Front Cover
 Title Page
 Table of Contents
 Uganda place names: Some European...
 A question about the Bachwezi
 The Amagasani of the Abakama of...
 Speke and Grant
 The birds of Bwamba -- further...
 Two mammals from Bwamba County,...
 Cattle earth licks with some observations...
 The blood of the Karamojong
 The population of Karamoja
 Tsetse Fly in Ankole: A Hima...
 Index to Volume 17 (1953)
 Back Cover

xml version 1.0 standalone yes
PreviousPageID P220

Digitization of this item is currently in progress.
The Uganda journal
External Link ( Journal issues, 1999-2002 )
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00080855/00034
 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: UF00080855:00034
 Related Items
Related Items: Uganda journal

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Title Page 1
        Title Page 2
    Table of Contents
        Table of Contents
    Uganda place names: Some European eponyms
        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
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
    A question about the Bachwezi
        Page 135
        Page 136
        Page 137
    The Amagasani of the Abakama of Bunyoro
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 144a
        Page 144a
        Page 144a
        Page 145
    Speke and Grant
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
    The birds of Bwamba -- further additions
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
    Two mammals from Bwamba County, Uganda
        Page 166
    Cattle earth licks with some observations in Lango
        Page 167
        Page 168
        Page 168a
        Page 169
        Page 170
        Page 171
        Page 172
    The blood of the Karamojong
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
    The population of Karamoja
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
    Tsetse Fly in Ankole: A Hima song
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
    Index to Volume 17 (1953)
        Page 203
        Page 204
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text

The Uganda Journal


VOLUME 17, No. 2


Uganda Place Names: Some European Eponyms
Lake-flies W. W. MACDONALD
A Question about the Bachwezi DR. ROLAND OLIVER
The Amagasani of the Abakama of Bunyoro DR. K. INGHhAM
The Birds of Bwamba-Further Additions
Two Mammals from Bwamba County, Uganda J. E. HILL
Cattle Earth Licks with some Observations in Lango
The Blood of the Karamojong MARGARET W. STANIER
The Population of Karamoja DR. P. H. GULLIVER
Tsetse Fly in Ankole: A Hima Song - FORD
Early Treaties in Uganda -
Whistled Signals -
Derivations of some Teso Place Names - -
Professor G. D. Hale Carpenter - -
Birds of Eastern and North-Eastern Africa. Vol. I (b% C. W.
Mackworth-Praed and C. H. B. Grant) J. G. WILLIAMS
Some African Milestones (by H. F. Varian) C.sA. C. R. S. PITMAN
The Indigenous Trees of the Uganda Protectorate (by W. J. Eggehng
and I. R. Dale) A. L. GRIFFITH
The Pleistocene Geology and Prehistory of Uga'ida. Part II (by
Professor C. van Riet Lowe) - k. NIARSHALL
Social Science Conference in Uganda - -
Notice to Contributors -
Index to Volume 17 of The Uganda Journal -

Published by
Price Shs. 10 (10s.)











His Excellency Sir Andrew Cohen, K.C.M.G., O.B.E.
President: Vice-President:
). Jameson, O.B.E. Dr. Audrey I. Richards

The President
The Vice-President
The Hon. Secretary
The Hon. Treasurer
The Hon. Librarians
The Hon. Editors
Mr. J. C. Dakin
Mr. C. M. S. Kisosonkole
Hon. Secretary:
Hon Treasurer:
Hon. Librarians:


Mr. S. W. Kulubya, M.B.E.
Prof. E. Lucas
Mr. D. K. Marphatia
Mr. M. A. Maybury
Dr. H. C. Trowell
Dr. K. P. Wachsmann
Prof. Fergus Wilson

Mr. D. G. Thomas
Mrs. B. W. Ray
Mr. and Mrs. C. H. J. Wild

Hon. Editors: Dr. W. H. R. Lumsden
Mr. J. V. Wild
(To replace Mr. J. V. Wild as Hon. Editor in 1954: Dr. A. W. Southall)
Hon. Auditor Hon. Legal Adviser
Mr. J. L. Bray Mr. C. L. Holcom
Corresponding Secretary at Jinja: Mr. T. R. F. Cox
Corresponding Secretary at Mbale: Mr. P. A. Whiteley
Corresponding Secretary at Serere: Mr. J. D. Parsons
Hon. Vice-Presidents:
H.H. Edward Mutesa II, Kabaka Sir Mark Wilson
of Buganda Mr. E. B. Haddon
R. A. Tito Winyi II, C.B.E. Mr. H. B. Thomas, O.B.E.
Sir E. F. Twining, K.C.M.G., M.B.E. Dr. A. W. Williams
Sir John Milner Gray
Past Presidents:

Sir A. R. Cook, C.M.O., O.B.E.
Mr. E. J. Wayland, C.B.E.
Dr. H. H. Hunter, C.B.E., LL.D.
Mr. H. Jowitt, C.M.G.
Sir H. R. Hone,
K.B.E., M.C., K.C.
Mr. J. Sykes, O.B.E.
Mr. N. V. Brasnett
Captain C. R. S. Pitman,
C.B.E., D.S.O., M.C.
Mr. S. W. Kulubya, M.B.E.

1943 -44

Mr. E. A. Temple Perkins
Mr. R. A. Snoxall
Dr. K. A. Davies, C.M.G., O.B.B.
Mr. p. H. E. Hopkins, O.B.E.
Mrs. K. M. Trowell, M.B.E.
Dr. W. J. Eggeling
Dr. G. ap Griffith
Rev. Fr. F. B. Gaffney, w.F.
Dr. A. W. Williams
Dr. J. B. Hutchinson,
C.M.G., F.R.S.

Mrs. K. M. Trowell, M.B.E.

Mr. B. K. Mulyanti

Mr. G. P. Saben

Secretary: Mrs. B. W. Ray

Mr. J. D





Uganda Journal



No. 2


J. V. WILD Hon. Editors
(assisted in England by H. B. THOMAS, O.B.E.)

Published by



Uganda Place Names: Some European Eponyms
Lake-flies W. W. MACDONALD 124
A Question about the Bachwezi DR. ROLAND OLIVER 135
The Amagasani of the Abakama of Bunyoro DR. K. INGHAM 138
Speke and Grant SIR JOHN MILNER GRAY 146
The Birds of Bwamba-Further Additions
Two Mammals from Bwamba County, Uganda J. E. HILL 166
Cattle Earth Licks with some Observations in Lango
The Blood of the Karamojong MARGARET W. STANIER 173
The Population of Karamoja DR. P. H. GULLIVER 178
Tsetse Fly in Ankole: A Hima Song J. FORD 186

Early Treaties in Uganda 189
Whistled Signals 189
Derivations of some Teso Place Names 191

Professor G. D. Hale Carpenter - 193

Birds of Eastern and North-Eastern Africa. Vol. I (by C. W.
Mackworth-Praed and C. H. B. Grant) J. G. WILLIAMS 195
Some African Milestones (by H. F. Varian) CAPT. C. R. S. PITMAN 196
The Indigenous Trees of the Uganda Protectorate (by W. J. Eggeling
and I. R. Dale) A. L. GRIFFITH 197
The Pleistocene Geology and Prehistory of Uganda. Part II (by
Professor C. van Riet Lowe) K. MARSHALL 198

Social Science Conference in Uganda 200
Notice to Contributors 200

Index to Volume 17 of The Uganda Journal -


THE explorers and the pioneers-missionaries, soldiers, and administrators
-who put eastern Africa 'on the map' in the latter half of the nineteenth
century seldom had the time or the knowledge of local languages needed for
enquiries as to the indigenous names of places within the regions which they
were opening up. They required to distinguish the various features on the maps
which were the record of their activities, and in the absence of readily ascer-
tainable local names they frequently employed European proper names, of
persons great, or not so great, who for the most part had some association with
the country or with the map-maker's project. The less worthy motives of
sycophancy or self-advertisement are not commonly discernible. It is worth
remembering however that such names were seldom pronounced with rdclame
at the moment of discovery: frequently they were not devised until the prepara-
tion for publication of the explorer's maps was taken in hand on his return
to civilization.
The question of the use of proper names was exercising the minds of
geographers in England as early as 1875 when, in an article 'On the names of
places in geography' in Proceedings of the Royal Geographical Society, vol. 19
(1875), Lieut.-Col. William Ross King registered a plea against their use. But
it was not until 1901 that a first formal pronouncement discountenancing the
practice was made by the Society in an article 'On the giving of names to newly
discovered places', published in R.G.S. Yearbook and Record, 1901.2 This
guidance, embodied in subsequent editions of the Society's widely circulating
Hints to Travellers, has since gained such general acceptance among carto-
graphers that proper names are now only introduced on maps published in
English where-as for instance in Polar Regions or the interior ranges of British
Columbia-no other names exist.
In Uganda the practice was introduced by Speke, but only in a few instances,
for he was a conscientious recorder of vernacular names: it reached its culmina,
tion in Lieut.-Col. J. R. L. Macdonald's, 1 inch to 10 miles 'Map of Uganda' of
1899-1900 (I.D.W.O. No. 1429), hereafter referred to as 'Macdonald's map'
which, being a compilation from the maps of many earlier explorers, is liberally
1 It came to our notice that each of the authors was separately at work on this subject.
At our suggestion they readily agreed to pool their researches and we are now able to
present the results of their collaboration. They do not claim that their enquiries are
exhaustive, and hope that readers of the Journal may be able to supplement or correct
some of their present identifications and references.-[EDs.]
2 For these references we are indebted to Mr. M. Aurousseau, Secretary of the Permanent
Committee on Geographical Names.

besprinkled with such names. Though many of these have long since been dis-
carded in favour of indigenous place-names, they claim recognition by reason
of their appearance in the literature of this epoch of modern African discovery.
Some of them, however, such as Owen Falls and Lake Salisbury seem likely to
survive permanently.
Other important general maps of our period-that is maps other than those
illustrating some explorer's particular journey-are the map, basically the work
of Ernest Gedge, accompanying E. G. Ravenstein's paper on 'Messrs. Jackson
and Gedge's Journey to Uganda via Masailand' in Proceedings of the Royal
Geographical Society, vol. 13, April 1891, hereafter referred to as 'Gedge's
map'; I.D.W.O. map No. 991 (July 1893) in Intelligence Division War Office
Handbook of British East Africa, 1893; I.D.W.O. Map No. 1012 (Jan. 1894)
prepared to illustrate Sir Gerald Portal's report and published as Blue Book
Africa No. 3 (1894); and the 'Map of Uganda and Unyoro' (1895) by Lieut.
C. F. S. Vandeleur, which appeared in Geographical Journal, vol. 9 (April
1897) and is reproduced in his Campaigning on the Upper Nile and Niger (1898)
hereafter referred to as 'Vandeleur's map'. Macdonald's map summarizes the
cartographical information available regarding Uganda at the end of the nine-
teenth century. A note indicating the widely scattered sources from which he
drew will be found in Thomas and Spencer's A History of Uganda Land and
Surveys (1938).
In certain instances names have found their way on to maps as a group. These
and some other aspects of the nomenclature on Uganda maps are discussed
below as an introduction to the succeeding alphabetical list.' Such are the
names of peaks to the west of the Nile which came into use following Dr.
Junker's 'farthest south' in the Lugbara country in December 1877; the forts
erected in western Uganda by Lugard and his immediate successors; and the
nomenclature of Ruwenzori. Relevant also is the use of the names Nyanza and
Luta Nzige for lakes which we now know as Lakes Victoria, Albert, Edward
and George.
Of merely ephemeral interest are the names-mostly of German functionaries
connected with the organization of his expedition-fastened by Dr. Carl Peters
in 1890 upon various features in eastern Uganda (see Peters' New Light on Dark
Africa (1891)); and of some of the peaks and lakes of the Mufumbiro region
(now for the most part in Belgian territory), which Ewart S. Grogan in 1899,
dubs with intimate family Christian names (see Grogan and Sharp, From the
Cape to Cairo (1900)). These gained no currency beyond the books in which
they originated: their perpetuation would be valueless and apt to lead to con-
fusion: and they are accordingly omitted from our list.
The names of many of those commemorated, such as Speke, Baker, Stanley,
Emin, Lugard and Macdonald, are almost household words in Uganda. Regard-
ing them biographical details can be found in works of reference or in the
voluminous literature of Uganda: and for them no more than a brief note of
identification will be necessary. Thus some less proihinent personages may
receive fuller treatment in our list.
1 For the derivation of some of the vernacular place names of Uganda reference should
be made to Mr. R. A. Snoxall's article 'Some Buganda Place-Names' in Uganda Journal,
vol. 10 (1946).

JUNKER. Towards the end of 1877 the explorer Dr. William Junker, a Russian
national of German extraction, accompanied an Egyptian expedition which
entered what is now the West Nile District from the north-west. At his farthest
south in the Lugbara country, perhaps a few miles north of the present Arua
station, he described and named three mountains in the near distance, Jebels
Gessi, Gordon and Baker. Their identification is discussed in Uganda Journal,
vol. 15 (1951), p. 171 (R. N. Posnett, 'Some Notes on the West Nile Hills and
History'). Junker also saw, but does not name in the text of his book Travels in
Africa, "a mountain range rising up in the far south . beyond them still more
distant mountain peaks".
These last, which were for a while equated with Baker's 'Blue Mountains'
(The Albert N'yanza (1867), vol. ii, chap. III) would be in the highlands around
Mahagi in the Belgian Congo some fifty miles or more distant from Junker's
viewpoint: and to them were in due course appended names-Jebels Emin
Pasha, Chippendall (sic), Junker, Speke, Schweinfurth. The author of these
names is unlikely to be Dr. Junker himself and is more probably the German
cartographer Dr. B. Hassenstein. The first instance of their use seems to be in
the map prepared by Hassenstein (1880) to accompany Part II of Junker's
paper 'Die aegyptischen Aequatorial-Provinzen' in Petermanns Mitteilungen
(1880). All these names reappear in a map, also the work of Hassenstein, in
Junker's Travels in Africa, and they are reproduced en masse in Macdonald's
map-clear evidence that virtually nothing had been learnt of the region since
Junker's visit over twenty years before.

FORTS in western Uganda, 1891-95. Lugard, having marched through
Ankole, reached southern Toro in July 1891. Proceeding to Lake Albert he
deemed it necessary to protect his lines of communication. Accordingly he con-
structed stockades, or forts, at Katwe Salt Lake (Fort George) and at the mouth
of the Sebwe-Mubuku Valley (Fort Edward). Lugard had perforce to approach
Toro via Ankole, since the northern route by way of Singo was obstructed by
hostile Baganda Muhammadans under Nuhu Mbogo.
Having installed Kasagama as Mukama of Toro, and after enlisting Selim
Bey and his Sudanese followers, Lugard planned to protect Kasagama from the
attacks of Kabarega of Bunyoro by a new north-south line of forts along the
eastern frontier of Toro proper from Lake Albert to Lake Edward. Thus Forts
Wavertree (No. 1), Lorne (No. 2), Kivari (No. 3), Ntara (No. 4) and Grant (No.
5) were constructed during October-December 1891. Sudanese garrisons were
installed and Lugard's assistant, W. Fenwick de Winton, remained to supervise
affairs in Toro.
Even before returning to Kampala, Lugard had the idea of wheeling this
north-south line into a west-east line hinging on Fort No. 1: and so soon as
Mbogo's followers submitted in April-May 1892 this re-orientation was under-
taken. By July 1892 Fort Kivari had been replaced by a new Fort No. 3 (Fort
Briggs) and Fort Ntara by a new No. 4 (Fort de Winton): Fort Edward was
abandoned, its garrison moving to Fort Grant, while Fort George at the Salt
Lake was retained, and later occupied by the Chartered Company's armed

Swahilis. De Winton had died (31 March 1892) before these moves could be
effected, but they were carried through by the Sudanese who were without
European supervision until, in February 1893, William Grant was posted to
Sir Gerald Portal, concerned by the indiscipline of these Sudanese isolated in
the forts and not in fact in the employ of the Company or of anyone else, sent
Major Owen and Captain Raymond Portal to enlist and concentrate them in
Buganda. This was completed between April and July 1893. As part of the
reorganization Fort Gerry (now Fort Portal) was constructed to protect
Kasagama on his undertaking to provision it: and Fort Raymond (at Mityana)
to accommodate some of the evacuated Sudanese. For a while Fort George con-
tinued to be occupied but the other Toro forts were abandoned. An indispens-
able authority for this period is Bovill and Askwith, Roddy Owen. A Memoir
It was now necessary to protect western Buganda from Bunyoro, and during
September-October 1893 Fort Lugard and a new Fort Grant were erected near
the Singo-Buwekula frontier. These however became unnecessary when in
January 1894, all southern Bunyoro was cut off from Kabarega's control by
Colonel Colvile's Unyoro expedition. Fort Baranwa (on the Kafu River), Fort
Hoima, Fort Kitanwa and Fort Kibiro (on Lake Albert) then came into existence
as a line across the present Bunyoro. Fort Mruli (a stockaded post on the north
side of the mouth of the Kafu River, not the site of Gordon's old post on the
south bank) was built in April 1895 at the time of Major Cunningham's second
attack on Kabarega: and Fort Masindi was commenced in the following month
by Major Ternan. (Vandeleur, Campaigning on the Upper Nile and Niger, pp.
68, 82). In May 1895 Fort Nakabimba (Roddy) on the Muzizi River was
established by F. G. Foaker to prevent Arab gun-runners from German territory
from reaching Kabarega, (see, 'Imperatrix v. Juma and Urzee', Uganda Journal,
vol. 7 (1939), p. 71). At Foweira and Fajao posts were constructed by Ternan in
October 1896.
Nearly all these fort names appear on Macdonald's map; and, since Sir Harry
Johnston made use of this map, the sites of certain forts, Briggs, de Winton,
Roddy and the original Fort Grant in southern Toro, are recited as boundary
points in the descriptions of the administrative divisions of Toro in the Toro
Agreement of 26 June 1900; Fort Grant also appears in the Ankole Agreement,

RUWENZORI. In June 1891, Dr. Franz Stuhlmann, Emin's lieutenant during
the first stages of Emin's last fatal journey, distinguished from the western slopes
of Ruwenzori four principal groups which he named after prominent German
professors, Kraepelin, Moebius, Semper and Weismann. Sir Harry Johnston
(The Uganda Protectorate, (1902) vol. i, pp. 158-9) strongly objected to
"christening the heights of Ruwenzori for all time with the names of German
1 C. S. Reddie of the Chartered Company was in Unyoro and Usongora for nine months
from June 1892 (see Macdonald, Soldiering and Surveying in British East Africa (1897),
p. 218; also Reddie's route marked on Macdonald's map) and may have had some hand in
the choice of the new sites. No details of his proceedings can be traced, but he was clearly
not in command of the Sudanese.

worthies who have had absolutely no connection whatsoever with Africa or with
Africa's highest mountain". He proposed instead such names as Stanley, Stairs,
Emin, Bagge, Moore, Elliot, and Dr. Stuhlmann himself, all of whom had con-
tributed to the exploration of the mountain.
This proposal clearly influenced the Duke of the Abruzzi who ultimately
sponsored the greater part of the galaxy of European names which now
distinguish the peaks and passes of Ruwenzori. The Duke's expedition of 1906
provided the first clear picture of the inner topography of the range, and on his
return to Europe, following consultation with among others Sir Harry Johnston
and Dr. Stuhlmann, and with the acquiescence of the Royal Geographical
Society, the names of famous African explorers were adopted for the mountain
groups: while, for most of the peaks and passes, the names of members of the
Royal Houses of Italy and Great Britain, of previous explorers of the range and
of members of his own expedition were introduced. The names of Stuhlmann's
German professors were also utilized. Native names were however recorded for
its rivers. The identification and naming of the various features are discussed in
detail in Chapter VII of de Filippi's Ruwenzori (1908).
In a note following G. N. Humphreys' paper 'New Routes on Ruwenzori' in
Geographical Journal, vol. 69 (1927), p. 531, the Editor remarks that "The
nomenclature of the Ruwenzori Group is the last conspicuous example of the
practice which was once commended, but is now happily avoided, of affixing
European names to geographical features in Asia and Africa", and he tenta-
tively suggests the possibility of reconsideration. He recognizes that native
names were probably non-existent; but favours names with "the appearance of
belonging to the region", citing the practice of the Mount Everest Expedition,
which had, on the ground, adopted English descriptive names which were after-
wards translated to Tibetan, and so appeared on their maps.
The names for which the Duke is primarily responsible, noted hereafter as
'(Abruzzi, 1906)', are conveniently tabulated on the map accompanying
Humphreys' later paper 'Ruwenzori: Flights and Further Exploration'
(Geographical Journal, vol. 82 (1933)). It is doubtful if at this late stage a useful
purpose would be served by any extensive alterations in a group of names which
are now widely recognized.

NYANZA. When on 3 August 1858 Speke saw for the first time what we now
know as Lake Victoria, he enquired from the natives its name and they replied
Nyanza. This is the wide-spread Bantu root meaning a large extent of
water. Krapf, who reached the East African coast in 1844 soon heard of N'yassa,
doubtless the present Lake Nyasa, and Erhardt and Rebmann's 'Slug Map' of
1855 shows Nianja Mkuba and Ndogo at the southern end of their vast inland
sea. Livingstone also, in 1856, acquired the -name Lake Nyanja as the source of
the Shire River from the Portuguese, to whom it had long been known (cf.
Nanjaeja in Dr. de Lacerda's Journey to Cazembe in 1798, see Burton, The
Lands of Cazembe (1873)).
Burton and Speke's original sketch map of 1858 shows Speke's discovery as
'Lake Nyanza'; and in Burton's lengthy paper on 'The Lake Regions of Central
Equatorial Africa' (which comprises vol. 29 (1859) of the Journal of the Royal

Geographical Society and is the solid foundation of his more popular two-
volume work The Lake Regions of Central Africa (1866)) Lake Victoria is
referred to only as 'The Nyanza' or 'Ukerewe Lake'. The name 'Victoria
Nyanza' seems to have been introduced by Speke only towards the end of 1859.
Baker (1864) followed this use of 'Nyanza' for 'Lake' with 'The Albert
Nyanza': and the practice was continued by Stanley (1889) with 'Albert Edward
Nyanza'. Stanley had already, in 1876, ventured on 'Alexandra Nyanza'-
apparently Lake Mugesera in Ruanda-Urundi-see Through the Dark Con-
tinent (1878) chap. XVII-but that name was still-born.
But from the first 'Nyanza' was not consistently employed, and it has long
Been entirely discarded as accepted usage on Uganda maps in favour of the
simple designation e.g. 'Lake Victoria'.
LUTA NZIGE. Several lakes in Uganda have borne the local name of Luta
Nzige or some variant spelling (Luganda: 'it kills locusts', presumably because
swarms of locusts are from time to time engulfed). When Speke was with King
Rumanika of Karagwe in 1861 he obtained information regarding "the Little
Luta Nzige, which on the former journey (sc. 1858) I heard was a salt lake,
because salt was found on its shore and in one of its islands". The probability is
that he was then hearing of Lake Albert and the Kibiro salt springs rather than
of the Katwe Salt Lake. Baker in 1864 had recorded M'wootan N'zige before he
reached and named the Albert N'yanza. But the name was also applied at times
to what are now Lakes George and Edward. Stanley, when in Buganda 1875-6,
asked to be guided to Muta Nzige and was led to Lake George. And since Lakes
George and Edward were for some years not clearly distinguished the name was
at times attached to the latter.
A shadowy Lake Nzige near the 30th meridian of longitude had crept into
international maps by 1885 (Hinks, 'Notes on the Technique of Boundary
Delimitation', Geographical Journal, vol. 58 (1921), p. 418).
To General Gordon, who was surveying the Nile in 1876 (see Uganda Journal,
vol. 5 (1937-8), p. 284) is probably due another 'Gitanzege' or 'Gita Nzige' for
the eastern end of Lake Kyoga (see Gordon's map in Journal of R.G.S., vol. 46
(1876), p. 431), and this still had a place in maps accompanying Lugard's Rise
of Our East African Empire (1893).
All three may appear together; for instance in the map prepared by Ravenstein
for Emin Pasha in Central Africa (1888) are Mwutan Nzige (Lake Albert), Muta
Nzige (Lake George) and Gita Nzige (east Lake Kyoga).

ALBERT, LAKE-was named 'The Albert N'yanza' by Samuel Baker when he
discovered it in 1864 "as an imperishable memorial of one loved and mourned
by our gracious Queen", (The Albert N'yanza, vol. ii, p. 90),-the Prince
Consort, Albert of Saxe Coburg and Gotha (1819-61). In the use of the word
nyanza for 'lake' Baker was following the precedent of Victoria Nyanza (see
note NYANZA, p. 105 ante). But 'Lake Albert' was soon current; it became official
usage early in the present century and has now entirely replaced the original

ALBERT NILE-for the reach of the Nile between Lake Albert and Nimule has
come into general use since the end of the nineteenth century; the name does not
appear on Macdonald's map, though Stanley in 1889 (In Darkest Africa, 1890)
speaks of the Albertine Nile. During the Egyptian occupation and after it com-
monly appears as Bahr-el-Jebel. Also from the Prince Consort.
ALBERT PEAK-Ruwenzori, the north-west shoulder of Margherita Peak.
First climbed in 1932 and named by Comte Xavier de Grunne after Albert I,
King of the Belgians (1875-1934).
ALEXANDRA NILE-the Kagera River. Speke (Journal of the Discovery of the
Source of the Nile (1864), p. 262) records that "we reached the Kitangule-
Kagera or river" on 16 January 1862. For no conspicuous reason, H. M. Stanley
in 1875 elected to call this the Alexandra Nile, doubtless after Princess (later
Queen) Alexandra (1844-1925) who had married the Prince of Wales, later King
Edward VII, in 1863. The name has long been obsolete.
ALEXANDRA PEAK-Ruwenzori, a peak of Mount Stanley. (Abruzzi, 1906).
The second highest peak of the range, named in honour of Queen Alexandra,
the consort of Edward VII, the highest having been dedicated to Margherita,
the Queen Dowager of Italy. Alexandra Glacier on the north flank of Mount
Stanley was named at the same time.
ALICE, PORT-Entebbe, named by Sir Gerald Portal in 1893 (The Mission to
Uganda (1894), p. 231) after his wife nde Lady Alice Bertie. Following Sir
Gerald's death she married, in 1897, Major Robert Reyntiens of the Belgian
Artillery. Lady Alice Reyntiens died on 7 May 1950. The name originally
applied to Portal's station on Nsamuzi Hill: and to the bay to the west of Entebbe
Breakwater Pier.
ARTHUR BAY-Lake Victoria, east of Bukaleba peninsula, Busoga. Appears
on Vandeleur's and Macdonald's maps in the place of the inner reaches of the
earlier named Hannington Bay (q.v.). Lieut. L. R. S. Arthur (1864-1903) while
serving with the Sultan of Zanzibar's army was attached to Portal's Mission to
Uganda in 1893. In 1896, as H.B.M. Vice-Consul at Boma, he watched the pro-
ceedings in the Belgian Congo courts against Captain Lothaire for the murder
of Charles Stokes. Later Consul at Dakar and Colonial Secretary, Gold Coast.
BAKER, JEBEL-Possibly Luku Hill in West Nile District (see note JUNKER,
p. 103, ante). Stigand (Equatoria, 1923) always refers to Luku as Jebel Baker.
BAKER, MOUNT-Ruwenzori, one of the central massifs. (Abruzzi, 1906).
BAKER'S CAMP-a modem naming for the remains of the Egyptian fort at
Fatiko, Acholi District, established by Sir Samuel Baker in 1872.
BAKER'S VIEw-a modern appellation for a spot in Bunyoro on the escarp-
ment above Buhuka, which may reasonably be identified with the point from
which Baker and his wife first caught site of Lake Albert on 14 March 1864.
"'We toiled up the opposite slope. I hurried to the summit. The glory of the
prize burst suddenly upon me!" (Baker. The Albert N'yanza, vol. ii, p. 89).
Sir Samuel Baker (1821-93) was in the north of present day Uganda as a
private explorer, 1863-5; and as Governor-General of the Equatorial Nile Basin
for the Khedive of Egypt, 1869-73.
BEATRICE GULF-after Princes Beatrice (1857-1944), fifth daughter of Queen

Victoria, married in 1885 Prince Henry of Battenberg. The northern extension
of Lake George was so named by H. M. Stanley who reached the escarpment
above the lake, just south of the fall to the lake of the Mpanga River, in January
1876. Stanley's "chief captain" Manwa Sera (who had been present at the death
of Livingstone) actually reached the lake with a scouting party (Stanley,
Through the Dark Continent). The name did not persist: see GEORGE, LAKE.
BELL, PORT-Luzira, Kyodondo Country. When, in 1908, during the
Governorship (1905-9) of Sir H. Hesketh Bell (1864-1952), a first metalled road
from Kampala to Lake Victoria was projected, Luzira was selected as the Lake
port in place of the earlier landing at Munyonyo and received the name of Port
BERKELEY BAY-otherwise Sio Bay, Lake Victoria. The name is probably
due to the Railway Survey of 1892. It appears on map I.D.W.O. No. 991 of July
1893 and on the map illustrating Capt. J. W. Pringle's paper 'With the Railway
Survey to Victoria Nyanza' in Geographical Journal, vol. 2, August 1893.
Macdonald records the hospitality which the Railway Survey received on its:
arrival at Mombasa in December 1891 from Mr. E. J. L. Berkeley who was then
acting as the Chartered Company's Administrator.
BERKELEY, FORT-On east bank of the Nile at the Bedden Rapids now in the
Sudan. Established by Colonel Cyril Martyr about end of 1898 (see Uganda
Journal, vol. 16 (1952), p. 44). Soon abandoned in favour of Gondokoro (Uganda
Journal, vol. 16, p. 132) when the name became obsolete.
Sir Ernest Lennox Berkeley (1857-1932) accompanied Portal's Mission to
Uganda in 1893 and was later, 1895-9 Commissioner of Uganda.
BOTrEGO PEAK-Ruwenzori, northern-most peak of Mt. Gessi (Abruzzi,
1906). Vittorio Bottego (1860-97), Italian explorer in Somalia, western Abyssinia
and Lake Rudolf region. Killed by Galla tribesmen at Gobo near the Sudan
BRIGGS, FORT-Kyaka County, Toro: about two miles north of the main
Kampala-Fort Portal road between Kakabara and Matiri Gombololas; was
established July 1892, see note FORTS, p. 104, ante. Lugard (Rise of Our East
African Empire) refers to it only as Fort 3. One reference to it by Roddy Owen
in June 1893 as Fort Rosebery has been traced, but when he abandoned it in
July 1893 he speaks of Fort No. 3. The first use of the name Fort Briggs which
can be traced is in map I.D.W.O. No. 991 of July 1893, the only possible authority
for this portion of which seems to be Lugard.
Some interest attaches to the identification of this name. Some twenty years
ago Lord Lugard was asked its origin and he replied with his customary prompt-
ness "I have no recollection that I had called a fort after my very intimate friend
Dr. Briggs, who had greatly distinguished himself at Metemmeh and elsewhere
in the final dash to save Gordon". Lord Lugard's brother, Major E. J. Lugard
has recently expressed the opinion that Lugard himself must have originated the
name though he had in the course of time forgotten the fact. He remembered
Colonel Briggs, or Broun as he later became, as one of Lord Lugard's oldest and
closest friends from soldiering days in India.
Surgeon Lieut.-Col. William Hamilton Briggs (1854-1937) had joined the
R.A.M.C. in 1875 and retired in 1893. He is 'mentioned' in Lord Wolseley's

final despatch on the 1884-5 Sudan Campaign (see Colvile, History of the Sudan
Campaign (1889), vol. ii). He seems never to have had any direct contact with
East Africa, though he had hoped to join Lugard in Uganda. On his marriage
to Susan Georgiana, elder daughter of the 10th Earl of Dalhousie in 1894 he
changed his name by deed-poll to Broun; he died at Teignmouth 26 December
BROOKES' CORNER-Serere County, Teso, the junction of the roads from
Serere and from Soroti to Ngora. C. G. Brookes, sometime saw-miller in Busoga,
built a house here about 1930. He died in 1941.
CAGNI PEAK-Ruwenzori, "the southern extremity of a buttress which runs
between Mt. Baker and the South Portal Peak", de Filippi, Ruwenzori, p. 255.
(Abruzzi, 1906). Commander Umberto Cagni of the Royal Italian Navy was the
most experienced of the Duke of the Abruzzi's companions on the Ruwenzori
expedition having been with the Duke to Mount St. Elias in 1897 and to the
Arctic in 1899-1900. Cagni was the first to climb the peak on 8 July 1906.
CAVALLI PAss-Ruwenzori, between Mts. Emin and Speke (Abruzzi, 1906).
Dr. Achille Cavalli Molinelli of the Royal Italian Navy was attached to the
Duke of the Abruzzi's party as medical attendant.
CHIPPENDALL, JEBEL-see note JUNKER, p. 103, ante. Lieut.-Col. and Brevet
Colonel W. H. Chippindall, R.E. (1850-1942), served as a Lieutenant with
Gordon in the Sudan 1874-5. In March 1875 he had carried a survey from Dufile
to as far south as Wadelai. Chippindall (which is the correct spelling of his name)
died at Kirkby Lonsdale, Westmorland, 30 September 1942, aged 92.
DE WINToN, FoRT-Kyaka County, Toro, on the road to Nyamugura
gombolola, about 5 miles due north of Kyegegwa. (See note FORTS, p. 103, ante).
Named by Lugard after W. F. de Winton who died about twenty miles to the
east on the Buganda frontier on 31 March 1892. W. Fenwick de Winton, who
was a son of Sir Francis de Winton, Administrator of the Chartered Company
at Mombasa in 1890-1, had come to Uganda with Lugard in 1890.
EDWARD, LAKE-originally named Albert Edward Nyanza by Stanley in 1889
in honour of the Prince of Wales, later King Edward VII (1841-1910). A notice,
dated 10 December 1908 and printed in Uganda Laws 1910, p. 899, declares that
H.M. the King had been pleased to approve the change of name to Lake Edward.
EDWARD PEAK and GLACIER-Ruwenzori, the highest peak of Mt. Baker and
the glacier from its south face. (Abruzzi, 1906.) In honour of King Edward VII.
EDWARD, FORT-near Bugoye, Bunyangabo County, Toro. See note FORTS,
p. 103, ante. Named by Lugard in 1891 after his uncle the Rt. Hon. General Sir
Edward Lugard, G.C.B., P.C. (1810-98): (Rise of Our East African Empire,
vol. ii, p. 187).
EDWIN ARNOLD, MOUNT-Kabuga Hill, Kibale County, Toro. So named by
Stanley who saw it from the heights above Lake George in January 1876. The
name has long been discarded. Edwin (later Sir Edwin) Arnold (1832-1904),
poet and author of The Light of Asia, was editor of the Daily Telegraph when
that paper was sponsoring Stanley's 1874-7 Expedition.
ELENA PEAK and GLACIER-Ruwenzori, a peak of Mt. Stanley and the glacier
on its south-east face. (Abruzzi, 1906.) In honour of King Vittorio Emanuele
III's consort Elena (1872-1952), daughter of King Nicholas I of Montenegro.

EMIN, MOUNT-Ruwenzori, one of the northern massifs. (Abruzzi, 1906.)
EMIN PASHA, JEBEL-see note JUNKER, p. 103 ante. Emin Pasha (1840-92)
joined Gordon in the Sudan in 1876 and was Governor of Equatoria, 1878-89.
While Emin was isolated from Khartoum by the Mahdists Dr. Junker was one of
his few European companions.
FEILDING BAY-Busoga, the small bay running north-west from Jinja towards
Weiraka. The name is doubtless due to Macdonald, see his map; in memory of
Lieutenant Bertram F. P. Feilding, Wiltshire Regiment (1870-97), grandson of
the 7th Earl of Denbigh, killed in the assault on Luba's Fort, 19 October 1897.
He was an officer of the Uganda Rifles, not of Macdonald's expedition.
FLORENCE, PORT-Kisumu, Kenya, named after Florence, wife of Sir George
Whitehouse, K.C.B., Chief Engineer of Uganda Railways Construction, by
whom the site was selected in 1898 as the Lake Victoria terminus for the Uganda
Railway (see Uganda Journal, vol. 12 (1948), p. 5).
FRANZ JOSEF SPITZE-Mount Elgon, the peak on the crater rim northwards
of the Jackson's Summit, now known as Mubiyi (Uganda Survey Dept. map of
Mbale District, 1944, No. A949). Named by Rudolf Kmunke, leader of an
Austrian expedition which climbed Mount Elgon in December 1911, after Franz
Josef I, Emperor of Austria (1830-1916). Kmunke obtained the Emperor's
approval of the name (Quer durch Uganda (1913), p. 43-4), which was current
for only a few years.
FRESHFIELD PASS or CoL-Ruwenzori, between Mt. Baker and Mt. Luigi di
Savoia. (Abruzzi, 1906.) The name Hamugoma, recorded by Humphreys, 1932,
is now tending to take its place. Douglas W. Freshfield (1845-1934), alpinist,
President of the Royal Geographical Society 1914-7. He and A. L. Mumm were
the first expert mountaineers to climb on Ruwenzori, November 1905.
GEDGE, LAKE-the eastern extension of the Lake Salisbury system: now fre-
quently appears on maps as Lake Opeta. Ernest Gedge (died 1 August 1932) was
Jackson's companion on the journey to Uganda 1889-90. The lake seems first
to have been distinguished from Lake Salisbury by the Macdonald Expedition
1898 and makes its appearance on his map. The naming may not be inappro-
priate since Gedge with Jackson may well have been the first Europeans to see
it from the summit of Mount Elgon in February 1890, when they first caught
sight of Lake Salisbury.
GEORGE, FORT-at Katwe Salt Lake-see note FORTS, p. 103 ante. This was
the first of the Toro forts to be built by Lugard, and (apart from a withdrawal
for a few months after the Muhammadan rebellion of July 1893, see Uganda
Journal, vol. 17 (1953), p. 16) was garrisoned consistently until about 1899.
In his (unpublished) journal for 27 July 1891, Lugard describes the finishing
of the fort and says "I have called it Fort George in compliment to my friend
George Wilson, and to Mr. George MacKenzie. I was born in Fort St. George
(Madras) also". Mr. (late Sir) George Sutherland Mackenzie (1844-1910) was
the Chartered Company's Administrator at Mombasa intermittently between
1888 and 1891, and Lugard expresses appreciation of his understanding
support (Rise of Our East African Empire, vol. i, p. 221). George Wilson (1862-
1943) was later sub-Commissioner, Uganda and C.M.G.: for Lugard's tribute
to him, see ibid. vol. ii, p. 541-2.

GEORGE, LAKE-Lugard in 1891 was perhaps the first to recognize this as a
distinct lake separated from Lake Edward by the Kazinga Channel. He recorded
it as Lake Ruisamba and is followed by Macdonald. Sir H. H. Johnston (Uganda
Protectorate, vol. 1, p. 29), speaks of "a curious extension (of Lake Albert
Edward) which, on the whole, had better be called Lake Dweru". This last is
the wide-spread Bantu root for 'white' (cp. Kilimanjaro, Nkokonjeru, Lake
Mweru). Lake Albert has also at times borne this name, e.g. Stanley, 1889 (In
Darkest Africa) 'Ruweru'; and Lugard, 1891 (Rise of Our East African Empire)
'Dwera'. Map I.D.W.O., No. 1012, January 1894, gives both 'Dwera' for Lake
Albert and 'Dweru' for Lake (Albert) Edward. This confusion was resolved by
a notice dated 10 December 1908 (Uganda Laws 1910, p. 899) stating that H.M.
the King had been pleased to approve the name Lake George, in place of Lake
Ruisamba or Dweru. The name doubtless honours the then Prince of Wales,
later King George V (1865-1936).
GERRY, FORT-the original name of Fort Portal, Toro, commemorating Sir
Gerald Portal (1858-94). The fort came into existence about June 1893, see note
FORTS, p. 104, ante, and was so named by Major Owen. Captain Sitwell always
refers to his Toro headquarters (1896-8) as Fort Gerry and it is so in
Macdonald's map. Probably Sir Harry Johnston concluded that the more formal
name of Fort Portal was more fitting, for the Toro Agreement was signed 'at
Fort Portal' on 26 June 1900. The alteration was perhaps made just in time for
J. E. S. Moore who was in Toro earlier in 1900 refers to Fort Jerry! (To the
Mountains of the Moon, 1901).
GESSI, JEBEL-a peak, probably Liru, in the West Nile District, see note
JUNKER, p. 103, ante.
GESSI, MOUNT-the north-eastern massif of Ruwenzori (Abruzzi, 1906).
Romolo Gessi (1831-81), Italian administrator in the service of the Khedive of
Egypt. While under Gordon's command in the Equatorial Province he made
the first circumnavigation of Lake Albert in 1876.
GORDON, JEBEL-a peak, probably Wati, in the West Nile District, see note
JUNKER, p. 103, ante. General C. G. Gordon (1833-85) was Governor of Equatoria
1874-6: Governor General of the Sudan 1877-9. Killed at Khartoum 26 January
GORDON BENNETT, MouNT-Stanley's name for what he first saw of
Ruwenzori when on his way to Lake George in January 1876 (Through the Dark
Continent, chap. XVI) "a faint view of an enormous blue mass afar off, which
we were told was the Great Mountain in the county of Gambaragara. I named
it Gordon Bennett in honour of my American chief". At the time he was at least
fifty miles distant in hazy weather.
In 1889 (In Darkest Africa) Stanley indicates two outliers to the north-east of
Ruwenzori as Mounts Gordon Bennett and Mackinnon (q.v., post) and
attributes to each on his map a height of 15,000 feet. Possibly he caught sight of
some of the north-eastern summits such as Karangora or Musandama. But
further speculation is profitless. Lugard (1891) demonstrated their non-existence,
and the names are not accordingly on Macdonald's map.
James Gordon Bennett (1841-1918) was proprietor of the New York Herald
and one of the sponsors of Stanley's 1874-7 Expedition.

GRANT BAY-Kyagwe, Mengo District: named by Stanley in 1875 after
Speke's companion in the expedition for the discovery of the Source of the Nile,
1860-3 (Through the Dark Continent, chap. XII).
Colonel James Augustus Grant, C.B., C.S.I. (1827-92).
GRANT GLACIER-Ruwenzori, north face of Mt. Speke (Abruzzi, 1906). Also
after Colonel J. A. Grant.
GRANT, FORT-see note FORTS, p. 103-4, ante. There were two Forts Grant. The
first, also known as No. 5, was built and named by Lugard in December 1891.
The site is in the extreme south of Kibale County, Toro, close to Kicheche
Gombolola. It was soon abandoned and in May 1893 Major Owen noted that
the site was overgrown and could not be found. A second Fort Grant was built
by Owen in September 1893. This is on Bulu Hill, Buwekula, Mubende District,
about six miles east of Bukumi R.C. Mission.
It cannot be doubted that the name here commemorates William Grant,
Lugard's loyal companion from the Coast and to Kavalli's. Of his work and
comradeship Lugard speaks in the highest terms (Rise of Our East African
Empire, vol. ii, p. 521). He became a Provincial Commissioner in Uganda and
C.M.G.: retired 1904 and died 28 October 1919 aged 57.
GRAUER ROcK-Ruwenzori, a prominent rocky peak on the ridge of Mt.
Baker above the Mubuku (Moore) Glacier. First climbed by the Austrian
mountaineer R. C. Grauer in company with the missionaries H. E. Maddox
and H. W. Tegart in January 1906, shortly before the Duke of the Abruzzi's
expedition. Grauer named it King Edward Peak (after King Edward VII) but
the Duke adopted the name Grauer Rock, thus avoiding confusion with Edward
Peak, the highest summit of Mt. Baker.
GREEK RIVER-north-east of Mt. Elgon, Sebei County, Mbale District. The
name is applied to the crossing of the Kelim River by the main track from Mbale
to Karamoja, rather than to the river itself. It was unknown to Macdonald, and
seems to have come into use colloquially in the early years of this century.
Possibly the name is a reference to some Greek trader temporarily settled there.
W. D. M. 'Karamoja' Bell (in a recent unpublished letter), recalls a wandering
Greek from Maji (Abyssinia) who, dressed in skins, was handed over to him by
the Dodoth, and was passed on to S. Ormsby, the Collector at Mbale.
The name has more recently come into use for a river in Usuku County, Teso:
here the suggestion is that it may be derived from Ateso, korik, 'to drag across'.
HANNINGTON BAY-In south Busoga east of Buvuma Island. The name
appears on Gedge's map of 1891, and it is thus probably due to Jackson or Gedge
who had passed nearby early in 1890. It is, however, omitted from Vandeleur's
and Macdonald's maps, which show only 'Arthur Bay' (q.v.). Hannington Bay
appears on modern maps.
Bishop James Hannington (1847-85) was first Bishop of Eastern Equatorial
Africa. The headquarters of Chief Luba near which he was murdered were
situated a few miles north of the upper reaches of the Bay.
HARDY FALLS-Mt. Elgon: just below Bulago camp on a small tributary of
the Simu River. This name was current on sketch maps for some years c. 1910-
20. No one of this name has been identified as being in Uganda at this period,
other than W. McHardy of the Colonial Audit, and later of the Kenya and

Uganda Railway Service. But there is no information associating him with this
part of the country.
HOBART, LAKE-on the Koki-Ankole boundary: the local name Kachira has
now prevailed. Lieut. C. V. C. Hobart, Grenadier Guards, who had reached the
lake in 1897, himself so named it (Geographical Journal, vol. xii (1898), p. 189).
He was an officer of the Uganda Rifles, and took a distinguished part in the
battle at Kabwoko on 20 July 1897, and in the subsequent operations of
February/March 1898 in Kabula.
Sir Vere Hobart, 2nd Bart., D.S.O., O.B.E. (1870-1949), died in the Isle of
Wight 29 November 1949.
HOBLEY FALLS-Mount Elgon. On the Namagyo River, a tributary of the
Namatala rising from Nkokonjeru, near Mbale. C. W. Hobley, C.M.G. (1867-
1947), joined the I.B.E.A. Company in 1890 and retired as Provincial Com-
missioner of the Kenya service in 1921. When stationed at Mumias in 1896 he
made a circuit of Mount Elgon, "a beautiful waterfall is a feature of the western
bluff of Busano and it was named after me by a subsequent explorer" (From
Chartered Company to Crown Colony, (1929), p. 93). Hobley's Busano would
be Nkokonjeru. The "subsequent explorer" is presumably Macdonald, on whose
map the name appears.
HUNTER ROCK-an isolated rock in Lake Victoria, south of Buvuma Island
and west of Kisumu; administratively in Buvuma County, Mengo District. It
appears on Admiralty Chart, No. 3252. 'Victoria Nyanza (Northern Portion),
surveyed by Comm. B. Whitehouse, R.N. and Mr. C. S. Hunter, Uganda Rail-
way Surveys 1900-01': published 1902. Doubtless named after the latter who
(1899-1904) assisted Whitehouse with the hydrographical survey of Lake
IBRAHIM, LAKE-now Lake Kyoga. Discovered in 1874 by Lieut.-Col. C.
Chailld-Long, of the Egyptian Army, an American on General Gordon's staff;
see Central Africa: Naked Truths of Naked People, (1876), p. 164. In his later
book (My Life in Four Continents (1912), vol. i, p. 103) Long says "On August
11th I discovered a vast lake which the Khedive (sc. Ismail Pasha) subsequently
named 'Lake Ibrahim' in honour of his father the soldier Ibrahim Pasha".
Ibrahim Pasha (1789-1848) was the eldest son of Muhammad Ali; he took a
leading part in the conquest of the Sudan in 1821.
IOLANDA PEAK-Ruwenzori, the highest peak of Mt. Gessi. (Abruzzi, 1906.)
Princess Iolanda (b. 1901) eldest child of King Vittorio Emanuele III of Italy;
married 1923 Conte Giorgio Carlo Calvi, conte di Bergolo.
ISOLT, LAKE-Lake Wamala, Mengo District. Lugard, marching towards
Unyoro, records that on 15 April 1891 "I discovered the little 'Lake Isoldt'"
(Rise of Our East African Empire, vol. ii, p. 121). His map gives 'Isolt': but the
vernacular 'Wamala' continued to be used by missionaries. Macdonald's map
prints 'L. Isolt (Wamala)': and the latter name has prevailed since its adoption
by Sir Harry Johnston (see map in the Uganda Protectorate, 1902). According
to Baganda legend the lake was formed by the leaking water-skin of the god
Wamala (Roscoe, The Baganda, p. 314).
The Rev. A. B. Fisher, who established the C.M.S. station at Mityana over-
looking Lake Wamala in April 1893 less than twelve months after Lugard's

departure from Uganda, confirms that at that time Lugard's naming of the Lake
was known, and was assumed to refer to Iseult, "more fair than foam or dawn
was white", the story of whose mutual love for Tristram of Lyonesse has held
the imagination of western Europe for the past thousand years.
Lugard had come to Africa three years earlier after a very disturbing love
affair. Just before he first saw Lake Wamala a long-delayed caravan with letters
from England had reached him. As he searched his mail-bag a realization that
the past was irrevocably gliding out of his life must have come upon him. The
shimmering lake was lit up with all the magic of a sunlit African dawn in the
rainy season. Deeply impressed by its beauty his thoughts were perhaps a reflec-
tion of Algernon Swinburne's lines
As the sun's boat of gold and fire began
To sail the sea of heaven unsailed of man,
And the soft waves of sacred air to break
Round the prow launched into the morning's lake,
They saw the sign of their sea travel done
It was in this mood of high romance that Lake Isolt was named. It seems unlikely
that Lugard ever met Swinburne, who was however an intimate of Sir Richard
JACKSON'S FALLS-Mt. Elgon, on Sisi River above its junction with the Simu
and just south of Buligenyi. The name appears on Macdonald's map, but has
passed out of use.
JACKSON'S SUMMIT-Mt. Elgon; is the rocky knob, prominent from the west,
which was climbed by Jackson and Gedge in 1890: and is the usual goal of
modern climbers, though it is not, by a few feet, the highest point of the crater
rim (see Uganda Journal, vol. 2 (1934-5), p. 250). The name was first applied,
with Sir Frederick's permission, by the Austrian explorer Kmunke who climbed
Mt. Elgon in December 1911 (c.p. FRANZ JOSEF SPITZE ante). This name, which
is an exceptionally appropriate one, is now well established.
F. J. (later Sir Frederick) Jackson (1860-1929) led the Jackson-Gedge expedi-
tion to Uganda in 1889-90. He was Governor of Uganda, 1911-17.
JACK'S MOUNT-this is not a serious contribution to geography! In August
1875 Stanley (Through the Dark Continent, chap. XII) refers to the death of his
dog 'Jack' while he was camped at the foot of Ziba Hill, Kyagwe, which he forth-
with christened 'Jack's Mount'. It lies to the east of the Mill Hill Fathers' Nkoko
Njeru mission station.
JOHNSTON PEAK-Ruwenzori, peak at the southern end of Mt. Speke
(Abruzzi, 1906).
Sir H. H. Johnston (1858-1927), Special Commissioner for Uganda Protector-
ate 1899-1901. He ascended the Mubuku Valley in 1900.
JUNKER, JEBEL-See note JUNKER, p. 103, ante.
Dr. Wilhelm Junker (1840-92) was in the Sudan 1876-86. He spent some years
among the Azande and Monbuttu peoples and was with Emin Pasha during
much of the Mahdist period.
KIRKPATRICK, LAKE-Teso-Karamoja. It was mapped by Macdonald in 1898

as the source of the Nakadokoi River, but can only have been a temporary flood-
ing near the present Lorengikipi dam, and it no longer exists as a lake.
Captain R. T. Kirkpatrick, D.S.O., Leinster Regiment (1865-98) was a mem-
ber of Macdonald's expedition. He was surveying from the Nakwai hills when
he was attacked by natives and killed on 26 November 1898. Macdonald arrived
on the scene three days later and must then have seen what he took to be a lake.
KRAEPELIN PEAK-Ruwenzori, the northern peak of Mt. Emin (Abruzzi 1906).
Name first introduced by Stuhlmann in 1891: see note RUWENZORI, p. 104,
Dr. Emil Kraepelin (1856-1926). German psychiatrist. He had many works
translated into English.
LAWSON, MOUNT-a hill in Ankole seen by Stanley on his journey to Lake
George in January 1876. Since he refers to it (Through the Dark Continent,
chap. XVI) as near the source of the Rushango River it may reasonably be
identified with Nyabushozi which is the outstanding peak in this direction.
Mr. Edward Levy Lawson, later 1st Lord Burnham, (1833-1916) and his
father Mr. J. M. Levy were the proprietors of the Daily Telegraph which was
sponsoring Stanley's 1874-7 expedition.
LORNE, FORT-Mwenge County, Toro, about four miles south of Butiti; see
note FORTS, p. 103, ante. Erected by Lugard as Fort No. 2, "Nyakorongo, which
Mr. De Winton subsequently named Fort Lorne" (Rise of Our East African
Empire, vol. ii, p. 244).
The Marquis of Lorne (1845-1914) was at the time a director of the Imperial
British East Africa Company. He was the eldest son of the 8th Duke of Argyle
whom he succeeded in 1900 as 9th Duke; in 1871 he married Princess Louise,
4th daughter of Queen Victoria. Young de Winton had been in Canada with his
father Sir Francis, who was Secretary to the Marquis of Lorne when Governor-
General of Canada, 1878-83.
LUGARD, FORT-near Ntwetwe, Singo, Mengo District: see note FORTS, p. 104,
ante. Refer to Roddy Owen-a Memoir "October 11th (1893) wrote Foster
names of forts, Buru-Fort Grant; Lemba's-Fort Lugard: Makwenda's-Fort
Captain F. D. Lugard, later Lord Lugard (1858-1945), commanded the
Chartered Company's force in Uganda 1890-2.
LUIGI DI SAVOIA, MOUNT-Ruwenzori, the most southerly of the snow-
covered massifs. It had been the intention of the Duke of the Abruzzi to call this
Mt. Thomson after the explorer Joseph Thomson (1858-95). (It is misleading in
an East African context that de Filippi, Ruwenzori, p. 199, refers to him as
"J. Thomson, whose work in Nigeria is well known". Thomson had led a suc-
cessful expedition to Sokoto in 1885 but his claim to recognition in East Africa
rests upon his journey through Masailand in 1883.) But the Duke bowed to the
insistence of the Royal Geographical Society that it should be named after him-
self. However the Duke retained Thomson's name for the glacier on the south
face of Mt. Luigi di Savoia.
Luigi Amedeo, a prince of the house of Savoy-Aosta, Duke of the Abruzzi
(1873-1933) was a grandson of King Vittorio Emanuele I of Italy. He devoted his
life to geographical work and died near Mogadishu, Somalia, 18 March 1933.

MACDONALD BAY-Lake Victoria, Busoga. The name is perhaps due to
Vandeleur on whose map of 1895 it appears.
Major-General Sir J. R. L. Macdonald, K.C.I.E., C.B., R.E. (1862-1927). He
had commanded the Railway Survey to Uganda in 1892 and was in Uganda
until 1894. His later expedition, during which he was engaged in the suppression
of the Sudanese Mutiny, was from 1897 to 1899.
MACKINNON PEAK-makes its first and almost only appearance in Stanley's
In Darkest Africa (1890), vol. ii, p. 293-4: cf. GORDON BENNETT, MOUNT,
ante. He had presumably mapped it from the north-west around Kavalli's early
in 1889; and refers to Ruwenzori being "further defended by isolated outlying
forts like Gordon Bennett Mountain 14,000 to 15,000 feet high, and the
Mackinnon Mountain of similar height". Stanley's map places it much in the
position of Fort Portal. To lend verisimilitude to this description he provides an
impressionistic birds-eye view with two isolated snow-capped cones standing
well to the north-east of Ruwenzori. The name ceased to have any significance
when two years later Lugard traversed the area and proved that these peaks did
not exist.
Sir William Mackinnon, Bart. (1823-93) was founder of the British India
Steam Navigation Company and Chairman of the Imperial British East Africa
Company. He was Chairman of the Emin Pasha Relief Committee which had
organized Stanley's 1887-9 expedition.
MARGHERITA PEAK-Ruwenzori, the highest peak of Mount Stanley and of
the range (Abruzzi, 1906). Named in honour of the Queen Dowager of Italy,
Margherita, Princess of Savoy-Genoa (1851-1926), consort of King Umberto I
of Italy and mother of King Vittorio Emanuele III.
MOEBIUS PEAK-Ruwenzori, a peak of Mt. Stanley. Name first introduced
by Stuhlmann in 1891; see note RUWENZORI, p. 104, ante.
August Ferdinand Moebius (1790-1868), eminent German astronomer and
MOLONY BAY-Lake Victoria, Kyagwe, now Gobero Bay north of Koja
Peninsula; see Macdonald's map.
Captain C. A. Molony, Royal Artillery, died of wounds on 25 February 1898,
following the attack on the Sudanese mutineers in Kabagambe Boma on the
previous day. He was in the Uganda Protectorate service.
MOORE PEAK-Ruwenzori, north end of Mt. Baker and at the head of the
Mubuku or Moore Glacier (Abruzzi, 1906).
J. E. S. Moore, zoologist, was leader of a scientific expedition for the study
of the fauna of the great equatorial lakes of Africa. In March 1900 he ascended
the Mubuku Valley and was the first to confirm the existence of genuine glaciers
on Ruwenzori. No information regarding his career after 1908 can be traced.
MURCHISON BAY-Lake Victoria, the reach leading up to Port Bell. Named
Murchison Creek by Speke who first saw it in April 1862 (Journal, p. 391).
Stanley in 1875 speaks of Murchison Bay, which thereafter obtains.
MURCHISON FALLS-on River Nile, before entry into Lake Albert. Named
by Samuel Baker who discovered them in 1864. (Not to be confused with the
Murchison Rapids on River Shire, south of Lake Nyasa.)
Sir Roderick Impey Murchison (1792-1871) distinguished geologist and

geographer. For many years (with a few intervals) between 1843 and 1871 was
President of the Royal Geographical Society, giving unflagging encouragement
to African exploration.
NAPOLEON GULF-Lake Victoria, the arm leading up to Jinja and the Ripon
Falls. Named Napoleon Channel by Speke in 1862, "in token of respect to the
French Geographical Society for the honour they had done me, just before leav-
ing England, in presenting me with their gold medal for the discovery of the
Victoria N'yanza" (Journal, p. 469). Napoleon Channel was current for the next
30 years, but Lugard (1893) has Napoleon 'Gulf' and this last has since become
Napoleon III, Emperor of the French (1808-73).
NORMAN, LAKE-appears on Macdonald's map in a position approximating
to the arm from the Mpologoma River running northwards towards Agu and
Lake Salisbury. The name never attained currency. Doubtless named by
Macdonald after his brother Lieutenant Norman A. Macdonald, 14th Sikhs,
who was killed before Luba's Fort on 11 December 1897.
ORMSBY FALLS-Mount Elgon, on Dirigana River, a tributary of the Upper
Siroko. The name, which came into colloquial use after the opening up of Mbale
station in 1904, seems never to have found its way on to printed maps, and is no
longer in use.
S. Ormsby was one of the first administrative officers in the Bugishu county.
He was an agent of Boustead, Ridley & Co. when Major Ternan met him at
Nandi Fort on 2 May 1897 and engaged him temporarily as transport officer.
Was with Ternan in Buddu in the operations against Mwanga and proved most
useful. Appointed to Uganda Civil Service in 1902; he died as Collector Mbale
21 January 1909.
OSBORN FALLS-Mount Elgon, on Kaptokoi River, Sebei County, near
present Kapchorwa camp. The name, now obsolete, is on Macdonald's map.
Lieutenant P. B. Osborn, Oxfordshire Light Infantry, of the Uganda Service
was attached to Macdonald during the Mutiny operations. Wounded in the
assault on Kabagambe Boma on 24 February 1898, he was unable to accompany
Macdonald's expedition towards the Sudan.
OWEN FALLS-River Nile, near Jinja below Ripon Falls. The name seems
to make its first appearance on Macdonald's map and may reasonably be
attributed to him.
Major E. Roderic Owen, Lancashire Fusiliers (1856-96), came to Uganda with
Portal's mission in 1893. He reorganized the Sudanese garrisons in Toro and
served in the Unyoro Campaign of 1894 and was closely associated with
Macdonald. As a gentleman rider he had won the Grand National in 1892. He
died of cholera at Ambigol Wells, Sudan, during the Dongola Expedition,
1896. Owen only crossed the Nile near Jinja twice-on his way to and from
Uganda-on both occasions in great haste. Any evidence that he himself dis-
covered or even visited the falls is lacking. Nevertheless there is an incidental
appositeness in this name: for Owen was, for a short time in 1884, aide-de-camp
to the very Marquess of Ripon then Viceroy of India, who is commemorated by
the adjacent Ripon Falls.
PEARSON CHANNEL-the northern entrance of the Nile below the Murchison

Falls into Lake Albert. The channel was so named by Commander G. D.
Brown of the Railway Marine after Pete Pearson, big-game hunter, who had
discovered it. Pearson had hunted in the Lado Enclave during the Belgian
occupation: and with the (then) Prince of Wales in 1928 (see Chalmers, Sport
and Travel in East Africa, (1934)). He died in 1929 and his memorial stands by
the Masindi-Butiaba road on the scarp over looking Lake Albert.
PILKINGTON BAY-Lake Victoria, north coast of Buvuma Island. The name
first appears on Macdonald's map.
G. L. Pilkington (1865-97) of the Church Missionary Society was killed
11 December 1897 while acting as interpreter under Macdonald's command at
the siege of Luba's Fort.
POLISH LAKEs-Ruwenzori, Bujuku Valley. Named by Baron J. Golcz,
leader of a party of Polish mountaineers, who in June 1943, made the first
ascent of Alexandra Peak direct from the Bujuku Valley by way of the east
PORTAL PEAKS-Ruwenzori, to the north of the lower Bujuku valley. When
in 1900 Sir Harry Johnston saw them from the Mubuku valley he was not aware
of the existence of the Bujuku valley and included under this name Kinyangoma
to the south of the latter river. The name which is now confined to the line of
peaks north of the Bujuku, is due to Johnston, "a long serrated ridge .... they
serve as awful portals of the snow .... I venture to call them the Portal Peaks
and the name will serve to commemorate the services of Captain Raymond
Portal .... "(Uganda Protectorate, vol. i, p. 163-4).
Captain Raymond Melville Portal, Loyal North Lanes Regiment (1856-93),
elder brother of Sir Gerald Portal. He accompanied the Mission to Uganda in
1893 and died at Kampala 27 May 1893.
PRINGLE BAY-Lake Victoria, Busoga coast, opposite Sigulu Island. Name
probably due to Vandeleur, on whose 1895 map it appears.
Captain J. W. Pringle, R.E. was with Macdonald on the Railway Survey to
Uganda 1891-2. As Sir John Pringle, C.B. (1863-1938) he became Chief Inspect-
ing Officer of Railways in England.
RAYMOND, FORT-near Mityana, Singo; see note FORTS, p. 104 ante. Named
after Captain Raymond Portal-refer to PORTAL PEAKS, ante.
RIPON FALLS-River Nile, near Jinja. Discovered by Speke 28 July 1862, and
named by him after the 3rd Earl de Gray (1827-1909) who became 2nd Earl
(1859) and 1st Marquess (1871) of Ripon. He was President of the Royal
Geographical Society for the one session, 1859-60, when "my expedition was
got up" (Speke, Journal, p. 469).
RHODES BAY-the Kitabi bay immediately east of the Entebbe peninsula. The
name appears on Macdonald's map, and doubtless commemorated Colonel
Frank Rhodes (1850-1905), elder brother of Cecil Rhodes. He had accompanied
Portal's mission to Uganda in 1893, when he would have become known to
Macdonald. Name long disused.
ROCATTI PASS-Ruwenzori, between Mts. Emin and Gessi. (Abruzzi, 1906.)
Dr. Alessandro Rocatti was responsible for the scientific collections of the
Duke's expedition.

RODDY, FoRT-Kyaka County, Toro: on south bank of Muzizi River, about
two miles north of Nyamugura Gombolola; see note FORTS, p. 104, ante. Built
early in 1895 by F. J. Foaker of the Uganda Administration and then known as
Fort Nakabimba. Vandeleur (Campaigning on the Upper Nile and Niger, p. 87)
writing of June 1895 says "Mr. Foaker had been sent to establish the
Nakabimba post on the Msisi River, with a view to intercepting the large Arab
caravans which were known to be bringing in arms and ammunition". (cp.
Uganda Journal, vol. 7 (1939), p. 71). A small Sudanese garrison seems to have
been maintained until 1897. Captain Claude Sitwell, then in Toro, notes in his
diary, 9 April 1897 "Got mail from Kampala, everyone off to Nandi (i.e. Major
Ternan's Kamasia Expedition): things at Nakabimba to come to Toro": and on
1 January 1899 "Reached Nakabimba. Whole place overgrown." It is not clear
who devised the name Fort Roddy, which does not seem to appear before
Macdonald's map. For Major Roddy Owen (who is clearly commemorated),
refer to OWEN FALLS above. But Owen seems never to have been concerned
with the site, though he and Villiers may have crossed the Muzizi at this point
when they marched from Hoima to Toro direct in February 1894.
ROSEBERY CHANNEL-Lake Victoria, between the Kyagwe mainland and the
islands from Dwaji to Buvuma. Macdonald with Captain Williams had passed
through this channel in January 1893 on their way to attack the Buvuma
(Soldiering and Surveying, chap. x) but does not then seem to have named it: and
it makes its appearance in Vandeleur's map of 1895.
The 5th Earl of Rosebery (1847-1929) was Foreign Secretary in 1886 and
1892-4: and Prime Minister 1894-5.
ROSEBERY, FORT-see FORT BRIGGS ante. This name attained no currency in
Uganda. It was, perhaps, learnt that a'Fort Rosebery' had already been founded
in what is now Northern Rhodesia. (See 'Report by Commissioner Johnston ...
British Central Africa,' Blue Book Africa, No. 6 (1894), p. 28, "Fort Rose-
bery. .The site of this station was well-chosen by Mr. (sc. later Sir Alfred)
Sharpe in 1892.")
SALISBURY, LAKE-Teso District. Was first seen by Jackson and Gedge from
Mount Elgon in February 1890. It appears on Gedge's map of 1891 (cf. Uganda
Journal, vol. 12 (1948), p. 130).
The 3rd Marquess of Salisbury (1830-1903). He entered Parliament in 1853
and resigned as Prime Minister in 1902. For the period 1887 to 1892 he was
successively Foreign Secretary and Prime Minister.
SALISBURY CHANNEL-Lake Victoria between Bunjako and the Sese Islands.
This seems to have first been named by Vandeleur in his map of 1895: also after
Lord Salisbury who was again Prime Minister and Foreign Secretary, 1895-
1900, continuing as Prime Minister until 1902.
SCHWEINFURTH, JEBEL-See note JUNKER, p. 103, ante.
Dr. Georg Schweinfurth (1836-1925), German naturalist and explorer. He was
particularly concerned with the southern Sudan from 1868-71 when he reached
the Zande Country and the Welle River.
SCOTT BAY-Lake Victoria, between Zinga and Bunjako Islands. Appears
on Macdonald's map. Doubtless commemorates William Neil Scott of Camp-
beltown, N.B., Engineer of the Uganda Government's steam launch Victoria,

who was murdered, with Thruston and N. A. Wilson, at Luba's Fort 19 October
1897, aged 35 years.
SCOTT ELLIOT PAss-Ruwenzori between Mts. Stanley and Baker. (Abruzzi,
G. F. Scott Elliot, later Professor of Botany at Glasgow (1861-1934). In
1894-5, though ill-equipped, he showed great tenacity in making five explorations
towards the summits of Ruwenzori. His greatest height was reached from the
Butagu valley on the western side of the mountain.
SELLA PEAK-Ruwenzori, a peak of Mt. Luigi di Savoia (Abruzzi, 1906).
Vittorio Sella was photographer of the Duke's Ruwenzori expedition. He
accompanied the Duke to Mt. St. Elias in 1897 and to the Karakorums in 1909.
SEMPER PEAK-Ruwenzori, a peak of Mt. Baker. Name first introduced by
Stuhlmann in 1891; see note RUWENZORI, p. 104, ante.
Gottfried Semper (1803-79), Professor of Architecture at Dresden 1834-49;
he taught in London 1851-53.
SMITH BAY-west side of Buvuma Island: now on maps as Tome Bay. Named
by Macdonald in 1893 (see maps in Soldiering and Surveying). Major Eric Smith
was with Captain Williams and Macdonald in the expedition against the Buvuma
in January 1893.
Lieut.-Col. A. F. Eric Smith, C.B. (1857-1942), 1st Life Guards, served with
the Chartered Company, 1890-5; he founded Fort Smith near Nairobi.
SOMERSET NILE-In The Albert N'yanza, vol. ii, p. 291, Samuel Baker says
"On the maps given to me by Captain Speke (i.e. when they met at Gondokoro
in February 1863) he has marked the Victoria Nile below the Ripon Falls as the
Somerset river ... I also adhere to the name Somerset river for that portion of
the Nile between the Victoria and Albert Lakes; this must be understood as
Speke's Victoria Nile source". Nevertheless Speke's Journal and map refer only
to the White Nile or River.
It seems therefore that Baker may be responsible for the original dissemina-
tion of the name Somerset Nile or river (though he reverts to 'Victoria Nile' in
his later Ismailia). The name recurs from time to time on early maps, e.g. that
illustrating the journey of Emin Effendi in 1877 in Petermanns Mitteilungen,
(1878), where it applies both to the Nile below the Ripon Falls and to the reach
below the Murchison Falls. Gordon's map of 1876 gives Somerset River for the
reach below the Ripon Falls only. Dr. Junker, who spent the greater part of
1885 at the Egyptian post of Foda, between the Karuma and Murchison Falls
speaks always of the Somerset Nile. There are however no grounds for thinking
that the originators of the name intended to confine it to this (Foweira-Lake
Albert) section of the river-a use which has of late been coming into currency,
and is proving convenient.
For general usage the name has long been replaced by 'Victoria Nile'. Almost
certainly Speke gave the name Somerset in honour of his home county, cp. his
name Jordans Nullah for Mwanza Gulf, after Jordans, the seat of the Speke
family in Somerset.
SPEKE, JEBEL-see note JUNKER, p. 103, ante. It may be the hill Erusi.
SPEKE, MOUNT-Ruwenzori, one of the central massifs (Abruzzi, 1906).
John Hanning Speke (1827-64) reached the south end of Lake Victoria in

1858. His later expedition (1860-3) with J. A. Grant led them through Uganda
and resulted in the discovery of the Ripon Falls as the source of the White Nile.
STAIRS PEAK-Ruwenzori, a peak of Mt. Luigi di Savoia (Abruzzi, 1906).
Lieut. W. G. Stairs, R.E. (1863-92) was one of Stanley's officers in the 1887-9
Emin Pasha Relief Expedition. In June 1889 during the return journey via the'
west of Ruwenzori he climbed to an estimated height of 10,677 feet beneath
snow peaks which, the Duke of the Abruzzi suggests, were Mount Emin. Later
he commanded a Belgian expedition to Katanga and died at Chinde 9 June
STANLEY, MOUNT-Ruwenzori, the central and highest massif (Abruzzi,
STANLEY, FORT-near Lutoboka, Bugala Island, Sese. Established in 1901
by James Martin at the instance of Sir Harry Johnston (see Jackson, Early Days
in East Africa, p. 72).
Sir Henry Morton Stanley, G.C.B. (1841-1904), famous African traveller. His
journeys to Uganda were in 1874-7: and with the Emin Pasha Relief Expedition,
THRUSTON BAY-Lake Victoria, Busoga coast, to the north of Bukaleba
peninsula. Appears in Macdonald's map.
THRUSTON, FORT-Bukaleba, Busoga. This is not Luba's (Lubwa's) Fort in
which Major Thruston was murdered. It lies a few hundred yards north-west of
Luba's Fort, and was built shortly after the 1897-8 Mutiny. The name appears
on the maps accompanying Sir Harry Johnston's Report of 10 July 1901 in Blue
Book, Africa No. 7 (1901).
Brevet Major A. B. Thruston, Oxfordshire Light Infantry (1865-97) of the
Uganda Rifles. Murdered in Luba's Fort 19 October 1897.
TRACY FALLS-Mount Elgon, see Macdonald's map; now known as Sipi Falls.
Major the Hon. Algernon Charles Henry Hanbury-Tracy (1871-1915) brother
of the 5th and father of the 6th Lord Sudeley. He was a member of Macdonald's
expedition, 1897-9, and commanded the base camp at Save (Sebei) during the
period, August-December 1898, of Macdonald's journey northwards to Latuka.
UMBERTO PEAK-Ruwenzori, highest peak of Mt. Emin (Abruzzi, 1906).
Since the highest peak of Mt. Gessi is named after Iolanda, the eldest child of
King Vittorio Emanuele III, it may be assumed that this is in honour of her
brother Umberto, (b. 1904) the then infant Crown Prince. Upon his father's
abdication and until the declaration of the Italian Republic he was for a few
weeks (9 May-13 June 1946) King Umberto II of Italy.
VICTORIA, LAKE-see note NYANZA, p. 105, ante. It is a common inference that
the lake was named Victoria Nyanza by Speke upon its discovery in August
1858, but the name does not seem to occur until over a year later. In Black-
wood's Magazine of October/November 1859 is an article by Speke where, in a
footnote to his reference to the discovery of 'The Nyanza', he states "This
magnificent sheet of water I have ventured to name Victoria after our gracious
Sovereign"; this note is reproduced in Speke's What led to the Discovery of the
Source of the Nile (1864), p. 307. It was soon referred to alternatively as the
Victoria Lake or Lake Victoria. This latter name gained predominance in the

early years of the present century; and is now accepted as the official designation.
The change over in official use was taking place at about the time of the transfer
of Uganda from Foreign Office to Colonial Office control in 1905 and may have
been associated therewith.
VICTORIA NILE--f. SOMERSET NILE above. The name was appropriated to
the Nile between the Ripon Falls and Lake Albert, probably by the Royal
Geographical Society, after the return of Speke and Grant in 1863. The name
does not appear in Speke's Journal and to both Speke and Grant at the time it
was 'The White Nile'.
VICTORIA, PORT--n the east coast of Berkeley (Sio) Bay, now in Kenya. The
originally proposed lake terminus of the Uganda Railway, it was located by
Lieutenant P. G. Twining, R.E. of Macdonald's 1891-2 Railway Survey.
'Victoria Station' appears on the map accompanying Captain Pringle's paper
'With the Railway Survey to Victoria Nyanza' in Geographical Journal, vol. 2
(August 1893).
H.M. Queen Victoria (1819-1901).
VILLIERS, LAKE-now Kijanebalola, Koki. Lieutenant Villiers, Royal Horse
Guards, accompanied Portal's Mission to Uganda in 1893. In August/Septem-
ber 1893 he cleared Koki of rebel Muhammadans and was then the first Euro-
pean to see the lake which Macdonald named after him (Soldiering and Survey-
ing, p. 291).
Lieut.-Colonel. C. H. Villiers, C.V.O. (1863-1947). In 1895 he was involved
in the Jameson Raid. Later for many years he was a member of the Corps of
VITTORIO EMANUELE PEAK-Ruwenzori, the highest peak of Mt. Speke
(Abruzzi, 1906).
King Vittorio Emanuele III of Italy (1869-28 December 1947). He had
abdicated in 1946.
WAVERTREE, FORT-Burahya County, Toro, about 5 miles north of Rusekere.
Established by Lugard October 1891 as Fort No. 1; see note FORTS, p. 103, ante,
"'at Ireti's capital, Nsororo, I selected a site on which to build our first fort"
(Rise of Our East African Empire, vol. ii, p. 234). Named after Sir John Kirk's
home 'Wavertree' at Sevenoaks, Kent, where he settled after his retirement in
1887; and died in 1922. A tea estate now surrounds the remains of the fort.
WILSON BAY-Lake Victoria, between Zinga and Busi Islands. Appears on
Macdonald's map. Doubtless commemorates Norman Alexander Wilson of
Castle Park, Huntly, Scotland. Assistant in the Uganda Government Service,
murdered at Luba's Fort 19 October 1897, aged 25 years (cf. ScoTT BAY).
WILLIAMS BAY-on the west side of Bugala Island, Sese. The name has not
been traced prior to its appearance on Macdonald's map.
Captain W. H. Williams, was in Uganda 1890-3 as Lugard's second-in-com-
mand and successor. In February-March 1892 he commanded a boat expedition
to Sese Islands and ascertained for the first time that it was not a single large
triangular island but consisted of one main and an archipelago of smaller islands
,(Macdonald, Soldiering and Surveying, p. 90).
Colonel W. H. Williams, C.M.G., R.A. (1857-1938).

WEISMANN PEAKI-Ruwenzori, the highest peak of Mt. Luigi di Savoia. Name
first introduced by Stuhlmann in 1891; see note RUWENZORI, p. 103, ante.
August Weismann (1834-1914). Professor of Zoology at Freiburg-im-Breisgau,
well known as opponent of the Lamarckian theory of transmission of acquired
WOLLASTON PEAK--Ruwenzori, peak at northern end of Mt. Baker (Abruzzi,
Dr. A. F. R. Wollaston (1875-1930), Fellow and Tutor of King's College, Cam-
bridge. He had, as a member of the British Museum (1905-6) Expedition,
reached one of the summits of Mt. Baker in April 1906, shortly before the Duke
of the Abruzzi's arrival.

AKE-flies need no introduction to those who live near the shores of Lake
Victoria, especially in the vicinity of Entebbe. It is not uncommon to see
large dense swarms of these flies being swept in from the lake and their invasion
of houses and offices makes them a well-known pest. The flies belong to the
family Chironomidae and to another group of non-biting flies, the Chaoborinae,
which are closely related to mosquitoes; both groups belong to the order
Diptera. Like other holometabolous insects, four stages in the life-history can
be defined, egg, larva, pupa and adult; most of the life of the lake-fly is passed
in the larval stage in the water. Much work has been done on the flies and their
larvae in Europe, chiefly by Prof. A. Thienemann and Prof. F. Lenz in Germany,
Dr. M. Goetghebuer in Belgium and the late Dr. F. W. Edwards in Britain. As
a result a groundwork has now been established for the study of their biology
and ecology. However, the many different species are not easy to identify and
very little has been attempted in the tropics so that nothing is known of the
number of kinds of flies occurring nor of the effect of the relatively high
temperatures on their life-cycles. There are, however, some descriptions of adult
flies available. These are mainly the work of Dr. Goetghebuer who named and
described flies sent to him from the Belgian Congo; nothing is known of their
larval stages.
The importance of chironomids and chaoborids in fresh-water biology and
fisheries lies in their abundance and their widespread distribution. Choronomids
in particular are abundant not only in numbers but also in species. There are
some 2,000 different kinds known, but in addition there are many others await-
ing description and classification. This is especially so for African forms. The
Chaoborinae, on the other hand, are not abundant in species but vast numbers
often occur in lakes and ponds. The remarks in the following pages apply chiefly
to the chironomids except where Chaoborus is specially mentioned.
Larvae may be found in almost any lake, pond or river and very often they
are present in astonishingly large numbers. In Lake Victoria there are commonly
three or four thousand chironomid and chaoborid larvae per square metre of
lake bottom but even higher figures have been recorded from some other lakes.
In Lake Mendota (U.S.A.) 18,000-30,000 Chaoborus larvae per square metre
have been recorded; in another lake an American worker estimated a maximum
population of 71,000 per square metre. The occurrence of dense swarms com-
posed of millions of flies is therefore not surprising; nor are such large swarms
confined to the tropics. Sharp, in 1899, described a swarm in New Zealand as
having been three-quarters of a mile long and twenty feet high. Records of flies
constituting a plague by invading houses situated near a lake are not uncommon;
a case is on record of a fire-brigade being called because a swarm of flies around
a church tower was mistaken for smoke. Probably David Livingstone was the
first to mention that some tribes on the shores of Lake Nyasa collect the flies
by swinging baskets amongst a swarm and then compress the catch into 'kungu'

cake. Edwards examined specimens of 'kungu' and named the fly Chaoborus
edulis Edw. A similar or closely-related species has been collected at Entebbe.
One reason for the formation of such large swarms lies in the relatively
simple feeding-requirements of the larvae. In general, Chaoborus larvae feed
on plankton, i.e. the countless microscopic plants and animals which float and
drift in the water. Chironomid larvae, however, have a much wider range of
feeding habits. A large number, living on the bottom of lakes and ponds, feed
on decaying organic matter and are in some ways comparable to earthworms
in that they burrow through the mud and aerate it; rhythmic undulations of the
body bring aerated water down into the burrows. Many larvae are browsers and
grazers on plants but others are less active and construct tubes in which they
live, either stretching from the tube and feeding on the detritus, algae etc.
around the tube opening, or drawing food particles into the tube by water
currents. Filter-feeding mechanisms have been evolved in some groups and in
these cases a net to capture the food organisms is constructed near the mouth
of the larval tube. In the case of stream-dwellers the current brings down the
food particles but in the case of larvae living in ponds and lakes the larvae
themselves cause currents to pass through the tube by undulatory movements
of the body. The process of net-spinning and feeding has been studied by
Dr. B. M. Walshe who found that after the net is spun and the food captured
the larvae eat both the net and the catch. A new net is then spun and the process
repeated, the whole operation lasting only a few minutes. Some larvae are
known to cause damage to aquatic plants by burrowing into the tissues and
feeding on the plant material. Another larva, a stream-dweller, has the notable
habit of feeding on the larvae of Simulium (mbwa fly). As might be expected,
the family contains predatory members which attack and feed on other larvae
but probably in most cases this habit can be modified and larvae will accept
vegetable matter if no animal prey is available.
Thus, by virtue of their feeding habits, the larvae perform the useful function
of converting much organic debris and vegetable matter into potential fish food
and hence form an important intermediate link in food chains in lakes and rivers.
Chironomid larvae and pupae have long been recognized as an important food
supply of fish. In 1908, Prof. Thienemann showed that of 24 wild fish in
Germany, 12, at some time or place, feed principally on chironomid larvae.
Prof. Johannsen in the U.S.A. wrote, ". . the larvae and pupae of the
Chironomidae are the chief constituents of the food of many freshwater fish,
and in some cases they constitute almost the sole diet of brook trout." In Lake
Victoria a number of fish feed to some extent on the larvae and pupae but
probably few fish anywhere consume as many as does Mormyrus kannume
Forsk. (the 'elephant-snout fish'). The stomach of one fish was estimated to
contain the remains of nearly 5,000 larvae. Thus, examination of fish stomachs
can be a very effective method of collecting larvae as well as for finding the
species that are most important as fish food. It was also found that regular
examination of the fish food from a standard locality threw some light on the
biology of the larvae. The next most common fish eating lake-fly larvae in
Lake Victoria, apart from other mormyrids such as Gnathonemus, are various
species of Haplochromis.

An interesting outcome of their importance as fish food has been the develop-
ment of what might be called a 'Chironomus industry'; successful attempts
have been made to cultivate larvae in order to supply the requirements of
aquarium keepers. As a result 'farms', where the popular large red Chironomus
larvae are grown in ponds and shallow lakes, have been operated. Water polluted
with organic waste matter is very suitable and such water is usually of little use
for anything else. In Moscow this industry is about a hundred years old and the
producers of the larvae have their own jealously-guarded methods for ensuring
good production by fertilization of the water. In an efficient farm 25,000 larvae
per square metre may be grown. In some cases the management has passed
from father to son, together with individual methods for high productivity. At
the end of the nineteenth century ten farms were busy in Moscow and in 1933
the industry produced about 16,000 kg. of Chironomus larvae, i.e. more than
15 tons. A kilogram cost 35-40 roubles, purifying, another 7 roubles; it was
estimated that a net profit of 150,000-160,000 roubles was made. Similar figures
were estimated for Leningrad whilst in Odessa the price was lower. The industry
thrived also in Germany. One ornamental-fish breeder obtained 3 litres of clean
larvae from 12 litres of mud and in three consecutive years received 682 Rm.,
2,000 Rm. and 3,262 Rm. for larvae, showing a profit of 60 per cent. Investiga-
tions have been carried out in America on the fertilization of small artificial
basins and the effect on the breeding of Chironomus tentans Fabr. Thus lake-
fly larvae have proved commercially profitable.
One of the most notable features of the family as a whole is the wide range
of habitats which its members have populated. Although they are principally
fresh-water dwellers, both land and sea have been colonized to some extent.
The terrestrial and marine forms have had to overcome many barriers. The
former are in danger of desiccation and the latter have to become adapted to
the high salt content of the sea. These are the primary problems but there are
others; hence relatively few species have succeeded along these directions.
There are also a number of interesting micro-habitats which have been popu-
lated. Within the giant Lobelia, so characteristic a feature of the mountain
slopes in East Africa, larvae are to be found living in the sticky viscous fluid.
Dr. K. Strenzke, investigating the fauna in the small amount of water collected
between the leaf and stem of the plant Scirpus sylvaticus Linn., found a sur-
prising collection of animals including several species of chironomid larvae.
During a German expedition in 1928-9 Prof. Thienemann found a population
of insect larvae living inside coconut shells in East Java. The coconuts were
found lying under the palm trees with a round hole in the shell uppermost, and
inside the shell was rain water. It was found that when the nuts were on the tree
they were attacked by squirrels which gnawed a round opening on the under-
side thus allowing the milk to pour out; they then crept inside and ate the nut
clean. When the nut fell to the soft ground below, the side with the opening,
being lighter, always landed uppermost. Rain water collected inside the nut
and a micro-habitat was provided which was soon utilized by insects to lay
their eggs.
Whilst different species may often be found together it is also the case that
closely-allied species sometimes have vastly different habitats and charac-

teristics. For example, a species of Cricotopus may be collected from hot springs
in Iceland, the temperature of the water being 41 C. T. A. Stuart, on the other
hand, working on the chironomids of marine pools in the Clyde Sea Area, found
a closely-related species which survived freezing in a block of ice for one or
two hours. Another example of larval tolerance can be quoted from Northern
Nigeria where a species has been found living in rock pools which are liable to
dry up in the hot season. The larvae have become adapted so that they can
survive dehydration for as long as 18 months, after which they can be reactivated
by immersion in water. Dr. H. E. Hinton found that they could be dried and
reactivated a number of times, and even after the temperature was raised as
high as 200" C. some of the dehydrated larvae lived for a few minutes when
they were immersed in water.
Although the family as a whole populates a wide range of habitats and some
members are tolerant of extreme conditions of life, most of the species occupy
limited environments. Hence certain species in a lake or pond have become
associated with the presence of certain physical and chemical conditions in the
water or mud and have, therefore, been termed 'indicator' species. In Germany
in 1910, when the bottom fauna of a number of lakes was being investigated, it
was observed that in one group of lakes the dominant bottom-dweller was the
red larva of Chironomus, whilst another group of lakes was characterized by a
small red Tanytarsus larva. In one case only a narrow crater-bridge separated
two lakes of different groups and it was soon evident that depths and tempera-
tures played no part in the difference. Further work revealed that during the
summer months the deep water of the lakes containing Tanytarsus larvae con-
tained abundant oxygen whilst that of the lakes containing Chironomus showed
a large oxygen deficiency. It was soon found that there were other marked
differences between the lake types and to distinguish them the terms Chironomus
lake and Tanytarsus lake were applied. When additional lakes had been studied
it became necessary to subdivide types further and a classification based on the
chironomid fauna emerged. The name of the typical larva in the lake was used
for the lake as a whole, e.g. Sergentia lake and Stictochironomus lake. The
system, however, suffers from several defects and is now seldom used.
Since larvae are to be found in almost every body of fresh water there are
many adaptations, both structural and physiological, which may be related to
the mode of life. Perhaps the most interesting feature, typical of a number of
forms, is the presence of the red blood pigment erythrocruorin which is closely
related to haemoglobin. The occurrence of this respiratory pigment in many
common larvae has resulted in the name 'blood-worms' being sometimes used.
As yet the functions of erythrocruorin are not fully known. The pigment can
bind sufficient oxygen for only about 12 minutes normal larval respiration so
that when there is no oxygen present in the water the pigment is of limited value.
It is, however, of great value when the oxygen pressure is low (i.e. at pressures
equivalent to between 5 and 44 per cent saturation at 17' C.). At low oxygen
pressures the erythrocruorin helps the larvae to maintain almost normal
respiration. This is important to larvae which inhabit the bottom mud of lakes
and ponds where the oxygen content may drop at times very low. With a few
exceptions the larvae that live in the deep or profundal regions of a lake are

equipped with erythrocruorin. Nevertheless, when oxygen becomes totally
absent from the water, larvae may continue to live for a long time without
oxygen, i.e. anaerobically. According to Prof. Harnisch, who has done much
work on this problem, these anaerobic processes require an aerobic phase for
the undisturbed continuance of metabolism. This may be accomplished by
increased oxygen uptake when conditions are favourable, i.e. the larvae can
repay an oxygen debt which was accumulated during the unfavourable anaerobic
conditions or, in the case of the environment remaining without oxygen, the
larvae may obtain oxygen from peroxides formed in the body. This latter process
Prof. Harnisch calls endoxybiosis. However, many stream-dwelling larvae also
possess erythrocruorin and in streams the oxygen content seldom, if ever, falls
below the optimum requirements. In experiments under conditions of low
oxygen concentration the stream-dwellers behave differently to the lake-dwellers.
Thus there are still many questions to be answered before the full significance
of the blood pigment can be realized.
Other interesting features in larvae of the genus Chironomus are three pairs
of appendices on the 10th and 1th segments. There are two pairs of ventral
tubuli on the 11th segment and one pair of short lateral outgrowths on the 10th.
These appendices were at one time thought to have an accessory respiratory
function, but Prof. Munro Fox, investigating this possibility, found no evidence
for it and their function is still unknown. The appendices show great variation
in different types of Chironomus larvae; the lateral outgrowths may be absent,
the ventral tubuli may be reduced or very long or one pair or both may be
entirely lacking. The so-called 'anal-gills' in chironomid and in chaoborid larvae
have no respiratory function despite the name. They are concerned with the
exchange of salts between the larva and its environment and ensure that a proper
osmotic balance is maintained. In Chironomus water is taken up through the
whole body surface and from the water, of course, oxygen is extracted. Salt,
however, is principally absorbed through the anal papillae or 'gills'. Larvae
living in an environment where the salt content of the water is very low have
enlarged anal papillae to provide a bigger area for absorption whilst con-
versely, larvae in the sea, where the salt content is high, have very reduced
papillae. This close correlation between the size of the anal papillae and
the salt content of the medium was first noted in 1923 by Martini whilst
working with Aedes (mosquito) larvae. Mosquito larvae differ from those
of Chironomus in that the body surface is not very permeable to water and they
obtain oxygen by swimming to the surface and drawing air into their tracheal
system through spiracles. Salt and water are taken up through the anal papillae.
Chironomids have a closed tracheal system and do not obtain oxygen from the
atmosphere. It is interesting to note that there seems to be a close correlation
between the development of the tracheal system and the presence of erythro-
cruorin. Larvae without respiratory pigment have a much better developed
tracheal system than those in which it is present. There is also some correlation
with habitat. For example, T. A. Stuart working on marine pools, found an
interesting series in one pool; three different larvae, of species of Chironomus,
Anatopynia and Cricotopus, differed in habitat, amount of blood pigment, and
development of the tracheal system. Chironomus was blood-red in colour

and had a rudimentary tracheal system; it lived in the mud and seldom left
its burrow. Anatopynia was pale red and its tracheal system was partially
developed; it lived on the surface of the mud and, being predatory, it was fairly
active thus coming into contact with a greater volume of water from which it
could obtain oxygen. Cricotopus lacked the blood pigment entirely but possessed
a well-developed tracheal system; this larva lived near the surface of the water
where aeration was greatest and, being very active, was constantly bathed in
well-aerated water. Possibly the well-developed tracheal trunks had a hydro-
static function in addition to a respiratory one, enabling the larva to maintain
its semi-planktonic habit. In this connexion it may be mentioned that the air-
sacs of Chaoborus, which act as hydrostatic organs, are the sole vestiges of the
tracheal system.
Very little is known of the biology of chironomids and chaoborids in the
tropics although they undoubtedly play an important part in the general
economy of tropical lakes. A number of common mud-dwelling larvae have been
studied, therefore, particular attention being paid to the growth rates, number
of generations per annum, hatching times and the relationship with Mormyrus
kannume, the fish which feeds on them to the greatest extent. Two species of
Chaoborus, Tanypus guttatipennis Goetgh., Procladius umbrosus Goetgh., and
a species of Chironomus were studied. It was established for each of these species
with the exception of the Chironomus species, that the life-cycle lasts approxi-
mately two months. All the larvae were collected regularly from Ekunu Bay,
Busoga, about 12 miles south-east of Jinja, Lake Victoria, where Mormyrus
were netted weekly. With few exceptions larvae were collected at weekly
intervals for a period of nine months and from the records of the lengths and
numbers of the various larvae it has been possible to build a picture of some
aspects of their biology.
The average density of larvae in the mud zone of Ekunu Bay is about 3,000
per square metre and undoubtedly the most common larvae are those of
Chaoborus. These comprise 75 per cent of the total population. Samples from
other bays and from the main lake confirmed this dominance of Chaoborus.
Much more sampling is required, however, before a map can be drawn showing
the general distribution and abundance of each species in the lake. Few regions,
if any, will have no larvae present but there may well be areas where conditions
of depth, type of lake bottom, food supply etc., allow a higher population of
larvae to live.
It seems best to consider the biology of one species of Chaoborus and then
the remainder of the Ekunu Bay species can be compared with it. It must be
borne in mind, however, that there are very many other species in the lake,
which were not investigated, and that it is important not to draw general con-
clusions from one small region for such a large lake.
Chaoborus (Neochaoborus) anomalus Edw. is a common larva in waters of
depths up to 10 or 12 metres where there is a mud substratum, but larva may
also be collected in deeper water in small numbers. Larvae were collected,
examined and measured at weekly intervals from Ekunu Bay, and, as in other
culicids, four larval instars could be determined. The whole life-cycle lasts
approximately two months. However, a few words are required now on a general

feature of the biology of Chaoborus which has attracted attention in many
countries. Chaoborus has the distinction of being classed both with mud-
dwelling and with planktonic animals. As a rule the larvae are to be found in
the bottom mud of lakes and ponds during the day but when night falls they
leave the mud and migrate into the open water, remaining there until about
daybreak. The larvae feed in the open water, being predatory on other plank-
tonic animals. Within the body are two pairs of 'air-sacs', an anterior and a
posterior, and these gas-filled sacs are said to have a hydrostatic function,
enabling the larva to remain suspended in the water at any depth. One advan-
tage of this mechanism is that the larva may rest motionless in the water until
its unsuspecting prey is within reach then with a swift, sudden movement the
larva captures the prey and holds it with its prehensile antennae. An interesting
aspect of the feeding is seen when the remains of the prey are ejected. This takes
place through the mouth and in fact the whole pharynx can be everted and
retracted. In a dish the larvae may be carnivorous on one another and often in
such cases the entire prey cannot be contained in the pharynx so that half of it
remains extruding from the jaws of the attacker.
Chaoborus anomalus, therefore, is to be found in the mud during the day.
But only the third and fourth instars are found, i.e. the half- and full-grown
larvae. The young first and second instars are entirely planktonic. Only when
larvae reach the third instar do they enter the mud, returning to the open water
at night for feeding. When the larvae are mature, i.e. after about seven weeks,
they start pupation, the relatively quiescent stage when the larval tissues and
organs are broken down and the adult organs and features develop. The life of
a pupa is only a few days, and the adult also, after emergence, probably only
survives a few days at the most. The adult flies do not feed.
Whilst Chaoborus is unique in making diurnal vertical migrations from the
mud to the open water, the other lake-flies studied were similar in the important
matter of length of life-cycle, i.e. two months. In temperate climates there are
commonly one and sometimes two generations per annum, hatching usually
occurring in the summer. In Lake Victoria, however, where there are no strong
seasonal changes and the conditions are relatively uniform, hatching occurs
throughout the year. Nevertheless, hatching is not continuous but shows a
marked periodicity. It was found in Ekunu Bay that all the individuals of a
species, e.g. Chaoborus anomalus, were, at any one time, all of approximately
the same size, that is to say, they grew, matured, pupated and finally hatched
together-the result being a large swarm of flies. It seems quite likely that this
synchronized growth and hatching is a general feature of lake-fly biology since
general observations show that mass hatches over different parts of the lake are
common phenomena and similar swarms have been recorded from .other
tropical lakes. A point of interest, however, emerged after examination of pupal
skins collected on the water surface of Ekunu Bay. The skins represented as
many as 10 or 12 species, showing that swarms may comprise more than one
species of fly. A few samples of adults from swarms were also found to be of
mixed species, confirming this point.
It is known that in some species of fly the male and female flies hatch at
different times. This would seem to be the case with some of the Lake Victoria

flies. Two collections of several hundreds of Chaoborus anomalus were found
to be entirely of females. Another collection, of Chaoborus (Sayomyia) pallidipes
Theo., made as the flies were rising from the surface of the water, was also
found to be of females only.
Clearly, therefore, there is some underlying factor or mechanism which
controls or influences growth and hatching, although at present there is too
little evidence to suggest the character of this unknown factor.
Before discussing the swarms themselves there is one other aspect of the
general biology to be mentioned. It has been stated that the life-cycle of the lake-
flies studied lasts two months. However, after measurements of the larvae,
e.g. of Chaoborus anomalus, had been plotted for several months it became
apparent that there were at times two distinct size-groups present in the mud.
When one group was mature and about to pupate the other group was only
half-grown. As was mentioned earlier, young Chaoborus are planktonic and
enter the mud when they are half-grown and these half-grown larvae take about
a month to mature and pupate. Hence the two size-groups represent a difference
of a month in age and the second group or population will mature, pupate and
hatch a month after the first. As the second population matures, the next
generation of the first population will have passed through the planktonic stage
and will be entering the mud. Thus, the relative positions of the two popula-
tions are reversed monthly and instead of a hatch every two months there is a
monthly hatch.
This interesting feature of temporally-separated populations (populations
separated in time) was found to obtain also in the cases of Tanypus guttatipennis,
Procladius umbrosus and other undetermined species of Chaoborus. In the case
of the latter, the behaviour was identical. In the case of Tanypus and Procladius
early third instar larvae appeared in the mud at stations where the depth was
6-10 metres, when another group of larvae were mature and near pupation.
Unfortunately, it was not clearly determined where the young first and second
instar larvae were living. They were not in the plankton nor in the mud in the
6-12 metre zone. It seems likely that the eggs are laid inshore and the young
larvae live there and migrate outwards at a certain age, but this point requires
The remaining common larva in Ekunu Bay belonged to the Chironomus
plumosus group but the growth-rate and life-cycle were not determined. How-
ever, weekly examination of the stomach contents of Mormyrus showed that
at four- or five-weekly intervals the stomachs contained large numbers of
Chironomus pupae. This clearly points to monthly hatches of these flies also
and suggests that the length of the life-cycle is a multiple of one month. These
larvae grow to almost twice the size of the Tanypus or Procladius larvae but
clear size-groups could not be distinguished.
As might have been expected, the stomach contents of Mormyrus captured
in Ekunu Bay showed some regular variations which could be correlated with
the changing state of the insect food. This was particularly so in the case of
Tanypus larvae which were often the principal food in this bay. More gener-
ally, however, Mormyrus are affected very greatly by the presence of two
temporally-separated populations. It has already been pointed out that not only

do the individuals of one species hatch together but that many species may hatch
about the same time. Therefore, if there was only one population it would
mean that for a considerable period there would be no insect larvae in the mud.
With two populations, however, the fish are assured of a constant food supply,
when one population hatches the other being already available. A point of
interest regarding Chaoborus remains. Although they are the most common
mud-dwelling larvae they are relatively unimportant as food for Mormyrus by
reason of their habit of migrating into the open water at night. The evidence
available shows that Mormyrus feeds at night on the bottom when Chaoborus
has left the mud, thus missing an important food supply. S. H. Deathe found
that ". . Mormyrus can be caught as well as Haplochromis by deep water
trawling at night, though in the same area no Mormyrus were caught by day-
time trawling." It has previously been observed that the fish are quiet by day
and active by night. Capt. Flower, keeping Mormyrus kannume in an aquarium
at Gezira, wrote of them: "Naturally they spend the day quietly on the bottom
of the tank, but after nightfall become very active, searching energetically for
food." He also observed that they soon learn to feed by daylight in the
aquarium, examining the bottom and the stones for pieces of chopped earth-
worms which were fed to them.
It must be emphasized that these observations on lake-fly biology were carried
out on a particular bay and it is uncertain just how far one may generalize.
Samples of larvae from adjoining regions have shown that the life-cycles follow
a similar path but relatively few species have been examined. During a short
visit to Lake George and the Kazinga Channel, some samples were collected and
larvae of Chaoborus, Tanypus and Procladius were found. They were of species
similar or closely related to those present in Ekunu Bay and again two size- or
age-groups could be distinguished suggesting that here also were two popula-
tions behaving in a manner similar to those in Lake Victoria. Strangely enough
during the two-day visit the insects were pupating and hatching, and previous
and later observations on the Ekunu Bay flies showed that they also hatched
about the same period.
In order to gather some information on the dense swarms which are so charac-
teristic a feature of at least the northern part of Lake Victoria, and of some other
tropical lakes, records were kept of the occurrence of swarms as they were
observed by various people. Thanks are due mainly to the Fisheries Officer,
Entebbe, Mr. J. D. Sampson (formerly of Dagusi Island) and to the Air Traffic
Control Officer, Entebbe, for records of hatches which they noted. These records
are by no means complete but well over a hundred swarms were observed and
recorded. As a result it can be said that hatching is not confined to a few days
of each month nor does it bear a simple relationship to either full or new moon,
but the interesting point emerged that 95 per cent of the swarms occurred during
the first and second quarters of the moon, i.e. when the moon was waxing. The
few swarms outside this period usually occurred only a day or two before the
new moon appeared. This is unexpected and suggestive but many more records
are required to test this apparent periodicity and in this case reliable negative
evidence would be just as valuable as positive. Taking into consideration the
regularity of hatching in Ekunu Bay, the possibility of a lunar influence, direct

or indirect, cannot be ignored although at present it is difficult to conceive how
such an influence could be felt by these aquatic insects. The evidence from
Ekunu Bay of synchronized hatching of more than one species at regular
intervals undoubtedly suggests some form of control and the records of lake-
fly swarms are also suggestive. What is certain is that many more data are
required, especially information on the species-composition of the swarms so
that in time it may be possible to map out the general distribution of the various
flies and their larval habitats.
It seems at present that the large swarms all originate from the relatively deep
water, perhaps where the water is over 5 metres in depth, but there remains
a large group of chironomid larvae living inshore in water up to 5 metres deep.
This group, dwelling amongst sand, stones, water-plants etc., contains a large
number of species, utilizing the wide range of habitats available in shallow
water. In order to capture these insects as they emerged, several traps were set
and then examined and emptied daily for two months. From the results there
appeared to be no special time of the month when hatching was markedly heavy.
Almost every day some flies had hatched and sometimes there were many flies in
the traps but it was not possible to correlate the catch with any known factor.
Perhaps a longer series of results would have shown some regular variation but
at present it may only be said that the inshore insects hatch almost continuously,
unlike the deep-water insects.
The last problem to be considered, but not the least, is that of control.
Perhaps the most interesting aspect of lake-fly biology to many people, especially
to those who have experienced an invasion of flies, is possible means of control.
Two lines of approach may be considered, either control of the aquatic stages
or control of the adults after their emergence. The former method has been
employed several times. Near Port Elizabeth, in South Africa, in 1934, a success-
ful attack was made on the larvae by pumping sea-water into the lake where the
larvae lived. The lake was used as a supply dam for the local power station and
a pumping plant had been installed to pump sea-water to the lake should the
level fall below requirements. However, sea-water was not admitted unless it
was very necessary, and the salt content of the water was normally low.
Chironomus larvae thrived and outbreaks of midges were serious. Copper
sulphate was added to the water to give an approximate concentration of one
part per million, but only temporary relief was obtained. Oil-spraying was
unsuccessful since the larvae, unlike mosquito larvae, do not come to the
surface to breathe and the pupae are only in contact with the surface for a short
time. Finally, salt water was pumped into the lake until there was a chloride
content of at least 3 per cent. The larvae died and the outbreaks were stopped.
In the New York Fair Ground some ponds, from which large numbers of flies
were emerging, were treated with a solution of derris powder of 5 per cent
rotenone content made up to a concentration of 6-10 parts per million. At the
same time copper sulphate was used to kill the algae which provided food and
shelter for the larvae. Treatment such as this, however, is only practicable when
the water is not required for drinking and where there is no important fish
population which may be damaged or eliminated. Furthermore, the size of the
pond or lake is a decisive factor.

With regard to the adults, control experiments have been made in some
places. Most methods make use of the fact that light has a tremendous attraction
for the flies and various traps have been devised with this in mind. Suction-fan
traps and electrocuting light-traps were used in America with some degree of
success after considerable experimentation to prevent clogging due to the vast
numbers of flies which were attracted. Experiments were also carried out to find
whether the flies-in this case Chaoboris lacustris Freeborn-showed a selective
colour response. Various coloured lights, such as red, green, light and dark
blue, blue-violet and ultra-violet (all discharge tubes) and a white incandescent
lamp were set along the shores of the lake from which the flies came. Associated
with each lamp was a trap and all the lamps were of approximately the same
intensity. At each station there was a control lamp and trap to compare with
those in the experiment thus ensuring that local temperature and wind variations
were taken into account. The result of this experiment showed that roughly the
same number of flies were attracted to each light and hence that there was no
selective colour response. A similar experiment, using lamps of different intensi-
ties, however, showed that the flies reacted differently. The number of flies
captured increased rapidly with increased light intensity when lamps of 25 to
300 watts were used and the numbers decreased when lamps of 1,000 watts and
over were used. In practice it was found that a 1,500 watt lamp attracted approxi-
mately the same number as a 200 watt lamp. These observations were made, of
course, on a particular species of fly and one may not deduce that similar results
would be obtained with flies from Lake Victoria. However, it seems likely that
the behaviour of the latter would be comparable and at any rate the American
experiments suggest means of relief. On a basis of light attraction a good deal
of protection might be obtained by arranging a light-barrier around a house. It
is important that the lamps should be facing the house and shaded on the other
side in order to attract only the flies around the house, and, of course, the lamps
outside should be of a higher intensity than those inside.
The question whether the lake-flies can be controlled during the daytime or
before they reach the shore remains. It has often been observed that the flies on
Lake Victoria usually form a distinct and definite column and are not dispersed
at random. This feature of their behaviour would simplify aerial spraying of
insecticides. The swarms are big enough to be seen many miles away but an
aerial attack would be expensive, probably too expensive to be practicable.
However, the successful and economical control of an insect depends to a large
extent on a clear knowledge of its biology and life-history and at present there
are still many gaps in this knowledge. The task of ridding the lake completely
of its lake-fly population is impracticable, and, in the light of what has been
said in the previous pages, undesirable, but although immunity from attack
cannot be obtained much could be done by means of traps to lessen the nuisance
of the lake-fly invasions.
It is hoped, however, that in this outline of the biology of lake-flies, enough
has been said to justify their existence, and to cause them to be regarded not
simply as a pest found near lake shores but as the final stage of a complex and
interesting insect life-history.

N view of the widespread interest aroused by the Rev. Gervase Mathew's
recent survey of archaeological sites in the Masaka District and the subsequent
location of other sites in the Mubende District by Mr. Lanning (see Uganda
Journal, 17, p. 51), it may be permissible to anticipate the publication of a new
collection of Ankole traditions compiled by Mr. Katate (which is at present
being revised for the press in Mbarara) and to invite comment on a claim which
it makes about the Bachwezi, to whom both sets of sites are traditionally
Hitherto most of our knowledge of this mysterious dynasty has come to us
through the traditions of Bunyoro, in particular through the notable contribu-
tions of K. W. to earlier numbers of this Journal and through the Abakama ba
Bunyoro Kitara of Mr. J. W. Nyakatura. From these we learn that before the
coming of the present Babito dynasties to Bunyoro and Buganda approximately
19 generations ago, there was a brief period when nearly the whole of southern
and western Uganda was united under the rule of two Bachwezi kings, Ndahura
and Wamara. The traditional site of Ndahura's capital is Mubende Hill. The list
of his dominions starts with Bulega to the west of Lake Albert and a province of
unspecified size to the north of Lake Kyoga; it covers the whole area between
the Victoria and Somerset Nile and the Katonga River; it stretches east through
Busoga to Mount Elgon; south of the Katonga it includes Busongora, Bwera,
eastern Ankole, Buddu and the Sese Islands. By the time of Wamara the empire
has contracted, and its centre has shifted south from Mubende to Bwera, the
district lying to the south of the middle Katonga, in which are the earthworks of
Bigo and Ntusi,
Apart from the territorial extent of their empire, however, the Bachwezi are
remembered throughout western Uganda as the originators of a pattern of social
organization and a religion which were sedulously, if uncomprehendingly,
imitated by the successor dynasties, whose subjects looked back to the Bachwezi,
much as mediaeval Europe looked back to the Roman Empire, as the golden
age of civilized and pre-barbarian government. The Bachwezi left their regalia,
their drums, their copper spears, their beaded crowns surmounted by tall copper
cones and underhung with masks made from the skin of the colobus monkey,
distributed about the country in the keeping of the district chiefs who had served
under them. They left their reed palaces, which were faithfully copied upon the
sites of subsequent capitals. They left their slave artisans and even, it is said, their
palace women, from whom their Babito successors learned the pastoral duties
of royalty. They left, in all probability, a working system of administrative
officialdom accustomed to ruling small districts as the local representatives of a
centralized monarchy, and a regimental organization, later developed in the
Bitongole, under which young men were conscripted into the military service of
the king and were maintained by the peasants occupying areas of land designated

for the support of the army. They left, finally, some notions of a religion which
was ordered according to the phases of the moon, and which recognized the
power of the spirits of the dead to take possession of living people and to give
oracles through their mouths.
Of the fall of the Bachwezi and their supersession by the Babito, Bunyoro
tradition, if read in the light of Father Crazzolara's researches on the Nilotic
migrations, gives us a fairly adequate explanation. The Bachwezi were a ruling
clan, a tiny minority, who maintained their dominant position mainly through
the prestige inspired in the local populations by their apparently superhuman
knowledge and skills. That prestige, we are told, they were already fast losing
when the Nilotic invaders appeared on the northern frontier. As Mr. Kamuhigi
of Rwengoma described it to me, "They were a just people among themselves,
but they discriminated in their legislation between themselves and others. They
lost face when it was gradually discovered that they were human like other
people, and they also became unpopular because they tried to monopolize the
wealth of the country." Thus, while the earthworks to the north of their capitals
at Mubende and Ntusi seem to stand as evidence of their intention to defend,
first the line of the Nkusi and then that of the Katonga, against the Nilotic
invasions, it may well be, as tradition tells, that apathy among their soldier sub-
jects prevented them from using these fortifications, and that the Bachwezi clan,
after gathering for defence, was forced to make an ignominious escape either to
the south or to the west.
Where Bunyoro tradition is quite unhelpful, however, is in giving any indica-
tion of where the Bachwezi came from or how their empire in southern Uganda
originated. "The Bachwezi," says Nyakatura, "were white people who came
from the north-east. They were more powerful than the natives, and they made
themselves chiefs over the Lango and the Acholi before they crossed the Nile."
Having carefully differentiated them from the Bahima, he then goes on to tell
the story of how Isaza, the last of the pre-Bachwezi rulers of Bunyoro, received
an embassy from Nyamuyonjo, the king of 'the Underworld', of whereabouts
unspecified. At the end of this story Isaza is lured away to the Underworld, where
he weds Nyamuyonjo's daughter Nyamate, and becomes the father of Isimbwa,
the first Muchwezi. According to the Bunyoro version, Isimbwa was not a king-
for, on the disappearance of Isaza from Bunyoro, his minister Bukuku usurped
the throne. It was only when Ndahura, the son of Isimbwa by Bukuku's daughter
Nyinamwiru, assassinated his grandfather, that the Bachwezi began to rule in
Bunyoro. Even then, says Nyakatura, Isimbwa lived on, a peripatetic voluptuary,
to become in some mysterious way the ancestor of the Nilotic Babito some two
generations later.
Against this confused account, which seems to be concerned primarily to
supply a fictitious continuity between the Babito dynasty and its predecessors,
the new collection of Ankole tradition by Mr. Katate, while confirming the
stories of the embassy of Ruyonga (Nyamuyonjo) and the marriage between
Nyamate and Isaza, roundly declares that the Underworld was Bwera; that
Ruyonga, Nyamate and Isimbwa were all Bachwezi; that Ruyonga and Isimbwa
were both kings, the former ruling from Kishozi, on the north bank of the
Katonga, opposite Bigo, the latter from Byezigoro bya Mugenyi in Bwera, also

near Bigo; further that all these Bachwezi were the descendants of Rugaba and
Ruhanga, both praise-names of Imana the Supreme Being of Ruanda, but also,
according to Father Pages, used of the mythical ancestors of the Batutsi, who,
"left the sky with their cattle and after walking for a long time arrived at
Mubari", a well-known ford on the Kagera River between Karagwe and Ruanda.
Finally, in Mr. Katate's version, Ruyonga is linked to Rugaba through another
"Nyamate Omuchwezi", a man this time and a king, who had his capital at
Katikabayagi in Nshara county, near the modem boundary between Ankole and
In other words, thinly as it is preserved, Ankole tradition seems to provide us
with an account of the origins of the Bachwezi hegemony which is at least more
consonant with the mighty legacy of that dynasty to its successor states in western
Uganda than the truncated account preserved in Bunyoro. From it, the Bachwezi
emerge as a royal clan among the Batutsi aristocracy which subjugated Karagwe,
Ndorwa, Gissaka, Mpororo, Ruanda, Ankole and finally Bunyoro. In the story
of Isaza and Ruyonga what is happening is that the Bachwezi dynasty, already
established for some generations in Bwera, is making its first attempt to extend
its sway northwards, over the region between the Katonga and the Nile.
Ruyonga, it appears, succeeds in making a prisoner of Isaza, but Bukuku
manages for another generation to defend the northern kingdom from Ruyonga
and Isimbwa. With Ndahura, the son of Isimbwa, the extension is at last
achieved; but with Ndahura's son, Wamara, the advance is permanently checked
by the arrival of the Lwoo from the north, the dynasty collapses, and its civiliza-
tion survives only at second-hand, becoming increasingly degenerate as the
centuries pass.
It is true that such a claim, even if substantially correct, does not tell us where
the Bachwezi originally came from. It also leaves open the whole question of
the pre-Bachwezi kingdom of Isaza between the Katonga and the Nile. It may
be that the 'sky' of Batutsi legend refers to this 'upper kingdom' in just the same
way as the 'Underworld' of Bunyoro tradition evidently refers to the area south
of the Katonga. It may be that the Bachwezi of the lower kingdom came ulti-
mately from the north via the upper kingdom; and it is possible that the Buganda
and Busoga legends of the migration of Kintu and his followers from beyond
Mount Elgon may refer to a movement which was the parent of both the upper
and the lower kingdoms. Where Mr. Katate's claim is important, however, is in
connecting the Bachwezi civilization, immediately, with the south-with
Karagwe, Mpororo and Ruanda. It suggests the urgency of a search for the
earlier Bachwezi capitals in that direction, which, if found and excavated along
with the sites in Bwera and Mubende, would surely add a wealth of material
evidence to support and qualify the allegations of tradition. Indeed, with the con-
tribution of archaeology, it should be possible to add a chapter to the history of
Uganda which would be of significance for the whole mediaeval history of

THE amagasani, or shrines, of the Babito dynasty of Bunyorot are an interest-
ing survival and doubly so since their future is by no means secure. They are
scattered over western Uganda, many of them in areas which are difficult of
access and most of them outside the restricted frontiers of the twentieth century
Bunyoro kingdom. Some are well looked after, others are partially or wholly
neglected. Some are well-built and impressive, like the egasani of Kabarega near
Hoima; others, like the shrine of Kyebambe I, Omuzikya, in Toro have nothing
to show but the feeble framework of some former hut in the midst of a clearing
which will soon belie its description. Meanwhile, elephants have destroyed the
shrine of Olimi I, Rwitamahanga, and there is no trace of the burial of
Omukama Nyarwa save an uncertain legend of his having been murdered by
his brother, Chwa I, Rumomamahanga.
The death of an Omukama was kept secret as long as possible in the hope of
avoiding a struggle over the succession. On occasion the two most important
chiefs, the Mugema and the Bamuroga, even attempted to keep the secret by
impersonating the actions of the dead ruler in the doorway of the royal hut,
choosing a time when no one outside the court circle was sufficiently near at
hand to recognize the deceit.
The Mugema was the greatest of the saza chiefs and governed the saza of
Nyakabimba, now incorporated with Kyaka in Toro. The Bamuroga, who was
of rather less importance, was a household official, head of the royal servants
and keeper of the drum Kanumi. These two chiefs supervised the proceedings
leading up to the burial of the Omukama and were responsible for the mainten-
ance of the shrines. It was under their direction that the Abaswato, servants of
the dead king who during his lifetime had looked after his personal possessions,
removed the rusaya, or jaw-bone, from the body in accordance with a custom
which the Babito are said to have brought with them to Bunyoro. They also
removed the stomach and to preserve the body they dried it over a fire, the whole
proceedings taking place in absolute secrecy. Meanwhile, if the secret had been
well kept, the two chiefs sent for the eldest son of the former Omukama and in
his presence placed grains of millet in the hand of the dead ruler. The son then
ate the millet as a symbol of his succession to the office held by his father.
When it was either impossible or unnecessary to maintain the secret any longer
a young nephew of the dead Omukama was ordered to climb the roof of the
royal house, the Karuzika, carrying a container of milk. When he reached the
top he seized the spear, the galengera, which had remained fixed on the peak
of the hut throughout the king's lifetime as a symbol of the royal presence and
authority, and dragged it from its position. Simultaneously he poured out the
milk he was carrying and called out in a loud voice, Obwire bwaira kabiri amata
gayatika Omukama atulize--"Darkness has come in the daytime, the milk is
spilt, the Omukama is gone." Then, with his task completed, the youth descended
1 The Bakama of the Babito dynasty are listed in 'The Kings of Bunyoro-Kitara' by
K.W. (Uganda journal, iii (1935-6), p. 155-6).

:and was speared to death by the royal executioners. The royal fire and all other
fires in the vicinity were extinguished and all work ceased until after the burial.
If the Mugema and Bamuroga had managed to get hold of a popular
successor to the dead ruler the next step was to appoint another son as
Omuragwangoma to act as Omukama until the new ruler took office. At the end
of that interim period the Omuragwangoma was put to death. If, however, they
had failed to secure the succession, the eligible sons of the Omukama sought
refuge with any chiefs who would befriend them and the country lapsed into a
state of chaos whilst the brothers made war on each other until one had
triumphed over the rest. Only then did they turn their attention to the task of
burying the old Omukama. If the Omukama had died in war his body might
have been buried immediately and only the rusaya reserved for ceremonial
burial. Generally, however, both body and jaw were kept until the succession
was secure, when the saza chiefs decided on the place of burial.
The site of the egasani where the jaw was to be buried was, traditionally, one
which had witnessed some important event during the Omukama's lifetime, or
it might be in the vicinity of the royal house. The body was usually buried some
miles distant. Chopi in the northern part of the present Bunyoro district is said
to have been the area in which the earliest Babito kings were buried although
there are now no traces of amagasani there. As the Banyoro pressed south-east-
wards in conflict with the Baganda, however, the centre of the kingdom shifted
to what is now Singo and northern Mubende. The jaw of Winyi I,.Rubem-
bekantara, was buried at Kibulala in Singo where the egasani is still to be found.
'The body of his successor, Olimi I, Rwitamahanga, fifth of the Babito kings, is
also said to have been buried in Singo although the egasani, having been over-
run by elephants, cannot be traced. The amagasani of Nyabongo I, Chwa
Rulemu, are said to be in Singo and Mubende, but they are no longer to be
found. The jaw of his son, Winyi II, Rubagirasega, was first buried in Singo and
was later moved to Mubende District where the egasani still stands. The jaw of
the eighth Omukama, Olimi II, Ruhundwangeya, is known to have been buried at
Burongo in Singo County, but the spot is marked only by a clump of 'elephant
sugar-cane' (Dracaena fragrans (L.) Ker-Gawl.; Lunyoro, mugorogoro) such as
forms the living fence which surrounds a number of the amagasani. The egasani
which houses the body of Olimi II is in Mubende.
The struggle between Nyarwa Nyamulirahaiguru, ninth Omukama, and his
brother and successor, Chwa I, followed by the far-ranging wars of the latter,
carried the centre of the kingdom southward into what is now Toro. The murder
of Nyarwa is thought to have taken place there. Chwa himself was buried near
the royal hill at Rubungo'in Bukoba District, Tanganyika, where his egasani,
consisting only of a clump of trees, has been discovered by Father Nicolet of the
White Fathers' Mission. Chwa's sister, Masamba Omubitokati, who ruled in his
absence and for a short time after his death, is said to have been buried in
Mubende although her egasani is not known. Her. successor, Chwa's son,
Kyebambe I, Winyi Omuzikya, was buried in Toro where his egasani is now in
a state of extreme neglect. The headquarters of Winyi III, Rugurukumacholya
were at Ruwembuba on the borders of Singo and Bugangadzi, but his egasani is
in Bunyoro on the northern bank of the River Kafu.

After Winyi III it is difficult, on the evidence available, to disentangle the
order of succession among the Abakama. Nevertheless, it is clear that the
Buyaga and Bugangadzi sazas of Mubende became the centre of the Bunyoro
Kingdom and also the main burial area. Circumstances induced Omukama,
Kyebambe IV, Kamurasi, to take up residence further north in what is now
Bunyoro, but he was buried in Mubende and an attempt was made, even in the
twentieth century, to arrange for the burial of Chwa II, Kabarega, in the same
area. However, owing to a delay in the negotiations with the Buganda Govern-
ment, the attempt failed, and Kabarega, like his son (Andereya) Ruhaga II, was
buried in Bunyoro near Hoima.
The first stage in preparing the egasani was to dig a hole some thirty feet deep
and large enough to accommodate comfortably a reclining body. The hole was
spoken of as an iziba (Fig. 2b), normally meaning a well, as an indication that the
Omukama was as the water supply of his people. It was, too, in keeping with the
Nyoro practice of using special words to describe the persons and actions of their
Abakama. Many people were employed in digging the iziba and the work took
several weeks to complete. On the day fixed for the burial the rusaya, wrapped
in a cow's skin and placed on a rectangle of wood rather like a door, was carried
to the egasani by the Abaswata along with the Omukama's personal possessions,
his ebikwato. Barkcloth was spread over the bottom of the iziba and one of the
abago, or wives, of the late Omukama, either the eldest or the best beloved, took
her seat on the bark-cloth, facing north, with the rusaya, still wrapped in the
cow's skin, in her lap. Onlookers were then seized, their arms and legs were cut
off and they were thrown, still living, into the iziba on top of the Omugo
(principal wife) until the hole was full. There is a tradition that if the iziba was
full before all those who had been seized for sacrifice had been thrown to their
fate any surviving women were given as wives to the surviving men. Kyebambe
IV, Kamurasi, who died in 1869, was the last Omukama to be buried to the
accompaniment of human sacrifices of the sort described here.
To mark the iziba a raised rectangular tablet of mud (Fig. 2a), was con-
structed round the edge, partly covering the hole. Meanwhile, in preparation for
the burial, the largest bull in the royal herd had been killed and its skin was
now stretched over the open mouth of the iziba and pegged down with the heads
of nine hoes, known as enfuka (Fig. 2d). Nine, like seven, was regarded as a
lucky number. Normally one hoe was placed at the head, or southern end, of the
iziba, two at the foot and three along each side, although that order was not
rigidly observed. I have been unable to discover whether the hoes have any
ritual significance in this context or whether their use was due merely to their
handy shape. They are pointed at the end which was used to pierce the hide,
whilst the upper end is broad like the spade symbol on a playing card and acts
like the head of a huge nail to hold the hide to the ground (Fig. Ig). Sometimes,
too, the spear which had acted as galengera was placed upright in the iziba with
its head projecting through the hide and was clearly meant to imply that the
Omukama was in residence in his new home. This was done in the amagasani of
Kyebambe III, Nyabongo II and Kamurasi (Fig. 2c).
Next, a hut was erected with great expedition over the iziba, the work having
to be finished before nightfall in order to protect the iziba from rain or dew.

Although the existing huts may bear no resemblance to the original amagasani
it would seem that there was no special design for them, a pattern common to the
time and the district normally being adopted. Some, like those of Winyi I and
Kyebambe III, are shaped like half-an-egg, with a bamboo frame and a thatch
of grass reaching to the ground. Others, like that of Olimi II and Nyabongo II
are circular mud-and-wattle huts thatched with grass. The amagasani of Winyi
II and Kamurasi are made of the same materials, but are roughly square in shape
while, to judge from the existing framework, the egasani of Kyebambe I was
unique in being markedly oblong. None, however, is of imposing dimensions
save that of Kabarega, the splendour of which well befits the character of that
redoubtable warrior-king. The majority average from twelve to fifteen feet in
height and have a floor area of from 200 to 300 square feet. It was also customary
to have two doors, one facing north and one south, but now that practice is
rarely observed.
Within the hut the iziba was still further enclosed by curtains of bark-cloth
which surrounded the edge of the raised tablet. The practice of suspending a
curtain horizontally, like a roof over the others, as seen in Kabarega's tomb, is
a modern innovation and is said to have been copied from a mosquito net.
Although intended to provide greater privacy for the iziba the curtains do not
seal off a holy of holies which may not be entered. The royal leopard skins, upon
which the late Omukama had been accustomed to sit, might be spread actually
on the edge of the raised tablet round the iziba or might be used as mats on the
floor of the hut. In the latter case no one must stand on them.
Finally, the ebikwato were placed in the hut. There is, however, no limit to
what might be considered ebikwato. The egasani of Nyabongo II, which con-
tains the best collection, has, amongst other things, an iron carpentry-tool
resembling a chisel whose wooden handle is missing, and a small metal bell, five
inches long, with which the Omukama was wont to summon his servants after
finishing his solitary meal. There are also strings of beads, red, pink, royal blue,
pale blue and white. One string is adorned by the four canine teeth of the
Omukama and another by two leopard's claws. At Nyabongo's egasani, as in
that of Winyi II, there are also ebirongo byomunsika, wooden objects shaped
like an English decorative shield eight inches high and six inches wide, with
three metal prongs at the top and two metal hooks projecting in front (Fig. If).
These could be fixed to the top of a forked stick, some three feet long, which was
stuck in the ground, and from the prongs and hooks skin milk-buckets and
shields could be suspended. Whilst eating utensils and ornaments might be
included to serve the needs of the buried Omukama, the basic ebikwato com-
prised the personal spears, shield and drums of the Omukama. This may be con-
cluded both from general evidence and more particularly from the case of
Kabarega's egasani where every facility was available for including a wide range
of articles, but only spears, shield and drums are to be found. The number of
spears is not fixed, although three is not an uncommon number to find. In the
egasani of Nyabongo II there are three spears, two with cumbersome iron heads,
twenty inches long and straight edged (Fig. la), and one similar in shape with
:a fifteen-inch blade. At the other extreme of design there is the neat, three-inch-
long, elliptical, brass spearhead in the egasani of Kabarega (Fig. Ib). Between

these two extremes must be placed the ornate spears in the egasani of Winyi II.
One has a triangular iron head seven inches long with a A-shaped prong:
immediately below it (Fig. Ic). Another, with a six-inch-long triangular head,
has a series of receding prongs below the head like the bones of a fish's spinal
column (Fig. ld). The main qualification, however, is that the spears should
have been the personal weapons of the Omukama and not part of the royal
regalia which remains in the palace of successive rulers. Often one of the spears
is marked with a spiral groove in the metal below the head indicating that it was
given to the king by his father when he reached manhood (Fig. le), a practice
common to all youths of the Babito clan. In apparent contradiction to this state-
ment, however, the guardian, mutaka, of Winyi II's egasani referred to the spears
in his keeping by the names Kaitantahi, Muhungyo and Kiinegena and the
mutaka at the egasani of Kyebambe I spoke of one of his spears as Kimuli
Kylkya Mahanga (The torch which burns up all nations). All these are names of
spears which form part of the permanent royal regalia and which would not be
removed from the palace of the reigning monarch. It is probable, therefore, that
the guardian of Winyi's shrine had come to regard the spears as the regalia of the
dead Omukama and had named them accordingly, whilst the mutaka at
Kyebambe's egasani may well have been influenced by the copper head of his
spear in attributing to it the name of one of the spears in the royal Karuzika.
The engabo, or shields, which are to be found in some amagasani are
elliptical in shape and from. 20 to 24 inches in length. They are made of hide, but
in certain cases have a boss, or centre piece, of some other material. Kabarega's
engabo has a pointed wooden boss. Of Winyi II's shield there remains only an
elliptical centre piece of iron eleven by eight inches.
The drums are of two kinds, the long ngaija and the short ngoma. An
Omukama usually made two of the smaller drums soon after his accession, one
of which remained in his house and the other (there might be more than one
of these) accompanied him to war or on a journey. On the King's death the
former was deposited among the drums of his ancestors in the house of the new
Omukama while the latter accompanied the dead king to the egasani. The long
drums were a different matter. There is one official ngaija in the royal regalia
known as Tomunju, that is, "Everything belongs to me". But others were given
as presents to the Omukama and it was these which were placed in the egasani.
There might only be one, as in Kabarega's egasani, or as many as three, as in the
shrine of Nyabongo II.
There is no special place for the ebikwato in the egasani. The spears, with
their heads in bark-cloth bags, usually stand inside the curtains which surround
the iziba. The remaining ebikwato, also wrapped in bark-cloth, lie nearby. At
the egasani of Kyebambe I the ebikwato are kept in the hut of the mutaka
since the shrine itself is in ruins. On the morning after each new moon, however,
the ebikwato are still arrayed outside the northern door of the egasani and
remain there till sunset if there is no rain. Then, too, a fire is lit in the egasani as
a relic of the custom, which passed with the coming of the Europeans, that a fire
burned continually in the shrine as it had done in the royal house during the
Omukama's lifetime.
To be joint ministers at the shrine there were chosen an omugo and a mutaka,

each living in a separate hut in the vicinity of the egasani. The omugo was
responsible for keeping the interior of the shrine clean and maintaining the
ebikwato. On her death she was replaced by her daughter, her sister or some
other female relation of the same clan, either the Abalisa or the Abakwonga.
Since the borders of Bunyoro were fixed in a manner which excluded the greater
number of amagasani, however, no new abago have been appointed to look after
the shrines. One old omugo, Victoria Kaganda, still lives at the shrine of
Kamurasi, but she will probably not be replaced when she dies. At Kabarega's
egasani, close by the present capital of Bunyoro, there is still an omugo, and the
widow and daughter of Ruhaga II lived near his egasani which also is not far
from Hoima.
The mutaka was chosen from one of a number of clans which were not dis-
pleasing to the Omukama and he was given wide powers over an area, about
three miles in radius, round the egasani. Even the saza chief in whose county
the shrine was sited had no jurisdiction over the egasani land. No one was
allowed to be buried there and no one was allowed to pass near the shrine carry-
ing any article on his head. The mutaka was responsible for the maintenance of
the egasani and for the rebuilding of the hut every eight or ten years, the latter
task always being completed between dawn and sunset. The villagers in the
vicinity were required to give their labour on demand and they also supplied the
mutaka with food. On his death the guardian was replaced by his son or brother
or by any other male member of the same clan. The Mugema or Bamuroga, or
even the ruling Omukama himself, saw to it that the guardian fulfilled his duties.
A bull was given to the egasani and was allowed to wander and eat at will. Its
skin was used to replace the old one which covered the iziba when it became
The new Omukama, immediately after the burial, went to drink from a well
named Wamala in Mubende before proceeding on an expedition to Ankole to.
capture cows. On his return, he went to the egasani of his predecessor where he
milked the cows which he had brought back with him, putting the containers
into the iziba where the milk was ostensibly drunk by the dead king, but in fact
provided food for the omugo. Then, he chose some of the younger and more
attractive abago of his predecessor to be his own wives, leaving the rest, if they
had no relatives to maintain them, to live near the egasani in company with the
omugo in charge. The king then proceeded to the erection of his own palace,
leaving a number of cows at the egasani to provide milk for the mutaka and the
dead Omukama, the latter's share being consumed vicariously by the omugo.
The cows were replaced as was necessary, but they might not be killed without
the Omukama's permission.
Most of the Abakama had a second and less important egasani, the egasant
ebijwaro, that is, the shrine in which their bodies and garments were buried.
Omukama Isansa requested that his jaw-bone should not be separated from his
body and in more recent times Kamurasi, Kabarega and Ruhaga II have all been
buried intact. This was not the normal practice in former times, however. Usually
on the day on which the jaw was buried the body of the dead Omukama was
wrapped inside the skin of a cow, together with the clippings of his nails and his
hair cuttings, both of which had been carefully preserved during his lifetime.

An iziba similar to that prepared for the jaw was also dug for the body which
was lowered down the hole and placed in a lying position on its right side. The
clothes of the Omukama were then laid on top. No omugo was buried with the
body nor were any other people sacrificed there. Otherwise, the egasani was
similar to the more important shrine where the jaw was buried. A raised mud-
tablet partially covered the mouth of the iziba and the hole in the centre was
covered with the hide of a bull pegged down with nine enfuka as before. Bark-
cloth curtains were hung round the iziba, a hut was built to protect it and a few
of the dead king's spears might possibly be put inside the curtains. An omugo
and a mutaka were also appointed to superintend the care of the tomb and a bull
and cows were given by the new Omukama to wander round the egasani and to
provide food for the dead king and for the guardians of his shrine. Once again
it was the duty of the Mugema and Bamuroga to visit the egasani and see that
it was well maintained.
To the description of the shrines set out above the egasani of Ruhaga II who
was a baptized Christian, offers an interesting contrast. There the iziba is com-
pletely covered in and sealed by a raised mud-plaque which is surmounted at
,one end by a cross. Nevertheless, old custom survives in the shape of the curtain,
made of cloth and not of bark-cloth, round and over the grave, whilst a fine
hut protects the shrine and three spears stand near at hand as if to serve the
Omukama in his new environment.
The long struggle between Kabarega and the British and the ultimate defeat
sof the former (1899) led to the curtailment of Bunyoro's boundaries and resulted
in the restriction of many of the practices relating to the amagasani. The office
of Bamuroga disappeared and the Mugema lost his position as leading saza chief.
In 1902, however, Ruhaga II induced the Protectorate Government to permit
one man to reside in Mubende District, by then part of Buganda, to supervise
the egasani in that area. That man henceforward was to bear the title of Mugema,
but he no longer possessed the powers which formerly went with the title. The
Native Treasury in Bunyoro paid his salary and also provided money for the
upkeep of the amagasani wherever possible. At the time of the Mailo Survey of
Buganda five acres of land were also allotted to each egasani in Mubende
District to enable a guardian to maintain himself whilst looking after the shrine.
The office of mutaka still remained with one family, but the office holder was no
longer held in such high regard by his neighbours as had been his predecessors.
When poll tax was introduced the Government of Bunyoro paid the tax to
Buganda on behalf of the guardians until 1950. Since then the guardians have
received a monthly salary of Sh. 22 from the Bunyoro Government and have had
to pay their own poll tax. The Mugema collects the salary on behalf of the
.guardians and distributes it among them to augment their income from their
five acres of land. Cows are no longer granted to the shrines as in former days.
The guardians' enthusiasm for their work seems in many cases to be declining
and, possibly owing to the difficulty of getting labour now that Mubende is no
longer in Bunyoro, only the egasani of Kyebambe III, which is near the home
of the Mugema, is in really good repair. This apparent neglect may also be due
in part to the fact that the old Mugema, who died in 1952, admitted with regret
that he had been unable, through infirmity, effectively to carry out the super-

a b

c d f


Scale 1:10
FIG. 1
Objects from the amagasani of the Abakama of Bunyoro.

Not to scale
FIG. 2
The rectangular tablet of mud covering the iziba.

.c ,. ". \ ',..

SI17 1

n\ 2- '- "\

IMubende ,. District boundaries
o000.00. S ozao boundaries
t/ > eb *' o \

-*". .. I.yegeqwoa I oads---- Eqosaoni z

FIG. 3
Sketch map of north-western Uganda showing locations of the amagasani of the Abakama
of Bunyoro.

1. Winyi I (Jaw-bone). Egasani in fair condition.
2. Winyi II (Jaw-bone). Egasani in fair condition.
3a. Olimi II (Jaw-bone). Marked by a clump of trees.
3b. Olimi II (Body). Egasani in fair condition; unfenced.
4. Chwa I (Body and Jaw-bone). Not shown on map, but is at Rubungo, south of
Bukoba, Tanganyika Territory; marked only by a clump of trees.
5. Kyebambe I (Jaw-bone). Egasani in total disrepair.
6. Winyi III (Whole body). Egasani in fair condition.
7. Isansa (Whole body). Said by some authorities to be same person as Olimi III; egasani
in fair condition.
8a. Ruhaga I (Jaw-bone). Egasani in good condition.
8b. Ruhaga I (Body). Egasani destroyed by recent bush fire.
9. Olimi IV (Whole body). Egasani in fair condition.
10a. Kyebambe 11I (Jaw-bone). Egasani very well preserved.
10b. Kyebambe III (Body). Marked on Survey Dept. map of Mubende District as 'Tomb
of Nyamutukula'; egasani overrun by elephants and never repaired.
I la. Nyabongo II (Jaw-bone). Egasani well-kept.
I b. Nyabongo II (Body). Marked on map of Mubende District as 'Tomb of Mugenyi';
egasani in fair condition.
12. Olimi V (Whole body). Said by former Mugema to be egasani of Rueru who ruled
only for a few months; marked on map of Mubende District as 'Tomb of
13. Kamurasi (Whole body). Egasani in fair condition.
14. Kabarega (Whole body). Egasani very well cared for.
15. Ruhaga II (Whole body). Egasani very well cared for.
16. Marked on map of Mubende District as 'Tomb of Kabagonza'; Kabagonza was a
brother of Kabarega but was never Omukama; an ant-hill fills the hut.
17. The Butemba area is the burial ground of a number of descendants of Kabarega,
none of whom was ever Omukama. There are no huts, but small clearings mark
the whereabouts of several graves.
18. Marked on the map of Mubende District as 'Tomb of Kanumi' but was, in fact, the
place where the royal drum, Kanumi, used to be kept.

vision of the more distant shrines, especially that of Winyi I at Kibulala which
also fell within his jurisdiction. This latter shrine presented a particular prob-
lem. At the time of the distribution of Mailo land the area on which the egasani
of Winyi I stands became the property of Omumbeja (Princess) Nkinzi of
Buganda. There was, therefore, no land available in the vicinity of the shrine to
maintain a mutaka and the Princess demanded rent for the land occupied by the
shrine. The situation was made easier, however, when Samweli Kaliga, who
owned land near the egasani, undertook the task of acting as guardian and when
he died in 1949 the heir to his land, Paulo Lubega Lugwa, in turn became
mutaka. Attempts are, however, being made by the Bunyoro Government to
purchase land around the shrine in order to ensure that its maintenance is not
left solely to the goodwill of a neighboring landowner.
The isolation of the egasani of Kyebambe I in Toro has also led to its serious
neglect. The present mutaka, Kagoro, succeeded his father thirty years ago. He
occupies a small area of land adjoining the egasani and formerly there were,
attached to the shrine, cows which he was free to milk. The last cow died fifteen
years ago and since then no new ones have arrived, nor does the mutaka receive
any pay for his work which he continues to carry out partly because his father
passed on the responsibility for the egasani to him and partly also because he is
in awe of the spirit of Kyebambe. He still lays out the ebikwato, three spears
(sent there by Kabarega to replace others which had been lost) and three leopard
skins, at the door of the egasani after every new moon; but the hut itself has not
been repaired during his tenure of office.
Even in what remains of the kingdom of Bunyoro it is not easy to maintain
the amagasani. The shrines of Kabarega (d. 1923) and of Ruhaga I (d. 1924),
both comparatively recent in origin and near to Hoima, are kept in good repair
by labour supplied by the Native Government. The amagasani of Winyi II and
of Olimi IV, however, being less accessible, are more difficult for the Assistant
Mugema, who lives in Hoima, to supervise, also it is difficult to obtain the ser-
vices of conscientious guardians now that more remunerative occupations are
readily available. It would, however, be a serious loss if the remaining shrines
were allowed to disappear from sight as a result of any of the difficulties which
have been mentioned. It would be well if those which are now no longer in a
good condition could be restored to their original state. Only the shrines located
in Mubende appear to be marked on readily-accessible maps and not all of those
are accurately named. The accompanying map shows the whereabouts of all the
amagasani which I have been able to trace and also rectifies some of the errors
in the 1:250,000 map of Mubende District (Uganda Survey Dept. Map No.
A962, 1945).
The information contained above is far from complete although it has been
collected from many different sources. I should, however, like to mention my
indebtedness to Mr. Nyakatura, to Bishop Balya, to Omubito Karubanga, to the
Rev. Ezekyeri Binyomo and to the guardians of the various amagasani, all of
whom, in the course of long conversations have given me invaluable assistance.

T is unnecessary to inform readers of The Uganda Journal that John Hanning
Speke and James Augustus Grant were the first Europeans to reach the king-
dom of Buganda and that each of them wrote a book describing his experiences.
But there is a tendency to believe that the two, in so far as Central Africa was
concerned, merely flitted across the stage and were afterwards no more seen nor
heard; to regard them rather like the cosmopolitan globe trotter, who goes to
London, visits the principal sights, and then returns to his own country with no
more interest or affection for that city than he has for Paris, Rome, Vienna or
any other capitals to which his sightseeing has taken him. Speke's untimely
death in 1864 prevented him from giving many proofs of his continued interest
in the peoples of the lands which he had discovered, but none the less, as I will
endeavour to show, he had a very real interest in them. Grant outlived Speke by
many years, and as will be seen, right up to the day of his death, his interest in
the peoples of Uganda never waned.
Amongst the archives now in the custody of the Peace Memorial Museum at
Zanzibar is a bound letter book, containing original correspondence, which is
labelled "Captain Speke's Expedition". It is disappointing, however, to have to
add that the label also bears the words "1863-4, vol. 3". What has become of
vol. 1 and vol. 2 I know not. Vol. 3 is mainly concerned with matters connected
with the winding up and settling of the purely business affairs of the expedition,
but it also contains some interesting personal letters addressed by Speke to his
brother officer, Captain (afterwards Colonel Sir) Robert Lambert Playfair, who
was at that date British Consul at Zanzibar.
The earliest of these letters was originally dated "Gondokoro 26th February,
1863", but this was altered later to "Khartum, 30th March". It was addressed to
"H.B.M.'s Consul, Zanzibar". It enclosed a list of forty-two names with the
statement that the persons in question had been recruited at Zanzibar and had
deserted from the expedition in the heart of Central Africa, eighteen of them
with guns. The writer asks the Consul to take measures to effect their arrest
with a view to punishment.
Another letter of 30 March 1863, from Khartum is addressed to Playfair's
predecessor in office, Colonel (afterwards Major-General) C. P. Rigby,1 "or
whoever is in office now". The letter deals largely with business matters and asks
for news from Zanzibar. It then proceeds as follows:
"I have 19 men with me who will receive pay, a present and wife each,
and a Free-men's Garden for having accomplished the whole journey to
Egypt. They all know the run-aways and could detect them. In proportion
1 Other letters addressed by Speke to Colonel Rigby, who was British Consul at Zanzibar,
1858-61, will be found in General Rigby, Zanzibar and the Slave Trade (1935), by his
daughter, Mrs. C. E. B. Russell, pp. 233-41.

as I am rewarding these men I wish the others to be punished and I think
the best way of doing so would be to put them in prison for three years with
hard labour, which labour might be turned to the benefit of the Faithfuls,
tilling their garden, etc., but more of this when I arrive at Cairo and send
the Faithfuls home to Zanzibar. At present I have limited the garden to
1,000 $, not knowing the value of the land, but I should be glad if you would
inform me how many acres it would purchase, when I might increase it, if
not sufficient to keep them on an equality with free men, which is my aim.
The terms will be framed at Cairo."
On 29 May 1863, Speke sent to the British Consul a "Nominal Roll of
Bombay's Faithful children who reached Cairo entitled to two years pay" with
instructions as to the payments they were to receive on arrival at Zanzibar and
a request that land might be purchased for them to be held by them as "a Free
Men's Garden". The paper also sets out the conditions upon which they are to
enjoy the use of that land.
Another paper of the same date is headed "Nominal Roll of Baraka's unfaith-
ful children, deserters, entitled to three years' imprisonment". The roll contains
twenty-six names, of whom all except two are stated to have deserted in
Bunyoro. Twenty-one of them are said to have carried off a carbine apiece and
one a spear.
Those who wish for fuller particulars of Bombay's Faithfuls and Baraka's
Unfaithfuls will find them in Appendix A to Speke's Journal of the Discovery
of the Source of the Nile, but many readers may well fail to realize from the
pages of that book how very close to disaster the expedition was brought by
these wholesale desertions in Bunyoro and at other earlier stages of its journey.
This is what Speke has to say about this matter in a note appended to the list:
"Baraka, father of the deserters, tried to frighten all my men from advancing in
Sorombo,' but failing to break down the expedition then looked into the
Uganga;2 was honoured and not dismissed only because he held influence over
the men, which by his own shameless acknowledgement was sufficient, had he
chosen to exert it, to leave me manless in the country only with such "infatuated
fools" as Bombay, etc., who had sold their lives with ours. He was finally sent
from Karague to Kamrasi's in Unyoro in company with Uledi, where, having
obtained my permission, he received a tusk of ivory as large as he could carry,
but never joined the expedition after. His pay, however, with his companion's
Uledi, was sent to him by Msalima from Unyoro by an order on the British
Consul dated 30 October 1862, which also included Msalima's pay, the three
being separately, Baraka 100 $, Msalima 60 $, and Uledi 50 $, and which is more
than they deserved, excepting Uledi who was always accounted a very steady
honest working man".
"I have termed the above list of deserters Baraka's Unfaithfuls, because they
never ceased to mutiny until they ran away and Baraka could never have gone
on beyond Bogue,3 had not Bombay sided with me and put him to shame. This
is the opinion of every man composing the expedition. So strong indeed were
1 In Uzinza territory between Unyanyembe and Karagwe.
2 Sc., had recourse to witchcraft.
3 Also in Uzinza.

Baraka's opinions with these men, that they no sooner heard of Baraka's return
to Karague than they refused to enter Unyoro at all, and had Mtesa, King of
Uganda, not ordered them to follow me, I believe I never should have been able
to keep them together so long as I did. I sent a letter after the desertion suggest-
ing Mtesa to seize all my carbines and transport all the faithless wretches to an
island in the Lake until such time as he might hear further from me."
As Speke's Journal shows, Baraka and Bombay were constantly quarrelling.
In the circumstances therefore it may be that not all the allegations made against
Baraka and believed by Speke were entirely true. But there is no doubt that
Baraka was more than once most insubordinate. In a letter to Playfair, Grant
says he "was very smart, but too cheekey". It was almost inevitable that in the
end Baraka would refuse to accompany his employers any further and should
try to induce others to follow his example. Those, who were persuaded by him
or were influenced by his example to desert, carried off more than half the
expedition's guns just at the moment when Speke and Grant were about to enter
the territories of the Nilotic tribes to the north of the Victoria Nile, who had a
reputation for savagery and hostility to strangers. In the circumstances it is not
surprising to learn how strongly Speke felt about Baraka's conduct. To a
military man such conduct was tantamount to incitement to desertion in the face
of the enemy and deserving of punishment as such.
Some of these deserters doubtless took service with local rulers, who were
only too glad to enlist men in possession of such a valuable weapon as a firearm.
Others appear to have set up in business in these inland regions, making use of
their stolen weapons to procure ivory and other merchantable goods. Twenty-
four years later, when Stanley was proceeding up the river Aruwimi in the
Congo to the relief of Emin Pasha, he found the man Uledi, whom, as already
mentioned, Speke has described "a very steady, honest working man", estab-
lished in business as a slave and ivory trader under the name of Uledi Balozi
(the Consul's Uledi) alias Ugarrowwa. Readers of Stanley's In Darkest Africa
and Parke's My Personal Experiences in Equatorial Africa will learn in detail of
the abomination and desolation which he was then causing in those regions.1
In course of time a homing instinct took others of the deserters, including
Baraka, back to Zanzibar. Doubtless they hoped that the fact that they had
deserted Speke would either be unknown, or at any rate go unpunished. But that
was not the case. On 14 February 1864, Playfair reported to the Bombay Govern-
ment that eleven of them had reached Zanzibar and that, "as I felt convinced
that no subsequent traveller would be able to penetrate Africa with any degree
of safety, should these men escape without punishment, I laid a statement of the
case before His Highness (sc., the Sultan of Zanzibar), who condemned them to
imprisonment during his pleasure. He further informed me that he would not
release them till I should intercede on their behalf." A resolution of the Bombay
Political Department informed Playfair that his proceedings were approved.
But if Speke felt vindictive towards these deserters, his gratitude to the Faith-
fuls was unbounded. Their names can be found in the Appendix to Speke's
1 On 27 Rabi-el-Akhir, 1306 (31 December 1888), Uledi, "freed slave of the British
Consulate", sent a letter reporting his dealings with Stanley to the British Consul at Zanzi-
bar. A copy of the letter is in the Zanzibar Archives and a photostat thereof in the Zanzibar
Peace Memorial Museum.

Journal and their portraits are to be found (reproduced from a photograph
taken at Cairo), facing p. 611 of that work.' At the foot of the next page of that
book are to be found the portraits of the four women who also travelled from
Bunyoro to Cairo. Two of these were wives of the headman, Bombay, who is
seated in a chair of honour in the group already mentioned, and the other two
were wives of other Faithfuls.
True to his promise, Speke sent money to Playfair to purchase land for them
in Zanzibar and those of the Faithfuls, who decided to remain in Zanzibar, were
put into joint possession of the shambas at Bububu and Ziwani on the outskirts
of the town. Subsequent correspondence shows that Speke was constantly
inquiring after "my poor men", sending them presents and asking Playfair in
what other way he could help them. After his untimely death Speke's brother
also wrote to ask Playfair "if there is anything you may think he would have
wished to have further done with what he called 'his Faithfuls' ".
In a letter of 17 November 1863 to Playfair, Grant says "Speke has merely
mentioned having heard from you that they were generally drunk every day or
enjoying their return like true Africans". On 7 February 1864, Speke himself
felt constrained to write the following apologetic remarks in regard to Bombay:
"I was delighted by yours of the 27th Septr: to find you had made such v.
good arrangements for my children, though extremely sorry to hear that
Bombay's head had been turned by his elevation. On the journey I had often
occasion to wig him for getting drunk, but always made it up again, as he
was the life and success of the expedition. You will see by the book what he
did for me."2
In 1864 the Royal Geographical Society sent a silver medal for Bombay and
seventeen bronze medals for distribution amongst his companions. Whatever
the failings and shortcomings of each one of them may have been they fully
deserved all the praise that Speke bestowed upon them and it was altogether
fitting that they should receive from the Royal Geographical Society this recog-
nition of the invaluable and loyal aid which they had rendered to the service of
exploration and discovery. Speke's enthusiasm for them and their devotion have
proved infectious and I have tried to follow up their after careers. It is interest-
ing to learn that not a few of them afterwards served with other explorers such
as Livingstone, Von der Decken, Cameron, Stanley and Joseph Thomson. Three
of them-Ulimengo, Mabruki and Manwa Sera-were later to be awarded
another medal by the Royal Geographical Society for their devotion to Living-
stone on his last journey. Many an old campaigner would feel proud to have
earned the right to wear medals for services such as were rendered by these few.
1 "Since my arrival I have had all my men photographed by a French man. They were
not well done but will do to form a frontispiece to the book, which plague be to all the
pigs, I must write"-Speke to Playfair, Cairo, 31 May 1863.
2 Bombay subsequently served Stanley in the Livingstone Relief Expedition and
Cameron in his journey across Africa. Neither of these explorers had the same high opinion
of him as had Speke and it is to be feared that he did become somewhat swollen-headed in
his later years. For a few months in 1876 he was with the C.M.S. missionaries, Shergold
Smith and Mackay, on the exploration of the Wami and Pangani Rivers; but he was getting
past work and thereafter retired having been awarded a small pension by the Royal
Geographical Society, which he enjoyed until his death.

In addition to settling the affairs of his Faithfuls and the winding up of matters
connected With his expedition Speke's letters to Playfair are full of his plans for
the further exploration of the Dark Continent and for the opening up of Africa.
On 18 February 1864, a meeting was convened at the house of the Marquess
Townshend at No. 6 Grosvenor Place, London, to which fifty persons "inter-
ested in the cause of Africa" were invited to attend so as to hear Speke expound
his scheme "for opening Africa". Speke circulated in advance two printed
handbills respectively entitled Considerations for opening Africa and Scheme
for opening Africa.
In the Considerations Speke agreed that the experiences of Livingstone, him-
self, and Arab traders had shown that there was a belt of fertile land in
equatorial Africa, which "is the only part of Africa worthy of serious considera-
tion" and that European countries should concentrate on opening this particular
"To reach this land of fertility, at present the best route is via Suakin on the
Red Sea to Berber on the Nile, and thence up the Nile which is navigable to
within 300 miles of the equator, as it soon will open an extensive trade with
Egypt, and the facilities of transport are easy."
"To those regions especially in the kingdoms of Unyoro, Karague and Uganda
all Church Mission Societies should especially direct their attention, as they will
find Wahuma Kings, descended from the ancient Christian Abyssinians, power-
ful enough to protect them, richness in soil sufficient to support them, and a
people affluent enough to spare their children to be educated. These three
specialities have particularly to be regarded, as it was from want of them our
Missions have hitherto failed."
"Now with regard to the slave-trade. Our present system for suppressing it has
failed because we have not made use of the Negroes. Indeed we have done a
previous (sic) harm to the cause we are advancing, by running down slaves. No
man in the interior of Africa knows what we are doing. The native chief finds the
price of slaves rise by the demand, and sets to work to meet that demand by
fighting his neighbours. The horrors of the middle passage are by this rendered
light in comparison with the fearful devastations committed in the interior. An
example of what I mean may be instanced in a few words. A few years ago, the
Zanzibar ivory-traders frequented the line leading from Kilwa to Lake N'Yassa,
and found friendly nations and plenty of provisions wherever they went. Since
then, the slave-trade having been diverted from the west coast to the east, that
line has been completely ruined owing to the incessant wars carried on between
the native chiefs, who fight against one another for slaves."
The Scheme for opening Africa set out eight 'propositions' which were
designed to meet the points raised in the Considerations and to "turn our
African experiences to good and useful account". The propositions were as
"1. That our Government be petitioned to use its influence, conjointly
with the Egyptian Government, to suppress the illegitimate tendencies of
the White Nile trade; and, for that purpose, to establish an alliance with the
Bari Negroes at the foot of the cataracts above Gondokoro,"
"2. That our Government be petitioned with the European Govern-

ments, to support a United Church Mission, to be sent via Suwakin on the
Red Sea to Berber on the Nile, and thence up the Nile to the foot of the
cataracts above Gondokoro; whence, by land, they would march up the
Nile to the kingdom of Unyoro,"
"3. That the Missionaries composing the Mission be selected not for the
purpose only of preaching to the Wahuma, but for general instruction; and
that they shall be bound to be self-supporting after the first two or three
years, or until a trade can be instituted with Egypt,"
"4. That after a certain time, and (sic) the King of Unyoro can see and
understand that legitimate trade is the best thing for the maintenance of his
Government, and the prosperity of his people, detachments of Missionaries
should be sent further on to the kingdom of Uganda and Karague,"
"5. That our Government be petitioned to make arrangements with the
Sultan of Zanzibar, to put a full stop to the slave-trade in his dominions,"
"6. That our Government be petitioned to recognize all persons con-
victed of taking part in slavery as conniving at murder, and to treat them
accordingly; for without bloodshed slaves cannot be caught,"
"7. That our Government be petitioned to form a chain of Negro Depots
round the East and West sides of Africa, in sufficient numbers to half man
our men-of-war, and yet to have a strong reserve at each depot; who shall
all be educated and brought up for the holy purpose of liberating their fellow
country men from the thraldom of slavery; as it is obvious that the great
sums of money now spent, with a view to suppressing slavery, are doing
more harm than good,"
"8. That, as much as possible, Negroes should be educated and
employed in all British Services, an- taught to abhor the slave-trade, which
they have hitherto been taught to consider legitimate from the fact that
they are purchased with European articles of merchandise."

I have not been able to find any reference to this meeting in the contemporary
press. Speke was a very poor and unready speaker at public meetings and per-
haps he failed to make a good impression. Possibly his audience showed a polite
interest in what he had to say and then departed, fully convinced that something
ought to be done in regard to the opening of Africa by somebody other than
themselves. But it is interesting to note that over eleven years before Stanley's
well-known call for missionaries for Uganda was published in the Daily Tele-
graph John Hanning Speke had at this meeting in London urged the sending
of missionaries to the Court of Kamurasi. In a book Tramps round the Moun-
tains of the Moon, (1908), a missionary, the Rev. T. B. Johnson, wrote of
Stanley as "a later English traveller with a nobler ideal than that of Speke". It is
far from wise to pass sweeping condemnations on persons without full inquiry.
The author in question failed to grasp the fact that two of the many differences
between Stanley and Speke were that the former had almost unlimited funds at
his disposal and a good press behind him, whereas Speke had neither of these
things. There was another difference besides this to which I shall refer later.
Speke never returned to Africa. On 15 September 1864, he was out partridge
shooting on his uncle's land at Neston Park, Bath. He proceeded to climb over

a stone wall with his gun at full cock. It was one of those momentary acts of for-
getfulness which can overtake even the most wary. One barrel was discharged
into his body and he died a quarter of an hour later. In Sir Reginald Coupland's
words, "in the course of a few years he had proved himself one of the greatest
and noblest of African explorers". (Exploitation of East Africa, 1939, p. 109).
His books are modestly and attractively written. His correspondence reveals a
very simple and unaffected and a very lovable personality. Letters from brother
officers in the Indian Army show that he was extremely popular and held by them
in high esteem. When news of the tragedy reached Zanzibar, Bombay wanted to
go on a pilgrimage to his master's grave.

"You have had a long walk, Captain Grant." Lord Palmerston, that irrecon-
cilable enemy of the slave-trade, was addressing Speke's travelling companion,
when the two were introduced at a meeting of the Royal Geographical Society.
Captain Grant agreed with his Lordship and made Palmerston's saying the
leading title of his book. Like Kirk, Mackay, and so many other early pioneers
in East Africa, James Augustus Grant was a son of the manse. He was born at
Nairn, of which place his father was a parish minister, on 11 April 1827, and
was thus only twenty-three days older than Speke. Grant joined the Indian Army
in 1846. He tells us that
"My acquaintance with Captain Speke commenced as far back as 1847,
when he was serving in India with his regiment. We were both Indian
officers, of the same age, and equally fond of field-sports, and our friend-
ship continued unbroken. After his return from discovering the Victoria
Nyanza, he was, as is well known, commissioned by the Royal Geographical
Society to prosecute his discovery, and to ascertain, if possible, the truth of
his conjecture-that the Nile had its source in that gigantic lake, the
Nyanza. I volunteered to accompany him; my offer was at once accepted;
and it is now a melancholy satisfaction to think that not a shade of jealousy
or distrust, or even ill-temper, ever came between us during our wanderings
and intercourse."
Before going to East Africa Grant had seen service in the Sikh War (1848-9),
and throughout the Mutiny, having been wounded in the operations for the
relief of Lucknow. After his return from his walk across Africa he served in the
intelligence department in Lord Napier's campaign in Abyssinia. After that cam-
paign he retired from the service with the rank of Lieutenant-Colonel.
Sir Samuel Baker first met both Speke and Grant at Gondokoro in the last
stages of their journey in 1863. He was to come into contact with Grant more
than once in later life. About a year after Grant's death he wrote this of him:
"Speke was a splendid fellow in every way .... Grant (his companion)
was one of the most loyal and charming creatures in the world. Perfectly
unselfish, he adored Speke, and throughout his life he maintained an
attitude of chivalrous defence of Speke's reputation ... He was the most
unselfish man I ever met: amiable and gentle to a degree that might to a

stranger denote weakness, but, on the contrary, no man could be more
determined in character or unrelenting when once he was offended."'
On his journey with Speke, Grant took with him a photographic apparatus,
but the tropical heat of the dark tent in which he had to work rendered him so ill
that he had to abandon photography. He then took to sketching and many draw-
ings from his sketch book have been reproduced in Speke's Journal. Many of
these sketches are surprisingly accurate and might be accepted as faithful repro-
ductions by an artist of to-day of the same scenes. Sir Albert Cook once pointed
out to me that the sketch drawn by Grant, entitled "View of the Murchison
Creek", which is reproduced on pp. 390-1 of Speke's Journal, might well have
been drawn only the other day from just outside Sir Albert's house at Makindye.
Grant also painted in water colours. Sir Harry Johnston says he "did more than
anybody else, down to a quite recent date, to put before our eyes some idea of
the beautiful coloration of African wild flowers". He published "The Botany of
the Speke and Grant Expedition" (1872), in vol. 29 of the Transactions of the
Linnean Society. "Although most of these flowers were drawn for him by
scientific draughtsmen, his own sketches supplied means for an accurate colora-
tion which could be no longer ascertained from the dried specimens. In this
particular Grant has made an important contribution to African research."
(Johnston, op. cit. pp. 132-3).
Until his retirement from the Army, Grant's military duties kept him fully
occupied and he had little time for inquiring into African affairs. Only one letter
of his appears in Playfair's letter book, in it Grant inquires after Bombay and
the other Faithfuls and sends 100 for distribution amongst them.
In 1872 with some hesitation he complied with a request from Sir Roderick
Murchison that he should contribute an article to the Journal of the Royal
Geographical Society (vol. 42 (1872), pp. 243-342). That article has the some-
what cumbersome title of Summary of Observations on the Geography, Climate
and Natural History of the Lake Region of Equatorial Africa, made by the
Speke and Grant Expedition 1860-1863. It contains some information which is
not to be found in My Walk Across Africa or either of Speke's books. In that
same year an American subject, Henry Morton Stanley, returned to England
from his expedition in search of Livingstone. He had found Livingstone and had
brought him much needed assistance at an opportune moment. It had been a
great achievement and, let it be said, that English public opinion had been over-
grudging in giving it a due measure of praise. But Stanley himself was largely to
blame for this. His own complacent description of some of his exploits had quite
rightly shocked that same public opinion. In particular, as he explained at length
in Chapter 8 of his How I found Livingstone, Stanley had assisted the Arab
slave traders of Unyanyembe in a war against the local chief, Mirambo. English
public opinion knew enough about slavery and the slave-trade in Africa to
regard such conduct on the part of a European explorer with decided aversion.
With his first-hand knowledge of the consequences of these tribal wars, Grant
raised his voice in protest in his Summary of Observations
1 Baker to T. Douglas Murray, 22 August 1895; printed in Sir Harry Johnston, The Nile
Quest (1903), pp. 131-2.

"I would therefore here emphatically protest against any foreigner taking
part in the fights of a country through which he may be travelling for
scientific and other information, for I feel confident that his neutrality is
his truest and safest course."
There is one other passage in the Summary which should be quoted, as it
shows that, despite his military duties, Grant had not lost his interest or touch
with African affairs.
"The Mahomedan Government of Egypt is, I regret to say, extending its
influence in those parts by large acquisitions of territory; and it becomes its
duty to control the desire of its subjects to make themselves masters of the
slaves of Abyssinia and Central Africa. Annexation by this power would
be a serious evil; and, for the sake of the fine and independent races of
Uganda and Karagwe and their fertile country, I hope and trust that civiliza-
tion may be introduced among them by Christian, not by Mahomedan
races, who would turn the whole country into a market for slaves. The trade
of the east coast of Africa is being developed more rapidly now since the
opening of the Suez Canal (sc. in 1869), but the interior should be penetrated
to obtain its rich products, and foreign traders should push on from the east
coast to Egypt, protecting the people from Mahomedanism, forming trading
depots at different points and showing their intolerance of slavery."
Those words were written in 1872. Though at the time he wrote them Grant
could not have known of the event, on 14 May of that year Sir Samuel Baker had
proclaimed the annexation of Bunyoro to Egypt, in pursuance of a fiiman of
the Khedive. Grant's fears for the future of the Lake Regions were therefore by
no means idle.
In 1874 Stanley returned to Africa and started on his first journey across the
continent. Periodically letters from him reporting his progress were published
in the Daily Telegraph. On 29 November 1875, Grant read a paper to the Royal
Geographical Society on Mr. H. M. Stanley's Exploration of the Victoria
Nyanza. (Journal R.G.S., vol. 46 (1876), pp. 10-34). In it he paid a generous
tribute to Stanley's achievement declaring it to be "one of the most important
and brilliant that has ever been made in Central Africa, or, indeed in any
country". At the same time he had perforce to express his disapproval of the
punitive measures taken by Stanley against the people of the Island of Bumbire.
Stanley had once again done that very same thing which Grant had deprecated
when addressing the Society three years before. "I think," he wrote, "that Mr.
Stanley committed a serious mistake in making this attack and in meddling with
the old quarrels of his allies-the Waganda-for such demonstrations leave their
mark, and check the prospect of friendly intercourse for years." Nevertheless he
fully endorsed all that Stanley wrote about Uganda. Calling it "this bright spot
on the Equator", Grant said, "I can echo his sentiments by adding that it
deserves a kind protection and a happy future".
On 15 November 1875, exactly a fortnight before this paper was read the
Daily Telegraph published Stanley's well known letter appealing for Christian
missionaries to be sent to Uganda. The Church Missionary Society took up the
challenge and set to work to equip a special expedition to be sent to Mutesa's

capital. When organizing that expedition, they very naturally got into touch with
people who had some personal knowledge of the regions to which the mission-
aries were to be sent. These included Colonel S. E. Gordon, brother of Colonel
Charles Gordon, and Lieutenant (afterwards Colonel Sir) Charles Watson, who
had served under Gordon in the equatorial regions of the Nile. Needless to say,
Grant was also consulted. Gordon's and Watson's personal knowledge of the
means of approach to Lake Victoria was by way of Khartoum and it is therefore
not surprising that they advised that the missionaries should proceed by the
Nile route. On the other hand, Grant, who knew both the Nile route and the
route from the east coast, "urged strongly that, as Colonel Gordon was regarded
in Central Africa as an invader, a Mission approaching Uganda from north-
ward under his auspices would seem identified with a policy of annexation;
whereas the liberal tendencies of the Sultan of Zanzibar, who had that year
visited England, and was really doing his best, under Dr. Kirk's influence, to
suppress the slave-trade and to foster the growing commerce between his
dominions and the interior, pointed to the East Coast as a natural starting point
for an expedition to the Lake regions ...."
"..... Colonel Grant also urged, in view of the probable uncertainty of
Mutesa's character, that the Mission should aim first at the kingdom of Karagwd,
to the west of the Victoria Nyanza, where he and Speke had found the king,
Rumanika, of a singularly noble and gentle disposition."'
Once again, Grant was not making a wild statement in regard to the threat to
the Lake Region from the north. Baker had been called upon by the Khedive to
perform a superhuman task. He had been told to suppress slavery, but had found
his Egyptian troops to be worthless and his Egyptian officials to be working
hand in glove with the slave-traders. In 1873 he had resigned in despair. In the
following year the Khedive had appointed Colonel Gordon in his place. On
22 August 1875, Gordon had written to Sir Henry Rawlinson, President of the
Royal Geographical Society, informing him that in furtherance of the Khedive's
aims he proposed to "continue his march up the Nile (Speke's Somerset River)
to Foweira and Rionga, above the Karuma Falls, which posts he would
strengthen for defensive purposes, using them as a base for further operations
against Kabarega at M'Rooli, and ultimately, if necessary, against M'tesa of
As Edward Hutchinson, Lay Secretary of the Church Missionary Society,
wrote at the time, "in spite of Colonel Gordon's pacific desires and intentions,
the gradual advance of the Egyptians is regarded with suspicion.... Public con-
fidence has been much shaken in the supposed enlightened views of that Govern-
ment by recent events".3
It was therefore decided to accept Grant's advice and to send out the first
missionary party by the east coast route, but it was also resolved to keep the
precise locale of the Mission an open question to be decided by the missionaries
themselves when they reached Lake Victoria. As is well known, the advance
party, consisting of Lieutenant Shergold Smith and the Rev. C. T. Wilson,
1 Eugene Stock-The History of the Church Missionary Society, III (1899), pp. 96-7.
2 Proceedings of the Royal Geographical Society, vol. 20 (1875-6), p. 53.
3 Hutchinson-The Victoria Nyanza, a Field for Missionary Enterprise (1876), pp. 59, 61.

reached Rubaga on 30 June 1877. Grant had greatly assisted the party with his
advice as to equipment and many other matters of vital importance to an
expedition setting out at that date for the heart of Africa, but that was not all.
On 17 December 1875, he gave a donation of 100 to the Victoria Nyanza
Mission Fund.' Thereafter he maintained a constant interest in the activities of
the Church Missionary Society in Uganda.
Shortly after the missionary party set out for East Africa Grant was filled once
more with misgivings in regard to the attitude of the Egyptian Government
towards Uganda. At the end of 1876 Gordon had tendered his resignation to the
Khedive and departed for England "with the sad conviction that no good can
be done in those parts, and that it would have been better if no expedition had
ever been sent".2 But pressure was brought to bear on him in both England and
Egypt to withdraw his resignation. As Gordon told Burton, "the whole of the
future exploration, or rather opening, of the Victoria Lake to Egypt has not a
promising future to me, and I do not like a bit the idea of returning. I have been
humbugged into saying I would do so, and I suppose I must keep my word".'
When the news that Gordon was returning to Egypt was known, Grant wrote
a letter to The Times, dated 30 January 1877; It opens with the following
"The letter in the Times of to-day on Colonel Gordon's position with the
Khedive of Egypt hits the mark exactly. The chief object of his return is
the annexation of the Victoria Nyanza. All who know Colonel Gordon
must feel the highest confidence in his administrative power; but what will
become of the country as soon as he leaves it."
"On this ground I would strongly protest against the occupation of any
portion of the Lake by the Khedive, as, if it be annexed, it would remain
a thorn in the side of civilization, and immensely increase the difficulties in
the suppression of slavery."

In a letter to Burton, written during a fit of impatience at any criticism of, or
opposition to, his plans Gordon rather disagreeably referred to Grant as "that
old creature Grant, who for seventeen or eighteen years had traded on his
wonderful walk".4 But Gordon knew neither Buganda nor Mutesa, whereas
Grant knew both and continued to raise his voice against any attempt at annexa-
tion of the country by Egypt. Thus, on 3 June 1878, at the close of Stanley's
lecture to the Royal Geographical Society on his latest journey across Africa,
Grant once more called attention to this danger and said:
"Mtesa was in great hopes that he could get assistance from England
from the east, and be able to hold his own against any aggression from the
north. He was sure that, if Mtesa was helped by England in any way, he
would assist in opening up the whole of the country, so that the southern
routes might be connected with the Egyptian routes."5
I Hutchinson-ibid (1st edition), pp. 105-7.
2 Birkbeck Hill-Colonel Gordon in Central Africa, p. xlii.
3 Wilkins-The Romance of Isabel, Lady Burton, pp. 652-3.
4 Gordon to Burton, 19 October 1877-Wilkins op. cit., p. 661.
5 Proceedings of the Royal Geographical Society, vol. 22 (1877-8), p. 409.

"That old creature Grant" was not by any means the only person who lifted
up his voice against Egyptian aggression in central Africa. The Aborigines'
Protection Society, the British and Foreign Anti-Slavery Society, and the Church
Missionary Society each sent a memorial on the subject to the British Govern-
ment. Dr. Kirk, British Consul-General at Zanzibar, made his protest, as did a
number of other prominent persons, in the columns of the British press.
Eventually the British Foreign Office was induced to make representations to
the Egyptian Government with the result that the latter Government eventually
gave an assurance that they "have no intention whatever of making any further
annexations in Central Africa nor of encroaching upon King Mtesa's
In the meantime Grant had received an interesting letter from the Rev. C. T.
Wilson. It ran as follows:
"I am forwarding to you a Uganda shield and two spears, which Mtesa
has asked me to send to you with his salaam. You will be pleased to hear
that the prospects of the Mission there are very hopeful. Mtesa is very
anxious for Englishmen to come, and says he wishes to have no more deal-
ings with the Arabs. These latter have of course opposed us very strongly,
and have tried to poison Mtesa's mind against us with all sorts of stories.
Mtesa is terribly afraid of the Egyptians and looks with the utmost
suspicion on anything and everything connected with them, which I fear
will greatly retard the opening up of the Nile for traffic. On the other hand,
Mtesa is desirous of opening up a road to the East Coast through the
The present of the shield and the two spears is interesting. Speke had received
the like present at the time of his and Grant's visit to Buganda. It was something
more than a mere courtesy present. It was the presentation of what was really
the insignia of an order of merit, which was bestowed by the Kabaka only upon
a privileged few. On 8 December 1893, Colonel (afterwards Major-General Sir)
H. E. Colvile tried to obtain official recognition for the order. He wrote to the
Acting Consul-General at Zanzibar recommending the institution of such an
order for Baganda chiefs. He submitted a sketch of the proposed insignia, con-
sisting of a Ganda shield with three spears, two crossing each other and the
third standing upright behind the shield. In this letter Colvile stated that the
design was based on a flag formerly flown by Mutesa. Nothing came of the
proposal at the time, but in 1931 the late Kabaka instituted an Order of the
Shield and Spears with somewhat different insignia which is bestowable upon
his subjects for meritorious and distinguished service.3 There can be no doubt
that the sending of this shield and these spears to Grant was clearly intended by
Mutesa as a very high honour indeed.
On 26 June 1880, Grant once more renewed his personal acquaintance with
people from the land which he and Speke had explored eighteen years before.
I H C. Vivian to Lord Salisbury, 13 June 1879 (Zanzibar Archives).
2 The letter is printed in Proceedings of the Royal Geographical Society (New Series),
vol. 1 (1879), p. 138. Its date is not given, but it was written by Wilson whilst at the southern
end of Lake Victoria at the end of 1877.
3 Twining-'Uganda Medals and Decorations', Uganda Journal, vol. 2 (1934-5), p. 222-4.

These were the three Baganda envoys whom Mutesa had sent on a mission to
Queen Victoria. They had arrived in England under the escort of Rev. C. T.
Wilson and Dr. R. W. Felkin of the Church Missionary Society. Their spokesman
was one Namukade, who held a small chieftainship in Buganda. His com-
panions were subordinate chiefs under him and were named Kataluba and
Magigo.' On 14 June 1880, in company with Wilson and Felkin, and Edward
Hutchinson, Lay Secretary of the Church Missionary Society, they were received
in audience by Queen Victoria at Buckingham Palace. Namukade presented a
letter from Mutesa to the Queen as well as a number of presents. Unfortunately,
I have not been able to find any record of Her Majesty's impression of the
envoys. Mackay has given us those of Magigo.2 When Magigo and his com-
panions found themselves in the presence of the members of the Royal
Geographical Society, Wilson had to explain that "he was afraid they were a
little overawed at finding themselves before such an august assembly", but
Hutchinson said he "could testify to their power of adapting themselves".3 They
returned to Buganda in the following year, but their journey was barren of any
permanent results. The stories of the envoys were regarded with the utmost
incredulity. Namukade's son tells us "people were suspicious and said that
Namukade never went to Europe; perhaps he came back whilst on the way
there. But recently Sir Apolo Kagwa, K.C.M.G. Katikiro, assured us that on
his journey to England he slept in the very same house as Philip Namukade".4
Thus more than twenty years later the veracity of the envoys was at long last
When Mutesa died in 1884, Grant wrote for the Royal Geographical Society
a short obituary notice in which he stressed the fact that the deceased ruler "has
proved to be a true friend to us geographers".5
Naturally, Grant was very much distressed by the events which occurred in
Uganda shortly after Mutesa's death. He was also alarmed by German activities
in East Africa and their possible repercussions on missionaries and explorers in
those regions. On 4 November 1888, he addressed a long letter to Lord Salis-
bury, in which he referred to a letter from Pere Lourdel which had been pub-
lished in The Times five days previously and which had emphasized the fact that
Arab slave-dealers in Uganda had been inciting Mwanga against English and
German alike. In his opinion not only were French and English missionaries in
Buganda in very great danger, but so was Stanley's expedition for the relief of
Emin Pasha, which would, it was believed, be making its way at an early date
down to the East Coast. He declared that the Germans "have shown great
impatience and blundering in their first attempt to colonize in East Africa. My
1 Magigo appears in Mackay's letter and in the Royal Geographical Society's records
as Sabaddu, which was in fact only the title of his chieftainship.
2 Mackay of Uganda by his sister, Chapter 7.
3 Proceedings of Royal Geographical Society (New Series), vol. 2 (1880), pp. 366, 383.
4 Yosefu Kazamiti, Namukade's eighth son, wrote an account of his father's journey,
which appeared in Munno (1921) pp. 184-5 and (1922) pp. 53-5. I think the sending of
envoys of relatively humble rank was because Mutesa did not care to expose more
important chiefs to the risks which he believed to attach to a mission into an unknown
land. He therefore chose in accordance with his usual practice "slight unmeritable men
meet to be sent on errands". A further account of the envoys' visit to England can be
found in the C.M.S. Gleaner (1880), pp. 62, 91.
5 Proceedings of Royal Geographical Society (New Series), vol. 7 (1885), p. 338.

suggestions to any Power who wished to colonize in Central Africa would be to-
adopt a method more practical, pacific, and inexpensive, one which would be
of ultimate benefit to the native and the Europeans."
The man, who penned those words, had shown that he could practise what
he preached. One morning thirty years before whilst making his way along the
western shores of Lake Victoria, he was told that "the natives were up in arms".
On learning this, "I ordered the guns of the Seedees to be filled with shot-sized
pebbles instead of bullets but", as he wrote with relief, "we did not require to
fire them". There was a parley with the natives and eventually Grant learnt that
his escort had been guilty of kidnapping two children, whom he at once restored
to their parents. (A Walk Across Africa, pp. 211-2.) It was acts of restraint such
as this which earned him the very high and lasting reputation which he enjoyed
in Central Africa and which also gained for him the lifelong respect of Mutesa.
Grant died at his birth place, Nairn, on 11 February 1892. Like his companion
Speke, he had not been found wanting in his interest in the welfare of the
countries through which he had once passed, but had made an honourable con-
tribution to the furtherance of their well-being.

It is customary to think of Stanley's appeal for a Christian mission to be sent to
Kabaka Mutesa of Buganda as having come to the notice of the Church Missionary
Society in November, 1875, as an inspired bolt from the blue.
Sir John Gray here shows that Speke had ventilated such an idea more than 10
years before during the short period between his return to England in 1863 and his
sudden death.
It transpires that Speke had sought to interest the C.M.S. in Uganda as early as
1861, some months before he had even reached the confines of Mutesa's kingdom.
Thus the notion must have been within the purview of the inner councils of the
C.M.S. some fourteen years before Stanley's letter was published.
Through the interest of Mr. H. S. Cobb, the Archivist of the Church Missionary
Society, and with the permission of the Society we are able to print below the text of
a holograph letter from Speke to Colonel Michael Dawes, R.A., the Lay Secretary
of the Society which has recently come to light among the Society's records.
Colonel Dawes was known personally to Speke. Both had taken part in the Sikh
War of 1848-9 under Sir Hugh Gough and both were present at the battle of
Chilianwallah. After a distinguished military career including service in the Indian
Mutiny, Colonel Dawes retired from the Army to become Lay Secretary to the
C.M.S., 1859-66; and died in 1871.
One may properly remark on the good will and the complete absence of aggressive
intent with which Speke approached this then mysterious kingdom in the heart of
Central Africa.
10 March 1861.
You will probably be surprised to see my hand writing, but I have a crotchet in
my head which cannot be worked out without your instrumentality and even with it
I do not know how the matter is to end. In two or three days I shall probably start

for (the country of) Uganda, where there is a powerful Monarch governing a multi-
tude of men in the very heart of Africa.
With this King I expect to fnake great friends, appearing in his country as a
British Agent come to open up the land to commerce and civilization. The King's
name is Mtesa, he is young, but, I believe, reasonable and desirous of knowing a
little of the handiworks of the outer world.-I think therefore it might suit my ends,
that is to say the discovery of the source of The Nile, and a passage by it to England,
if I proposed taking one or more of the Princes of Uganda with me to England to
have them instructed by the Church Missionary Society, as I believe that Society is
wont to instruct the heathen. It is needless for me to tell you what other advantages
I expect, will be derived from bringing such persons to England, for your own mind
must comprehend the matter at a glance in its every bearing. Suffice it, therefore, to
say I have fully made up my mind to make the attempt of bringing young Wagandas
to England, and I trust if I do succeed they will be welcomed by you.
By the same post that conveys this letter to you I shall write to Sir Roderick
Murchison informing him of the matter, and telling him that, as missionaries and
travellers under all denominations necessarily assist one another in working to the
:same grand end so I hope the heads of those Departments at home will join together
in their mutual support when as is the case now I shall have acted without any
precedent and on the assumption only of discretion that what I am doing is for the
Truly yours, (signed) J. H. SPEKE.

THE following observations on the avifauna of Bwamba County, Toro,
Uganda, were made by the authors during a visit of three weeks to the area
(23 September-18 October 1951). This paper is confined to notes on species
seen or collected additional to the information in previous papers in this Journal
by V. G. L. and G. R. C. van Someren (1949) and J. G. Williams (1951).
During the visit we lived chiefly at Hakitengya, in the house originally built
for the Yellow Fever Research Institute. We also camped near the hot springs
at Mongiro, and, for two days, on the Semliki River near its junction with the
Lamia. Our collecting expeditions were therefore largely confined to the area of
cultivation, to forest remnants near the road, and to the forest and swamp at
Mongiro. Our object was not only the collection of birds but small vertebrates
in general.
Like others before us we did not find collecting easy in Bwamba. The dense
undergrowth, the prevalence of man, the elusive nature of forest species and the
limited time available, all prevented us from making as complete a collection of
birds as we could have wished.
Our thanks are due to the Uganda Game Department and to the District
Commissioner, Fort Portal, for permission to use the house at Hakitengya, to
Mr. J. G. Williams and Captain C. H. B. Grant for the identification of our
specimens, and to Doctors A. J. Haddow and W. H. R. Lumsden and Mr. E. H.
Winter for invaluable assistance in many ways.
During our visit we obtained or saw eighteen species which are new to the
list of birds recorded from Bwamba. In a locality so diverse and relatively little
known and containing in a comparatively small area so many different types of
habitat, this fact is not very surprising, but it seems worthy of record. Those
species which are new records for Bwamba are prefixed with an asterisk.
The nomenclature followed is that of Sclater (1924, 1930), and Mackworth-
Praed and Grant (1952).

We saw a few dabchicks on the Nyangasa Lake on the 13 October. This lake is
about 200 yds. long by 50 yds. wide and is covered with duckweed. The forest reaches
right down to the edge of the water. At the time there were also a few moorhens,
darters and cormorants.
*BUTORIDES STRIATUS ATRICAPILLUS (Afzelius). Green-backed Heron.
We saw a few birds on the Semliki River below Sempaya, and one bird, almost
certainly of this species, in the hot swamp near the Mbuga Hot Springs on 6 October.
FALCO CUVIERI Smith. African Hobby.
We obtained a male near the Kirimia River on 28 September. It had been noticed
sitting on a branch of a dead tree for two or three days. Another bird flew over
Bundibugyo on 24 September.

Two specimens obtained at Mongiro had been feeding on oil palm husks (Elaeis
guineensis Jacq.), although we shot them scavenging near our camp.
A bird, almost certainly of this species, was seen near Hakitengya on 30 September.
GYPOHIERAX ANGOLENSIS (Gmelin). Vulturine Fish Eagle.
This species was extremely common near Mongiro. Often half a dozen were
visible at one time. One stomach contained oil palm husks.
CIRCUS sp. Harrier.
Birds were seen at Mongiro and also on the open Semliki Plain during September
and October, but unfortunately no specimens were obtained and we cannot say
which species occurred.
*TRINGA HYPOLEUCOS L. Common Sandpiper.
Numerous at Mongiro.
*TRINGA OCROPHUS L. Green Sandpiper.
One or two seen with the next species.
*TRINGA GLAREOLA L. Wood Sandpiper.
Common at Mongiro, in the swampy ground round the hot springs which provide
ideal feeding grounds for waders. This species was the commonest present and
several flocks of six or eight were seen.
*TRINGA STAGNATALIS (Bechstein). Marsh Sandpiper.
One at Mongiro on 6 October.
*TRINGA NEBULARIA (Gunnerus). Greenshank.
One seen at Mongiro on 16 October was definitely identified.
*LARUS FUSCUS FUSCUS L. Lesser Black-backed Gull.
A native brought to us an adult bird of this species while we were camped at
Mongiro, saying that he had trapped it in his shamba. It is one of the most unlikely
birds to find in Bwamba, but we also picked up a primary feather of some species of
gull, probably this one, in the hot swamp near the Mbuga Hot Springs. It is possible
that the bird could have wandered south from Lake Albert, and if so, this is a new
record for Bwamba.
COLUMBA ALBINUCHA Sassi. White-naped Forest Pigeon.
We carefully searched for this bird among the large flocks of pigeons which feed
in the thick palm forest near the Mongiro hot springs. We shot a considerable
number of pigeons for food but all were C. a. arquatrix. We did, however, obtain
one juvenile of C. albinucha on 7 October, which was shot in the branches of an
ironwood tree (Cynometra alexandri C. H. Wright) about 50 yds. from our camp in
the "fly" clearing at Mongiro. This was the only example that we saw, and is the first
specimen yet obtained in immature plumage. Its stomach contained small seeds.
STREPTOPELIA CAPICOLA TROPICA (Reichenow). Ring-necked Dove.
It appears to us to be worthy of record that, whereas these doves were very
numerous to the east of the Buranga Pass, extending right up to the top of the pass,
none were ever seen on the Bwamba side of the mountain. This appears to be a
clear example of the faunal barrier presented by the Ruwenzori massif.
TRERON AUSTRALIS UELLENSIS (Reichenow). Uelle Green Pigeon.
A fledgling was brought to us by the natives on 1 October. They said they had
found it near Hakitengya in a nest 3 feet above ground in a bush.

CHRYSOCOCCYX FLAVIGULARIS Shelley. Yellow-throated Green Cuckoo.
We collected an adult female in full breeding condition in a tree beside the Tokwe
River near Hakitengya on 27 September. Its stomach contained the remains of hairy
lepidopterous larvae and Coleoptera.
CERYLE RUDIS RUDIS (L.) Pied Kingfisher.
Previously only recorded on the Semliki River, where we also found it abundant,
but we saw one bird hovering over the stream at Mongiro near the Mbuga Hot
Springs. The attraction here was probably the small fishes which live just upstream
of the hot springs, where the stream leaves the forest.
*MEROPS APIASTER L. European Bee-eater.
Flocks of bee-eaters were often seen over Hakitengya at a considerable height,
attention being called to them by their unmistakable notes. On one occasion a flock
of about 60 birds was hawking termites in the village and we obtained two adults
and a juvenile, the remaining members of the flock being so intent on their food
that we could have secured the lot.
BYCANISTES ALBOTIBIALIS (Cabanis and Reichenow). White-thighed Hornbill.
This hornbill occurred commonly, generally in pairs, but on one occasion a flock
of six or seven was seen. A pair shot near Bundibugyo on 27 September had been
feeding exclusively on coffee berries.
We collected a female in the hills above Sempaya on 15 October. Its stomach con-
tained arachnids as well as hemipterous and coleopterous remains. This species does
not appear to have been previously recorded within Bwamba County and it was, in
fact, only just in the area.
We shot one male on the road near Mongiro after dark. It had been feeding on
flying termites. We did not see any other nightjars at all, despite constant night
CYPSIURUS PARVUS PARVUS (Lichtenstein). Palm Swift.
This species was the only bird we saw in the Nkarara palm grove, an area of high
elephant grass with scattered borassus palms (Borassus aethiopum Mart.) occurring
in the midst of the forest. It was exceedingly numerous there, as also at Mongiro,
and a few were seen near Hakitengya.
A juvenile male was brought to us by natives on 4 October.
CAMPETHERA NIVOSA HERBERTI (Alex.). Ubangi Buff-spotted Woodpecker.
A native brought us a female of this species together with two eggs which he said
he had taken from a hole four feet up in a dead tree. The stomach was crammed full
of small ants. The eggs were white, very transparent, and measured 22-3 x 17-0 and
21-2x 16-4 mm.
TELACANTHURA USSHERI SHARPER (Neumann). Spine-tailed Swift.
A pair of these birds, which were first recorded for Uganda by Williams (1951),
was found roosting after dark, clinging to the wall of the Court House in the village
of Hakitengya, near the old nest of some species of swallow. The male was obtained
on the 26 September and found to be in breeding condition, hut the female was not
seen again.

MOTACILLA AGUIMP VIDUA Sundevall-African Pied Wagtail.
Large numbers of these birds were seen roosting in thick beds of elephant grass at
dusk in the Kirimia swamp. It was impossible to count the numbers, but there were
probably two or three hundred, mostly roosting off the ground in the edges of an
extensive patch of elephant grass. This behaviour is not unlike that of the British
species, Motacilla alba yarelli Gould. They associated with Lamprocolius splendidus
*MUSCICAPA STRIATA STRIATA (Pallas). Spotted Flycatcher.
We saw a few birds in and around the villages of Bundibugyo and Hakitengya.
Although we did not secure a specimen, we are confident that our identification is
correct, since we are familiar with the species in England.
*TCHITREA NIGRICEPS SOMERENI (Chapin). Black-headed Paradise Fly-
We obtained one male and saw others in thick forest near Mongiro.
HIRUNDO RUSTICA L. Common Swallow.
Very large flocks of this species were seen over the Mongiro swamp in early
October; amongst them were a few sand martins (Riparia riparia riparia (L.)).
*HIRUNDO RUFULA EMINI Reichenow. Red-rumped Swallow.
One bird from a small party at Hakitengya was shot on 29 September. It was seen
to perch several times in a dead tree.
CORVULTUR ALBICOLLIS Latham. White-necked Raven.
In September we saw several of these birds scavenging, in company with marabou
storks, kites and hooded vultures, around the Uganda Fish Marketing Corporation
Camp at Bweramule.
Two females were secured from a flock on the banks of the River Semliki where
it emerges from the forest.
As already noted under the African Pied Wagtail, we saw communal roosting of
this species near the Kirimia River. The birds gathered before dusk in nearby trees,
to the accompaniment of much melancholy whistling, and, shortly before dusk,
flocks would wheel round and round over the tops of the elephant grass, calling all
the time and displaying astonishing agility and speed. They finally dived into cover
as dusk fell. The whole process was not unlike that of English starlings at their
roost. We estimated that several hundred birds gathered to roost in this particular
place, and on several occasions birds were also seen assembling in trees near
Bubukwanga. All of three birds shot at the latter rendezvous at 6 p.m. had empty
POEOPTERA LUGUBRIS MAJOR Neumann. Ituri Narrow-tailed Starling.
A few small flocks of this species assembled in some very tall trees near
Bubukwanga, in company with the Splendid Glossy Starlings, but were far out-
numbered by them. They also departed in the same manner.
*PLOCEUS NIGRICEPS NIGRICEPS (Layard). Spot-backed Weaver.
An adult female, in full breeding condition, was obtained on the Tokwe River
near Hakitengya.

Four nests examined on 17 October contained fresh eggs, two nests containing
two eggs, and two containing one. The nests were built round two stems of elephant
grass at a considerable height off the ground. The birds did not appear while their
nests were being examined.
We secured a male bird which was building a nest in the "fly" clearing at Mongiro
on 5 October. The nest was an untidy structure of fine grasses, with the entrance on
one side, placed at the very end of a drooping branch of an ironwood tree. The bird's
stomach contained only oil palm husks.

MACKWORTH-PRAED, C. W. AND GRANT, C. H. B. (1952). Birds of eastern and
north-eastern Africa. London: Longmans.
SCLATER, W. L. (1924, 1930). System avium ethiopicarum. London: B. O. U.
VAN SOMEREN, V. G. L. AND VAN SOMEREN, G. R. C. (1949). 'The birds of Bwamba.'
Uganda J., 13, Supplement.
WILLIAMS, J. G. (1951). 'The birds of Bwamba: some additions.' Uganda J., 15, 107.

By J. E. HLL
British Museum (Natural History)
AMONG the mammals collected by Lord Richard Percy, H. E. Percy and
M. W. Ridley at various localities in Bwamba County, Uganda, the follow-
ing constitute range records of some interest. The specimens have now been
presented to the British Museum (Natural History).
Hopkins (unpublished MS.) suggests that material from Toro should be referred
to the geographically nearer H. r. semliki (Thomas). A specimen collected at
Mongiro, 2,400 feet, clearly referable to nyansae and not to the more deeply
coloured semliki, indicates that the range of nyansae extends in a westerly direction
probably to the Semliki River.
A specimen caught in a gin trap at Hakitengya, 2,800 feet, is apparently the first to
be recorded from Uganda, although Hylarnus has been collected as far east as the,
Semliki River. This is the first complete adult specimen received in the Museum
collection, the race hitherto being represented by the type skull, an immature study
specimen and three native skins.

By D. H. L. ROLLINSON, PH.D., B.Sc., M.R.C.V.S.
THE eating of earth by cattle, wild herbivora and humans is widespread in
Africa and the presence of salt licks, i.e. a cutting in the soil licked and eaten
by herbivora with beneficial results has for generations been recognized by
natives as an asset in grazing.
The tendency to eat or lick earth is known as geophagia and is common to
several types of specific mineral deficiencies and may also occur in certain
diseases. The subject is both vast and complicated and is as yet scarcely
explored. This has caused two fairly widespread assumptions which are not
borne out by fact. The first of these is that the licks so much liked by cattle in
Africa are taken because of their common salt content (sodium chloride). This
is erroneous since only a very small minority of the licks that have been
examined show any sodium chloride. The second assumption is that when cattle
eat earth they are necessarily deficient in the element phosphorus, a deficiency
of which is characterized by a depraved appetite.
It was in fact the abnormal appetite in cattle that gave rise to the earliest
reports of these conditions, but typical phosphorus deficiency results in a craving
for cattle bones or decaying carcasses. Thus about 1780 it was noticed in
Namaqualand that when the grass was harsh the cattle searched out bones to
chew but when the pasture was good this never occurred. Similarly in 1838 it
was reported that wild cattle in Paraguay ate the dry bones of animals which
died. These were, however, isolated records and the first scientific observations
on phosphorus deficiency as such were recorded by Hutcheon, Veterinary
Surgeon to the Cape Colony, between the years 1884 and 1903. Since that time
much work has been done in South Africa on the specific deficiency of phos-
phorus and the means of eliminating its effects, but in East and Central Africa
cattle have continued to eat large quantities of earth.
It is almost impossible to draw a clear line between the accidental ingestion
of earth whilst grazing and deliberate geophagia, whatever its cause; thus
French states "the more rational outlook is to regard geophagia in animals,
other than the unavoidable ingestion of soil during feeding, as either:
(a) the satisfying of a passing fancy for something different or of condimental
(b) a habit acquired through copying other members of the herd;
(c) the satisfying of a craving for something missing in the diet; or
(d) the expression of an unsatisfied craving as a result of ill-health or serious
In Africa edible earths are usually found at the edges of dried-up lakes or
swamps, along river banks or near dams. They may consist of old ant-hills;

frequently they are said to have originated as ant-hills, although there is now no
sign of a hill. This is of interest since there is some evidence to show that a
difference of chemical composition exists between the ant-hill earth and the
surrounding soil. The chief element concentrated in ant-hill earth is calcium.
From Java, den Doop has reported that ant-hills in flat, grey land act as
chimneys during the rainy season; water continually evaporates through them,
on account of their high, dry and exposed position, resulting in a movement of
soluble elements from the surrounding soil into the base of the ant-hill,
especially the lower internal portions. At the same time the land surrounding the
ant-hill becomes robbed of these substances. This causes a concentration of
calcium and magnesium in the base of the ant-hill and the opposite effect in the
case of potassium and phosphate. Somewhat similar effects were found by
Griffiths in Uganda. However, the factors causing these effects are various and
do not seem to be well understood.
Owing to the desire for common salt, widely known to be exhibited by live-
stock in European climates (where salt feeding is an old custom) it has been
assumed that edible earths in Africa were attractive because of that salt content.
This, with few exceptions, is not so; thus of 25 Kenya samples reviewed by
French none contained sufficient sodium chloride to make it probable that they
were eaten to satisfy craving for this salt.
Further studies of the composition of edible earths were made in Nigeria
(Godden, 1929) and also in Kenya (Orr and Holden, 1931) and in neither study
was the presence of common salt outstanding. In many there was variation in
composition, and there was no specific factor to indicate why geophagia should
be so common. It was further concluded by Hudson (1944) that the lime and
phosphoric acid contents are too small to be of much value to the animal.
Although it has been assumed that certain natural licks do contain soda, lime,
iron, or other compounds of use to stock, the reason why the majority of them are
so attractive is still a mystery. In Tanganyika, salt deposits and nearby water
were analysed and it was found that the same salts were consumed in drinking
water as were sought after in the edible earths, so that, in one area, the intake of
sodium in the water drunk would be greater than that in the earth lick assuming
that two ounces per head per week of earth were consumed.
In Africa the sodium content of the grass for grazing is lower in the dry
period than in the wet, but earth-eating occurs where pasture contains adequate
sodium. There is, therefore, no correlation between earth-eating and the sodium
content of the forage. Seven of the twenty-six analyses reported from Kenya
showed the quantities of soluble sodium and potassium to be so low that they
could not be of significant value in improving the intake of potash. Neal (1941)
has further pointed out that plant growth would fail before the potash content
in the plant became inadequate for stock, which suggests that pastures could
never be deficient in potash.
French concluded "that craving for licks is prevalent in most districts, even
on the fertile pastures of volcanic ash soils. That the composition of the lick does
not reflect the nature of the deficiency causing the craving is also obvious, since

Photo by D. H. L. Rollinson

FIG. 5 Photo by D. H. L. Rollinson
Cattle at earth licks near Aloi, Moroto County, Lango District.

FIG. 4

the licks analysed vary from almost water-insoluble soils to rich salt 'cakes' and
it is possible that stock suffering from a mineral deficiency would eat no earth
containing that mineral, but the fact that the craving for licks is prevalent even
on good pasture suggests that another cause may be operating."
Most of the analyses quoted, however, make no allowance for the presence of
trace elements. It has been the investigation of earth-eating that has led to the
recognition of cobalt as a cure in enzootic marasmus in Australia while attention
was also drawn to copper deficiency in this way. Further, it has only recently
been found that cobalt forms 25 per cent of the molecule of vitamin B12, the
animal protein factor, although it had long been known that cobalt cured
enzootic marasmus when given by mouth, but not by injection.
Russell (1944) concludes that depraved appetite is usually an indication of
deficiency. When this takes the form of osteophagia, phosphorus deficiency is
indicated, but in other cases, the real deficiency may not be apparent from
analysis of the materials eaten. For instance, animals may appear to seek earths
or salts high in sodium or iron or consisting mainly of silicates, while their real
objective may be some trace element.
On the other hand Maynard (1944), when referring to man states, "It is worth
while to remember that consumer demand is not identical with nutritional
requirements. If man cannot choose his foods to correct the deficiencies in his
diet, why should domestic animals be credited with these powers." This becomes
more interesting when one considers that phosphorus-deficient cattle will crave
bones, but when allowed access to calcium phosphate or other unfamiliar
sources of phosphorus which will alleviate the symptoms no craving is shown.
On the other hand, some animals, once cured of phosphorus deficiency by the
addition of bone meal, continue to show a craving for bones. In the case of
copper and cobalt deficiency, many objects which are unlikely to satisfy the
demand for these elements are eaten. In this case, the instinct is blind, but it is
possible that if animals by trial and error, can find an antidote, that this material
may be eaten exclusively.
The mineral requirements of young animals depend upon their rate of growth..
The faster the growth, the richer must the diet be in all food constituents, includ-
ing the inorganic salts which are needed as constructive material for the forma-
tion of new tissue. This is strikingly illustrated by comparisons of the average
rate of growth of the young of different species with the concentration of mineral
matter in the milk of the species; faster-growing species have milk containing
a higher percentage of ash, and a higher proportion of ash to energy-producing
The rate of growth of cattle in the natural state is slow compared with modern
standards and the amount of milk produced by cows is limited to that required
by the calf. The improvement in cattle through selective breeding has produced
types with a greater rate of growth and greater capacity for the production of
milk. Modern cattle grow at double the rate of the type from which they
originated and they reach maturity in about half the time. The domestic cow,

under good conditions of feeding, has her first calf when about three years old.
'Native' cattle breed much later. Milk production is similarly increased.
This improvement of breed should run hand in hand with improvement in
pasture and pasture-management, and when animals of an improved type bred
on a mineral-rich pasture are transferred to a district with poor pastures, the
low mineral content of the pastures is liable to be insufficient to support the
rate of growth or the capacity for milk production; consequently malnutrition
due to mineral deficiency is likely to occur. Orr maintains that this is the explana-
tion of the poor results obtained by grading up 'native' cattle with high-grade
When pastures are depleted of minerals by removal of animal or plant pro-
ducts without compensating return to the soil and if the forces of nature are
allowed to operate, there would be a gradual elimination of the faster-growing
types with the evolution of a type with a slower rate of growth, adapted to the
impoverished pastures. If selective breeding and importation of new blood takes
place, the quality of the stock is maintained or even improved, and the gap
between the requirements of the animal and the supply of these minerals from
the pasture increases.
Alternatively the elements may be present in the soil or herbage, but may not
be available to the animal. Thus much soil phosphate is rendered unavailable
to the plant by the presence of other chemicals. Similarly the absorption of
calcium and phosphorus from the foodstuff via the animal's intestine is governed
by the presence of vitamin D and the state of the intestine. The low level of
copper in the blood and liver of cattle in England and Scotland is due to a failure
of absorption by the animal since the soil and grazing contain adequate copper.
Orr (1929) reported that "In a certain area in Uganda, a disease termed
rickets is said to occur in young cattle. There are usually other signs of malnutri-
tion, such as sterility in cows, stunted growth and high mortality in calves." He
concluded, "It is probable that if the necessary investigations were carried out
in those districts from which pica and malnutrition are reported, pathological
conditions affecting the bones would also be found".
Apparently nothing further was published until Fiennes (1939) reported a
disease resembling phosphorus deficiency near Kapujan on Lake Salisbury,
Teso District. Here symptoms were seen only in the dry season; the cattle showed
debility, crossing of the knee joints, a turning outward of the hind legs and a
jerky gait. Over-growth of the horn on the medial claw of each front foot was
also seen. Pica was not marked, but calves were in very poor condition. Soil
samples showed low levels .of potash, phosphorus and calcium.
In August, 1951, reports were received from Lango District of a large herd
ravenously engaged in chewing, licking and consuming as much as they could
obtain of a yellowish earth. It was reported from Mbale District soon afterwards
that stunted cattle occur in Bulucheke in South Bugishu; in this case large
quantities of matoke (banana) and coffee are continually being removed from
the area in just the manner to deprive the soil of mineral reserves. There is no
doubt that this earth-eating by cattle and other ruminants is common to most

parts of the Protectorate, although its degree, frequency and season vary some-
what. Buffalo and elephant also partake regularly of earth licks.
The eating of earth by cattle is seen in trypanosomiasis but this is probably
due to the fact that the disease causes an increased elimination of nitrogen bases,
chloridess and phosphates; thus if the food supply has only barely sufficient
mineral elements the trypanosome infection will make an animal deficient and
geophagia may follow. However, many of the reports of earth-eating come from
areas in which trypanosome infection is not common or does not occur, so that
whilst the two are inter-related, trypanosomiasis is not the whole explanation.
Earth-eating is frequently seen in Lango District in the area within the
triangle Aloi, Omoro and Orumo in Moroto County. At the time of the investi-
gation blood slides from herds in this region failed to show any evidence of
trypanosomiasis, but cattle were in poor condition. Further enquiries in Lango
District revealed that most owners allow their cattle to eat earth for a beneficial
effect on the animals. Earth licks are almost universally used and, as it is
estimated that from one to five hours a day may be spent there, it is clear that the
animal's grazing time suffers; hence any beneficial effects of the earth are
nullified by fall in food intake. Many cattle owners can distinguish 'good' and
'bad' licks; those along the north shore of Lake Kyoga are mostly regarded
favourably. Generally cattle so relish the licks that they have to be forcibly
driven from them.
The earth relished by cattle consists usually of sub-soil frequently in the spill-
way of a dam or at the edge of a drinking place. It is usually light grey in colour,
frequently mixed with water to a semi-liquid mud, which also is taken by the
cattle with relish. The quantity of earth eaten by the cattle does, in certain
seasons, cause death from impaction in the stomach; this by-product of the con-
dition is a minor cause of loss. No accurate estimations of the quantity of earth
consumed have been made, but from watching cattle at the licks, it appears that
two ounces of earth per animal per day would not be an excessive estimate.
Both the phosphorus and copper levels have been estimated in blood samples
from cattle in this area but so far no indication that these elements are deficient
has been obtained; further investigations are in progress. More recent work in
Nigeria has allowed Garner to suggest that calcium deficiency is the principle
factor in this condition. This agrees with the suggestion that calcium is con-
centrated in ant-hills especially in their lower regions (which are in fact most
relished by cattle in Lango), and also with present work at Entebbe.
Animals require food to obtain energy and materials which are used for
growth, to replace worn-out tissues, for movement and locomotion, to produce
hormones which regulate the animal to its environment, and for the production
and nourishment of young. Water is essential to maintain the equilibrium of the
body, to aid excretion and to help regulate the body temperature by evaporation.
For the food animals to provide the greatest amount of meat and milk in the
most economical way, their food must be available within a reasonable distance

(else excessive energy is wasted during the search), it must be palatable and
satisfying to the animal and contain adequate protein, fat and carbohydrate. In
addition to these constituents, adequate but minute amounts of many inorganic
salts are necessary.
Pasture is the natural food of the species of herbivora which have been
domesticated. It is, therefore, the natural raw material of the most important
animal products, namely milk, meat, wool and hides, so that a large part of the
world supply of these primary necessities is drawn from animals whose food
consists entirely of pasture, either grazed or preserved as hay. Grazing animals
in the natural state depend entirely on the vegetation growing on the normal
grazing grounds. The type, quality and quantity of grazing depend in turn, on
the soil and the water supply. Much of Uganda, and indeed of Africa, consists
of volcanic rock masses covered to a variable depth by soils, many of which are
volcanic in origin and which have, in addition, been leached by centuries of
violent rains. Of the basic compounds which form vegetation, hydrogen and
oxygen are obtained from water, carbon and nitrogen from the atmosphere and
the soil. These four elements are the basis of the great organic compounds which
constitute by far the greater part of the food substance. The mineral elements,
such as phosphorus, calcium, sodium, chlorine, iodine etc., if they are not present
in the soil in sufficient amounts, and in a form available to the plant, cannot be
obtained in appreciable quantities from rain or the atmosphere. It is, therefore,
very likely that the absence of elements beneficial to the animal body will be
widespread and that these deficiencies will limit the rate of growth, the health
and the production of cattle, sheep and goats. Furthermore, animals will, by
trial and error, endeavour to repair such deficiencies. The study of these
deficiencies thus assumes importance at the present time when increased pro-
duction is of such value to mankind.

ap Griffiths, G. (1938). A note on termite hills. E. Afr. agric. J., 4, 70.
den Doop (1938). Utilization of sisal waste in Java and Sumatra, Part II. E. Afr.
agric. J., 4, 89.
Fiennes, R. N. T. W. (1939). A suspected mineral deficiency among cattle in Uganda.
E. Afr. agric. J., 5,68.
French, M. H. (1945). Geophagia in animals. E. Afr. med. J., 22, 103 and 152.
Godden, W. (1929). Notes on certain edible earths and native salts from Nigeria.
W. Afr. med J., 2, 187.
Hudson, J. R. (1944). Notes on animal diseases. XXIII-Deficiency diseases. E. Afr.
agric. J., 10, 31.
Maynard, L. A. (1944). War-time and post-war nutrition problems. Chem. andInd.,
63, 402.
Neal (1941). Quoted by French (1945).
Orr, J. B. (1929). Minerals in pasture and their relation to animal nutrition. London:
H. K. Lewis & Co.
Orr, J. B., and Holm, A. (1931). The Economic Advisory Council. Sixth Report of
the Committee on the mineral content of natural pastures.
Russell, F. C. (1944). Minerals in pasture-Deficiencies and excesses in relation to
animal health. Imp. Bur. anim. Nutr. tech. Comm., No. 15, 68.

PATHOLOGISTS and biochemists are surprised, on first coming to Uganda,
to find that the so-called 'normal values' for various constituents of the blood
do not seem to apply to Africans. A man or woman, apparently healthy and with
no complaints, may have blood with a red-cell count of only 3-5 million/cubic
millimetre. The normal values for the red-cell count are 5-6 million/cubic
millimetre. Here is a man, happily digging his shamba, with a red-cell count
which, in England, might make him a candidate for a blood transfusion. It is the
same with several other elements of the blood. The figures for normal levels
given in the text-books of medicine and physiology are based on the average of
a large number of healthy Europeans and Americans, and apparently-normal
Africans are in several respects 'not like the book'.
The properties of the blood of special interest to the Physiology Department,
Makerere College, to which I am attached, are (a) the red-cell count, (b) the
concentration of haemoglobin, the red protein present in the red-cells and (c)
the concentration of the proteins present in serum, the two most interesting
being albumin and gamma-globulin. These constituents have been measured in
the blood of a large number of men of three groups: firstly, Europeans living in
Kampala; secondly, students of Makerere College, who are known to be living
in healthy conditions and eating an adequate diet; and thirdly, an assorted col-
lection of Africans, representatives of some eleven tribes, living in or near
Kampala and having no complaint suggestive of anaemia or serum protein
disturbance. The average values for each group of men were found and com-
The normal red-cell count varies with altitude above sea-level, a physiological
response to the lowered oxygen-pressure of high altitudes. The red-cell count of
the Makerere students, 5-9 million/cubic millimetre, is the count appropriate to
the altitude of Kampala, four thousand feet. The average red-cell count of the
Europeans was only a little lower. The average for Kampala Africans, however,
was considerably lower, 5-4 million/cubic millimetre. Similarly, with the
haemoglobin concentrations, the Europeans and the Makerere students had
haemoglobin levels far higher than that of the mixed bag of Kampala Africans.
As to the serum albumin, the highest mean figure was that of the European
group, 4-25 grammes/100 millilitres; the Makerere students' value was only
slightly less, 4-19 grammes/ 100 millilitres; the Kampala Africans had much less,
only 3-27 grammes/100 millilitres. One possible explanation for these findings
is that many Africans eat a diet extremely low in protein; plantains (matoke),
cassava and sweet potato, which form the greater part of the diet of many
Kampala Africans, are notoriously poor in this essential foodstuff. The bodies
of these Africans are thus short of the materials necessary for forming the pro-
teins of the red-cells and the serum. It is well known from animal experiments
that a protein-deficient diet causes mild anaemia and lowered serum albumin

level; and this may be, at least, one of the causes of these differences between
European and African blood.
The gamma-globulin levels are harder to explain. The highest mean level
(nearly 2 grammes/100 millilitre) was found in the Kampala Africans: the
Makerere students had 1-6 grammes/100 millilitre, and the Europeans only
just over 1 gramme/100 millilitre. A raised gamma-globulin level is a symptom
of certain types of liver disease, and numerous Africans are found to have slight
liver damage which is thought to be caused by a low-protein diet in childhood,
or possibly (an explanation suggested by experiments on animals and humans)
the raised gamma-globulin is the direct consequence of a diet rich in vegetable
foods and poor in animal foods. Again, the antibodies of the blood, made by the
body when stimulated by parasitic invaders, are associated with the gamma-
globulins of the serum: could it be that Africans are particularly rich in the
antibodies to some universally-occurring parasite? Levels of gamma-globulin
high by text-book standards have been found in most of the black races of the
world (West African and American negroes, South African Bantu and Fijian
and Samoan Islanders) but it seems, a priori, unlikely that the high gamma-
globulin is merely a racial character.
Clearly some of the questions raised by these findings could be answered by
studying the blood of Africans who habitually eat a diet rich in protein, and who.
are known to have eaten such a diet from childhood. The Karamojong tribe are
Africans of this kind. They drink the milk and defibrinated blood of their cows,
supplementing this food with maize, millet and beans, all of which have a high
protein content as compared with roots and plantains. When their cows or goats
die, they eat meat but they do not kill them for food. When they eat meat, they
eat it in great quantities, and the entire carcase, except for the hide, horns and
hooves, is consumed. Infants are given cow's milk from the age of six months
when the breast-milk becomes inadequate. There is little doubt that the
Karamojong eat more protein than any other tribe in Uganda. It seemed, then,
that if blood samples from healthy Karamojong men could be obtained and
studied, they would help the interpretation of the values found normal in
Accordingly, in January 1953, a party of four visited Karamoja District (by
courtesy of the Provincial Commissioner, Northern Province, and the Uganda
Medical Department) to collect blood samples from men there. There were
Holmes-a physician, Croot-a surgeon, Kyobe-a Muganda technician, and
myself. We took camp gear, laboratory equipment and a few medicines, in the
back of Holmes' 15-cwt. truck, and set out from Kampala with enough food for
three weeks. The heaviest item of equipment was a petrol-drive generator which
produced current for the laboratory apparatus. Fortunately we did not have to
take tents, as these were lent to us by the Medical Department at Moroto.
Our base was Lotome Protestant Mission. A Karamojong sub-chief living
nearby spoke Swahili and attached himself to our party as guide and interpreter.
We offered medical treatment, for minor complaints, to the men (and sometimes
their wives and children), and in return we took about 10 millilitres of blood from
each man. People came along with trivial complaints, a headache, or cough, or
constipation. We gave them aspirin, a dose of cough mixture, or cascara tablets,

and they were quite prepared to let us take a little of their blood. We had brought
cigarettes, tobacco and sugar with us to Karamoja, by way of payment for blood,
but it soon became obvious that aspirin was more popular. Many people who
had nothing wrong with them came along with dramatic complaints ("Pains all
over, Doctor") just for the entertainment of having a stethoscope put over their
chests. We took blood from an arm vein, by syringe or open needle, into three
sterile tubes: one sample was for the red-cell count and the measurement of
haemoglobin; another was allowed to clot, and then the serum was separated in
sterile conditions and stored, ready for bringing to Kampala for serum protein
measurements; the third sample, a few drops, was for blood-grouping. The latter
was not part of our investigation, but a Kampala pathologist, who is interested
in the distribution of the blood-groups in different tribes, had asked us to bring
back blood samples for this purpose. The blood-count and haemoglobin
measurements could be done on the spot, either the day the blood was drawn or
the following morning; we had taken an electro-haemoscope, which gives these
measurements quickly and accurately by means of a lamp, a photo-electric cell,
and a galvanometer. The remaining tests had to await our return to Kampala.
After about a week collecting blood samples from men living around Lotome,
we made expeditions to collect blood from men in the Nabilatuk area, further
south. Here there are dry-weather grazing-grounds, and a number of the strong
young men from the Lotome area had taken most of the cows and goats and
moved to cattle-camps around Nabilatuk, until the next rainy season. So we
went out on day-safaris to collect blood. We took in the truck our medicines,
test-tubes, syringes and a folding camp-chair. Our procedure was to park the
truck under a tree at a spot chosen on the advice of the Karamojong sub-chief.
We then brought out the camp-chair, and a table on which the medicines and
syringes were set out. In a very short time, a crowd collected, and our sub-chief
described what we proposed to do. After a few minutes someone would start
the ball rolling by sitting down in the chair. His blood was taken under the gaze
of a fascinated crowd of spectators, and he was duly examined and given some
medicine. From then on, there was no stopping them; as soon as someone left
the chair, the next man seated himself in it. We recorded, for each man, his
name, approximate age, and height, the state of his teeth, gums and eyes, and
whether his spleen was enlarged. Very few of the men had any serious disease;
to those who were genuinely ill, we gave such treatment as we could, but took no.
blood, because we wanted blood of normal men for purposes of comparison.
We gained a view, though of course only a superficial one, of the medical state
of these Karamojong. Partial blindness, due probably to trachoma and con-
junctivitis, was shockingly prevalent. We rarely saw malaria in an adult, but a
number of sick children who might have had malaria were brought along. Worm
infestations were among the complaints, but we could not tell how common they
were. Enlarged spleens were far less common than among Kampala Africans.
In all, we collected blood from 129 persons in about two weeks. For the first
week, the party camped in a dry river-bed and the laboratory was a tent. Later,
the missionaries kindly lent us their house at Lotome while they were on local
leave, and gave us the use of a little school-room nearby as our 'laboratory' and
clinic. There was always a number of spectators for our laboratory work. Little.

boys peered in through the windows to watch the centrifuges go round. Alto-
gether, our party gave great entertainment, and we were flattered, on our
departure for Kampala, by the earnest requests to come again.
Our own entertainment (apart from guinea-fowl shooting and climbing Mount
Moroto), was to watch the Karamojong watering their cattle, and drawing their
.cows' blood for food. The waterholes in the stream-bed are surrounded by a
fence of cut thorn, and the gateway is unblocked only when the herds are being

FIG. 6
Karamojong taking blood from a calf. One man holds the animal's ears and lower jaw,
another its foreleg; a third shoots the arrow and a fourth collects the blood in a calabash.

watered. A wooden trough is put in the gateway and filled by a calabash from
the water-hole and the cows are driven up, two or three at a time, to drink. At
the height of the dry season, the goats are watered only on alternate days. The
procedure of bleeding the cows is most interesting. It is done, as one of the
Karamojong pointed out to us, "the same way as you take our blood", (except
that a different vein is used, and a different instrument). A strip of hide is tied
round the cow's neck, close against the shoulders, to make the jugular vein stand
out. Then, while two boys hold the cow's head and shoulders, a man shoots at
the jugular, at the top of the dewlap, with a bow and arrow, from a range of
about six inches. The jugular is pierced and the blood is collected in a calabash.
When enough has come out, the strip of hide is untied, and the animal stops
bleeding almost immediately. The blood is stirred with a stick until the clot of

fibrin separates out. This is removed and given to the dogs in times of plenty,
but in the dry weather, the men cook and eat it themselves. The liquid blood
which remains (it was 36 fl. oz. in volume on the occasion which we witnessed),
is mixed with an equal volume of milk and makes a meal for one man. The
mixture is not cooked in any way; it is simply drunk. In the rainy season, when
the cows have plenty of grazing, this meal is taken every day but in the dry
weather, only once or twice a week. Millet and posho appeared to compose the
remainder of the dry-weather diet. It was satisfactory to see this for ourselves,
to assure ourselves that the Karamojong really did take the diet attributed to
The results of this study of Karamojong blood are still incomplete. The mean
red-cell count of the tribe is considerably higher than that of Kampala Africans;
in 123 Karamojong men it was 5-85 million/cubic millimetres, about the same
figure as had been found earlier for Makerere students. Unless one makes the
unlikely assumption that there are inborn tribal differences in red-cell counts,
this finding supports the view that the low mean count of Kampala Africans is
due to a protein deficiency.
Analysis of serum proteins takes some time and all the figures are not yet
available. It is interesting, however, that in the sera so far analysed, the gamma-
globulin level does not appear to be any lower than in the sera of Kampala
Africans. If this result is confirmed, the raised gamma-globulin level of Africans
would seem unlikely to be of dietary origin. It cannot be due to a predominantly
vegetable diet, and is probably not due to liver damage either, unless one sup-
poses that the Karamojong, as well as Kampala Africans, are prone to liver
damage, which can hardly in their case be due to a protein-deficiency in child-
hood. There remain the antibody hypothesis, which is a difficult one to test, or,
an alternative explanation, that the high gamma-globulin level of Kampala
Africans and of the Karamojong have completely different causes.

/ *O4o

/i.' 3 I -"
/'/ "o + '2. %'
/ +'2 ci

] *- .. .
;.".**, /, A ,"
I - "
A ."


... .../
7 -.~.

/ ,

SI Settlement
4 /boundary
S ~ .... Modern
/ 6" road
i :j'"' 4 '" y

S. District
S .... ritu al rve
ro !'' ..'""' I bo0aory

Sketch map of the indigenous districts and settlements of the Jie, Karamoja District.
Only the settled (central) part of Jieland is shown; the rest of the country is common pasturage. Boundaries of
districts and settlements are inserted for clarification only, for there are no set boundaries of any kind. All the
homesteads and the bulk of the garden land of the inhabitants of a unit lie within the boundaries as here drawn.
A. RENGEN D. KOTIDO 4. Ngiriemiriem
1. Caicaum 1. Lothorgut 5. Loijem
2. (Central Rengen) 2. Lother 6. Tarban
3. Lokatap 3. Lokwori 7. Komukwin
4. Kapelok 4. Lokatap 8. Lorikitai
5. Kadoketa 5. Lokaicil 9. Ngileto
6. Niriwo 6. Locerimu G. KAPELURU
1. Lominit 1. (Northern Kanawert) 2. Wertakow
Rest of district not 2. (Southern Kanawert) 3. Kadoica
surveyed F. PANYANGARA 4. Potongor
1. (Western Lothilang) 2. Kaputon No subdivision; this is really
2. (Eastern Lothilang) 3. Natelo a sub-district of Rengen.
2. (Eastern Lothilang) 3. Natelo a sub-district of Rengen.


T HROUGHOUT East Africa the enumeration of the African population in
1948 was based primarily on administrative areas. Thus in Karamoja the
smallest units taken were the areas of wakungu, i.e., the lowest order of Govern-
ment-appointed chief. These areas were then grouped together into 'sub-
counties' and 'counties'. This procedure was doubtless dictated by practical
convenience as well as by its obvious administrative value. In Karamoja, how-
ever, these administrative areas do not always coincide with tribal territories,
let alone with the indigenous units within those territories. This census pro-
cedure has therefore made it almost impossible, to ascertain the exact numbers
of each tribe and of each native unit within the tribes. In Karamoja this general
difficulty has been increased because of the way in which tribal names have been
used in the Census Report. A heterogeneous collection of tribes has been
grouped together as 'Karamojong'. It can be understood, that the Jie and Dodoth
have been classified as 'Karamojong'; though in fact both are independent tribes
with some differences (especially considerable in the case of the Jie) in social
organization and culture. On the other hand there seems to be no good reason
for the inclusion of the semi-pygmy Teutho, the mountain-dwelling Tepeth, the
Labwor, and the unclassified Nangia (Ngiangeya) and Ngiakwai in this category.
Each of these five small groups differs from the Karamojong, Jie and Dodoth in
language, culture and indigenous political organization.
In Table I an attempt is made to give the basic outlines of the populations of
the eight main tribes of the Karamoja District. The assumptions upon which the
table is based are as follows-
Karamojong.-This tribe is taken to comprise the counties of Mathiniko,
Bokora and Pian, with the exclusion of the sub-counties of Tepeth (Mathiniko),
Nyakwai (Bokora) and Kadam (Pian).
Jie.-The four sub-counties of the Jie proper, viz., Kotido, Kapelimuru,
Pangyangara and Rengen, are included.
Dodoth.-The three sub-counties of the Dodoth proper, viz., Kaabong.
Loyoro and Morungole, are included.
Teutho. The Teutho sub-county of Dodoth county is included.
Ngiangeya.1-The sub-counties of Kakwanga and Napore of Dodoth county
are included.
Tepeth.-83 per cent of the sub-county of Tepeth in Mathiniko county, and
the Kadam sub-county of Pian county, are included.
Ngiakwai.-The sub-county of Nyakwai of Bokora county is included.
I The Ngiangeya people are closely related to the peoples on western Nyangeya and
Rom mountains to the west in Acholi District. It is not known whether those living in
Karamoja comprise a single tribal unit. They are often referred to as 'Napore' but this
appears only to be the name of the locality where the jakait of that sub-county lives.
179 6B

Labwor.-The sub-counties of Alerek and Wiawer (now Abim) of Jie county
are included.
Suk.-The Upe county and 16 per cent of the sub-county of Tepeth are
It is realized that certain errors exist in these assumptions, but it is believed
that they are not considerable and therefore that Table I gives a reasonable.
account of the 1948 population.
Estimated populations of eight Karamoja tribes in 1948; for explanation see text.

Males Females Totals
Adult Total
Jie ... ... 3,413 8,671 9,540 18,211
Karamojong ... 11,552 26,321 28,375 54,696
Dodoth ... ... 4,037 9,761 10,394 20,155
Teutho ... ... 270 586 554 1,140
Ngiangeya ... 730 1,607 1,639 3,246
Tepeth ... ... 1,728 1,728 3,456
Ngiakwai ... 324 689 808 1,497
Labwor ... ... 937 2,409 2,821 5,230
Suk ... ... 8,502 9,329 17,831

In 1919 officers of the King's African Rifles carried out a census of adult males
in the tribes of the District as then constituted. The general results of this are
given in column (a) of Table II. By taking the ratio of the number of male adults
to the total population as given in the 1948 Census (Table II, column (b)) and
assuming that this ratio was roughly the same in 1919, estimates can be made
of the total population of the tribes at the earlier date (Table II, column (c)).
Finally, an estimate can be made of the percentage increase in population during
the 29 years between the two censuses, Table II, column (d)).
It will be seen that there were considerable differences in the tribal increases
over the 29 years. In the case of the figure for the Dodoth (much lower as com-
pared with the Karamojong and Jie) it is possible that the K.A.R. officers
included some or all of the Ngiangeya group, thus artificially inflating the size
of the tribe in 1919. The smaller rate of increase of the Teutho is not perhaps
surprising in view of their remoteness and their conservatism in the old life of
hunting and gathering, with a little agriculture. As a personal opinion, it is sug-
gested that probably neither the 1919 nor the 1948 Census covered the whole
of this tribe, whose homesteads are scattered sparsely along the Rift Valley
escarpment, north of Morungole Mountain, as far as the Didinga Hills in the
Sudan. No estimate of the increase in the Tepeth tribe is attempted here, for it
would appear that the 1919 figures must have been much too low.
In any case the figures given in columns (c) and (d) of Table II must be treated
with reserve because of the inadequancy of the data and the nature of the
assumptions involved. They should be treated as indications only. It is probable
that the 1919 figures are rather too low in all cases, although the detailed results
of males per individual homestead throughout the country are quite impressive,

and show the considerable pains that the officers and their assistants took. It is
likely also that the ratio of adult males to the total population was lower in 1919
than in 1948, for at the earlier date warfare had only just ceased and therefore
the losses of men in fighting would still be important. If this was the case then the
estimates of total population for 1919 are all too low, although increases of the
scale indicated in column (d) of Table II are not unknown in the modern world.
It is clear, however, that there have been very considerable increases in the sizes

Comparison of the populations of some Karamoja tribes in 1919 and 1948; for explanation
see text.

(a) (b) (c) (d)
1919 1948 1919 Percentage
Total increase
Males Ratio (estimate) 1919-1948
Jie ... ... 1,869 5-3 9,906 83-8
Karamojong ... 5,298 4-7 24,900 119-5
Dodoth ... ... 2,674 5-0 13,370 50-8
Teutho ... .. 180 4-2 756 50-8
Tepeth* ... ... 167 4-7 785 -
*On Moroto Mountain only.

,of these tribes in the last three decades, which is also the conclusion drawn from
a study of family genealogies recorded among the Jie by the present writer. It
should be remembered that at the time of the first Census the tribes of Karamoja
were only just beginning to recover from the disastrous rinderpest epidemic of
the end of the nineteenth century, which indirectly caused a serious diminution
-of population. Even without the influence of modern conditions, therefore, it is
likely that there would have been considerable increases. With the gradual build-
ing-up of the herds, augmented by the cattle brought in by traders of the time to
buy ivory, general prosperity was improving, and thus most probably the popula-
tion were also rising. Of modern influences perhaps the most important in this
region have been the end of inter-tribal warfare and the introduction of external
food supplies. The end of war meant not only the end of many deaths in fighting
(and these involved all sections of a tribe, though mainly the younger men), but
it also meant the end of conditions whereby families were often reduced to
destitution by the sudden loss of their herds and flocks. The introduction of new
food supplies has meant that a higher standard of living has become possible,
and, more importantly, that at the periodic times of famine and near-famine
following animal epidemics and unreliable rainfall there has been the alterna-
tive supply of food. Indeed it is one of the responsibilities of Government to
ensure this. The sales of posho, beans and other foods in the dukas (shops) have
been rising steadily for some time now.

Enumeration in the 1919 Census was based on two indigenous units, the
homestead and the district. Among the Jie the homestead, today as then, con-
tains a group of (usually) agnatically-related people within a continuous, high,

palisaded fence. Inside this fence are the separate, enclosed, yards of each wife,
each yard containing her huts, granary baskets, fire-places, etc. These yards
surround a central kraal space which can be subdivided to make the goat and
cattle kraals of the various owners. Members of the same clan normally live close
together and their homesteads form a distinct hamlet. A number of these clan-
hamlets form a settlement, the chief mark of which is the communal possession
and use of a ritual grove. A number of settlements form a district and seven
districts comprise the total tribe (Table III). Each district was, and still is, a
unit for large scale ritual activity particularly in connection with rain-making.
In the old days, raiding parties were made up by distinct contingents from each
district, and each district took a proportional share of the final booty. Such a
share was distributed amongst the members of the district (not only the members
of the successful raiding party) by an acknowledged leader. There are no fixed
boundaries to a hamlet, settlement, or district, for they are primarily groups of
homesteads and their inhabitants, and not areas of land. In many cases there is
no discernible boundary between the homesteads of different units. Neither are
any of these units land-owning groups in any way.

A summary of the results of the 1919 Census. The second row of figures for each district
gives the numbers of adult males in each homestead.
Kotido 21 homesteads containing 176 men (average: 8-4 men)
8 (headman), 9, 13, 5, 7, 11, 8, 8, 6, 5, 10, 7, 7, 6, 8, 5, 14,
7, 9, 16, 7.
Kanawert 10 homesteads containing 142 men (average: 14-2 men)
42 (headman), 8, 11, 17, 8, 7, 14, 9, 12, 14.
Lothilang 9 homesteads containing 103 men (average: 11-4 men)
12 (headman), 18, 10, 12, 7, 12, 13, 10, 9.
Kapelimuru 32 homesteads containing 428 men (average: 13-3 men)
60 (headman), 9, 7, 5, 19, 11, 9, 15, 6, 10, 9, 11, 12, 18, 14,
14, 14, 12, 8, 7, 7, 20, 10, 18, 23, 9, 8, 12, 11, 12, 14, 14.
Panyangara 40 homesteads containing 493 men (average: 12-3 men)
12 (headman), 12, 17, 25, 16, 14, 11, 10, 8, 8, 14, 16, 24, 7,
11, 10, 12,14,9,11, 11, 6, 20,8,11,15,112,12,12, 7,11, 18,
7, 9, 6, 27, 9, 7, 13, 11.
Rengen 28 homesteads containing 330 men (average: 11-8 men)
Kotiang 21 homesteads containing 197 men (average: 9-3 men)
25 (headman), 8, 9, 6, 9, 11, 9, 5, 9, 11, 8, 8, 10, 7, 5, 4, 8,
16, 9, 13, 7.

In 1948 Kotido district contained 50 homesteads1 with 292 adult males (over
16 years of age in the Census Report). Thus there was an average of 5-8 men per
homestead, compared with 8-4 per homestead in 1919. Similar figures for
Lothilang district show 6-2 men per homestead (46 homesteads) in 1948 as com-
pared with 11-4 men per homestead (9 homesteads) in 1919. Thus not only have
homesteads increased in number with the rise in population, but individually
they are becoming smaller. This is, in fact, a process that has been going on ever
since the establishment of British rule, and it is still going on. Formerly a home-
stead contained all the members of one clan, or occasionally of two clans. These
1 Homesteads were actually counted in 1951, but those built since 1948 have been

large homesteads had already begun to break up in 1919, though not, for
instance, in Lothilang district. In 1951 a few of these old type clan-homesteads
remained in Wertakow settlement, Kapelimuru district, and in Caicaum settle-
ment, Rengen district.1
In Table III, the homestead of each headman (iakait) is noted. These head-
men were appointed by the military administration of the day. Doubtless at that
time the native districts were more readily identifiable than at the present time,
for then the Jie had only just acceded to the compulsory cessation of warfare;
but it should not be thought that their importance is negligible today. As far as
can be ascertained, it appears that the first-appointed headmen were the district
war-leaders, and that the 'assistant headmen' were the subsidiary war-leaders.
These men not only frequently went out in charge of raiding parties, but they
were acknowledged to have special magico-religious powers which enabled
them to foretell the right and propitious times for raids, and also enabled them
to strengthen and purify their followers and weaken the enemy. They were the
men who distributed captured livestock amongst the members of their districts.
It is most probable that such men rose to their position by virtue of personality,
skill, ambition and success, but not by a hereditary claim of any kind. They
tended to become wealthy as a result of the larger shares of booty which they
retained for themselves at the times of distribution. For instance, in Kotido
district today the sons and nephews of the war-leader Cowno, (headman in 1919),
are now almost the wealthiest in that area.
The 1948 Census, as already mentioned, was based on administrative areas.
In most cases, though not all, such areas coincided with indigenous native units,
but not all of the latter were of the same order. Some of the headmen's areas
involved a single settlement, some a whole district; a few involved several settle-
ments of no particular pattern. Unfortunately the outsider cannot ascertain what
sort of native unit is involved in any one case because no account has been avail-
able of the indigenous structure of the tribe. A further complicating factor arises
because, in the 1948 Census, in most cases the headman's area was named by
the clan-hamlet to which the headman himself belonged. By 1951 some of the
wakungu headmen, in office in 1948, had been replaced by men from other clan-
hamlets in the same area. For instance, the Kotido district (under one headman)
is named as Lokwori, the clan-hamlet in which that headman lived. In 1951 his
appointed successor lived in Logwele hamlet in another settlement. He did not
recognize the name Lokwori as having any relevance to his post, in which, of
course, he was quite right. Now, whilst it may not be considered necessary from
the administrative point of view that all wakungu areas should coincide with any
one type of native unit (especially since units vary considerably in the numbers
of inhabitants), yet it is desirable that there should be a fixed, conventional,
name for each headman's area, which can be readily identifiable and referred to
in the future. Such conventions have in fact already been accomplished in the
case of sub-county names throughout Karamoja District. A conversion table

1 The largest clan-homestead of which I have record broke up less than 10 years ago.
It contained the clans of Lotome and Lopotha in Panyangara district, and measured some
300 x 250 yards. In Wertakow settlement in 1951 there were at least two clan-homesteads
which contained more than 30 wives, each with her own yard.


(Table IV) gives a list of native names for the various wakungu areas with
reference to the ad hoc names to be found in the 1948 Census Report. By using
Table IV it is possible to establish a direct comparison, Table V, between the
two Censuses, since the earlier one was based directly on the native districts.

Showing the native names for the various wakungu areas with reference to the ad hoc
names to be found in the 1948 Census Report. The actual type of native unit involved, in
each case, is given in brackets.

Kotido (comprises the whole of
the native district of Kotido,
Kanawert and Lothilang)

Census name
Lokwori (clan-hamlet)
Longelep (clan-hamlet)
Nayethe (clan-hamlet)

Native name
Kotido (district)
Kanawert (district)
Lothilang (district)

Nakapelimuru (should be Kape- Longerep (settlement) Wertakow (settlement)
limuru district) Loriu (clan-hamlet) Kadoica and Potongor
Ngolemuru (clan-hamlet) Lokuruk (settlement)

Panyangara (comprises the native
district of Panyangara and the
sub-district of Kadokin)
Rengen (comprises the native
districts of Rengen and Kotiang)

Kadokin (sub-district)*
Rukruk (clan-hamlet)

Lukwori (clan-hamlet)
Lokaidi (clan-hamlet)
Chamchaun (settlement)

Panyangara (district)

Kotiang (district)
Northern Rengent
Caicaum (settlement)

The sub-district of Kadokin is ritually affiliated to the district of Rengen, from where
its people moved some generations ago; but it exercises more independence than other
settlements since they are now separated by other districts. Apart from current administra-
tive organization Kadokin has no connection with Panyangara, although the two units are
geographically adjacent.
t Northern Rengen comprises five distinct settlements for which there exists no com-
prehensive native name.

Comparison of numbers of adult males in certain districts of Jieland in 1919 and 1948.

District 1919 1948
Kotido ... 176 292
Kanawert ... 142 220
Lothilang ... 103 285
Kapelimuru ... 428 709
Panyangara ... 493 914
Rengen ... 330 744
Kotiang .. 197 249
1,869 3,413

In 1919 the same basis of enumeration was taken in these two tribal areas as
amongst the Jie; the figures for the indigenous districts are given in Table VI.
The present writer has not sufficient knowledge of either of these tribes to be
able to extract comparable figures from the Report of the 1948 Census.


The Adult male population in certain Karamojong and Dodoth districts in 1919.
Karamojong: Modern County
Ngipian ... ... 1,414 Pian
Ngimthiniko 1,238 Mathiniko
Ngimuno .80 ko
Ngitome ... 292
Ngimosingo ... ... 196
Ngingoloriet ... 1,044
Ngimerimong ... 348 Bokora
Ngipei ... ... 188
Ngikopo ... ... 498
Total 5,298 adult men
Dodoth: Modern sub-county
Nginkorituk ... 1,261
Ngitiiti ... 426 Loyoro
Ngikasimeri ... 575
Ngimeris ... ... 412 Kaabong
Total 2,674 adult men
No information concerning the modern sub-county of Morungole is available.

These indigenous districts are each composed of several settlements, but unlike
the case in Jieland it is probable that they are not further sub-divided into clan-
hamlets such that all members of a single clan inhabit one hamlet. In both tribes
clans appear to be scattered more or less throughout the country. On the other
hand, as with the Jie, districts are ritually autonomous, each owning and using
its particular ritual grove.'
[Information on the 1919 Census, which is derived from official files, is pub-
lished by courtesy of the Chief Secretary, Uganda Government. EDS.]

1 The names of districts are given here (Table VI) as they appear in the 1919 Census
Report. The prefix ngi is the normal masculine plural, indicating here, 'the people of... '.
Today these names are more usually prefixed with the locative lo-, e.g., Lotome, rather
than Ngitome. The locative prefix gives the name of the district itself, whereas the other
gives the name of the inhabitants of the district. It may be noted that although, for example,
Ngitome comes from the word for 'elephant', there is no totemistic or other special signific-
ance implied.
The origin of the county name, Bokora, is not known; it does not appear in early travel
literature or maps and may have come from the name of a single settlement or other par-
ticular spot. Cf. the map drawn by Capt. Chidlaw Roberts, 4th K.A.R., 2 November 1919,
on which these districts are shown for southern Karamoja.

THE infestation of south and south-east Ankole by tsetse flies, Glossina
morsitans, began about 1908 (Simmons, 1929), seventeen years after the
immense devastation of the Great Rinderpest. The result was that the Bahima
and their herds, which were slowly recovering after the rinderpest, were forced
out of the pastures of Ishingiro and Nshara (Karo Karungyi, "the good land")
into the drier, less-favoured country of Nyabushozi. When, in 1945, I was
stationed in Ankole to work on the tsetse problem, I asked Mr. Bulasto Kabenge,
then employed by the Tsetse Research Department, if there were no songs
recalling these disasters. He kindly produced the translation which I have para-
phrased below.
I had little time in which to discuss details of the song, and the brief notes with
which I conclude are my own interpretation of the events recorded. The old life
of semi-nomadic wanderings with the long-horned cattle is rapidly disappearing
from Ankole under the pressure of the tsetse fly and of the changes induced by
the British Administration in a sustained effort to bring the Ankole cattle
industry into line with modern economy. Many of the old customs have already
been well described (I append a short bibliography) but this additional record
of a not ignoble way of life may be worth preserving. The title of the song is that
given by Mr. Kabenge. Banika I take to be a plural form of nyinyeka, the owner
of a large kraal (Williams, 1938); enshubi are favourite cattle and enkurikizi,
tsetse flies. The latter are often called ebivu though this name includes all blood-
sucking flies. Enkurikizi originally referred to Glossina palpalis, a tsetse which
has existed on the shores of Lakes Edward and George from time immemorial.
Since the recent arrival of G. morsitans and G. pallidipes in Ankole, it has been
applied to them as well.

(The Lords of Ishingiro, their country being infested with tsetse flies.)
1. Magabi and Ntantamuki are burnt and Rwenkwanzi is bare.
The fires came down in Obufubuto and Nyakarinzi.
Mwanga and Mugugu are burnt.
The fires came down to Rwaisheremba.
Obusheka, Obugyemeko, Obwengyerero are destroyed.
Bisyoro is burnt and Kengoma, Nkyerengye and Minyangarara.
Obwibura in Buhunga is destroyed.
The flowers on the rocks of Nyamarungi are gone, that delighted the eyes of
the people.
Izhumuriro and Nyamiyonga are burnt.
To Rwachwa and Rwaisheremba the fire came down.
Murambiro and Nyanyindo are burnt and Bitete and Gabyalubambi are in

Mugugu and Rwaisheremba are destroyed.
The villages at Obugyemeko and Rwambaga are burnt.
The fires came down to Muchwa and Rwenkwanzi.
Rwomugina and Magabi are burnt.
Ntamtamuki is destroyed.

2. Mazinio, Mazinio, you whom the musicians followed!
Ikaza Misinga has been given Ntantamuki.
Rugonyesa is displeased with the beaten in battle
And displeased that the Bairu are blackening the towns.
The lion-like fighter does not fear, when face to face with foes.
He never shirks the battle.
Rukarangirwa's arrows, protecting parents, pierce the bodies of men.
Rugyereka trusts them, who at Mugugu and Rwaisheremba kept guard of
You fought eagerly, Banika! Tell me how you shared the land.
Igasira has won Magabi and Ntantamuki. Rutinampara now is lord of

3. You are lost, Magabi and Ntantamuki.
Rwenkwanzi you are barren.
At Gabyalubambi, poverty strikes the Banika lords.
Rutimburwa's cattle in Obusheka are dwindling, one by one.
The land is beset with flies.
At Izhumuriro and Nyamiyonga the vultures feast on the slaughtered herds.
The herds from Obwibura and Nyamiyonga met and died.
They were skinned by the Muslims and the Bairu.
Ishingiro and Kyangabukama are infested with flies.
The Bahima have nowhere to graze their herds.
Let us leave Magabi and Ntantamuki, they are beset with flies.
Let us go to other lands.
The enjoyments of your riches shall be yours no more.
Where once your calves were pastured the Muslim butchers stand and flies
swarm round their heads.
I commemorate Magabi and Ntantamuki, Nyakaigo and Nshungyezi,
Where now the Muslims bathe in the water holes,
Where once Rugaiha watered his enshubi, the enkurikizi breed.

The song clearly falls into three parts. The first part describes the grass fires
which occur annually in Ankole, in their progress from one pasture to another.
These fires, which usually are ignited by the Bahima themselves to provide
fresh grass, are mentioned in another song already recorded by Mackintosh
(1938), while Lukyn Williams (1938) describes a song dealing with a large fire
which occurred in 1900. I conclude, then, that there is little historical signifi-
cance in the first part of the song. The second part may concern a quarrel over
grazing rights or water-holes, or it may refer to rewards given following Ankole
raids into neighboring countries or, perhaps, since the defence of parents is

mentioned, to Banyaruanda raids which, in 1896, penetrated into Ishingiro
(Lukyn Williams, 1945). The "Bairu blackening the towns" are, of course, the
aboriginal Bantu of Ankole who were released from serfdom with the coming
of the British. The status of the Bairu in earlier times is described by Oberg
(1940) and doubtless there were many light-skinned Hima aristocrats like
Rugonyesa who disapproved of freedom for the black-skinned Bairu.
The third part of the song describes the effect of the advancing tsetse. Of
particular interest is the line about meeting the herds from Obwibura and
Nyamiyonga. It describes what must frequently have happened and, indeed, still
happens. A herd infected by contact with tsetse moves to other pastures. Here
it meets other, clean, herds. These in turn, become infected, not by the bite of the
tsetse, but by direct mechanical transmission of trypanosomes from beast to
beast by other blood-sucking flies. This part of the song definitely dates it at
later than 1910, when Glossina morsitans was first found at Nsongezi
(Nshungyezi) having arrived there via Nyakaigo in its advance westward up the
valley of the Kagera. From Nsongezi it branched north to enter the basin of the
lakes of south Ankole and, by 1930, Karo Karungyi was no more.

Mackintosh, W. L. S. (1938). Some Notes on the Abahima and the Cattle Industry
of Ankole. Entebbe: Government Printer.
Oberg, K. (1940). 'The Kingdom of Ankole' In African Political Systems edited by
Fortes, M., and Evans-Pritchard, E. E. Oxford: Oxford University Press.
Simmons, R. J. (1929). 'Notes on the tsetse belt in western Uganda.' Bull. Ent. Res.,
19, 421.
Williams, F. Lukyn (1938). 'Hima Cattle' Parts 1 and 2, Uganda J., 6, 17. 87.
Williams, F. Lukyn (1945). 'Nuwa Mbaguta, Nganzi of Ankole.' Uganda J., 10, 124.

May I correct a statement regarding Bulemo Ruigi, chief of Kitagwenda, which
appears in my 'Early Treaties in Uganda, 1888-1891', at pp. 28 and 41 of Uganda
Journal, vol. 12 (1948)? The original Bulemo or Bulemu was a grandson of
Olimi V, who was supplanted as Mukama of Bunyoro by Kamurasi c. 1852.
(H. H. Karubanga, Bukya Nibirwa, p. 11). This Bulemu was succeeded by a
son named Ruigi, who is the person who is alleged to have made a treaty with
Stanley in 1888. Ruigi died of small pox some time after Lugard's visit to
Kitagwenda in 1891, leaving an infant son, Kambere, who was subsequently
given the name of Bulemu. It was this last who was deposed in 1907.
May I mention also that there is in the Uganda Secretariat Archives a treaty,
on a Foreign Office standard printed form, "at Kasenyi in the Wavira country
on the West Shore of Lake Albert" on 12 December 1895, between Lieutenant
H. J. Madocks and Kavalli? It has never been published-probably because it
merely duplicated that made between Dr. R. U. Moffat and Kavalli on 18 April
in the preceding year (see H. B. Thomas-'More Early Treaties in Uganda',
Uganda Journal, vol. 13 (1949), p. 173). J. M. GRAY.
The High Court, Zanzibar.
January 1953.
I was much interested in Dr. Haddow's article on this subject (Uganda
Journal, 16 (1952), 164). The problem is not so much how conversation by
whistling is possible but rather why only one old man and his family in Bunguha
and another family in Kalegyalegya are the only Bakonjo who can practise
this art.
That there is a difference between a conventional or preconcerted signal and
a sentence made up of signals conveying information which can be varied at
will by the sender is clear. Bugle calls, conventional drum-beats, and the use of
a particular tune to convey information to the initiated are examples of the
former. Morse code or semaphore spell out words, naval code flags spell out
code-words and belong to the second category. 'Drum language' or 'talking
drums' also belong to the second category but employ a different principle.
It is well known that in talking any African language the correct tone and
stress is quite as important in speech as the correct word, and in many languages
the use of an incorrect tone is just as confusing to the hearer as using an incorrect
syllable. When Africans shout to each other from ridge to ridge it is difficult to
believe that they hear the consonants or vowels of the words used, but they do
hear the musical tones and the rhythm of the sentence. One can test this for
oneself, using for example the Luganda word Otyano, but care must be taken
not to distort tones or stresses in an effort to make one's voice carry a long
distance. By tone is meant that each syllable of a word has its own relative pitch,
high, low or intermediate. For.a long time the principle on which talking drums
in West Africa work has been known. Using two drums, one high, the other

low, in pitch, or a gong emitting a high tone when struck on one side, a low one
on the other, the operator beats out the tones and the rhythm of the spoken
words comprising the message..A recent book (Carrington J. F. (1949). Talking
drums of Africa. London: Carey Kingsgate Press. Reviewed in African Affairs,
48 (1949), p. 252) may be recommended to those who wish to enquire further
into this matter.
In 1927 I went on leave through the Belgian Congo and had the opportunity
of seeing and hearing talking drums in action amongst both Sudanic and Bantu-
speaking peoples, and could compare it with the 'howling', or more euphemis-
tically, 'yodelling', by which the Lugbara converse with each other when many
miles apart in the stillness of the night. My conclusion was that the Lugbara
yodelling is the vocal counterpart of the talking drums but whether this is now
accepted as a fact, I do not know. I was told then that the original home of the
Lugbara was a considerable distance west of the Nile and well within the talking
drum area.
Haddow mentions that the Bakonjo are "adept at yodelling and 'throwing
the voice' in various ways", and one wonders whether they use this art to the
same extent as do the Lugbara, for I have not myself heard them 'chatting' to
each other at night as one hears the Lugbara do in West Nile District, Kampala
and elsewhere.
Drum language is used by Bantu-speaking tribes in the Congo, but I believe
these languages are more dependent on tones than those of Uganda and East
Africa generally. Acholi is a tonal language and the Acholi are said to whistle
information to each other whilst hunting, but whether this is an elaborate
system of conventional signals or actual 'conversing', I do not know. The Madi
do not, I think, yodel. I have thought that one section of the Baamba and a
tribe on the southern slopes of Ruwenzori who speak in a sing-song fashion
(? the Banyabindi) speak languages which might be capable of being Psed in
this way. It would be interesting to know how far this art is developed in Kigezi
and on Mount Elgon. In West Africa and the Congo one knows that, whilst all
members of a tribe know a good deal about the drum language, some are more
skilled than others in transmission and reception, a point I can well appreciate
for I am myself very nearly linguistically 'tone deaf'.
If drums can be used, so then could a two-note whistle (instrument) and
perhaps it is, though it occurs to me that a whistle probably does not carry so
far as shouting. I imagine lip whistling is less likely to disturb game than the
use of instrumental whistling; on this supposition can we infer that lip-whistling
is a part of a hunter's equipment?
Thus the points requiring clarification seem to be: (a) whether this means of
conversing by whistling is a lost art of the Bakonjo, retained only in these two
families, or, (b) is an independent discovery by them, or (c) if these families stem
from immigrants from another tribe who practise this art. Haddow's observa-
tion that the Bakonjo yodel suggests that Lukonjo is sufficiently tonal in
character to allow speech-tones to be transmitted in this way.
The whole question of rapid dissemination of news in Africa is fascinating
and facts rather than theories are needed. My own experience has been that
there was usually a rational explanation, but many people prefer to believe in

talking drums in places where they do not appear to exist, and a distinguished
anthropologist friend of mine once told me that he thought such communication
was done by thought-reading mediums. Personally I am extremely sceptical
when people postulate supernatural means to explain phenomena which we do
not understand. ERNEST B. HADDON.
28 Barton Road,
Cambridge, England.
15 December 1952
Mr. Haddon refers to two Bakonjo families who can practise this art. It should be
noted that Haddow failed to confirm the statement that a family capable of
whistling signals lived at Kalegyalegya. [EDs.]

The Mill Hill Fathers' contribution at p. 168, Uganda Journal, vol. 16 (1952),
mentions the Ateso names of certain trees and grasses; and, as was there sug-
gested, the botanical names corresponding to some of these can now be supplied.
Place name Ateso name of tree or Botanical name
grass from which it is
Abata ebata Albizzia zygia (DC.) MacBride
Abia ebiya (ebia) Imperata cylindrica Beauv.
Abule ebule Ficus brachypoda Hutch. or F. platyphylla
Aderut aderut (ederut) Flueggia virosa Baill. (and other spp.)
Adodoi edodoi Kigelia aethiopica Decne. or K. moosa,
Ajanit ajan (ejanit) Sporobulus pyramidalis Beauv.
Akore ekore Gardenia jovis-tonantis (Welw.) Hiern
Alecer elecer (? = elecoroi) Crossopteryx febrifuga Benth.
Aminito eminit Acacia hebecladoides Harms
Amuria emuriai Carissa edulis Vahl
Aojabul ebule see Abule above
Apeduru epeduru Tamarindus indica L.
Atira etirir Acacia sieberiana DC.
Kapelebyong ebyong Ficus glumosa Del.
Kasilang esilang Ziziphus abyssinicus Hochst.
Kumi ekum Diospyros mespiliformis Hochst. ex A. DC.
Obalanga abalang atol Vernonia cineria (L.) Less., possibly
Odukait edukut Borassus aethiopum Mart.
Okere ikere Harrisonia abyssinica Oliv., possibly
Okoboi okoboi Terminalia spekei Rolfe
Pallisa eparis Grewia mollis Juss.
[Spellings used by Mr. Stephens, where they differ from those used by the Mill Hill
Fathers, are given in brackets. Ateso plant-names mentioned by the Mill Hill Fathers
but not yet related to botanical names are: amunyir (Amunyir); ecomai (Ocomai);
ekunguru (Okunguro); idodo (Dodos); ikeriau (Akeriau); ikorom (Akoromit); and
elwa (Olwa). EDS.]