Front Cover
 Title Page
 Table of Contents
 Status conflict: Cattle and education...
 Mobility and village composition...
 The clan system among the...
 The struggle for dominance in Bufumbira...
 The grasses of Queen Elizabeth...
 The maidenhair ferns of Uganda
 A fossil incisor from Lake Edward,...
 Uganda bibliography, 1969-1970
 Back Cover

xml version 1.0 standalone yes
PreviousPageID P34

The Uganda journal
External Link ( Journal issues, 1999-2002 )
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00080855/00053
 Material Information
Title: The Uganda journal
Abbreviated Title: Uganda j.
Physical Description: v. : ill. ; 22-24 cm.
Language: English
Creator: Uganda Society
Publisher: Uganda Society etc.
Uganda Society etc.
Place of Publication: Kampala etc
Frequency: irregular[1976-<1995>]
semiannual[ former 1934-]
annual[ former <1948>-1973]
completely irregular
Subjects / Keywords: Periodicals -- Uganda   ( lcsh )
Genre: serial   ( sobekcm )
periodical   ( marcgt )
Spatial Coverage: Uganda
Dates or Sequential Designation: v. 1- 1934-
Numbering Peculiarities: Suspended May 1942-Mar. 1946; replaced by the Society's Bulletin.
Numbering Peculiarities: One issue published every few years, v. 38 (1976)-<v. 42 (1995)>
Issuing Body: Vols. 11-13 published in London.
General Note: Journal of The Uganda Society (formerly The Uganda Literary and Scientific Society).
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - ABS8002
oclc - 01644239
alephbibnum - 000301510
issn - 0041-574X
lccn - 52026895
System ID: UF00080855:00053
 Related Items
Related Items: Uganda journal


This item has the following downloads:


Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
    Status conflict: Cattle and education in Ankole
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
    Mobility and village composition in Bunyoro
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
    The clan system among the Banyankore
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
    The struggle for dominance in Bufumbira 1830-1930
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
    The grasses of Queen Elizabeth National Park
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
    The maidenhair ferns of Uganda
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
    A fossil incisor from Lake Edward, and a method for its determination
        Page 75
        Page 76
        Page 77
        Page 78
        Page 78a
        Page 78b
        Page 78c
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
    Uganda bibliography, 1969-1970
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text

The Uganda Journal


VOLUME 34 LT1 1970

IN BUFUMBIRA 1830-1930 --- P. MATEKE 35
Area handbook for Uganda (By A. B. Herrick) R. T. CURLEY 89
Uganda timbers (By C. H. Tack) F. WOOLFENDEN 90
Field guide to the butterflies of Africa
(By J. G. Williams) A. W.R. McCRAE 91
SOCIETY NEWS - - - - - - 47
Compiled by B. W. LANGLANDS 95

Published by


His Excellency the President of Uganda, Dr. A. Milton Obote.

Professor B. W. Langlands

The President
The Vicer-President
The Hon. Secretary
The Hon. Treasurer
The Hon. Editor
The Hon, Librarian
Dr. W. B. Banage
Mr. H. J. Bevin
Mr. J. E. Compton
Mr. J. L. Dixon

Hon. Secretary:
Hon. Treasurer.
Hon. Editor:
Hon. Llibrarran:

Hon. Auditors:

Messrs Cooper Brothers & Co.

Vice -Preside nt.
Dr. F. Sempala-Ntege


Professor H. EI-Abd
Mrs. D. hM. Etoori
Mr. R. Frankum
Mr. S. C. Grimley, M.B.E.
Mr. W. S. Kajubi
Mr. D. Kavuli
Mr. J. Kingdon
Mr. A. W. R. McCrae
Mrs NM. Macpherson
Mr. H. Sassoon

Mr. J. P. OcILtu
Mrs. J. Bevin
Professor B. W. Langlands
Mrs. F. Wapenyi

Hon. Legal Adviser:

Mr. R. A. Counihan

Hon. Vice-Pre'dents & Hon. Lile Members:

Sir Tito Winvi Galabusa C.B.E.
Reverend Father J. P. Crazzolara
Captain C. R. W Pitman, c B.E., D S.O., M c.

1933-34 Sir A. R. Cook. c.M.o., O.B.E.
1934-35 Mr. E. J. Wayland, C.B.E.
1935-36 Dr. H H. Hunter, C.B.e., L.L.D.
1936-37 Dr. H. Jowitt, c.Ml.o.
1937-38 Sir H. R. Hone, K.O.S.,K.B.C.,
M.C., Q.C.
1938-39 Mr. Sykes, O.B...
1939-40 Mr. N. V. Brasnett
1940-41 Captain C. R. S. Pitman, C.B.E.
D.S.O., M.C.
1941-42 Mr. S. Kulubya, C.B.E.
1942-43 Mr. E. A. Temple Perkins
1943-44 Mr. R. A. Snoxall
S1944-45 Dr. K. A. Davies, c.M.o., o.B.e.
1945-46 Mr. G. H. E. Hopkins. O.B.E.
1946-47 Mrs. K. Mh. Trowell, M.B.E.
1947-48 Dr. W. J. Eggeling
1948-50 Dr. G. ap Griffth

Mr. E. B. Haddon
Mr H. B. Thomas, o.B.E.
Professor A. W. Williams. c.B.r.

Past Presidents:
1950-51 Rev. Fr. F. B. Gaffine, w.F.
1951-52 Professor A. W. Williams, C..E.
1952-58 Sir J. N. Hutchinson, C.M.F., F.R.S.
1953-54 Mr. J. D. Jameson, o.B.e.
1954-55 Dr. Audrey I. Richards, c.B.E.
1955-56 Rev. Dr, H. C. Trowell, o.B.E.
1956-57 Mr. D. K. Mlarphatia. M.B.E
1957-58 Mr. M. Barrington Ward
1958-59 Dr. H. F. Morris
1959-60 Professor A. W. Southall
C1060-61 Mr. J. C. D. Lawrance, o.n.E.
1961-6" Mr. B. E. R. Kirwan. M.BL.
1962-64 Mr. W. S. Kajubi
1964-65 Dr. NM. Posnansky
1965-66 Dr. W. B. Banage
1966-67 Professor S. K. Baker .B.E.
1967-68 Dr. AM. S. NM. Kiwanuka
1968-69 Mr. J. L. Dixon
1969-70 Mr. P. Zirimu

Mr. P. N. Kavuma

Mr. E. Kironde

Rev. K. H Sharpe

Secretary: Mrs. J. Bevin



The Journal of the Uganda Society


Published By

P.O. Box 4980

Including Uganda Bibliography
published in association with
Milton Obote Foundation.

Uganda Society 1970

Vol. 34

Part 1


Area handbook for Uganda (By A. B. Herrick) R. T. CURLEY 89
Uganda timbers (By C. H. Tack) - F. WOOLFENDEN 90
Field guide to the butterflies of Africa
(By J. G. Williams) A. W. R. McCRAE 91
SOCIETY NEWS - - - - - - 47
Compiled by B. W. LANGLANDS 95

Uganda Journal, 34, 1 (1970) pp. 1-14



In 1960, for the East African Institute of Social Research, A. J. Maleche
and F. K. Kamoga undertook a survey of wastage in primary schools in
Buganda, Teso, Toro and West Nile. Their aim was to locate as far as
possible the fundamental causes of wastage; and they listed, as the main
general causes, ignorance, poverty, shortage of good calibre teachers, and the
lack of an effective administrative educational policy.1 The present study is
based on a different approach to the problem, for its central concern is the
local community rather than the school; and it is intended to indicate some
factors of home background which encourage or inhibit school attendance.
The material was collected during July to September 1969 from three separate
communities within the same general area of Rubaare Gomborora in central
Kajara, Ankole. This is one of the principal grazing areas of Ankole and so
allowed comparison between cattle-keepers and cultivators.
The object of the investigation was to study variation in school attend-
ance, firstly to see how far individual attendances within a single community
could be correlated with family background and occupation, and secondly
to determine the effect of location and environment as reflected in attendance
differences between communities., The v village (ekyaro) was an appropriate
community unit for study because it has definitive natural boundaries and
because it is usually sufficiently small for a complete homestead survey to
be practicable. The choice of Kagongi, Kabuye and Nyakasa villages (see
map) was, however, fairly arbitrary.2
In the course of the study there emerged what seemed a relatively
high incidence of parental refusal to send children to school. This made
invalid an early assumption 'that education was a scarce good in general
demand. In some areas, in fact, it was clearly more sensible to account for
cases of attendance rather than cases of failure to attend school. Antipathy
towards education is clearly a problem in the context of a developing nation.
The following exposition is concerned primarily to indicate some correlates
of such antipathy in rural Ankole and secondarily to account for consider-
able variation in school attendance within a fairly small area.
There are two prevalent attitudes which help to explain popular response
to education. One derives from tradition; the other from a widespread be-
lief about the purpose of education. To understand the first it is necessary
to give a brief historical account.3 The present district of Ankole was for-
merly divided into a number of small semi-autonomous kingdoms in which
there was a common pattern of social stratification: the pastoral Bahima
dominated the agricultural Bairu. This pattern is explained and perpetuated
in a Hima myth, variants of which are widespread in the Interlacustrine
area. Ruhanga, the Creator, had three sons, Kakama, Kahima and Kairu.
He set them a test4 so that he could judge their respective fitness to rule
after him; and each according to his performance was assigned an irrever-
sible occupational role: Kakama represents the ruling dynasty, Kahima the


pastoral aristocracy and Kairu the cultivating serfs. The term Bairu was
applied by Bahima to the indigenous Bantu whom they subjugated, and
it carries a number of ascriptive labels such as short, heavy, dark-skinned,
in contrast to the tall, slender, lighter-skinned Bahima. In order to perpetuate
the relationship Bairu were not allowed to own productive cows, i.e. they
could neither accumulate wealth nor freely inter-marry with Bahima. The
term therefore has a double reference: 'servants' and 'those who have no
cattle.' This explains the traditional equation of lack of cattle with low
social status, which has a significant influence on attitude to education. It also
explains why 'Bairu' is used only as an external reference term, and cultiva-
tors today refer to themselves as 'Banyankole.' The term 'Bairu' is used here
only to distinguish from 'Bahima' and is not connoted with inferiority in any
The other prevalent attitude is summed up in the belief that the only
purpose of education is to enable the child to accumulate wealth later in life.
The logical corollary of this is readily apparent: those who wish to better
themselves by educating their children in many cases cannot afford it. The
figures for school attendance might be expected to reflect this, and indeed
a popular stereotype is that the son of a rich man idles his time away at
home awaiting his inheritance, but the successful secondary school boy comes
from a poor home. Detailed discussion below however will show that some
practical qualification of this stereotype is necessary.

Tables I and II give the ethnic distribution per village of homesteads
and population, respectively, in July 1969.

Table I Ethnic distribution of homesteads.

Total Ihu Tutsi Hima Other

Kagongi ... ... ... 31 23 3 4 1
Kabuye ... ... ... 36 34 1 1
Nyakasa ... ... ... 62 59 1 2 -
Total ... ... ... 129 116 5 6 2
Percentage ... ... ... 100 90 3.9 4.7 1.4

Table II: Ethnic distribution of population.

Total Iru Tutsi Hima Other

Kagongi ... ... ... 206 128 28 47 3
Kabuye ... ... ... 341 316 13 12
Nyakasa ... ... ... 470 439 17 14 -
Total ... ... ... 1017 883 58 61 15
Percentage ... ... ... 100 86.8 5.7 6.0 1.5


It can be seen that Bairu heavily predominate numerically. Some Bahima
have recently moved from this part of Kajara to parts of Nyabushozi where
ranching schemes have been initiated following tsetse clearance. The Ba-
tutsi have been established for some time. They are immigrants of at least
one generation and lead a similar life to the Bahima, though they maintain
distinctive dress and speech. Comparison of Tables I and II indicates that
on average Hima and Tutsi homesteads are larger than Iru ones, which is
consistent with the typical residential unit in each case. The cattle-keepers
generally live in extended family groups looking after one large herd; the
cultivators are usually self-sufficient nuclear or compound5 families.
Kagongi lies in a shallow east-west valley 1km off the Rubaare-Rwa-
shamaire road (see map). It is divided, according to agreements made with
the muruka chief, into cultivating area the bottom of the valley and the
north side and grazing area the south side. There are several large
herds of 50 cows or more, owned by Hima and Tutsi; otherwise plantations
(sing. ekibanja) are small and no coffee is grown. A partial explanation of this
is in the soil type and rainfall. In this part of central Kajara, on the lower
ground away from the hills (e.g. Kagongi) with a base of granite and gneisses,
the soil is usually sandy and thin, and does not hold enough water for surface
crops to survive during the two to three months' mid-year dry season. The
available rainfall figures for the area suggest a deficiency of moisture for coffee
(i.e. less than 100 cm year) at least two years out of five. Kagongi is about
4 km equidistant from Rugarama (Trading Centre and government primary
school P1-7) and Ruyonza (Trading Centre and government primary school
P1-4). Physical distance from schools therefore does not in itself prevent
attendance, though the lack of facilities whose operation is obvious to the
village does inhibit an awareness of education.
Kabuye lies below the high ridge of hills that marks the Ankole-Kigezi
District boundary. The hills probably get rather more rain than the lower
area where Kagongi lies, although exact statistics are not available. There
are extensive alluvial fans and rock waste brought down the slopes by violent
wet season storms. The usual soil type is an unsorted stony clay mixture
in which coffee growing is successful.6 This appears to account for the marked
prosperity in Kabuye which is evident from the greater average size of eki-
banja, the frequency of mabati (corrugated sheeting) roofs and from the high
frequency of polygny, with the resulting large homestead. There is a
small part of the village, however, which contrasts with this pattern in that
the homesteads are obviously poor. They lie on or at the foot of steep screes
where the soil is too rocky to support coffee. The reason for the main differ-
ence between Kagongi (no coffee) and Kabuye (much coffee) was given by
a Kagongi informant as 'climatic' and by a Kabuye informant as 'ignorance'
(on Kagongi's part) both points of view have some truth. Reference to
the map will show that Kabuye is well epuipped with primary schools within
easy walking distance. This has clearly had some effect in encouraging
village awareness of education.
Nyakasa is enclosed within a long valley whose northern end opens out
into the swampy area surrounding Lake Nyabihoko. It is separated from
Ruyonza by a long ride which curves round the southern end. The village
had a much higher concentration of Bahima in the past, but recently most of
these have moved away to Nyabushozi following tsetse reclamation there.



mpumo I' Sub-
. r County
*. .. .. *
3 I I*



&* .***, *.... ,*\
IKIGA. Rh '' Sub-County Kbole

.. -District boundary r i Swamp
. .....Sub-County boundary [ Trading
w Villages 0 1 2 Centre

M Ile

* Government
0 Church School


There is a 'hang-over' of the Hima attitude to cattle in that many Bairu with
cattle in the village now expressed reluctance to sell cows to pay school fees.
Some coffee is grown by individuals but is not as successful as in Kabuye
because annual rainfall is less. Ruyonza primary school (P1-4) is only 2 km
from the southern end of Nyakasa, and most of the children who attend
school go there. Few go to Kanyampumo church school (C.O.U. P1-4), 2 km
from the northern end, because young children fear the buffalo which inhabit
the swamp and incidentally ravage the village crops at night. There is some
difficulty about attendance in classes above P4, since the nearest P7 school,
Rubaare, is 8 km. as the crow flies. Nevertheless four boys do the long return
walk, in practice about 20 km, every day.

Homesteads may be divided into those with children of school age and
those without children of school age. The latter category consisted of home-
steads in which the parents were elderly and married offspring were living
elsewhere, and homesteads in which the children were not yet old enough
for school. The limits of 'school age' are somewhat arbitrarily imposed, as
fathers who 'know the value of education' often send their children to school
at a young age, sometimes as young as six years, whereas in another family
the fact that a child of ten is not yet at school does not necessarily mean
that his father is unfavourable. Seven years seems a reasonable average of
the youngest at school, and the oldest was aged twenty years.

Table III: Number of homesteads per village. A with children of school age
(7-20 years) and B without children of school age.

A B Total
Kagongi ... ... 23 8 31 Number
Kabuye ... ... 31 5 36 of home-
Nyakasa ... ... 48 14 62 steads

Table IV: Distribution per village of children in school 1969.
A: number of children of school age;
B: number of children in school 1969;
C : number of children in school as a percentage of all children
of school age.
Boys Girls Total
Kagongi ... ... A 40 31 71
B 8 8
C 20.0 11.3
Kabuye ... ... A 63 60 123
B 24 11 35
C 38.1 18.3 28.5
Nyakasa ... ... A 92 92 184
B 23 9 32
C 25.0 9.8 17.4


Table V: Distribution per village of homesteads with 0-6 children in school.

Number of child-
ren in school 6 5 4 3 2 1 0 Total
Kagongi 1 6 16 23 number
Kabuye 1 4 5 7 14 31 of home-
Nyakasa 3 3 2 7 33 48 steads

The homesteads without children in school 1969 contained either children
who had never been to school or children who had started but dropped out,
or some of both.

Table VI : Reasons for drop out or failure to attend among homesteads
with children of school age but with none in school 1969.

A: Father too poor to afford;
B: Father refused to send, or withdrew, children;
C: All girls;
D: Children refused to, go, or continue;
E: Children will go, but are still 'too young';
F: Examination failure.

A B C D E F Total
Kagongi ... 7 3 5 1 16 Number
Kabuye ... 5 4 1 2 1 1 14 ofhome
Nyakasa ... 13 8 5 4 3 33 steads

In some areas, where a mission or school has been long established, it
may be assumed that in general people will favour education and will send
children to school when they can afford it. This is because the benefits of
schooling have become obvious over the years in terms of visible return:
white-collar job, financial help or simply high status. Elsewhere, however,
in areas where there have been no missions or schools, there are many people
who are 'ignorant' or 'just don't know the value of education' because the
boundaries of their experience are spatially very limited. They measure
wealth in traditional terms, i.e. cattle, and do not understand the alien
standard of education. It is possible, then, to speak of villages with a
'favourable climate' or unfavourablee climate' of opinion with regard to the
value of sending children to school. There appear to be three main factors
which generate the former climate: 1) long-standing mission influence; 2)
proximity of school facilities; 3) ability to afford education. The first two
are self-evident and often go together. The third requires some explanation
as the operative factor is not actual wealth but the conception of wealth
and the way it affects attitudes. The importance of differential resources as
a determining influence is investigated in some detail below. Meanwhile
it is adequate to suggest that given all three conditions one might expect a
relatively high incidence of attendance for a given village.


Without the first two conditions the probability of school attendance
is hard to assess since it will depend largely on accidental factors such as
local government intervention through the muruka chief or persuasion by a
relative or friend who has personal experience of schooling. This can be
illustrated from the 18 cases of school attendance in Kagongi village, which
has had no mission or school, from 1950 to 1969. Table VII gives the year
of starting, and the number who have dropped out, together with the reasons
for this.

Table VII: The 18 cases of school attendance in Kagongi village from 1957
to 1969.

Year Number Number Number Reasons for drop out
started started continuing dropped out A B C
1957 5 1 4 3 1
1960 8 4 4 2 2
1963 1* 1 1
1965 1* 1 1
1966 2 2 -
1969 1* 1 -

Reasons for drop ouit: A: Father or child refused;
B: Lack of fees;
C: Examination failure.
There were two years, 1957 and 1960, in which the muruka chief, himself
an educated Muhima, came round the village and forced fathers to send
children to school, using the threat of (illegal) imprisonment. The figures
clearly reflected this. Of the 5 who were forced to go in 1957, 3 had left
the same or following year; and of 8 sent in 1960, 2 had left the same or
following year. The 2 who started in 1966 did so because there were close
relatives and neighbours (1 Hima and 1 Tutsi) of 1 boy, a Tutsi, who had
started in 1960 and who passed his P6 leaving examination at the end of
1965 the first tangible result of education in Kagongi. The individuals
marked with an asterisk are all children of the muruka chief who was favour-
able in the first place. It is ironic that his are the only examination failures.
There has been much wastage, although there are now 3 boys in senior
secondary school (all started school in 1960), and 2 others have started in-
dependently. Of the total 18 who began, 13 did so as a direct result of
pressure from the muruka chief, 3 were his own children, and the fathers
of the other 2 were persuaded by a relative. A church school was started
in 1960, with the 8 recruits, but folded up at the end of the year owing to
lack of support. Since then there has been no government intervention to
force children to school. In addition the muruka chief was transferred else-
where in 1967 and the new one, uneducated himself, has no interest in
coercing parents. This renewal of apathy in Kagongi suggests that continuous
propaganda is vital to an appreciation of the importance of basic education.
Kabuye, on the other hand, has three church schools within easy walking
distance: Kashenyi P4 (started 1954), Ngomba P1 (1957) and Murambe
P4 (1960). And there has been a school at Rugarama, within sight of Kabuye,


since the 1930's. This became a government school in 1962 and is now one
of the two complete primary schools (P1-7) in Rubaare Gomborora. Kabuye
is also relatively prosperous from coffee which has developed increasing im-
portance in the village since it was introduced in 1927. Thus conditions exist
for expectation of a high school attendance rate. According to the data,
more than half (17/31) of the homesteads with children of school age have
at least one child in school, which compares very favourably with 7/23
from Kagongi and 15/48 from Nyakasa (see Table V). 3 out of the 4 cases
of parental refusal (see Table VI) can be accounted for by temperamental
particularity in that the 3 fathers concerned are known as excessive drinkers;
and 1 of the 2 cases of child refusal was an only son wanting to get married.
Response in Nyakasa is affected by a previously high concentration of
Bahima. The attitude has remained behind them that schooling is a poor
substitute for many cows. One informant, who happened to be the principal
emandwa practitioner (diviner) in the village, lamented his fellows' ignorance
in this respect, while drawing a substantial income in cattle in payment for
his services. He was demonstrating his own 'progressiveness' in spite of making
a living out of traditional belief. The attendance figures for Nyakasa in
relation to wealth are considered in some detail below. Kagongi and Kabuye
have been treated here at greater length because they represent extremes:
11.3% and 28.5%, respectively, of children of school age are in school (see
Table IV). Nyakasa lies between these two extremes with 17.4% at school.

Education and Wealth

In order to relate school attendance to family resources7 it was necessary
to find a method of comparing wealth from a cash crop and wealth in cattle.
The only realistic method seemed to be compare poll tax payments. Every
male over 18 years who is neither too old nor in school is liable to pay poll
tax, which is assessed annually on his capital resources: in this area these
consist mainly in cattle and coffee.8 The minimum payment is Shs. 60/-
per year.9
It is suggested that in Kabuye a climate favourable to educa-
tion prevails, so that the better off a particular family is the more children
it is able to send to school. Table VIII gives the average of cows and the
approximate average number of coffee tree groups of homesteads with child-
ren of school age: group A, with no children in school; group B, with one
child in school; and group C, with two or more children in school (see Table

Table VIII: Wealth in Kabuye in relation to number of children in school.
Group and Number of Average no. Approx. average
number of children of cows per no. of coffee trees
homesteads in school homestead per homestead

A 14 0 3 53
B 7 1 6 146
C 10 2 or more 15 246


The table indicates a clear correlation between wealth and number of children
in school. When it is borne in mind that only 10 out of the 36 homesteads
do not grow any coffee and that these 10 are mainly in the poor section of
the village the proposition emerges that those who are sufficiently progressive
to adopt a cash crop such as coffee as a principal means of livelihood also
tend to send children to school.
In order to test whether good attendance figures for Kabuye could be
accounted for by the presence of a profitable cash crop rather than by the
proximity of schools it was necessary to attempt to relate attendance variation
within a single community to family source of wealth. Nyakasa was the obvious
test case, since wealth in this village is derived from both coffee and cattle. The
poll tax returns and the total resources in coffee and cattle of two groups of
homesteads were compared : group a, the 8 homesteads with two or more child-
ren in school, and group b, the 8 homesteads in which the father had specifical-
ly refused to send children to school (see Tables V and VI).

Table IX: Comparison of wealth in coffee and cattle of Nyakasa homesteads
a, favourable and b unfavourable to education.

Group and Total poll Total Approx. total
number of tax paid number of coffee
homesteads by heads of cows trees

a 8 Shs. 610/- 73 1490
b 8 Shs. 620/- 121 290

The head of one of the homesteads in group a has virtually no coffee and
30 cows, i.e. he is an exception to the prevailing pattern of wealth in this
group (much coffee and not many cattle). According to several informants,
although he owns in addition a shop at Kanyampumo and has 2 children in
the church school there, the rest of his 18 children (he has 6 wives) are un-
likely to go to school. The attendance of these 2 is merely a concession to his
favourite wife (she runs the shop), not a demonstration of enthusiasm for
education, because "the man is interested in cattle only." If his case is dis-
counted (his 30 cows being almost half the total in group a) the discrepancy
between the sources of wealth of the two groups is even more striking. The
homesteads with 2 or more children in school have more than five times
more coffee than the homesteads whose fathers 'refused,' whereas the latter
have three times more cattle than the former; but according to poll tax
assessment the total 'wealth' of the two groups is virtually the same. These
data would appear strongly to support the general thesis that coffee and
schooling go together, while cows and schooling do not.
There is further, indirect, evidence from Kagongi. Here there is a
definite poor response to education, partly because of poverty, but partly
because those who can best afford to send children to school are wealthy
cattle-keepers. Of homesteads paying more than the minimum Shs. 60/- poll
tax (9/31) all have sizeable herds and only two have any other source of
wealth: one is muruka chief, the other mukungu (village headman) both
salaried positions. It is clear that the negative response in Kagongi is a


combination of two facts: that Kagongi is definitely poorer than the other
two villages and that such wealth as there is is vested exclusively in cattle.
All of which is not particularly surprising, in view of a still marked
traditional orientation among cattle-keepers, especially Bahima, who retain
a conviction of their cultural self-sufficiency. But two points arise clearly
from the foregoing discussion : firstly that it is necessary to modify the common
stereotype about wealth and attitude to education which was referred to
earlier; secondly that attention should be drawn to the empirical non-conver-
tibility of cows into school fees. Cattle can be effectively regarded as a sphere
of exchange distinct from and more prestigious than the monetary sphere:
conversion downwards, i.e. selling cows for cash, is condemned because the
social value of possessing a cow is greater than its market value of about
Shs. 500/-. In other words, cattle represent appreciating and inconvertible
capital, whereas the proceeds from a cash crop such as coffee are by definition
fluid. Thus antipathy to schooling is not simply a result of backward think-
ing: coffee income is easily translated into school fees, so that a man who
depends mainly on a cash crop has school fees far more readily available than
a man who depends on cattle, though the latter may be wealthier in local
estimation. In terms of i:s effect on school attendance this is rationalized
as follows. Sons of a family with a large herd have guaranteed wealth, and
do not need to rely on schooling to acquire it; but wealth from coffee is in
the form of an annual but variable cash income which is easily dissipated
and cannot be transmitted as 'safe' capital. Sons of a family with coffee
must therefore be sent to school to acquire alternative and marketable skills.

Education and Status

In many parts cattle are still regarded as the outward indicator of standing
in the local community. This is particularly true of Kagongi, where there
are large herds, a specific grazing area and no alternative source of wealth,
and of Nyakasa, where many Bahima were settled a few years ago. 'What
the neighbours think' is as important here as elsewhere, and if a man has to
sell cows for cash he is bo:h actually and conceptually poorer, because a cow
represents lasting wealth and cash can be quickly drunk. Thus 'having cows'
is the most important traditional status index; and this is one factor which,
in the way indicated in the previous section, inhibits a man with many cows
from taking the initial step of sending sons to school. Another factor is his
greater need for young labour, for daily herding duty. Cattle as a status
index and the labour requirement of looking after them, both operate to
prevent children from cattle-keeping families from attending school. It is
difficult to assess priorities however: a man will often rationalize antipathy
towards education in terms of need for herding labour. This need is perhaps
less important than informants state.
Given this situation, it is possible to trace the evolution of village opinion
through time. Perhaps a typical series of events would be the following. A
few children from the village go to primary school, probably some miles away,
as a result of external pressure or persuasion; some drop out after P1 or P2
for one reason or another. But several continue, and when a father sees his
own child doing well his reluctance is overcome and he becomes more willing


to find fees. One boy passes his primary leaving examination and may be
accepted into secondary school. This is the first tangible result, and it per-
suades several other reluctant parents to send their children to school. But
the time-lag is inevitable: the point is that through the early primary years
the parents see much wastage both in fees and in that the child rapidly loses
his rudimentary literacy and numeracy. If a boy continues to P7 and then
drops out, he tends to develop a sort of resentful pride which makes him
unwilling to dig and often alienates him from his contemporaries who remain-
ed at home. He may well leave home and swell the mass of semi-educated
school-leavers loafing about the towns.
What, then, is the purpose of education? Many people have very little
conception of what a child does at school. It is an ill-defined way of earning
money. And only when a boy enters secondary school and begins to own
decent clothes, speak English and import new ideas do they realize there is
something in this education business after all. The parent who has persevered
with paying school fees then begins to feel very proud of having a son in
secondary school, and will tend to shift the credit for initially sending the
boy by the muruka chief. 'Secondary school' remains an alien concept to most
of the villagers, but it acquires a certain mystique and popular respect, so
that a new status index emerges which contrasts with the traditional one of
having many cows. After some time partially educated people begin to
describe rich cattle-owners and others who will not send children to school
as 'ignorant' or 'backward;' and the process of substitution is almost complete.
There is another side of the coin. When only a few boys have reached
secondary school their success is resented by many of their contemporaries
who have failed or dropped out for other reasons. Categorical resentment
in the village context is translated into interpersonal tension which may be
expressed in such terms as the following: "Look at this boy. He is becoming
just like a European. He will get a big job and then despise us. Is that good?"
This was said to be particularly true of inter-clan relations: people remark,
"It is destroying our culture" because it corrodes the traditional ethic of clan
egalitarianism. Such resentment appears to be fairly widespread. Since
jealousy of secondary school success might be assumed to imply awareness
of the importance of education it might have been less marked in areas with
limited attendance, such as Kagongi, than in areas with high attendance,
such as Kabuye. This was not obviously so. People apparently use a double
standard here. They may not be interested in sending their own children to
school, and may deride other parents for 'wasting their substance' in sending
theirs; but they are still able, somewhat inconsistently, to realize that others'
success in school will lead to a good job and to be jealous on this account.
Possibly it is a temporary and to some extent inevitable symptom of adapta-
tion to changing social circumstances. It should disappear with the passing
of time as secondary schooling becomes a less exclusive achievement.
Thus at an early stage in the period of transition the two indices cows
and education exist side by side. There is a conceptual opposition between
them which may be expressed, for example, in rival boasting at a beer party
of the type: "Your son knows nothing; mine is in secondary school and will
get a good job. Who will rule in future?" "But you, you have no cows. Look
at all mine. My son will be a rich man." Later, when secondary schooling
is appreciated but achieved by only a few individuals, there is conflict of


another sort: the 'have-nots' are jealous of the 'haves.' Eventually the 'edu-
cation' status index may replace the 'cattle' status index. In some villages,
with not many cows and a background of mission and school, it has nearly
done so already; in other villages, with many cows and no mission or school
influence, it will not do so in the foreseeable future.

Three variables then may be recognized: 1) response to education, ex-
pressed for each village as the percentage of children of school age who
actually attend school; 2) source of wealth, expressed as the degree to which
village resources consist in cattle on the one hand and cash cropping on the
other; and 3) criterion of status ascription, expressed (somewhat imprecisely)
as the degree to which status is accorded to individuals in respect of cattle
on the one hand and having children in secondary school on the other.
A fourth variable is proximity of school facilities. While it is clear that
this has some effect on village attitude towards education, it has been shown
that physical distance from schools does not in itself prevent attendance by
children from any of the three villages, except possibly classes P5-7 in the
case of Nyakasa. It cannot therefore be regarded as a decisive factor in
explaining variation in attendance figures between Kagongi, Kabuye and
Nyakasa. In any case people are affected more by the frequency with which
they see evidence of a school in operation, i.e. children in uniform walking
to and fro with books, than by mere spatial separation between school and
village. Since an index of variation between villages in this respect is statistic-
ally indefinable, it is not included in the diagram below.
In a particular village at a moment of time it has been suggested that
attendance figures are a function of source of wealth. In the diagram, then,
response to education and source of wealth are represented synchronically
along the right-hand vertical axis and the horizontal axis respectively. The
villages are plotted with a cross according to these two variables. The left-hand
vertical exis represents changing criteria of status ascription through time,
and the dotted lines indicate that position on this diachronic scale reflects
degree of response to education, as described in the section on 'Education and
Status.' The downward slope of the dotted lines is intended to suggest the
time gap between a certain level of 'attendance achievement' and recognition
of this in terms of individual status, owing to the conflicts generated by the
presence of differential status criteria and by disparity of achievement. The
validity of this representation is questionable, but it remains a convenient
summary of the inter-relationship of the critical variables as they have been
presented in the discussion. It shows, for example, that the dominant status
index in a village will be a function not only of the qualitative factor of
wealth but also of the quantitative factor of school attendance as it varies
through time.
The following points emerge from this brief exercise in applied sociology.
Firstly the study confirms, ten years later and from a different angle, some
of the conclusions drawn by Maleche on the basis of his survey of primary
school wastage in other districts of Uganda, in particular that ignorance,
poverty and the lack of an effective educational policy are important causes
of wastage and of failure to attend school. Secondly it suggests that, in this


education positive

r-.. KABUYE 28.5%
Criteria of Response to
status education
ascription (SYNCHRONI
.._.. ,-7 NYAKASA 17. 41.

KAGONGI 11.3%.
cattle __cash crop
cattle source of wealth negative

part of Ankole at least, such 'ignorance' derives not only from simple lack
of knowledge about schools and their wider implications, but also from social
inhibitions which have two sources. Firstly the traditional conception of
cattle as a unique measure of wealth and status, rendering education irrelevant;
and secondly the associated reluctance to send children to school through
fear of local derision for 'wasting substance.'
The main assertion of this paper is that in certain rural areas response
to education depends on the facility of conversion of items of wealth into
school fees and on the primary motivational role of status considerations. At
any rate, the findings here presented show the relevance of social background
to an understanding of the problems of educational efficiency. They may
also be of relevance in the planning of education policy.

The writer would like to acknowledge financial support from the following:
the Mary Euphrasia Mosley Fund, University of Cambridge; the Worts
Travelling Scholars Fund, University of Cambridge; Mitchell Cotts Group
Ltd.; Fitzwilliam College, Cambridge; D. L. Patel Press (Uganda)
Ltd. He is indebted to Mr. B. H. Mottram for geological and climatic in-
formation relating to Kajara. He would also like to thank the many indivi-
duals who gave freely of time, information and comment during the course
of the study.



1. Maleche, A. J-, Wastage among Primary School Leavers in Uganda. E.A.I.S.R.
conference paper, December 1960.
2. These villages were the homes of students at a secondary school near Rwashamaire
who were known to the writer through previous teaching experience there and
who proved invaluable as interpreters. Complete homestead surveys were carried
out in all three villages with their help.
3. A full account is contained in Morris, H. F., A history of Ankole, East Africasn
Literature Bureau, 1962. For further background information see Roscoe, J.,
The Banyankote, C.U.P. 1923, and Oberg4 K., The kingdom of Ankole in Uganda,
in African political systems, edited by M. Fortes and E. Evans-Pritchard, O.U.P.
4. For details of the myth see Morris, H. F., op. cit. p. 6.
5. A nuclear family consists of a man, his wife and their unmarried children. A
compound family consists of a man, his several wives and their unmarried children.
6. There are, in fact, two varieties of coffee grown: Arabica and Robusta. Arabica
is normally grown above 1,500 metres and Robusta below, i.e. much of southern
Ankole is marginal. The cash yield of Arabica is almost double that of Robusta,
but Robusta is an annual crop whereas in this area Arabica usually crops once
every two years. In the long run therefore the cash value of the two varieties is
approximately equivalent.
7. An estimate of the annual cost of keeping one boy in a government primary
school is given below, for classes P1 and P7. Building contribution is an amount
agreed by parents which depends on the state of the school in question: the
figure given is therefore approxmate only

P1 P7
Fees 24 sh. 80 sh.
Building Figures in
books etc.) 100 sh. E.A. shillings
contribution 30 sh.
Total 154 sh. 210 sh.

The 1959 rural per capital income was 76 sh. (Source: Gross Domestic Product
of Uganda 1954-59, E.A. Statistical Dept. Uganda .Unit 1961, Table 31, p.35),
It is apparent that education can be a heavy financial burden on the average
householder. But the figures are not reliable for comparison for several reasons:
they are not contemporaneous, the range of variation in cash income is very
wide, some having none at all although they may be wealthy in stock; and
the extended family network ensures that many boys' schooling is paid for by
different relatives.
8. The number of cattle a man has is recorded in a book kept by the muruka chief
and amended monthly if necessary. Cash crop acreage is assessed by a group
of appointed assessors: for a particular muruka these are the gomborora chief
and two muruka chiefs of whom one is an outsider; for a particular village these
three men are accompanied by the mukungu (paid village headman).
9. Individuals are often unsatisfied about the amount of tax assessed, but the
possibility of appeal to higher authorities and subsequent re-assessment, all ac-
cording to fixed grades, ensures that tax figures are a fairly reliable guide to re-
lative wealth, from whatever source it is derived.

Uganda Journal, 34, 1 (1970) pp. 15-27.



In 1966 as part of a study of immigration and immigrant settlement in
northeast Bunyoro. I studied the indigenous village of Kiminal in Kigumba
sub-county. In attempting to analyse the composition of this village a kind
of paradox arose. Although the village had had a definite and fairly recent
beginning, in 1919, its members were linked by a dense network of kingship
and affinal ties, just as Beattie found for the inhabitants of Kihoko village
in southern Bunyoro in the early 1950's. More impressionistically, as a
community it possessed an air of stability and permanence, arising in part
no doubt from this elaborate inter-relatedness. Yet this stability was clearly
in a sense illusory, for the population both of the village itself and of the
area generally was highly mobile. The paradox to be resolved was therefore
how this real instability could accompany features which might normally be
taken to indicate considerable stability.

The object here is therefore to try to define the nature and extent of
stability and instability in this setting and to try to produce a kind of model
of village composition which will allow a better understanding of who lives
in such a village. Necessarily in a situation of flux it is processes of moving
in and moving out recruitment and its opposite that attract attention,
more than any static consideration of rules of residence and categories of
resident. This analysis and model is based on one village only; but it would
be surprising if there are not in Bunyoro, as Audrey Richards has recently
shown that there are in Buganda,2 villages of a number of fundamentally
different types. The village analysed is in the northeast of Bunyoro. Though
not in the Palwo (Chope) area, the area is in several senses peripheral to the
centre of the former kingdom, and the village itself no doubt has, as do all
villages, its own idiosyncratic features. In the absence of any discussion of
the principles of village composition, even so tentative and inadequately
grounded an account as is provided here may be useful, at least to prompt
further discussion and study. Certain hypotheses are presented which may
contribute to a more satisfactory future analysis of the village in Bunyoro.

There is no altogether unambiguous term in Bunyoro for what is common-
ly termed a village. It is indeed the translation in use for both the terms
kyaro (plur. byaro) and mugongo (plur. migongo), but while either may
mean the essentially social unit to which I am referring, they do not neces-
sarily do so. Concentrating attention on kyaro, Beattie has translated this
as 'settlement area'.3 Undoubtedly this is a good representation of its general
significance, for kyaro, or the name of a particular kyaro, many like 'settlement
area', indicate a number of differently defined units. It may, for example, be
an administrative division of the sub-county for tax purposes; a geographical
territory defined by natural features, particularly valleys and swamps; or it
may be a social unit, a community of people who live together. Each .rplies
an area but not necessarily the same area.


lack of physical nucleation and social cohesion. Significantly he has also
chosen in his published reports to approach the analysis of local relations
through ego-centred studies of kinship, affinity and neighbourliness, rather
than through the examination of any definite social unit.4 In the fluid
Bunyoro situation there is great merit in such an approach. Yet there is no
doubt that such a unit does exist and that it is frequently the primary signific-
ance attached to a kyaro name. Beattie's approach may have the drawback
of obscuring the problems presented by the need to define this unit.
In the first place the village is normally a discrete settled area, a cluster
of homes and fields, even if it does lack a definite pattern of spatial organiza-
tion. Only in the few areas of high population density is this physical separation
of one village from the next not apparent. Though the village has no corporate
control over resources, there is a territory within which its members normally
make their fields and from which they may occasionally act to exclude others.
It is probable that the level of social interaction within the village is very
much higher than outside it. The great majority of people seen within a
village and attending every-day events there are residents of the village. It
is not surprising therefore that it is a distinct unit of self-identification for its
members. Lastly, the village has an officially recognized 'elder', the mukuru
w'omugongo. He is elected by the villagers and acts as an intermediary
between them and the chiefs. The villagers are supposed also to bring their
disputes to him, and these he deals with through an informal village court
consisting of himself and a few other senior men of the village on whom
he calls. As an indication of village size Kimina has 38 households early in
1966 and was one of the most substantial of the indigenous village in the
area.5 There were 41 in Beattie's Kihoko.
The village exists as a social unit, but its membership is considerably
unstable. From a sample census carried out in Kigumba sub-county in 1966
on the basis of household lists compiled a year previously, one in every six
or seven indigenous households appeared to have moved in during that inter-
val; The effect of such a level of instability can be seen in the origins of
households in Kimina. Only one in four had been created within the village
by the maturation of a son or daughter of the village and their establishing
an independent household. The rest had, over the 47 years of the village's
existence, moved into it, this despite the fact that Banyoro regard it as proper
for sons to establish their households near their fathers' and for brothers
similarly to live together. Half the total households dated in fact from the
years since 1960, an indication not that the village had doubled in size in
that period, but simply a symptom of the general instability. The longer
the time since a household arrived in the village, the less the probability of
its still being there. This is what accounts for the preponderance of new
Obviously relevant to this high level of mobility is the ease of movement.
Neither land nor housing present any great difficulty to the would-be migrant.
Land itself is neither bought nor sold. Until mechanical cultivation becomes
common, relatively small areas are cultivated and do not require any very
elaborate clearing to prepare them to be hoed and planted. In these circum-
stances no great value is commonly attached to the land in use and it is
readily abandoned. At the other end of a move, plentiful land has meant,


at least until very recently, that there has been no difficulty about obtaining
it wherever one wanted to settle. Land, the basic resource of an agricultural
people, has therefore provided no serious check to movement, at least in the
areas under immediate discussion. Houses are readily borrowed, built or
bought at little expense since most materials for a simple house are such as
to be obtainable free. Help from neighbours is desirable in building, but
neither practical nor social requirements make it essential.
In terms of social relationships, leaving one village and moving into
another is generally no more difficult. Neighbours in Bunyoro need to depend
on one another in only the weakest of senses, and what dependence there is
does not rest on any specific, particularistic type of relationship. What is
important and valued in Bunyoro is, rather, a diffuse and universal neighbour-
liness, a general helpfulness and mutual interest based on nothing more than
residential proximity.6
This is the negative side of the matter. It explains why people are able
to move so readily, but it does not explain why they should want to. Naturally,
if the effect is easily achieved, the force required to bring it about need not
be very great. People move for a variety of reasons. In Kigumba they have
moved away from the north of the sub-country particularly to take advantage
of amenities and economic opportunities along the line of road further to the
south, as well as to avoid damage done by increasing elephant herds. Of the
households in Kimina, three moved in in the early 1960s as a result of a
progressive evacuation of the villages immediately to the north.
But this is a once-for-all kind of moving. In Bunyoro as in many other
small-scale societies, personal difficulties give rise to a more continuing mobility.
Difficulties of all kinds are often thought of as the result of some kind of evil
influence at the place where one is residing., They are particularly likely
to be attributed to the sorcery ur the enmity of neighbours, concepts often
not sharply distinguished, though ancestral spirits or embandwa spirits may
also be identified as responsible. A person who feels himself threatened in
this way, either personally or through members of his family, is likely to
resort to moving.1
Whatever the reason for moving, unless it is over a very small distance,
people commonly choose to go to a place where they already have kin, parti-
cularly agnates; or affines particularly, men married to their female agnates;
or personal friends. There may be no more to this than that the presence of
such people in a village distinguishes it from the large number of villages
to which they might in the free circumstances of Bunyoro move. Their
presence is a basis for choice where choice has to be made, yet other criteria
are lacking. But again, this is the negative aspect. Particularly where a feeling
of insecurity has led to the removal, a place where one can feel secure is
what is required, and firm, pre-existing relationships are seen as likely to
provide this. This basis for choiccris crucial for the composition of villages.
It means at least that in spite of the high mobility, most people live in villages
in which they have relatives.
Though mobility is high and seems a striking feature of the indigenous
population, most households are not constantly on the move and a few move
only very rarely. The household in Bunyoro is a conceptually distinct unit, in its
ideal type, a kind of little kingdom ruled by its head, the nyineka. This status


is inherited, ideally by a younger son, as might also be the widows of the
deceased head, at least until the recent past. The survival of the household
is not dependent therefore on the survival of any single member of it. In
this way, the first household to arrive in Kimina in 1919 was still there in
1966, in spite of the death some thirty years earlier of the nyineka who had
brought it there. Though no other household could rival this longevity,
two others had been in the village for thirty years. There is thus an element
of stability; not all is in equally rapid flux.
The problem to be faced is whether differences in mobility have to be
regarded as matters of chance, or whether there is in them any systematic
element to be discovered. I want to suggest that there is, in the form of what
may be termed, focii' of immigration.
The crucial fact that people generally move to be near particular others
has already been observed. If one considers whom it is that people move to
be near, the result, at least in Kimina, is remarkable. Most people never
attract others in this way. In the fortyfive years of Kimina's existence there
seem to have been eight men only who have done so, of whom only half
were 'active' in 1966. It is these men who appear to have given and to
continue to give, the village its shape, who are responsible for the directions
taken by the flux, and who provide what stability there is. These eight and
the part they had played in creating the village up to 1966 are examined
The first was Uma, the founder of the village. He acquired and kept
.three wives, but only the first ever had children, and only one son, James,
reached adulthood. He failed therefore to achieve a large family himself
but at least managed to retain James' allegiance. James never left Kimina
and was his father's heir and successor when he died in the mid-1930s. To
join Uma in the village came Opio, friend, kinsman (his father's mother's
brother's son) and affine (his wife's sister's son). There were doubtless others
of whom I never heard, who perhaps did not stay long and it was in any
case long ago. Opio similarly managed to keep his two sons with him; if
a man's own sons do not stay with him, as is thought proper, it is most
unlikely that anyone else will be attracted to live with him either. Opio's
maternal half-brother and his family joined, also Odyek aind his family,
a friend but not a kinsman. But such links were soon forged; he married
his daughter to one of Opio's sons, and one of his sons married a daughter
of Opio's half-brother. Odyek was the third in this trio of those who attracted
others. He likewise retained his two sons, and another man with a family
whose father had been his friend moved in nearby.
In both the cases of Opio and Odyek, as in Uma's, there were probably
others who have been forgotten or missed. Nevertheless these three were
clearly small-scale and unexceptional in their attractiveness and hence in
the cluster which gathered around them. Not such was Bacumirwa. He
came to Kimina in 1936 but he had already, while living on the next ridge,
gathered the basis of a substantial cluster about him. He moved in with his
sister and her husband and his father's brother's son's son, Kasigwa, and his
family. His own two sons and their families he kept with him, and others
soon began adding themselves to the cluster. First were his sister's daughter
and her husband and when he died their daughter as her father's heir,
together with her children. Bitamara and his wife came, with his mother and


sisters, maternal kin of Kasigwa, affines to Bacumirwa. Gradually the focus
of the cluster shifted from Bacumirwa himself to his eldest son, Peter, but
people continued to add themselves to it. Kaahwa, a paternal half-brother's
son of Kasigwa, married to a sister of Bitamara arrived, but the way that
he settled almost within the homestead of Peter and Bacumirwa showed
that it was to them rather than to his affines or his closer agnates that he
was attaching himself. Related in the same way to Bacumirwa and Peter
were Kasigwa's wife's children, his own and. his elder brothers' from whom
he had inherited her. These likewise remained in the village, four of the boys
being adult by 1966, three with wives and three with children. The two
daughters had married within the village, but one of the marriages had
broken down and she has subsequently married elsewhere. Peter's own adult
son had also stayed in the village. Such an extensive cluster as this is certain-
ly exceptional, yet it is only a product of the same factors which produce the
more usual small groupings exemplified by all the other cases in Kikunya.
Nevetheless, settled as they were at the very centre of the village these Bagon-
ya, for that was their clan, and their kin inevitably imposed a special charact-
er upon the village.
Five of the eight have been considered. Of the remaining three, one,
Bitamara, who has already been mentioned was himself within the large
cluster though not himself a Mugonya. In spite of this he also, in a small
way, attracted incomes, such as a paternal half-brother and two mother's
brothers, who all lived on the fringes of his own household.
Ezira, the seventh, was brought to the village as a child by his mother
when she was married there. By 1966 he was in his late 30s and had as yet
no grown-up children. In spite of this he had attracted at least three lots of
people: his sister had retired from her husband fifteen years previously,
taking her children with her, and, though she had had children since, showed
every sign of having forsworn marriage permanently. Their mother's sister
with her husband and children had also come, though the husband had
left again after a few years. The father's brother of a second wife Ezira had
had for a short time also come to live nearby. This man had further connec-
tions in the village; he was a brother of one of the wives of Uma, inherited
by James, but it was apparently .o Ezira rather than to James that he had
Finally, James took over his father's position and inherited his two
surviving wives. He managed to retain two out of his three sons, and his
daughter returned with her children after the death of her husband, adopting
the same 'free' stance as Ezira's sister, a not uncommon one in Bunyoro
today. In addition to these immediate kin, two brothers of one of the old
wives had joined the cluster, the daughter of one of them marrying a son
of Kasigwa, and the sisters son of the other old wife also came. He had two
sons, one of whom later married a daughter of Kasigwa though this union
did not last long. This family later moved to Bacumirwa's part of the village
after trouble with elephants in James' more exposed section in 1960. Lastly,
in 1966, there was staying at James' a couple with the grown daughter of
the wife. These had left their home in a nearby village on account of mis-
fortunes there and had come to James because the man's father and Uma
had been friends, or 'brothers' or he insisted. He also insisted that he would
stay there permanently, but others judged that before long he would return





/ NBdcumirwa

A 0 A 0
r P^eK

S /A ~~-Kaahwa
IA 0 Bitamara
0.0' AA' 0 a
O jA v T

A People not present or dead


to his own home. He was apparently the type of person others move to be
with, rather than the type who themselves move. By 1966 James' part of
the village had acquired a name of its own, Kisonga, from the clan of its
inhabitants, the Basonga.

In Kimina there had been eight foci of immigration, of whom four
were in father-son successions, the pre-eminent were Bacumirwa and Peter,
and James and his father Uma. The four other foci were all of much smaller
scale, Ezira, Opio, Odyek and Bitamara. What were the ingredients of the
attraction exercised in such varying degrees by these people and totally absent
in the, case of most others?
A study of these Kimina foci suggests that social standing above the
common level, preferably combined with an advantageous economic position,
was most significant. In considering such status it has to be borne in mind
that the traditional Bunyoro was a highly inegalitarian society. Beattie has
used the idea of a 'premise of inequality' to point to the fact that a usual
feature of role-relationships was, and to a large extent remains, the identifi-
.ation of superior and inferior. Ideas of 'ruling' even penetrate into kinship
roles.8 The seeking out of those of higher status has to be seen against this
There may of course be practical advantages which a man of higher
status is able to secure .for those who have claims on him, such as the kin,
affines and friends who commonly move to his vicinity. His wider experience
and contacts make him a valuable source of advice and influence, a buffer
between the ordinary person and the outside worlds of administration and
law; but these practical advantages should not be over-stressed. In moving,
a person is not usually looking for a patron but for a place where he feels
he can live secure.
Three main fields from which higher status could be derived are to be
distinguished: firstly, the traditional political sphere, particularly association
with the kingship; secondly, the domestic and kinship sphere, as the head of
a substantial family; and thirdly the modern occupational sphere, with posi-
tions of authority in local employment and increasing occupational stratifica-
tion. As this new basis of status differentiation has established itself and
expanded, the relative significance and even the definition of the other bases
has necessarily undergone some readjustment.
Traditional status derived from association with the kingship is
manifested in the history of Kimina only in the person of Bacumirwa. He
was a parish chief for twenty years, during which time much of the Bagonya
cluster built up around him. As Beattie's analysis has made clear, as a parish
chief he would never have been a part of the true elite of Bunyoro;9 but
nevertheless from the point of view of the ordinary peasant he was clearly
a person of some standing. He would have had the ear of the sub-county
chief and have frequently been called upon by the common people to hear
their disputes. He would never have been highly paid, but he did have a
regular salary in days when this was even rarer than at the present.
Closely associated with this traditional source of status differentiation is
the matter of kibanja-ownership to which Beattie has given considerable


attention.10 Soon after the reforms of 1933 had made it possible, Bacumirwa
obtained a Certificate of Occupancy for land in use, that is to say he registered
his kibanja. He was not however alone: Uma, Opio and Odyek all did
likewise at various times during the 1930s. Uma claimed about 100 acres
at the eastern end of the village, while Bacumirwa obtained 75 acres in a
block to the west of this. Though the other two bibanja were considerably
smaller, these figures show that in practice little attention was paid to the
original intention to limit Certificates to the land the claimant was himself
actually farming. Such large holdings often in fact included several house-
holds and the claim was then tantamount to an attempt to approximate,
albeit usually on a small scale, the status of a traditional land-holding lord,
with his estate and tenants. The fact that all four original claimants were
in their time foci for immigration is not fortuitous; the kibanja-holding was
clearly initially invested with a considerable charge of actual or at least
claimed social superiority. We need to enquire therefore as to its significance
by the 1960s when the system was largely a relic from the past.
It had in fact, in the area studied, come to be nothing more than a
source of mild prestige. There was no significant relationship between a
kibanja-holder and those living on the kibanja. Indeed it was possible, and
exemplified in Kimina, for a person not to know whether or not he was within
a kibanja. Nor did it matter who was or was not legally the holder. There
was one man generally credited with a kibanja the Certificate for which had
necessarily lapsed when his father, the original holder, had left the village.
Another was generally thought of as belonging to the village elder although
it was actually his younger brother who, as his father's heir, had inherited
it. Most clearly of all, the lack -of practical significance is seen in the settling
of a party of Alur immigrants within the large kibanja James had inherited
from Uma.
In 1965 a small group of Alur went to a leading Munyoro of a neighbour-
ing village to Kimina to ask whether there was land in his part where they
could settle. He pointed out to them the empty hillside across the valley
from his own kibanja, and they went there and began to build. This was
within James' kibanja in Kimina. When he found them there sometime
later he considered that they had infringed his rights by simply building
without consulting him. But it was not against them that he took action.
They told him who it was who had sent them there and it was against him
that James complained to the sub-county chief. The Alur were not regarded
as being at fault, nor was any serious attempt ever made to remove them.
They therefore stayed, but James never had any dealings with them. They
soon had a leader of their own, recognized as such by the local chief, and
who took upon himself the allocation of land to newcomers. James himself
never troubled himself even to fix boundaries.
As this preceding intances show, kibanja-holding was in the 1960s
a source of prestige, which might be slighted, but nothing more. As a
potential element in village organization or a determinant of social relations,
the kibanja was not significant. Still less was it in itself a basis of attraction;
only two of the four regarded as kibamja-ho:ders were also foci of imm;gra-
tion. Obtaining land for cultivation was not a problem.
Modern occupational status seems relevant to three of the four con-
temporary foci and one of the earlier ones. Uma was a headman at a time


when few more elevanted occupations were available. He was also a catechist.
His son, James, was a road headman first with the Public Works Department
and later with he Bunyoro Kingdom Government for which he looked after
a long stretch of minor roads in the country. James' position undoubtedly
meant less than his father's had previously, but equally surely it still meant
something. Bitamara also was a headman and the same can be said of him;
as higher level occupations became commoner,. he social standing to be
derived from such lowly supervisory occupations becomes less.
Nevertheless the standing of only one person in Kimina was significant-
ly higher. Th's was Peter who was on the fringes of the new Bunyoro elite.
In schooling, with seven years to his credit, Peter was the most educated
adult of the village, though soon to be eclipsed by the generation still at
school. On the strength of his schooling he had been employed by East
African Railways and Harbours trained as craftsman, a painter and
songwriter. This involved being sent to Nairobi and various other parts of
East Africa before returning to the Masindi depot. It was a pensionable
post in an East Afr'ca-wide organization, placing him near to, if not actually
within,the Bunyoro modern elite of the late 1940s and early 1950s. In 1951
he made a 'ring' marriage with the daughter of a man working with him
in the railways, now a successful business man, but she had no children and
in other ways proved unsatisfactory. He therefore replaced her in 1958 with
the daughter of a successful retired teacher, living near Hoima. In 1959,
however, he was struck down with polio and so incapacitated that he had
to be pensioned off in the following year. But for his illness he might, or
might not, have gone further. As it was he remained on the fringes of the
elite. Through his former associates and his own affines he retained some
contact with the wider world beyond the village. His daughter had been
married by tre assistant manager of a nearby Group Farm, a successful
young man belonging firmly to the modern, English-speaking, motor vehicle
owing class. Peter's education and his connections, supported no doubt by
his small but regular pension, gave him indubitably the highest social standing
in the village.
In addition the income which normally attaches to modern ocupations
attracting higher status is also important. What appears to be significant,
however, is not so much the overall level of income as its regularity. Nor
does it appear that regular income in itself is sufficient, if not linked with
higher status. The people of the area are cotton-growers and this is the
staple source of cash, received each year over a period of about three months
starting at Christmas. Some people supplement this by local employment of
some kind, most often casually when the opportunity of a job coincides
with a need for money, but sometimes fairly regularly for a period each year
or even full-time. The main local sources of employment are on the roads or
on a local Agricultural Deparment seed multiplication farm, both of these
paid high government rates with a minimum in 1966 of 2 shs. 90 cents per
day. Non-governmental employment, for the local co-operative society or
for longer scale farmers, indigenous or immigrant, is sometimes obtained
but the rates for such work may go as low one shilling per day. Most people
in Kimina were able to obtain at least occasional employment at the higher
rates. Those with a full-time occupation and regular income fall into a
separate category. These were James as a road headman, and three of the


Bagonya cluster on the Agricultural Department farm, a headman, a trainee
tractor driver and an askari (watchman).
Cotton and employment are not of course the only sources of money
in the village but they are the main ways in which money gets into the village
from outside. By various means money way be redistributed before leaving
it again. The sale of beer and distilled spirit is significant in this redistribution.
Those with regular salaries have particular significance since they bring
money into the village throughout the year. They are able to maintain more
steadily than others a higher, more money-dependent standard of living;
they are able more often to help people in emergencies; they are able occas-
ionally to employ others; and they make a particularly important contribut-
ion to maintaining the now largely cash-dependent round of village sociability,
even at periods when most participants are without money. A regular income
is therefore a factor of definite potential relevance to the attraction of others
to live with one. Indeed in some cases it may be the most important factor.
It was probably largely to his income that Bitamara owed the presence of
the old men who lived on the fringes of his household since the higher status
deriving from his employment was marginal. In James' case, income was
perhaps a contributing factor, as it was for the Bagonya. Peter himself was
not, it is true, in employment but he did have a small pension, and the trainee
tractor driver was his son. Ezira, though he worked on the Agricultural
Department farm for a few months each year, showed, however, that a
full-time income is not essential, while the askari among the Bagonya showed
that it is not sufficient to attract others to live with one.
The third source of higher status, after traditional political and modern
occupational status, is in the domestic and kinship sphere. Heading a large
family has retained considerable contemporary significance, at least at the
village level. The basic consideration here is that a man should have sons
and that, when these marry and establish their own households, they should
remain near their father. If a man cannot keen his own sons he is unlikely
to attract others to live with him. The foci have, in so far as they have had
adult sons, usually been able to retain them, and this is an important part of
the phenomenon of foci and attraction. But the point to be made is that, in
order to be able to retain sons, just as to be able to attract other kin and
affines, such people have first of all to exist. In Bunyoro this cannot be taken
for granted. Uma's three wives and one son excite no surprise in Bunyoro
where both men and women without offspring at all are not rare. For many
any possibility of becoming foci is lessened seriously by a shortage of people
to be attracted, and a shortage of sons is certainly the most crucial, since it
is likely to prevent a cumulative process getting started. Status derived from
having others living with one is likely to contribute to bringing in others.
From the preceding it may be seen the way people move, and hence the
composition of the community, depends to a considerable extent on the pre-
sence of a small number of people with the capacity to attract others. In
the inegalitarian society of Bunyoro it is social standing above the ordinary
which is crucial in this respect. This may be more important than any directly
economic advantage. Two possible economic factors, income and landholding,
have some role but it may finally be asked whether there are any other such
resources which might be significant. Permanent crops and livestock deserve
comment. Of the former, the only thing of significance in this area is the


plantain groves which on the one hand take several years to get established,
and on the other, provide the basis of a popular type of beer which allows
beer parties to be held. While the possession of extensive areas of bananas
is certainly an asset, Kimina provides no evidence that it is an asset possessed
by those who constitute foci to a greater extent than by others. Almost every
well-established household has its grove, and even if it did not, beer could
still be provided without difficulty since grain-based beers are at least equally
common. The most prestigious type of all, indeed, is masohe, which should
be made from millet, while the commonest is kwete made from maize. As
for livestock, a few goats, sheep and hens are kept but there are no concent-
rations, and even if there were they would not represent a significant resource
likely to attract others. Cattle which well might do so are scarcely to be found
in indigenous hands in Kigumba; in Kimina there are none. It appears
therefore that the matter of status is the major determinant of attraction.
Thus the focii' mould the local community by attracting newcomers into
it, but they also contribute to its stability, both real and apparent. They
contribute in the first place by being far less mobile themselves than most of
their fellow villagers. This probably depends to some extent on factors which
do not distinguish them from other well-established householders. They
tend to be among those with a greater material investment in their homes
and land, with larger plots of plantain, more fruit trees and even such new
developments as a small plantation of trees for building purposes as owned
by Ezira. But with the ease of sale and purchase of such 'developments' in
Bunyoro, it is not perhaps these which are crucial. Of more particular signi-
ficance are aspects of a person's situation in one village which are not readily
duplicated elsewhere. The prestige of holding a kibanja can now only be
had by remaining on one's existing holding. Although the case of Odyek
who moved away from his, shows that possession is not always sufficient to
prevent movement, it is nevertheless certainly a factor contributing to stability.
More directly, the fact of being a focus and having others living with one
is itself a position only achieved gradually and with luck, hence it is not
something that can be readily recreated in another village after moving. For
the 'focus,' other villages are much less equivalents in social terms to the one
he is already in, and unless he can take the cluster of those around him
with him, as Bacumirwa did in moving to Kimina, he is therefore much less
likely to move than an ordinary person.
It seems probable that focii' also contribute to stability in a second way.
The basic phenomenon is the attraction of individuals into the community
by particular others. Where many are attracted and remain, however, this
results in something much more like a long-term structuring of relationships
within a section of the community. A definite cluster with a life of its own is
generated. This is a concentration of significant relationships which does
not depend at all exclusively on its initial or primary focus. The effect of
this is to extend the range of those for whom it is not readily possible to dupli-
cate elsewhere their present social position. To be a Mugonya in another
village is not the same as to be one in Kimina in which the leading people
are one's kinsmen and clansmen, numbers of one's other kin live around one,
and in which one belongs indeed to the cluster with which the village tends
to be identified. In consequence it appears that where sizeable clusters are
generated, this again leads to a higher degree of stability, at least for their


more central members.

Finally, if misfortune, giving rise to a feeling of insecurity towards
one's neighbours and their hostility and possible sorcery, is an important
reason for moving in Bunyoro, it is likely that this will affect focii' or central
members of large clusters less than others. Are they not equally likely to
suffer misfortunes? In fact, although there is no evidence at all on this matter,
and it is perhaps hardly practical even to think of obtaining it, it does seem
very likely that such people would tend to react differently to misfortune.
There is in Bunyoro, as elsewhere, always a certain latitude in the interpreta-
tion of troubles and it seems likely that whether simply by jumping to conclu-
sions or by some process of divination, alternative possibilities to the sorcery
of neighbours would tend to be adopted. Attribution to ancestral or
embatndwa spirits, God, or even Western 'bad luck' are alternative possible
reactions. Even if sorcery is decided upon and a neighbour is identified as
responsible, there is still an alternative to removing oneself. Even though
to accuse somebody of sorcery is in law a punishable offence, in practice
informal pressures, or even outright violence, such as occurred in the Kimina
Alur settlement, can often be brought to bear with impunity to cause a person
to move. Of all people, cluster focii' would seem most likely to be able to
act in this way.

On the basis of some knowledge of one village, certain hypotheses which
may lead to a better understanding of the composition of villages in Bunyoro
generally, have been advanced. As Richards has said of villages in Buganda,
lack of any distinct lineage structure does not mean that villages resemble
British housing estates in their principles of recruitment and composition.12
The composition of all communities may be expected to be related to the
particular complex of principles of social organization and culture within
which they are found. Despite a complex pattern of kinship and affinal ties
within the village, the population is unstable, and so it is therefore to the
directions in which people move and the reasons for which they do so, that
one must look to find any systematic element which may exist in the compo-
sition of villages. People appear to move particularly in search of security
and in any case move to the vicinity of others with whom they have pre-
existing ties of kinship, and friendship. In practice there are few only
who have the capacity to attract and therefore guide movement in this way.
The composition of any given village is therefore largely the result of the
attractions exercised to varying degree by its various focii.' It appears that
it is essentially their higher than normal social standing, deriving from tradi-
tional or modern occupations and from heading a substantial family, that
qualifies certain people as focii.' To say this does not imply that they and
they alone must necessarily attract others. People to be attracted must be
available, and no doubt personality differences are both positively and
negatively relevant to the actual outcome. Finally it seems that focii' and
certain others around them may be expected to be less mobile than others.



1. In order to preserve the anonymity of informants as far as possible, proper names
have been changed throughout.
2. Richards. A.1., The changing structure of a Ganda village. Nairobi, East African
Publishing House,, for E.A.I.S.R. 1966, (East African Study, no. 24).
3. Beattie, J. H. M., Bunyoro: an African kingdom. New York, Holt, 1960, p.9. "The
Nyoro term kyaro (a place where people stay) refers to the territorial aspect of
the community." Also p. 51. "A number of scattered homestead make up a settle-
ment area or 'village'."
4. Ibid. Chapter 6 and Beattie, J. H. M., Understanding an African kingdom. New
York, Holt, 1965, pp. 14-15.
5. Seven of these 38 households were of other than Nyoro origin: 4 Lango, 1 Alur,
1 Kikuyu, 1 Teso. The Alur settlement mentioned later was socially external to
the indigenous community and has been. excluded, so that this essay does not
discuss the position of the non-Nyoro community.
6. Beattie, 1960. op. cit. pp. 61-66.
7. Beattie, J. H. M., Sorcery in Bunyoro, in Witchcraft and sorcery in East Africa,
edited by J. Middleton and E. Winter, London, Routledge, 1963.
8. Beattie, 1960, op. cit. pp. 51-52.
9. Beattie, J. H. M., The Nyoro, in East African chiefs, edited by A. I. Richards,
London, Faber, 1960, p. 119,
10. Beattie, J. H. M,, The Kibanja system of land tenure in Bunyoro. (Journal of
African Administration, Vol 4, 1954).
11. Beattie, 1960, (Richards) op. cit., p. 110.
12. Richards, op. cit. p. 95.



Simon Charsley is a, lecturer in the department of .sociology at Glasgow
University, having spent two years as a research fellow at the Makerere
Institute of Social Research working in north Bunyoro and prior to which
he had studied the Nyakyusa of Tanzania on the history of whom he has
recently had a book published.

Father Krommenhoek had taught for two years at Namilyango College, but
has since re urned to teach in the Netherlands; he has published a note
on his fossil finds at Kazinga in a, previous issue of the Uganda Journal.

John Lock has recently completed six years as a research fellow with the
Nuffield Unit for Tropical Animal Ecology at Mweya, in Queen Elizabeth
National Park and has obtained a Ph.D. for his work on the grasses of the

Kaare Lye is from the Botanical Institute of the Agricultural College of
Norway from which he spent two years on secondment to the Herbarium at
Makerere on a Norwegian Aid scheme, and is expected to, return again in

Patrick Mateke graduated in April 1970 from Makerere and is awaiting an
opportunity to continue with his research interests in Kigezi. The paper
presented here is a modified and abbreviated version of an undergraduate
research paper.

Colin Murray taught for two years at Kitunga School in west Ankole
where he collected the information for this paper which he has worked on
as a research fellow at Fitzwilliam College, Cambridge under the direction
of Dr. Audrey Richards.

Musa Mushanga is the Regional Representat:ve of the Department of
Continuing Education of Makerere at Fort Portal where he is engaged on
a study of crime in western Uganda.

Larry Robbins is an Associate Professor of Anthropology at Michigan State
University, East Lansing, from which he has spent a period of secondment to
the Cultural Division of the Institute of Development Studies at Nairobi

John Wilson is a resident of Moroto who during the 1950s and 1960s spent
long periods with the agricultural department in Karamoja working mainly
as a research officer on the soils and vegetation of the area .

Uganda Journal, 34, 1 (1970), pp. 29-33



Historically speaking there was no such term as "Banyankore," for as
H. F. Morris says "The Kingdom of Ankole, as it exists today is the creation
of the British Administration."' Ankole had comprised a little more than
the present saza of Isingiro. Even today to most people of Ankole, "Nkore"
means only the sazas of Isingiro, Kashari and Nyabushozi; for people in
Igara, Shema and Kajara, "Nkore" means Mbarara. Present-day Ankole
comprises ten counties of which only one is the original Ankole. The most
densely populated counties of Igara, Kajara, Shema; the larger part of
Rwampara, and the less densely populated sazas of Buhweju, Ibanda and
Bunyaruguru were brought under the state of Ankole at the end of the nine-
teenth century. Ankole proper was a thinly peopled, dry and open country,
suitable for cattle grazing.
The Banyankore are of two groups, the dominant Bairu made up
92%-96% of the Banyankore population of 519,000 in 1959. These were
the original inhabitants of the country. The remainder, the Bahima, seem to
have invaded three or four centuries ago. These people were herdsmen; but
recently some, especially the more educated, have abandoned their old
migratory habits. The Bairu do not seem to know how they received their
name. Taylor translates Omwiru (sing. of Bairu) as a peasant,2 but it can
also be used to mean servant, serf or slave.
Among the Ankole people, Bairu as well as Bahima, the social principle
of descent is agnatic. The clans are totemic and exogamous. As a collection
of individuals they never assemble or take common action. The word for
dan is "Oruganda" (pl. "Enganda"). The only writer who has said anything
plausible about the concept of Oruganda was Oberg. He stated that the
year starts with "Okurya amwaka" ("to eat the year"). This is the first meal
of the harvested millet, "Omweza." At this meal it is ceremonial practice
to take a portion of the millet meal and paste it on the main poles that support
the house, "Emiganda." When millet is being harvested, it is gathered in
bundles that are called "Enganda." After millet has been made ready to eat,
it is first pasted on the Emiganda, and usually members of the oruganda
(clan) gather to "eat the year" together. The common root, "ganda" in
these three words was seen by Oberg to have symbolic significance. Thus
enganda (millet) supports life, emiganda (poles) support the house and oruga-
nda (clan) supports and gives security to the individual member of the clan.3
The dan structure among the Bairu is more less the same as it is among
the Bahima. This point, however, had not been recognized by earlier writers.
Roscoe for instance thought that the clans were different and those of the
Hima of a higher status.4 This possibility was referred to again much later
by A. I. Richards, though she recognized that the view "has been queried by
subsequent investigators."5 The misconception that "there are mixed clans
as well as pure Hima clans, and interclass marriages do take place" has recently
been perpetuated.6


Other writers have also sown seeds of confusion as to the number of
clans in existence. Roscoe wrote that there were three main clans the
Abahinda, the Abashambu and the Abagabe. C. Taylor also shared this
view, though they differed on the number of sub-clans. Roscoe claimed that
the Abahinda clan had 32 sub-clans, but Taylor only gave 14; for the Aba-
shambu, Roscoe had 38 and Taylor 54; for the Abagabe Roscoe had 30 and
Taylor 42. The counting of sub-clans, however, is particularly dubious, since
most of the names given by Roscoe and Taylor are unknown today. Morris,
however, counted four main clans, by adding the Abaishekatwa, and Stenning
is claimed also to have counted four.7
The Oruganda (clan) is a reference ca egory, members of which do not
know each other. All that clan members have in common is a totem ("Omu-
ziro") and all members of the clan observe the exogamy rule. Clan member-
ship is important in that a member gets help, fair treatment and hospitality
when away from home. Roscoe wrote that marriage was permitted between
sub-clans. There are no sub-clans among the Bairu; but sub-clans and second-
ary totems are found among the Bahima. Marriage on a sub-clan level seems
to have been instituted to afford the Bahima a chance of marrying between
themselves so as to keep as 'pure' as possible; for if they had insisted on
exogamy there would be no such people as the Bahima today. This can be
seen in Buganda, Toro and Bunyoro where the rule of exogamy was apphed
to the mother's clan as well, and where typical Bahima are not found.
This factor may account for the institution of sub-clans, secondary and
tertiary totems and the use of the word "Beene" amongst the Hima. The
word "Beene" means "sons" or "descendants." The Bairu do not have sub-
clans and no-one refers to himself as belonging to "Beene" so and so sub-
As already seen Roscoe referred to differences between the Bahima and
Bairu clans. This may have been so once; but now the Bahima and the
Bairu who belong to the same clan (i.e. have the same avoidances, "Omuziro")
call themselves by different names. Roscoe thought that there were clans
that were exclusively Bahima. Audrey Richards quotes evidence from the
Bahaya, however, that "social promotion" of a Bairu was possible. If a Bairu
acquired cattle he could rise in the social scale, if a Muhima lost his cattle
he could be demoted. These two conditions created the "Abambari;" a
hybrid as result of Iru-Hima union.
The following is a list of clans and their totems. Clans with a similar
avoidance have been grouped as one, and all dissimilar avodances have been
regarded as being of different clans. Thus the criterion for recognizing the
existence of a clan is "avoidance" ("Omuziro"):-
Clans (enganda) Totems (emiziro)
Abairu Abahima
Abahira Abanzira Epus (breastless woman) others
enkyende (small black monkey)
Abasingo Abagina Omurara (a cow with a white strip
running from head to tail).
Abahweju Abashambu Enkyende
Abayangwe Abahinda Epu
(the ruling clan)



Abariisa 9


Amashereka (breast milk)
Ente ngoobe (cow with many spots of
different colours)
Ente ngoobe others say they avoid
Engabi (bush buck)
One group avoids Epu, another avoids
ngoobe (cow).
Akafute (cow or goat born feet first).
Oburunga (seeds of coral trees).
Enjojo (elephant).
Ensenene (grasshoppers).

There are people who have given unknown names for their clans, and
these have been included in literature by previous writers. Such clans include
the Abanyazi, Abaami, Abatwe, Abaranzi, Abatyaba, Abakungu, Abahambi
and a seemingly endless list of others.10 It should also be noted that some
clans have second totems. According to Roscoe the Abahinda are said to
Among the Banyankore, especially the Bairu, women are given new
clan names. At marriage a woman is given a new name by her husband and
as she settles down into her new family of procreation, usually after she has
had several children, the women folk begin to call her by her clan name.
Such names are:-

Abasiita / Abahambi


Some birds in Ankole are identified with some clans, and a few of these

Entuuha crested crane
Ekikoona crow
Eshande-Rukakaare weaver bird
Kanyamunyu bird with a long tail
Efunzi a very small bird
Empungu mareere kite-hawk
Ekiteera-nkumba a bird fond of eating millet
Oruyongoyongo- heron
Enyawawa green ibis (hadada)
Enkooherwa hoopoe


It is unfortunate that early writers about the people of Ankole never
recorded joking relationships, "obukumbi." These have tended to disappear






especially in areas where people of other tribes are gathered, like trading
centres and towns. Joking relationship as described by Professor Radcliff-
Brown, "as a relation between two persons in which one is by custom permitted,
and in some instances required, to tease or make fun of the other, who in
turn is required to take no offence."11 The joking relationship, in Ankole is
what Professor Radcl'ff-Brown called 'symmetrical', that is each of the two
persons teases or makes fun of the other. This is on a clan basis, and the
Banyankore know that a joking relationship exists between such and such
clan, examples are:-

Abatsyaba joke with Abahweju
Abasingo ,, ,, Abairuntu
Abayangwe ,,,, Abakimbiri
Abaitira ,,,, Abaikizi (Abayangwe)
Abanzira ,, ,, Abatsyaba
Abanyari ,,,, Abasingo
Abazigaaba ,,
Abatorogo ,

The joking relationship is an ambiguous combination of friendliness and
antagonism. The behaviour is usually confined to equals of either sex. Joking
could take the form of confiscating property or consumable goods, such as
pots of beer, bananas or meat. The owner of the goods would not bother
to recover his goods, in most cases he would wait to reciprocate the behaviour.
A type of joking relationship exists between a boy and his maternal uncle
("nyinarumi" or male mother). The boy enjoys closer friendliness with his
maternal uncle than with his paternal uncle with whom no joking exists.
Although he is more friendly with his maternal uncle, and can joke with him,
yet he seems not to be warmly received at his maternal uncle's home because
of this joking relationship which could be uneconomic on the part of his
uncle. For example, in the course of his joking he could touch a goat and
demand it to be slaughtered, and according to convention, his maternal uncle
cannot refuse this demand. Another kind of joking relationship exists between
grandparents and grandchildren. Old men call young girls their wives, and
one often hears an aged woman call a boy of two years "my husband."
Very often an old woman will be seen handling the genitals of a young boy
and bringing her hands to her mouth, demonstrating that she is kissing the
genitals of the boy.
Kinship structure and function is weakening among the Nkore people,
but clan membership is still valued because it defines the categorical status
of an individual.



1. Morris, H. F., The making of Ankole, Uganda J., 21, 1957, p., 1.
2. Taylor, C. Runyankore-English dictionary, 1962.
3. Oberg, K. Kinship organisation of the Banyankole. (Africa, 11, 1938, pp. 129-138).
4. Roscoe, J. The Banyankole. 1923.
5. Richards, A. I. East African chiefs, 1960, p. 31,
6. East African Study Material, Department of Political Science, Makerere University
College., Part 1, page 14 (1969?).
7. Taylor, B. K. The western lacustrine Bantu, 1962.
8. "Epu" or "Epa" is a breastless woman,j it is also used as a form of emphatic
assertion, either in the affirmative or for denial.
9. This is a clan of Nyoro origin found mostly in Buhweju.
10. Taylor, B. K. op. cit. 10.2, footnote.
11. Radcliffe-Brown, A. Structure and function in primitive society p. 90.



Sir John Gray, a former Chief Justice of Zanzibar and a pioneer of
research into the history of East Africa, died on 15 January 1970.
John Milner Gray, the younger son of Dr. Arthur Gray, Master of
Jesus College, Cambridge was born on 7 July 1889. His maternal grand-
father was the Rev. John Philip Gell, one of Dr. Arnold's star! pupils at
Rugby and the model for 'Old Brooke' of Tom Brown's school days; whilst
his grandmother was the only child of Sir John Franklin, the Arctic explorer.
Educated at Dover College, Perse School and King's College, Cambridge, he
had qualified as a solicitor when the 1914-18 War took him to France, where
he was twice wounded. He recalled an impressive meeting as a young staff
officer with the ageing General Marchand (of Fashoda), and it may be that
this turned his thoughts to Africa, for in 1920 he joined the Colonial Admini-
strative Service in Uganda. He was soon engaged in magisterial work and,
having been called to the Bar by Gray's Inn, he was in 1934 transferred to
the Gambia as Judge of the Supreme Court. Here he twice acted as Governor,
returning in 1943 to East Africa as Chief Justice of Zanzibar. He was knight-
ed in 1944. He retired in 1952 but remained in Zanzibar to pursue researches
in the Zanzibar Consular archives, the while assisting the Government, pre-
siding over a number of inquiries and undertaking a revision of the laws. In
1960 he returned to England to continue the life of a scholar at Cambridge
where he died. He was unmarried.
When Gray went to Uganda in 1920, East Africa had engendered a
considerable travel and missionary literature, but no professional historian
had turned his attention to a co-ordinated study of the past of this rapidly
unfolding region. Gray with his academic background was quick to recognize
this lacuna. For some years he was posed in isolated out-stations, and here
he acquired that intimate knowledge of untouched Africa and its peoples
which so enhanced the value of his later work. He was at pains to gain
a knowledge of local languages and was the first to attempt the collation of
oral traditions with a virtually unknown vernacular literature, distilling there-
from a coherent whole,


In 1934 a small body of like-minded officials, prominent among whom
was a young district officer, la'er Lord Twining, combined to publish The
Uganda Journal. In this appeared Gray's first findings, seminal contribu-
tions which are among the foundation stones upon which the studies of the
now numerous schools of African history, all then non-existent, have been
raised. His subsequent explorations of Arab, Portuguese, French, German,
Egyptian and American sources yielded a rich harvest of discoveries, which
wi:h unconstrained generosity, he handed on to any who shared his devotion
to historical integrity. These appeared for the most part in The Uganda
Journal and Tanganyika Notes and Records. In addition to these many
scores of contributions, all scholarly and informed, his major publications
including a History of the Gambia, The British in Mombasa, 1824-1826,
Early Portuguese Missionaries in East Africa and History of Zanzibar from
the middle ages to 1856.
On the occasion of his seventieth birthday, his portrait, an appreciation
of his services to African history, and a bibliography of his writings to 1958
appeared in Tanganyika Notes and Records, 53, 1959, pp. 148-53. It is
hoped that a complementary bibliography will be published in The Uganda
Journal in the near future.


(with acknowledgement to The Times.)


The Uganda Society mourns the tragic death in London on 21 Nov-
embsr, 1969 two days after the 45th anniversary of his birthday of its
Honorary Vice-President and at one time Patron, Sir Edward Mutesa. K.B.E.,
thirty-fifth Kabaka of Buganda and first President and Commander-in-Chief
of Uganda.
In a poignant letter to The T:mes (27 November, 1969) Dame Mar-
gery Perham than whom no one has followed more understandingly the
unfolding events of his reign speaks of "the courage and dignity" which
he displayed in the "impossibly difficult situation" to which he succeeded
upon the death of his father, Kabaka Sir Daudi Chwa, in 1939.
He first accepted office as Honorary Vice-Pres;dent of the Uganda Society
in 1950. As president of Uganda, he was patron of the Society from 1963
to 1966, and he remained an Honorary Vice-President until his death.
Many will recall his dignity, courtesy and cordiality upon the occasion
of his opening on 23 September, 1963, 'of the new headquarters of our
Society, which he referred to as "one of the most outstanding and well-known
of Africa's cultural Societies". His address on this occasion is printed in the
Uganda Journal 29, 1965, pp. 112-3.


Uganda Journal, 34, 1 (1970) pp. 35-47.



Batutsi conquest.

In order to understand the political organisation in Bufumbira one has
to see what was happening in Rwanda. The Batutsi came to Rwanda before
1500 A.D. They were cattle keepers. They found the Bahutu organised on a
dan basis. Each clan had its own paramount chief known as Omuhinza (plural
Abahinza). The Abahinza were usually associated with the fertility of the
soil and the coming of rain. So long as rain fell and so long as the crops did
not fail all was right with the Abahinza.
The first Batutsi are supposed to have arrived in Rwanda around 1000
A.D. or thereabouts. In the process of time they started undermining the
power and respect enjoyed by the Abahinza. In order to get rid of the Bahutu
paramount chiefs (Abahinza) they are said to have employed the prevailing
status attached to cattle. Some Bahutu are supposed to have desired to own
cows. So they became clients of the Batutsi. As time went on the loyalty of
the Bahutu was divided between the Abahinza and the Batutsi. After some-
time the Abahinza were stripped of their positions by the Batutsi either by
force or by persuading the Bahutu to withdraw their support from their tradi-
tional chiefs. Some of the Abahinza did not like the Bututsi interference and
subversion and so they decided to move out of Rwanda supposedly before the
fifteenth century and headed towards the present Kigezi district. Some of
these immigrants settled in the then Rukiga area which is now divided into
two counties, Ndorwa and Rubanda. The rest settled in Bufumbira. The exact
date of the arrival of these immigrants is not known.
In Bufumbira the political organisation was an imitation of the earlier
clans in Ruanda. Before the establishment of strong political institutions as
they had existed in Ruanda prior to the Batutsi interference another conquest
took place. The Ruandese authorities sent several expeditions to. re-subjugate
the Bahutu who had fled Batutsi domination. Batutsi chiefs were sent from
Ruanda to Bufumbira to rule the Bahutu. The first Mututsi to be sent to
Bufumbira is remembered as Mpama, son of Bumbogo. He was sent by Mwami
Yuhi IV Gahindiro in the 1830's. The second Mututsi chief to be sent to
Bufumbira is known as Muleganshuro son of Mwami Yuhi IV Gahindiro.
This was during the reign of Mutara II Rwogera (1840-1853). Not much is
remembered about these early Ba'utsi chiefs.'
The Bahutu in Bufumbira did not accept the new Batutsi domination
quietly. While some clans collaborated with the Batutsi chiefs, several clans
tried to terrorize the whole of Bufumbira so as to get rid of the Batutsi rulers
and their collaborators. The Abatongo group living just below the Kanaba gap
can be said to be a fairly representative example of Bahutu's resistance to
foreign Batutsi rule. The Abatongo were Bahutu who inhabited the valley
below the Kanaba gap. They were not one clan but clans such as Abungura,
Abacyaba and Abazigaba. Their leader was Ngirabanzi the son of Nyangabo.


Ngirabanzi was a chief priest of the religious cult called Biheko which gave
him added authority and may explain why he was able to unite a number
of clans against the Batutsi invaders., Later in the history of Kigezi it was
again shown that religious leaders could call upon a wider area of loyalty
than could clan and political leaders.
Ngirabazi and his Abatongo carried out a series of campaigns all over
Bufumbira and even went so far as to attack Bukamba, a province of Ruanda
to the south of Bufumbira. They defied the invaders to the extent of killing
two Batutsi princes, Rutaremara and Mpezambihigo, sons of Mwami Yuhi,
father of Mwami Rwabugiri. When Rugyira, a clan head of the Abasinga clan
of the Bahutu, went to the help of these two princes he was killed as well. In
fact the Abatongo established their rule almost all over Bufumbira. Several
fighting men were sent against the Abatongo by the Batutsi but in vain. Buki,
a Mututsi chief appointed by Mwami Rwabugiri, looked very desperate, for
the whole of Bufumbira was being shaken by violent activities of these
Following the death of Rugyira, his nephew, Mushakamba son of Bivange,
was appointed chief over the Abasinga clan.. Mushakamba's major goal was
to defeat the Abatongo. Since he could not do it alone he sent for the Batwa
living on the western side of Lake Bunyonyi and feasted them on beer and
meat. Then one morning the Abatongo found themselves encircled by the
Batwa and Abasinga warriors. The Batwa were experts at shooting with bows
and arrows. After a few hours nearly all the Abatongo men were dead. They
were just taken unawares for they were encircled at dawn. Those who suc-
ceeded in escaping were finished off by a famine which took place immediately
after the fighting. From then on Mushakamba's popularity was enhanced.
He was given a popular name, Rutsinsura, because he defeated the Abatongo
who challenged Ruandese rule.
But why were the Batutsi able to re-subjugate the Bahutu? The Bahutu
in Bufumbira were not united and inter-clan conflicts were frequent. If all
the clans in Bufumbira had united it would have been impossible for the
Batutsi to have established their rule. If all the clans for instance, had joined
behind Ngirabanzi the course of Bufumbira's history before the colonial
period would have been quite different. The Batutsi were very few and they
would not have been able to conquer a united Bahutu. Because the Batutsi
were few they were compelled to use inducements and diplomacy to establish
their rule. They sought collaborators in return for cattle and important Bahutu
clans like the Abasinga and Abacyaba under Mushakamba and Birahira,
respectively, became their tools of conquest. Finally, it appears as if the support
of the Batwa for the Batutsi cause was of crucial importance. The Batwa,
as the original inhabitants, may have resented the Bahutu immigrants and
were prepared to support the Batutsi against their old enemies the Bahutu.
Though the Batutsi established their rule in Bufumbira, most Bahutu were
not happy because they preferred to be ruled by their own leaders. In fact
Batutsi chiefs like Buki and Nyirinkwaya failed to reconcile the Bahutu to
Batutsi rule. As a result the Batutsi found it politically expedient to rule the
Bahutu indirectly, that is through Bahutu chiefs. Since the Batutsi conquest
had been successful only through the assistance of Bahutu leaders, it was
only natural that these leaders be rewarded.


Bahutu chiefs like Mushakamba, Birahira anl Mutesi, could not initiate
their own policies. They had to take instruction from above. Their duties
were to collect tributes and send people as unpaid labour to local chiefs. How-
ever, areas which were predominantly settled by Batutsi were given Batutsi
chiefs. For example, Nshizirungu was in charge of Jomba, while Hagumaka-
mwe and Nyirimpunga were in charge of Busanza.
Indirect rule and the use of Bahutu chiefs, however, came slowly. It only
became obviously necessary as resistance to Batutsi over-rule demanded that
the Batutsi invaders seek Bahutu collaborators. Very little is known about
the first two Batutsi rulers of Bufumbira. The first was Mpama sent by
Mwami Rwogera, the second was Muleganshuro appointed by Mwami Mutara
II Rwogera,(1840-1863). The third was Buki, son of Muhabga. He was sent
to Bufumbira by Mwami Rwabugiri (1853-1895). Buki was instructed by the
royal court in Ruanda to build a royal residence in Bufumbira. This was
built at Mabungo. When this royal residence was completed Mwami Rwabugiri
went to Bufumbira where he spent several days being entertained by the
Bafumbira. Buki is mostly remembered for his oppressive rule. It was during
his time that there was a great deal of unrest all over Bufumbira. The Abatongo
group gave Buki much trouble. People who had cattle were always in constant
fear. In fact his rule was so harsh that he was recalled to Ruanda. It would
seem that he put self-interest first and then Ruanda's second. He was criticised
as being a woman, meaning that he was weak and tactless.
Buki was succeeded by Nyirinkwaya son of Giharamagara. As his rule
was as oppressive as that of Buki he was also recalled to Ruanda. Mwami
Rwabugiri decided to send his daughter Berabose assisted by Muvunandinda
to rule Bufumbira. This marked the end of one period and the beginning of
another. One informant summed it up thus: "The Bahutu accepted Berabose's
rule. Her rule was not oppressive. During her time in Bufumbira most Bahutu
accepted her rule and paid due allegiance to the Mwami of Ruanda. Her
time in Bufumbira was a very peaceful one." Tradition has it that she over-
heard some top secrets of the Ruanda kingdom when Rwabugiri was consulting
his royal adviser known as Abiru. It was found out that Berabose had been
in a house where these secrets were being discussed. As a result she was
another royal family, the Ruandese royal family would be overthrown. In order
commanded by her father never to marry. It is believed that the secrets she
overheard concerned the succession and that when she married a prince from
to comfort her she was asked to go to Bufumbira as a provincial ruler.
There was speculation among top officials in Rwanda that Berabose could
not manage the affairs of Bufumbira effectively, thus a Mututsi known as
Muvunandinda, son of Gafirigi, was appointed to assist her, but Berabose
was the one vested with full authority. Fundamental problems were settled
by her and Muvunandinda. It was generally agreed by nearly all informants
that Muvunandinda was the most liberal Mututsi chief in Bufumbira since
the Ruandese authority had started encroaching on the sovereignty of the
Bahutu. Muvunandinda and Berabose were the ones who encouraged the
appointment of local chiefs as Ibisonga or sub-chiefs. There was no single
case of Muvunandinda forcefully seizing people's cattle. As an informant
put it:- "Muvunandinda was very strong, tall and massive. He wanted people
to fear him but he was a good ruler and the Bufumbira liked him." He is also
remembered for his courage and reserved disposition. At least his attitude


and behaviour towards the Bahutu was quite different from his predecessors.
In fact when he left Bufumbira most Bahutu wanted him to stay on. He was
a victim of his lowly birth, for when Nyindo grew up he succeeded Barabose
as the ruler of Bufumbira.
The provincial Batutsi chiefs in Bufumbira were assisted by Ibisonga (sub-
chiefs). Some of the most important sub-chiefs were Mushakamba, Birahira,
Mutesi, Hagumakamwe, Nshizirungu, Nyirimpunga and Rwacyendera. These
in turn had lesser chiefs below them, for instance, Mushakamba's lesser
chiefs were Rwenya, Habizana and Rwanyonga. Sub-chiefs were usually
appointed by a provincial chief and confirmed by the Mwami. Birahira was
an exception because he was directly appointed by Mwami Rwabugiri when
he was touring his kingdom. Birahira was appointed chief because he frequent-
ed the provincial royal residence at Mabungo and on various times he acted
as a nightwatchman and attendant to Buki. His visits to Mabungo were
interpreted as a sign of loyalty. In the case of Mushakamba he was known
to be very influential in the Abasinga clan. He succeeded his paternal uncle
Rugyira, who was killed while fighting against the Abatongo., Mutesi too was
very powerful among his dlansmen, the Abarihira. The Batutsi chiefs, as
opposed to the Bahutu, discussed above, had to be close clients of either a
provincial chief in Bufumbira or they had to be already known by prominent
officials at the court in Ruanda. All the chiefs had to possess strong character
and the ability to lead. Each sub-chief ruled several clans. He had a represent-
ative from each clan under his jurisdiction. At times members who did not
belong to the clan of a sub-chief attempted to rebel against this leader. This
was shown by the fact that Mushakamba was constantly opposed by the Aba-
tongo group who consisted of several clans, while Birahira of the Abacyaba
clan was at times opposed by the Abazigaba clan in Nyarusiza.
Batutsi sub-chiefs like Hagumakamwe, Nshizirungu and Nyirimpunga
were in charge of areas where the Batutsi population was large. The areas
inhabited by the Batutsi were mostly plains and grazing areas like Jomba,
but grazing areas were not necessarily the determining factor in locating
Batutsi chiefs, what mattered was the size of the Batutsi population.
The Bahutu too were asked to provide tribute labour to the provincial
chief and sub-chiefs. While the Bahutu provided tribute labour and agricultural
products the Batutsi subjects provided jars of milk. All the tribute was taken
either to Muganza and Mabungo where royal residences were, or to the royal
court in Ruanda. Apart from collecting tribute the sub-chief had the duty
of providing men to defend Bufumbira. Every chief had to provide able-
bodied men to fight against the Belgians when they invaded Bufumbira.
When the Abatongo rebelled every sub-chief provided men to fight. When
the Batwa at the foot of Muhavura mountain revolted, every sub-chief in
Bufumbira had to provide soldiers for Berabose.
(igeri IV Rwabugiri (1854L1895) is remembered as one of th- most
remarkable and powerful Mwami of Ruanda. He is mostly remember-I for
his military conquests. He waged many military conquests against area- sur-
rounding Ruanda. During his time as the Mwami of Ruanda, Bufumbiri was
closely controlled from Ruanda. This was because he sent his daughter
Berabose who proved an able administrator. She consolidated Ruandese control
over Bufumbira by a system of indirect rule somehow acceptable to the people.


Around 1873 Mwami Rwabugiri carried out military campaigns against
the Bakiga east of Lake Bunyonyi; he proceeded to Bubare and Bugeri and
here a powerful Mukiga called Bugi, son of Bujuri from Abasigi dan, was
killed by Rwabugiri's forces. Around 1882 Rwabugiri attacked Ndorwa of
Mpororo. It is believed that he seized a good number of cattle from people
inhabiting Ndorwa. The cows increased the number of cattle that had escaped
rinderpest in Ruanda. In 1885 he attacked Rujumbura and seized a herd
of cattle. About 1890 he attacked a place called Ruhinda and many people
were killed. He even went to attack Nkole under Ntare V. In fact Rwabugiri
has no parallel in the history of Ruanda. Given this military king, revolt in
Bufumbira was likely to be a hopeless failure.
After his death in 1896 Ruandese influence in Bufumbira declined for
various reasons. In 1896 there was a succession dispute. Musinga, a brother
of Rwabugiri's wife called Nyirayhi, usurped the throne from Mibambwe
Rutarindwa who was the heir-apparent. It seems likely that Musinga
became a Gterman puppet so that they would protect him against rival
claimants to the throne. Though he had killed most of Rwabugiri's sons at
Rucuncu there were some Batutsi from the royal clan Abanyinginya who
were against him. Moreover Rwabugiri's wife, Muhumuza, had fled from
Ruanda to Ndorwa where she tried to mobilize the people to fight Musinga
so that her son Buregyeya might become Mwami of Ruanda. Furthermore,
the Bahutu of the eastern Ruanda were in favour of Buregyeya as Dr. Louis
suggests. 2 Under these circumstances it was impossible for Musinga to send
military help to Bufumbira to suppress Batwa rising or to oppose Belgian
penetration in the first decade of the twentieth century.
The Batwa were a minority social group in Bufumbira., They were scorned
by both the Bahutu and the Batutsi and yet they were always called upon
by the authorities in Bufumbira to fight against rebellious groups, as for instance
when the Batwa assisted Mushakamba and Buki to fight against the Abatongo.
In fact the Batwa on the eastern margin of Bufumbira were so powerful that
Mushakamba used to send tribute to Basebya, their leader. They were the
most powerful group militarily and yet were denied political and social recog-
nition in Bufumbira. They excelled groups at shooting with bows and arrows.
They failed to understand how they could remain under the yoke of the
Batutsi and Bahutu permanently. They decided to claim what they considered
as their rights. A group of Batwa at the foot of Muhavura realized their
strength in relation to the other two groups. A Mutwa informant said "they
realized that it was a high time they established their own rule by overthrowing
of the provincial leader, Berabose, and her chiefs like Birahira and Mushaka-
mba. These Batwa had confidence in their ability to shoot with their arrows
and they knew that once they fought against the legitimate authority they
would win".
The leader of the Batwa at the foot of Mount Muhavura was an old
woman called Nyirakote. Her immediate assistants were Mugogo, Mabyiyihene,
Kanyarwanda, Majege and Karerabana. They sent a message to Birahira, a
Muhutu sub-chief of Nyarusiza requesting him to give them one of his most
beloved and beautiful daughters, Nyiramazuru. They wanted Nyiramazuru
to marry Nyirakote's son, Kanyarwanda. They sent another message to Nyindo
to ask whether his half-sister, Berabose, a provincial chief of Bufumbira
would marry one of the Batwa men. They were in other words claiming


equality with the Batutsi and Bahutu. The failure of the two groups to inter-
marry with the Batwa was the sign of Batwa inferiority. Moreover the Batwa
could not become chiefs in Bufumbira. Furthermore intermarriage into the
ruling families would provide the Batwa some political influence. The Batwa
were demanding at least token integration into the social and political system
of Bufumbira.
Birahira and Nyindo took small notice of the Batwa demands and the
latter decided to take up arms. They attacked Birahira's army and defeated
it. For a time it appeared as if the Batwa would take over all of Bufumbira.
Nyindo who since 1898 had accepted Belgian overlordship, called upon the
Belgians for assistance. Some Batwa fled to the southeast of Bufumbira and
some to west of Lake Bunyonyi where they joined Katuregye to raid the neigh-
bouring Bakiga. Batwa leaders like Kanyarwanda, Mabyiyihene and Majegye
joined Katuregye.
Katuregye was supposed to be one of Nyindo's sub-chiefs, but his
allegiance to Nyindo seemed to have been nominal. He was separated from
Bufumbira proper by thick bamboo forest and was so independent in his area
that at times he attacked another of Nyindo's sub-chiefs, Mushakamba
However, Nyindo and Katuregye found a common enemy in the British and
this hatred made both of them collaborate and fight against the British in
1914-1915. But Katuregye did not want to challenge Ruandese authority
having seen the barbarous methods used by Mwami Rwabugiri against the
Another area of Batwa unrest became the scene of an insurrection in the
first decade of the twentieth century was Rugezi in the southeast of Bufumbira.
The Batwa leader was called Basebya. Basebya was a very powerful Mutwa
whose mother was a Mututsi but his father a Mutwa. When Basebya's mother
Nyirantwari was found with an illegimate pregnancy she was chased to a
forest where she was found by a Mutwa and then took her his wife. Then
she gave birth to Basebya. When he grew up he became a leader of the Batwa
in Rugezi. He became so powerful that he became a headache to surrounding
areas like Bufumbira.
One of Mushakamba's sons spoke about Basebya thus: "This was a leader
of the Batwa who rose against Mwami's authority in Rvugezi and he killed
Mahiryori, son of Muvunandinda ... chiefs like Mushakamba sent tributes
to Basebya. At one time Mushakamba's tribute carriers were going to be killed
by Basebya because Mushakamba had spent a very long time without sending
tributes to him."
Combined forces of Bahutu and Batutsi were always defeated by Basebya's
men. They were able to do this because they were skilled bowmen and when-
ever they realized that odds were against them they retreated to inhospitable
places like forests and swamps. The Germans in Ruanda failed to cultivate
good relationships with Basebya and his Batwa fighting men. Before 1911 the
Batwa activities in Bufumbira caused a state of fear and insecurity since Base-
bya used to send his men to collect tributes and at times seized cows and goats
by force. Mushakamba's subjects appeased some of these Batwa by placing
provisions at their disposal. Mushakamba's relatives like Munganda used to
hold beer parties in the honour of these Batwa. On such occasions bullocks,
goats and cows were killed. The Batwa revolt came to an end when Basebya
was sentenced to death by the Germans in 1911.


Batwa activities seemed to indicate that they might have become military
leaders in Bufumbira. They did succeed because the Belgians in Bufumbira
and Germans in Ruanda were against them. The Batutsi and Bahutu could
not have defeated the Batwa, since the latter were experts at shooting with
arrows and their military techniques excelled those of the Bahutu and Batutsi.
It is often forgotten that the Batwa played an important part in the history
of Bufumbira before the effective imposition of colonial administration.

The European scramble for Bufumbira.
The Belgians came from Congo to Bufumbira around 1898. Almost the
whole of present Bufumbira belonged initially to the Belgians for they claimed
all the west of 300 Latitude before the Kivu-Mufumbira Conference of 1910
which finally demarcated the boundaries. The Belgians arrived when Barabose
was ruling Bufumbira. Muvunandinda was assisting her but left for Ruanda
during the first half of the first decade of the twentieth century. Nyindo was
becoming mature enough to take over the administration of Bufumbira.
Nyirimpunga who was a Mututsi sub-chief of Jomba informed Berabose
of the Belgian advance. On hearing of it Muvunandinda, on the advice of
Berabose, decided to gather a fighting force to go and ward off this foreign
invasion. All sub-chiefs were thus ordered to gather all able-bodied men.
Muvunandinda and his men were no match for Belgian bullets and his army
were forced to retreat having suffered several casualties. In the process of
carrying out these deadly raids the Belgians, with their askaris, seized anything
valuable they came across. Life became insecure and many people made caves
their dwelling places. In such a situation no economic activity could be carried
on. During this time of uncertainty and fear, family life in Bufumbira was
disrupted and a great famine locally known as Urwamavuta cleared very many
people and this great human suffering made trade between Ruanda and
Bufumbira flourish. This famine is still remembered by men of over sixty
years of age. Many Bafumbira lives were lost during this period of terrible
suffering. As a result of this some people resorted to exchanging their cattle
for food from Ruanda. Some of the Belgian followers who committed these
atrocities were Mavuta, Gapiganyi, Kamandi and Cyatambehe.
During this reign of terror the provincial leaders of Bufumbira fled.
Berabose left with her brother Nyindo and the remaining sub-chiefs did not
know which policy to pursue. Chiefs like Birahira, Mushakamba and Nyiri-
mpunga bowed to Belgian demands. Tribute in the form of beer, honey, cows
and goats were sent to Rutchuro which was a Belgian centre. Meanwhile
the Belgians sent for Berabose and Nyindo. They decided to rule through
Nyindo. So long as he and his chiefs sent tribute to Rutchuro all was calm.
Thereafter several Belgians with their Congolese askaris would be seen here and
there all over Bufumbira. They started to build small administrative posts in
such places as Nyagisenyi, Kisoro, Chihe and Muhiga. However, they did
not set up a regular administrative system. They were replaced by the British
after the Kivu-Mufumbira dispute had been settled in 1910 and boundaries
delimited. The Belgians held Jomba which had been under a sub-chlief.
Though the Belgian period in Bufumbira was quite short, about ten years,
their rule was very harsh and caused terrible human suffering and humiliations.


Their followers raped African women frequently and there was a lot of human
degradation. Free food was demanded from Africans in Bufumbira. Captain
de Courcey Ireland who participated in the Kivu-Expedition of 1909 agrees
with most Bafumbira informants. "The Belgian soldier in these parts is a
professional looter, and is capable whether on his own account or by order
of his superiors, of sucking as much out of an acre as professionals born and
bred to the work."3
The Belgian occupation of Bufumbira contributed towards the decline
of Ruandese influence in Bufumbira. Tributes which used to be taken to
Ruanda were diverted to Rutchuro. The Batu.tsi chiefs in Bufumbira were
now ruling for the Belgians. Mwami Musinga was unable to save Bufumbira
possibly because he was still confronted with a lot of problems at the capital.
Naturally with the succession rivalries at the capital the border sections of
the empire would be left to themselves.
After 1909 a few British officials like Bgana Kute (Coote) visited Bufu-
mbira. They came with Nubians, Indians, Banyankole and Swahili. After a
series of meetings between these British officials and the Belgians in Bufumbira,
the latter packed up their bags and luggage, to the relief of the Bafumbira.
One informant has this to say about the British; "Unlike the Belgians they
did not come demanding goats and cows. They never shot anybody for no reas-
on at all. They were welcomed by Mushakamba in Nyakabande... they told
Nyindo that they had come to stay and that Bufumbira was theirs." In October
1912 Nyindo was confirmed as the leader of Bufumbira and the chiefs who
had been under him continued ruling areas they had been in charge of. Nyindo
was given the title of Intebe, an equivalent of a County chief, while his sub-
chiefs' title became Gomborora.
Nyindo was allowed to visit his half-brother, Musinga, in Ruanda and he
used to send tribute to the royal court in Ruanda. Some Batutsi clients are
said to have overtly resumed taking tribute to Ruandese authorities. But the
amount of tribute sent to Musinga from Bufumbira was not as enormous as
in the period before Belgian occupation of Bufumbira. Nyindo was made to
understand that sending tribute to Ruanda'did not mean that he was supposed
to forget his obligations and duties to the British government. Since Nyindo
seemed to be a very understanding man he decided to comply to British
A Muganda, Abdulla Namunye, was appointed a British agent to advise
and supervise Nyindo and his subordinate clients. This was supposed to be
a temporary measure. It was planned that once the indigenous people could
be trusted then Namunye would quit Bufumbira. Up to 1915 the British did
not introduce major innovations apart from instituting a poll tax. Each
individual of taxable capacity had to pay three rupees.. If a person failed
to get three rupees, he was asked to pay three goats, but since there was
no cash economy, unpaid labour was encouraged. Some Bafumbira used to go
to Kabale to build official buildings and constructed roads instead of paying
three rupees. 4
After 1913 the relationship between Nyindo and Nyamunye was not very
cordial, may be because Nyindo realized that the power rested with Nyamu-
nye for all orders were coming from him. Moreover Namunye had be-
friended nearly all the chiefs that mattered and this was really a threat
to Nyindo. But this could not have precipitated a revolt. The British


officials in Kigezi were nevertheless surprised and disappointed when Nyindo
rebelled against them. Initially they thought that if they appointed Nyindo as
a paramount chief of Bufumbira he would not be won over by his half-brother,
Musinga in Ruanda. It seemed to be a sort of appeasement by giving him
such a responsible post. However, they were mistaken since he found it more
convenient to obey his half-brother.
Nyindo, would not have revolted at the time that he did had it not been
the outbreak of the first World War. The Germans in Ruanda asked Musinga
to advise his half-brother, Nyindo, to rebel against the British in Bufumbira.
Since Musinga was a sort of German puppet he had no other alternative but
to comply. One of Nyindo's clients has this to say: "Musinga urged Nyindo
to rebel against the British in Bufumbira. Musinga asked Nyindo where he
would go after the Germans had defeated the British... some Batutsi in
Bufumbira advised him to' support the British against the Germans......
Musinga said to. Nyindo. 'let us work together' ".
Nyindo opened his revolt when Abdulla Nyamunye asked him to provide
porters to carry the luggage of a British official. In fact Nyamunye's mes-
senger narrowly escaped death. When Nyamunye personally went to look
for carriers he was attacked by Nyindo's clients and sympathisers. There
was a short but decisive battle between Nyindo's men and Nyamunye's
fighting force called Ibitaramukanya. Meanwhile Nyamunye went to Ikumba
to report Nyindo's insubordination to the Acting District Commissioner,
Sullivan. Before this incident, Nyindo had already rebelled against the
British authority in a disguised manner since he usually boycotted meetings
with Sullivan. Then Sullivan came to see what had happened to. Nyindo who
pretended to be sick.
Nyindo's revolt was not supported by all subordinate chiefs. Most chiefs
sided with Nyamunye. Mushakamba, Mutesi and Hagumakamwe stayed
wi"h the British, while Birahira supported Nyindo. Mushakamba and Mutesi
were probably influenced by the fact that Nyamunye had married two
daughters of the former and one of the latter. Birahira possibly chose to
support Nyindo because he had been given large herds of cattle by Mwami
Rwabugiri and Nyindo. When Rwabugiri visited Bufumbira during the time
of Buki he found that Birahira was a royal attendant of the royal residence
and so he (Birahira) was given very many cattle.
While Nyamunye was at Ikumba, Nyindo's men harassed Mushakamba
and his people. They burnt several houses belonging to Mushakamba's sub-
jects who supported British rule. Kanyamihigo's house was burnt and this
Manyamihigo was Mushakamba's son. When Nyamunye returned with
Sullivan and a handful of policemen, Nyindo's men came to attack. Sullivan
threatened these Batutsi attackers by shooting whoever came near him.
This took place in 1914. Meanwhile a very big group of Batutsi from Murela,
under Nyindo's leadership came to attack Sullivan at Lake Kigezi. Sullivan
under pressure from Mushakamba decided to disperse these attackers.
They were ultimately pushed back into Ruanda. During this campaign the
Batwa on Nyindo's side gave the British their greatest trouble for their fight-
ing technique excellent that of Batutsi and Bahutu. "The Batwa seem to
have no respect for rifle, and are adept at taking cover, crawling from


mound to mound, wriggling like snakes, firing arrows and crawling away
again, hence they are difficult to hit."5
Since German influence over Nyindo was a very serious threat to the
British hold over Bufumbira, Sullivan expected that soon he could expect
a serious attack from Ruanda. He therefore began to fortify strategic points,
like Murora near Lake Chahafi and made trenches on the top of Murora hill.
Early in 1915 an invading force of Germans heavily supported by Nyindo's
supporters crossed over from Ruanda and attacked the British at Chahafi.
An informant reported that:- "Before the battle took place, the British had
already made some preparations for they had dug trenches. Then one morning
at dawn the Germans began to shoot. The British were hiding themselves
in these trenches. When German ammunition became exhausted the British
started to reply by shooting several Germans and Banyarwanda fighting men.
... The Germans were defeated and went back to Ruanda."
At the same time Katuregye was disturbing British communication across
Lake Bunyonyi. It is said that he was a spy of the Germans. It could be so, be-
cause Katuregye was extremely anti-British because they had arrested his
mother, Changandushi, who was a chief priestess of Nyabingi. She was deport-
ed to Mbarara and died on her way back. Katuregye trusted the military
strength of his Batwa supporters, but he was shot when attempting to hide
himself in the bamboo forest and died shortly after.6
British soldiers crossed into Ruanda with the intention of seizing Nyindo.
While in Ruanda the soldiers began seizing people's cattle. Musinga became
alarmed and advised Nyindo to go back and die for his country. So Nyindo
decided to go back and gave himself up. Having surrendered to a British
officer, Sullivan, Nyindo was deported to Kampala where he met Nyiragahu-
muza (Muhumuza). Muhumuza was Rwabugiri's wife but fled Ruanda when
there was a succession conflict. She wanted her son, Buregyeya, to assume the
throne of Ruanda instead of Musinga. So she ruled over the inhabitants of
the British Ndorwa with the military assistance of the Mutwa chief, Basebya.
Muhumuza was arrested by the British and deported to Kampala. Nyindo
was taken to Fort Portal where he was kindly received by Mukama Daudi
Kasagama. Nyindo was set free on the understanding that he would return
to Ruanda and not Bufumbira. When he reached the Kigezi-Ankole border at
Rwentobo he died of heart failure.
Why were most people in Bufumbira ready to fight alongside the British
instead of helping their chief, Nyindo? Nyamunye's activities in Bufumbira
so far as creating loyal subjects were concerned, had been very effective. He
already had very substantial support from chiefs like Mushakamba, Mutesi,
Haguhakamwe who were prepared to follow his leadership; one of whom
advised "I recommend to your notice the work of Agent Abdulla throughout
this trying period and also chiefs Mushakamba and Mutesi who remained loyal
to us throughout." Since the subordinate chiefs had experienced Belgian ad-
ministration which had been very oppressive they were ready to side with the
British whose intervention in Bufumbira affairs at this time was minimal.
Given the fact that the Bahutu chances of independence were nil, it was best
from their point of view that someone other than the Batutsi should rule;
someone who might be expected to be impartial and treat Bahutu and Batutsi
alike. Since the Germans were identified with the Batutsi and given the recent
experience of the Belgians, the British appeared the least of a number of


evils. Thus the majority of Bahutu, but not all, supported the British. On
the other hand the Batutsi generally identified themselves with Nyindo, the
Ruanda court and the Germans, but again not all of them did so. For example,
the Mututsi chief, Hagumakamwe, did not support Nyindo. The Batwa were
at least consistent, they fought against every new imperialist into Bufumbira
and the British were merely the latest of these.

Ntokibiri's revolt 1915-1919.
Ntokibiri was a Muhunde who seems to have been living among Hunde
tribesmen. His home was in Nyamikumba, Nyabgishenye. This Nyabgishenye
is just on the Uganda-Congo border. He was a very short man and one of
his hands lacked three fingers. His other name was Bicubirenga because when-
ever the British and the Belgians tried to trace him they failed, for
his movements were like those of clouds (Ibicu). He could be said to be one
of the champions of the African cause in Kigezi because he tried with all his
might to get rid of all the British and Belgians in Bufumbira. He, like any
African nationalist, resented European encroachment upon the sovereignty
of Africans. He was a very able leader and he had a very strong following
from all social groups in Bufumbira. He brought members from different
clans together so as to fight against colonialism.
His sheep, Nyiramatabaro, was believed to be in possession of mysterious
powers because it always went with Ntokibiri. It was always said that no
bullet could kill this sheep. Whenever it was shot at by the British fighting
men it is said that it jumped up and down. So this sheep was a rallying force.
Moreover it was believed that nobody could kill Ntokibiri by employing
maxim guns.
When Ntokibiri heard that Nyindo had fled Bufumbira, he found it a
grand opportunity to go and fight against imperialists.8 He went to Ruanda
where he found Nyindo. He told Nyindo that he was prepared to go and
fight for him and he assured Nyindo that he would defeat the British. He
told the same story to the Germans in Ruanda. Ntokibiri insisted that so
long as Nyindo gave him men to fight with him, he would accomplish his
overriding motive of defeating the British in Bufumbira. Nyindo reluctantly
agreed to send fighting men under Kayijamahe, a Mututsi. Many of Nyindo'j
clients who fled to Ruanda with him went with Ntokibiri to fight. In Bufumbira
he faced Nyamunye's fighting force called Ibitaramukanya. Ntokibiri achieved
an initial success when he killed Sayidi and Kantukundi, Nyamunye's leading
fighters. Ntokibiri captured three rifles and ammunition. Though his force
was very large, he was overwhelmed by a British force which made him
flee to Nyabgishenye forest.
It is very interesting that the people who joined him had varied motives.
Some may have followed him because they thought he had the power of
Nyabingi, but it is difficult to identify Ntokibiri with Nyabingi. Instead a
majority of his followers thought his sheep, Nyiramatabaro, was possessed
with mysterious power. Thus this sacred sheep rallied some strong support,
able to withstand the power of the British for about three years. Yet some Ba-
nyaruanda from Mulera and Bukamba followed Ntokibiri because they thought
that once he defeated the British, Nyindo would go back' to Bufumbira and
that Bufumbira would become a German area. Hence it can be clearly seen


that some people hoped Nyindo would rule Bufumbira under the Germans
when the British failed to defeat Ntokibiri. Though the original motive in
joining Ntokibiri was to fight against the British, other motives came into
play during the course of events. Some realized that they would get rich if
they joined Ntokibiri since looting activities were very common. So such
people did not care whether the British were victorious or not, though it is
natural one wants one's own side to win. Then there were some people who
joined Ntokibiri through sheer panic. Some came to believe that Ntokibiri
was a great man with unequaLed powers since he was challenging people
whom Nyindo and the Germans failed to defeat. Thus Ntokibiri's prolonged
activities against the British drew to him very many supporters.
As the movement advanced, Ntokibiri became overwhelmed by British
and Belgian forces. When he realized that he could not withstand maxim
guns of the British he escaped to northern Bufumbira. He hid himself in
Kayonza forest. The British chased him in vain. His failure in Bufumbira
is believed to have been caused by his marrying a woman against gods' wishes.
That he became lustful and so god deserted him. The fact is that the British
mobilized all their available military resources against his movement. From
Kayonza he went to Rubanda. His intention was to go and attack the British
at Kabale. Before he could do this he gathered support among the Bakiga
of that area. He started instituting blood pacts among the leading Bakiga.
The essence of this was believed to be that a blood-brother could not betray
his fellow blood-brothers. Nevertheless his goal was shattered when a Mukiga
chief, called Bikaku, betrayed him and told the British his whereabouts. He
was found in Bikaku's house where he was shot. His sheep was also killed
and burnt to ashes. His head was taken to Britain while his hand with two
fingers was taken to Bufumbira by my informant, Sempigyi. This was to
show the Bafumbira that if anybody tried to resist the British he would suffer
Ntokibiri was defeated because there was no unity of action among the
Africans. There was lack of nationalism or patriotism. If, for instance, all
the Bafumbira and Bakiga supported him he would have been able to resist
the British a little longer. Most Bahutu in Bufumbira had been won over
by Nyamunye through chiefs like Mushakamba and Mutesi and one Mututsi
chief, Hagumakamwe. In Rubanda a Mukiga chief, Bikaku, betrayed him.
Moreover the British maxim guns were effective in making some of his sup-
porters fly away from him. He is however remembered as a man who resisted
the Europeans for a long time.
Unlike Nyindo who collaborated with the Germans, Ntokibiri was not
prepared to compromise with any foreigner. He was driven by a sense of
purpose of restoring Africans' sovereignty. He might have changed the course
of Bufumbira's history had he managed to get total support of Africans. His
death is a landmark in the history of Kigezi.



1. A list of the k!ngs of Buanda is found in Inganji Karinga by A. Kagame.
2. See W. R. Louis. Ruanda-Urundi 1884-1919. London, 1963. p
3. De Covrcey Ireland as quoted in W.R. Louis, Ibid p. 18q.
4. Ngologoza, P. Kigezi N'Abantu Baamwo, Kampala, 1967,
5. Thomas, H. B. Kigezi operations, 1914-1917. (Uganda Journal, VoL SO, No. 2,
1966,% p.167.)
6. This appraisal of the role of Katuregeye relies on research material, as yet un-
published, of Mr. Rwabihigi.
7. Ntokibiri's revolt is written up in detail ,n F. S. Brazier, The incident at
Nyakishenyi,, 1917 (Uganda Journal, Vol 32, No. 1, 1968, pp.. 17-27); but
neither in this nor in M. J. Bessel, Nyabingi, (Uganda Journal, Vol. 6, No. 2,
1938, pp, 73-86) is the role of the sheep (lyiramatabaro) accurately portrayed.
These writers identify the sheep with the Nyab:ngi cult but this was not exactly
the case. Rather its significance lies in bringing various clans and different social
groups together under Ntokibiri.
8. Thomas, op.cit. p.166 poses the suggestion that local historians might be able to
establish what relations existed between Nyis'lo and Ntokiribi (Ndochibiri). This
paragraph throws some light on the problenr


For 1969 membership was much the same as for 1968, with a slight
increase in the number of institutional members but with a further substantial
increase in the number of associates. Tne fact that in one year nearly a
hundred people left the Society indicates a very rapid turnover of member-
ship. The financial resources of the Society were substantially reduced during
the year due to high costs of printing, maintaining the premises and running
the Society office. Also the sale of back numbers of the Journal lessened in
1969. As in the previous year the Journal benefited from a generous donation
from the Milton Obote Foundation towards the cost of printing the Uganda
The Society maintained its programme of regular monthly lecture meet-
ings. The year had also been an especially active one for the Natural History
section. Numerous excursions were held, and of particular note were two
week-end meetings, one to lake Nabugabo and one to Budongo forest. In
addition a bird study group was formed which has organised group bird
watching meetings and which produces a monthly newsletter of lists of birds
spotted. There has also been a monthly lecture meeting on natural history.



28 January
26 February

28 February-
2 March
18 March
19 March

22 March
16 April

30 April
20 May

21 May

24 May
28 May
14 June
17 June
18 June

15 July
16 July
13 August
19 August

29-31 August
23 September

24 September
15 October
15 October
16 October
2 November
5 November
12 November

Plant quarantine-Dr. J. Mukiibi.
Oxen, age-mates and enemies: music making in Karamoja -
Mr. A Gourlay.

Expedition to Budongo forest.
Research on U'ganda rodents Professor M, D. Delaney.
Different traditional musical cultures in Ankole Rev. Fr. P. van
Bird watching excursion to Kajansi-Mr. R. Frankum.
Myth, memoir and moral admonition: historical writings in Luganda
-Dr. J, A. Rowe.
Evangelists and sub-imperialists-Dr, M. L. Pirouet.
The pied kingfisher Mr. R. J. Douthwaite and a film 'Feathered
safari' .Dr. G. Whitehead,
A social history of Kibuli: continuity through change-Miss R. M.
Excursion to Nkumba peninsula
Annual General Meeting, sherry party and film 'The permanent way'.
Excursion to Zika forest.
Pottos: facts and fables-Dr. A. C. Walker.
Westminster Abbey, or the growth of a vision at Mbale-Rev. K.
H. Sharpe,
Angling in Uganda-Mr. B. Cole.
Factional conflicts at Mwangai's court-Mr. D. Kavulu
Ten's company; Makerere Travelling Theatre-Professor D. Cook.
The need for conservation cropping in National Parks-Mr. R. G.
Expedition to lake Nabugabo.
Current aims and techniques in the management of Uganda's indig-
enous forests-Mr. T. J. Synnott.
The lyrical poetry of Bahima women- G. Nshemerierwe.
Birds of he Kampala area-a discussion.
The cultural basis for a literature of Africa-Mr. M. N. Okenimkpe.
Excursion cruise on Murchison Bay,
Birds of the Kampala area-a discussion.
African oral literature-Professor J4 Mbiti.
Population and protein: a new legume project for Uganda-Mr. C.

Uganda Journal, 34, 1 (1970) pp. 49-63.


A Preliminary Annotated Check-List.


Queen Elizabeth National Park was established in 1952, and since then
much scientific work has been carried out, mainly connected with the high
hippopotamus populations. In 1961 the Nuffield Unit of Tropical Animal
Ecology began work and collected plant specimens from the start. These
have been accumulated into an herbarium which now contails over a thousand
Most of the collections have been made by the writer, and his collections
have been identified at the East African Herbarium. C.R., Field of the Nuffield
Unit also collected and identified many grasses during his studies of the
comparative feeding habits of the large herbivores (Field 1968). Other
collections have been made by D. D. Thornton, A. J. Brooks, Dr E. M. Lind,
and H. K. Buechner. This list is based mainly on collections and observations
made by the author between 1963 and 1969.
The Park lies in the western rift valley of Uganda, south of the Ruwenzori
range and adjoining Lakes Edward and George. The equator passes through
the northern part. The climate is of a dry equatorial type, with two distinct
rainy seasons each year, and the mean monthly maxima and minima tem-
peratures lie near 280C abnd 180C respectively, throughout the year. The
annual rainfall varies from 1200 mm near the escarpment in the Ankole
section, to 600 in areas near Lakes Edward and George. Areas described as
having high or low rainfall approach these extremes; other areas are inter-
mediate. Soils are mostly relatively young, developed on volcanic ashes or
lake sediments of Pleistocene age. They have been described by Harrop (1960)
and his nomenclature is used here.
A subjective scale of grazing intensity is used in the descriptions of areas.
It is based on the normal appearance of the grassland which depends on the
relationship between primary productivity and removal by grazing. Over-
grazing, or excessive grazing, caused particularly by hippopotamuses, damages
the grassland and produces bare patches with erosion of soil by rain and wind.
Heavy grazing reduces a grassland to a lawnlike structure, preventing the
flowering of taller species except in very favourable seasons. Moderately
grazed areas show little inhibition of flowering and taller grass growth; but
there may be patchiness due to selective grazing. Lightly grazed areas are not
obviously affected by the few animals seen to graze them. The frequency and
intensity of fires is inversely proportional to the grazing intensity, and most
lightly or moderately grazed areas are burned at least once every two years.
For recording purposes the Park has been divided into eleven blocks.
The primary division is a political one, into the three districts of Toro, Ankole
and Kigezi. Use of the county boundaries for further subdivision is impractical,
as much of the Park lies in only two counties, Busongora and Bunyaruguru.
Further division is therefore along convenient natural or artificial boundaries


into blocks which are roughly of equal size. Some of these are ecologically
uniform, as well as convenient, others are less so.

T1. Nyamugasani Kayanja This is a lakeside trip, mostly less than
100 feet above the level of the lake (3000 feet a.s.l.). The area is generally
flat and undissected. At the south end a series of old craters has been broken
into and filled by the- swampy delta of the Nyamugasani river. Much of
the delta is occupied by the Kihabule Local Forest Reserve, mainly dense
scrub of Capparis tomentosa, Euphorbia candelabrum, Turraea robusta, and
Securinega u.rosa, with some trees of Crolton macrostachyus and Acacia kirkii.
At the north end of the area there are extensive stands of Acacia sieberiana
woodland, and some thicket along the Mpondwe river, the boundary with the
Congo Republic. There is a narrow marshy strip along the whole shoreline
with some lagoons near Kayanja. Fires are frequent when there is enough
grass to burn. Rainfall increases northwards from very low values at Nyamu-
gasani, to moderate at Mpondwe. Grazing is generally heavy, particularly
near the lake where there is a high hippopotamus population. The soils are
back clay-loams of the Nyakatonzi Series, derived from volcanic ashes.
T2. Craters. This is a hilly area, including about eighty explosion craters
about 10,000 years old (de Heinzelein 1957). There are very steep slopes and
cliffs round some crater rims, and the land rises to 4480 feet at Kyamutuma,
the highest point in the Park. Five of the craters contain salt lakes, and some
contain xerophyllous woodland of Olea chrysophylla, Cordia ovalis and
Euphorbia candelabrum; this woodland is largely unexplored botanically. The
Nyamugasani river runs in a deep gorge along the northwest border. Much
of the area is covered with Themeda triandra or Cymbopogon afronardusl
Imperata cylindrica grassland, with scattered trees of Acacia gerrardii. Soils
are mainly Kyamatoma Catena with some Nyakatonzi Series, both derived
from volcanic ashes. Grazing is generally light, and fires are frequent, although
efforts are now being made to check them with firebreaks.
T3. Mweya Katunguru. This, the most intensively studied section of
the Park, is a gently dissected area, with rolling ridges in the south, and a
broad plain, sloping genlty up to the crater foothills, in the north. The south-
ern part is Sporobolus pyramidalis grassland with many Capparis tomentosa
- dominated thickets, and the northern part is a mixture of Themeda and
Imperata grasslands, with more widely scattered thicket clumps (Lock 1967).
Grazing is moderate, becoming heavy or excessive near the Kazinga Channel
and large permanent hippopotamus wallows. Fires occur regularly in the
northern part, but very rarely in the Sporobolus grassland. Soils are Nyakatonzi
Series, tending towards Chambura Series in the south (Lock 1967).
T4. Kasenyi Hamukungu. This is probably the driest part of the Park.
It is rather level with gently undulating ridges and valleys. Much of the
drainage is internal into broad shallow valleys, and these and the shores of
Lake George support extensive swamp communities. The two craters in the
area both contain salt lakes. Grazing is moderate or heavy, becoming excessive
near Lake George where there are large numbers of hippopotamuses. Fires
are common in the west of the area, but become scarcer in the east where
there is not usually enough vegetation to support them. Hyparrhenia


filipendula/Heteropogon contortus grassland with thicket clumps covers much
of the area; Themeda is absent except in the extreme west and around Lake
Bunyampaka. Soils are Nyakatonzi Series, but are much richer in clay than
other areas so mapped and become very sticky and temporarily flooded in
wet weather.
T5. Kamulikwezi. This block is composed of alluvium washed down from
the Ruwenzori foothills. The strip along the northwest edge slopes gently
and is dry, with Acacia gerrardii scrub and Balanites aegyptiaca wooded
grassland. At the foot of this slope there are alkaline seepages in several places.
The lower, main part of the area is an almost level silty plain with traces of
many old river courses. Much is covered by short mixed grassland with some
patches of Acacia sieberiana woodland, and extensive thickets of Capparis
tomentosa. There are extensive swampy grasslands along the lake shore and
also large open patches of Sporobolus range grassland on alkaline areas. The
area is generally heavily grazed, with overgrazing near the lake. Fires are
frequent in the northwestern part of the area; but elsewhere there is rarely
enough grass. Soils are generally fine-textured and sedimentary in origin, of
the Kasese Series or Sebwe Series.
T6. Kyondo Dura. This is a large, heterogeneous, inaccessible and
underworked block. In the western part there is tall grassland with much
Imperata cylindrica; and the only Borassus aethiopium palms in the Park.
Much of the rest of the area is very wet with much swamp of Cyperus papyrus,
and swamp forest and with large stands of Phoenix reclinata and Raphia sp.
To the east there is a narrow strip of Hyparrhenia / Themeda grassland below
the escarpment. Grazing is rather light over much of the area and absent
in the very swampy parts, but isolated herds of hippopotamuses cause local
heavy grazing, even of Imperata. Soils are of several types; the grassland
below the escarpment is on Bubandi Series loams, the central swamps on
Papyrus Peat, and most of the western part is on Sebwe Series sandy clay-


Al. Lion Bay Kazinga. This block is similar in many ways to T3, which
lies opposite it on the north side of the Kazinga Channel. The main vegetation
is rather short grass with scattered thicket clumps. In the northeast, there is
an area of denser thicket of Capparis tomentosa, and thickets are also denser
in the southwest where Tarenna graveolens with Capparis is common. Most
of the area is a gently rolling plateau around 200 feet above lake level. There
are no permanent streams. Fires occur regularly, decreasing in frequency
near Lake Edward. Grazing is excessive along the Kazinga Channel, and was
formerly also near Lake Edward before the removal of the Lion Bay hippopo-
tamus population in 1962-63 (Laws 1963). Rapid changes have taken place
in the vegetation near the Lake since then (Field 1968). Soils are mainly
Nyakatonzi Series and Rwanga Series with some Chambura Series and
Ishasha Series. All are rather fine-grained. The relative complexity of the
soils in this area reflects the thinning out of the ash deposits southwards, with
Nyakatonzi Series and Chambura Series soils being derived from the volcanic
ashes from the Busongora and Bunyaruguru crater areas, and the Rwanga





+* Iternational
- Block

Win Main Rad
S Salt ake
0 Fmrwater Lake
R Swamp
-w- Rolahay


and Ishasha Series being derived from the underlying lake sediments to the
A2. Rutanda. The Katunguru Ishasha Road -.onveniently divides this
block from Al, coinciding as it does, approximately, with the boundary
between various shorter grasslands in Al and longer Cymbopogon afronardus
/ Imperata cylindrica grasslands in A2. There is also a steep rainfall gradient,
from being high near the escarpment to lower in the west, which is reflected
in the vegetation differences between Al and A2. The topography is similar
to Al. One semi-permanent stream, the Rutanda river, meanders northwards
through the block, passing through the Kibona swamp on the way. The
Kibona swamp includes some Cleistopholis patens forest which has not been
visited. Grazing is generally light or moderate. Fires occur frequently and are
often intense. Soils are fine-grained loams, Chambura Series in the northeast
and Rwanga Series to the southwest.
A3. North Maramagambo. This large area of forest reserve lying within
the Park has not been adequately studied. Few tracks lead into it, but it is
probable that few species remain unseen except perhaps on the grassy hills
near the escarpment. There are various forest types including some Cynometra,
but much of the forest is known only from the aerial photographs. The
topography is more deeply dissected than that of the grasslands to the north
and there are several permanent streams. Grazing is heavy near the forest
lakes where there are many hippopotamuses, but light elsewhere. Fires occur
occasionally on the grassy hills near the escarpment. Soils are Rwanga Series
below the escarpment and Ishasha Complex near the lake. Both are developed
on rather fine-grained sediments.


K1. South Maramagambo Bitereko Rwenshamai. This block is made
up of two separate parts. The southeast section is forest, including some mature
stands of Cynometra alexandri, and also some grassy hills at Bitereko Ranger
Post. These support the only tall grassland on stony hill slope soil in the
Park. This type of grassland is widespread in western Uganda and it is
unfortunate that the only protected examples of it have now been much
damaged by cultivators encroaching on the Park. A recently cut track just
outside the Park boundary has allowed visits to this area and to the forest,
and records from along this track are included in the Park list. The ground
is well dissected in places, paridularly rear the escarpment and in the
Cynometra forest. Rainfall is highest near the escarpment and is high through-
out. The northwest section is drier, with much scrubby forest, degenerating
in places into thicket clumps. Soils are mainly Rwanga Series, with Ishasha
Complex on the low plain near the lake. Grazing is light in the forest, but
moderate near the lake. Fires occur occasionally in the grasslands in both
K2. Ishasha. This is a large and varied block with more species recorded
than any other, but for convenience it is not subdivided. There are riverine
forests including extensive stands of Euphorbia daWei along the Ishasha and
Ntungwe rivers. On the silty lacustrine flats there are extensive thickets of
Capparis tomentosa and C. fascicularis and there are also large swamps along
the lake margins. The ground slopes gently towards the lake. Near Kikeri


and between Ishasha Camp and Katoke there are large areas of open grassland
on rather sandy soils with low rainfall. These open grasslands are heavily
grazed, but elsewhere grazing is light or moderate. Soils are of the Ishasha


I am grateful to the staff of he East African Herbarium for their help,
and particularly to Miss C. H. S. Kabuye who has identified all my grass
collections and has always been most helpful in answering queries and checking
nomenclature. I am also grateful to Dr S. K. Eltringham, Director of the
Nuffield Unit, Professor C. D. Pigott, C. R. Field and R. G. Strugnell who
read draf s of this list and assisted in its compilation. Most of the work has
been done during the tenure of a Ford Research Fellowship and whilst
attached to the Nuffield Unit of Tropical Animal Ecology.

Brooks, A. C. (1957). Notes on some ecological studies in the Queen Elzabeth National
Park with particular reference to grasses.
Unpublished Report, Uganda Game and Fisheries Department.
Field, C. R. (1968). The food habits of some wild ungulates in Uganda.
Unpublished Ph.D. Thesis, University of Cambridge.
Harker,, K. W. and Napper. D. M. (1961). An Illustrated Guide to the Grasses of
Uganda. Government Printer, Entebbe.
Harrop, J. F. (1960). The soils of the Western Province of Uganda.
Memoirs of the Research Divis:on, Department of Agriculture, Series 1, No. 6.
Heady, H. F. (1966). Influence of grazing on the composition of Themeda triandra
grassland, East Africa. (Journal of Ecology. Vol. 54. pp. 705-747.)
de Heinzelein de Braucourt, J. (1957). Les Fouilles d'Ishango.
Exploration Pare National Albert Mission J. de Heinzelein de Braucourt
Fasc. 1.
Laws, R. M. (1963). The Nuffield Unit of Tropical An'mal Ecology Report for 1962-63
In Report and Accounts of the Trustees of the Uganda National Parks, 1961-62.
Lock, J, M. (1967). Vegetation in relation to grazing and soil in Queen Elizabeth
National Park, Uganda. Unpublished Ph.D. Thesis, University of Cambridge.
Napper, D. M. (1965).Grasses of Tanganyika, with keys for identification. Tanzania,
Ministry of Agriculture, Forestry and Wildlife: Bulletin No. 18.



Nomenclature follows Napper (1967), except for certain genera which have
been revised for Flora of Tropical East Africa: Gramineae (Clayton; in prepn.).
Where this differs from that in Harker and Napper (1961), the name used in
the latter work is given in brackets. Areas where the species has been recorded
are listed after the name; areas where the species is a conspicuous component
of the vegetation are marked with an asterisk. The collections cited here
(L. = J. M. Lock) are ones which have been checked by the East African
Herbarium, Nairobi. All except those in bracke's are represented in the
collections at the Nuffield Unit, and at Nairobi.
Acroceras zizanoides (H. B. K.) Dandy. L.855 A3, Kl.
Rare, in moist places in forests.
Andropogon canaliculatus Schumach. vir. fastigiens Stapf
L.69/99. Al-3, K1,2.
Uncommon, in Themeda / Hyparrhenia grassland, and also in
shorter patches in moister areas.
Aristida adoensis A. Rich. L.69/234. Al-3, K1,2,
Frequent, in dry heavily grazed grassland. A dense tussock forming
perennial which appears to replace Sporobolus pyramidalis in very
dry situations.
Aristida adscensionis L. L.536. T1-5. Al. K1,2.
A frequent small annual of bare places and tracksides. Often round
Aristida barbicollis Trin.. & Rupr. (A. lommellii). L.69/230. K2.
Sandy patches in Themeda / Hyparrhenia / Brachiaria platynota
grassland at Ishasha; rare.
Beckeropsis uniseta (Nees) Robyns L.69/223. T6., A2,3. K1.
Tall perennial of drainage lines in Cymbopogon / Imperata grassland
in A2; more widespread in the wettest parts of the Park.
Bothriochloa insculpta (A. Rich.) A. Camus (Field s.n.)
Bothriochloa pertusa (L.) A. Camus (L.11)
Bothriochloa radicans (Lehm.) A. Camus Brooks 53.
These species have been much confused throughout this study, and
it is not possible to give definite ranges for each one. The genus as
a whole occurs in all blocks. Such evidence as there is suggests that
B.pertusa is the commonest and most widespread species. B.radicans
occurs with it in heavily grazed areas, and B.insculpta appears in
higher rainfall areas.
Bothriochloa glabra has been recorded in error in the past.
Brachiaria brizantha (A. Rich.) Stapf L.69/212. T1-3, A2, K1,2.
Frequent in taller and moister grasslands.
Brachiaria eminii (Mez) Robyns (B. decumbens)Tl-5. Al*, A2*,3. Kl,2*


Common, particularly in heavily grazed areas, although it is itself
heavily grazed. In T3 it often forms a narrow zone round thicket
clumps, and has some preference for sandy soils and shade. It is
variable in size, particularly in moister areas.
Brachiaria platynota (K. Schum.) Robyns L.69/227. A1*,A2*,A3. K1,K2*.
Common, in drier grasslands, to the south of the Kazinga Channel
only. It is heavily grazed and forms small tussocks. Its absence from
the north of the Park is probably an accident of dispersal; a plant,
accidentally introduced, flourished for some years in a lawn at
Brachiaria rddicans Napper L.69/53. T1,3,4,6. Al. K1,2.
Frequent in moist places, often on lake shores with Panicum repens,
and in abandoned wallows in Al. A recently described species.
Brachiaria scalaris (Mez) Pilger L.69/392. A3, K1.
Rare, on disturbed ground in the wettest parts of the Park.
Brachiaria soluta Stapf L.683. A2.
Rare, in moist well-grazed hollows in Cymbopogon / Imperata
Cenchrus ciliaris L. L.520. T1-5. A1,2.
Frequent in the drier parts of the Park, but appears to be absent
from the extreme south. It occurs in lightly grazed Sporobolus
pyramidalis grassland, and around termitaria, and also on dry slopes
where it is always stunted. It is always heavily grazed where it
Chloris gayana Kunth L.521. T1,2*,3*,4*,5*,6. Al*,2. K1,2-
Widespread, in the drier grasslands. It is commonest in the
Sporobolus pyramidalis grasslands, and it is also common in slightly
saline or alkaline conditions near salt lakes, and in T5, In Themeda
grassland it is less common. It is always heavily grazed.
Chloris pycnothrix Trin. L.519. T1-5. Al-3. 11-2.
Frequent small aanual of dry places, on bare ground, roadsides and
disturbed ground, where it varies considerably in abundance from
year to year.
Chloris roxburghiana Schult. (C.myriostachya) L.535. T1-5. Al. K2.
Frequent, in dry grasslands. The inflorescences are conspicuous.
Often found in dry microhabitats, such as the tops of erosion terraces,
in otherwise moist situations.
Chloris virgata Sw. L.534. T1,3,4. A1,2. Kl-2-
Uncommon, in similar habitats to Chloris pycnothrix.
Chrysochloa orientalis (C. E. Hubbard), Swallen
L.69/199. T1,2,3,4,5. Al. K2.
Locally abundant in eroded and overgrazed areas. Apart from its
perennial habit, it can be distinguished from Chloris pycnothrix,
which it resembles greatly when sterile, by the asymmetrical leaf
tips. (C. R. Field, p.c.) Early collections from the Mweya area
were wronglynamed as Axonopus compressus.
Cleistachne sorghoides Benth. A3, K1.
Rare, on disturbed ground in moist tall grasslands.


Coix lachryma-jobi L. L.68/11. T1.
One record from a riverside sandbank. An escape from cultivation.
Commelinidium mayumbense Stapf L.69/141 K1.
By forest tracks in deep shade, where it is frequent.
Cymbopigon afronardus Staf L.708. TK1,2*,3,5,6- A1*,2*,3- Kl*,2-
This large tussock-forming species is abundant and widely dominant
in the wetter areas of the Park. It flowers copiously after burning,
and scarcely at all in later seasons. It is little grazed except by
elephant. In Lutoro/Lunyankole the name of this species is Omutete.
One crater in T2 is called Rwamtete (place of tete), suggesting
that this species has been a feature of the place for some time.
Cymbopogon excavatus (Hochst.) Stapf L.593. T2,3,5.
Uncommon, usually in small patches in moist Themeda grassland.
Cynodon dactylon (L.) Pers. L.69/191, 69/222. T1-6. A1-3. K1-2.
Common, and extremely variable in size. The large forms occur in
the higher rainfall areas, in grassland and on roadsides. The smaller,
more prostrate forms are very common on lake shores, termite
mounds, roadsides, and disturbed ground generally. It is heavily
grazed, particular the small forms.
Cyrtococcum chaetophorum (Roem. & Schult.) L.69/295. K1.
One record, in shade in mixed forest. Not previously recorded from
East Africa.
Dactyloctenium aegyptium (L.) Beauv. L.525. T1-4. Al-2. K1-2.
Uncommon, on roadsides and disturbed ground.
Digitaria ascendens (H.B.K.) Henr. L69/359. K2.
Only recorded from river sandbanks and sandy roadsides, but is
probably more widespread but overlooked.
Digitaria diagonalis (Nees) Stapf var. L.69/213. T1-3,5,6. Al-2. K2.
A frequent tussock-forming species in the higher rainfall areas, where
it usually occurs with Imperata cylindrica. Var. hirsuta (De Wild.
& Th. Dur.) Troupin occurs in tall grassland at Bitereko (K1).
Digitaria longiflora (Retz.) Pers. L.481. T1,3. Al. K2.
Uncommon. Usually on eroded slopes and tracksides, but also in
heavily grazed dry grassland in K2.
Digitaria melanochila Stapf L.69/338. A3. K1.
Rare, on disturbed ground in tall most grassland.
Digitaria scalarum (Schweinf.) Chiov. L.573. T1-4. Al-3. K1-2.
Uncommon in rather dry grasslands, where it is easily overlooked
as it occurs as isolated shoots among other species. The Digitania
which forms almost pure stands on lacustrine flats in K2 and T5
appears to be this species and not D. herpoclados as was first thought.
A small, almost glabrous form occurs in seasonally moist depressions
in Al and A2.
Dig:taria ternata (A. Rich.) Stapf L.854. T1,3,4. A1-3. K1-2.
Uncommon, usually on moist muddy patches by roads and tracks.
Digitaria velutina (Forsk.) Beauv. L.531. T1-6. A1,2. K1-2.
This is an uncommon weed of tracks and disturbed ground, but is
commoner in the shade of thicket clumps with Setaria kagerensis.


Diplachne caudata K. Schum. L.634. A3. K1-2.
Rare, usually in abandoned water-filled wallows.
Diplachne fusca (L.) Beauv. L.69/214. T1,2.
Rare, on very wet muddy shores of Lake Edward.
Echinochloa colonum (L.) Link. L.69/194. T1,3,4,5. Al-3. Kl-2.
Frequent, in muddy wallows and on seasonally flooded muddy
Echinochloa crus-pavonis (H.B.K.) Schult. L.569. T1,2. A3. K1-2.
Occasional, on seasonally flooded sandbanks by rivers.
Echinochloa pyramidalis (Lam.) Hitchc. & Chase L.69/207 T1-5, Al.
Uncommon, on wet lake shores. Sometimes floating, thus escaping
the heavy grazing which it usually receives.
Eleusine coracana (L.) Gaertn.. L.69/221. Al-3. K1-2.
Casual on roadsides. Millet.
Eleusine indica (L.) Gaertn. L.532. Tl-6. Al-3. K1-2.
Frequent, on trampled and disturbed ground, and on roadsides.
Enteropogon macrostachyus (A. Rich.) Benth. L.530. T2,3,5.
Recorded from three si.es only, all very dry, where it is locally
Eragrostis aethiopica Chiov. L.68/67. T4.
Rare, on sandy tracksides. Ondy one or two plants seen in T3.
Eragrostis cilianensis (All.) Lutati TI-5. Al. K2.
Frequent on roadsides and eroded or overgrazed areas. The com-
monest small short-lived Eragrostis in the Park, formerly named as
E. tremula and E. m.ldbreadii, belong here.
Eragrostis ciliaris (L.) R.Br. L.69/232. T3. K2.
Rare on sandy tracksides. Only one or two plants seen in T3.
Eragrostil exasperata Peter L.57-2. A3. Kl, K2.
Not uncommon in short, heavily grazed, seasonally moist grassland
in the south of the Park.
Eragrostis heteromera Stapf L.67/159. T4,5. K2.
Uncommon. Occurs in seasonally wet hollows and valleys where it
forms a distinctive community with Sporobolus pyramidalis.
Eragrostis macilenta (A. Rich.) Steud. L.69/369. K1.
Weed in shade of abandoned banana gardens.
Eragrostis pilosa (L.) Beauv. L.574. Tl,3,4. Al-3. K1-2.
Frequent, on sandy roadsides.
Eragrostis poaeaodes Beauv. L.651. T2-4. Al-2.
Uncommon, on sandy roadsides, often with preceding species.
Eragrostis racemosa (Thunb. Steud. L.69/229. K2.
Uncommon, in dry Hyparrhenia filipendula / Themeda triandra
grassland on sandy soil.
Eragrostis superba Peyr. T3.
One record only, from the Mweya ferry landing. It seems likely
that it was introduced in mud on a visitor's car, as is not recorded


from Western Uganda by Harker (1961), and the area where it was
found is used for car washing.
Eragrostis tenuifolia (A. Rich.) Steud. L.523. T1-6. Al-3. 1K-2.
Common in overgrazed areas, tracks, and disturbed ground. Usually
in fairly moist sites. It is very similar to Sporobolus pyramnidalis in
its habitat preferences, and is often succeeded by it.
Eriochloa nubica (Steud.) Thell. T4-6. K2.
Uncommon, in moist hollows and seasonally flooded grasslands.
Harpachne schimperi A. Rich. L.522. T1-5. Al. K2.
Frequent, in heavily grazed short grassland, particularly on sandy
Hemarthria natans Stapf. L.848. T1,3. Al. K2.
Rare, on wet lake shores and riverside sandbanks.
Heteropogon contortus (L.) Roem & Schult. L.533. T1,2*,3,4,5* Al- K2-
Abundant, and very variable. The two main forms are a densely
tufted glaucous hairy leaved type, and a glabrous, less glaucous,
loosely tufted type. They occur together in some areas, particularly
in T3. It usually occurs in Themeda/Hyparrhen.a filipendula
grassland, but it is also common on very dry steep slopes in T2
(Craters). It is much grazed when young and accessible.
Hyparrhenia cymbaria (L.) Stapf L.655. A3. K1.
Frequent in moist tall grassland in a few places on the eastern
Hyparrhena diplandra (Hack.) Stapf L.69/374.K1.
Thin stony hill-top soils at Bitereko only.
Hyparrhenia filipendula (Hochst.) Stapf var. pilosa (Hack.) Stapf
L.69/211. T1*,2,3*,4,5,6- A1,2*,3- K1,2*
Abundant in the drier grasslands of the Park. It usually occurs with
Themeda triandra (except in T4.) but because its inflorescence is
relatively inconspicuous it is very easy to underestimate its impor-
Hyparrhenia pilgeriana Hubbard L.69/395. K1.
One record from by a path in tall grassland at Bitereko.
Hyparrhenia rufa (Nees) Stapf L.69/198. Tl-6. A1-3. 1K-2.
Frequent throughout the Park, but much less common than H. filipe-
ndula, and usually in moister situations.
Hyparrhenia variabilis Stapf L.712. T6.
Rare, in moist Imperata cylindrica grassland.
Hyperthelia dissoluta (Steud.) Clayton Hyparrhenia dissoluta.
L.69/196. T1,2*,3,4,5- A1,2- K2*-
Locally common,, particularly round crater rims in T2, and in
Themeda grassland in K2, Occurs both on dry crater slopes and in
shallow valleys in T3; its habit preferences are not fully defined.
Imperata cylindrica (L.) Beauv. T1,2*,3*,4,5*,6- A1,2*,3- K1*,2*-
Abundant throughout the Park, particularly in moister areas. It is


spreading rapidly in T2, and T3, and probably elsewhere. It flowers
copiously after burning, but much of its spread is vegetative,
forming large circular patches. Mature Imperata patches are almost
pure stands. Field (1968) showed that this species was only palatable
for a very short time after burning. Its continued spread into more
palatable grasslands is a severe threat to the Park's grasslands.
Leersia hexandra Sw. L.69/198. T1-5. Al. K1-2.
Common at lake and swamp margins, sometimes forming floating
mats with Hydrocotyle ranuncutoides and Enhydra fluctuans.
Leptaspis cochleata Thwaites L.69/275. K1.
Frequent, in deeply shaded forest undergrowth.
Leptochloa obtusiflora Hochst. L.529. T2-5. A1,2. K2.
Frequent, in the driest parts of the Park, on slopes and round termi-
Melinis minutiflora Beauv. L.726. A3. K1.
A rare scrambling species of forest clearings.
Microchloa kunthii Desv. L.528. T1-5. A1,2. K2.
Common, in drier grasslands. In Themeda grasslands it forms a
scattered understory with Sporobolus stapfianus, and flowers after
burning before the taller species elongate.
Miscanthidium violaceum (K.Schum.) Robyns (A3).
A small swampy crater in the northeast of the Maramagambo
Forest appears, from the air, to contain this species, but it has not
been collected.
Odyssea jaegeri (Pilg.) Robyns & Tournay T2.
Frequent on the saline mudflats of Lake Katwe and Lake Munyan-
hange. Occurs on open mud which is occasionally flooded, below
the zone dominated by Sporobolus raingei and Sporobolus spicatus.
Olyra latifolia L. L.69/42. K1.
Uncommon, in moist forest undergrowth.
Oplismenus hirtellus (L.) Beauv. L.69/217. A3. K1.
Frequent, in rather dry shady places in forest, often with Pseude-
chinolaena polystachya.
Oryza eichingeri Peter L.69/219. A3. K1.
Frequent, in moist places in forests.
Oryza longistaminate A. Chev. & Roehe L.69/319. T4.
One record from a seasonal pool.
Oryza punctata Steud. L.69/220. Kl.
One record, from seasonal pools on clay soil in forest.
Panicum atrosanguineum A. Rich. L.650. T1-4.
Frequent weed of tracksides and disturbed ground. Occurs occasion-
ally on overgrazed ground, and by termitaria.
Panicum brevifolium L. L.68/210. K2.
One record, from moist riverine forest.


Panicum sp.nov.aff. chusqueoides Hack.
(= Verdcourt & Greenway 261). L.69/26. A3.
One record, from moist forest undergrowth.
Panicum cyrtococca.des Napper L.69/342. K1.
One record, in deep shade in forest. A recently described species.
Panicum deustum Thunb. L.719. T2,3,5. A1,3. Kl.
Frequent, on wooded slopes and dry thickets, particularly where
protected from grazing.
Panicum infestum Anderss. L.69/200. T3,5. K1-2.
Occasional, in dry grasslands. Heavily grazed where it occurs.
Panicum maximum Jacq. L.69/209. T1-6. Al-3. 1K-2.
Common. It appears to thrive where the soil is rather moister than
elsewhere, either in hollows, or near water. It also occurs in and
around thicket clumps, and is always heavily grazed when animals
can reach it.
Panicum meyerianum Nees L.67/178. T4,5. K2.
Rare, in seasonal pools, sometimes saline, and on river sandbanks.
Panicum porphyrrhizos Steud. L.67/175. T4.
Rare, in seasonally moist hollows with Eragostis heteromera and
Sporobolus pyramidalis in the eastern part of T4. The plants are
smaller than most in the East African Herbarium.
Panicum repens L. L.69/190. T1-6. Al. 11-2.
Common on sandy lake shores and river banks, where it forms
dense, almost pure stands. It does not appear to be much grazed.
Panicum robynsii A. Camus L.68/130. A3. K1.
Uncommon, in moist forest undergrowth.
Panicum subalbidum Kunth (P.glabrescens). L.69/353. K2.
Very rare, on riverside sandbanks.
Panicum trichocladum K. Schum. L.711. ,T1,6. K2.
Rare, in tall moist vegetation by rivers.
Paspalid'um geminatum (Forsk.) Stapf L.517. T1,3,4. Al.
Frequent, on wet lake shores. Also grows in shallow water, where it
usually remains sterile.
Paspalum auriculatum Presl. L.69/204. T1,2. K1-2.
Uncommon, on moist sandy ground by rivers.
Paspalum commersonii Lam. L.69/224. T5. A1-3. K2.
Uncommon, in moist grassland, often where disturbed.
Paspalum conjugatum Berg. 69/205. T1,3,5. A3. K1,2.
Common, in moist shady places, particularly in forest, where it is
often much grazed by hippopotamuses to form a close sward.
Pennisetum purpureum Schumach. L.69/206. .T1,2. Kl.
Occurs on sandbanks in the Nyamugasani River, but is very much
commoner in abandoned cultivation outside the Park to the east, and
also at Bitereko. The few plants within the Park are always grazed.
Perot:s patens Gand. L.69/228. K2.
Frequent in the Ishasha area on sandy soil by tracks and in grassland.


Phragmites mauritianus Kunth (L.552) T1-4,6.
Rare, at lake edges and on river banks.
Pseudechinolaena polystachya (H.B.K) Stapf L.69/218. A3. K1-2.
Frequent in rather dry places in forest, often with Oplismenus
Rhynchelytrum repens (Willd.) C. E. Hubbard L.69/233. T1,2,5. K2.
Rare weed of sandy or rocky roadsides and disturbed ground.
Saccharum biflorum Forsk. L.69/321. T1.
Banks of the Mpondwe River only.
Setaria caudula Stapf L.69/226. T2. K1.
Rare, in moist places in forests.
Setaria chevalieri Stapf L.69/225. K1. A3.
Moist places in forest. Often with the preceding species.
Setaria homonyma (Steud.) Chiov. (S.aequalis) L.653. T3.
Rare annual of disturbed ground.
Setaria kagerensis Mez L 68/292 T1-3. A1,3. K1-2.
Frequent in shady places, and nearly always present at the margins
of thicket clumps. Formerly misidentified as S.plicatilis.
Setaria sphacelata Stapf & C. E. Hubb. sensu lato
L.69/210. T1-6. A1-3. K1-2.
Frequent in shady places, and nearly always present
at the margins of thicket clumps. Formerly
misidentified as S. plicatilis.
Common in moister grasslands, particularly in hollows where the
taller grasses are excluded by seasonal waterlogging or other factors.
The species is variable in the Park, and one specimen was deter-
mined as S. anceps. Miss C. H. S. Kabuye of the East African
Herbarium says that all are best called S.sphacelata unit the group
is revised.
Setaria trinervia Stapf & C. E. Hubb. L.69/362 T1. K2.
Locally common in open grassland on dry sandy soil in K2. A few
plants on a roadside in T1 had probably been introduced with road
Setar:a verticillata (L.) Beauv. L.652. T2,3.
Rare annual weed of disturbed ground.
Sorghum verticilliflorum (Steud.) Stapf L.69/208. T1,2,3,5. A2. K2.
Uncommon, on roadsides and riverbanks.
Sporobolus consimilis Fresen. (S.robustus) L.69/203. T1-5.
Occasional, by alkaline lake shores and seepages, where it sometimes
forms tall dense pure stands.
Sporobolus festivus A. Rich. L.654. T1-4. K2.
Frequent, in dry, open, heavily grazed grasslands. Annual, or short-
lived perennial, often confused with S.stapfianus, but lacks the
latter's dense basal mat of old hairy leaf sheaths, and has smaller
Sporobolus filipes Napper L.68/308. A3. K1.
Frequent, in trampled forest undergrowth.


Sporobolus pyramidalis P. Beauv. L.67/141. T1*,2*,3*,4,5- A1,2,3- K1,2*-
Abundant tussock-forming perennial grass in areas now, or formerly,
heavily overgrazed. There is strong evidence for the spread of this
species in the last ten years (Lock 1967). It can resist waterlogging,
and is more vigorous in valleys than on slopes or ridges. Although
not a preferred grazing species, it occurs in such quantity that it
makes up a large proportion of the diet of many animals. It is very
variable in size, with large and small forms occurring together in the
same stand. Specimens from the Park have been determined as
S.natalensis and S.jacquemontii by W. D. Clayton at Kew. He
comments that for everyday purposes all specimens could be clas-
sified as S.pyramidalis. Records of S.pellucidus (Brooks 1957) and
S.angustifolius (Heady 1966) refer to this species.
Sporobolus range Pilg, (S.homblei) L.69/201. T1-5. Al.
Frequent, on saline and alkaline lake shores. Also on alkaline patches
of soil by freshwater lakes.
Sporobolus spicatus (Vahl.) Kunth L.69/202. TI-6. Al.
Frequent, on slightly alkaline lake shores. Usually stoloniferous,
forming dense swards when grazed, but can form tussocks when
ungrazed (e.g. L.69/138).
Sporobolus stapfianus Gand. L.527. Tl*,2,3*,4*,5- Al-3- K2*
A common small tussock-forming species, occurring with Microchloa
kunthii in Themeda grassland, and, like it, Nowering after burning
before the dominant species become tall. In areas heavily grazed by
hippopotamuses it forms dense lawn-like stands, again with
M.crochloa. It can be distinguished vegetatively from Micro chloa
by the absence of scattered long hairs on the leaves, and by not
becoming reddish when dry.
Themeda trandra Forsk. L.69/179. T1*,2,3*,4,5*,6 A1,2*,3- KI-2**
The most conspicuous species of the drier, less heavily grazed parts
of the Park, where it covers large areas. As in other parts of Africa,
it appears to thrive on regular burning. It is grazed heavily only
when regrowing after burning.. Flowering appears to reach a mixim-
um in the second season after burning, and almost ceases after four
seasons without a fire. As in other parts of Africa, the leaves vary
in the degree of hairiness and glaucousness.
Tragus berteronianus Schult. L.69/193. T1-5. Al. K2.
Frequent, in dry overgrazed areas, and on bare disturbed ground.
Urochloa panicoides Beauv. L.524. T1-4. Al.
Uncommon annual of disturbed ground and overgrazed areas.
Vossia cuspidata (Roxb.) Griff. T3,4,5,6. Al.
Frequent, floating in the Kazinga Channel and Lake George. Flow-
ers only very rarely.

Uganda Journal 34, 1 (1970) pp, 65-73


By K. A. LYE

The maidenhair-ferns, i.e. the fern genus Adiantum L.,are the favourite
plants of many gardeners. These ferns are all relatively small, but they have
a very elegant and delicate habit. All the indigenous Ugandan species are
of some horticultural value, but the most beautiful are probably A. oatesii,
A. thalictroides, and A. raddianum. At present the Departrment of Crop
Science and Production at Makerere is beginning the cultivation of maiden-
hair-ferns for export to Europe.
In a paper on the genus Adiantum in East Africa Verdcourt (1962)
records only 2 species from Uganda present in the East African H barium,
but 8 species trom East Africa as a whole. Recent collecting and studies in
other herbaria have revealed many more species in Uganda, including the
European Maidenhair-Fern Adiantum capillus-veneris L. This paper includes
9 Ugandan species (with notes on 3 additional East African species not yet
found in Uganda), of which 3 are recorded for the first time in Uganda, viz.
A. oatesii, A. soboliferum, and A. vogelii. Botanists and naturalists should be
encouraged to collect these interesting plants whenever they are found; only
then can a proper understanding of their climatic requirements and habitat
preferences be acquired.
This paper includes all collections present in four major East African
herbaria, viz. Makerere University Herbarium (MHU), Kawanda Agricultural
Research Station Herbarium (KAW), Forest Department Herbarium, Entebbe
(ENT), and East African Herbarium, Nairobi (EA).

Key to the species of Adiantum
1. Frond once pinnate 2
1. Frond 2-4 pinnate 5
24 Rhachis and petioles not winged 1. A. soboliferum
2. Rhachis and petioles winged 3
3. Rhachis glabrous 2. A. ihilippense
3. Rhachis more or less pubescent 4
4. Longest petioles 2-4 mm long 3. A. confine
4. Longest petioles less than 2 mm long 4. A. incisum
5. Fronds pedate palmatee with side-lobes divided) 5. A. oatesii
5. Fronds not pedate 6
6. Fronds regularly bipinnate 6. A. vogelii
6. Fronds at least tripinnate 7
7. Base of most pinnules rounded 7. A. thalictroides
7. Base of all pinnules cuneate 8

8. Most pinnules 5-10 mm long
8. Most pinnules 10-20 mm long

8. A. raddianum
9. A. capillus-veneris


Description of species

1. Adiantum soboliferm Hook. (Fig lc-d). A simply-pinnate fern with short
creeping at the apices. Rhachis blackish, winged for its entire length. Petioles
1-3 mm long with strongly undulate wings. Pinnae alternate, 15-30 mm long
and up to 20 mm broad, becoming successively smaller above, rectangular,
pinnatifid on upper and outer margin; surface glabrous and with prominent
On sandy banks of forest-rivulet, 700 m (2300 ft). Very rare. TORO
(only recorded from the Semliki Forest). Specimen seen: Lye 4275
(MHU,EA). Also one record from Tanzania (cf. Verdcourt 1964).
This species is immediately recognized by its winged stems and petioles.
2. Adiantum philippense L. (Fig. la-b). A simply-pinnate fern with an erect
frond about 30 cm long. Rhachis and peticle dark brown and shiny, glabrous.
Lower petiolules 10-15 mm long. Pinnae alternate, 20-30 mm long and 10-15
mm broad, rectangular to fan-shaped, when raature often breaking from the
peticule; surface glabrous and with prominent veins.
On shayd, steep and rosky river-banks and forest-edges, 900 m (2940ft)
Very rare. ACHOLI (only recorded from Payere Reserve and Thallanga
Forest). Specimens seen: Sangster 393 (MHU, ENT); Thomas 1589 (KAW).
Also recorded from Tanzania.
This species is easily recognized on its long petioles and glabrous stem.
3. Adiantum confine Fee (Fig. 2a-c). A simply-pinnate fern with 20-60 cm
long erect or creeping fronds. Basal scales pale brown, very shiny. Rhachis
dark brown to blackish, densely hairy. Lower petiolules 2-4 mm long, blackish
and densely hairy. Pinnae alternate, 20-25 mm long and about 10 mm
broad, becoming successively smaller above, rectangular below but becoming
obliquely triangular above, dentate on upper and outer margin; base of lower
pinnae truncate, base of upper pinnae cuneate; surface glabrous or with a
few scattered hairs, with less prominent veins than A. incissum.
On sandy roadside bank, 700 m (2300 ft)., Very rare. TORO (only rec-
orded from the Semliki Forests). Specimens seen: Faden 69/1279 A (MHU,
EA); Liebenberg 925 (KAW); Lye 4276 (MHU). Also recorded from the
Usambara mountains in Tanzania.
This species is closely related to A. incisum, but it has longer petiolules
and less pinnatifid pinnae. It also differs in having the lower scales paler
and unicoloured.
4. Adiantum incisum Forsk. (Fig. 3a-d)) A simply-pinnate fern with 10-40
cm long erect or creeping fronds, usually rooting at the apices. Basal scales
linear, up to 5 mm long and 0.2 mm broad, blackish with a narrow pale
brown margin. Rhachis and petiole dark brown, densely hairy, in lower half
often with narrow pale brown scales. Petiolules 0.1-1.0 mm long, dark brown
and hairy. Pinnae alternate, 7-20 mm long and 4-8 mm broad, becoming
successively smaller above, rectangular to obliquely triangular, shortly or deep-
ly pinnatifid on upper and outer margin; base of pinnae cuneate or rarely
rounded; surface densely hairy with white or brown hairs, rarely with a few
scattered hairs only; vein usually very prominent.


Fig. 1. a-b: Adiantum philippense; c-d: A. soboliferum.
From Tardieu-Blot 1953, 1964a, 1964b and Verdcourt 1962,


Fig. 2. a-c: Adiantum confine; d-e: A. oatesii.
From Tardieu-Blot 1964a and Verdcourt 1962.


On shady roadside and river-banks, under boulders or between rocks,
800-1500 m (2600-4900 ft). Not uncommon. BUGISU, BUKEDI, BUNYO-
RO, BUSOGA, TORO, WEST MADI. Specimens seen: Cooper 80 (MHU);
Dawe 839 (KAW); Eggeling 878 (ENT), 4036 (ENT,EA); Faden 69/1252
(MHU,EA); Jex-Blake EA 10098 (EA); Lock 68/270 (MHU,EA); Lye 1723
(MHU,EA), 3188 (MHU, EA), 3189 (MHU), 4030 (MHU, EA); Meinestyhe-
gen 9472 (EA); Osmas'on 1972 (ENT); Sangster 39 (ENT), 296 (MHU);
Thomas 2308 (KAW); Wood 540 (ENT). Also in Kenya and Tanzania.
This species is closely related 'o the much more rare A. confine, but has
almost sessile and usually much more pinnatifid pinnae. Its basal scales
are very dark with a much paler margin.
5. Adiantum oatesii Bak., sometimes caled A. patens Willd. subsp. oatesil
(Bak.) Schelpe (Fig. 2d-e). A pedately branched fern with a short creeping
rhizome, and with only a few fronds at any one time. Petiole 15-30 cm
long, brown and shiny, glabrous. Fronds branching dichotomously, each
branchlet having the habit of an A. confine frond. Petiolules alternate, 1-3
mm long, brown, glabrous. Pinnules alternate, 15-25 mm long and 7-10 mm
wide, subrectanguJar, dentate or shortly pinnatifid on upper and outer
margins; terminal pinnule subtriangular, slightly larger than the pinnules
immediately below; surface glabrous.
On shady banks of pathway, 1500 m (4900 ft). Very rare. BUGISU
(only recorded from near Busano). Specimen seen: Lye 3186 (MHU,EA).
Also recorded from the Buha District in Tanzania (cf. Verdcourt 1964).
The dichotomous main branching but pinnately arranged pinnules makes
beautiful species easy to recognize. A Tanzania species, A. his-
pidulum, has a similar branching but smaller pinnules and hairy stem.
6. Adiantum vogelii Keys. (Fig. 3e-f). A sub-palmately branched fern with
a well developed creeping rhizome. Petiole 5-25 cm long, dark brown, usually
angular and hairy. Branching pinnate with branches and pinnules alternate.
Pinnules usually 5-15 mm long and 5 mm broad, subrectangular, sharply
dentate on upper and outer margin; surface glabrous.
On sandy banks of forest-rivulet, 700 m (2300 ft). Very rare. TORO
(only recorded from the Semliki Forests). Specimen seen: Lye 4277
(MHU,EA). Also recorded from Zanzibar. (cf. Alston 1959).
This is the only species in Uganda with a regular bipinnately branched
frond and is consequently very easy to recognize. A second species
with similar dwarf tree-like habit, viz. A oatesii, has glabrous rhachis
and pedately arranged branches.
7. Adiantum thalictroides Schlechtend. Sometimes called A. poiretii
Wikstr. (Fig. 4a-c), A much-branched fern with creeping rhizome and erect
20-60 cm long frond. Main axis and branches dark brown and glabrous, but
with brown lanceolate scales at the base of the petiole. Branching irregular;
main branching usually subpinnate, smaller branches often dichotomously
arranged. Petiolules usually 2-5 mm long, brown and glabrous. Pinnules
irregular in shape, often triangular or fan-shaped, 5-12 mm long and 5-15
mm broad, frequently broader than long; when mature often breaking from
the petiolule; base of pinnules truncate or rounded, rarely cuneate (but never
in all pennules). Sori curved, in marginal sinuses.
On shady road and river-banks, but also in other sheltered habitats,
1200-2500 m (4000-8200 ft). Not uncommon. ACHOLI, BUGISU, EAST



a e

Fig. 3. a-d: Adiantum incisum; e-f A. vogelii.
From Tardieu-Blot 1964b and Verdcourt 1962.




d e

Fig. 4. a-c: Adiantum thalictroides; d-e: A. raddianum; f-h: A. capillus-
veneris. From Tardieu-Blot 1953, 1964a and Verdcourt 1962.


MENGO, KARAMOJA, KIGEZI, TORO. Specimens seen: Chandler-
Hancock 2546 (KAW); Eggeling 811 (ENT); Hazel 260b (KAW,EA), Jack
5190 (EA); Kertland JUN63 (MHU); Lock 68/140 (MHU,EA), 68/151
(MHU,EA), 68/164 (MHU,EA); Lye 3185 (MHUJ); 4751 (MHU); Osmaston
3132 (ENT); Purseglove 3197 (KAW); Thomas 281 (KAW), 1166 (KAW),
1784 (KAW), 2542 (KAW); 3606 (KAW,EA), Verdcourt 785 (EA), Wilson
569 (EA). Also in Upland Kenya and Tanzania.
This species is similar to A. raddianum and A. capllus-veneris in habit
and branching. It is, however, separated on the basal part of the pinnules,
which is (at least in most pinnules) truncate or rounded.
8. Adiantum raddianum Preslstet (Fig. 4d-e). A much-branched fern with
erect fronds 20-50 cm long. Main axis and branches dark brown and glabrous.
Branching irregular, but main branches often dichotomously arranged.
Petiolules 1-3 mm long, brown and glabrous. Pinnules triangular to fan-
shaped, 5-10 mm long and 4-12 mm broad; base of pinnules cuneate. Sori
curved, in marginal sinuses.
On roadside-banks and ditches, 1140-1200 m (3800-4000 ft). Not
common. EAST MENGO, WEST MENGO. Specimens seen: Chandler
1574 (KAW), 1834 (KAW); Cooper 43 (MHU); Eggeling 17 (MHU); Lye
5185 (MHU). Also recorded from Upland Kenya and Tanzania.
This species is similar to A. capillus-veneris but has much smaller pin-
nules. It is separated from A. thalictroides by having the base of the
pinnules cuneate, never truncate of rounded. A. raddianum is an Ame-
rican species which has become naturalized in various parts of eastern
and southern Africa.
9. Adiantum ctpillus-veneris L. (Fig. 4g-h). A branched fern with erect
fronds 5-30 cm long. Main axis and branches dark brown and glabrous.
Branching irregular, but at least the branchlets dichotomously arranged.
Petiolules 2-8 mm long, brown and g:abrous. Pinnules usually 10-15 mm long
and 6-15 mm broad, usually longer than broad and with crenate margin, but
extreme shade-forms have often 20 mm long lobed pinnules; base of pinnules
luneate. Sori usually less curved than in A. thialictroides, and not in marginal
On sandy roadside and river-banks, sometimes on wet rocks in caves
or near water, 1100-2500 m (3600-8200 ft). Not uncommon above 1500m.
Specimens seen: Dawe 495 (KAW); J. K. G. 150 (MHU); Hazel 260a
(KAW); Lye 808 (MHU), 3187 (MHU,EA); Milburn 143 (MHU); Osmaston
1993 (ENT), 2106 (ENT); Sangster 433 (MHU,EA); Wood 680 (ENT). Also
in Upland Kenya and Tanzania.
This species is similar to A. raddianum but has larger pinnules. It is
distinguished from A. tha.ictroides by having the base of the pinnules
cuneate, never truncate or rounded.


Additional species

The following species have been recorded from Sudan, Kenya or Tanza-
nia, but not from Uganda. Since they may be found in Uganda I have in-
cluded them here.
10. Adiantum reniforme L. is a very distinct species with simple, reniform
frond on a long petiole. As ye: only recorded from Mt. Kulal and Marsabit
in Kenya.
11. Adiantum h 'spidulum Sw. has fronds subpalmately divided into several
pinnate pinnae and a hairy stem. It has the branchng of A. oatesii, but is
easily separated on the hairy stem. It is recorded from Mt. Kilimanjaro and
the Usambara mountains in Tanzania.
12. Adiantum schweinfurthi. Kuhn is similar to A. philippense in having
fronds once-pinnate and a glabrous rhachis, but differs in having pinnules 3-4
times as long as broad. In East Africa only recorded from Sudan.


I would like to thank Mr. R. B. Faden at the East African Herbarium
and Dr. M. E. S. Morrison at Botany Department, Makerere for reading the
manuscript and suggesting many amendments.


Alston, A. H. G., (1959): The ferns and fern-allies of West Tropical Africa.
Tardieu-Blot, M.-L. (1953): Les Pteridophytes de l'Afrique Intertropicale
Francaise. Mem. Inst. Franc. Afr. Noire, No. 28.
Tardieu-Blot, M.-L., (1964a): Flore du Cameroun. 3. Pteridophytes.
Tardieu-Blot, M.-L., (1964b) : Flore du Gabon. B. Pteridophytes.
Verdcourt, B., (1962) : Two interesting plant records from East Afrira.
journ. East Afr. Nat. Hist. Soc. 24 (105): pp. 37-40.
Verdcourt, B., 1964): A not on Adiantum pedatum Peter. Ibid. 24 j108):
p. 73.

Uganda Journal, 34, 1 (1970) pp. 75-78



Early in December, 1968, one of us (W. K.) collected a fragment of a
left mandible a Hippopotamus imaguncula Hopwood (1926) from washed-
out Kaiso clay with ironstones near Kazinga village on the shores of Lake
Edward. A small fragment of a tooth was observed sticking out of the clay
near the place from which the fossil hippopotamus mandible (Plate C in
Krommenhoek, 1969) had been recovered. Despite intensive subsequent search-
ing during 1969 and 1970, nothing more was found. The whole collection
made at Kazinga is kept in the Uganda Museum, Kampala. The age of the
Kaiso series is considered to be early Pleistocene, according to Wayland (1925,
1926), Bishop (1965) and Cooke and Coryndon (in press). The deposits appear
to have a lacustrine or fluviatile origin.

Method followed for a determination
The form of the tooth, which consists of a small fragment of a root and
a comrete, very worn crown leaves no doubt that it is an incisor. The abraded
surface of the crown (Plate 1, F) shows a number of concentric dentine areas,
such as may be encountered in worn incisors of either primates or ungulates.
In view of the extremely short remaining fragment of its root, which makes
a further determination very difficult, it is not easy to ascribe the incisor to
a member of the Primate order or to one of the super-orders Masaxonia or
Paraxonia. Fortunately the structure of mammalian dental enamel may show
appreciable differences when teeth of each of the mentioned categories are
compared (Boyde, 1965).
During the fossilization of the tooth every single microscopical cavity
and interstice was filled with siliceous material. As a result the enamel and
the dentine have acquired a very dense structure in which no water is probably
any longer present. On the surfaces of the fossil the deposition of silica seems
to have continued somewhat longer so that the characters of the original
exterior, when seen weakly enlarged under a binocular microscope, have dis-
appeared to make way for a thin and lumpy siliceous coating. It is therefore
impossible to recognize anything which might be termed an orientation of
enamel prisms on this surface, without its being prepared or treated in some
Fossil material is usually prepared for microscope examination by cutting
off a small slice and by subsequent grinding and polishing this thin section.
The smallness of the fossil incisor fragment did not permit such a procedure
if it had to be kept intact as much as possible. The tooth was therefore
imbedded in automatically solidifying resin; after this the mass of resin and
the tooth were ground on a wet polishing disc until a surface of approximately
3 mm2 of the fossil enamel, more or less at right angles to the direction of the
prisms, had been obtained in a morphologically less important area. This


surface was then polished with the aid of a polishing machine, after which it
was etched with hydrofluoric acid. In order to obtain an object of compar-
ison a recent human tooth was treated in exactly the same manner, only here
the etching was carried out with diluted hydrochloric acid. Although an
attempt had first been made, this latter agent caused no alterations on the
polished surface of the fossil tooth; this formed the reason why hydrofluoric
acid was used later. The degree of dilution of the acid and the time during
which it is allowed to act are very important, but no data were known with
reference to these matters, so that a number of different concentrations were
The etched surfaces were accurately washed with distilled water to pre-
vent a re-crystalization of dissolved material on the treated surfaces (notwith-
standing this precaution such re-crystallizations did occur, but not to a serious
degree). These were subsequently dried with care and an impression was taken
from them in which the Collodiumrreplica method described by Pantke
(1956) was followed.
Collodium celloidinn wool, Hopkin and Williams) is dissolved in a 10%
solution of amylacetate. A drop of this solution is placed on the surface from
which the impression has to be made. Some of the collodium solution has
meanwhile been heated continuously in a Petri dish until so much has evap-
orated that a thin membrane has formed. This may then be lifted from the
dish by adding water to it. A number of narrow strips may now be cut from
this membrane. One of these is selected and it is then closely pressed against
the still liquid drop of collodium-aceta'e solution on the etched surface.
After several hours the strip, which in the meantime has formed a whole with
the now soldified drop of solution, may be pulled off from the surface. Upon
inspection it contains an exact negative impression of the etched surface.
This impression may be cut out of the strip with as little surrounding smooth
areas as possible, after which microscope study, of course with the negative
impression uppermost, can follow with aid of dark field or phase contrast
Apart from several scratches, the etched surface of the tooth shows no
clear structure over a large areas; an observation which holds good also
for a small unpolished external area of the tooth, on which some etching
experiments were previously carried out. On the place where the hydrofluoric
acid appears to have had some effect a number of pits have come into exist-
ence, but there is no really recognizable structure. The dissolved material seems
to have partly re-crystallized in the form of strongly reflecting small crystals,
probably of calciumfluoride (CaF2), which adhere to the surface of the pulled-
off strip.
There were some rare instances in which narrow bandlike regions halfway
between the pitted areas and the unaffected places on the polished surface
could be detected, where the concentration of the acid, or the duration of
its action, or both these causes working together, quite luckily had resulted
in the appearance of some of the original prismatic structure of the enamel.
This clear section of a bandlike region was photographed as it was, instead
of grinding and polishing the whole surface anew and then etching it by using
the correct concentration and time. The already small fossil would otherwise
have been damaged too much.


The photographs (Plate 2, 3-b, 3a-d) show the etched fossil area and
'etched enamel of incisors of Homo, sapiens L., Syncerus caffer (Sparrman,
(HCI-) etched enamel of incisors of Homo sapiens L., Syncerus caffer (Sparr-
man, 1779 and Equus burchelli Gary (1825). It has already been stated in the
case of the human tooth that the procedure has been the same as that followed
the only difference being that HC1 instead of HF was used. The structure
of the prismata in the incisors of a buffallo and a zebra clearly differs from
that in Primates. The palissade-like longitudinal rows of prisms are separated
by inter-row sheets of interprismatic 'substance' (Boyde 1965), forming what
has been. termed the 'pattern 2 structure'. There is a clear difference between
buffalo (smooth prismatic sections) and zebra (wrinkled prismatic sections),
In both kinds of structures, however, the remark made by Boyde (p. 165) is
clearly illustrated that the "inter-row sheets are thickest, and the separation
of the rows of prisms best marked, in the Ungulate and Macropodidae (Mar-
The recent human tooth (Plate. 2, b) shows that the prisms (or again
perhaps the columns of interprismatic substance) have been etched away
somewhat more than their immediate surroundings. The prisms have been
ground through at right angles to their long axis. As long as the microscope
enlargement is not excessive it appears that each of the prisms is hexagonal.
This, and the interrelation of the prisms, repeats the situation encountered
in a honeycomb. A dark line surrounds each prismatic section, and a white
.band ('interprismatic substance') keeps these dark lines apart from each other.
Very strong microscope enlargement shows that the prims do not form a closed
hexagonal area but something more akin to a horseshoe, open at one end.
In. each case a single prism is always surrounded by six. others. One might
say that a 'flowerhead arrangement' (as in a marigold) exists, a'structure which
has been emphasized here by delineating one such group of seven prisms
(Plate. 2, b). This arrangement is known as the 'pattern 3 structure'.
Plate 2 a, is a photograph of the bandlike area in the etched part of the
fossil incisor. Here also two groups of seven prisms have been chosen at random
and their contours delineated with ink. One may observe that exactly the
same flowerhead arrangement is present. The fact that such a great resemb-
lance exists between the arrangements in the fossil tooth and in the recent
human tooth, while very important differences are present when the prismatic
structures in two mutually also differing recent African ungulates are comp-
ared with the fossil, forms the decisive argument for the contention that the
fossil incisor has to be ascribed to a Primate.

The incisor has been figured in its different aspects on Plate 1, a-f.
The following measurements were taken from the tooth before the etching
had taken place:
Height of the enamel in the centre on the anterior
buccall) side 9.6 mm
Mesiodistal width of the marginal enamel edge (the 'incisal
edge,) on the anterior side 11.0 mm
Maximal dimensions of the abraded surface: mesiodistally 10.5 mm
buccolingually (=a.p.) 9.6 mm

Height of the incisor on the distal (?) side at facette:
crown height 7.4 mm
height of remaining root stump 3.3 mm
As has been stated already, the incisor is considerably worn, so that no
clear details are observable from which the former presence or absence of
lingual tubercles, perikymata or raised margins might be deduced. The thick-
ness of the enamel appears to be more or less constant all around. The form
of the nearly flat, slightly concave incisal surface is decidedly asymmetrical,
with a distal (?) buccal angle of nearly 90 degrees on one side and a rounded
mesial (?) buccal angle on the other. The distal (?) side of the incisor can
be recognized by the presence of a clear facette, which can only have been
caused by the continuous wear against it of a lower canine. This provides an
indication for the supposition that this is a lateral maxillary incisor of the
left side. In its form, however, it much resembles a worn (and in that case,
right) central incisor. However, it is hard to explain the presence of the
facette when it is assumed that the tooth is a central incisor; the possibility
is somewhat greater that it is a lateral maxillary one of the left side, of a type
which resembles a central incisor (see for instance Robinson, 1956, p. 26-27,
and Le Gros Clark and Leakey, 1951, p. 39-40).
The broadly spatulate tooth, which has a dark brown colour with some
mahogany coloured spots on the abraded incisal surface where concentric dent-
ine zones can be observed, possesses a slightly swollen basal tubercle (with
a few whitish spots on it) near the cervical border at its lingual side. The
nearly flat buccal enamel face of the incisor has four to five very slight vertical
furrows or rather striae. The small remaining mass of the root, which also
has a dark brown colour, can be seen to consist of a number of concentric layers
around the central canal. These layers, of a varying black to grey and yellow
colour, are perpendiculary crossed by a large number of dentine radii.
The description and additional date given above together form a suffic-
ient base for the contention that the tooth found at Kazinga is an early Pleis-
tocene Pongid maxillary incisor, probably a lateral one of the left side. No
more precise determination can be given.
Bishop, W.W., (1965) Quaternary and geomorphology in the Albertine rift valley,
Uganda. Geological Society of America, Special Paper, 84, pp. 293-321.
Boyd, A., (1965) The structure of developing mammalian dental enamel. Reort of the
Proceedings of an International Symposium, London Hospital Medical College,
April 1964. Bristol, Wright. pp. 163-167.
Le Gros Clark, W. E., and Leakey, L. S. B., (1951) The Miocene Hominoidea of East
Africa. Fossil mammals of Africa, No. 1. British museum (Natural History),
Cooke, H. B. S., and Coryndon, S., Pleistocene mammals from the Kaiso Series and
other related deposits in Uganda. (in press)
Krommerhoek. W., (1969) Mammalian fossils from the Pleistocene of Lake Edward and
Kazinga channel. Uganda J, 33, pp. 79-84,
Pantke, H., (1956) Een Verbeterde replica-techniek. Tijdschrift voor Tandheelkunde,
63, pp. 269-275.
Robinson, J. T. (1956) The dentition of the Australopithecinae. Tranvaal Museum,
Pretoria, Memoir no. 9.
Wayland, E. J., (1925) Petroleum in Uganda. Geological Survey Memoir No. 1,
and others, (1926) The Geology and palaeontology of the Kdiso bonebeds.
Occasional Paper no. 2, Entebbe.

Plate 1 : Several aspects of the fossil incisor from Kazinga:
(a) seen on the root; (d) lingual aspect;
(b) from in front: (e) right lateral aspect;
(c) left lateral aspect (note the facette); (f) crown surface.

Plate 2: Microscopic view of small areas of collodium replica:
(a) of the etched enamel surface of the fossil too h;
(b) ,, ,, ,, ,, ,, a recent human tooth.
In both specimens a flowerlike "cell" of seven hexagons has been
outlined in ink.

- I

Plate 3: Microscopic views (2 different enlargements each) of collodium
replicae of the etched enamel surfaces of incisors of an African
buffalo (a & b) and of a zebra (c & d). Some of the typical columnal
elements have been encircled in ink.


In January of 1970 a Michigan State University archaeological expedi-
tion initiated fieldwork in the vicinity of Kadam Mountain, a 10,067 foot
peak in southern Karamoja. Kadam Mountain is rich in caves and rock-
shelters which have been inhabited since at least Late Stone Age times. Many
of the caves and rock overhangs are still being used by Pokot and Tepeth
peoples today.
Duringthe course of an archaeological survey along the southern slopes
of the mountain, a cave with wall paintings of two white giraffes was dis-
covered at the Pokot locality known as Napeduh hill (Uganda 1: 50,000,
series Y732, sheet at 45/3 edition 3-U.S.D., reference XS936 871). Napeduh,
which is adjacent to the main mass of Kadam, appears as a single conical
hill from a distance. The cave is cut into the bedrock at the top of the hill
about 300 feet above the level of the plain surrounding Kadam. After
a steep ascent, one enters the cave from near the bedrock face and passes
through a natural archway which divides the cave into two chambers. The
paintings are located on the northwest wall of the second chamber, just after
the archway.
The paintings were executed in a position that would be difficult to reach
from the floor of the cave, although an adjacent shelf provides an awkward
bench for a painter. The paintings are in a reasonably good state of preser-
vation, as one can distinguish the horns and ears on the left hand animal.
While this giraffe appears -to be looking at the viewer over its shoulder, the
lankier right hand animal is looking straight ahead.
Although it will never be known for certain what motivated the artist to
to paint the giraffes on the cave wall, it is worthwhile to point out that the
cave is an excellent lookout station for a hunter. One can easily see for a
distance of 15 miles through the main window of the cave, and a view of
'Kadam can be seen when looking through the archway. Sizeable herds of
giraffe and eland roam the plains below the cave, and the locality is widely
used as a hunting ground today. Perhaps the paintings were executed as a
hunter watched and waited for the game animals to appear below.
It is difficult to estimate the antiquity of the paintings. There are no
dated animal paintings of this type described for Uganda, and, as stated
earlier, the caves of Kadam have been used since at least the Late Stone Age.
In terms of more recent history, this cave may have been used by the Pokot,
Tepeth, Karamojong and possibly Didinga refugees from the southern Sudan.
As far as is known, none of these people do rock paintings of animals today.
Lokut, the Pokot tribesman who very kindly guided the party to the Napeduh
paintings, did not know of any information concerning their origin. They
were there as long as he or other elders that he had talked to could remember.
The only modern East African instances of rock painting that I am aware
of are among the Tanzania Masai and in Karamoja. Dr. Alan Jacobs (per-
sonal communication) has observed the Masai casually paint human stick
figures, giraffe, eland and domestic cattle on cement bore-hole reservoirs as
well as shield symbols on rock overhangs. Recently, I saw geometric dung
paintngs:o a. ceremonial rock just north of Kaabong. These paintings consist
primarily of large circles bisected by a central cross. According to a local elder.


the circles represent cattle enclosures and are drawn by travellers passing
by to ask God's blessing.
Unfortunately there is no stratigraphic way to date the Napeduh paint-
ings. However, there is prehistoric cultural debris on the talus below the rock
face not far beneath the cave. Collected artifacts include a number of decor-
ated potsherds, an obsidian crescent and a broken bored stone. The slope
itself is littered with quartz flaking debris. All of the other sites tested or
excavated in the vicinity range from Late Stone Age through Iron Age in
antiquity. Since the present people are probably not responsible for the paint-
ings, it is suggested that they belong to this general period.
Thus far no other rock art has been discovered on Kadam, although a
four row bao game board was seen ground into a large boulder at the locality
of Teklelat near Moruita. Elsewhere in Karamoja geometric paintings have
'been reported at Magisi,1 and a series of spiral engravings has been reported
at Loteteleit.2 However, the Napeduh giraffes appear to be the first instance of
naturalistic animal paintings reported for Uganda.3 The nearest general simi-
larities are with the Mount Elgon animal paintings of Kenya. Elgon can be
clearly seen from the southern slopes of Kadam,. the Elgon cave paintings
are mainly of domestic cattle and are done in both red and white. There is also
an enigmatic antlered animal that has been described as a deer.4 While it is
possible that there is a general relationship between the two groups of paint-
ings, there is a need to find additional paintings associated with stratigraphic
evidence before much more can be said about these sites.5
1.: Posnansky, M. and Cole, G. Recent excavations at Magosi, Uganda:
a preliminary report. (Man, 63, 1963, pp. 104-106.)
2. Morton, W. H. Rock engravings from Loteteleit, Karamoja. (Uganda J.
31, 1967.)
3. See for instance, Posnansky, M. Rock paintings on Lolui island, lake
Victoria (Uganda J., 25, 1961, pp. 105-111) and Posnansky, M. and
Nelson, C. M. Rock paintings and excavations at Nyero, Uganda Azania,
3, 1969. pp. 1-20.)
4. Cole, S. The prehistory of East Africa. London, weidenfeld and Nicolson,
5. I am grateful to the National Science Foundation for providing generous
financial support for the fieldwork and to Dr. A Jacobs, Research
Director of the Cultural Division, University College, Nairobi, for facil-
itating the administration of the project. Special thanks are also due to
the following people who helped in the discovery and description of the
paintings:- Pat Robbins, Joseph Akumoit Longorikero, Lokut Aowei
Lokeriss, Robert Soper and Knut Odner.
N.B. The next issue of the Uganda Journal will report the discovery of another
giraffe painting in Karamoja by Mr. J. Wilson. (Ed.)
N. B. The Editor apologises to Professor Robbins and to the readers of the
Journal that contact prints and negatives for the illustration of this note have
been lost in the process of preparing the material for printing.
The cave is located on the right hand side of a massive rock outcrop
at the top of the hill. Inside the cave is a painting of two giraffes; one 28
cm by 20 cm and the other 28 cm by 24 cm.





The Karimojong have adapted for domestic and 'ornamental usea
number of naturally occurring minerals. This is not widely known outside
of Karamoja so the following list may be of interest.
Clay. Clays are widely used for the manufacture of cooking pots and water
jars, and also for body decoration. For the manufacture of pots and jars,
clay of a fairly uniform consistency is sought and may be obtained from
deposits of clay as such, but also may be obtained from a particular soil
which meets the requirements. Such soils' include Opopwa clay loams. Clay
used for the manufacture of pots and jars is known to the Karimojong by
the general name of elupe. Clays for decoration of the body, such as those
used on ceremonial occasions, may be obtained from the sources listed above
but also from alluvial deposits and calcareous soils. Such clays are known
under the general name of amunyen, but particular varieties distinguished
according to colour include the following; eru~s which is green, lomunyeni-
yang which is yellow aproreit which is red, and lacate which is white. The
red clay aporeit, also finds use as a pigment for head-dresses.
Quartz and Quartzite. Clear quartz is known to the Karimojong as alili
and quartzite as nakwanga, both these minerals are still occasionally fashioned
into lip plugs, usually cylindrical in outline. One individual, for instance,
sports a superb specimen fashioned from the variety amethyst.
Ilmenite. Ilmeni'e sand is fairlyy widespread occurrence in Karamoja
but is particularly abundant around the Labwor hills, and is known to the
Karimojong as ngakup. This is collected, ground up, then mixed with oil
and applied to the hair by women and girls as a dressing which enhances
the lustre.
Asbestos. Asbestos2 of the amphibolite group is found in scattered localities
mainly in eastern Karamoja. The crude fibre being dug up and roughly
ground, it is then mixed with pot clay as a reinforcement and the article
after firing is stronger and more durable than articles made without it. It is
commonly used among the Matheniko and Bokora peoples and is known as
emilimil. Its adaptation by the Karimojong as a reinforcing material is inter-
esting, as it parallels its use in industrial countries, and the question must
arise as to whether Karamojan use pre-dated that of the former. It is also
worth noting that its presence in pottery shards indicates that the pottery
is of Karimojong origin, as its appearance appears to be contemporary with
the Karimojong occupation of the district, since it does not occur in the
pottery of the earlier inhabitants, the Oropom3.
Talc (Soapstone). Talc4 deposits are of fairly widespread occurrence in
eastern Karamoja, the.mineral being known to the Karimojong as elupe, the
same word as clay, and is used in the manufacture of pots. For this purpose
it may be incorporated with the clay as a powder in which case it gives a
characteristic fine-grained, strong bisque, or it may be applied as a slip


imparting a smooth, glossy surface. The use of talc appears to have been
borrowed by the Karimojong from the Oropom people, whose remnant pot-
tery fragments clearly show its incorporation in many instances.
Gneiss and Marble. Outcrops of gneiss are of widespread occurance in
Karamoja and these are a source of flat slabs which find use as grindstone.
Marble4 is employed similarly. The former is kown to the Karimojong as
ekukuse and the latter as atapim.
Iron Ore. This mineral occurs fairly widely in small outcrops usually in
the form of magnetite but heavily ferruginised laterite was alsoiexploited.
These minerals were until the recent past exploited by the Tobur (Labwor)
and Napore people from which a crude iron was smelted and from which
both spears and implements were fashioned. It is very doubtful if they are
utilised at the present day since scrap metal is in plentiful supply and is
widely used as the source material for the manufacture of spears. It is worth
recording that at one time iron, as evidenced by the occurrence of slag heaps,
was smelted over a wide area, the author having found sites both in and to
the east of the Labwor-Napire hills, and in the east of the district from
Mt Zulia in the north, proceeding south-wards through such places as Mt
Morongole; east of Loyoro; near Rupa; at Alale in the Karasuk mountains;
and towards Mt Elgon at Kacheliba, Kanyarus and Greek River. The occur-
rence of these slag heaps in such a north-south pattern is suggestive of the
passage of iron-working people, whose industry stretched from the Sudan
boundary in the north to the Kenya boundary in the south. It is also of int-
erest to record that even up to the present day both the Karimojong and
the Pokot still largely depend on the Tobur people as the manufacturers of
spears, as iron working is still a skill not generally practised outside Labwor.
Chalcedony. The use of chalcedony is at the present day very rare but it
is recognized by the Karimojong as a distinct and useful mineral and is
known by various names according to locality, these include the following;
ekabultu.nut, ekatunatunut, and ekabuttutu. Chalcedony flakes and frag-
ments, being the tools and waste materials of former occupants of the district,
are remarkably common in occurrence. These are used as the occasion war-
rants by individual Karimojong as scrapers for producing a smooth finish
on stools or spear shafts. Occasionally such fragments may be employed as
a blade for cutting string if a knife is not available.


1. Wilson, J. G. The soils of Karamoja district. Agricultural Department
Research Memoir. Series 1, No. 5, 1959.
2. Annual Report, 1953. Geological Survey Department.
3. Wilson, J. G. Unpublished work.
4. Annual Report, 1954. Geological Survey Department.




A note in Uganda J. 32 (1968), pp. 217-219, suggested that Austin may
not have known of Macdonald's secret instructions regarding the destination
of the Juba expedition. Austin's personal diary shows not only that this was
not the case, but also that he was "extremely sorry that Macdonald did not
push thro' at all events to Lado, when he was so close at Tarrangole".' It
seems that Austin did not draft With Macdonald in Uganda until after
Macdonald's 'Uganda in Revolt' had been banned by the Foreign Office,
so it can be assumed that Austin omitted any direct references to the ex-
pedition's true objectives in order to save his book from the same fate as
Macdonald's manuscript.
The manuscript of With Macdonald in Uganda was submitted to the
Foreign Office at the beginning of 1902. Although Macdonald was still hope-
ful that the embargo on 'Uganda in Revolt' would eventually be removed,
he encouraged Austin to seek approval for his version of the same events
which Macdonald had described in his suppressed account. Macdonald did
so because he was anxious that the allegations made against conduct of affairs
in J. W. Gregory's The Foundation of British East Africa towards the end
of 1901 should be rebutted without delay. Macdonald considered Gregory's
charges were libellous but neither Sir Thomas Sanderson, the Permanent
Under-Secretary, nor Lord Lansdowne, the Foreign Secretary, shared this
view. Sanderson did suggest, however, that any reply to Gregory's criticisms
would come better from a member of Macdonald's staff than from Mac-
donald himself.
Macdonald took the hint and discussed with Sanderson the proposal
that Gregory's "partisan" and garbled account "the most disastrous period
in the last decade of (Uganda's) history"2 should be countered by Austin's eye-
witness account of the mutiny operations and the Juba expedition. When
Macdonald left London for India shortly after his discussions with Sanderson,
he was satisfied that the Foreign Office would raise no objections to the
publication of With Macdonald in Uganda. Sanderson took a different view
of the discussions with Macdonald, however, and the section of the book
dealing with the Juba expedition was sent back to the author soon after
Macdonald's departure. When this apparent breach of faith was reported
to Macdonald, he sent Sanderson an urgent protest about his conduct. The
Under-Secretary justified his action on the grounds that he had not been
told that the rejected section had been written by agreement with Macdonald
or at his request, and that it was unfair for Austin to be permitted to publish
in view of the fact that Macdonald's own account of the expedition had been
banned. Sanderson's decision suggests either that the section on the exped-
ition in Austin's manuscript was not mentioned during Macdonald's discus-
sion at the Foreign Office which seems most unlikely, or that Sanderson
thought it politic to suppress Austin's account of the expedition, if it was
at all possible for him to do so. If the second suggestion is correct, Sanderson
evidently did not consider it wise to press the point in the face of Macdonald's


representations, and the embargo on the offending section was removed at
the end of March 1902.
Sanderson's manoeuvres give some indication of the relative importance
which the Foreign Office attached to the two phases of Macdonald's activities
in Uganda between 1897 and 1899 the suppression of the mutiny and the
Juba expedition. 'No objection was raised to what on the face of it was likely
to prove the more tendentious of the two issues: the rebuttal in Appendix B
of With Macdonald in Uganda of Gregory's allegations. This was based on
Macdonald's comments on a catalogue of mis-statements in The Foundations
of British East Africa. Publication of "a narrative account" of the Juba ex-
pedition was viewed in a different light. Despite the author's tactful avoidance
of references to the real objectives of Macdonald's secret mission, as well as
the omission of critical comments in his diary on Martyr's Nile reconnaissance,
Berkeley's duplicity, Cavendish's conduct in Turkana and other controversial
issues, Sanderson undoubtedly tried to prevent Austin from bringing the ex-
pedition to public notice. The inference is that even five years after Salisbury's
confidential instructions to Macdonald had been drafted, the Foreign Office
was anxious to avoid giving grounds for renewed and embarrassing specula-
tions as to their true content.3
J. P. Barber in Uganda J. 28 (1964) pp. 2-4, discusses the misunder-
standing that arose between Berkeley and Macdonald about the route to be
followed when the Juba expedition resumed its march to the north. Austin's
diary seems to confirm Berkeley's contention that Macdonald changed his
plans after he left Kampala, but gives no indication of the reasons fnat prompt-
ed him to lead the main column towards the Nile, and to, leave the Rudolf
investigation to Austin. In addition to day to day accounts of the mutiny
operations and the progress of the Juba expedition, the diary also contains
a number of items of more general interest, including a lengthy, well-written
essay on "Events that led up to the Mutiny", and the regrettable way in
which military officers were inclined to take punitive action in order to earn
distinction and rewards.4


1. Austin's Diary, 10 December 1898. The first volume of the diary up to 2 February
1898 is missing, so it is not known when the secret instructions were divulged
to Austin.
2.. Gregory J. W, The Foundation of British East Africa. p. 236..
3. Correspondence in FO 2/523. 598, 5991 For a fuller account of Macdonald's
service in East Africa and the controversies it aroused see the writer's introduc-
tions to forthcoming reprints of Soldiering and Surveying in-British East Africa
and With Macdonald in Uganda in Dawson's Colonial History Series.
4. Austin's four African diaries covering the railway survey, the Juba expedition,
the Sobat River survey, and the journey from Omdurman. to Mombasa are in
the McMillan Memorial Library, Nairobi, I ani grateful to Major R. E. Austin,
and to Mr. R. G. Opondo of the McMillan Memorial Library, for permission
to consult and quote from the diaries.



..By H. B. THOMAS..

SWriting of the Rev. F. G. Lugard, Lord Lugard's father, Margery Perham
(Lugard the years of adventure, 1956, pp. 6-7) states that his second wife,
Emma Cameron, and his third wife, Mary Jane Howard, who was Lord
Lugard's mother, were both at the time of their marriage working in Madras
uiider the Church Missionary Society. But neither name appears in the
Society's registers; both were in fact members of the Society for Promoting
Female Education in China, India and the East, later commonly referred
to as the Female Education Society or F.E.S. This select little body, its recruits
were mainly gentlewoman, was founded in 1834 and could claim to be the
oldest women's missionary association in England. After over sixty years of
worthy service it was decided in 1899 to dissolve the Society, when the greater
part of its work in the Holy Land, India and the Far East was handed over
to the C.M.S.
The Society's minute books and surviving papers are among the C.M.S.
archives in London. Emma Cameron's name does not emerge, but it is on
record that "Miss Howard" went out to the Ladies' Institution in Madras
in 1847.
The Rev. F. G. Lugard (1808-1900) married three times:
(i) 1834, Grace Price Morgan, divorced about 1838.
(ii) 1848, Emma Cameron, died 1851.
(iii) 1855, Mary Jane Howard (1818-1865). He died at 5 Rosemont Road,
Acton, 31 May, 1900.
Sir John Gray has told me that his father Dr. Arthur Gray's grandmother,
Anne Elizabeth Gray (n6e Howard) was Mary Jane Howard's aunt. He re-
membered her as a formidable old lady who 'scared him stiff in his early child-
hood. Both families were staunch adherents of the evangelical tradition of
the Clapham Sect; as were both Lord Lugard's father and Bishop Dealtry
of Madras from whom Lugard received one of the Christian names. Arthur
Gray who was six years the elder, could recall meeting Fred Lugard in York,
their home, when both were children, but there is no record of their having
met again.



In January 1969, whilst digging a pit latrine for his family, A. Asiimwe
discovered a stone lance head near his home at Murambi in Kajara County,
Ankole. The site of the pit is grid reference 773817 on the 1:50,000 scale
map, sheet 94/1. It lies about 50 yards southeast of what is shown on the
map as a permanent stream called Numumbai (it is in fact, seasonal and
local people give it no name). The site is on the Karagwe-Ankole phyllites,
but is only about 50 yards from the junction with the granite/gneiss which
forms the basement of the Lugalama arena. The site is thus at the edge
of the arena overlooked by Rwabirorwa hill.


The lance head was 17 crs. long, 4.5 cms. broad and 2.5 cms. thick.
The specimen is composed of vein quartz anid has been flaked to provide a
sharp point and two cutting edges. There has been no attempt to polish or
to grind the blade. There is a possible notch at the rear end of the blade. It
has been suggested that the lance is of the Sangoan culture.
The lance head was found at a depth of just over 8 feet in the pit (5 feet
by 3 feet). Further digging brought the pit to 9 feet 6 inches, but no further
artifacts were found. At this depth the pit seemed to have reached solid rock.
The soil in which the pit was dug appeared to consist of a grey-rust coloured
clay, unstratified except for a surface layer of about 2 inches of darker grey
soil. There were very few stones in the soil, except for one small piece of
quartz at depth of 41 feet. Small holes of roughly the size of a fist appeared
to occur irregularly at all depths, but there seemed to be no obvious explan-
ation for them.
A second pit at 771818 revealed nothing, but was 9 feet deep below
which lies the granite gneiss rocks. Again there was no stratification in the
red friable sandy material, and only a few small fragments of quartz were
found. It does not seem therefore as though pieces of quartz suitable for tools
are very common in the area.
The site where the specimen was found is some 45 miles west of the
artifacts found at Nsongezi, and as far as we know, no other stone implements
have been found anywhere else in the area.



The following notes are derived from the Natural- History section news-
let'er which for most of 1969 was compiled by Dr. W. Van Eck.
Research on Uganda rodents:a talk by Professor M. J. Delaney on 18 March.
The food habits of different rodents have been analysed from the stomach
contents of trapped specimens. It was found that Articanthus and Otomys
feed primarily on grasses while others such as the multimammate mouse
Mastomys use both vegetables and insect food material. Two closely related
species of Lophuromys differ quite widely, one species being almost entirely
vegetarian, the other primarily insectivorous. Such food habits allow dif-
ferent species and individuals to occupy simultaneously a relatively small
area. Small and large mammals are often complimentary in the same habitat,
as small mammals, such as rodents, thrive when grasses are long and leave
when large mammals graze the area down only to re-occupy the site as grasses
grow back. Breeding studies of eight species in Queen Elizabeth Park show
that their cycles closely follow the wet and dry seasons, although there are
specific differences. Interesting observations involve the introduced common
rat Rattus rattus which had not spread to forests and grassland but competes
with the native species Mastomys around human habitations, the latter usual-
ly disappearing from the habitat.
Anglbng in Uganda a talk by Mr. B. Cole on 15 July.
Tilapias are ubiquitous in Uganda and can provide good sport. The
Tiger Fish, restricted in Uganda to waters below the Murchison Falls, prov-
ides some of the best sport in the world but is currently rather neglected by
anglers. The Nile Perch, the current favourite, is thriving in the Nile above
Murchison, as well as in Lakes Kioga and Victoria where it was first intro-
duced in 1955-6 and 1960-62 respectively. The musty or earthy flavour of
these fish from Lake Kioga which was complained of in the mid-1960s now
seems to have disappeared, and more recent attempts to find affected fish
in order to track down the sources of this trouble have drawn a complete
blank. The Barbels provide excellent sport, but share with the Tiger Fish
in making poor eating, while the Catfishes, notably the Semutundu, may
give good sport although their culinary virtues are a matter of taste. Turning
to introduced sporting fish, the Black Bass in Kigezi lakes seem to have gone
into an inexplicable decline, while Rainbow Trout in the Sipi on Mt. Elgon
and in the Mubuku on Ruwenzori continue, but are fairly scarce and usually
require a spinner or live bait rather than the customary sporting fly for cap-
ture; the Fisheries Department issues licences on request. Mr. Cole showed
several living Tilapias, tackle and lures including a feather lure, for example,
which anglers seldom if ever use in Uganda but which is highly effective in
taking Nile Perch on Lake Rudolf. The talk was completed with slides of
fish and fishing sites. He made a special request that anglers should report
their record catches, particularly of the less-fished species, as Uganda's records
are in many cases considerably smaller than those of other countries in Africa.
The need for conservation cropping iit National Parks: a talk by Mr. R.
Wheater on 19 August.
Few casual visitors or short-term residents appreciate the ecological problems
which arise from restricting big animals to small ranges in National Parks.


When natural enemies or adversities are eliminated populations of game may
build to intolerable numbers causing widespread restrictions of habitat. If the
purpose of a National Park is to preserve as diverse a series of habitats as
is natural to the area, animal population needs to be controlled. The hippo
cropping served as ah example. Started in .1959, aerial surveys found 10,000
hippo on 55 miles of Nile, nearly twice the acceptable density. (of 25 per
square mile) for habitat preservation. The exercise of killing over 4,000 hippo
was not only an extremely valuable scientific experiment but turned out to
be profitable to the Park (over 200/- sh. per hippo) as meat, skin and ivory
were readily sold to local butchers and regional processors.
Excursion to lake Nabugabo on 29-31 August.
Excursions on Saturday led through the famous Loudetia grasslands to Birinzi
where we explored the shoreline vegetation of Lake Mayanja with its remark-
able floating Sphagnum bogs where the flower stalks of Drosera madagas-
cariensis grew literally waist high while Utricular:a flowered abundantly in
pounded depressions. This lake. and surroundings deserves very detailed ecolo-
gical study and its remoteness is hopefully maintained. A very productive
ornithology trip was had late on Saturday to the now flooded isthmus at
Bukakata where a wide variety of water and marsh birds (among them Goliath
heron) were seen. The sighting of a Madagascar bee-eater among the count-
less little bee-eaters, of a bateleur eagle and of a blue-headed coucal are
noteworthy, as was the kingfisher colony at Busama. There was a useful
exploration trip to the actual Nabugabo settlement on a cliff-like corner of
the lake that bears the same name. Discussion with White Fathers in the
shade of huge native Podocarpus trees dispelled any chance of the lake being
50 feet above Victoria as has been reported. Fears for crocodiles in the lake
were unfounded but the sight of a swimming python is a warning for due
caution. The intention is to repeat this outing several times this and next
year in order to continue the inventory of this unique corner of Uganda.
Population and protein: a new legume project a talk by Mr. C. L. A. Leakey
on 12 November.
Those concerned with public health, and especially child health, have for
some time been aware that its improvement is closely tied to agricultural
development, especially to the acceptance by farmers to grow cheap sources
of high-protein foods. Of these, pulse-crops provide the greatest yield at the
smallest cost, e.g. a tenth the cost per pound of protein compared to that
from animal sources. Research is developing through selection and breeding
improved varieties of pulse crops and cereals. The once held assumption
that higher protein content was coupled to lower yield has been proved
-incorrect; In the Uganda research on bean varieties has been developed with
50% .increased yield and 25%. increase in protein.and methionine content
while maintaining disease resistance. A new soybeann variety yield. 2800 Kg/
.ha or .60.% ..overlocal strains and has 43% more protein and 20% more
.methionine., A special warning was sounded on groundnuts which have only
.a. low. methionine-content. and often contain the dangerous aflatoxin spores.
It is .unfortunate that the new-pulse crop varieties do .not always get to the
common farmer because of lack of multiplication and. distribution, orbecause
of traditional resistance to change.



By A. B. HERRICK and others.
Washington, U.S. Government Printing Office, 1969, 456 p. Price U.S. $3.50

This volume is meant to serve as a compendium of information on Uganda.
The forward states that it is "one of a series of Handbooks prepared by
Foreign Area Studies (FAS) of the American University, designed to be use-
ful to military and other personnel who need a convenient compilation of
basic facts." (p. iii) It is thus directed at non-scholarly readers who are largely
unfamiliar with Uganda and the published literature pertaining to it. Un-
fortunately, the book makes no mention of the training, background, or
professional affiliation of the co-authors, and it is difficult to determine who
actually wrote the various sections of the book. Six co-authors.are listed on
the title page, but the Preface states that each of the following additional
persons wrote one of the chapters: M. Choumenkovitch, A. L. Habermacher,
J. L. McLaughlin, B. C. Skapa, and M. S. Wells. However, the individual
chapters are not signed, and all we know for certain is that the handbook is
a collaborative effort involving at least eleven writers.
The book is divided into 4 sections: Social, Political, Economic, and
National Security, and the sections are further divided into chapters, there
being a total of 26 chapters. No explanation is given as to why these particular
topics were chosen, but it seems that their choice was dictated by editorial
policies which apply to the entire series of area handbooks.
The section entitled :"Social" serves as an introduction to basic facts
about Uganda and contains 12 chapters. on such topics as "Ethnic Group
and Languages", "Social .Structure", "Historical Setting", and "Social Val-
ues". This is probably the least informative of the book's four major sections
since it includes less statistical information than the other sections and tends
to rely on outdated ethnographic sources in an attempt to present general
statements about the peoples of Uganda. The results are less than successful,
for the section contains many factual errors and misinterpretations of the
published sources. Thus, in a discussion on prehistory, we read that "recent
archeological investigations confirm that man had his origins over 20 million
years ago in the apes that later moved onto the East African plains," (p. 31)
Since the handbook does not contain citations to reference works, we do riot
learn the source from which the authors derived this startling bit of infor-
mation. Unfortunately, many further examples of this sort could be cited,
and the reader can only conclude that the handbook was rather hastily com-
piled by authors who had little real expertise in their subject matter. .
The sections dealing with politics, economics, and national security are
more useful than the section on the social realm. These last three sections
include much statistical, information and objective descriptions of political
and economic features. They provide a particularly good coverage of the
post-Independence period and are based on a wide variety of published works.


On the whole this is a useful book in, that it contains information which
is not readily available elsewhere. It presents 34 tables and 8 figures and
these are all drawn from official publications and scholarly works. While it
is certainly true that the errors in the book could be dangerously misleading,
most of these errors can be spotted rather easily by the reader who is already
familiar with Uganda. This sort of reader will find the handbook a conveni-
ent reference work and will value it for the statistical information which it
The handbook also contains 32 pages of bibliographic references in fine
print, and this is perhaps the most valuable portion of the book for those
who want to learn more about Uganda. The bibliography is arranged top-
ically and contains both recent and early publications although its coverage
of recent literature is more complete. Lastly, in contrast to the expensive
editions which seem to be a hallmark of African studies, this handbook is
quite reasonably priced and well worth the price of purchase if one does not
already have access to statistical and bibliographic information on Uganda.

Department of Anthropology. R. T. CURLEY
University of California,-Davis.


by C. H. TAOK
Entebbe Government Printer 1969, 137 p. Uganda price, 22 Shs.

S In compiling this book, which is dedicated to "an increased and better
use of Uganda's many beautiful and useful timbers", Mr. Tack has drawn
not only on his own extensive experience as a forest engineer, and that of
his colleagues in the Forestry Department, but has recorded the final results
of years of research, and has drawn upon the relevant portion of over 26
other publications, with the result that all of the more important basic infor-
mation regarding Uganda timbers is now available in one very compact and
concise volume. Sufficient data is thus readily available concerning timber
in general and over 60 of the most prevalent species, for the use of engineers,
designers, and constructional contractors.
It: should not, however, be assumed that the contents of the book are
of interest only to technicians. Advice is included on the selection of suitable
timbers for various practical and decorative purposes. Furthermore the book
is of considerable commercial interest as an indication of the trend in demand
for local consumption, and, possibly more importantly, the development and
exploitation of Uganda's forests in relation to the production of a valuable
export commodity. These together account for the remunerative use of some
7% of the total rural area of the country, and provide an enormous employ-
ment potential. In this latter context, the book would be of considerable
interest :to younger people involved in the difficult task of selecting an inter-
esting, and promising career.


All aspects of the local timber industry are adequately described includ-
ing milling, treatment, and grading techniques, and a particularly interesting
section deals with the relative importance of the minor natural defects which
are apparent in even the choicest of timber. From this it is possible to decide
whether, for example, various types and sizes of "knot", which produce eye-
pleasing grain patterns, are likely to have any serious effect on the structural
strength of the timber. Generally speaking this is not so, but the more serious
defects which must be guarded against are carefully defined.
The "Export and Grading Rules" are fully quoted, and, apart from
confirming the government's intention that the high esteem which Uganda
timbers enjoy abroad is maintained and enhanced, include a useful glossary
of technical terms associated with the material. The book is very well indexed
ror quick reference, and includes a bibliography which would be invaluable
to those wishing to make a more intensive study of the potential uses of timber,
or the characteristics of the various types, grades, and species.
Reference is not made to the use of waste materials, which is of great
importance when one realises that only about 30% of the volume of grown
timber is converted to constructional use. Particle board is already in product-
ion, and experiments are in progress on the production of wood-wool, and
these materials which, together with the charcoal industry could virtually
eliminate this wastage factor, should not be overlooked by prospective users.
In conclusion, this is a book which should be in the possession of all those
whose work or interest involve the selection and use of timber, such as archit-
ects, engineers, contractors and even the many exponents of the "do it yourself"
cult, and is extremely good value. The author must be congratulated on having
succeeded in producing a concise and easily understandable handbook on what,
particularly to the layman, must appear to be on obscure and complicated

Ministry of Works, Communications and Housing F. WOOLFENDEN


Collins, London, 1969, 238 pp., 24 colour plates, U.K. price Shs. 45-

One glance at this copiously' illustrated book and one wonders why it
had not been done years ago (it had; by Aurillius in Seitz, Vol. XIII, 45 years
ago; which is costly, bulky, scarce and out of date). This is indeed a book
for which so many have been waiting for so long, and emphatically it deserves
wide sales and popularity.
The 21 page introduction 'gives an elementary account of butterfly
morphology and biology, basic advice on the equipment and techniques to
assemble a collection and a short glossary. The meat of the book then follows-
193 pages giving descriptions and distributions of 404 species (not 436 as
stated on the cover). Each species is figured in black and white and 278 of
these plus 12 subspecies are also figured in colour. Food plants are given for