and Macdonald, R.A. Geology of the area surrounding the Bukwa mammalia
fossil site, Sebei. (In press-JJlahure, 1968.)
3. Hamilton, A. Some plant fossils from Bukwa. (Uganda J., 32, 1968, pp. 157-164,--this
4. Whitworth, T. Fossil mammals of Africa, No. 15. Miocene ruminants of East Africa. London,
British Museum (Natural History) 1958, pp. 1-48.
5. Bishop, W.W. The later Tertiary in East Africa; volcanics, sediments and faunal inventory.
Essay in Background to evolution inl Africa edited by W.W. Bishop and J.D. Clark, 1967, pp.
6. Van Couvering, J.A. and Miller, J.A. Miocene stratigraphy and age determination, Rusinga
island, Kenya. (In press-Nature 1968).
7. Bishop, W.W. and Miller, J.A. (In press).
8. Hooijer, D.A. Fossil mammals of Africa, No. 21. Miocene Rhinoceroses of East Africa.
London, British Museum (Natural History), 1966, pp. 119-190.
Figure to illustrate Bukwa fossil site, page 155.
Reconstruction of the upper molar/premolar series of the Bukwa Chilotherium
with a third molar outline added as reconstructed from Loporot, Kenya
Uganda Journal, 32, 2 (1968) 157--164
SOME PLANT FOSSILS FROM BUKWA
By A. HAMILTON
Several types of fossil plants have been found at Bukwa. The flora contains
determinable flowers, which, being of very rare occurrence elsewhere, make
the site of particular palaeobotanical interest. Associated with the flora is an
important fauna, and it is hoped that future workers will be able to build up
from the fauna and flora a picture of the environment of the time.
The best known Miocene flora in East Africa has been described from
Rusinga Island by Chesters (1957). As at Bukwa, plant and animal fossils are
associated. Many fruits and seeds have been identified. On Elgon, several sites
containing plant fossils have been described from volcanic rocks or from sedi-
mentary rocks underlying the volcanics. Fossil wood is abundant on Mr. Elgon*
Visitors to caves and travellers along the mountain road in Sebei often collect
pieces as souvenirs. Bancroft (1935) considered some specimens to be of Dipter-
Ocarp (group Dipterocarpoideae) affinities. This is of phytogeographical
significance since the Dipterocarpoideae are tall trees characteristic of rain-
forests in Southeast Asia and are unknown in Africa at the present time. Chaney
(1933) examined plant fossils from sandstones of the Bugishu series, which in
places underlie the Elgon volcanics, and reported the presence of fossil wood,
leaves and fruits. Twelve plants were identified and these constitute an assemblage
which Philips (Chaney 1933) considers might have come from an "Open
Woodland" or "Woodland", of a type common in East Africa today, where
the rainfall is from 25 to 50 inches per annum. In a letter to Dr. Lind of the
Botany Department, Makerere, Wayland (1959) mentions finding the stem ofa
fossil tree-fern, Cyathea, on Elgon, lying at about 2,150 metres. The annual
reports of the Geological Survey of Uganda (1922, 1932, 1933) record fossil
wood and leaves on Elgon.
The fossils at Bukwa fall into two assemblages; those for which there is no
evidence of derivation from plants growing in situ, and those that have been
preserved in their positions of growth. The first includes the fossil wood and
the fossils from the various leaf beds. As little or no internal structure is present,
the fossils wood is not identifiable. However, the leaf beds contain abundant
leaves, fruits and flowers, some of which are moderately well preserved. The
second assemblage includes various monocotelydonous remains; the Grass
Bed described below contains fossils of this type.
The fossils are preserved as casts partly filled with calcite. Little anatomical
structure can be distinguished. Treatment with hydrochloric acid dissolves
the calcite and leaves, in some examples, an insoluble residue containing thin
sheets of pigmented crystalline material on which structures reminiscent of
cell-walls can be seen. Immersion of whole fossil flowers in concentrated hydrol
chloric acid for about 15 minutes "clears" the calcite and enables interna-
structures to be seen through external. Longitudinal and transverse sections of
the flowers were cut, and some of these show the flora organs in some detail.
No pollen has been found in either the basal clay or in the plant fossil beds.
158 UGANDA JOURNAL
Description of individual sites and of fossils
1. UPPER LEAF BED
The fossiliferous horizon is about 5 cm. thick. Abundant fossil stems,
leaves, flowers, and fruits are preserved in a fine-grained calcareous ash. About
6 square metres of this deposit were exposed, but the boundaries of the fossili-
ferous beds were not reached.
Stems--Fossil stems are common. The largest found was 1 .5 cm in diameter.
As no internal structure is preserved, they are not identifiable.
Leaves--At least four types of dicotyledonous leaves are present. Of these,
three are simple leaves. As they are of types common amongst Ugandan trees
today, they are not identified. The fourth type of leaf is, however, of a shape
uncommon in Ugandan plants today, being particularly characteristic of some
tree genera of the Euphorbiaceae and Sterculiaceae (e.g. Cola). The petiole is
long (8-13 cm) and is divided at the base of the lamina, into 6, 7 (to 10 ?) main
veins. Each of these veins leads to a leaf-lobe, the largest (13 cm) of which is
opposite the petiole with those on each side decreasing steadily in length. The
leaf-lobes are fused to their neighbours about half way along their length. The
venation is prominent, being at right angles to the main veins in the lower part
of the lamina.
Flowers and fruits.--Several types of fruits, flowers and flower-buds are
present on this horizon. Six are described below.
(a) Small zygomorphic flower. (Figs. 2-6) This is by far the most abundant
flower in the Upper Leaf Bed. Its structure is complex and no specimen shows
more than a few aspects of the floral arrangement clearly. The most noticeable
feature is the staminal tube and this, together with other features, such as the
zygomorphy and the few stamens on their thick filaments, suggest that this
plant is a member of the Melianthaceae. If this is so, then judging from Thonner
(1915) and Verdcourt (1958), it is closest to Bersama abyssinica Fresen. The
specimens differ from material of B. abyssinica in being comparatively narrow,
but otherwise their external appearance and their morphology agree. B.
abyssinica displays, incidentally, 'quite remarkable variation' which is 'sometimes
so intense that it is seldom that two populations are found which are exactly
alike' (Verdcourt loc. cit p. 1).
Bersama aby1ssinica is a tree, usually up to 10 m tall, found in a variety of
habitats. It occurs in rain-forest, forest-edges, riparian vegetation and grass-
land from 1,100 to 2,800 m.
Pedicel 1.2--1.8 mm long, about I mm in diameter, suddenly expanded into a flat-topped
receptacle. Sepals 5 (possibly variable), thick, unequal, with a maximum length of 7 mm
and occasionally partially fused for about half of their length. The corolla is indistinct on
most specimens, being thinner and more fragile than the calyx. There are probably five
free petals, alternating with the sepals. The petals project about 3 mm above the sepals,
and the stamens about 3 mm above the petals. The thick filaments are united in a con-
spicuous tube for part of their length, but are free above and probably free below. Specimen
UMP. 68.64 (Fig. 4) is probably a receptacle bearing a pentagonal disc, four staminal
bases and a young gynoecium. Four stamens can be seen in some specimens and there is no
conclusive evidence that more are present. The anthers are about I mm long, dorsifixed
and two-lobed, each lobe having a single longitudinalslit. The ovary is probabliysuperior,
rounded and with a single terminal style.
(b) Small regular flower. (Fig. 1) This small flower, with its five thick, part-
ially-fused sepals and its staminal tube bearing two whorls of nearly sessile
stamens, is probably a member of the Sterculiaceae and, judging from German
(1963), and Hutchinson and Dalziel (1927), from either the genus Cola or
Pteryg~ota. Of the Ugandan representatives of these genera today, only Cola
BUKWA FOSSIL PLANTS 159
ordifolia (Cay.) R. Br. matches the flower structure closely.
Cola and Ptergota are both, in the main, forest trees and both are much
better represented in the Congo basin and in West Africa than in Uganda.
C. cordifolia is a widespread Ugandan tree, up to 35 m. tall, occurring in rain-
forests below 1,500m.
Pedicel about 2.5 mm long by 2 mm diameter. Calyx-tube about 6 mm in diameter and
3.5 mm tall, consisting of five thick triangular segments, fused for about 2/3 of their length.
In specimen UMP. 68.54 where one side of the calyx-tube is broken, a central structure
can be seen in the centre of the calyx-tube. This is c. 3.5 mm in diameter and 2 mm tall,
and is attached by a very short stalk I mm in diameter. On the central structure are a
number of (predominantly) bi-lobed anthers, which seem to be arranged in two whorls.
Each anther is about I mm long and 0.5 mm across, the upper whorl being connected-
to the central structure by thick filaments about 0.8 mm long. At the top of the central struc-
ture is a slight depression containing a small mound, which is divided into four or five
parts of uncertain morphology.
(c) Srnall "Knobbly"fruit. This fruit is the commonest in this horizon. The
large number of vertical septae and the mode of dehiscence into mericarps
indicate that it may belong to the order Malvales.
Specimens UMP. 68.54-57 are slightly depressed spherical in shape, about 7 mm long and
8 mm in diameter and divided by vertical constrictions into about 12 compartments.
Each of these is again divided by transverse constrictions into 6-8 parts. Specimen UMP.
68.58 is longer than it is broad, being 5 mm long and 3.5 mm in diameter. It is divided into
six vertical compartments, each restricted into 4-5 parts. Fragments of this fruit show
that dehiscense is along both the vertical and transverse septae into (probably) one-seeded
(d) Fruit. This is one third of a trilocular fruit. It is 10 mm long with a
maximum width of 7 mm. Several families (e.g. Euphobriaceae) have fruits
of this type.
(e) Flower. This flower consists of a short, four lobed calyx-tube, 4 mm
tall and 5 mm in diameter, and four free petals alternating with the calyx lobes
and projecting 6 mm above them. Its identification is uncertain.
(f) Flower. This zygomorphic flower has a fused calyx of two short members
(4 mm long) and two (three?) long members, (6 mm long). The corolla is
tubular, of fire (?) fused members, and projects up to 4.5 above the calyx.
A number of structures, possibly including stamens are visible within the corolla-
tube. Its identification is uncertain.
2. MIDDLE LEAF BED
The matrix of this bed consists of an ash of a coarser texture than that of
the Upper Leaf Bed. Unfortunately, this bed proved to be very friable, and
fractured badly, so that collection of good specimens was impossible. Fossils
found in this horizon include dicotyledonous leaves, possibly a pod of a member
of the Leguminoseae, casts of snail shells and a tooth of M~yokyrax.
3. LOWER LEAF BED
This was found by Dr. Walker and Professor Haines in January, 1966, but
could not be re-located. The ephemeral occurrence of this leaf bed, which
from reports was of a similar type to the Upper Leaf Bed, suggests that leaf
beds are of a patchy occurrence in any one horizon.
4. GRAss LEAF BED
This bed is a fine-grained ash about I m thick, containing abundant
monocotyledonous fossils. In the lower part of the deposit, the plant remains
lie predominantly in a horizontal position, but, in the upper part, they are
orientated more or less vertically. In some cases, the fossils can be traced from
root-systems in the lower part of the bed to leaves in the upper, indicating that
160 UGANDA JOURNAL
these fossils are preserved in their positions of growth. The distribution of
various fossils in this bed is not even, different types predominating in different
Probably the commonest fossil overall in this bed is a grass. It has a height
of 1 m or more. The roots were not located, but it seems from the orientation
of the lower parts of the shoots that the plant was not, or only a little, tufted.
The largest stems found were 3 mm in diameter, being very slightly, to dis-
tinctly, oval in section. In one specimen, it can be seen that the leaf sheath is
about 6 cm long and that the leaves are set at intervals of about 3 cm. No ligules
were found. The length of the leaves is probably well over 15 cm, with a maximum
width of about 9 mm. The ends of the leaves taper gradually to fine points. In
most specimens, no clearly defined central nerve is visible, but in some, there
are indications that one or perhaps several leaf-veins are more prominent
than the others. There are many extant grasses with similar morphology.
Abundant in parts of this bed is an unidentified monocotyledon, consist-
ing of numerous shoots arising from a densely tufted stock. The shoots are
circular in section and about 2 mm in diameter. Professor Haines has found and
identified the rhizomes ofJuncellus laevigatus (L.) C.B.CL. on this and other
horizons. This is one of the commonest plants occurring on the margins of
alkaline lakes in East Africa, either growing partly under water or on land
close to the lake shore.
The volcanic sediments at Bukwa contain identifiable fossils. There
are two distinct plant assemblages present. The first, for which there is no direct
evidence of derivation from locally-growing plants, is represented mainly by
the leaf beds. The presence of wood fragments, the size of the leaves, the abund-
ance of f lowers and the absence of monocotelydonous remains indicate a forest
origin for these flowers. The "identifications" made also support this hypothesis.
The second assemblage consists, of plants preserved in their positions of growth.
If fucellus laevigatus is correctly identified, then at least some of these plants
were growing in, or near the margins of, an alkaline lake. Present-day examples
of such lakes in East Africa are Lakes Katwe, Nakuru and Manyara. Because
species found in alkaline lakes are adapted to rather peculiar ecological condi-
tions, it is unlikely that further research on fossils of this assemblage will give
much information on the overall vegetation at Bukwa during the period of
I would like to thank Mr. Boston and Professor R. W. Haines for help
with the extraction of these fossils. Dr. P. Dixon kindly prepared most of the
fossil flower section. Methods for studying the fossil flowers come largely from
the ideas of Mr. T. R. Milburn. Dr. C. Ferrira, Dr. M. E. S. Morrison and
Miss A. C. Tallantire helped with the identifications.
Annual Reports of the Geological Survey, of Uganda, 1922; 1932; 1933; Entebbe, Government Printer.
Bancroft, H. The Dipterocarps in Africa. Empire Forestry. Journal 14, 1935, pp. 74-75.
Chaney, R. W. A Tertiary flora from Uganda. ournal of~eology Chicago. 41, 1933 pp. 702-709.
Chesters, K. I. M. The Miocene flora of Rusinga Island, Lake Victoria, Kenya. Palaeonfographica,
101, 1957, pp. 30-71.
German, R, 1963 Flore dur Congo, du Rwranda et du Burundi. Vol. 10. Bruxelles, 1963.
Hutchinson, J. and Dalziel 1927, Flora of west tropical Africa. Vol. 1 London,
Thonner, Fr. The flowering plants of Africa. London, 1915.
Verdcourt B. The flora of tropical East Africa, the Melianthaceae. London, Crown Agents, 1958.
BUKWA FOSSIL PLANTS 161
Fig. 1 UMP.68.54 Staminal tube of a member of the (?) Steruliaceae.
A. View from above.
B. View from the side.
Fig. 2 UMP.68.65 Filaments and anthers of a member of the (?) Melia-
Fig. 3 UMP.68.65 Posterior (A) and anterior (B) views of an anther
showing attachment of the filament and lines of
Fig. 4 UMP.68.64 The receptacle and associated structures of a member
of the (?) Melianthaceae. See text for description.
Fig. 5 UMP.68.93 Two views of the flower of a member of the ( ?) Melian-
thaceae. Sepals not lettered. Petals lettered A-E. F
is probably the staminal tube. G is probably the
Fig. 6 UMP.68.94 Two views of the flower of a member of the ( ?) Melian-
thaceae. Showing staminal tube (shaded in 6A) and
petals (shaded in 6B) lying behind the sepals.
Uganda Journal, 32, 2 (1968), 165--182
THE EXCAVATION OF AN ANK(OLE CAPITAL SITE
By M. POSNANSKY
In 1959 on behalf of the Uganda Museum the writer excavated the
Ankole royal capital site of Bweyorere, 17 miles southwest of Mbarara and a
little over three quarters of a mile from the Mbarara-Kikagati road along the
Omwibare borehole track (fig. 1). The site was located by Professor R. Oliver
in 19581 during a survey of the royal capitals of Ankole and was selected for
excavation because of the presence of extensive earthworks and surface pottery.
A preliminary account of the excavation was published in 1962 following an
excavation from June I to July 7, 1959.2
According to the traditions of the Banyankole, Bweyorere had been used
as a royal settlement several times and one of the aims of the excavation was to
ascertain whether the traditions are a reliable guide for archaeology in Uganda.
In this respect the excavation was highly successful in that at least two phases
of settlement were indicated and the date provided by the traditions was
confirmed by two rather ambiguous radiocarbon determinations. Though not
exactly a "compact town site"3 Bweyorere represents a large pastoral settle-
ment with evidence of a palace site far larger than those encountered by the
nineteenth century European travellers to Ankole and with the earliest trade
goods yet known from the lacustrine region.
The Site (fig. z)4
The site consists of a series of low banks stretching for a quarter of a mile
along the western side of the top of a low hill in the undulating Masha plain,
less than a mile from the steep slopes of the Isingiro hills. In all there are
forty-nine banks, mostly slightly arcuate and rarely more than three feet
high. The banks, in groups of twos and threes form small enclosures open
towards the centre of the hill. A larger enclosure, about 150 by 100 feet across,
is made up of five large banks each up to four feet high which is clearly inter-
pretable as the ekyikanis, or royal enclosure which would normally be found at
the centre of an Ankole royal kraal or orurernbo. None of the smaller enclosures
directly face into the large enclosure. The plan of the banks seems to fall into
two units. One with the ekyikari and a string of eight smaller more open
enclosures immediately behind and to the south of the ekyikari. The other to
the north of the ekyikari is smaller, but constitutes a better defined enclosure.
This second area contains a particularly large bank over forty yards long and
nearly as high as the ekyikari banks. It is not inconceivable that this northern
group of banks formed a separate orurembo from the larger one to the south.
Most of the banks are deeply riddled with rodent burrows and from the
upcast and pitting it was readily apparent that the banks contained a large
amount of cattle dung. There were no ditches around the banks. &elow the
banks on the western slope of the hill slight terracing caused by former cultiva-
tion was observable.
BWEYORERE EXCAVATION 167
Besides the banks, nine hollows were picked out, three of which lay behind
the large enclosure. The largest of these was 25 by 15 feet across and two
A series of 31 staggered, five feet wide trenches were dug across the central
enclosure and through two of its banks, across four of the other enclosures
and two of the hollows. All the trenches were dug down into the natural
undisturbed bedrock which consisted of either highly weathered shales and
pink coloured phyllites of the Karagwe-Ankolean series or to an overlying
'laterized' murram. In all the cuts, the buried ground surface was easy to
pick out on top of one to three feet of brown compact stoneless soil and was
often marked by a compacted turf line. The trenches across the ekyikari enclo-
sure revealed that the earth for the banks had in part been scraped up from
the area within the surrounding banks as nowhere was there more than six
to nine inches of soil above the 'natural; whereas test pits outside the enclosure
revealed a soil profile of up to two feet. There was no trace of any super-
structure either along the banks, in the gaps between the banks or across the
entrances. This is not surprising as the mound soil is in all cases loose, highly
disturbed by rodent action and if thorn bush branches had been used for a
fence, as is the present-day Bahima practice, no evidence of them would
The sections (figure 2) indicate that the smaller enclosures (sections '12
and 26) were largely made up of soil and dung, distinguishable by its light
weight, whiteness and lenticular structure,6 Which suggests that they were
thrown up during the use of the enclosure. It is the present Bahima practice
to heap dung around the enclosure against the thorn bush fence with the
houses on the inner side of the dung heap (Plate 1). The large banks of the
ekyikari (sections 1 and 11) contained far less dung and it was evident that the
earth had been intentionally thrown up in order to give added height to the
normal accretion of dung. What earth lensing there was, suggested that they
had been thrown up from the centre of the enclosure which explains the
absence of top soil within.
In all cases the buried ground surface beneath the banks indicated settle-
ment prior to the erection of the banks, with more than a third of the pottery
coming from undisturbed contexts beneath the bank. In addition to the pottery,
there was burnt bone and traces of fire and burnt dung all clearly predating
the erection of the large mounds. It is possible, though, that an initial occupa-
tion took place and small banks of dung and earth were thrown up to be followed
by the rather more substantial banks indicated in section 1.
Trench I and its extension (figure z) revealed on the old ground surface
a series of 15 post holes (Plate 2) in two groups. One group formed a rectangle,
five feet by four, in which there were four large post holes from 12--18 inches
in diameter and 18-24 inches deep. Between and around three of these larger
post holes were eight smaller holes about six inches in diameter. The three
other post holes were situated twenty five feet to the east and were also about
six inches in diameter. To one side of the large post holes was an irregular
area of ash, four feet by six at its widest extent. This covered a regular oval
area of reddened ground two feet by three. It was probably a hearth and the
reddened area the actual fireplace. The deep discoloration of the soil indicates
that this had been a hearth for some considerable time and was not just a
temporary fire, which would perhaps blacken the surface of the soil but not
168 UGANDA JOURNAL
cause deep soil oxidation. A turquoise-blue, spherical, cane-glass bead was
found to one side of the hearth.
On the old ground surface a large amount of ash and charcoal was found,
above which was a mass of burnt dung, red in colour, in places up to a foot
thick. The red dung can often be seen on top of the banks brought up by
burrowing creatures. When plotted in trenches 1, 15 and 31 (the square to the
north of Trench 1), it was apparent that burnt dung covered a circular area a
little over 50 feet in diameter of which the square of post holes was at the
true centre with the three isolated post holes at the perimeter. It was noticeable
(section 15) that the greatest thickness of the burnt red dung was situated
towards the edge of the circle forming a ring. The evidences suggest a circular
structure, made of organic material, on which a large amount of dung had
been plastered. In such an interpretation, the large post holes at the centre
would be structural posts supporting the building and the smaller perimeter
post holes would peg down the roof, a feature of traditional western lacus-
trine Bantu houses and typical of pre-colonial Buganda in which walls and
roof were of one piece forming a huge pointed dome. Structurally, such build-
ings are like upturned baskets as the grass is worked as in the making of a
round basket. Had the structure been of any other form, the perimeter walls
would have had to support a greater weight and consequently would have
been larger. The presence of plastered dung is rare in such buildings but
with a predominantly pastoral people it is not surprising and is a practices
employed on the humbler homes of peoples like the Masai. The post hole
contained both ash and charcoal indicating destruction by fire rather than
the removal or slow decay of the posts.
As the bank covered the post holes and attendant features it was clearly
a later erection, though how much later it is impossible to tell as there were
no recognisable soil surfaces or turf lines in the mound. The cistern to the
west of the bank was also clearly later as it cut through the western perimeter
of the circle of the presumed building.
In other cuttings, though the buried ground surface contained pottery
or bone there were no other recognizable habitation' features. Settlement did
occur in an earlier phase but it was nowhere as evident as beneath the high
banks of the ekyikari. As the pottery was neither very abundant on the whole
site nor showed significant differences between the layers, it is impossible to
say for certain which enclosures were contemporary with the ekyikari.
Two trenches (12 and 27) through the hollows revealed that they were
cut between four and five feet deep into the weathered rock. The marks of the
two and half inch wide blades, presumably of iron, which had been used for
digging them were apparent around the relatively steep sides of the hollow.
From the angle at which the blows had been struck in their excavation it was
clear that the blades must have been hafted in some kind of handle with an
elbow-bend like a hoe shaft. The hollows must have been dug for cisterns and
when full after the rains would have contained upwards of 20,000 gallons
between them. They were certainly not dug as sources of earth for the banks,
as their upcast is immediately around, and, in the case of the one sectioned in
Trench I1, to the southern end of the hollow and not on the banks.
A series of samples were taken from the buried ground surfaces for pollen
investigation, but the soils Proved to be too aerobic for pollen preservation.
Two charcoal samples were submitted for isotopic dating. The first one
170 UGANDA JOURNAL
from one of the post holes in Trench 1, was sent to Geochron and a date of
1640 f 95 A.D. (GXO 520) was obtained. The other was from the buried
ground surface below the bank in Trench I and was sent to Yale University
Radiocarbon Laboratory and provided a date of 1820 f 80 A.D. (Y.-1394).
Taking the statistical error into account, which means that there are only two
chances in three that the date actually falls within the plus and minus range,
it can be seen that for post 1500 A.D. sites radiocarbon dates provide little
accurate dating evidence as their methodological standard of error allows a
dating spread over at least a hundred and fifty years. It could be inferred that
as the sample from the post hole provides a mid-seventeenth century A.D.
mean date it is acceptable, whilst the sample from the ground surface could
belong to the end of a period of occupation not necessarily contemporaneous
with the post holes. Archaeologically there is very little evidence to separate
the two samples and for dates falling within the seventeenth century the de
Vries effect, related to the variation in the natural concentration of radio-
active CO<, can allow at least three widely differing dates for samples of the
Pottery--Altogether a total of 814 sherds were found of which 700/ were
undecorated. A large number came from the mound soil and presumably
represent broken pottery thrown up in the course of occupation. Virtually no
pottery came from the enclosures which must have been scraped periodically
of dung and occupation detritus. The amount of pottery was surprisingly
small considering the large number of trenches excavated, though it accords
with the paucity of pottery normally found on a pastoral site. A far larger
excavation at Bigo yielded 3459 sherds, whilst at excavations at Lanet in
Kenya only 400 sherds were found. It is presumed that the occupants made use
of calabashes and wooden milk pots and pottery was employed only for cooking,
serving and for drinking vessels.
The most distinctive characteristic of the pottery (figures 3-7) is the pre-
valence of decoration with knotted-grass roulettes, generally around the
upper part of the neck, though occasionally on the inside of the neck (fig. 5, 6)
or on the top edge of the rim (fig. 5, 12). Few large pots such as would be
used for storage were found (fig. 4) and only a handful of what could be
construed as water jars (fig. 4, 4-6). One massive base of an undecorated pot
stand, or fumigator (fig. 4, 3), was found. Most of the collection consisted of
parts of relatively small bowls hemispherical; spherical with banded everted
rims (as fig. 6, 2, a form common at Bigo);i and dish-like bowls (fig. 5, 2, 6, 9).
As hardly any two vessels were alike it is impossible to describe the whole
collection in detail though the number of forms, as distinct from minor varia-
tions of form, was limited. The pottery was well-fired, mainly using broken
pot as temper and varied in colour from red, through greys to nearly black,
depending on the amount of oxidation in firing.
From the old ground surface, entirely in the ekyikari area, were found
forms which are not present at all in Bigo culture contexts; pots with very
constricted necks (fig. 3, 1, 4, 5) reproduced in pottery, invariably black, the
from of the cultivated calabash. Calabashes if left to grow naturally have a
nick which bends as the calabash fruit gains in weight. If the fruit is supported,
the neck can be elongated and artificially constricted and such long necked
calabashes are still in great demand in Ankole for beer containers. Several of
the narrow rims or neck fragments separately found (e.g. fig. 7, 2, 7, 8) pro-
BWEYORERE EXCAVATION 171
bably form parts of the necks of such flasks. The decoration on these calabash
forms, though rouletted, is arranged in bands (fig. 3, 4) or in large chevrons
(fig. 3, 1, 5; fig. 6, 5) occasionally haphazardly arranged in bands (fig. 6,
10, 12), and it is tempting to see in some of these a copying of the string bags
used for hanging milk pots in a Hima hut. These black calabash forms may
owe their ultimate origin to Kinyoro prototypes where shiny surfaced, graphite
slip, calabash forms were made for the abakama. In Buganda, from the
eighteenth century, similar black pottery, probably also in imitation of the
Kinyoro, were was made. In a history of raiding from Bunyoro, and with
numerous occasions when a tributary subservience to the major kingdom in
western Uganda was politic, it is not surprising that influences from Bunyoro
should appear. Only one small piece of pottery definitely from Bunyoro was
found; a piece with a graphite slip which could conceivably have been a
royal gift, since sources of graphite are geographically restricted to the Lake
Albert region of Bunyoro.
The finest pottery, thin, highly burnished, often with a light brown slip,
well-fired and made into simple symmetrical calabash forms (fig. 3, 2, 6, 1)
and mostly with slightly beaded rims, has only been found at capital sites and
must represent a royal Ankole ware. Similar royal wares have been in Urundi
and Rwanda and are displayed in the Butari Museum. A few pieces of excep-
tionally fine bowls (fig. 6, 11) had close incised decoration below the rim.
Also found at Bweyorere (fig. 3, 3), though always in late contexts, on
the banks and never on the old ground surface, are fragments of what Hier-
naux calls Boudine ware.'o It is of a coarse fabric and the clay has been folded
over in much the same way as pie pastry. It is always grey black in colour
with a rough surface texture. Hiernaux described it as associated with earlier
wares including dimple-based pottery and it could well represent a local
survival of a pre-Bigo ware. It has been found in the Kagera valley and on
Parts of five pipes were found (fig. 7, 1, 3, 4, 5) all had straight sided
bowls with finely executed decoration below with in one case a beaded rim
and in two cases bevelled rims. The bowls were rounded at the base and the
angle between the bowl and the stem was about 450. Insufficient pipes have
been found in Uganda to attempt a typology, though on West African analogy,
where pipes provide a valuable dating medium, such simple forms would be
thought of as early, as they follow what must originally have been European
Two round cylindrical pieces of baked clay were found (fig. 7, 10-11)
which may have been handles or lugs.
Stone--Two grindstones were found, one of which was on the old ground
surface of Trench 1. This indicates that some millet grinding may have been
undertaken, being the work of Bairu serfs; though for such a large area the
number is insignificant compared with sites such as Lolui in eastern Uganda
where literally thousands of querns have been found in an area the size of
Several pieces of ground down haematite (ochre) were found and must
have been used for red colouring as well. Small lumps of kaolin occurred and
kaolin is still used for whitening drums, cattle horns, bodies and making
symbols on milk-hut walls.
Beads--Four beads were found, two turquoise blue of re-melted cane-
172 UGANDA JOURNAL
glass (6 mm. long) and two much smaller (3 mm.) pearly white spherical beads.
The blue beads were found on the old ground surface of Trench I by the
hearth and the other in the cistern sectioned by Trench 27. The two white
beads were found on the old ground surface of Trench 11. They are far too
indistinctive to be dated, even in relation to the East African coast where
similar turquoise blue beads are fairly time-transgressive and can hail from
sources as far removed as Venice and India. Only by spectrographic analysis
of the original glass used in the manufacturing could their provenance be
more precisely located.
Iron--Though traces of iron were found, all the objects were in the last
stages of disintegration, being masses of rust with no iron core left. The frag-
ments suggest parts of arrows, knives, rings and armlets though none were
reconstructable or illustratable.
Bones--Dr. B. H. Fagan, then of the Livingstone Museum in Zambia,
studied the bones, but was only able to identify species from teeth, horn cores
and mandibles. Limb bones which are not so susceptible to identification he
sub-divided into small, medium and large bovidae; with domestic cattle being
grouped among the large bovidae, whilst sheep and goats were classified with
the small bovidae. Out of a total of 367 bones, 163 were classified; of which
31% belong to the large group, 45% belong to the medium group, and 24%
to the small. The following species were identified on the basis of teeth, horn
cores or characteristic bones:
Domestic ox:-teeth, horn cores, mandible.
Zebra:-1l complete metacarpal.
Haare:-Pelvis, scapulae and tibia
(?) Eland:--3 doubtful immature upper molars.12
Dr. Fagan writes that; "it is probable that a good proportion of the limb bones
are from game species, rather than domestic animals. There seems indeed to
be a contrast between Bigo and this later site, where hunting was more
common; if this small sample can be relied upon. Neither small stock nor
dogs could be identified positively. Immature bones are common, and of the
four cattle jaws found all are of young beasts and one is very young indeed's
It may be that hunting was concentrated on immature or old beasts. No
deduction can be drawn about the species or size of the cattle from the site".
One whole cow horn core was found and was over eighteen inches
long, suggesting that the long-horned cattle, for which Ankole is still famous,
were then present. Unfortunately this specimen (Plate 3) was very fragmentary.
The absence of sheep or goats is not surprising on a Hima site as sheep are
not kept for food, but only for good luck" whilst goats are of relatively recent
Bweyorere is mentioned three times as a royal capital in Katete's version
of the Ankole royal traditions collected by Nuwa Mabagatu and L. Kamungu-
ngu.'5 The first occupation was of Omugabe Kasasira who lived some twelve
generations back from the present; later of Karara, a son of Macwa whose
brother Omugabe Kahaya I, was a contemporary of Kigeri III of Rwanda who
died in either 1781 or 1792;'6 and lastly of Omugabe Rwebishengye who was a
contemporary of K~abaka Kamanya of Buganda who reigned sometime in the
period 1810-30. The only recorded details of the occupations concerns Karara,
who when murdered by his brother and immediate successor, Karaiga, was
BWEYORERE EXCAVATION 173
burnt in his own palace by his grief-stricken sister, the Rubuga Karuganda. The
evidence thus indicates three occupations; in the mid-seventeenth century; in
the last part of the first quarter of the eighteenth century, and in the latter
part of the first quarter of the nineteenth century. The first occupation is
datable by reference to calculations based on a generation of 27 yearsl7 and
the later occupations by cross-dating with the genealogies of Rwanda and
The first two abagabe mentioned are credited with no other capital site;
but Rwebishengye had four others. This is perhaps not so significant as pre-
viously thought'" as most of the later arbagabe are mentioned as having several
capital sites; whilst it is probable that only the most significant sites were
remembered for the earlier rules. The physical nature of Bahima kraals makes
it unlikely that one kraal could serve for more than a few years.
There are strong local traditions and also the view of the then Omugabe,
that Bweyorere was also a capital of Omugabe Macwa.20 There is no firm way
of confirming this ascription or even of deciding from archaeological evidence
which of the three possible occupations of Karara, Macwa and Rwebishengye
is to be preferred, since all three reigned within a period of three generations;
that is about half the normal variation provided by a radiocarbon determina-
tion which is presently the most accurate way of dating Uganda's Iron Age
sites. It is also conceivable that all four abagabe lived at Bweyorere.
From the traditions one can also get clues as to the cisterns. Cwarnali of
Bunyoro is reported to have dug reservoirs in Ankole and Omugabe Kahaya to
have made bathing wells at Bugaba. The cisterns could be interpreted as more
permanent cattle troughs with water drawn from the nearby Rubaga stream,
but it is of interest that the idea of digging holes to contain water was not
Both the excavation and the traditions indicate that Bweyorere was a
site occupied more than once. Basically it was a large Bahima kraal and like
present day ones, it had dung heaps wherever possible on the lower sides and
entrances on the upper.21 The houses, of which there is now no trace, pre-
sumably had grass floors and withy walls. Though large in area there is no
need to assume either a large population, probably no more than a few
hundred people, nor even occupation over a considerable period at any one
time. Even though the abagabe were pastoralists, the relative paucity of pottery
and bone would argue against Oliver's idea that a "single site was occupied
for a whole reign,"22 unless of course the reign was short. The bone evidence
indicates that game was hunted, a practice still attested to in the early twentieth
It was not a town and there is no trace of the practice of any crafts. It
is difficult to know whether the surface plan of the site represents one or more
capital, though certainly at least two are evidenced. The large structure,
loosely labelled 'palace' on the plan, could represent the large meeting hut of
a mugabe (the nyarunzhu rweterekyeTO24 or rwemihunda.25 The evident destruction
by fire of this structure could support an ascription to Karara and if this is
the case and the tradition is taken literally there could be a possibility of
finding his charred bones in an unexcavated part of the site. The mean of the
first radiocarbon date would certainly support an earlier ascription of the site
174 UGANDA JOURNAL
Considering that in West Africa a native tradition of making tobacco
pipes only developed in the seventeenth century, even though the Ghana
coast had been in continuous contact with tobacco smoking Europeans for
more than a century, it is more tempting to ascribe the first kraal with tobacco
pipes and beads to the early eighteenth century, either Macwa or Karara.
After the defeat of the Bunyoro by Ntare IV the abagabe would perhaps have
been more inclined and able to live in Kinyoro style, with royal pottery and
calabash forms copied from Bunyoro. It is not impossible that Kasasira had a
capital for a time on a hill in the more exposed Masha plain, though the gen-
eral practice before Ntare was to keep to the defendable uplands of Isingiro.
It is not inconeivable, prior to the relative decline of Bunyoro around 1700,
Ankole rulers could have had some buildings akin to their more power-
ful neighbours to the northet8y the early twentieth century however capitals
had become transitory affairS26 and Bishop Tucker's descriptions of Kahaya
II's kraal hardly matches with the ekyikari banks and cisterns of the later
developments at Bweyorere.
The wide minor variations in such a small collection of pottery, limited to
relatively few forms, would suggest that the pots were acquired, with the
exception of the fine ware and calabash forms, from a variety of sources,
probably relatively local and the work of Bairu groups. The beads attest
trade, though their scarcity suggests its limited nature, probably through the
general intermediacy of tribal contact, rather than on any formalized trade
basis. The presence of tobacco pipes is a further indication of a break in the
isolation of western Uganda. It is of interest that Hiernaux also noted the
intrusive nature of tobacco pipes in his Kibiro sequence which appear at a
similar time to the graphite calabash forms and often bear similar decorative
motifs.28 Until the eastern Congo and central Tanzania are extensively sur-
veyed it will be impossible to elaborate the story of the spread of tobacco.
The excavation of Bweyorere with its links with the earlier Bigo in the
ekyikari enclosure'" and the persistence of basic pottery forms has highlighted
some of the problems still facing the archaeologist of the later Iron Age of
Uganda. Only an intensive study of a large number of capital sites assignable
to specific rulers, who can be roughly dated from traditional evidence, will
enable a more accurate pottery chronology to be developed. The paucity of
pottery on such pastoral sites, and the persistence of basic pottery traditions,
demands large excavations, if statistically significant pottery assemblages are
to be used for differentiating between sites occupied for short periods, and
all datable to within two and a half centuries at most. It would appear that in
the absence of a vertical stratigraphy at one particular site, close analysis of
the artefacts from traditionally "dated" sites may provide an interpretable
horizontal stratigraphy. Because of the paucity of pottery on such pastoral
sites however, and the persistence of basic potting traditions over long periods,
large excavations are essential if statistically significant pottery assemblages
are to be recovered. Even with such large collections, typological analysis and
interpretation of minor variations can prove to be difficult in such a situation
as Bweyorere where all the capitals were occupied for brief durations and all
of them are datable to a period of less than two hundred and fifty years, dur-
ing which there may have been little stylistic change. The excavation of
known nineteenth century sites such as Nyamirima and Rukoma should
provide some indication of the expansion of trade following the movement
of Muslim traders into Unyamwezi and Karagwe.
- prJt-- t.
jlk ~~;L LI
( _)?;;~5~C Od
O 4 b I 1
3 c O ~-I 2 3 4NCHES
O 2 4 6 8 O0 CENTIMETRES
O 2 3 INCHES
o 2 4 6 CENTIMETRES
180 UGANDA JOURNAL
Fig. 1. Map showing location of site and plan of earthworks and excavations.
2. Cross-sections of selected trenches.
3. Pottery. 1, 4-5 'Calabash' black ware, 2 Fine 'royal' ware, 3 surface
sherd ofBoudine ware. 1, 4, 5 polished, 5 with fine slip.
4. Pottery. Large pots. 1-2, 4, 5, 6 variations of cord rouletted ware.
3 part of ? pot stand.
5. Rim sherds.
7. Pipes and Pottery. 1, 3, 4, 5 pipes. 10-11 ?handles.
Plates 1. Present day Hima kraal near Bigo showing dung heaps around
lower perimeter of enclosure.
2. Post holes and hearth on old ground surface of Trench 1.
3. End of large horn core of Ankole cow on old ground surface
BWEYORERE EXCAVATION 181
The excavator is extremely grateful to the former Ankole government
for permission to excavate the site and especially to the Enganzi, the late Kesi
Nganwa, for his deep interest and assistance at all times. The saza Chief
Isingiro, Mr. Mugenyi provided local amenities and rest house accommodation
free of charge. Financial assistance was provided by Makerere University
College, the Uganda Government, and the British Academy. The Geological
Survey, the Lands and Survey, Game and Fisheries and Public Works depart-
ments gave help in kind or services. Mr. Patrick Y. Bulenzi was site super-
visor and to both he and the following staff and former students of Makerere
University College I am deeply indebted for their labour and cheerful compan-
ionship, Dr. J. Thurston, Misses B. Sloane, F. Mukasa, R. Lever, S. Bosa and
Messrs. J. Muthama, S. Mathae and M. Bulumna. Figure 3 first appeared in
the Journal of Alfrican History,so and is here reproduced with the permission of
the editors. The other pottery drawings are all by Mr. G. Kakoza of the
Margaret Trowell School of Fine Art at Makerere.
1. Oliver, R. Ancient capital sites of Ankole, Uganda J., 23, 1959, pp. 51-64.
2. Posnansky, M. Some archaeological aspects of the ethnohistory of Uganda, in Actes du
IVe Congres Pan-Africain de Prehistoire et de l'etude du QLuarternaire, vol. 3, Tervuren, 1962,
3. Oliver, of. cit, p. 57
4. In the original map in Posnansky, 1962, of. cit, the position of Bweyorere on inset map
II is incorrectly shown.
5. According to Oberg, K. The kingdom of Ankole, in African political systems, edited by
M. Fortes and E. Evans-Pritchard, London, 1940, pp. 140-141, the term ekyikari is to be
preferred to orurembo, previously used by the writer to describe the central enclosure as
orurembo more correctly describes the whole capital.
6. Later confirmed by chemical analysis in the Uganda Museum.
7. Very kindly studied by Dr. M. Morrison of the department of botany, Makerere Uni ver-
8. de Vries, H. L., Proc. Koninkl. Nhed. Akad. Wetenschap, 61, no. 2, 1958 or for a fuller account
of the problems of radiocarbon dates see Polach, H. A. and Golson, J. Collection of speci-
mens for radiocarbon dating and interpretation of results, Australian Institute of Aboriginal
Studies, Manual no. 2, Canberra, 1966.
9. All finds are now in the Uganda Museum, Accession numbers A 60/3-23, 25-36 and
10. Hiernaux, J. and Maquet, E. Cultures prehistoriques de 1' Age des metaux au Ruanda-
Urundi et au Kivu (Congo Belge), Academie Royale des Sciences d'Outre-Mer Classe des
Sciences Naturelles et Medicales, Brussels, 10, no. 2, 1960, fig. 28.
11. Chapman, S. Kantsyore island, Arania, 2, 1967, pp. 165-191.
12. These were initially suggested as kudu, but as kudu are not present in the area they
may possibly be of elands, which the Bahima often hunt. (See Williams, F. Lukyn, Hima
cattle, Uganda J., 6 1938, p. 19.)
13. Bull calves are often eaten by the Bahima to thin the herd. (Williams, ibid., p. 28.)
14. Williams, ibid, p. 26.
15. Katate, A. G. and Kamungungunu, L. Abagabe b'Ankole, Kampala. Eagle Press, 1955,
16. Date based on a choice of date for an eclipse occurring in Kigera III's reign. (Vansina, J.,
L'evolution du Royaume Rwanda des Origines a 1900, Brussels, Academie Royale des Sciences
17. Oliver, op. cit, pp. 51-52.
182 UGANDA JOURNAL
18. Posnansky, M. The history of the kingdoms of the western lacustrine Bantu and Rwanda
as seen from their traditions and archaeology, Proceedings of the Svnrteenth Conference of the
Rhodes-Livingstone Institute, Lusaka, 1963, fg. 2.
19. Posnansky, 1962, of. cit, p. 376.
20. Oliver, of. cit, p. 57.
21. Williams, of. cit, p. 21.
22. Oliver, o. cit, p. 62.
23. Williams, of.cit, p. 27.
24. Oberg, of. cit, p. 141.
25. Oliver, op. cit, p. 57.
26. Roscoe, J., ~The Balnyankole, London, 1923, p. 36
27. Tucker, A. R. Eighteenyears in Uganda and East Africa, London, 1908, vol. 2. p. 239.
28. Hiernaux, J. Recent research at prehistoric sites in Ruanda, in the Belgian Congo
(Katanga Province) and in Uganda (Kibiro), in Discovering Africa's past, edited by M.
Posnansky, Kampala Uganda Museum, Occasional Paper no 4, 1959, pp. 26-30).
29. Posnansky, M. Kingship, archaeology and historical myth, Uganda]., 30, 1966, pp. 1-12).
30. Posnansky, M. Pottery types from archaeological sites in East Africa, journal of African
History, 2, 1962, p. 194.
The Wenner Gren Foundation generously made a grant for a translation
of Katate's Abagabe b'Alnkole by Mr. Francis Kasura and Mr. John Kazoora,
which forms the basis for the discussion of the traditions.
The Editors of the Uganda Journal wish to acknowledge the receipt of
a generous grant from the Trustees of the Uganda Museum towards the cost
of reproducing the illustrative material for this article, and express their
gratitude to the Trustees.
Uganda Journal, 31, 2(1968), 183--188
THE ORIGIN OF LAKE NYABIHOKO,
KAJARA COUNTY, ANKOLE
By B. H. MOTTRAM
Lake Nyabihoko lies in the floor of the abandoned former Kagera valley,
about 5 miles south of Rwashamaire and about 55 miles southwest of Mbarara.
Evidence is offered that this lake, which lies on the water-parting between
drainage east to the Kagera (and Lake Victoria) and west to Lake Edward
across the western rift valley scarp, has come into being since the final reversal
of drainage. It is probable that two tributary streams have built delta fans
across the old valley floor and trapped water on the valley floor to form the
present lake. Interest in the origin of this lake was aroused by being told the
local legend about the drawing of Mutumo's kraal and land, of which remnants
remain as islands within the present lake. An English version of the legend is
given here, as told by some pupils of Kitunga High School, whose homes are
near the lake.
Lake Nyabihoko (also given as Lake Karengye on printed maps but not in
fact known locally by that name) lies in the floor of an otherwise nearly dry,
flat-bottomed valley that crosses southern Kajara county from southeast to
west-northwest. This valley is almost certainly the line of the proto-Kagera
before reversal.' Much of the information of the accompanying map is from
published topographical and geological maps, but the vegetational details
round the lake and small modifications to the alluvium boundaries are based
on field observations. Today the surrounding country is extensively grazed by
Bahima and Banyaruanda pastoralists, of whom most of the latter came in
about twenty years ago. The lake lies in an anomalous position astride the water
divide between reversed drainage east to the Kagera and Lake Victoria and
west across the western rift valley scarp. This poses a problem as to why the
lake is where it is and why there are no other lakes in similar positions else-
where, though extensive areas of swamp with drainage out at both ends do
also occur on the Kafu and Katonga rivers.
The general setting of Lake Nyabihoko has been described by Doornkamp
and Temple2 who review the previous literature of the area. These authors
note that "as uplift continued, the effect would be progressively to reduce the
competence of the drainage network to carry load; aggradation would therefore
begin before final reversal took place (about 60,000 years ago), and would
continue during and after the period of actual reversal. The main valleys east
of the upwarp are invariably choked with infill of alluvium or vegetation (papy-
rus) or standing water". In general this is so, though in the drier part of the old
Kagera valley, through Kajara, papyrus is partly or completely absent along
some stretches of the abandoned valley. Even so this does not go far towards
accounting for the actual location of Lake Nyabihoko, almost excatly on the
present divide. More local causes have to be found. It has previously been
noted that in the narrow deeply incised gorges through the hills of the Karagwe-
Ankolean series, steep-surfaced lateral delta fans, at the foot of the steep ridges,
have certainly contributed to the final blocking of through westward flow, and
184 UGANDA JOURNAL
this is especially true of the Kirimbi divide of the Rwizi, northeast of Rwasha-
maire.3 The truncated west flowing Kasharara and Rwamabondo swamp
streams still flow west on the crest of the upwarp. It seems likely that similar
local delta fan blacking contributed to, or was mainly responsible for, the precise
location of Lake Nyabihoko where it now is. Accordingly a study of the rivers
flowing into this lake and associated land forms was made.
Though Doorkamp and Temple also write: "those affluents, whose course
run roughly north-south, parallel to the rift wall, and drain but slowly and
deviously to the rift, show little signs of rejuvenation below the general
level of the Lake Victoria drainage"4; in the case of some of the north bank
tributaries of the proto-Kagera in Kajara, this is not always true. The two south
flowing streams from the Nyakisa hills that flow into the present lake have
relatively steep gradients. This is also true of the Mugoya-Kahengye river that
forms the Ankole-Kigezi boundary. This river falls over 100 feet a mile north
and south of the Rwashamaire to Rukungiri road, and is mostly a rapid, fast-
flowing river over a rock botton. The smaller Rwamunaba, further east, south
of Rwashamaire, has an irregular profile. Southeast of Kitunga it is papyrus
filled, but has a gradient of six feet in a hundred. In past times it must have
been clear of swamp vegetation for sub-angular quartzite boulders have been
dug out from under grey clay when digging the intake channels for the Kitunga
School water supply near the Katoma road. Three miles southwest of the
saza headquarters at Rwashamaire, the Nyakabare falls on this river have a
drop of over 40 feet and there are further falls and rapids before the river reaches
its delta fan at the west end of the lake, at Ki~jonji. This and other valleys appear
to hang above the main through-valley of the former proto-Kagera. Atributary
of the north flowing Chobugombe river, which joins the lake at its eastern end,
has cut a gorge over 30 feet deep before it debouches onto lower ground south of
Royonza about 5 miles south of the lake. Evidently violent flash floods can be
powerful erosion agents and bring down sufficient quantities of silt and mud
from the phyllites of the hills, to build out extensive delta flats on lower ground.
The writer has recorded up to I( inches of rain in little over an hour. When
falls of this order come, the sheet floods, pouring off the grassy bare-sided
Nyakisa hills (formed of a deep syncline of Karagwe-Ankolean rocks), rise
very quickly and high enough to flood the roads at their swamp crossings on
both sides of these hills.
When these tributary valleys are traced to the lake, there are several delta
flats of clay. Levelling of some of them revealed the following: (i) Rwamunaba,
nearest to lake, 53 feet rise in 66 feet, easing off further away from the lake to
5) feet in 160 feet, before meeting the valley slope; (ii) Rwibango, a steady
slope of 53 feet in 480 feet; (iii) Chobugombe is flatter, with only a gradient
of 1 foot in 130 feet (this delta flat is also gently cone-shaped and not fresh looking
and could pre-date the lake).
There are actually more such delta fans of clay alluvium than are shown
on the geological map prepared by Coombe. These are shown on the map
accompanying this text. These delta fans are advancing out into the lake, but
except at each end have not altered markedly the original lake shore. Most of
the lake shore has steep slopes that continue into the lake with no apparent
change of slope at water level. The lake floor has been reported by those poling
a raft to be of clay and only about 10 to 15 feet deep. The shore-line appears
'drowned' and there are the offshore islands ofrock, testifying to their drowning,
after earlier formation as small 'tors' on the former valley floor. One of these,
400 yards south of the Nkongoro fishing village, is known as Mutumo's Island,
LAKE NYABIHOKO 185
but the map omits another a further 300 yards further south, nearer the southern
shore. Sand could not be found anywhere along the old proto-Kagera valley or
around the lake, though it has been looked for, for building purposes. Sand,
derived from rotted granite-gneiss, is present higher up some of the valleys.
During heavy rains the short streams flowing into the lake can raise its
level at least 23 feet, judging by measured high water marks on rocks. When
this happens there is a simultaneous flow out from both ends of the lake, east-
wards towards the Kagera and Lake Victoria; and westwards into the Kanono
swamp and thence into the Bubira gorge and this into Lake Edward. The
Kakono swamp has extensive tall marsh grass as well as papyrus, and also
contains a smaller lake that is now larger than shown on the survey map, hased
on 1955 air photographs. It is likely this lake became enlarged during the 1961
and 1963-454 wet period, at the same time the Rwashamaire to Rubaare
road was flooded where it crosses the outflow from the east end of Lake Nya-
Above the present lake old benches were looked for, but only two rather
indefinite bevels were found near Kijonji, about 200 feet above the present
water level. They are very degraded and thought to be associated with the
former proto-Kagera, when flowing west, and unconnected with the present
lake. No sub-angular gravels were found but only angular rotted sub-aerial
The precise origin of the lake is still uncertain, but the following tentative
suggestions can be made. First, the lake occupies almost the whole valley floor,
with little or no fringing swamp. The lake is not part of a larger swamp-lake
complex, like the Koki-Nakivale system, which has resulted from the ponding
of the head-waters of the proto-Oruchinga following the final drainage reversal.
The lake lies astride the water divide (from which there is flow in opposite
directions). There is not any discernable evidence of papyrus blocking. The
outflow channels pass through more marsh grass than papyrus, especially at
the western end. Also the lake is certainly deeper than 6 feet to enable papyrus
to root. The lateral delta fans alone have not directly blocked the two outlets,
even if they have contributed. The Chobugombe delta, at the east end, does
not extend right across the former through-valley and nor has the shorter
Rwamunaba, at the west end, extended its delta right across. One pointer
could be that a series of sudden floods in this area from the surrounding streams,
draining from the nearby hills poured into this part of the valley floor, before
marsh vegetation had built up to levels seen east and west of the lake. Such a
sudden series of floods could kill off already existing vegetation and prevent
new from growing. Such a series of sudden floods could account for the Mutumo
legend. The floods could have been local and could have happened at any time,
since the first cattle people came into this area four or five centuries ago. The
steep shores suggest the drowning can have been relatively rapid, but the water
was trapped long enough for it only to escape slowly east and west, where
vegetational aggradation has continued. The Kakono swamp, to the west,
shows a more common swamp-lake associated and is cited by Doornkamp and
Temple as characterisite of aggraded and reversed streams, following the
final stages in the uplift and reversal.
A further factor to be borne in mind is that much of the present drainage
is superimposed and antecedent to the present landscape,5 and that this accounts
for the hanging valley and knick-points already referred to. It was only at the
final stage of the uplift east of the rift scarp that the former west flowing rivers
were unable to continue down-cutting, as the slow rise continued. For a time
C ~ Mostly grass swonp
Open water ( Lake water
height Is just over LS00ft)
A v I Puvieltflats and de~tts
lH Hot slprig
Pe rmanent flow
LAKE NYABIHOKO 187
during the late Tertiary, while down-cutting was continuing, the larger main
rivers would be able to cut down more effectively than weaker tributary streams.
Knick-points would advance back up tributary valleys according to rock structure
and volume offlow. In this way the irregular profiles of the Kasharara-Kahengye,
Rwamunaba and Chobugombe rivers could be produced. All have rapids and
falls on them and their lower reaches appear rejuvenated, though both the Chobu-
gombe and the Rwamunaba now join a slow east flowing stream to the Kagera.
Where these tributary valley join the main valley there are lateral delta fans,
which are also very common along the Rwizi further northeast in Ankole and
held to be a contributory cause of blocking the formation of Lake Nyabihoko in
its present position, about 4, 500 feet above sea level and on the crest of the
upwarp east of the western rift valley.
1. Temple, P. H.., personal communication.
2. Doornkamp, J. C. and Temple, P. H., Surface, drainage and tectonic instability in part of
southern Uganda. (Geographical Journal, z31, 1966, pp. 238-252).
3. Mottram, B. H. The Rwizi drainage basin of southwest Uganda in relation to the Oruchinga
and Kagera rivers. (Ugandca Journal, sy, 1963, pp. 177-186).
4. Doornkamp, J. C. and Temple, P. H., of. cit., pp. 247-248.
5. Ibid., p. 244.
THE LEGEND OF LAKE NYABIHOKO
By J. KARIISA and E. M. KAHWTITE
Where Lake Nyabihoko lies now it is thought that there once lived a man
called Mutumo. He had only one cow which produced many calves and made
him a rich man. One day his cow said to him, "W~hen I die, bury me, do not
eat my meat". Mutumo went and told his family what the cow had said to him,
but Mutumo's wife said, "That is a miracle! How can a cow talk ?" Mutumo
insisted to his family that the cow should be buried when it died.
Mutumo's cow increased and he wanted a good bull for them, so he went
with his youngest son to Rwanda to look for a bull. While he was in Rwanda
his cow died. His wife told her children that there was nowhere on earth where
a cow had ever talked and told her sons that they could eat the cow, sell some
of the meat, also burn the bones and bury the ashes.
Next day, while sweeping the sitting room a girl heard a splash of water,
but when she looked around she could not see it. She was amazed that she did
not know where the sound had come from. She told her mother, who was
shaking milk at the time to make butter, about what she had seen; but her
mother cursed and beat her and asked how could the water have come under
the house. The following day the mother heard it for herself and also did not
know where it had come from or what was happening. In the evening where
the cows returned towards their kraal the elder son saw that there was water
between the cows and the kraal. Then the cows disappeared into the rising
water and all the area became flooded that same evening.
188 UGANDA JOURNAL
When Mutumo came back from Rwanda he found that the whole of his
land had turned into a lake. He went to Mwamba, who was a rich man and who
lived in what is now called Kashari, but was then known as Nkore. Mwamba
had as many cows as Mutomo had before his property was destroyed and
Mutumo was given the job of looking after Mwamba's calves. But Mwamba's
children soon heard Mutumo talking to himself, saying that though things
were better they were not as good as they had been. Mwamba's children went
to their father and told him what they had heard Mutumo saying to himself.
Then next day Mwamba went secretly to where Mutumo was tending Mwamba's
calves and to listen to what Mutumo had been saying to himself, but without
Mutumo knowing Mwamba was listening. After hearing Mutumo talk to
himself, Mwamba later asked Mutumo why he always felt so miserable. Eventu-
ally Mutumo told Mwamba the whole story.
Later Mutumo and Mwamba became very friendly; so Mwamba decided
to share his friend's misery by both drowning themselves in the lake. The two
friends went to Lake Nyabihoko with their children. In order not to arouse
suspicion among their children the two fathers decided to prepare a feast at the
lakeside. During the feast Mwamba told the children to share his property
equally after his death, and this was to include Mutomo's son. Then Mwamba
told all the children to return to the Nkore home to collect more supplies, another
cow and more beer. When Mwamba saw that the children were out of sight,
he told Mutumo to go with him to where Mutumo's property had been. Then
they put their arms round each other's shoulders, because the party was being
given by Mwamba in order that Mutumo could forget everything about his
lost property. Mwamba felt that it was not right for Mutumo to go on being a
poor labourer. Then, while they held their arms round each other, they fell into
the lake and that was the end of Mutumo and Mwamba.
Later the children returned to find that their fathers had disappeared.
But because no one had lived near Mutumo before the lake came into being,
the children had no one to ask where their fathers had gone. They collected the
remains of the feast and returned to Nkole.
In this way Lake Nyabihoko came into existence and the islands are all
that remain of Mutumo's kraal. Indeed the island nearest the north shore (and
marked on the map) is known as Mutumo's island to this day.
The legend as recorded here was told to B. H. Mottram and written by the two authors.
It is understood that the legend has been published in a Runyankore reader, but this is not
available and has not been seen by the authors.
Uganda Journal, 32, 2 (1968), 189--198
THE DIARIES OF EMIN PASHA--EXTRACTS XIV
Edited by SIR JOHN GRAY
(These extracts from Die Ta~gebiicher von Emin Pasha edited by Dr. Frann
Stuhlmann, vols. i, ii, iii, iv, and vi (Braunschweig: Westermann, 1916-1927)
have been translated and provided with introductory notes by Sir John Gray.
They have appeared 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. Extracts I to X
appeared in successive issues of the Uiganda Journal commencing with Extracts I
in U'ganda Journal, as, 1961, no I until Ug~anda Journal, ag, 1965, no 2. Extracts
XI appeared in Uganda Journal, 3o, 1966, no. 2. Extracts XII in Uganda Journal,
3x, 1967, no. 2, and XIII in Uganda Journal, 32, 1968, no. 1. This present
extract brings the series to a close. Readers of the Journal are grateful to Sir
John Gray for his unremitting zeal in fulfilling this enormous task.-EDS.)
XIV--EMIN'S JOURNEY THROUGH ANKOLE
4 July to 3 December 1889
With these Extracts, Emin's Odyssey to the Indian Ocean comes to an end.
The last five months of the Relief Expedition's journey were for Emin a period
of sickness and frustration. Latterly he was far from Uganda, and only a few
extracts have been included, but these suffice to indicate his morbid sensitivity
and growing disillusionment. The 'unfortunate fall' at Bagamoyo which is
referred to in Stuhlmann's conclusion, and which incapacitated him for some
months, occurred on the evening of 4 December 1889. It is described by Stanley,
In Darkest Africa, vol. II, pp. 413-418.
The full text of Stuhlmann's edition of Volume V of the Tagebikcher,
running from 26 April l890 to 23 October 1892 was duly advertised by German
publishers, but seems never to have appeared in print; nor has a copy of the
script been located. But the diary is quoted at length in Volume II of Georg
Schweitzer's Life of Erin Pasha (1898). This was the period of Emin's service
under the German Government, and his last tragic journey towards the Congo
and his death, probably on the day of the last entry in his diary.
For Uganda readers special interest attaches to the months between
November 1890 when he reached Bukoba and December 1891 when, westwards
of Lake Albert, he directed his companion, Stuhlmann, to return to Bukoba.
Emin was during this period within or near the confines of Uganda's sphere of
influence, and for part of northwest Ankole and Kigezi he has left the very
first written record.
190 UGANDA JOURNAL
Extracts from Emin's Diaries
July 4 1889. We are all smitten with fever. At noon we halted in a some-
what neglected banana plantation. The place is called Kitete and belongs to
Ankole. I do not know what lies ahead, as fever overpowers me despite my
July 5. At 10.30 a.m. we halted in a banana plantation on a hill site.
The place is called Kiwiga. During the last three days my temperature has
never gone down. Parke has given a subcutaneous injection.
July 6. (Reached Busimba). For a week I have had no sleep because of
July 7. Rashid bin Omar, the headman, and the Zanzibaris, have been on
their way to the Kibuga, the residence of the King of Ankole. They have come
to the village where the King's mother resides and were ordered by her to stop
as King Ntare is at present in his possessions in Ruanda.
July 8. Today Stanley mustered all the armed men, Zanzibaris and my
own and told them they had permission to go out and take bananas, maize,
beans or sweet potatoes wherever they would find them, but they must respect
the other property of the inhabitants.
July 9. (Marched until midday). Soon afterwards Dr. Parke came and
informed me that eight armed Baganda had come with a letter to Stanley,
who was in secret consultation with them. At 4 p.m. some of them, led by
Stanley's servant, came to me. One of them, who claimed to have been
Mackay's servant in Buganda, said that all my old acquaintances with some
others whom he did not name, were dead. They were the Katikiro, Kyamba-
lango, Kitunzi, Gabunga, Kangawo, Masudi etc. He referred me for details
to Stanley, to whom he had told everything. I think these people are fugitives
from Buganda and have settled with King Ntare and have perhaps been sent
by him to Stanley. They are certainly not ambassadors from Mwanga.
Our night quarters are called Kitega. The village chief, whom I have
not seen, is called Mwedusi.
July 10. (Reached Katura). Two days more and the third brings us to
Kagera, the Alexandra Nile and thence into Karagwe.
The Baganda have told Parke that Mwanga was deposed a year and a
half ago and is a prisoner on an island in the lake. His brother is Regent in
Buganda. Civil war--daily. Europeans driven out. Daily massacres. Two thou-
sand people with weapons have fled to Ankole.
July 11. Our night quarters are called Ulingo. The chief, who remained
invisible, is called Bagongera.
It was nearly evening when Mackay's servant came with Stanley's servant
Salem as interpreter, and wished to speak to me, as he was the headman of the
Christians in Ankole. He gave me an account of the sufferings of the Christians
and their followers . .. ... In his conference with Stanley he had been told
that he (sc. Stanley) could not set foot in Buganda without a mandate from
Queen Victoria, but he promised that from Usukuma he would telegraph,
and if the reply was favourable either he himself or myself would go there.
July 12. (Started early and camped at Mpogo).
July 13. Igomera, Ntare's son, to whom the village belongs, has gone away.
Today two Baganda came, one being Mackay's servant Sitefano. They
are going to accompany us. Zakariya,' the chief of the Baganda Christians,
has sent Stanley three cows as a present.
July 14. Stanley has caught a cold and has shortened the march for that
DIARIES OF EMIN PASHA--EXTRACTS XIV 191
reason. At 11 a.m. we halted at a large banana plantation. I was told that the
place is called Ruganda and that the chief, Wasikani is a relation of Ntare.
At 4 p.m. I received the following note from Lieutenant Stairs
"I am directed by Mr. Stanley to tell you for your guidance that tomorrow
the column halts here (15th). The next day we march (16th). On the
next--the 17th-we halt. Then there are two days marching to the river,
19th. He expects we shall be two days in crossing all hands over. Once
over we march direct to the King's capital, seven days-Karagwe. For
these seven days there is no food at all on the road.-Therefore, Mr
Stanley wishes you kindly to tell all your people to prepare seven days
food between this place and the river so that when they cross they will
have food right on to the capital. This is to be a reserve ration over and
above the ordinary every day's supply".
July 18. The place where we now are is called Kasozi. Fever.
July 19. We are still here. Stanley is waiting for a son of the King with
Ntare's son, for whom we are waiting, has not come. Stanley has again
had a bad bout of fever. So has L~ieutenant Stairs. At 5 p.m. Stanley sent
Captain Nelson to tell me that we will stay here tomorrow to wait for the
King's son. In other words he is ill.
July 20 and 21. (A number of Emin's followers announce that they are
unable to march any further and want to settle in Ankole).
July 22. I have visited Stanley and he has talked to me about his campaign-
ing in Abyssinia. He is the best story-teller I have ever heard. When he was
told that the Prince of Ankole (Mulangira)2 had come, he left me. To my
surprise Stanley has let me know that we will stay here tomorrow.
July 23. (The entry in the diary occupies three lines of print and is confined
to details of diseases amongst Emin's people. There is no mention of Stanley
having on this day gone through the ceremony of blood brotherhood with
July 24. (Expedition reached Mawane).
July 25. We set out early and reached the bank of the Kagera at 1.30 p.m.
July 26. We took down our tents and huts early and at 1.30 p.m. we were
on the other bank of the river in Karagwe.
July 27. Stanley has crossed over. We marched off in the morning.
July 28. The district from here to the river is called Bugomera by the
inhabitants. That on the left bank of the river where we first camped is called
Kahandakamu. In front of us lies Mtangata, the district of hot springs.
July 29. (Reached Mtangata).
July 30. (Nelson and Jephson are both ill).
July 31. Casati is quite ill, but is so obstinate that he refuses to ride on
either my or Stanley's donkey.
This place is called Kibale according to Stanley's interpreter, but the
natives call it Kasasi.
August 1. Nelson tells me that Stanley proposes to march the whole day
until evening. I have reduced my 33 loads to 15. Nelson has just come back
and says Stokes has taken Mwanga to Buganda and is fighting the Arabs.
Many fugitives from Buganda have come to Karagwe.
August 2. (Reached Kiwuna about noon). Casati is so ill that he must
be carried on the march. Stanley suffers from rheumatism.
August 3. (Sighted Grant's Lake Windermere on the march and reached
192 UGANDA JOURNAL
I went to Stanley who had pitched his tent in the ruins of Ahmed bin
August 4. Yesterday King NIdagara, a lad sixteen years of age sent to
Stanley to welcome him as a friend of his grandfather.
August 5. Stanley and Jephson to visit the King and asked if I would
accompany him. I thanked him for the offer of this hill climb, but declined it.
Jephson has come back disgusted and declares all African kings to be humbugs
and swindlers of travellers.
(Emin gives a list of losses amongst his people between 2 July and 4 August
1889, totalling 141 people in one month. Terrible.)
August 6. The King's much exaggerated present has been reduced to
twenty clusters of bananas and a cow.
When Rumanika5 died, he was succeeded by his eldest son, who died
during a rebellion with the present boy king. Another brother followed, who
died in a dispute with his chief. A third brother followed who soon died.
There remains a fourth and last brother, who refused the offer of the kingdom
and is content to act as guardian of his nephew and still does so to this day,
although in two years' time the nephew will attain his majority.
Stanley has today sent me some milk, a sheep (a present from Kiwengo,6
Speke's old Kirongozi, who lives here), a hen and some bananas. We march
August 7. We came to Kiwengo's Zeriba, a large thoroughly dirty place.
W~e stopped for half an hour whilst Stanley conversed with Kiwengo. We
reached camp at 11 a.m. King Ndagara has let it be known that he will not
allow our sick persons to remain behind.
August 8. We left Kyaairakari (the name of the village). Speke calls it
Rusaka that being the name of the ferry. At midday we camped at Butenga,
otherwise Speke's Uthenga. Stanley sent word to me that we will stay here
tomorrow and the next day.
August 9. Stanley let it be known that he will march tomorrow. A very
pleasant conversation. I had not received a present from him since Kavalli.
Today he sends me a very pretty mantle for Ferida.
Stanley tells me that one of Mlwanga's Baganda chiefs got permission to
proceed and passed by us at night when we were sleeping at Rusaka.
August 10. We reached Speke's Lake Urigi and have camped in a w~ilder-
August I1. We are camping in the village of N~ryabikukuru and are now
out of Karagwe and in Ihangiro.
August 13. An inhabitant tells me that the name of the ferry we have
reached is Bahimiro and belongs to Ihangiro. Here they speak a sort of Runyoro.
At sunset Stanley sends word to me that we will stay here tomorrow. Why?
I saw him today and he did not say a word about it. This change of mind
is characteristic of him. I have been informed that Stanley had thrown five
(others say ten) boxes of ammunition into the lake. It would have been better
to have taken them to Msalala (twelve days' march) and sold them 'for the
good of the Expedition' to Stokes.
August 15. (Reached Kimw~andya which belongs to Buzinza).
August 16. Stanley told me that on the day he reached Msalala, he will
at once send a courier to Zanzibar and there will be an opportunity for me to
write to the Committee (sc. of the Relief Expedition).
August 17. (Reached Kisinga).
DIARIES OF EMIN PASHAA-EXTRACTS XIV 193
August 18. After leaving Kisinga and climbing on a hill we had a glimpse
of Lake Victoria and all its islands, Stanley cried 'Hurrah' and so did the
Zanzibaris. I also had to do so.
August 22. The following note (in English) regarding beads reads as
"The rations are for 4 days commencing tomorrow August 23. Each per-
son receives 2 Kette (? strings) of red beads and one each of blue and white
beads. Your number is of 304 people i.e. including your own rations. You will
receive 26 cups of red beads, 7 cups of blue beads and 5 of white beads a total
of 38 cups, each cup of red making 30 Kette, of blue 57, of white 75".
August 27. At 2 p.m. we came to a place indicating that we were near to
civilisation. A Boma displayed a pretty wooden tower encased in boards.
At 2.10 p.m. we reached the French mission station in Usambiro.
(Emin gives a description of the station and its buildings which had
been abandoned by the White Fathers).
August 28. I mounted my donkey in a miserable state for the four and a
half hours' journey to Mackay's station. The four and a half hours became six.
At last we reached the English mission station and were welcomed in a most
friendly and hearty fashion by Mackay and Deekes.
I was soon put into a cool house as was Stanley. I have received a whole
heap of old letters and newspapers.
August 29. I do not know what happened yesterday or today. They have
have passed and I have done no work, except looking at letters and news-
papers, but doing no orderly reading of them, but studying one after the other
seated and idling at the table. And it will be like this tomorrow.
I am not well but I must not let anybody know this.
Mackay tells me that five cases of effects for me were lost in Buganda and
amongst them my clock and hundreds of letters &c.
August 30. I am gradually getting my papers in order. Mackay has been
with me most of the day and has given me a lot of interesting information.
He has sent to Stokes, who can arrive here in five or six days with the latest
news about Buganda. A relief caravan of some more of Mackinnon's people
should be four days to the east on their way to Wadelai. Mlackay wants to
send people to them. We can then all go to Buganda, annex it and go home.
August 31. Began to collect my things. They are few enough.
September 1, 1889. I went out today to collect (sc. botanical specimens).
September 2. Stanley has promised to give me one of his twenty Unyamwezi
donkeys, but they are all very miserable creatures.
Mackay no longer comes to see me or to inquire as to my wants. I cannot
understand the reason for this.
September 3. Yesterday Stanley told me that, if my letter was ready, he
would send it by the post which was going early today. I am making three
copies of every letter because the route is so uncertain. Today the post left at
September 4. Late yesterday evening two French missionaries arrived from
They were Phres Girault and Schynse (a German Alsatian).
(They brought fruit for Stanley and Emin, who bought a donkey from
Schynse for 120 dollars and also paid 125 francs for a saddle).
September 5. The gentlemen set out today--five hours by water--equal
to two days by land.
The boat sent to Mr. Stokes has returned, not having found him. The
194 UGANDA JOURNAL
messenger says two Catholic and two Protestant missionaries have gone to
September 8. Mackay has not been to see me, although he knows that I am
unwell. He is always having a great many conferences with the officers.
September I I. (Had a long discussion with Stanley about geology).
Stanley is enthusiastically employed in making preparations for our depar-
September 13. Stanley has today distributed the cloth. For me and all
my people two good porter loads. At 11 a.m. Nelson came and brought me
nine Zanzibaris as porters, but later took one of the porters away. So they are
now eight. I need eighteen loads for myself and two for the child (Ferida).
Mackay has generously made up his account as amounting to 73 dollars
and wishes all the gifts he gave to me to be treated as presents.
September 14. It is interesting to listen to the discussions between Stanley
and Mackay. The latter is patient, obliging and well mannered. Stanley on the
other hand is impatient, disobliging, pretentious, stubborn and lacking in
September 16. Yesterday was the last evening of our sojourn with Mackay.
For the occasion he took the opportunity of putting a flask of good port wine
on the table. We were all surprised when Stanley proposed a toast as if in the
the leading part of a Gordon. I proposed the second toast and therein paid
ambiguous compliments and the hope that I would return soon to the renowned
land of Africa. Then followed Mackay's toast to the greatest of all explorers.
Then Stanley's to the Church Missionary Society and Mackay and then mine
to Mackinnon. Music followed as we sat at our tables. Mackay said the long
expected post might come in the morning. He suggested to Stanley that he
might postpone his departure for one or two days. Stanley said it would be
better if the people took a trial march and that they could return after proceed-
ing two or three miles.
We were all ready early today and had to wait five minutes, which seemed
like an hour until Stanley's people were drawn up in order of march. When
Nlelson's company had set out, Stairs came and told me that we were to march
behind Nelson. Stanley sat on his donkey and stayed behind. At 10 a.m. we
halted at a village called Urima. Here Nelson told me that Stanley would
not be coming today, but would come tomorrow because he was waiting for the
post. We were superfluous guests at the station, but my trusty Stairs, Jephson
and Parke remained at the station.
September I 7. At 9 a.m. Stanley came with the officers' and men, exchanged
a few words with Nelson and went on his way. Soon afterwards Nelson came
over to me and told me that no post had come and that Stanley was camped
not far from us.
We had given up our huts and in ten minutes reached our new huts.
Shortly afterwards the three officers who had arrived with Stanley came to
inquire of us.
Parke told me that he was surprised that I had not come back. I said
(in English) "I was told to go on, not to come back."
Jephson more unexpectedly and rashly greeted me (in English) with the
words "And so you have been chiselled out yesterday."
Stairs did not say anything.
September 19. (At 2 p.m. Nelson came and told me that Stanley had
rebuked him for disobeying a standing order and had bidden him reverse an
action which he had taken. Nelson had then said "I shall obey orders and do
DIARIES OF EMIN PASHA--EXTRACTS XIV
it--so help me the God Almighty").
September 23. Nelson during the march was at my side and in the course
of conversation said "I rather like this way of travelling. It is like rabbit shooting
and Stanley is just up to it. He is pleased like a boy who shoots his first rabbit."
September 29. (Expedition reached Nguru, referred to by Stanley as
Usongo N., where Charles Stokes had a station).
October 1. Stokes' Agent here paid me a visit. H~e is an Indian, Mission
bred and speaks English fluently, and gave me information as to the way to the
coast. Stanley is ill.
October 2. Stanley is seriously ill.
Stokes' man has given me a pretty spear, belonging to the Wakanga.
(He also supplied Emin with detailed information as to the camps on the
route to the coast).
Fever has again broken out. Parke and Stairs are both bed ridden. Stanley
appears to be still ill.
October 4. (Emin, at Usongo, Stoke's principal station, submitted to
Stanley a return of the Egyptian force at this date; this shows a total of 311.
Since leaving Kavalli's on 4 August 140 persons had been lost. At Usongo two
had died and eight sick remained behind. The expedition now consisted of 294
people having lost 17 persons in one month.)
October 8. Stokes wife, the daughter of the chief of this place, a young
brown house-wife with a somewhat provocative manner has paid a visit to
my little one (Ferida) and offered one of her two companions, a brown maiden,
aged about fifteen, who first arrived here from Tabora, where she was handed
over by an Arab to the Christian Mr. Willing who paid six doti for her.
I have given a blow to this and I hope we will get away from here quite soon.
October 18. (At Kungu). Today the French missionaries came here. They
are the same persons as visited us at Usambiro (sc. Fathers Schynse and Girault).
They have at once visited Stanley. Towards evening they came to me. They
came in sixteen days of marching from Bukumbi and are going with us to
Zanzibar. They came by the west route (Msalala).
October 19. The Frenchmen have paid a long visit to me. The Alsatian
Pbre Schynse knows the route quite well and gave me a lot of information.
He works on his mapping with a pretty little theodolite.
October 31. We had just got into the thickest part of a forest when we
heard the beat of a drum and soon after the head of great caravan came in
sight. There were twenty to thirty porters who gave the German greeting
'Guten Morgen' (good morning) and one of them said clearly 'Arab Bagamoyo
Kaput'. The women also greeted us in strict military fashion. They were two
united caravans and I was not a little surprised when the two chief Wanyamezi
came and brought a letter to me. It was from the Imperial Commissioner of
the German Colony in East Africa, Herr Wissmann, dated 15 October 1889
from Mpwapwa. This was a surprise.
We had started again when we heard a gun shot. The people who had
been sent with our post from Usambiro, Mackay's station, had come by the
southern route from Nyangwira, had heard that we were here and stopped us.
They only brought a friendly letter from Wissmann for me and some letters
for Mackay. Stanley was so friendly as to let me read through Reuter's tele-
grams before sending them on.
November 1. We sent the post to Mackay yesterday evening. The march
was very long. We first left the forest after a six hours' march.
196 UGANDA JOURNAL
November 5. Stanley has paid me a long visit and we had an animated
conversation. I do not know why everybody has made up their minds that the
German Government will employ me in East Africa.
November 7. I have made a list of my people, which is as follows the
numbers in brackets are those of the census made in 4 October at Usongo.
30(30) officers, soldiers, clerks etc, 70 (70) women and concubines, 59 (61)
children, 131 (132) servants of both sexes, 13 (18) porters.
November 3. (The following entry appears in the diary of the Alsatian
Missionary August Schynse):
"Emin Pasha is ill. We offer him wine we have for the celebration of mass.
He returns it untasted. 'I shall ask for it again one of these days on behalf of
some sick member of the caravan. Till then please keep it for me.' It is an
enigma to me how he lives and stands the journey. .. .. There is the march,
during which he never gets off his donkey, . .. His time is given to science:
What is left of it to his little daughter. She is always carried before him in her
November 10. We had arrived within about an hour of Mpwapwa at
Kisokwe, the English station which is situated on a mountain which lies
another half hour off the road, when a European introduced himself. He was
accompanied by some boys. "I am Rev. Price. 9 Are you Emin Pasha ?"
-"Yes"-"Much pleased. I am going to preach in Kisokwe. In the afternoon
I shall come back and shall preach in Mpwapwa. I have therefore no time for
you, but tomorrow I shall see you." Therewith this extraordinary fellow
disappeared. He had had a similar meeting with Stanley, to whom he appeared
with a spear in his hand.
(Reached Mpwapwa and found Medem, the station chief, suffering badly
November 13. (Expedition left Mpwapwa and reached Tubugwe).
November 16. (Expedition reached Kisaga).
I visited Stanley. Quite suddenly he questioned me as to what I intended
to do with my Madi porters. He said it would be best to hand them over to the
English East Africa Company as up to the present time they had been fed at
Mackinnon's expense. It was in any event only a 'hint' and I must do as I
wished. I should, however, as a German, perhaps deliver them over to the
Germans. I should, however, remember that England had sent the Expedi-
tion that Germany had first remembered in the eleventh hour. In my
present position I could not of course settle about their future, as they were
after all subjects of the Khedive. He had written to the Consul-Generals at
Cairo and Zanzibar and informed them as to the exact number of my
people. I should approach the English Consul-General at Zanzibar who
would in any event give me the best advice. He had also written to him so
that quarters etc, would be ready for them and so that they should not be
treated like shipwrecked people on a foreign island. A true Yankee ending of
December 3 1889. On the way came newspapers and a letter to Stanley
saying that the steamer Oriental was ready for me and my people. The English
are sending over a man-of-war to ferry them across from Bagamoyo and the
German Sperber is likewise available. The English do not want Wissmann
to get in touch with my people.
After a march of four and a half hours we encamped in a wood near
Mbiansi. Plenty of good water far off. The nearest water is stinking. I found here
two cases of conserves sent to mle by Kanji Hansraj, a rich Indian merchant,
DIARIES OF EMIN PASHA--EXTRACTS XIV
as a present.
(After this Stuhimann adds the following note:-
"'Emin's diary closes here. On the last day of Stanley's journey, where he
was welcomed by Wissmann at the Kingani River, he entered Bagamoyo,
but he made no further entries. Then he had an unfortunate fall from a window
and thereafter made no entries until he set out with the German Expedition
on 26 April 1890")
Letter of Emin to R. W. Felkin dated 28 August 1889. Only extracts are
given here of the letter which Emin wrote to R. W. Felkin on the day of his
arrival at Usambiro. The original containing over 1,200 words, is printed
in full Schwertzer, Life of Emin Pasha, vol. 1. Introduction, pp.xli-liii.
Usambiro, English Mission Station,
As you will see from the above, I arrived today at this station with Mr.
Stanley, and hope after a few days' rest to start from here for Zanzibar. What
has led me to this decision, which is certainly inexplicable to you, and probably
unwelcome, it is impossible to explain to you within a few lines; you must
therefore content yourself with what follows. You are the only one to whom
I write upon these sorrowful events.
(Emin here sets out a history of the events which befell him after his
first meeting with Stanley and the latter's return into the Congo to bring up the
main body of the Expedition).
The officers who went with me to Stanley entered into an undertaking within
twenty days to bring to him all their belongings, as well as the soldiers who were
willing to leave the country. Time passed, however, and only a few clerks and
officers, mostly the refuse of the province, came. After much trouble, I obtained
an extra twenty days from Mr. Stanley, and sent and told the people that at
the end of this time we should absolutely march, without waiting longer for
them, and to the same effect Mr. Stanley wrote himself. The twenty days
passed without our hearing a word from them, I was compelled, as I had
pledged my word--with a heavy heart it is true--to march, without knowing
where the people were, or what had kept them back. Concerning the journey,
excuse me writing to you; others will be able to do this better than I. All my
hopes are shattered, and I return home half-blind and broken down. I indeed
hope that I shall not be judged too hardly.
(Emin asks to be remembered to various of his correspondents in England
And now, last but not least, my thanks to Miss Felkin for two letters
received by the Congo. I still hope some day to come to England, and to be
able to thank you in person--your wife, your sister, and yourself, for all your
friendly sympathy with my affairs, and for the energetic work you have done
Enough for to-day. My greetings to you each.
Yours truly obliged.
I have just received packets of letters from you written in 1888. A thousand
thanks. They shall be answered by next post.
198 UGANDA JOURNAL
1, Zakariya Kizito became one of the Regents of Buganda during the minority of Daudi Chwa.
He died in 1917.
2. "Mulangira" is merely an honorific title meaning "prince". The actual name of the chief was
3. For an account of the ceremony see F. Lukyn Williams, Blood-brotherhood in Ankole,
Uganda J., 2, 1934, p. 40, and Early explorers in Ankole, Urganda J., 2, 1935, p. 200. See
also Parke, T. H., My personal experiences in equatorial Africa, p. 459. For the treaty which
Stanley claimed to have made "in this moon May 1888" see Sir John Gray, Early treaties
in Uganda, Uganda J., 12, 1948, pp. 31-32. A further comment is given in M. Doornbos,
Stanley's blood-brotherhood in Ankole, Uganda J., 30, 1966, pp. 209-210.
4. A biography of Ahmed bin Ibrahim appears in Ulganda J., gl, 1947, pp. 90-97.
5. Rumanika was the ruler of Karagwe at the time of the visit of Speke and~ Grant. He died in
6. "Doctor" Kiwengo, or Kiyengo is often mentioned by Speke in Journal of the discovery of the
~Nile, and Grant in A walk across Africa. (See also Uganda J., 28, 1964, pp. 191-192, note 6.)
7. The White Fathers were Pdres Lourdel and Denoit. The Protestant Missionaries were
Cyril Gordon and Robert Walker.
8. "Mr. Willing", otherwise Moses Willing, would be the person previously referred to as
"Stokes' Agent". He was one of the trusted headmen listed by Stokes to receive a share
under his Will. (See Africa, No. 8, 1896, p. 89). A number of letters from him were found
among the correspondence impounded by the Belgians when Stokes was executed in 1895.
(Ibid., pp. 84-86).
9. The Rev. J. C. Price had reached Mpwapwa in 1879 and died there, 23 January 1895.
Ugpanda Journal, 31, 2 (1968), 199--203
REDEMPTION IN GANDA TRADITIONAL BELIEF
By A. M. LUGIRA
The Ganda people may be grouped with other Africans as an "incurably
religious people".' In spite of the majority being affiliated to Christianity or
Islam2, still in moments of crisis people are comforted by lapsing into practices
of Ganda traditional belief. It is therefore possible to study aspects of redemption
in Ganda traditional religion to see what the nature of this redemption is
and secondly what the agents of this redemption are.
In the eighteenth century Islam was introduced to Buganda by Arab
traders. In the following century, Christianity started striking roots in the
country through the missionary activities, partly of the Church Missionar~y
Society from England and partly from the Franco-Catholic Wrhite Fathers.
From this time on, society in Buganda which had been entirely traditional
in belief assumed a religiously oriented triple classification. There were now
Moslems, Christians and those branded as having no dini3. The last group,
whether justifiably or not, became known as pagans, heathens, or kaffirs--the
Bakaafu. Nevertheless, traditional ideas have survived and amongst these is
the concept of redemption.
The idea of redemption in Ganda traditional belief is expressed mainly
by means of such words as Kuwonya (to cure, to save--with all possible shades of
analogical meanings in the sense of a redeeming action); Kununula (to buy back,
to ransom, to redeem;) Kulokola (to save, to rescue). In general all these
terms have the significance of getting someone out of trouble, or out
of captivity. 'Trouble' in this sense may mean misfortunes that befall
mankind such as death, illness, lack of children in married life, or poverty.
This situation in which man finds himself, which he experiences as miserable
and not terminable by himself, turns a traditionally Ganda mind to superhuman
ultra-human or superior forces for help; the effective result of which is regarded
It is only after the arrival of Islam and Christianity that the Baganda
started recording their traditions in writing. Before this, much of their history
was enshrined in oral traditions which they passed on meticulously from gener-
ation to generation. The need for redemption can be traced in legends, myths
and proverbs. In a particular way the need for redemption is implied in the
aetiological myth of the Baganda. This myth briefly runs as follows: Kintu the
founder of the kingdom of Buganda went to heaven to receive a wife. Nambi
was the name of the woman given to him by God. Nambi had a brother known
as Walumbe (death). After marriage God admonished the couple to start
their journey to the earth without Walumbe's knowledge. They departed
in Walumbe's absence. But on the way the woman remembered sne had
forgotten the grain for her fowl. She then insisted on going back to heaven for it.
Kintu tried to dissuade her, but in vain. She said she would hurry back and
get it without anyone seeing her. On her arrival in heaven she found that
200 UGANDA JOURNAL
Walumbe was already back. She tried to steal away without him but he followed
her. So Walumbe joined Kintu and Nambi on earth and became the cause of
death and other misfortunes to their offsprings.4
In relation to this fact of Walumbe, troubles and misfortunes are con-
sidered, analogically, as 'Death.' A Muganda who happens to be in trouble will
exclaim 'Nfudde'--I am dying. Indeed in this sense 'dying', does not signify
a termination of physical life. 'Death' is referred to anything that comes about
as a misfortune. In such situations people traditionally have recourse to
spiritual agents whom they trust would help liberate the afflicted from troubles.
Thus, in Buganda, spirits play the role of the 'Heiland', the 'Salvator,' the
'Saviour', in so far as misfortunes are concerned. They are also to be considered
as 'Heilbringers' bringing benefits and cultural dispositions that help in the
maintenance of social order. The names lubaale, mzusambwua, jijembe designate
outstanding spirits of the Ganda society. Muzimu signifies spirit of the departed
Mackay, one of the first Christian missionaries to arrive in Uganda had
developed an intimate relationship with Kabaka Muteesa I. One day in Mackay's
presence the king commenced questioning his chiefs: "What is lubare" ?
Answer "Mulogo" (wizard or sorcerer). "What is a Mulogo?" Answer
"(Muntu") (a man). "Then if he is a man he is not a lubare, for a lubare is a
spirit, or a god."5 This kind of question and answer would even to-day be
reiterated if one had to interview a number of Baganda on the concept of
lubaale. People simply believed in lubaale without bothering themselves to
find out what this mysterious power exactly was. This power may be considered
as a unity in diversity. Students of religion will find the concept of lubaale
presenting itself in various refractions. For this reason the concept lubaale
required systematization and analysis in order to throw better light on it.
Synthetically lubaale may be considered as a pyramidal, hierarchy of powers.
Thus lubaale Katonda may be considered as the supreme power and in contrast
to other lubaale as minor powers. On the other hand lubaale may be considered
as a spirit associated with the mighty celestial phenomena, for instance the sky
is known as olubaale, as it is the abode of such powers as lightning and thunder.
The vast expanse of water traditionally known as N~nalubaale (Lake Victoria)
is also to be considered in this connection. These phenomena have occasioned
a certain Ergriffenheit,G to the effect that they greatly influence the Baganda
in their spontaneous religious considerations. One can speak of the lubaale of
the above, or of the below and of a horizontal order. Lubaale or balubaale
(plural) properly speaking constitute the deity of the Baganda. In the 'above',
that is in the firmament and beyond, Katonda is the 'Creator', Ggulu is the
heaven and Kiwanuka is the controller of lightning and thunder.
Katonda was traditionally known as lubaale. Although he was enumerated
among the rest of the Ganda pantheon, Katonda held a special standing
among the Baganda. He was spoken of as the father of gods, because he had
created all things. His official estate on earth is the village of Butonda in
Kyaggwe country. There still stands a temple built for him and those who
wish to petition favours from him may still visit this place. He is a kind and
generous power with whom people may enjoy joking relationships compar-
able to those between grand-parents and grand-children according to the
GANDA TRADITIONAL BELIEF 201
customs of the Baganda. He is not a God of Hell-fire but one from whom all
redemptive influence is supposed to flow.
From time immemorial, Katonda has been known under several other
names denoting his attributes.
Liisoddene -the Big eyed
Kagingo -The Master of life. This is supposed to be a very ancient
name for God in Buganda.8 It is accompanied by idiomatic
expressions stressing the power of God; such as Kagingo
z'agera tezinatuuka-'God's moment' (something to take
place, to die) 'has not yet arrived', which was said to show
God's power in controlling life and time; or Kagingo
tatangwoa-'God is never deterred by anything' which was
said in order to assert the omnipotence of God.
Mukama -the Master
Sewannaku -the External
Ddunda -the Pastor
Lugaba -the Giver
Ssebintu -the Possessor of all things.
To show his suppremacy and excellency it is to Katonda that the Baganda
paid special hommage by invocation on awakening in the morning. It was
from him that the head of a family asked for blessing on behalf of the house-
hold. Such prayers were directed to him: Alyi Kiatonda Ssewannaku, Kaging~o
N~nyini Ggulu, ggwe omuzadde ataboola, kumakuma abanabo, abatambuze badde nga
tebaliiko kabi . .. ..-"Oh, Katonda, Ssewannaku, Kagingo, Possessor of
Heaven, the parent who never disowns any of your children let the traveller
return safely home."g Even in case of consultations with other Ilubaale supplica-
tion was made first by invoking Katonda.'o
Unlike the deified heroes who were often approached with slavish awe,
people thought of Katonda with a childlike confidence. They knew Katonda
tatta-'he does not kill.' He is traditionally considered as a providentially
kind Master to his needy creatures, as the Baganda have preserved in their
Katonda ky'aterekera omulamu te~kivunda--'What God put in store for someone
never goes rotten.'
Ezinunula omunaku Katonda azitunga kiro--'God redeems the afflicted
according to his will.
Walumbe, in so far as he descended from heaven, as we have seen in the
aetiological myth, bringing death with him to earth, may be considered as the
lubaale or 'deity' of the below. His function in society is a destructive one.
He occasions people to turn to other spirits for salutary assistance.
Historically the lubaale of the horizontal order are mainly those powerful
men who in the traditions of the country are said to have lived in the Sese
Islands of Lake Victoria. The reason for their coming to the mainland was
strictly a redemptive one. Nakibinge the eighth King of the Ganda dynasty,
got into trouble with the Banyoro who had always been the enemies of the
Baganda. Nakibinge then had recourse to the authorities of Sese Islands for
help. It is on this occasion that those strong men found their way to the main-
land coming to rescue Baganda from the enemies. These same heroes when
they passed away were raised to the ranks of lubaale. Most outstanding of these
202 UGANDA JOURNAL
are M~ukasa, Wanga and Kibuka. For brevity's sake one may single out Kibuuka
as the best representative of the Ganda lubaale concept of redemption.
Before Kibuuka could leave the islands King Nakibinge had to hand over his
twin" to the authorities of the islands as a surety. This would be redeemed
after Kibuka's task had been terminated. Before sending him off Mukasa told
his brother Kibuka to be careful in his dealings with the enemies, and never
to let them know where he took up his position in battle. He further strongly
warned him not to have any dealings with the Banyoro women. Kibuka's
arrival in Buganda caused great pleasure to King Nakibinge who then felt
confident of victory as a redeemer had arrived. Kibuka took up his position as
general of the army; he flew up into a cloud and hovered over the enemy;
during the battle he shot down arrows and spears upon them, while the Ganda
army pressed them in front. The battle ended that day in favour of the Baganda.
Their enemies withdrew to a safe distance to consider what they should do
next. Among the war captives there were some women; one of these took
Kibuka's fancy. He had her sent to his residence. Kibuka become the Samson
of Uganda. At night the woman escaped having discovered who Kibuka was,
and where and how he posted himself during the battle. She carried the inform-
ation to her people who in the next battle sent a volley of arrows into the
cloud by which Kibuka was mortally wounded. He fled way in his cloud to a
place called Mbaale in Mawokota county where he alighted on a large tree
and died. The tree is there still venerated even to day. This became a holy
place, a shrine was built for the remains of Kibuka. Kibuka became a symbol of
salvation to the nation. He became the national god of war. During the Christ-
ian era Kibuka's temple was burnt down, his remains were handed over to
a Missionary anthropologist, John Roscoe who in turn sent them to England
in 1906 where they had been on show in the Museum of Archeology and
Ethnology in Cambridge till they were returned to Buganda on 2 July 1962,
accompanied by the then Buganda Minister of Education.'2 As those remains
were returned on the eve of the country's political independence, it was inter-
preted by traditionally minded people as a redemption of the highest degree.
Today Kibuka lies in state in the Uganda Museum.
As hinted above, in times of crisis the Baganda have often turned to
traditional belief and practices for consolation if not for salvation. When in
1953-1955, the Kabaka of Buganda was deported to Britain by colonial
administrators for political reasons many traditionalists rallied around a
reincarnation of Kibuka in the person of Kiganira who claimed he would
redeem the Kabaka and eventually the country. The King's return to his
country in 1955 was attributed to the powers of the National god who in turn
received strength from the arch-lubaale, Katonda. From the following prayer
composed under political uncertainty of recent times one may sense the type
of Redemption the Ganda traditionalists have in mind.
"O Muwanga, who ordered this land we kneel and pray Thee so to
order it again as Thou didst order it in ages past. Guard our Kabaka
and the T~hrone, together with the ancient traditions of our "nation".
that they may again become as they were in the beginning. For ever and
O blessed Muwanga have mercy upon us.
O blessed Mukasa of the lake have mercy upon us.
GANDA TRADITIONAL BELIEF 203
O all ye blessed ones of old, who disappeared, have mercy upon us,
We beseech you.
O all ye blessed ones of old, who fled because you feared the atrocities
of the "foreigners",
Arise and join with us that we may restore our land and establish it -as it
was of old.
For ever and ever. Amen."
In this passage it is clear that supplication is for social order, social security
and peace according to traditional lines. In spite of the belief in immortality
as one observes it in the Ganda consideration of the spirits of the departed,
there is no clear indication as to whether redemption for life hereafter is part
of Ganda traditional belief. This may be seen only in an implicit way. The Ganda
concept of Redemption is not concerned with the individual person. It is
concerned with the community. Someone committing an outrageously exce-
ssive fault such as incest is considered to have polluted the community to which
he happens to belong. For that fact he is either disowned by his own people,
measures have to be taken whereby he may be redeemed back~ to society. This is
done by having recourse to the priest who in turn introduces the culprit to the
medium who speaks with god as intercessor on behalf of the sinner. The effective
act of such consultation with the god renders the person in fault acceptable
again into society and order is supposed to have been restored to the particular
It might be true that in contemporary Uganda, publicly, traditional
religion appears to be relegated to the realm of archaism. Yet a redemptive
efficacy is still to be sought through powers which can be considered as tradi-
tional beliefs. If a proper understanding of such words as Ujamaa, African
socialism, African Personality, La Negritude is to be developed, it is necessary
to have a clear grasp of what African traditional belief has to say on the concept
i Parrinder, G3. African traditional religion, p. 9.
2. Tourigny, Y. Eum~enism in Africa, Rome, 1966, p. 99 basing an estimate upon the 1959
census believes that "the Christian population in Uganda is slightly under 60% of the total
population of some 7,000,000 people; 33% Roman Catholic and 25% Protestant. Apart
from 5% who are Moslems, the rest of the population belongs mostly to the local traditional
3. Gardet, L. on Dini in Encyclopedia of Islam, vol. 2, edited by B. Lewis and others, London,
4. Kagwa, A. Engero z'Abaganda (Legends of the Baganda) pp. 1-8 and Basekabaka ba Baganda
(Kings of Buganda) pp. 1-2. and Roscoe, J. The Baganda, pp. 460-464.
5. A. M. Mackay, by his sister, p. 150 (modified).
6. Jensen, A. E. Myth and cult among primitive peoples, pp. 3-4. Ergriffenheit has been translated
as "seizure" in various English discussions of the work of Frobenius, one of the pioneer
serious writers on Africa.
7. Kagwa, A. Empisa z'Abaganda (Customs of the Baganda), pp. 215, and 266; Roscoe,
op. cit. p. 312; Munno, Katonda Waffe ow'edda mu Buganda (Our Katonda of old in
Buganda) 1916, pp. 176-177; Nsimbi, M. Amannya amaganda, passim, and Thomas, H. B.,
The doctrine of God in Uganda, in African ideas of God, edited by E. W. Smith, London,
1950, p. 202.
8. Gorju, J. Entre le Victoria, I'Albert et l'Edouard, p. 195.
9. Munno, of. cit, pp. 176-177.
10. Gorju, of. cit, p. 197.
I1. By "twin" is meant an umbilical cord of a king which was jealously preserved and consi-
dered as his "alter ego".
12. Welbourn, F. B. Kibuuka comes home, Transition, no. 5, 1962, pp. 15-17 and 20.
13. Welbourn, F. B. Religion and politics in Uganda, p. 44.
204 UGANDA JOURNAL
NOTES ON CONTRIBUTORS
John Hall was a Science master at King's College Budo from 1960 to 1964
when he became a lecturer in education at Makerere, since 1967 he has
been lecturing in education at the University of Newcastle upon Tyne.
He has completed a thesis on the biology of weaver birds and has written a
biology text-book for East African schools.
Mr. A. Hamilton is a research student in botany at Makerere University College,
and is working on aspects of the early vegetational history of selected parts
Eric Lanning is a regular contributor to the Uganda Journal, and has contributed
many studies on Mubende, where he served for many years during his
service with the Uganda Administration from 1948-1958.
Father A. Lugira is a lecturer in religious studies at Makerere College.
Brian Mottram is now the headmaster of Katunga School, Ankole, and had for
many years been the geography master at Ntare School, Mbarara. He has
contributed an article on an aspect of the geomorphology of Ankole in an
earlier issue of the Journal.
Merrick Posnansky has contributed to the Uganda Journal many times before,
and is a Past-President of the Uganda Society. Since last writing for the
Journal he has left Uganda to become the Professor of Archaeology at the
University of Ghana.
Alan Walker lectures in primatology in the Department of Anatomy, at the
Medical School, Makerere University College. After taking a degree in
geology he obtained a Ph.D. in Anatomy and Palaeontology with work on
the extinct fauna of Madagascar.
Uganda Journal, 32, 2 (1968), 205-211.
THE DISTRIBUTION OF CERTAIN COLONIAL WEAVER BIRD
SPECIES IN UGANDA
By J. R. HALL
This preliminary survey on the distribution of certain species of colonial
weaver birds was undertaken in May and June 1965. Although the work is
inconclusive, it may form a valuable starting point for more detailed study later.
The species concerned are Ploceus cucullatus (Muller) and P. nigerrimus
(Vieillot).L These are both colonial species and their ecology and general
habits are very similar. P. cucullatus, the Black-headed or Village Weaver, is
probably the commonest weaver bird over a large part of Uganda. This is
certainly true in the central area around Kampala. It is a typical yellow and black
coloured weaver in which the male in breeding plumage has the front part of
the head black as far back as the occiput. As in many related species the female
is smaller and duller than the male with no black in her plumage. In both sexes
there is a non-breeding plumage in which the bright yellow and black dis-
appears and males and females become difficult to distinguish. P. nigerrimus,
Vieillot's Black Weaver, is the only weaver common in Uganda in which the
male is entirely black. The female is a sparrow-like bird which the inexperienced
may find difficult to distinguish from the female P. cucullatus. There is no non-
breeding plumage in P. neg~irrimus.2
The very marked colour difference between the males is apt to suggest to
the uninitiated that the species are not closely related. In most other ways they
are very similar. Thus in the details of breeding behaviour they differ little and
the nests are much alike in shape and in manner of construction. They choose
similar nesting sites and indeed it is common to find the two species building
nest colonies in the same tree. Both are believed to be predominantly granivorous.
These weavers normally inhabit tree savanna or cleared forest areas, where
there are trees providing suitable sites for colonies and where the grass pre-
sumably provides the major source of food in the form of grain. The Black
Weaver requires a more humid environment than the other species. Thus
Crooks gives a map showing the continental distribution of the two species
and this shows that P. cucullatus is widely distributed over the whole of Africa
south of the Sahara from about 15oN. to at least 20"S., being absent only from
the most arid regions, whereas P. nigerrimus has a much more restricted range.
It is confined, roughly speaking, to those regions where forest occurs, or did
occur in the recent past. It will be shown that this difference is reflected in the
more local distribution of the species within Uganda.
In considering the evidence relating to the distribution of the birds which
is described below the possible influence of the following factors may be borne
in mind. First the large scale distribution which has just been described indicates
a relationship to the humidity of the environment which differs in the two species.
Secondly availability of suitable kinds of food might have an inf luence. It appears
that no detailed study of the food preferences of P. cucullatus and P. nigerrimus
has been made, but it might be supposed that preferences for the grain of certain
grass species might lead to concentration of populations in certain kinds of
206 UGANDA JOURNAL
vegetation. Thirdly, distribution may belimited in some cases by lack of suitable
breeding sites. Colonies are commonly found in the following types of site:
(i) thorn trees, (ii) trees near dwellings, (iii) trees situated in swamps or over-
hanging water, (iv) less commonly, in exceptionally tall trees. These all appear
to provide some measure of protection from predation, and colonies are rarely
if ever found in other situations. In the area surveyed, thorn trees (all species
of Acacia) form an important element in certain plant communities only and
over large areas where these weavers are common the tree-by-house type of
site is much the most frequent. This is typical of zones of more intensive cul-
tivation. The third type of site is noted much less often, but swamps, lakes and
rivers are widely distributed over the whole area.
Colonies near the road are easily observed when travelling by car at
moderate speed and the numbers seen in consecutive ten mile stretches of road
were counted. In this way an approximate measure of population density was
obtained, which was expressed as the number of colonies in each 10 mile unit.
Admittedly this can be no more than an approximate estimate, but it does
make it possible to distinguish between zones of relatively high and relatively
low concentrations of the weaver population with some degree of confidence.
The roads traversed in this survey are shown in the figure. Continuous lines
indicate where there were one or more colonies every 10 miles, the dotted lines
where there were fewer.
When observing colonies in this way it was possible to identify species
only when birds were present, since, although there are differences in nest
form, these are not easily seen when viewing a colony briefly from a distance.
Accordingly species were identified in a minority of cases only. Out of 55 colonies
where identifications were made only 2 contained species other than P. cucullatus
or P. nigerrimurs. Accordingly the survey may be regarded as providing infor-
mation about distribution of these species only.
There was evidence that P. nigerrimus colonies were restricted to the zones
in which forest occurred. The shaded areas in the figure represent the distribu-
tion of forest and forest-savanna mosaic.* The locations of identified nigerrimus
colonies are shown in the map. Every one of these existed within regions con-
taining forest. By contrast out of 44 cucullatus colonies identified 21 were not
situated in forested areas. In fact nowhere did nigerrimus exist in the absence of
cucullatus and even where it was most numerous it still formed a minority of the
colonies. Thus in the region of Kampala 29 colonies of the Black Weaver were
recorded compared with 52 of the Black-headed Weaver. Accordingly if the
distinction between the species is ignored the distribution of colonies revealed
in this survey may be taken to indicate the distribution of cucullatus.
(a) Kampala-Masindi-Gulu. From Kampala to a point about 40 miles
north the road passes through a region consisting mostly of forest-savanna
mosaic and weaver colonies are common. They were recorded at intervals of
about one mile. Further north the vegetation changes and the rest of this part
of the survey was remarkable for the extreme scarcity of colonies. For the 270
miles of road before Gulu only 20 colonies were noted and 5 of these were seen in
the first 30 miles, the remainder being scattered through the remaining dis-
tance. The last colony was one of Black-headed Weavers over the Karuma falls.
The most likely explanation of this scarcity of colonies was lack of suitable
breeding sites. The human population is very sparse throughout most of the
distance, so that there are few houses with neighboring trees. Only 2 colonies
were seen in trees near houses after the forest-savanna mosaic zone had been
left behind. No colonies were seen in thorn trees and the Combretumr savanna
DISTRIBUTION OF PLOCEUS CUCULLATUS AND P.NIGERRIMUS
(BREEDING COLONIES) IN RELATION TO FOREST
%Lyontonds Wosaka It
ROADS TRAVERSED DURING SURVEY
SOne or more colonies per 10miles
..... Less than one colony every 10miles
Forest and forest- savanna maosai
A- E Areas of tow rainfall
A APositively identified P.nigerrimus
Approximate scale: 0 20 40 60 80 100
208 UGANDA JOURNAL
which occupies most of this stretch of country is of a type in which thorn trees
may be present without forming a major component of the tree layer. Of the
remaining 8 colonies recorded 2 were in tall trees and the rest over water.
(b) Kam~pala-Fort Portal. From Kampala to Mityana the road again passes
through part of the central forest and forest-savanna mosaic zone where weaver
colonies were numerous. In the first 10 miles after Mityana, where the amount
of cultivation was considerable, colonies were seen frequently. Thereafter
cultivated land was replaced by scrub with very little cultivation and for the
next 30 miles approximately no weaver colonies at all were seen. The vegeta-
tion here consists of Combretum-Cymbopogon savanna, which contains few thorn
trees, and gives way to a type of moist Acacia savanna in the last 10 miles. Not until
cultivation and houses became more frequent on approaching Mubende did
more colonies become evident. At the same time the vegetation changes to
forest-savanna mosaic. Throughout the rest of the way until a point about 10
miles from Fort Portal colonies were seen at variable frequency, although they
never became very scarce. The vegetation is a patchwork of different communities
consisting mostly of forest and forest-savanna mosaic with some thickets and
savanna. Near Kyenjojo there is an area of mixed savanna containing some dry
Acacia savanna which is not present in other areas, and here colonies were most
frequent, occurring at intervals of about amile. in the immediate neighbourhood
of Fort Portal colonies were entirely absent. This is perhaps surprising because
the environment there is closely similar to that near Kampala and other formerly
forested regions where there is much cultivation and many dwellings, and
where these weavers were numerous.
(c) Fort Portal-Mbarara. Colonies were first seen about 25 miles south of
Fort Portal and from that point they were frequent for the next 35 miles. They
were particularly numerous in certain places near Kasese where as many as
20 colonies were counted in 4 miles. The great majority were in thorn trees and
the vegetation map shows that here the road passes along the boundary between
a zone of dry Acacia area, where it entered the flat plain of the rift valley, there
was a marked decline in the number of weaver colonies, which coincided with
a decrease in the number of suitable nesting sites provided by thorn trees or
trees near houses. In fact from a point about 30 miles north of Kasese to the
Kichwamba escarpment houses are very scarce and it is clear that the distri-
bution of weaver colonies is very largely determined by availability of thorn
trees. Above the Kichwamba escarpment the road passes through a valley
leading up into the west Ankole highlands and here weaver colonies were quite
numerous in an area of forest-savanna mosaic with much peasant cultivation.
After leaving the valley only one weaver colony was seen until a point about ten
miles east of Bushenyi, where there were four within a fairly short distance.
In the highland area there is a considerable amount of cultivation and the
vegetation is dominated by a post-cultivation community. There must be
numerous possible sites for weaver colonies provided by trees near houses and
their scarcity must be accounted for by some other factor. The last 20 miles of
the route was completed in darkness and no observations could be made.
(d) Mbarara-Kigezi border. This is an area dominated by grass savanna,
~Themeda-Chloris and Themeda-Loudetia. Near Mbarara there is widespread culti-
vation, whilst further south the road passes through cultivated land in the valley
bottoms. Weaver colonies were observed throughout, but their frequency was
somewhat variable. It is suggested that the presence of the weavers is related
to the savanna rather than to cultivation. Thus near Mbarara, where there is
little savanna left, weaver colonies were scarce, whilst elsewhere the area of
WEAVER BIRDS 209
cultivation available to any one colony must usually have been very small and
the savanna close at hand and extensive. On the other hand the houses associated
with cultivation are undoubtedly important in providing protected sites; only
3 out of a total of 22 colonies were noted as being definitely away from dwellings.
Colonies became really frequent only near the Kigezi boundary. It may be sign-
ificant that this coincided with the only area of dry Acacia savanna encountered.
(e) Kigezi. The boundary marks an abrupt change to moister, cooler
conditions with much cultivation in a formerly forested area. Weaver colonies
of both species were seen, but they were not very frequent, averaging about 2
every 10 miles.
(f) Mbarara-Masaka. Only one colony was seenl near Mbarara and this
was on the very edge of the cultivated zone. From here until the area of forest-
savanna mosaic near Masaka is reached the vegetation consists almost entirely
of Acacia savanna. It might have been expected, from what was seen in the rest
of the survey, that weavers would have been numerous in this region, but in
fact they were not. Indeed, until Lyantonde was reached only one more colony
was recorded. This was in a tree on an isolated plot. This is readily explained
by the fact that all savanna trees have been felled in the area in connection with
tsetse control operations, whilst at the same time there is scarcely any: cultivation.
From Lyantonde onwards the road passes through considerable areas of moist
Acacia savanna mostly in the valleys. It was noticed that although this vegetation
was full of thorn trees there were no weaver colonies in them at all. This contrasts
strikingly with the situation near Kasese described in paragraph (c) above.
The intervening areas of dry Acacia savanna were to quite a large extent cultivated
and here a few colonies were seen, always near houses. From a point about 20
miles from Masaka colonies became more and more frequent and about one
every mile was recorded in the forest-mosaic region which starts about 10 miles
from the town.
(g) Masaka-K~ampala. Almost the entire length of this section of road
passes through forest or forest-savanna mosaic. A notable exception is where
a very flat, low-lying area is crossed about 15 miles north of Masaka. Here
only one weaver colony (P. cucullatus) was seen in about 10 miles. Otherwise
they were recorded with a fairly constant frequency of about one per mile.
(h) Kampala- Tororo. Here weaver colonies were seen at a frequency of
about one per mile from Kampala to a point about 25 miles from Tororo, as
the road passes most of the way through the typical forest and forest-savanna
mosaic vegetation of central Uganda. None were seen after that point, and it
is no doubt significant that it coincides approximately with the edge of the
(i) Tororo-Mbale-Soroti. Colonies were seen at a frequency of about one
every three miles from Tororo to about 25 miles north of Mbale, but there-
after were infrequent. It seems impossible to relate this to changes in vegetation,
but possibly it may be related to conditions of higher rainfall and lower tempera-
tures near Mount Elgon.
The results of the above survey suggest possible relationships between
the distribution of colonies of Ploceus cucllatus and P. nigerrimus and various
environmental factors. It would seem to be fairly certain that the latter species
is confined to forested or formerly forested areas, whereas the former is more
widely distributed, though neither species inhabits the forest itself. Clearly the
distribution of suitable breeding sites is a controlling factor; but other relation-
ships are much less certain. Accordingly some of the remaining possible factors
will be discussed in an attempt to summarise the evidence relating to their