Front Cover
 Front Matter
 Table of Contents
 The Uganda Society
 The stamps of Uganda
 A guide to the snakes of Uganda,...
 The effect of cloud on the behaviour...
 Back Cover

xml version 1.0 standalone yes
PageID P296
ErrorID 4

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/UF00080855/00013
 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
Publication Date: 1938
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: UF00080855:00013
 Related Items
Related Items: Uganda journal

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Front Matter
        Front Matter 1
        Front Matter 2
        Front Matter 3
        Front Matter 4
    Table of Contents
        Page i
    The Uganda Society
        Page ii
        Page iii
        Page iv
    The stamps of Uganda
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 148a
        Page 149
        Page 150
        Page 150a
        Page 151
        Page 152
        Page 152a
        Page 153
        Page 154
        Page 155
        Page 156
        Page 156a
        Page 157
        Page 158
        Page 159
    A guide to the snakes of Uganda, part XI
        Page 160
        Page 160a
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 176a
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 182a
        Page 183
        Page 184
        Page 184a
        Page 185
        Page 186
        Page 187
        Page 188
        Page 188a
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 194a
        Page 195
        Page 196
        Page 197
        Page 198
        Page 198a
        Page 199
        Page 200
        Page 200a
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 208a
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 214a
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 226a
        Page 226b
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 244a
    The effect of cloud on the behaviour of Tsetse Fly
        Page 245
        Page 246
        Page 246a
        Page 247
        Page 248
        Page 248a
        Page 249
        Page 250
        Page 250a
        Page 251
        Page 252
        Page 252a
        Page 252b
        Page 252c
        Page 252d
        Page 253
        Page 254
        Page 255
        Page 256
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text

J A4 ViViR* 140


A AOW A', w



1|+_ V+
'',+ GOG,

JE/y r!CLQ~
101 IE R"NMM'A"'N',3DV'rI NO
No AU$O:.-H .


15(1 "V~~Tff~l I I



a fIght

Brewed and bottled by

East African Breweries

Established 1922.

a ckair







The Murchison Falls on the Victoria Nile are one of the wonders of East
Africa-Here the waters of the mighty Nile thunder down in giant cascades from
the lip of the cliffs. The roar of the waters fills the gorge with awe-inspiring sound
as thousands of tons of water pour through the narrow cleft to the pool beneath
and spray leaps up caught and transfigured in the Sun-But that is not all. The
great pool at the foot of the Falls stretching from bank to bank is alive. Hip-
popotomi break the surface with great grotesque pink snouts and rise like creatures
of another world. The young of the species disport themselves on the rocks or
move lazily in the sun. Great crocodiles float in scores, disappearing with a flurry
on sound or movement to rise a hundred yards away. Their great bulk hides half
the glistening sand of the tiny coves among the giant trees. With a rushing
sound a monster emerges from the undergrowth, moves at incredible speed for the
water and is gone in an instant with a splash. Here in this enclosed gorge is all
the wild river-life of old Africa, unspoiled by man's intrusion. The Murchison
Falls are one of Nature's most closely guarded secrets but
holds the key to this place of enchantment.
2, Maritime House, Loveday Street,


Have you ever considered the difficulty you may
be confronted within 10 to 15 years' time?
Let the "OLD MUTUAL" show you how, at
very little outlay now,, you may secure, whether
you are alive or not, the education you intend
your child to have.





(Incorporated by Act of Parliament in the Union of South Africa).


Sirona House,
Sadler Street,

P. O. Box 359.

Resident Manager for East Africa.


General Managers for UGANDA (KAKIRA) SUGAR WORKS, Ltd. (Incorporated in Uganda).
Associated Firms (1) KENYA SUGAR LIMITED (Incorporated in Kenya).
(2) Nile Industrial and Tobacco Co., Ltd. (Incorporated in Uganda).
SUGAR Manufacturers, GINNERS and COTTON Merchants,
And Tobacco and Cigarette Manufacturers.


KAKIRA SUGAR WORKS :-Holding about I ,ooo acres of land, mostly under
cultivation, at mile 9, Jinja and Iganga Road. Employing about 5,ooo
Africans, 200 Indians, Europeans, Mauritians. About 36 miles of Light
Railway. Water supply to the Factory by means of pumping plant on
Lake Victoria.

TELEPHONES: Kakira Factory 125; Jinja Office: 29, 121, 79.
P. O. Box 54, JINJA (UGANDA).
District) near Mombasa. Box 158, MOMBASA.

UGANDA-(1) Bukoboli, (2) Busowa, (3) Bubinga, (4) Kamuli, (5) Mbulamuti,
(6) Kakira, (7) Kabiaza, (8) Butiru, (9) Kabiramaido, (10) Pilitok, (11) Amaich,
(12) Aboki, (13) Chagweri, (14) Batta, (15) Jaber and (16) Kalaki.
KENYA- Malikisi. TANGANYIKA--Ruvu and Kiberege.

Other Plantations totalling about 4,000 acres Freehold Land.

Uganda Journal.


Vol. V. JANUARY, 1938. No. 3.



The Stamps of Uganda ... ... ...

A Guide to the Snakes of Uganda, (Part XI) ...

The Effect of Cloud on Tsetse Fly ... ...

... ... by F. H. ROGERS.

... by C. R. S. PITMAN.

... by C. W. CHORLEY.


A Prehistoric Inscription in Ankole

Do Birds Eat Butterflies ? ... .

... ... ... by E. J. WAYLAND.

. ... by G. D. HALE CARPENTER.



Patron :
President :
H. R. HONE, ESQ., M.C., K.C., LL.B.
Honorary Vice-Presidents:
Committee :
Honorary Secretary:
Honorary Treasurer:
Honorary Editor:
Honorary Assistant Editor:
Representative in Great Britain:
Honorary Auditor:


January, 1938.
Members may well be tempted to believe on receiving their January Journals
that an attempt has been made to provide something in the nature of a bulky "Christ-
mas Annual" or "Double New Year Number", but we hasten to assure them that
even in spite of the new cover, of which we hope you will all approve, and the
amount of reading matter in this number, this was not our intention. In order to
get Captain Pitman's long awaited book on to the market in the New Year it was
essential to include in this issue an extra long and final instalment of "A Guide to
the Snakes of Uganda".
The last quarter of 1937 has been chiefly remarkable from the Society's point
of view, for the offer by Government to the Society of some premises in Nakasero
Road, Kampala, at the opposite end of the building at present occupied by Toc. H.
The premises have not yet been taken over but it is gratifying to be able to record
the prospect of the Society finding at length some place in which to meet and where
it will be possible to deposit the growing library and the early records.
We take this opportunity of drawing our readers' attention to the change in
the address of officials of the Society, and all correspondence should in future be
addressed to the Uganda Society, Private Post Bag Kampala. In order to prevent
the confusion which at present exists owing to the different and erroneous Post
Box numbers appearing on the Society's stationery, all the headed stationery at pre-
sent in use is being called in immediately.
There has been a most satisfactory increase in membership as a result of Mr.
E. F. Twining's "drive", but the small number of African members is still most no-
ticeable, and we appeal to those who have already reaped some of the benefits ac-
cruing to members to do their best to improve this state of affairs, the Society will
never really merit its name unless it can include a good proportion of African
In view of the already large Journal for this quarter it has been decided
to omit the Notices which have usually occupied not less than three pages and we
shall therefore confine ourselves here to the acknowledgment with thanks of our
reciprocating contemporaries.

The Stamps of Uganda.


The story of Uganda's Stamps is an interesting one and the following notes
are an attempt to reconstruct it, not too briefly as in the new Handbook (1) nor
technically as in a philatelic periodical (2). It is hoped therefore that this via
media will appeal to others than those who collect stamps.
This not being a general treatise on Stamp-collecting, I need not discuss
whether collectors are "Philatelists" or "Philotelists" or even merely "Stamp-
collectors", but I would warn readers en passant that it is a serious occupation and
that some of us might even seem to be a little touchy on the subject. Who, for
instance, when engaged on a brilliant piece of research into the authenticity of a
certain overprint, would appreciate hearing a caller informed "Oh yes, he's in; he's
busy sticking in stamps?" Collectors however can afford to be magnanimous be-
cause if the hobby continues to spread as it is doing at present, non-collectors will
soon be in a minority. Who knows, but that this very article may inspire an en-
thusiasm in some noble breast and send its owner searching among early postal or
Secretariat archives -though I expect the latter have already been combed pretty
These notes have been compiled from various sources. Some of the infor-
mation given may not be as full as possible, but so far as it goes, efforts have been
made to check and substantiate it. The writer will be glad to receive corrections,
addenda and criticisms, which will be co-ordinated and published with due
acknowledgment. And so, with this introduction, let us proceed to our subject.


For many years (and I do not doubt that, in many quarters, the ignorance still
persists) the existence of Uganda was known to very few persons in England; these
were either specially interested in foreign missions, a select few in the Foreign and
Colonial Offices, or were Stamp-collectors.

(1) "Uganda," Thomas and Scott. pp. 240, 241.
(2) Stanley Gibbons Monthly Journal, Feb. and March, 1903.

Most of the residents in this country are aware of something of the early
history of the various Uganda missions and the part they played, both out here and
in England, in the retention of the country for the Empire (1), but probably few
know of the reasons which have brought it to the special notice of Philatelists.
As a simple means of effecting prepayment of the cost of transmitting letters,
those small adhesive labels called Stamps cannot be improved upon where ordinary
private, as opposed to commercial, correspondence is concerned. They are fre-
quently made to serve other purposes as well, such as defraying certain foims
of tax, advertisement or the commemoration of persons and events, but the first
stamps of Uganda were utilitarian rather than sentimental. Their issue marked
a definite stage in the peace and administration ol a country which had just passed
through troublous times. They indicate too how anxious were the Europeans to
introduce the advantages of civilised life by attempting to inaugurate a regular
postal system for correspondence, both locally and to the Coast. Until about 1894
there seem to have been no regular postal or letter services, "runners" being sent
off with letters as occasion required. Mails from the Coast arrived quarterly or
half-yearly (or occasionally not at all (2),) and were distributed on arrival at Head-
quarters, the messengers who took them on returning in due course with the
replies. It then fell to "Bwana Tayari" (Mr. George Wilson) to experiment with
a proper system, complete with stamps, as a result of which Uganda has the dis-
tinction of being associated with the only stamps made on a typewriter. This in
itself may sound a matter of small consequence, but there is more in it than that; the
production of these stamps and their immediate successors typifies the beginnings
of that sturdy independence which has since consistently marked the development
of the country.
In the making of stamps numerous methods have been adopted from time to
time, the commonest processes being engraving, lithography and sundry variations
of these forms of printing. Probably the rarest means used is that of the type-
writer, the only other recorded occasion being a typewritten surcharge from Tonga
a year later, in May 1896.
Now there are sceptics who deny the authenticity of these typewritten "stamps"
and they have sufficient justification for their attitude to make it worth while
examining-their reasons. "A purely local effort," they say "and not issued with
any authority." It depends on what is meant by "authority"; the first Uganda
stamps were issued with the approval (indeed, at the instigation) of the local
Government and that, one would think, would be adequate; but it does not satisfy
the critics. If membership of the Postal Union is necessary, in this case it is
lacking, as Uganda did not join the Union till October, g190 (s), and it is also im-
probable that the Home Authorities were consulted and their approval first obtain-
ed. What exactly constitutes sufficient local authority (in the view of the stamp-
collecting fraternity) does not seem to be agreed upon. For instance, in Mafia
Island, off Tanganyika Territory, in I915 and in Long Island, in the Aegean, in

(I) "Uganda," ibid. p. 33. "18 years in Uganda", Bp. Tucker. Ch. XIII.
(2) v. Tucker "18 years in Uganda" Vol. I. pp. 201, 202.
(3) Sir Albert Cook, Uganda Journal, Vol. II. p. 104.

t916, provisional stamps were issued by the respective military and naval officers
in command, and many non collectors must be acquainted with the Mafeking pro-
visionals issued by "B.P." These three issues are recognized in the Gibbons cata-
logue and some handsome values are set against them. On the other hand, ano-
ther equally good catalogue says of the Mafia stamps "They appear to be local issues
and are therefore not listed;" while, writing of the Long Island ones, it says "The
issue......was a private one, being unauthorised. This has been confirmed by the
Postmaster-General." Where then do the first issues of the Uganda stamps come
in ? They were a local issue, so local in fact in franking-value that if a letter
was sent beyond the confines of Uganda itself (which then extended as far as
.Naivasha) additional postage fees had to be added, not by means of Uganda
stamps, but with British East African ones affixed by some agent or friend at the
But why worry ? Do we, as Stamp Collectors need to concern ourselves with
the original "authority"? "Yes," and "No." "Yes," if one is a speculator or has
no use for amateurish efforts; "No," if one's hobby is all-embracing. A collector
of Beer Labels (and there are such) might be very cross at finding that a treasured
Allsopp "He-man" Burton-brand label was a base imitation produced in Holland;
he would be inclined to discard it at once-though, on second thoughts, it might be
retained as a "variety." Again, in the case of a collector of Match-box Covers (and
there are such; I know one), is a "3 Stars" label marked "Made in Sweden" which
turns out to have been produced in Japan, an authoritative Cover or not? It must
be, after all, entirely a matter of individual taste what amounts to an "authoritative
issue." On the whole therefore, I suggest that the criterion should be the genu-
ineness of the purpose for which a particular article was produced combined with
a need for it, and if sufficient local or other authority or cognisance is added, then
the local Uganda stamps satisfy all requirements and may be accepted as genuine
stamps. Others, even more catholic in their attitude, consider that everything
which has any bearing on a country's postal or philatelic history may safely be in-
cluded in the scope of their collection. Moreover I suggest that these early Uganda
stamps should have a special value for us here, quite apart from any speculative
one; they have historical associations (and "history" in Uganda is very concentrat-
ed) and a sentimental value of which none need be ashamed. In a general way too,
they serve to illustrate how stamps in any country might have come into being.
Let us therefore ignore the carpings of the critics and accept the first Uganda
stamps as stamps and learn what we can about them.
The need for them appears to have been felt in T894-95 when, as already men-
tioned, the country had begun to settle down and minds could be turned to other
things than those of a purely military kind. Experiments in planting were made
and, seeing what importance is attached to the interchange of news with friends,
it is not surprising that efforts to establish a regular letter service should also have
been made. The Government had its scattered posts as far west as Toro and as
far north as Dufile, and elsewhere, and probably had a fairly regular system of
communication with them by "runner" services. The missions also would be
anxious to keep in touch with their various stations, and an obvious thing would be
to combine the two needs.

it is difficult nowadays for us, used to railway facilities and smart red post-
vans running here and there, to realise the vicissitudes which the carriage of
letters in those days underwent. Coast mails were lost, stolen or wilfully des-
troyed, and wild animals took heavy toll of the runners themselves. In 1902 a
solitary mail-runner was killed by a buffalo on the way to Kasaka, and a par-
ticularly dangerous section of another service was near Kyazanga, on the Masaka-
Mbarara route where another runner was killed and eaten by lions as late as 1914;
and one can readily appreciate thd risks from elephant on the Masindi-Gulu run.
Yet these men were wonderfully regular; in the course of some three years' con-
tinuous residence in Gulu I never knew the post more than half-an hour late,
having come all the way from Kampala via Masindi on foot, by a series of relays,
a distance of about 270 miles. Punctually on Wednesday mornings at 5.45 the
familiar whistle would be heard in the distance and in due course, with morning
tea, would come one's letters. It was customary for all "matalisis" to announce
their approach while still at some distance from their destination by means of a
whistle of horn which gave out two notes of a pitch peculiar to these men; this
enabled the next relay-runner to be ready to carry on the mail-bag at once or pub-
lished to the Boma the fact that the mail had arrived; frequently the Post-call was
taken up by the Police bugler. Many must have thrilled at the sound of these
runners' whistles, and in passing, a tribute may well be paid to the real bravery of
these men who faced their regular 25-30 mile marches, whether by day or at night,
armed only with an ancient rifle and a hurricane-lamp. It may surprise some to
learn that during the War, for a short time, all Kenya and Uganda letters were
carried in this way from Kampala via Masindi, Gulu and Nimule to Rejaf where
they caught the Nile steamer. Telegrams and cables from Uganda were also car-
ried by runner between Nimule and Rejaf, a dangerous stretch by reason of buffalo
and elephant. But we seem to have digressed somewhat from our original subject.
The need for regular postal services having been felt, a means was sought
whereby prepayment of the cost could be defrayed, and stamps would be the obvi-
ous method. There being none, local ingenuity had to make good the deficiency.
The then Acting Deputy Commissioner in Uganda, Mr. George Wilson, C. B.,
approached the Rev. Ernest Millar, an honorary missionary of the C. M. S., who
possessed the only writing-machine in Uganda in those happy days. Millar fell
in with the idea and agreed to co-operate, and so "the ubiquitous and philatelically
"iniquitous typewriter" (1) was brought into use for the first time in the making
of stamps. No printing-press was available as Mackay's press had been destroy-
ed in 1888 when the mission-house at Natete was pillaged by Arabs. The machine
was subsequently taken to Bukaleba (with the type, very much in "printer's pie")
and repaired by Mr. (now Canon) Rowling; it now rests, in honoured retirement,
at the C. M. S. Headquarters in London.
Millar's task of typing sheets of stamps must have been a most monotonous
one (though perhaps after the heart of many of our present-day typists). His "de-
sign" was necessarily of the simplest (PI. I. Fig. i.), nevertheless a number of errors
and mistakes crept in, but "they were surprisingly few in number, considering the
method of production." Such errors were caused, for the most part, by pressing
wrong keys or by wrong spacing, e.g., 50 over 51, 40 over 44,"U A,"' "UG", and so on.
(1) "Postage Stamps in the making." F. J. Melville. P. 143.

Suitable paper was not available, so Millar fell back on his sermon paper
it was a thin, brittle kind known as "laid" or verge' batonne and foolscap in size; the
colour of the ink-ribbon was black. One source of information states that some
duplicating carbons were used, but I am inclined to doubt this as they are not listed
and in those days all letters and despatches were written and copied by hand.
Much ingenuity had to be exercised in order to get a maximum number of
stamps on a single sheet of paper. At first Millar was content with nine broad
stamps in a row (P1. I. Fig. i.) After he had typed twelve rows there was no room
on the machine to allow of another row being got in on the remainder of the sheet,
so it was then turned upside down and re-inserted, and a thirteenth row typed.
This expedient still left a small blank strip of clear paper and in this space was
typed a summary of what each sheet contained, thus- "9 @ 1o, 40, 50, 6o, 100oo; 36
@ 20 and 25." v. G. M. J. Plate. (Secretariat Library, No. 534).
The design consisted of U G in the upper corners, standing presumably for
Uganda Government and in the centre a figure denoting the value in terms of
cowrie-shells (nsimbi, which term is still applied to the modern Cents). There
was of course neither perforation nor gum, but the latter was available in its
natural state for those who wished to use it.
When these stamps or "primitive labels" first reached England we are told (1)
that they were received with something like contempt and held up to ridicule, and
it was the editor of a leading philatelic journal who seems to have originated then
the time-worn joke that "U.G." might stand for "You Goose" as well as for "U-
Gander". This attitude is not found nowadays however, and for the most part
Millar's stamps are accepted as such, save by those who are still unconverted, as
mentioned above.
While a sufficient supply of stamps was being made, a postal service was
being organised, and a notice was given to Millar to type:-

(Translation, from Luganda)

Letters will be collected from this box twice daily, at 9 a.m. and 6 p,m.
Letters will be despatched thus from Kampala:-
"For Entebbe, at 10 a.m. daily.
"For Gayaza, at 10 a.m. daily.
"For Kikabya, Buzinde and Bulemezi on Monday, Wednesday, and Friday at 10 a.m.
"For Ngogwe, Kyagwe, Tuesday and Saturday at 10 a.m.
"For Usoga, Tuesday at 10 a m.
"For Mityana, Kasaka and Singo, Thursday at 10 a.m.
"For Kinakulya and Buwekula, Mar. 21, Ap. 4 and 18, May 2 and 16 and 30, at 10 a.m.
"For Budu, Mondays at 12.0 noon.
"For Koki, Mar. 25, April 1, 15 and 29, May 13 and 27 at noon.
"For Sewaya's Island and Bukasa, Sese Island, Tuesday at 12 noon.

(1) The London Philatelist, Vol. ix. No. 103. July 1900.

"Rates of Postage.

"Entebbe, 10 cowries; Gayaza, 10; Kikabya, 20; Buzinde,
"Bulemezi, 20; Ngogwe, Kyagwe, 30; Sewaya's Island 40;
"Bukasa Island, 40; Usoga, 50; Mityana, 40; Kasaka, 50; Kinakulya, 50; Santa Maria, Budu
50; Koki, 60.
"Only one letter may be enclosed in each envelope,
"If more than one letter is put in an envelope, the whole will be confiscated. Letters
insufficiently stamped will not be posted." (sic. ? forwarded).
This notice is not dated but Millar's diary refers to it under date March 16th,
1895; in another place he says that the stamps were first issued on March 15th.
The value of the cowrie at that time was 200 to the rupee or 121 to a penny. It
is believed that there is no record as to how many stamps were made or issued;
Millar himself estimates about 2,ooo, but admits that this is mere guesswork. At
any rate, he seems to have been kept busy and soon found it necessary to try and
get more stamps on a sheet; this was effected by making them narrower and get-
ting eleven in a row; their length remained constant (PI. I. fig 2).

If careful comparison is made between figs. z and 3 on PI. I., it will be notic-
ed that there is a difference in the type, the letters "U. G." in No. 3 are each nar-
rower than in No. 2. This is explained by the fact that Millar got out a new type-
writer which evidently had type slightly different from that of his old machine.
The new one came out about the end of April, as he states that he sold the old
one to the mission on May 6th, and it was during May that the stamps with the
narrower letters first made their appearance.

We now have the broad stamps and two kinds of narrow ones, but before go-
ing further I must mention some provisionalss" which I have seen described as
very rare" and "valuable to very valuable." It may be that they are, but I think
that, as financial considerations here may enter into the matter, their authentica-
tion should be established. Mr. C. J. Phillips says (1) "A few of the wide black
stamps are known with the original value struck out and a new value written in
with black ink. Some of these stamps I have seen initialled 'G. R. B.'-the initials
of the Rev. G. R. Baskerville, who before he became a missionary in Uganda
was vicar of a church at Birchfields, Birmingham. I can only presume that at
some small place the values in greater demand became exhausted, and that values
in less demand were converted into other values as demand arose. I have met
with these stamps dated in pen-and-ink as early as '7.5.95', but all without the
name of town or district. These are all very rare. I have not heard of more than
a dozen copies in all the collections I know."

(In speaking of "towns" it is evident that the writer visualises Uganda in
terms of Britain.)

(1) Gibbons' Monthly Journal. Vol. xiv. No. 164. Secretariat Library No. 534.

Now here is a case, I think, where some sort of official sanction is definitely
needed if these provisionalss" are to be accepted as properly issued; if it is lack-
ing, they can have no special value, least of all be "valuable"; without adequate
authority, and in a similar way any of us can create "rare" stamps daily. For
instance, having no ij-cent stamps in the house, I recently had occasion to put a
2o-cents stamp on a local letter where a 15-cent one would have sufficed. Had I
written on my 2o-cent stamp "15 cents, F. H. R." it would have "cut no ice" as it
would still have cost me o2 cents to send the letter. But if one could procure, over
a Post Office counter, a o2-cent stamp for 15 cents (the stamp being so marked
either in print or pen-and-ink), that "provisional" stamp and surcharge would have
an enhanced philatelic value, as no Postmaster is likely to sell his stamps at less
than face-value unless he is authorised to do so by his Headquarters, in which case
there would be official sanction, but not otherwise. Therefore before accepting
these "very rare" provisionals at their catalogue value, it is necessary to be satisfied
as to their official standing. Who was "G. R. B." and was he acting in any
official capacity in writing-down the value of these stamps ? An error in Mr.
Phillips's account is his identification of the late Archdeacon Baskerville, whose
initials were G. K. B., with the late Canon G. R. Blackledge; both these gentlemen
were missionaries and highly respected officials in their particular walk of life, but
this did not include (so far as I am aware) the writing-down of postage-stamps.
I cannot trace any Government official of these initials and therefore suggest that
proper authorisation of these pen-and-ink surcharges should be more fully establish-
ed before the high prices asked for them are paid. These surcharged stamps may
be few, but their philatelic interest and value may even be nil.

On the other hand, Millar's errors may with more reason be looked on as
rarities; such e.g., as the one illustrated at the top of Plate I. Close examination
will shew that it is one of his third essays (narrow stamp, narrow type) but instead
of the usual "U G", these letters have become reversed by a sort of lapsus memories,
and the typist inserted "G U" in error. The mistake would be a genuine slip
(as all who knew Millar will agree) and may therefore be considered in the nature
of an official "variety" whereas the pen-and-ink surcharges dealt with above, so
far as I am aware, are not. After getting the stamp "vetted", for paper, ink, etc.
I was very content as a Collector to pay 30/- for this crumpled up scrap of paper
with one of Millar's stamps (and an error at that) on it; it is a piece of a religious
publication and an interesting feature (which was probably unknown to the experts
who decided that the stamp was genuine) is that the handwriting is Millar's own;
the addressee, Canon Leakey, died only recently. Ernest Millar himself was a
great character and took much interest in King's School, Budo. Every Saturday
afternoon he used to cycle out there from Namirembe, and it was the resulting
strain on his heart which ultimately caused his death during one of these visits.

I hoped at the time of my purchase that this stamp was unique, but a recent
advertisement (1) asked ,j25 for a used pair of Millar's stamps both shewing the
error "G U" for "U G". At any rate, I seem to have picked up a bargain, which
is always a satisfaction.

(1) Philatelic Journal of Great Britain, June, 1937.

We might now begin to tabulate the various issues (2) Issue No. r. (P1. I. Fig i.)
Broad stamps, 20 to 26 m/m wide, in 9 values from 5 to 60 cowries. First on sale,
15th March, 1895. The variation in width is due to the occasional insertion of an
extra dash (or to the omission of one) owing to the monotony of typing. The top
and:bottom of each stamp is composed of dashes, and the side dividing-lines of
apostrophes. Colour, black. Issue No. 2. (P1. I. Fig 2.)
April, 1895. Narrower stamps, 16 to 18 millimetres wide, in the same 9 values
and type as in Issue No. I. Colour, black.

Issue No. 3. (PI. I. Fig. 3.) May, 1895. The narrower stamps with narrow
letters due (as above mentioned) to the arrival of a new machine. The same 9
values, and still in black.
Towards the end of the year Issue No. 3 made its appearance in a new colour.
In answer to the query "Why were the colours (sic) changed ?" Millar replied
"Because I put a different ribbon in my typewriter." Which is concise and saves
further speculation on the matter, though at first there seems to have been doubt
in certain minds as to whether the new colour should be described as Violet, Green
or Dull Olive-green. The ink of the new ribbons was probably of a "copying-ink"
nature which appears greenish in some lights or if applied thickly. Its official des-
cription has settled down as "Violet." We therefore get Issue No.4. End of 1895,
similar to Issue No. 3 but in violet instead of black, and with the substitution of
1oo cowries value for 60.
In the catalogue there is a note to the effect that "Many of the values of the
above are known tee-biche," which is a French term applied to stamps printed up-
side down in reference to one another, or, shortly, head-to-toe. All it means in
this case is that sometimes two or more stamps have been cut off together from the
bottom of a sheet, and as we have already seen that the last line was typed upside-
down in relation to the rest, stamps cut off together from the two bottom rows
would be toe to-toe, with or without the summary of values between them.

Let not the uninitiated be deterred from taking up the science of philately by
all these seemingly complicated details. Other sciences are equally meticulous
in their classifications, and it is entirely optional as to how far one cares to pursue
varieties, whether of watermark, type, errors, paper, perforations or what-not.
As already stated, there are some who entirely refuse to acknowledge these type-
written stamps, but they are dealt with here in detail as they undoutedly form part
of the history of Uganda's stamps.

An odd circumstance appears to be the printing of two special values, 35 and
45 cowries, for.an individual, one Dr. Ansorge. Sir Albert Cook informs me that
this gentleman was a Government Medical Officer who came out in March 1894
and remained till 1898, and that for a short time, in 1895 or 1896, he was acting
as Collector in Kampala. It is difficult to suppose that he wanted these stamps

(2) For a complete check-list and detailed description of variations of type, etc., v.
G. M. J. ibid. (Secretariat No. 534).



;r r .

UPO. ~rU
i' *r D


as "curios" when the values already on sale were curious enough, but it is possible
that he had ideas of introducing new rates of postage which were subsequently
disallowed. At any rate it is on record that some stamps of these values were
made and they are not considered sufficiently authentic to be given catalogue status.
The next change in the local stamps occurred in June, 1896; we read (1) that
towards the middle of that year "the Uganda Postal service was taken over by the
British "Military Representative," and the result was a slight elaboration in the
design of the stamps, which must have added considerably to Millar's typing trials.
My only specimen (P1. I. Fig. 4) is a somewhat inferior one, but a whole sheet is
illustrated in "Gibbons Monthly Journal" (Secretariat Library No. 534). We now get
Issue No.5(Pl. I. Fig. 4) June, 1896. In violet, having "V. 96. R" and "Uganda" in
full. Ten values, still in terms of cowries.
As regards the market-value of these typewritten stamps it is impossible to be
dogmatic. They have a very limited market and, as we have seen, some people
look at them askance. On the other hand, given proper circumstances, quite fancy
prices appear to be realized, e.g., in 1933 for a "block" of 39 stamps (the whole of
the vertical side of a sheet, 3 stamps wide) 37.10.0 was being asked, though the
"catalogue value" was about 6i 15. In the purchase of single stamps it is not diffi-
cult to get them for half or one-third of what is usually being asked. The demand
is small and stamp-dealers (like book-dealers, often enough) are glad of the turn-over
or the space which they have taken up.
I have been unable to find out much about the actual postal service of those
days. When Sir Albert Cook arrived, in February 1897, there was still no Post
Office in Kampala and he says that they got their stamps direct from Millar him-
self (who may have been a sort of "Sub-accountant") and that probably others got
theirs from the Collector (known as the "D. C." nowadays). It would be interesting
to know more about the whole organisation, and if readers can furnish any detail
I should be very grateful.
Meanwhile Mr. Rowling had been making experiments on the repaired mission
press at Bukaleba, a C. M. S. station some 4 miles east of the Fort at Luba's (or
Fort Thruston, v. 1/1,000,000 map of Uganda, No. A. 530) on Napoleon Gulf and
opposite Jinja. He sent over some of his trials and as they were considered
satisfactory, the Government once again fell back on the good offices of the mis-
sionaries for assistance. I cannot do better than quote Canon Rowling's own
account of his efforts :-
"To save time and make forgery less easy, I tried setting some (stamps)
in type while near Luba's, Busoga, in 1896. A sample was approved in Kampala,
so I set a block of 16. That was as many as my type could supply and even
then two letters were short; one a small cap, "o" and the other a thick "I",
so each was put in from another 'fount' to make up the i6.
"All were printed in lots of 16; paper then turned round, and another
16 printed 'top to top'. This was merely because the paper was of that size,
ordinary printing paper slips, cut from large sheets when the 4 pp. sheets for
Gospels, Hymn books, etc., had been cut out. These were for all values up to
4 annas.
(1) The West-end Philatelist, Sept. 1912.

"Values of 8 annas and over were printed on India paper (my own writing-
paper) from A. and N. Stores. Also making forgery not easy."
Plate II. shews one of Canon Rowling's sheets. Though the two panes of [6
are upside down in relation to each other, unseparated pairs from the two inner
columns are not true tete-biche. The paper used is white wove, except for the three
highest values(8 annas, r and 5 rupees), which are printed on thin, buff paper with
faint lines (PI. I Figs. 5- 11). Plating, or the re-construction of a pane, is possible
by means of the vertical lines between stamps. To return to Canon Rowling's
letter :-
"For each value the stars, daggers, asterisks, etc., were varied, so that the
value could be known even if numbers were missing; but the main idea was to
make forgery less easy. Government approved: ordered printing, and all were
sent to Kampala when ready. I think a second "edition" was ordered in 1897, but
I have no details now at hand. 29,000 (about) were printed in all, so my old diary
"The 'design' was my own, mainly conditioned by type available; we had not
much choice; it took about 12 months to get any new stock, even if it arrived! The
press was that used by Mackay, which I had to repair (broken main column) on
arrival in Uganda, early 1894."
"As far as I remember neither Millar nor I was paid anything for the stamps.
In any case, Govt. would probably pay some small sum, which would go to Mission
credit; no individual would get it. It would merely be to cover bare cost of paper
and materials used. One of my 'boys'did part of the actual printing, i.e., he helped
me, but I was always there in charge, so that none could be stolen or printed surrep-
These new stamps came into use on the 7th November, 1896, and appended is
a copy of the notice which introduced them :-


"Notice is hereby given that from and after ist May, 1897, the following scale
of charges will be made for the conveyance of Postal Matter by the Government
"Between Stations in any two districts in the Protectorate:-
Letters 3 annas each per oz.
Newspapers i anna per 6 oz.
Books and parcels 8 annas per Ib.
"Uganda District Local Mail (to include Luba's):-
Letters 2 annas each per oz.
Newspapers i anna per 8 oz.
Books and parcels 3 annas per lb.

.U A *; U A NA D A U a A N D A VL' 18 D A
V ,VIR I I.VtR, j *VtR, | ,VtR,
N A N D a Nu a N j I 1U A N r-A U I, A n A
*VtRl, VtR, I VtR, I ,VtR+
Pi'rUCTOrlATIr PIoriTTLit.nTr Pir'TF.TubTi j POrTEI.TORaTE
U 0 A N D a I i Al N D A U 0 a N D j U u I rl. D
*VtR,. IVtR, VtR., I .VtR*
1 ANNA 1 ANNA i 1 aNNA 1 A1NNA
U 0 A N D L I u A N L A L' .3 A N UD t 0 A N D A
.VtR, I ,VfR, I VtR. I ,VfR*
1 ANNA I 1 'AN.1r 1 uANNA 1 ANNA

RATIBOZYBd | LLrctLIMORUAj l) I1.s11Ud10 G | AMoia-LUdJ
VNKV I vain r I VNKV I vav t
* A I A I *HA* I*aA*
DI .LSOd luDYVJ d t- I i TVLS'. aDY.I D l
VUNYDO a Y,1 ..II| i| a l v aN
RI51101jzi0Hd | 1jTI 11ii-l .aii.'ai REI rLa.i..*u'^ 31I BI L.1i0cL I
vJ-v I v N v TI Yr v I VNsv I
*a+A* .I-A* I I *a-A*
S39vYsod SavL.-od I ol'V.Cd ovLSLnd
1 .] KN 'd i ] v a k 0. 11 ] | V .1 v Ii N|I 1 i a
.ITYo .l,')A I aI1 i_ .1iAiL c | d"J.1N,)IJ ioUU d [I aivA-IL'OItOd
*VA Iv viT aY I jT v1 V I T VN I
* A* I *aA* I "itA* I lA*
aov.Lso I 30ov.IsOd nviS aov.Isod
V U Ir r l a I i N r 0 ai v U N v D u U N V o A
aA* I A* *HaA* 1 *4a+A*
4OVISoCd I IOViS. rd aVI -d I OViSOJ
> -I v I, -a |I > % a A U L Y D. I T U NY A



"From Stations in Uganda, Unyoro, Toro, Usoga, and Kavirondo to Kikuyu.
Letters 4 annas each per oz.
Newspapers i anna per 4 oz.
Books and Parcels 1 rupee per lb.
"From Stations in the Mau District to Kikuyu.
Letters 3 annas each per oz.
Newspapers i anna per 6 oz.
Books and parcels 8 annas per lb.
"In all the above cases similar rates will be charged for the Up mails.
"No parcel will be taken which exceeds 11 lbs. in weight.

H. M. Acting Commissioner and Consul General."
It will be noticed that values are now in terms of annas and rupees instead of
cowries. This may have been to fall into line with East African currency in gene-
ral, or to anticipate a possible change in the exchange value of the Cowrie. At
this time it was still 200 to the rupee, and remained so till about 1900 when it went
down to about r,ooo, and, later, to 1,200.
There were two printings of all values from this type except in the case of the
5 rupees; the first printing consisted of some 29,000 stamps and the second one
(about the middle of 1897) of a further 30,000.
Errors and varieties are pleasantly few, in fact the only true one is the small
"o" in "POSTAGE" in the ninth stamp; this is constant for all values and is easily
discernible in Plate II. There are two different "I's"(Pl. I. Figs. 5 and 6); Canon
Rowling says the short, thick "I" came in the first printing, but Mr. Phillips gives
reasons (1) for thinking otherwise, but it is not a matter of importance seeing that
about equal quantities of each were printed. We are left then with only one true
variety, viz., the small "o," but those who are meticulous in such matters may like
to consider the dropped "ATE" in the 4th and 2th, and the broken 2nd "0" in
"PROTECTORATE" in the i4th stamps (all constant repetitions) to be collectable
A sheet similar to the one shewn on Plate II. is described in "The West-end
Philatelist" of September, 1912, as follows:-
1896, i anna, 2 sheets, tete-beche; this piece contains 4 tete-beche pairs,
2 stamps with small "o" in "Postage" and other varieties, very rare, thus, mint
condition ... ... ... t 1o.
I paid 30/- in 1932.

(1) G.M.J. Ibid. P. 190.

Again, 1896, 5 rupees, block of 8 shewing variety raised "U" in "Uganda"
and slanting "Protectorate". This is a much undervalued stamp, ... 15.

(Dealers do not leave a great deal to chance.)

A feature which puzzled me for a long time was the "L" which appears on
stamp No. 13 on Plate I. It is said to, and may well, stand for "Local", but as the
Uganda stamps then were only local in their franking capacity, why should some
be more local than others ? It is suggested that "L" stamps were for use only in
Uganda, but we have already seen that letters to any destination outside Uganda
had to be re-stamped (cp. Pl. I. Fig. 12). The application of this "L" is explained
I think by the terms of the first two sections of Captain Ternan's Notice, "Between
any two Districts in the "Protectorate" and "Uganda District Local Mail".
Swahili was the official language then, as now, and "Uganda" is the Swahili form of
"Buganda"; cp. "The Uganda Agreement, 1900" which of course only relates to
Buganda. "Buganda District Local Mail" is meant therefore, and stamps over-
printed "L", as on No. 13, were intended for use in Buganda only and to Luba's,
just across Napoleon Gulf. This distinction in local rates of postage was soon dis-
continued. The fact that this "L" is only found on the one-anna stamps with the
thin "1" suggests that this type of"1" was the first printing after all; this at least is
Mr. Phillips's view.

Hitherto, and till fairly late in 1898, cancellation was effected by pen or blue
pencil (P1. I. Figs 8-xo, 12, 13) sometimes with the initials of the canceller, and/
or the place of origin and date (P1. I. Figs. 8 and io). Postmarks only came into
general use in 1898. On Plates I. and III. some of them are shewn. Chief Mumia
is still alive (Oct. 1937); his Station gave place to Kisumu, formerly known as
"Port Florence" and Luba's has completely disappeared, though its site is marked
by the Thruston Memorial, a monument which is well worth a visit, especially by
canoe. "Kampala" and "Mengo" are, so far as I have been able to ascertain, syn-
onymous (No. 26, 28). No. 29 is interesting as the postmark is forged; it reads
"KIKURGU", a misspelling of KIKUYU, which place was never in Uganda, and
I can only suggest it was applied in London by someone either with a view to au-
thenticating his stock or in order to supply both used and unused specimens; in any
case, that type of postmark (square instead of round) was never in use out here.
Forged postmarks on the old Uganda stamps are not uncommon, and as a rule are
readily discernible; a clear definition and irregular characters are the main tests.

Issue No. 6. In 1898 there was received from England a handsome set of stamps
(Pl. III. Figs. 22, 23) printed by Messrs. De La Rue & Co. These finally super-
seded the locally-made ones, though the stocks in hand of the latter were allowed
to be used up. The set was deficient in the I and z2 anna values, and as these
were found to.be necessary, in 1902 they were supplied by the expedient of over-
printing the name "Uganda" on the corresponding values of the current British
East Africa set (P1. III. Fig. 24). In the same year the i-anna stamp was re-is-
sued in a rather different colour, owing probably to the difficulty experienced by
the colour-men in mixing the exact shade.


r '9. I o
CWoauIinfy Sa/vr $tS~*svj irn E?, w AFr IcA, 1Vef- I4oo.





Plate III. (14-21) illustrates contemporary stamps in use in British East Africa
from May, 1890 till 1903. It may be inquired why these could not have been used
in Uganda too, if "Kampala Hill was ceded to Capt. (now Lord) F. J. D. Lugard as
Agent of the British East Africa Company by Kabaka Mwanga in 1891." (1)
This regime was short-lived however, as the country was taken over by the British
Government on the ist April, 1893, when Sir Gerald Portal hauled down the Com-
pany's flag and hoisted the Union Jack (2). There seems indeed to have been a
strong distinction maintained between the two Administrations, as Sir Albert Cook
speaks feelingly of a "No man's land" which existed somewhere between Naivasha,
the last station in Uganda, and the nearest one in B.E.A. some 20 miles away. It
appears that this gap necessitated an additional rate for postage and was an uncer-
tain element in the regularity of the service. Fuller information on this point
would be of interest. At any rate, the inconvenience seems to have continued till
October, Ig90 when, as already stated, Uganda came within the Postal Union.
This meant that letters from Uganda, once stamped, were prepaid all the way to
Europe without the need for being re-stamped at Mombasa; parcels also went at a
cheaper rate, telegrams were accepted and postal orders issued, and the "No man's
land" disappeared.

Let us glance briefly, then, at the stamps in use in British East Africa during
the period we have been considering, v. P1. II. Figs. 14-21. They too are of in-

Issue No. i. May, 1890. The ordinary English id., 2d. and 5d. stamps were
overprinted "British East Africa Company" and "Half anna," "I anna" and "4
annas" respectively, by Messrs. De La Rue & Co. (No. 14).

Issue No. 2. In October, i890, a lithographed set of "Company stamps" was
sent out (No. I5). The design embodied the Crown and Sun which appear on the
"Mombasa Rupees" of 1888, and the motto "Light and Liberty," doubtless refer-
ring to the recent emancipation of slaves in Zanzibar and the East African Coast
generally. The stamps are found both perforated and imperforated, and there are
varieties of shades to be had, each value having one or more. There seems to
have been an early shortage of the I and i anna values, as in 1891, at various
times between January and June, alterations were made in manuscript and ini-
tialled variously "A.D.", "A.B." and "V.H.M." The initials were those of Mr.
Andrew Dick (the Company's representative at Mombasa), Mr. Archibald Brown
(Cashier to the Company) and Mr. Victor H. Mackenzie (Bank Manager in Mom-
basa). It will be noticed that all these gentlemen had some sort of official stand-
ing in or with the Company, and it is on this account that the official position of
"G.R.B." in Uganda needs to be established before his, and other, manuscript
alterations can as readily be accepted as genuinely authorised changes in value.

(1) Uganda Journal, Vol. II. p. 80.
(2) Cf. Memorial Stone, Kampala Fort.

'the need for 5 and 71 anna values was apparent in 1894; surcharged ones (on
8-annas and x-rupee values) were issued in November and definitive issues came
out in December. Mr. T. E. C. Remington, a Postmaster, initialled some more pro-
visional J and i-anna stamps (on 3-annas) in February, 1895 A remarkable thing
about his manuscript alterations is the neatness with which they are carried out.
There must have been a large demand for them in order to justify such a step, and
the tediousness of altering and initialling whole sheets of 240 stamps each must
have been very great. (Stamps of the ordinary size are usually printed in "sheets"
which comprise four "panes" of 60 stamps each, in ten rows of six. Variations of
this occur when a different size of stamp is adopted, as in the case of our current
ones out here).
In July, 1895, the Company's administration gave place to that of the Imperial
Government, which produced Issue No. 3. This consisted of the Company's stamps
overprinted "British East Africa" in three lines (P1. III. Fig. 16). They were
handstamped at Mombasa and numerous mistakes occurred some of which are sub-
jects for forgery nowadays. Probably the services of native employees were
utilised thus producing many double and inverted overprints. This issue was a
temporary measure and served to use up remaining stocks of the Company's
Issue No. 4. Further supplies being required prior to the receipt of new
stamps from England, current Indian stamps were similarly overprinted in Nov-
ember, 1895 (P1. III. Figs. I7, I7A.)
Issue No. 5. In May, 1896, a handsome set was received from England,
printed by the same firm as before, and inscribed "British East Africa Protectorate"
(P1. III. Figs. 19 and o2). They comprise 15 values, from J anna to 5 rupees; the
design is neat and the colours pleasing. Some 18 months later the rupee values
were issued in a larger size (Fig. 21) which, in the coloured stamps, produces
a much less satisfactory effect; the neatness is lost and the colours appear garish.
Issue No. 6. For some reason there was issued in January 1897 a series of
six Zanzibar stamps overprinted as before (P1. III. Fig. 18). It is possible that
something had occurred to delay a fresh supply from England (cp. Coronation
mugs) and that these Zanzibar stamps were used to tide over a shortage; but how-
ever it might be, during the life of Issue No. 5 (1896-1903) there were in use
in British East Africa for a short time six of the Zanzibar stamps. It is a pleasant
feature to notice how accommodating were the two Administrations at Mombasa and
Zanzibar. In 1896 B.E.A. stamps (Fig. 19) were in use in Zanzibar, while in the
following year the reverse happens, and stamps, as Fig. 18, were in use in Mom-
basa. Issue No. 5 carried on till 1903 when they were superseded by the stamps
of the "East Africa and Uganda Protectorate", which brings us to the final part of
this paper.


In 1903 the postal services of the two East African Protectorates were com-
bined and a series of stamps issued for common use in both countries.

Issue No. i. 1903-04. A handsome set printed from two plates and mostly
in two colours; in terms of annas and rupees and with the profile of King Edward
VII. (PI. IV. Figs. 30, 31). Watermark, single Crown and C A. For the rupee-
values a larger size was adopted, fairly similar in design; watermark, Crown and
CC (Fig. 32).
Issue No. 2. The following year those responsible for the production of
stamps for the colonies decreed a change of watermark throughout the colonial
empire, from the single "Crown C A" to a "multiple" form of it, whereby several
Crowns and C A's appear on each stamp. There were periodical changes of water-
mark in this manner, the next one occurring in 1922 when the "script" watermark
came into use. In the case of the "multiple" Crown C. A, one of the reasons for
its introduction was to ensure that all stamps in a sheet received a watermark;
with the "Single Crown C A" some stamps, owing to faulty printing, were liable to
escape a watermark altogether. Further, with the Multiple Watermark the same
paper could be used for stamps of varying shapes and sizes, which was not the
case when one watermark was intended to fit each stamp; in the latter case water-
marks would have to be spaced according to the size of stamp to be printed, which
would both add to the cost of production and probably result in numerous errors.
The multiple Crown CA was extended to the rupee values also. The currency
was still annas and rupees.
Issues 1 and 2, being in two colours, had to be printed from two plates, a
"Head" or "Key Plate" and a "Duty Plate." This applied also to the four values
in only one colour as can be seen by the irregular centring of the king's head. In
191o the 6-cent red was printed from a single plate and is distinguishable from its
predecessor by the omission of four lines of shading to the right of the leaf nearest
to the king's beard; there are several other small points of difference too.
Two sets of issue No. 2 can be made up, one on ordinary paper and the other
on "chalky paper." Some papers are coated with a chalk or enamelled surface, a
device employed to render it impossible to clean off cancellations without destroy-
ing the impression of the stamp itself. The test of a chalk surface is to apply silver
to it, which leaves a grey or black mark on the prepared ground. For purposes of
this test I formerly used a 6d-piece; this is a clumsy method and as my staff mis-
took the coin for a thumuni, my stock of sixpences soon ran out. A tarnished old
silver fork with a sharp prong makes a much better job of it, and I am not continually
searching now. In issue No. I the chalky paper is confined to the rupee values, but
in the next issue (No. 2) all values are to be had with both surfaces except the 2'-
anna which is only found on ordinary paper, and the rupee stamps which are print-
ed only on chalky paper.
Issue No. 3. In 1907 the local currency was changed from annas to cents (of
a rupee) and this Was reflected in a new issue during 1907-08 (PI. IV. Figs. 33, 34).
The man-in-the-street now benefitted; hitherto one had obtained 16 one-anna (id)
stamps for a rupee and now 16 6-cent stamps (the approximate equivalent) were
still obtainable; but 16 x 6 = only 96 so an additional four cents, to make up a
rupee, were thrown in, a 4% profit as it were and a virtual "something for nothing"
which made it a satisfying proposition to deal with the Post Office! The large
rupee stamps of the previous issue were retained; Fig. 32.

Issue No. 4. A new reign means new stamps everywhere, and the Edward VII
issues were followed in 1912 by a George V. set (PI. IV. 35, 36), which lasted until
1922 without a change except for the new watermark already mentioned. More of
the values were printed in single colours and two of them (2; and 75 cents) on
varieties of paper, i. e., either coloured on the surface only or dyed throughout.
The rupee values were still of the larger size (Fig. 37).
In 1919 the rate of postage to England and the colonies was raised from
6 cents to io. Existing supplies of the orange lo-cent stamp were insufficient to
meet the new demand and the 6-cent one was almost superfluous, so some of the
stocks of the latter were surcharged "4 cents" with a series of bars obliterating the
old value (P1. IV. fig. 39A). Stamps of the ordinary size are usually printed in
Sheets of 240 stamps, divided into"panes" of 60 stamps each of which are separated
by "gutters". The surcharging was done locally and the 60 (or 240) overprints
did not tally exactly with the size of the stamps, so that whereas the top rows on
a sheet or pane were overprinted correctly, as on fig. 39A, further down the bars
and surcharges tended to get higher up on each stamp, as on fig. 39B, till on the
bottom rows the bars had disappeared altogether and only "4 cents" appeared on
them, as on fig. 39C, much as, at a given hour on consecutive nights, the position
of the rising moon varies. Fig. 39A and B are each catalogued at id. but 39C in
which the bars are "omitted" is priced at from 10/- to 30/- according to the cata-
logue used. Strictly speaking, on 39C they are not "omitted"; they have merely
got up into the stamp above, but from a single specimen from a bottom row
it is not possible to deduce this; it would require two rows to make it clear.
Some sheets got put into the press twice resulting in a double overprint (40/-
to i20/ ); others were put in upside down ("inverted overprint," 20/- to 8o/-), and
in a few cases the bars seem to have been genuinely omitted; perhaps there were
not enough. "Bars below figures" is merely the result of the gradual displacement
already referred to. One thus sees the "varieties" that a little carelessness may
result in and it is these shortcomings which give zest to the search for "rare pieces".
It was carelessness too that allowed the sheets of inverted and double overprints
to be put on sale instead of being withdrawn, as occurred in the case of the inverted
"G.E.A." on the io-cent stamp overprinted in 1922 for use in Tanganyika. One of
the defective sheets was sent by Mr. (now Sir) Claude Hollis to King George V. for
his collection and the others were destroyed under the supervision of a Board
which had been appointed to go through the sheets and see that no errors were
allowed to pass. I was unwise enough to buy one of these inverted overprints
before getting it "vetted" and have since ascertained that the overprint is a forgery.
It has however some interest and will find its way into the collection of forgeries
which is maintained by the Junior Philatelic Society.
Issue No. 5. The next change, which occurred in 1922, was an important one
and is within the memory of many who are out here now. It embodies the change
of currency from florins to shillings (P1. IV. Figs. 40 to 43). In 1919 the value of
the rupee rose, in terms of sterling, from is. 4d. to about 2s. 6d. It was not suffer-
ed to get as high as that in East Africa where it was stabilised at 2/- and in 1920
converted into a Florin. There were never any stamps of florin values because the
florin was shortly after split up into two shillings, as we have them today, and the


--- +I ---


44 46-


higher value stamps went immediately from terms of rupees to shillings. This appre-
ciation of the rupee was a windfall to those whose salaries were based on the pound.
but was correspondingly disastrous to others and resulted in the bankruptcy of not
a few estates. It also took with it our letter-rate from i d. to 21d. After a short time
the rupee, in India, receded to about Is. 6d., where it has remained ever since, but
as the local currency had already been changed, we were left with the florin, and
our stamps had to be altered accordingly.
By now most countries had departed from the system of "universal colouring"
whereby all stamps of the equivalent value of 1d. were green in colour, of id. were
red, and of 2 d. were blue, corresponding to the charges for postcards, internal and
foreign letters respectively. Rates of exchange had gone their several ways and
the economy of most countries had necessitated increased postal rates, much as has
happened in our own case. The present (1937) 5 cent and 30 cent stamps still ap-
proximate to the "universal" colour scheme.
In the 1922 issue there was some attempt to make the cent values aliquot parts
of a shilling. In the previous issue (No. 4) this had not been done(e.g., 3, 6, 12 and
15 are not aliquot parts of 0oo), the Post Office, in those days of financial stability,
having been content to translate the former rates into the new (cental) currency
but in the 5th issue, of the old values only the 12 and 30 cent values survived, the
former very soon dying an unmourned death. Some changes in colour were made
before this set (No. 5) finally settled down to its long existence of thirteen years.
As regards colouring, a singular anomaly has persisted out here since 1922 viz.,
that the one stamp which is most used for cancellation in ink has always been black.
In 1919 when the letter-rate was increased from 6 cents to ten the legal stamp-duty
for receipts remained at six cents which, under the new (shilling) currency, became
twelve cents. The 12 cent stamp was then black, and in response to representa-
tions that (i) a black cancellation is difficult to pick up on a black stamp, (ii) the ro
cent would be more convenient, as obviating the need for stocking two stamps of
about the same value for two different and common purposes, the stamp-duty was
reduced (though not till 1929) from 12 to to cents, and writing or other cancellation
was readily discernible on the then green io cent stamp. No sooner had this
change been effected than green was adopted for the 5 cent value (hitherto violet)
and the poor receipt (io cent) stamp became black again, thus reproducing the old
commercial difficulty. And in the pictorial issue of 1935 which succeeded, sure
enough the hideous 1o cent stamp (Fig. 44) is again blacker than any other in the set!
The. design of the shilling values is generally considered to be a poor one and
their tints are dull. It should be noted that the ink is very liable to run in the
case of the one-and two-shilling stamps; this is another safeguard against cheating
by the obliteration of postmarks. Apart from the really high values, io/- and up-
wards, there were three values, rarely seen, viz., the 2/50, 4/- and 7/50 notwith-
standing the fact there was a four-shilling parcel rate. In order to procure these
stamps, it was necessary to write to the Chief Accountant at Nairobi.
In 1928 tax-stamps (or "Postage Dues") were introduced (Fig. 47), the use of
which is to facilitate checking the amount received for understamped or taxed
letters. At first sight, seeing that letters are taxed double the deficiency, it is hard

to appreciate why some of these "postage due" stamps should be in odd, as oppos-
ed to even, values. The reason is that the sum to be collected is adjusted to the
nearest centime, the currency employed by the Postal Union, the clearing-house of
international postal accounts.
The making of Postage Stamps and the consideration given to new issues
or values must be a lengthy process; otherwise it is a curious thing how often there
emanate stamps of which the life is of short duration (sometimes only a few weeks, or
even days) or the need for which has ceased almost on the day of issue. Examples
of this are not far to seek and Issue No. i of this Part is a case in point; the local
pictorial George V. stamps provide another instance. In the first case, after less
than a year the paper was changed. The second one is that of the pictorial 65 cent.
This pictorial issue appeared on May ist 1935 and the 65 cent defrayed the cost of
an air-mail letter to England. On May 6th the Jubilee stamps were issued and as
these included a 65 cent stamp, the pictorial one was withdrawn. Shortly after,
on July ist, the air-mail rate Home was reduced from 65 to 50 cents. This
naturally greatly limited the demand for 65 cent stamps and the pictorial one might
never have been put on sale again having been withdrawn after an ephemeral
existence of less than a week. (As a matter of fact this did not happen; when the
Jubilee stamps were withdrawn at the end of 1935 the pictorial equivalents returned,
but the sale of the 65 cent one must have been small and in commercial value it has
naturally appreciated more than any other of that set.)
Again, in 1921-22 most of the 1912-22 issue was superseded by a new issue
similar in design but with the new "script" watermark. In 1922 the new shilling
currency stamps came out displacing the "script" ones which thus had a very short
existence. At the time of the second change I was serving alone in Gulu and, as
local Postmaster, had access to Post Office circulars. One of these announced the
forthcoming change in currency and design and as none of the intervening "script"
stamps had been supplied to Gulu (where sales were small) I at once wrote to
Masindi for some of them for my collection, only to receive the reply that their re-
maining stock had already been returned to headquarters. So in the end I had to
get my set of "script" stamps from a dealer in London. It is such short-lived
emissions as these which appeal to the heart of the more financially-minded of stamp
collectors, though even the others sometimes have to be on the quivive if they wish
their collections to be complete. But surely the change of currency, involving new
denominations in stamps, must have been already under consideration when the
192 i-22 "script" issue was being ordered or in preparation, so it is difficult to see
why the change was made at all. It must have been due to the absence of that
Liaison Officer who is such a necessity in every bureaucratic government.
For some time prior to 1935 suggestions were made continually that East
Africa should follow the example of other countries and issue stamps in pictorial
designs; the advertising value of such a set was emphasised and perhaps the mod-
ern craving for change and variety had something to do with it. However that
may be, in May, 1935, a pictorial set was issued and so we got Issue No. 6. (P1. IV.
Figs. 44, 45). Opportunity was now taken to announce the fact of the combined
postal services of Uganda, Kenya and Tanganyika. The size, designs and colora-
tion.of this-set have deservedly been much criticised (1) and the advertising value
(1) ep. Minutes of Association of Chambers of Commerce of Eastern Africa, August, 1936,

ean be but a modest one. Many regret the departure from the earlier idea of mak-
ing the sovereign's head the predominant feature of stamp designs. In the new
set the king's head, in most of the designs, is carefully tucked away in a colonial
landscape which artistically is of quite indifferent merit, as e.g. the 1/- value. The
best general design is probably that of the Ic., 2oc. and 1o/- stamps (crested cranes
supporting the king's portrait), and on the 3oc. and 5/- the king's head is given
reasonable prominence. The 5c. and 50c. illustrate a competitor of the Uganda Rail-
way in full sail, and many are the criticisms of the horrible design and colours of
the ioc. and Ii stamps. The 30c. value is notorious for the incompleteness of its
design, in that the road from the Jinja side of the bridge leads on to it and there
stops, there being no way off at the Kampala end. Commercial firms found the
similarity of the 5c. and i/- stamps confusing; each one serves a common rate of
postage and expensive mistakes were liable to occur. The 1935 series included
also a new set of "Postage Dues."
Issue No. 7. In common with 43 other Colonies, the Jubilee of the late King
George V was commemorated out here by a "short set" of four values, viz., 2oc.,
30c., 65c. and i/- which temporarily superseded the corresponding values of the 1935
pictorial issue. They were of a somewhat ornate design shewing Windsor Castle
and were on sale from the 6th May, 1935, until the end of the year. Of these four
stamps the 65c. value is likely to appreciate most as the air-mail rate to England
was reduced to 50 cents on the succeeding ist July and comparatively few of the
65c. must have been sold. In the Jubilee stamps there is only one constant variety
and that occurs on the 2nd stamp of the 4th row in the i/- value; it consists of an
oblique stroke through the "o" of "I910."
Issue No. 8. On Coronation Day, 1937, the 5c., 2oc. and 3oc. values of the 1935
pictorial set were temporarily withdrawn in favour of three commemorative stamps
of similar values and of a design common to all other Colonies. Locally there are
plentiful supplies of the used Coronation 5c. to be had, but the other two values are
comparatively scarce in used condition as their denominations defray the cost of
postage overseas and most of them are therefore sent away. A fair proportion for
the ordinary individual to come by is about half-a-dozen of each of the two higher
values per hundred of the 5c.
The pictorial 5c., aoc. and 30c. will be on sale again after January ist, 1938
but it is expected that a new George VI. set will shortly be issued, and it is tobe
hoped that not only will the new stamps be of a more reasonable size and of at-
tractive design, but that no changes will be required for many years to come.
This brings us to the close of our survey of Uganda's stamps and I hope it
has been shewn that it is not without interest. There is a close relation between
our stamps and the country's vicissitudes, whether political or financial. Make-
shift stamps were used in makeshift times, one sovereign follows another, financial
straits are reflected, and the development of the postal service is illustrated in
"Uganda," "Kenya and Uganda," and "Uganda, Kenya and Tanganyika."
I hope therefore that these somewhat discursive notes will be the means of
quickening an interest in our local stamps-if not in stamps generally-and that
some will find a new pleasure in what can be an absorbing hobby.

A Guide to the Snakes of Uganda.



Puff Adder.


(Plate XV, Fig. 1: Coloured Plate (R), Fig. 1).

Native names -So widely distributed, dangerous and conspicuous a species is
naturally known by an exclusive name in each of the dialects which are used in the
various parts of Uganda. Unfortunately it is not possible to give the full list. The
best known is probably the Luganda and Lusoga name "Essalambwa" denoting the
snake which has "to turn over or throw itself on its back before it can bite", and
refers of course to the instantaneous sideways and backward movement of the head
-swifter than the eye can follow-in order to get the long fangs into striking posi-
tion. In Lunyoro and Lutoro it is called "Mpoma": in the Bagungu region on
the north-eastern shore of Lake Albert, "Bulabundoo": in Lunyankole, "Mpiri" or
"Mpili:" in Lugishu, "Chikorviri" (Sebei) and "Ki-iri": in Teso, "Aipomiro" or
"Asigirikolongo": by the Karamojong, "Akipom": by the Lango, "Choichodo": by the
Acholi, "Olwero": and in Kiswahili "Boma" and "Moma" for adults, and "Kipili"
or "Kipiri" for juveniles. Corkill (1935, p. 27) mentions two Arabic names used in
the Sudan-"Dagar", and more usually "Nawama", that is the 'sleeper'; "this latter
name being obviously inspired by the Puff Adder's torpidity". In parts of Uganda
there seems to be a certain amount of confusion in the use of the term. "Mpili" or
"Mpiri" which in some localities is applied to cobras and in others to the Puff
Distribution-The Puff Adder is probably the most widely distributed African
snake ranging as it does from Southern Morocco and south of the Sahara generally
through. most of Tropical and South Africa to the Cape of Good Hope, also being
found in Arabia. It is equally at home at.an altitude of.7,500 feet as atsea level,
though it does not occur within the borders of the Rain Forest.


1. Bitis arietans.
2. Bit/s gabon/ca.
3. Bitis nas/cornis.

la. Lateral Section.
2a. Lateral Section.
3a. Lateral Section.

Presented by MC A.S.Vernay.


IF ft

: Occurrence mn Uganda-This well-known species is almost ubiquitous inf Ugh-
nda and is only absent from the higher altitudes, the larger forests, and a small
area in Western and South-western Kigezi (previously mentioned as being of
special zoological interest). It is unnecessary to record specific localities as it is
only its absence which is noteworthy.

Description-The Puff Adder is a massive, fat-bodied snake of bloated ap-
pearance which, in the parts of the Protectorate where both occur side-by-side, is
frequently confused with its near relative the Gaboon viper (Bitis gabonica).
The maximum size attained by the Puff Adder has been the subject of consider-
able controversy, and though it is claimed that on occasion a length of six feet has
been reached, and even exceeded, the only reliable published record of such antenor-
mous dimension is Sir Harry Johnston's reference in The Uganda Protectorate to a
Toro example longer than six feet. From personal field experience it can be stated
that Bitis gabonica not infrequently attains a size which is evidently exceptional
in B. arietans. In the East African Sportsman's Handbook (1934) a length "up.to
5 feet 5 inches" is mentioned, the maximum being abnormally large as can also be
considered any measurements of 5 feet and over.
Sir Samuel Baker in The Albert Nyanza (pp. 223-234) describes the killing of
a huge Puff:Adder in the Latooka country in the Sudan which measured 5 feet 4
inches in length and had a girth of 15 inches, the head was about 2, inches broad
and the fangs nearly an inch long.
The largest Puff Adder measured personally was a male killed in the Ba-
gungu region on the north-eastern shore of Lake Albert which taped 4 feet and 9
inches and weighed (stomach empty) 5 lbs. From the same locality three females
were collected, approximately 4 feet in length, 3 feet io0 inches (weighing 7 lbs.)
and 3 feet 81 inches (weighing 6 lbs., including its meal of five large rats), and a
male of 342 inches. Another large example obtained in Southern Ankole was a
little over 42 feet. Dr. G. D. Hale Carpenter, in A Naturalist on Lake Victoria,
refers to a specimen of 4 feet 6 inches which he saw killed on Damba Island in the
Victoria Nyanza.
Loveridge (in lit.) mentions:- "In former days it used to attain to a length
not far short of five feet, today three feet is a more usual size except in remote
districts." The same author (1936) refers to a Uganda or Kenya example:- "55
inches long, being the biggest I have seen. Though rather emaciated and its
stomach empty, it weighed 6 lbs. on a spring balance." Prior to this Loveridge's
largest specimens had been obtained respectively, at Kilosa in Tanganyika Ter-
ritory, a female of 43- inches (The Snakes of Tanganyika Territory), and then on
Ukerewe Island at the southern end of the Victoria Nyanza, a male of 451 inches
(Bull. Mus. Comp. Zool., Vol. LXXIV, No. 7, October, 1933)." Referring to the
latter he records:- "Big snakes ........ appear to be of normal occurrence on
Ukerewe for this was only one of three large Puff Adders-all males over 424 inches
-brought in together slung on a pole. Several others of similar dimensions were
also offered for sale but not purchased. The midbody circumference of the largest
male was 190 mm."

An athetintiated measurement of a female of enormous size, both in length
atndigirthi-obtained in the Emibu District of Kenya towards the Tana River:in 1935
is 5 feet exactly in the flesh, though the skin without pegging whilst being cured
was as much as 6 feet a2 inches. Even larger specimens are said to exist in this
Vide Ditmar's (1931, p. 177):- "Its length is up to four-and-a-half feet and
an example of this size would be nine inches in girth." Schmidt (1923, pp. 141-
142) referring to fifteen specimens he examined which had. been collected by the
American Museum Congo Expedition quotes the largest as measuring 808 mm. The
length of tail is diagnostic of the sexes, and according to the same author :-
"varies in the males from .io to. 14 of the total, in the females from .06 to .o8", from
which it will be understood that the tails are conspicuously longer in the males.
In a series of more than three dozen examined in Northern Rhodesia by the
write, the largest, a female, was 40 (tail 31) inches and it had a girth of nearly 13
inches; another very massive specimen 35 inches long had a girth of io inches.
According to Fitzsitnons (Snakes, 1932) :- "At birth the Puff Adder is six inches
Scales in 27-41 rows: ventrals 125-147: caudals I6-37. These data have
been obtained from more than a hundred and twenty examples collected mainly in
Eastern Africa. In a long series of Uganda and Kenya specimens the scales are
in 29-35 rows: ventrals 132-145: caudals 16-37. The greatest dorsal scale
counts come from Southern and South Africa, i.e. 41, 39 and 37.
Scale counts of sixteen Uganda specimens are as follow:-
Locality, Sex. Scale-rows. Ventrals. Subcaudals.
Kilembe, E. Ruwenzori male 33 132 31
Entebbe juvenile 31 145 18
W. Ankole juvenile 33 136 19
Uganda female 35 142 17
Entebbe half-grown 29 140 19
Entebbe half-grown 29 143 19
Semliki Valley juvenile 31 139 34
Semliki Valley juvenile 33 133 37
Entebbe male 31 133 29
W. Ankole male (half-grown) 29 138 32
Ongino, Teso male 33 137 37
Bulisa, Bagungu juvenile female 31 144 21
Bukalasa male 29 133 32
Jinja female 33 142 19
Kaiso, L. Albert male 31 136 33
Katwe, L. Edward male 33 142 32


Prominent characteristics are enumerated in the descriptive note on the genus,
but. this-speies is so well-knownt and familiar that furthl comprehensive. detailed
description is hardly necessary; It is a creature of terrifying appearance with a
very wide, thiekly-scaled head and a.fierce-looking snout; the nostrils ar directed
upwards. The breadth of the ugly flat head is accentuated by the .thi nejk
followed by a bloated, flat body, ir the adults excessively stout for its length. The
rounded ventrals, as is only to be expected, are extremely wide. The tail, which
ends in a tiny spike, is short and stumpy, being particularly abbreviated in the
females. The small, keeled scales which cover the head and body tend to give the
Puff Adder a rough appearance.
The vertically elliptic pupil varies from a malevolent golden-grey to a sinister
gleaming silvery-grey.
According to Boulenger the coloration may be:- "Yellowish, pale brown, or
orange above marked with regular chevron-shaped dark brown or black bars point-
ing backwards, or black with yellow or orange markings; a large dark blotch cqv-
ering the crown, separated from a smaller interorbital blotch by a transverse.y.eJlow
line; an oblique dark band below and another behind the eye; yellowish white
beneath, uniform or with small dark spots."
Quoting from Ditmars (1931, p. 177):- "Its brightly laid colors consist of
black chevrons separated by yellow crescents." Loveridge (1928) describes it as
a:-"handsome creature of very variable coloration, some being reddish, chocol-
ate-brown or lemon-yellow with various other colored markings superimpose. on
these ground tints, a characteristic feature is a series of V-shaped markings along
the whole length of the back".
Ditmars' description though conveniently brief and lucid gives little idea of the
brilliancy of coloration which imparts almost a handsome appearance to some ex-
amples of this repulsive species.
Field notes concerning an Entebbe male, the most vividly marked specimen, the
writer has handled, are:- "The light coloration is particularly brilliant, ranging
from rich orange-brown to a deep golden, the effect produced being most striking.
The coloration is usually bright in examples which have recently sloughed. In
reddish soils this ground-frequenting species is apt to acquire more or less per-
manently the coloration of its environment. In some specimens there are several
conspicuous light-coloured spots on the head. Newly-born juveniles are replicas
in miniature of their hideous parents.
Habits -The Puff Adder in Uganda is the savanna representative of the genus
Bitis and is equally at home in light woodland or plain, in grassland or amongst
rocks, and in the arid low-lying rifts or regions of marsh. Although normally it is
a ground-haunting species, which its peculiar shape at once suggests, there are re-
cords of its being found in the roofs of huts, as well as at an unusual height in a tall
tree-an exceptional achievement. It is most agile: in the water swimming with
remarkable speed and ease.
According to Fitzsimons (Snakes, 1932):- "PuffAdders are especially fond of
bathing.' They swim sluggishly, with an undulating movement of the body, which
is very buoyant. When tired of swimming, they often lie and rest on, the water.
The head and two-thirds of the body float, and a portion of the tail-end-hangs more

or less perpendicularly: in the water. Often the body is partly supported by water-
plants, and in this position the reptile will lie all day. However, they 'can float
freely without any artificial aid." In dry weather where common they will be found
to concentrate near the available water; and, under the prevalent arid conditions of
the Sudai, according to Corkill(i 935, p. 27):--"The species appears to be restrict-
ed to a riverain distribution."
Puff Adders frequent the vicinity of human habitations in order to feed on the
rodents, principally rats, which are also attracted by settlements, and thereby act
as benefactors to mankind. At certain seasons these snakes are accustomed to
enter gloomy huts and dark stores in their search for prey, and there is a publish-
ed record of one which found its way into a native's bed !
The Puff Adder is very sluggish in disposition, though on occasion when the
circumstances warrant, such as striking at an intruder in self-defence, it can be as-
tonishingly agile. It is not an aggressive species, but being extremely lethargic it
prefers when threatened to advertise its presence by the loud hissing from which
it derives its.popular name rather than seek safety in flight, and so gives timely
Warning of its whereabouts. The vigorous hissing is produced with each inhala-
tion and exhalation of the breath and is purely a warning demonstration, and not
as has sometimes been stated the prelude to an attack.
The noise seems to be a combination of.puffing and hissing and is the only
sound which can be uttered on account of the formation of this or any other snake's
Nevertheless, throughout Africa, wherever this species occurs, it is associated
by the natives with a booming or siren-like noise which is often heard amongst
the manifold sounds of the African night. It is indeed strange that this legend
should be so widespread, and the alleged cry of the Puff Adder as a source of con-
troversy is unlikely ever to lose its popularity. The booming noise uttered by the
crowned crane (Balearica regulorum) during the rainy season is also attributed by
some natives to the Puff Adder.
The following notes are contributed by a keen naturalist:- "As regards snakes
making 'a noise, I can assure you that Puff Adders can do more than hiss, though
I presume the sounds in each case are caused by expulsion of air. But the sound
is far from a hiss......(r) I had the misfortune to tread on a Puff Adder. The first
I knew was a whack over the shin by the reptile's tail, (for I trod on its head) and
a most peculiar noise, a sort of 'puff, long drawn out, resembling the sudden rush
fair from a motor tyre"......(2) "something had been making a peculiar noise in the
grass near hmy house, like a dying buck. It obviously could not be a buck, as the grass
was quite short, not long enough to hide a buck ......I decided to investigate.......
and quartering every foot of the grass could locate nothing and was on the point of
giving up the search when the weird sound was emitted again..... and going to
the spot, there lay coiled an enormously fat Puff Adder"..... (3) "I was told there
was a snake in the grass making a noise like a motor car. Almost immediately I
heard it, and. recognized the'Puff Adder call, though it was not quite the same as
the dying buck call ......and resembled more the noise horses make when forcing
air through their lips. The sound was madetime and again ....... believe it is a
breeding;call ", .... ....

Loveridge (1928) states:- "In captivity they are very quiet, well-behaved cre-
atures never attempting to bite after settling down," and Corkill (1935, p. 27) men-
tions:-"Placid and sluggish in disposition ...... it is not aggressive". On the other
hand Fitzsimons (Snakes, 1932) gives this snake a very bad character:- "However
long you might keep a Puff Adder in captivity, he can never be trusted because he
is suspicious and bad-tempered by nature."
Owing to the head being so much broader than the neck it is one of the easiest
snakes to pick up by the scruff of the neck, so is not difficult to handle. Great care
however must be exercised in catching these heavy-bodied creatures which cannot
be noosed and lifted by the snake-stick generally employed. A heavy-bodied viper
lifted by a noose around the neck will with one convulsive jerk dislocate its neck.
Normally a big viper is coaxed on to a crook at the end of a stick which is gently
inserted between it and the ground. If the operation is carried out skilfully the
snake can be lifted off the ground without it taking alarm and then dropped into the
bag held in readiness for it.
When being handled it is only natural that a Puff Adder does not always sub-
mit tamely to such an indignity, and care must be taken that the snake does not
manage to "spit" venom at its captor. Actually, it is not deliberate "spitting" as
is indulged in by the specialised cobras but is an involuntary squirting of venom
from the fangs as a result of the furious gnashing of the jaws. The poison is only
ejected for a very short distance, so it is only when one is examining a Puff Adder's
mouth that there is a possibility of receiving twin jets of venom in the face.
Puff Adders, like pythons, suffer a great deal from diseases. of the mouth which
then has to be disinfected, and it can well be imagined that operations on the
mouths of such difficult customers are by no means a pleasant business.
Puff Adders are often heavily infested with both ecto-and endoparasites. Snakes
personally examined have sometimes had several dozen ticks, a few of large size
and extremely swollen, attached to them: these ticks usually belong to the genera
Amblyomma and Rhipicephalus. According to Roosevelt (African Game Trails,
1910, p. 196):- "On the bigger puff adder, some four feet long, were a dozen ticks,
some swollen to the size of cherries; apparently they were disregarded by their
sluggish and deadly host".
Most Puff Adders are also the hosts of quantities of nematode worms, some of
extraordinary length, and in parts of Uganda these vipers harbour many of the cur-
ious Linguatulids or Pentastomids which will be described in greater detail in the
descriptive note on Bitis gabonica. Loveridge (1928) mentions nematodes of the
genera Ophidascaris and Thubunea as having been identified from puff adders.
According to Flower (P.Z.S., 1937, p. 35) a specimen kept in captivity in South
Africa was at least 13 years i months and 16 days old when it died.
Enemies-Nowadays with ever extending development of land and settlement,
man both directly and indirectly must constitute the Puff Adder's chief enemy.
Buzzards, eagles, the secretary bird and the ground hornbill are its principal
feathered enemies; and various species of mongoose and the warthog are its worst
foes amongst the four-footed beasts.

Buzzards and eagles seize their victims, carry them to a height and let them
drop on rock or hard ground; the secretary bird can easily stamp to a pulp the head
of the biggest puff adder; while the ground hornbill when tackling a big specimen
does not operate alone but acts in concert, the unfortunate snake being surrounded
by several of these huge birds which shielding their bodies with their wings deliver,
with their powerful beaks, a series of murderous blows. An ostrich in captivity,
once-picked up and ate a half-grown Puff Adder. The warthog's technique has been
referred to on an earlier page. The mongoose's methods have been described by
various observers but primarily this little creature manoeuvres to deliver its attack
from the rear while the snake well aware of its vulnerability from this direction
strives in vain to avoid assault from the dangerous quarter. A combat may last
as long as an hour. The mongoose relies for safety on its thick fur which mis-
directs the strike, its tough skin, alertness and strategy; and, owing to its agility,
patience and the eventual exhaustion of the snake, nearly always wins. Fitzsimons
(Snakes, 1932) gives a most entertaining account of the result of putting a big grey
rat one evening into a cage containing eleven puff adders. Next morning:- "Seven
burly puff adders lay dead, and the rat, with bulging sides, was coiled fast asleep,
in a recess in a tree stump. Inspection of the bodies showed the brains of the snakes
had been pierced by the rat's canine teeth and the tenderest parts of the heads and
bodies eaten."
In captivity puff adders are apt to come into conflict with other species of
poisonous snakes and occasionally fall victims to the bites of cobras, and, accord-
ing to Fitzsimons (1932):- "The Puff Adder fights long and energetically before
finally succumbing to the nerve-paralysing effects of the venom of its antagonist."
Boomslang (Dispholidus typus) venom is also lethal to the Puff Adder. There is
a published record of a Nile monitor (Varanus niloticus) in captivity which develop-
ed the habit of snake-eating: it recovered from one Puff Adder bite but subse-
quently succumbed to the effects of another.
Breeding-The Puff Adder is ovo-viviparous and the young develop fully in
the ovary of the female. Sometimes birth actually takes place inside the snake,
more usually it occurs as soon as the fully developed egg is deposited. The young,
of course, are entirely independent at birth.
The Puff Adder is prolific and produces astonishing numbers of young at a
time; 60-7o is by no means unusual, and the writer has examined about a dozen
specimens which contained between 60 and 72 ova with well-developed embryos.
A female collected in Northern Rhodesia contained about sixty developing ova in
each oviduct.
According to Loveridge (The Snakes of Tanganyika Territory):-"On Nov-
ember x8th, 1915, I killed an adult at West Kenya which held 38 eggs as large as
those of a domestic pigeon, there were 24 in the right oviduct and 14 in the left.
On November 28th, 1917, I chloroformed and examined a female which had 40 eggs
in the right ovary and 3 eggs in the left-a total family of 71 impending. On
February i6th, 1917, another female had 18 young adders in the right and 16 in
the left. These measured 7 inches each in length and were very near birth".
The same author on IIth November, 1933, obtained a female in Karamoja which
contained ova "just developing".


b'et-'The Puff Adder is primarily a rodent eater, and reference has been inade
previously to an example which contained five large rats (Arvicanthis sp.). It also
devours quantities of Rattus sp., Mastomys sp., Rhabdomys sp., and many other
species of rodents. It will feed on various types of small mammals when oppor-
tunity offers, and according to locality may consume freely toads and frogs. It has
also been known to eat orthopterans (cockroaches, grasshoppers and crickets).
Fangs and Venom -The erectile caniculated fangs of the huge vipers and their
deadly function have been described comprehensively on earlier pages, and further
repetition is unnecessary. In big puff adders the impressive fangs are fully three-
quarters of an inch in length. According to Fitzsimons (1932):- "The Puff
Adder's fangs are half to three-quarters of an inch long and strongly recurved
.........This formidable reptile, when about to bite, draws up its body, throwing the
head well back to obtain the necessary leverage, and makes a powerful drive for-
ward or sideways, but never backwards. However, when on the defensive it can
turn with surprising agility"; and, "The Puff Adder sometimes retains its victim,
but as a general rule it is bitten and immediately released"; also, "held his grip,
meanwhile pressing down at short intervals. He was pumping venom into the body
of his foe."
The venom though primarily a hemotoxin possesses certain neurotoxic pro-
perties and is viscid and straw-coloured. The yield from an average Puff Adder is
not as large as the size and weight of the snake would lead one to expect.
In the course of experiments and investigations carried out recently in South
Africa, it was ascertained that by manipulation of the glands and fangs a yield as
much as the equivalent of 0.75 gramme of desiccated venom could be obtained; but
the average weight from a considerable number of puff adders Imilked'for the first
time in captivity was 0.18 gramme. By allowing five days for a number of snakes
to accumulate venom after the original extraction, the average yield was reduced to
0.07 gramme, but in other instances a rest of three weeks enabled one to extract an
average of o.i gramme of (dried) venom.
Desiccated venom is readily soluble but the solution is cloudy with considerable
precipitation which is characteristic of most viperine venoms.
Quoting again from Fitzsimons:- "Puff Adder poison is extremely deadly to man
and beast; but it is slow in comparison to that of the cobra. A man may live for
twelve, and even twenty-four hours, after a bite by a Puff Adder, because a lethal
dose of the venom causes slow but extensive internal hemorrhage; and it takes
longer for the victim to die in this way than by the powerful nerve poison of the
cobra and mamba"..... "The venom of the Puff Adder causes hemolysis, and its
usual effect on the human subject is to induce such extensive internal bleeding that
the victim dies of exhaustion"..... "Puff Adder venom makes the blood-vessels porous,
and at the same time it dissolves the corpuscles. Profuse and fatal bleeding into
the abdominal cavity was imminent" ...... "The venom paralyses and kills all warm-
blooded prey such as rats and birds in a surprisingly short time. After being
bitten a large rat staggers for a few moments and dies painlessly. This is due to
the fact that the neurotoxin (nerve poison) in the venom produces an almost instan-
taneous fatal effect. In larger victims such as man, the neurotoxin principle in
Puff Adder venom is not sufficient in quantity to produce rapid death. Consequently,

the hamorrhagin, ot blood-poisoning, properties of the venom have sufficient time
to produce their characteristic symptoms, viz., haemorrhage into the tissues, dis-
coloration, and swelling more or less profound."
The same authority records that young puff adders:-"When born are provid-
ed with a fully developed apparatus and the glands are already charged with ve-
nom," and he recounts how a recently born Puff Adder when exploring the moat
(in the Snake Park) was swallowed by a carp, which it promptly bit in the stomach,
and the fish died after a period of abnormal activity "Five minutes after birth
these children of the serpent can kill a rabbit, guinea-pig, or large rat with one bite,
the victims dying in from one to two hours. A mouse dies almost immediately."
As previously mentioned there have been many cases of recovery from a Puff
Adder bite on the part of Europeans in South Africa after the injection of 20 cc. of
anti-venene: also, physical human activity after a Puff Adder bite very effectively
circulates the venom throughout the system. In cases of viper poisoning the admi-
nistration of alcohol even in small quantities is absolutely fatal.
A girl bitten by a Puff Adder in South Africa was injected with anti-venene
more than half-an-hour later, and "next morning was well enough to be taken home."
Fitzsimons has published full details in his brochure The Snake Park (1927) of
cases of Puff Adder bites, their effects and treatment when a European assistant
and the native attendant (on two occasions) at the Park were unfortunate enough
to be bitten; he also describes how a European farmer bitten in the calf by a Puff
Adder saved his life by a self-administered injection of serum.
Loveridge (1928) records the case of a native who "was struck by one fang on
the knuckle at the base of the index finger of his left hand" in which in spite of
cautery and treatment with potassium permanganate (rather superficially performed)
within five minutes of being bitten the man's condition thirty-one hours later was
precarious, but an injection of anti-venene saved his life and on the fifth day "he
was so far recovered as to be able to get up and wash, himself and thereafter con-
tinued steadily to improve. The doctor inclined to the opinion that he had received
a non-lethal dose and that he might have recovered without any treatment whatever."
Vide Hale Carpenter (A Naturalist in East Africa, p. 29):-"I do not think the
bite of a Puff Adder is always so serious a matter as is generally imagined. Re-
cently one of the porters on a safari came up quite calmly saying he had been bitten
by a snake, and there were two punctures on the top of his foot from which blood-
stained serum was oozing, but he seemed in no way disconcerted. I applied the
usual remedy, cutting through the skin at the site of the punctures and rubbing in
a crystal of potassium permanganate, but felt quite certain that, as it must have
been at least twenty minutes since the bite was given, a good deal of poison must
have been absorbed. The snake was seen, as it had been killed, and proved to be
a Puff Adder about two-thirds grown. But the porter turned up next day, quite
cheerfully, to carry his load." There is no doubt that little, if any, venom can have
been injected at the time of the bite-probably the poison glands were empty-or
it would have been a very different story.
Ansorge (Under the African Sun, pp. 291-292) mentions the case of one of his
servants who was bitten by a Puff Adder when on the march in Singo and who died
within two and a half hours.

- i6A

br. Burgess Barnett, Curator of Reptiles at the London Zoological Gardens
replying on 13th February, 1936, to an enquiry in East Africa writes:- "The
venom of a Puff Adder is far less toxic than that of a cobra, and though a much
larger quantity is injected, I do not think death would be usual in less than several
hours unless the fang of the reptile accidentally pierced a vein and introduced the
venom directly into the blood stream. The cases that recover would certainly suf-
fer from ulceration and gangrene, and one would expect to see hemorrhage from
many parts of the body, such as the stomach and kidneys."
It is remarkable that bare-footed natives are so seldom bitten. Cattle when
ploughing or grazing often get struck and in the absence of an injection of the re-
quisite serum rapidly succumb. Fitzsimons records that a mongoose died after
three hours, evidently expecting death, having been bitten in the tongue; and also
that Puff Adder poison will kill the big water lizard (Varanus niloticus).
In South Africa the Puff Adder is the classical example of African viper and
the cause of most fatalities and disabilities from viperine snake bites.
On earlier pages it has been mentioned that the pig (and presumably the wart-
hog) ishighly resistant to Puff Adder venom; that the South African grey mongoose
is not so resistant as its tolerance to Cape cobra venom would suggest; while the
meercat's reaction is in no way comparable with its amazingly high resistance to
colubrine venoms.
The African natives have various alleged cures, in which emesis plays a pro-
minent part, for a Puff Adder bite, and quoting from Loveridge (1928) "there is
something in their experience of snake cures I still believe, though there is a con-
siderable mixture of ignorance and charlatanry in their lore," this subject will
however be discussed more fully later.
On an earlier page under the heading Strange Use of Deadly Snakes has been
mentioned the sinister method, with the aid of a captive Puff Adder, employed by
certain African tribes in the past to obtain game meat.
Note-The Puff Adder, under the trade name of "Rhinoceros Viper" (West
Africa) is one of the species listed by the Advisory Committee on Hides and Skins,
Imperial Institute (1933) as being "already in commerce." The skin, however, is too
fragile to be satisfactory for making shoes, though attractive for "fancy goods."

BITIS GABONICA (Dumeril and Bibron).

Gaboon Viper or Puff Adder.


(Plate XVI, Fig. 1: Coloured Plate (R), Fig. 2).

Native names-The Baganda and the Basoga in parts of whose country this
species is common do not appear to distinguish it from Bitis arietans, and in con-
sequence it is called "Essalambwa" both in Luganda and Lusoga. In Lunyoro it is
"Nkondambogo"; and in Lutoro, "Mpoma" (the same as the Puff Adder).


Distribution-Bitis gabonica ranges throughout the Western Forest and is
also found outside the forest proper in Togo, Uganda, Tanganyika Territory, Mo-
zambique, Northern Rhodesia, Angola and Damaraland. It has also been obtained
on the island of Zanzibar.
Occurrence in Uganda-The Gaboon Viper is a common species in the inland
forests of Southern Buganda and in the lake-shore forests of the Victoria Nyanza
reaching its easterly limit somewhere along the Busoga coast though the precise
locality is not known. It is also common to the Budongo and Bugoma forests in
Bunyoro, and is believed to be widely distributed in certain of the western forests.
It possibly occurs on the Sese and other northern islands of the Victoria
Nyanza, though the writer is not aware of any authentic records. In the British
Museum- collection there is a specimen labelled "Victoria Nyanza" which may
however refer to a lake-shore locality. This viper ranges for a considerable distance
out-side its forest haunts.
Specific localities in which it has been collected include:- Entebbe, Kisubi,
Kampala, Bukalasa (30 miles north of Kampala), Kyagwe, Mabira Forest, Katebo,
Budongo Forest, Bugoma Forest, Murchison Falls and Semliki Valley (Toro).
According to Driberg (The Lango, 1923) it is found in Lango.
Description-Without any hesitation the writer claims that the Gaboon Viper
is the largest representative of the genus Bitis or in fact of any of the African
vipers. In support of this claim are quoted the measurements of a gigantic female
which was killed in the Mabira Forest in September, 1933. Length 5 feet 8
inches, in which is included the tail of 5| inches : greatest width of body 6j inches :
maximum girth 141 inches: length of head 4) inches: breadth of head 44 inches:
weight, the stomach being empty, x8 lbs. With enormous fangs of nearly two
inches in length this is indeed a nightmare monster. It was caught alive on the
outskirts of the Mabira Forest and brought into camp after dark, hissing furiously,
enveloped in a stout pig net and carried on a pole by a couple of men.
But this is by no means a solitary record of exceptional size, and the following
data indicate that in Uganda Bitis gabonica can be a really big brute:-
A specimen reported exceeding 6 feet in length from a forest in South
Kyagwe. The dried skin was more than 8 feet long and of great width.

A female 5 feet 2 inches long, weighing i 2 lbs. (stomach empty), examined
from a lake-shore forest near Entebbe.
A female 5 feet o inch long, weighing Io0 lbs. (stomach empty), from
Bukalasa, north of Kampala (authentic record).
A specimen exceeding 5 feet (no precise data) from the Murchison Falls
(authentic record).
A male 4 feet 9j inches.long (tail 64 inches), weighing 9 lbs., examined
from the Mabira Forest.

A male 4 feet 4 inches long (tail 64 inches), weighing io lbs., examined from
the Mabira Forest,
An example 4 feet 21 inches long, from Bukalasa, north of Kampala (au-
thentic record).

A male 4 feet 24 inches long (tail 6- inches), weighing 8 lbs., examined
from the Mabira Forest.
Examples 3 feet to 4 feet in size are common.

Loveridge (1928) refers to a Tanganyika example of 49 inches, i.e. i60
(1083+ 77) mm. An authenticated record of 4 feet, from Kawambwa, was given the
writer when in Northern Rhodesia: several other records verbally given of 4 feet
to 5 feet examples in Northern Rhodesia were not precise though undoubtedly re-
ferring to specimens of noteworthy size.
Boulenger's biggest measurement in a series of eight is only 170 (tail 70) mm.,
not particularly large, being less than 4 feet.

Schmidt (1923, p. 143) referring to material collected by the American Mus-
eum Congo Expedition quotes seven males ranging in length from 414 to .11oo mm.,
and six females from 443 to 1297 mm. The proportionate tail lengths in these
males vary from .09 to .12, and in the females from .05 to .06. He further states:-
"The sexes are sharply distinguished by this (tail) character and also by the number
of subcaudals, which is 18-20 in females, 27-32 in males".
Loveridge (1937) refers to a male of 1460 (300oo+ 6o) mm., collected by the
George Vanderbilt African Expedition in French Equatorial Africa, a measurement
a little in excess of 4 feet 9 inches. Scale-rows 35-43: ventrals 125-140: sub-
caudals 17-33.
These data have been obtained from published records and from examples
personally examined concerning in all about seventy specimens.

In a series of more than a dozen Uganda examples the variation is:- scale-rows
35-43: ventrals 129-139: subcaudals 19-32, which differs little from the maximum
Some scale counts of Uganda specimens are as follow:-

Locality. Sex. Scale-rows. Ventrals. Subcaudals.
Victoria Nyanza female 37 136 19
Entebbe juvenile 39 .131 19 (I)
Mboga, Semliki
Valley 35 130 30 (2)
Ituri Forest half-grown 41 137 .al (1)
Mabira Forest female 39 134 21

Mabira Forest male 39 129 30
Mabira Forest female 37 136 21
Mabira Forest male 41 129 29
Mabira Forest juv. male 39 131 31
Bukalasa female 37 139 22
Bukalasa female 37 137 20
Mabira Forest juv. male 37 130 32
Mabira Forest juv. male 39 132 31
Judging from the subcaudal counts (i) of above should be females and (2) a
Prominent characteristics have been detailed in the descriptive note on the
genus. Apart from the distinctive coloration it must be mentioned that the nostrils
are directed upwards and outwards, and, as described by Boulenger, there are:-
"a pair of more or less developed, compressed, erectile, triangular, sometimes bi-
or tricuspid shields, in contact with each other, between the supranasals, forming
a pair of nasal 'horns'."
The head as in other members of the genus is covered with small scales, moder-
ately keeled; the dorsal scales are strongly keeled-the outer row smooth--and
laterally slightly oblique. The huge flat head, pale above, combined with the pres-
ence of a pair of nasal 'horns' is an excellent diagnostic, differing as it does con-
spicuously from the coloration and markings of either B. arietans or B. nasicornis.
No better description than that of Ditmars (Snakes of the World, pp. 177-178)
will be found:- "The Gaboon Viper, B. gabonca, is the world's most frightful-
looking snake. An example less than four feet long is three inches in diameter with
a head as broad as across the four fingers of a man's hand. The incongruously
stout body suddenly tapers to a stubby tail and anteriorly tapers enough at the neck
to cause the head to look enormously wide and sinister. Upon this awesome form
the symmetrical pattern and really beautiful hues repel, rather than soften the
picture. The design looks more like a pattern in bizarre weaving than markings of
a serpent. There is a series of precisely oblong buff markings on the back, which
are enclosed within long ovals of rich brown. This combination is again enclosed
by a chain of purplish markings. On the sides are triangular blotches of purplish
or deep brown directed upwards. The background hue behind this pattern usually
has distinct tints of pinkish brown. A dark brown blotch begins as a point beneath
each eye and runs backward and downward to the jaw to form a vivid triangle.
The eyes are silvery. Some specimens have a blunt horn on the nose and a few
have a forked horn".
According to Boulenger, it is:- "Brown above, with a vertebral series of
elongate, quadrangular, yellowish or light brown spots connected by hourglass-
shaped dark brown markings; a series of crescentic or angular dark brown markings
on each side; head pale above, with a dark brown median line; a dark brown
oblique band behind the eye, widening towards the mouth; yellowish beneath, with
small brown or blackish spots."

Sir Harry Johnston in The Uganda Protectorate also gives an excellent and
lively description:- "The coloration of this puff adder is perhaps more vivid and
beautiful than in any other snake. It is like a carpet pattern of alternate black,
greenish yellow, mauve, and buff; while by the inflation of the body white edges to
the scales are often shown. Soon after death these colours fade away completely
and the dry skin gives no idea of the blooming tints of the live animal. I write
blooming, because the beauty of these colours is enhanced by a delicate bloom
which appears on the scales, and which softens the tints so that the whole design
might have been painted on velvet."

Loveridge (in lit.) mentions:- "beautifully patterned ...... Its pale brown head
has two conspicuous dark bands beneath each eye. The purples and browns of the
back render this handsomely-marked snake difficult to see when among fallen
leaves in a sunlit forest glade."

The juveniles are replicas in miniature of their hideous parents, the pattern
usually showing up brilliantly.

Habits-The forest association of the Gaboon Viper has already been men-
tioned and will be discussed further in the descriptive note on the next species B.

According to Christy (Big Game and Pygmies, p. 279) :- "Sometimes seen in
the underwood at a considerable distance from the ground."

In spite of the fact that Pasqual in the Nigerian Field (Vol. V, No. i, January,
1936, p. 46) accuses B. gabonica of being "aggressive" personal experience in the
field and the records of others indicate that this big, repulsive-looking brute pos-
sesses the redeeming feature of non-aggression.

According to Sir Harry Johnston (The Uganda Protectorate) :- "It is not, as far
as I can learn, a creature of aggressive malice, and is so far sluggish that the
specimen from which this painting was made was kept by me in captivity for some
time with very little objection on its part. It used occasionally to escape, and would
then allow itself to be picked up and brought back by the negro servants without
any attempt at biting."

The same authority also mentions:- "The vividly painted puff adders of the
Gaboon species are as common as the pythons, and although their bite is absolutely
deadly, they, too, do not seem to be much feared by natives".

Loveridge (1928) mentions that it has been "known to 'spit' its venom occasion-
ally", and doubtless Bitis gabonica, held by the neck, can when gnashing its jaws
discharge involuntarily jets of venom'from its great fangs but only for a short dis-
tance. There is no voluntary ejection of poison through muscular contraction as
in the case of Naja nigricollis; moreover the indications are that the venom flows
best when the fangs are buried to the roots, the gums pressing down hard, in the
object of attack.

The Gaboon Viper hisses just as loudly and furiously as the puff adder, and a
really big example can produce a noise reminiscent of escaping steam. Man is
probably this snake's chief enemy.
There is nothing in the records available to the writer concerning the breeding
of this species, which is of course ovo-viviparous.
Its diet is chief mammalian, and in the stomach of an exceptionally large speci-
men killed near Entebbe was found an adult giant rat (Cricetomys sp.). Other rodents
identified in the remains of meals made by B. gabonica include:- Rattus sp., Ar-
vicanthis sp., Mastomys sp., Thryonomys sp., Leggada sp. (field mouse) and a squirrel
(Athosciurus sp.). A specimen collected by the American Museum Congo Expedition
contained a rail as large as a pigeon. Toads and frogs are also favourites in the
fare of this viper.
According to Mr. J. D. Kennedy in,a letter to the Nigerian Field (No. 3, April
1932, pp. 25-26):- "The Gaboon Viper, another squat snake about six feet long
and as thick as a man's thigh, is found in the moister forests ......... The head is
triangular with the base equalling the sides in length, and very much flattened, about
2 inches thick. It feeds on squirrels, mice, rats and other rodents, and is consider-
ed a great delicacy by the Benis, who eat it in soup".
Nearly all examples of the Gaboon Viper examined from the Mabira Forest,
Kyagwe, and the lake-shore forests of the Victoria Nyanza have harboured numbers
of the curious Pentastomids (Linguatulids) which are worm-like creaturesofparasitic
habits. Specimens of these parasites submitted for expert determination have been
identified with Armillifer armillatus (Wyman) a very common species in this host.
These parasites are a dull yellowish-white in colour, some fully four inches in
length; the largest are 0.3 to 0.4 inch thick, and remarkably and heavily corrugated.
A single snake often contains quantities of Pentastomids, including nymphs and
adults of both sexes. Loveridge (1928) records finding a cestode in one stomach he
examined; and the species Proteocephaus gabonica has been recorded from examples
Obtained in Kenya and Tanganyika.
There appear to be no published records of ticks having been found on speci-
mens of either B. gabonica or B. nasicornis, and all examples examined by the
writer have been free of these unpleasant ectoparasites.

Venom-The venom of Bitis gabonica is of special interest for where mankind
is concerned it is evidently particularly deadly, being a combination venom of great
power, and besides producing the normal viperine hoematoxic reaction also exhibits
powerful neurotoxic properties. Ditmars in Snakes of the World (pp. 178-183)
records in detail the reactions of a European bitten by a Gaboon Viper, together
with the remedies applied and their result. Inter alia Ditmars mentions :- "The
poison of this viper is of remarkably high toxicity. Its prey is killed almost
instantly by injections driven deeply with the'long fangs......Usually with viperine
serpents, the hemolytic (blood-destroying) element predominates......We were to
learn that the poison of the Gaboon Viper not only produces a furious blood and
tissue destruction but has nearly as much power besides in attacking the nerve
centres, particularly the vasomotor system controlling the muscles of breathing, as

does that of the cobra; and that a wound from a single fang, only moderately
imbedded and injecting a far lesser amount of poison than would be imbedded at a
normal bite, can produce results dramatically severe which would probably term-
inate fatally unless every measure to combat the injury was at hand".
In view of the appalling destructiveness of this venom when introduced into the
human system, it is strange to learn that the results of recent laboratory experiments
in South Africa revealed that its potency where mammals and birds are concern-
ed is very much lower than that of the puff adder. Although very similar in
appearance to that of the puff adder it evidently contains its own extremely powerful
agent or agents, as this venom seems to be antigenically dissimilar to that of the
puff adder, for concentrated antivenene prepared with puff adder anavenom exerts
practically no neutralizing action on Gaboon Viper venom.
The poison of Bitis gabonica in the course of these experiments was found to
be very highly toxic to pigeons, but not so potent as that of the puff adder in the
case of mice, rats, guinea-pigs, rabbits and the domestic fowl.
In Northern Rhodesia a cat died as a result of eating the half-dried head of a
Gaboon Viper.
Sir Harry Johnston in The Uganda Protectorate describes graphically how a
Gaboon Viper he kept in captivity together with three enormous pythons killed its
cage mates after they had annoyed it by rolling their huge bulk over its body:-
"When they were particularly exasperating, it would turn and bite them, and a
bite with its formidable fangs would be followed by the spurting forth of two
little streams of blood. Nevertheless, the death of the pythons did not follow as
instantaneously as we knew would have been the case with warm-blooded
creatures; they only became ill, and lingered for two days before they finally died."
The same author without offering any precise evidence, however, also records:-
"Bitis gabonica, is very common in Uganda, and its bite is perhaps more rapidly
and surely fatal than that of any other venomous snake."
Annually there is considerable mortality amongst cattle in Kyagwe from
snake-bite, for most of which doubtless this species is responsible.
Legend-It is possible that the Gaboon Viper provides the foundation for some
of the amazing tales prevalent in Eastern and South-Central Africa on the subject
of the "crowing, crested cobra"! It hisses loudly and furiously, in the native mind
the nasal "horns" can speedily expand to the dimensions of a crest, and, moreover,
it is particularly deadly. The term "cobra" may be due to confusion with the
mamba, but more likely is convenient for producing a pleasing alliteration in the
popular European name! Doubtless this terrifying, deadly monster has a founda-
tion of fact, but like most 'mystery' beasts in Africa is probably composite, as the
writer has endeavoured to explain in A Report on a Faunal Survey of Northern
Rhodesia (1934, pp. 60-6 ).
Note-The large and often handsomely marked skins of Bitis gabonica, although
not yet collected commercially, have attracted the attention of the fancy leather in-
dustry, and in the Report by the Advisory Committee on Hides and Skins, Imperial
Institute (1933) are listed as "worthy of consideration,"


Rhinoceros Viper, Rhinoceros-horned Viper, Nose-horned Viper, Horned Viper,

Horned Puff Adder or River Jack.


(Plate XV, Fig. 2: Coloured Plate (R), Fig. 3).

Native names-It is curious that the Baganda do not seem to differentiate be-
tween this conspicuous species and the more prosaic puff adder, and refer to it by
the evidently generic name of "Essalambwa." In the Budongo Forest region of Bu-
nyoro, it is called "Nkondambogo," which is also the Lunyoro equivalent for the
Gaboon viper. In Lutoro it is known as "Mpoma Lugondo" or "Mpoma Rugondo"
which refers to its forest association, and in this connection it is interesting to
record that the Lutoro equivalent for tree (or bush) viper is "Lugondo" or "Ru-
gondo." In Lukonjo it is "Ekeli" or "Ekeri"; and in Kwamba, the language of
the Baamba, it is "Ki-iri," or "Ki-ili." This Kwamba equivalent is particularly
interesting as it is the same as the Lugishu name for the puff adder, and is also
the Lunyoro term for the variety of Dasypeltis scaber which is distinguished by
viper-like rhombs and markings and which in general coloration and pattern closely
resembles the night adder Causus rhombeatus.
Distribution-According to Schmidt (1923, p. 143) :- "Bitis nasicornis is
practically confined to the continuous Rain Forest but has been recorded from
Portuguese Guinea by Boulenger. It is unknown east or south of the forest, and
must be considered a typical Rain Forest form." In greater detail its range includes
Liberia, the Gold Coast, Togo, Nigeria, Cameroon, Portuguese Guinea, Belgian
Congo, Uganda and the Kakamega Forest of Kenya Colony (immediately south of
Mt. Elgon). Loveridge (1928) mentions that B. nasicornis has also been obtained in
the Kitui District of Kenya Colony: this record is based on an example in the
Coryndon Memorial Museum at Nairobi, and one cannot help wondering whether
the specimen is correctly labelled, the locality being so remote from this species' re-
cognised Kenya habitat. It does not appear to have been recorded from Angola and
is evidently absent from Tanganyika Territory. There is no authentic evidence of
its occurrence in Northern Rhodesia, though it has been suggested that it might be
found in the swampy forest region in the Mwinilunga District situated in the
extreme north-west corner of former North-western Rhodesia, adjacent to
Portuguese West Africa (Angola) and the Belgian Congo: however, it has probably
been confused with the closely allied B. gabonica which is known to occur there,
for the locality is very far south of the defined habitat of B. nasicornis. The
extreme southern limit is reached on the Congo river in the neighbourhood of Boma
(south-west) and at Lulenga (south-east), both localities being in the Belgian
Congo. In the north-east it seems to be fairly plentiful not far from Faradje (Haut
Uele, Belgian Congo).



2. Bilts nasicornis.


r. Bitis arielans.

Occurrence in Uganda-The distribution of the River Jack in Uganda and in
the adjacent Kakamega region of Kenya Colony is extremely interesting, and
wherever it occurs is an indication of a forest locality, liable to partial inundation,
closely associated with the vast western Rain Forest. One would expect the
respective habitats of B. nasicornis and B.gabonica to coincide, yet in Uganda and
Kenya the River Jack is found in areas of considerable magnitude from which the
Gaboon viper is absent, as well as ranging much farther east. For instance the
latter is not found in the Kakamega Forest in Kenya and is apparently absent from
the northern islands of the Victoria Nyanza and the Sango Bay (by the mouth of the
Kagera river) forest region in Uganda immediately west of this lake, in all of
which B. nasicornis is plentiful.

Included amongst the localities from which it has been specifically recorded
are:- Busoga, Mabira Forest, Kyagwe, most of the lake-shore forests of the Vic-
toria Nyanza, Bugoma Island, Kome Island, Damba Island, Tavu Island, Bussi
Island, Kisubi, near Kampala, near Entebbe, Katebo, Bugala Island, Fumve Island,
mouth of the Kagera river, forests of the Sango Bay region, Bugoma Forest and
Budongo Forest. Extralimitally it has been obtained in the Ituri Forest in the
adjacent Belgian Congo. At Bukalasa, thirty miles north of Kampala, where B. ga-
bonica is common it does not occur.

Description-In Uganda Bitis nasicornis does not reach anything like the huge
proportions so frequently found in local specimens of B. gabonica, and a specimen
4 feet in length would be considered remarkably large: the normal size appears to
be between two and two-and-a-half feet.
According to Ditmars (1931, p. 183):- "Its body is just as ponderous as that
of the Gaboon viper, but its head is smaller and narrower. This deficiency is fully
compensated by two high double horns standing erect upon its snout, and some-
times several smaller ones at their base".

Boulenger's largest measurement is 1250 (tail 125) mm., just over 4 feet in
length and probably a male. Schmidt (1923, p. 144) referring to a series of thirty
specimens collected by the American Museum Congo Expedition gives a:- "range in
length from 317 mm. to 1050 mm. The largest female measures 1050 mm., the
largest male, 944 mm. The proportion of tail length to total length is .07 to .08 in
females, mean .08, in the males, .io to .15".

Loveridge ( 936) in a series of eighteen collected at Kaimosi (Kenya Colony)
mentions:-"The largest male measures 760 (705 + 55) mm., the largest female
measures o015 (945 + 70) mm.": the tail in this male appears to be abnormally short.
In a long series of nearly sixty examples collected or examined by Loveridge o115
mm. constitutes the maximum length and specimens in excess of 900 mm. are
In a series of about two dozen specimens either collected personally in the
Mabira Forest and neighbourhood, or examined from the Budongo Forest and the
lake-shore and islands of the Victoria Nyanza, the greatest length, a female, is 37-

tail 3) inches : other measurements in excess of thirty inches are:-33 4 (tail 24)
inches, a female; 33 inches; 31 (tail 44) inches, a male; 31 (tail 31) inches, a male;
and 30o inches. The male of 3 1 inches weighed 3 lbs., and another male, 23A
(tail 5J) inches long, 3- lbs.
Scales in 31-41 rows: ventrals 17-140 : subcaudals 12-32 : these data are from
specimens personally examined and from the published records of various authori-
ties, a total in all of more than one hundred and twenty.
The variation in Uganda material is:-scale-rows 31-37 : ventrals 117-129:
subcaudals 19-29.
A few Uganda scale counts are as follow:-
Locality. Sex. Scale-rows. Ventrals. Subcaudals.
Mabira Forest half-grown 35 126 27
Budongo Forest female 36 130 20
Budongo Forest male 32 126 (damaged) 28
Mabira Forest female 35 129 19
Mabira Forest male 33 123 27
Katebo juv. male 31 120 29
Bugala Island (Sese) juv. male 31 117 27
Prominent characteristics are detailed in the descriptive note on the genus,
but diagnostic is the cluster of sprouting "horns" above each nostril, coupled with
the conspicuous, dark javelin-shaped mark on top of the head which is covered with
small, strongly keeled scales: the nostrils are directed upwards and outwards. The
dorsal scales are also small and. strongly keeled producing a very rough effect in
the skin. If ones hand is rubbed against the scales along the back, the rasp-like feel
of the rough scales is most striking.
Boulenger's description of the coloration is:- "Purplish or reddish brown
above, with pale olive and dark brown or black markings; a vertebral series of pale
dark-edged spots angularly notched in front and behind to receive a rhomboidal
black spot; an arrow-headed dark brown yellow-and-black-edged marking on the
head and nape, the point on the snout; sides of head dark brown, with a triangular
light marking in front of the eye and an oblique light streak from behind the eye
to the mouth; pale olive beneath, speckled and spotted with blackish, or blackish
olive speckled with yellowish."
Ditmars (193i,p. 183), however, provides the most graphic description :- "Or-
dinarily, as the creature crawls in and out of silty water its exceedingly rough
scales show nothing but a dingy pattern. Freshly shed, it is the most amazing snake
of Africa. The effect is of a body of velvet softly blending a dark carmine and olive,
with a pattern laid over that seems utterly out of place for a living creature. Down
the back is a row of large and nicked oblong markings of pale blue. Each of these
has a lemon yellow line down the centre. The blue markings are enclosed in
irregular, black rhombs. On the lower sides is a series of dark crimson triangles,
narrowly bordered with blue. The top of the head is blue with a vivid black
javelin-shaped mark pointing forward".

Schmidt (1923, p. 144) remarks:- "The color patterns of Bitis gabonica and
Bitis nasicornis distinguish them immediately from all other species of the genus,
and it is natural enough that brown, purple and bright yellow should replace the
more grayish hues of the Savannah species. In the case of vipers so formidable
as these two species, their patterns are probably to be considered as a warning
coloration, although in their natural habitat they are doubtless much less conspicu-
ous than would be supposed".
Personal experience with both these species in their natural haunts indicates
that they are most inconspicuous, and it would seem that the coloration is protective,
not from the point of view of the serpent's own safety, but as an asset in procuring
its food for the coloration merging so well with the surroundings enables the snake
to take up its position in a locality where it is likely to obtain a meal, by the meal
coming to it, with little fear of detection.
Loveridge has recorded (1928) how very difficult it is to detect Bitis nasicornis
in its natural surroundings. The same author (1936) when discussing "sexual di-
morphism" in this species in a series from Kaimosi records:- "Subcaudals in males
are 25-30, in females 16-19. Males up to a length of 476 mm. have the belly
beautifully marbled and mottled as in all females; adult males, however, have the
belly uniformly dirty white in sharp distinction to those of the females." Again he
(in lit.) mentions:- "The Rhinoceros Viper is even more marvellously coloured
(than the Gaboon viper), its appearance is that of a strip of richly-woven, velvety
carpet. It may be recognized by two horns on its snout and a purplish marking
shaped rather like a spear point on its head."
The newly-hatched and small juveniles are vividly marked miniatures of their
gaudy parents.
Vide Hale Carpenter (A Naturalistin East Africa, p. 28):- "The colours of this
snake are extremely beautiful when the skin has been recently shed: it has a pat-
tern like a carpet made up of patches of creamy-white, grey, pinkish-brown, and
purplish tints which render it invisible among fallen leaves and low vegetation in
the chequered light of the forest."
Habits-This viper derives its name of River Jack from its partially aquatic
habits and there is no doubt that an abundance of water or swampy conditions in its
forest habitat is an essential.
It is peculiarly placid and inoffensive in disposition, and most reluctant to
bite. According to Copley (in lit.):- "This snake is looked on with contempt by
the natives of Kakamega, where it is extremely common, and is very seldom known
to strike. Two prospectors brought me in a huge beauty which they had kept as
a pet. It was perfectly quiet and could be handled with ease. Unfortunately it
had a bad mouth and died."
It has been stated by Lang that a specimen obtained by the American Museum
Congo Expedition was able to "spit," but as previously explained this must have been
involuntary ejection due to annoyance and not voluntary, controlled "spitting."
Loveridge (1936) found:- "This big viper so abundant at Kaimosi (Kenya
Colony) that eighteen were brought in by natives in three days...............'con-
fident that I could have obtained a hundred during the month we were at Kaimosi,"

The same authority (1936) with reference to Enemies mentions:-"Nose-horned
vipers were recovered from the stomachs of a civet (Civettictis c. schwarzi) and
mongoose (Ichneumia albicauda ibeana)."
This species is able to produce a loud hissing in common with the other mem-
bers of the genus Bitis. Specimens from the Mabira Forest and other forests of
Buganda have been found to harbour plenty of the curious parasitic Pentastomids
which have been previously described. All those submitted for expert determina-
tion have been identified with Armillifer armillatus. Referring to a Kaimosi spe-
cimen, Loveridge (1936) records --"One three-quarter grown snake held twenty-
five large Linguatulids (Armillifer grandis) in its intestines, stomach and in
the viscera just behind the head. The stomach of the same animal held many small
The largest Uganda specimen (375 inches), a female collected in the Mabira
Forest on 31st August, contained 22 (12+ i0) eggs measuring 36 x 29 mm. in which
were well-developed embryos. Loveridge (1936) records that:-"Thirty-eight
large embryos were present in each of two big females brought in (at Kaimosi) on
8th February, 1934."
According to Schmidt (1923, p. 144):-"A female taken at Gamangui (Belgian
Congo) February i6th, 1920, contained thirty-one foetal young, arranged in two
rows of fifteen and sixteen respectively. These young, coiled tightly on one side
of the remains of the yolk, measure 200 to 21o mm. in length and show the color
pattern very distinctly."
The diet of B. nasicornis is unlikely to differ much from that of the other Uganda
species of Bitis, though its close association with water and the fact of its poison
having in addition to hemotoxic, a powerful neurotoxic, action does suggest that
frogs, toads and even fishes may possibly be included more commonly in its fare
than is the case in the other two; but, as all the material personally examined in
Uganda has been devoid of identifiable remains of meals the preceding suggestion
is mainly conjectural. On the other hand the few published records available refer
mainly to mammalian fare-Mastomys sp., Leggada sp., Lophuromys sp., and Croci-
dura sp., (a shrew). Loveridge (1936) found:-"A toad (Bufo r. regulars) in a young
viper", and at the same time recorded: "Rodent fur was present in most stomachs."
Venom-The fangs of B. nasicornis are not as enormous as in either B. gabo-
nica or B. arietans, and rarely exceed half-an-inch in length.
There appears little on record concerning the toxicity of the venom of this
viper though it certainly consists of a powerful combination as is found in B.gabo-
nica. No doubt the Pasteur Institute could provide information on this interesting
According to Ditmars (193 p. 183):-"It has been asserted that the venom of
the Rhinoceros Viper, Bitis nksicornis, is even more highly toxic than that of the
preceding species" (B. gabonica).
Note-The handsomely marked skins of Bitis nasicornis, although not yet
collected commercially, have attracted the attention of the fancy leather industry,
and in the Report by the Advisory Committee on Hides and Skins, Imperial Institute
(1933) are listed as "worthy of consideration,"

Genus ATHERIS Cope.
Atheris is a genus of curious Tree Vipers or Bush Vipers which is restricted
to Tropical Africa, two of its representatives being found in Uganda. They are
slender-bodied, wide-headed serpents which are very difficult to detect amidst the
leafy surroundings which they frequent, particularly as when disturbed they
remain motionless hoping thereby to escape detection instead of the more usual
resort of seeking safety in flight. Their behaviour in order to escape detection
has been described as an "eccentric habit," and according to Ditmars (931, p. 185):-
"They are leaf-green, barely a yard long, and live in low trees or bushes. They
appear to occur along mountain ranges. They may strike in defence, but seem
more commonly to try tactics to escape observation and stiffen the neck at odd
angles in an endeavour to imitate a twig. In these manoeuvres they simulate the
actions of the walking stick insects."
Uganda species though at times having an association with mountain ranges
up to an altitude of 7,500 feet are equally at home, in the case of one in the lake-
shore forests of the Victoria Nyanza at 3,800 feet, and in the other in the elevated
papyrus swamps at 6,000 feet and lower. Although at a distance, and superficially,
resembling other tree haunting species of snakes which are also green in colour,
the wide, triangular head of Atheris thickly covered with strongly-keeled imbricate
scales should at once suffice to reveal its identity. As in the members of the genus
Bitis the tail length is usually diagnostic of the sexes, in most males being markedly
longer than in the females.
Prominent characters include :- the head very distinct from the neck, thickly
covered above, as in the case of Bitis, with small, strongly-keeled imbricate (or over-
lapping) scales, instead of the usual shields; large eye with vertical pupil; lateral
nostrils; slightly compressed body; keeled scales with apical pits, the laterals more
or less oblique and smaller than the dorsals and the outer row; rounded ventrals;
moderately short tail which is prehensile; and single subcaudals. In one species not
found in Uganda, there are several erect, small, horn-like scales above the nostrils.
Little is known about the venom of these snakes, though possibly it is not
highly toxic to man ; it probably contains a fairly powerful neurotoxic agent in order
to deal swiftly with the cold-blooded prey on which these snakes to a great extent


Tree Viper or Leaf Viper.


(Plate XVI, Fig 2: Coloured Plate (S), Fig. 1).

Native names-Vernacular names so far ascertained are "Kisigosogo" which
according to Loveridge (1936) is the Lugishu equivalent; "Nalukonge" and "Kyo-
zima" in Luganda; and "Muryan'kunga" in Lutoro. It is surprising that endeavours
to obtain the Luganda and Lusoga names of a species which is so strikingly con-
spicuous when. handled have met with no success,

Distribution-This is a West African species which is found throughout the
main area of the Rain Forest, ranging from Togo and the Cameroons through the
Belgign Congo to Angola in the south, and Uganda and the Kakamega Forest region
(Kaimosi) of Kenya Colony in the east; and in certain highland localities in Tanga-
nyika is replaced by the closely-allied Atheris barbouri. It is absent from Liberia,
the western limit of the Rain Forest.

Occurrence in Uganda-Organised intensive collecting will probably reveal that
this species is widely distributed throughout the Uganda forests from Mt. Elgon
to the Budongo Forest and other forests of Western Uganda. It has been obtained
at an altitude of about 6,000 feet near Sipi on the north-western slopes of Mt. Elgon;
in the Mabira Forest; in a lake-shore forest at Kasiriye, on the Kyagwe coast of
Buganda; in the Budongo Forest, in Bunyoro, where it is evidently not uncommon;
ontheeastern slopes of the Ruwenzori range; and in the Toro District, precise locality
not known. In its forest haunts amidst its leafy habitat it is a species more likely
to be overlooked than noticed, unless it is the object of special search. Extralimit-
ally it has been collected at Rutshuru, in the western rift, in the adjacent Belgian

Description-Athough generally examples of Atheris squamigera are slender
and, with the exception of the distinctive head, in appearance most unlike vipers,
some of the larger specimens obtained in the Budongo Forest have had the typical
bloated body which one is accustomed to associate with adders.

This species does not grow to a great size, and a length in excess of two feet
is noteworthy: the average length of adults seems to be about 18 inches.

Boulenger's greatest measurement is 550 (tail 1oo) mm. In a series of forty-
one collected by the American Museum Congo Expedition, according to Schmidt
(1923, p. 145):- "The maximum length in twenty males is 657 mm., in twenty
females, 712 mm. The proportion of tail length to total varies from .15-.19 ......
in the males, and from .14-.18 ...... in the females."
From these figures it will be realized that proportionate tail length is not
a feature whereby the sexes can be distinguished. In this connection Loveridge
(1936) has recorded :-"Apart from the fact that adult females attain a larger size
than adult males while the latter have tails proportionately longer on the average,
the sexes cannot be distinguished either by their ventral or their subcaudal scale

The same authority (1936) referring to a series of forty-nine obtained at
Kaimosi (Kenya Colony) quotes:-"The largest male measures 595 (492+103)
mm.," and a female from Sipi, Bugishu, "701 (590+ 111) mm."

Measurements of 712 mm. (28 inches), 701 mm. (approximately 274 inches) and
657 mm. (nearly 26 inches) evidently represent outstandingly large specimens.
Data have been taken from a few examples personally examined, and a large num-
ber of published records, totalling in all about one hundred and twenty.




JohnB e Sons & Curnow tC Lo,1o,

I.Atheris squamigera.
2.Atheris nitschei.
3.Atractaspis irregularis.
4.Atractaspis aterrima.

Presented by M A A. M. Martin.

la. Lateral Section.
2a Lateral Section.
3a. Lateral Section.
4a. Lateral Section.

Ib. Ventral
2b. Ventral
3b. Ventral
4b. Ventral


Scales in 15-25 rows: ventrals 148-173: subcaudals 4o-65: the dorsal scale,
rows are often even. In the British Museum collection there is a male from Ru-
wenzori which has the scales in 15 rows, 159 ventrals and only 35 subcaudals,
altogether a somewhat abnormal example. Some Uganda scale counts are as
follow :-
Locality, Sex. Scale-rows. Ventrals. Subcaudals.
Budongo Forest ? 21 159 52
Budongo Forest ? 19 156 54
Budongo Forest hgr. 17 156 55
Kasiriye, Kyagwe juv. 21 156 48
Mabira Forest female 19 156 56
Mabira Forest male 17 163 61
Kajansi Forest (mile 8 on
Kampala-Entebbe road) ? 23 156 45
The prehensile tail, combined with the distinctive coloration, constitute good
distinguishing features. Most of the prominent characters have been detailed in
the descriptive note on the genus, but it can be conveniently emphasised that the
body scales are so small and rough that they give this snake quite a spiny appea-
rance. The anal is entire.

The gape of the mouth is enormous compared with the head and in some of the
smaller and more slender specimens the sideview is suggestive of a grotesque and
sardonic grin.
According to Boulenger the coloration is:- "Olive above, uniform or with more
or less regular, narrow, yellow cross-bands, or yellow with small green spots; pale
olive beneath, marbled with darker or with yellowish spots or uniform yellow!

Ditmars' (1923, p. 145) description is:- "The coloration is very uniform, and
indeed characteristic. The dorsum is dark green, with yellow-tipped scales arranged
in about thirty-two crossrows on the body. These crossrows are frequently
entirely obscured, but usually persist in a pair of light spots at the edge of the
venter. The venter is dark green like the dorsum, the throat yellow".

Specimens from the Budongo Forest and the one from Kasiriye are of a pleas-
ing shade of sage-green or light verdigris green with pale yellow markings as just
described: others from the Toro District have been dark olive-brown. The indica-
tions are that the tiny juveniles are brightly marked miniatures of their rathercomical
parents, but the writer has not had the opportunity of examining any newly-hatched
An unsexed example measuring 524 mm., collected in a forest near Entebbe
has the terminal portion of the tail ivory-white for a length of Io mm.: the inter-
stitial skin is black, this, however, is only noticeable if the skin is stretched
or where the body has swollen as a result of a blow.

Habits-Most of the attributes and the "eccentricity" of this active, arboreal
species have already been mentioned. The only published reference available in
connection with breeding is that of Loveridge (1936) referring to the Sipi female
which :-"killed December i8th, 1933, was bloated with very small embryos but
so damaged as to be uncountable." In the Mabira Forest a pair were found mating
on 5th October.

There seem to be few available published records concerning the diet of this
species, and in a series of fifty examined by Loveridge (1936) "forty-four stomachs
were empty," in the others were found unidentifiable rodent fur, a tree frog
(Hyperolius rossii), a tree mouse (Dendromus i. inisgnis), a mouse (Mastomys c.
tinctus) and two pygmy mice (Leggada g. grata). Mammalian remains are recorded
by Schmidt in five Congo specimens. One cannot on this scanty evidence assert
that this Tree Viper's diet is principally mammalian although this would appear to
be the case.
Venom-Little is known about the venom of this snake though possibly it is
not highly toxic where mankind is concerned, and probably the storage capacity is
extremely limited.


Nitsche's Tree Viper, Tree Viper, Bush Viper, or Green Viper.


(Plate XVI, Fig. 3: Coloured Plate (S), Fig. 2).

Native names-In western and south-western Kigezi where this species is
abundant it is called "Rugondo" by the Bakiga, and this name is also used by
the Bakonjo and the Batoro in its habitat along the eastern slopes of the Ruwenzori

Distribution-The range of this species seems to be restricted to an elevated
region in south-western Uganda, part of the eastern slopes of the Ruwenzori, the
vicinity of the north end of Lake Kivu, Ruanda-Urundi, and, according to Sternfeld
(1910), Bukoba, which presumably refers to the district and not the town, and to a
locality formerly in German East Africa but now incorporated in the Belgian

Occurrence in Uganda-This Bush Viper is common in the Mobuku valley on
the eastern slopes of Ruwenzori, up to an altitude of about 7000 feet, but the
limits of its distribution on this range are not precisely known; it is also abundant
in the swamps and humid valleys of south-western and western Kigezi especially in
the neighbourhood of Lakes Bunyonyi, Mutanda and Mureyhe, and along the Kisha-
sha valley where it has been found at an altitude of 7500 feet.


1 2


I. Bitis gabonica.

x2 x2

2. Atheris sqtanmgera. 3. Athenrs nitschei.


Descrlption-Like A. squamigera Nitsche's Bush Viper does not grow to A
great size though it is much more heavy-bodied. In a series of about sixty examined
in Kigezi, where hundreds could have been easily obtained, the largest example,
a female, measured 283 (tail 41) inches. All the biggest specimens are females and
out of twenty-eight well-grown adults no less than twenty-two are females. Out of
thirteen Kigezi examples which exceed two feet in length only one is a male, it
measures 26 (tail 41) inches.

In a series of twenty-one in the Field Museum of Natural History, Chicago,
collected by Heller around Lake Bunyonyi, the largest male measures 611
(523+88) mm., the largest female 651 (567+84) mm.
Several specimens of a Bush Viper were collected by the British Museum
Ruwenzori Expedition and described as new by Boulenger in 1906 under the
name of Atheris woosnami: the largest example is 630 (tail 85) mm. Tornier,
however, had described the same species a few years previously from (present-day)
Belgian Ruanda-Urundi as Atheris nitschei which name therefore has priority.

Scales in 25-32 rows, sometimes even : ventrals 141-162: subcaudals 35-52.
These data have been obtained from about one hundred and twenty specimens
mainly personally collected, and to a certain extent from the published records of

Some Kigezi scale counts are as follow:-

Locality. Sex. Scale-rows. Ventrals. Subcaudals.
Muko male 143 44
Muko male 147 46
Muko male 151 46
Muko male 156 44
Muko male 145 45
Muko male 159 55
Muko male 153 47
Muko female 155 46
Muko female 151 45
Muko female 150 45
Muko female 151 42
Muko female 153 47
Muko female 156 41
Muko female 159 38
Muko female 158 46
Muko female 154 42
Muko half-grown 158 39
Muko half-grown 158 42

Muko juvenile 150 4
Muko juvenile T5 50
Muko female 155 42
Muko half-grown 149 47
Kishasha Valley female 152 39
Kishasha Valley male 152 46
Muko female 25 I54 42
Kishasha Valley female 25 152 40
Kishasha Valley female 31 i6o 45
Kishasha Valley female 29 153 43
Muko ? 32 155 39
Muko ? 30 149 42
Muko female 28 152 41
Muko juvenile 28 148 49
Muko juvenile 28 145 44
Kishasha juvenile female 28 156 41 (tip off)

In a series of more than thirty specimens the counts are, in the males vent-
rals 143-159 and subcaudals 44-75; in the females, ventrals 150-159 and subcaudals
38-47. A newly hatched juvenile found with many others emerging in the oviduct
when a gravid female was slit open measured 158 mm. and had 27 scale-rows, 155
ventrals and 39 subcaudals: the tail is ivory-tipped for a length of 8 mm. The
striking coloration, combined with the prehensile tail is diagnostic.

Quoting from Boulenger's description of the type and paratypes of Atheris
woosnami:- "In coloration as well as in form and scaling this Atheris departs less
than its congeners from the typical viper pattern, the characteristic zigzag dorsal
band and the reversed initial V on the head being present in some specimens. The
ground colour varies from olive-green to bright grass-green above, from yellowish
to pale green beneath; the keels of most of the scales are black and the upper
head-scales are edged with black; there is usually a dorsal series of large black
rhombs, which may be confluent into a zigzag band, and a lateral series of smaller
black spots; a A- or A shaped black marking on the top of the head, from above
the nostril to above the last labial shield; the end of the tail is black or blackish."
This comprehensive description requires no amplification as a result of the exa-
mination of several dozens of these vipers personally collected. The newly hatched
juveniles are dull-looking miniatures, greyish-green in colour with dark markings,
of their parents : these youngsters are vicious and demonstrative at birth, while
the adults are much more placid. According to Loveridge (1936) :-"the tips of the
tails in these young vipers are ivory-white".

Any sunny day if one paddles quietly in a dug-out along the reed and papyrus
fringe of either Lake Bunyonyi or Lake Mutanda, particularly the former, one can
see plenty of these snakes in characteristic attitudes entwined high up amongst the
stems or coiled on the papyrus heads. At first they are extraordinarily difficult
to detect, but once the eyes have become accustomed to picking them up amidst the
surrounding vegetation, it is amazing to find out how plentiful they are,

According to notes made by the Ruwenzori Expedition several of these snakes
were obtained in the Mobuku Valley between 6,000 and 6,500 feet, and they may
sometimes be seen coiled up round the stem of elephant grass ten feet above the
Unfortunately the numerous specimens collected were not carefully examined for
parasites: but no ticks were found. No precise records have been kept of the num-
bers of eggs in the large quantity of gravid females (October-November) handled,
but they are not small and in one example eleven were counted. The ova are rather
spherical in shape, the largest, with embryos on the point of hatching, nearly as big
as a small marble: in a few instances many completely developed juveniles which
had emerged were found in the oviducts.
Most of the stomachs of the Kigezi specimens were devoid of the remains of
meals, though two juveniles each contained a tiny tree-frog (Hyperolius sp.), a juve-
nile and a female each contained a mouse, and a big female 27 inches in length and
a big male 261 inches long each contained a large rat. This seems to indicate a
preference for a mammalian fare.
Venom-Little seems to be known about the venom of this species, though it
is believed to be not highly toxic to mankind, an assumption supported by the lack
of snakebite mishaps in a locality such as Lake Bunyonyi where this species liter-
ally swarms.
This is a Tropical and South African genus whose representatives quoting
from Loveridge (1928):- "are so totally different from all the rest in their outline
that there is a very real danger in their being mistaken for harmless snakes. An
examination of the mouth, however, will reveal the huge fangs which, like those
of all other vipers, rotate with the maxillary bones so as to fold back when the
mouth is closed.......These snakes are known as the burrowing vipers and all the
three species with which I am acquainted are plumbeus in colour......In conformity
with their subterranean habits their bodies are slender and the heads no broader
than their bodies. Like the blind snakes they come to the surface after rain and in
travelling over the ground hold their sharply-pointed (in some species rounded)
snout at a peculiar angle as if ready to dig down into the soil at a moment's notice.
They are surprisingly quick."
Further, Ditmars(i93i, p. 185)mentions that this :- "genus of African vipers
is of particular interest as it is hard to surmise what has influenced the members in
acquiring enormously developed poison fangs, as they are burrowers. They have
small eyes and poor sight, appear to feed on small prey, and short fangs would
seem far better adapted to their purpose. These snakes form the genus Atractaspis.
They are a foot and a half to two feet long, slender, with heads no wider than their
bodies and covered with symmetrical shields. The body scales are smooth. The
eye is minute, with round pupil......All of them lay eggs."
From the foregoing it will be understood that this genus is quite one of the
most interesting in Africa, but little is known about its members owing to their
comparative rarity and the difficulties attendant on their acquisition. In addition

to the enormously developed poison-fangs and other characteristics previously
mentioned the head, as in Causus and unlike Bitis and Atheris, is covered with
large symmetrical shields; the body is cylindrical; scales smooth, without pits, in
17 to 37 rows; ventrals rounded; tail short; anal divided or entire; and subcaudals
single or in 2 rows.
According to Boulenger:- "This genus is remarkable as representing the most
extreme specialization in the Viperine direction, the poison-fangs being as large
in proportion as in any other form and the solid teeth on the palate and mandible,
which are much reduced in number in many of the Crotalines, having almost disap-
peared". Loveridge (in lit.) states :- "They are quick to use their enormous fangs
if molested, and their poison is highly toxic."


Burrowing Viper, Mole Viper or Black Burrowing Viper.


(Plate XVII, Fig. I : Coloured Plate (S), Fig. 3).

Native names-This species, or in fact any one of this genus, seems to be gener-
ally unfamiliar to the local populace, also it is inconspicuous, sufficient reason for
the lack of distinguishing names in the vernacular. In Lugishu it is called "Itimu. "
Distribution-Quoting from Schmidt (1923, p. 137):- "The distribution of the
species Atractaspis is interesting, although the scarcity of individuals and records
introduces an element of uncertainty into" deductions of range, counterbalanced,
perhaps, by the recognized value of burrowing forms for the discussion of distributi-
onal problems. Atractaspis irregularis appears to be a forest border species, ranging
from Liberia and Togo to Uganda and thence to San Salvador du Congo and Chin-
choxo on the southern border. The two localities added by the present collection
(American Museum Congo Expedition) link the Nigerian records with those from the
lake region."
In greater detail it appears that this species ranges from its western limit to
the neighbourhood of the Congo mouth in the south but without reaching Angola;
throughout the Belgian Congo (with the exception of the Katanga region) though
neither recorded from Northern Rhodesia nor the Anglo-Egyptian Sudan ; Uganda
and in parts of Kenya Colony, Tanganyika Territory and Zanzibar.
Occurrence in Uganda-This Burrowing Viper seems to be widespread in Uganda
and is equally likely to be found in the semi-arid Nile valley at an altitude of less
than 2000 feet as in the humid, lake-shore region of Victoria Nyanza at 4000 fee
and over.
According to Driberg (The Lango), evidently referring to a species of Atrac-
taspis, amongst the most poisonous snakes in Lango is "a thin black viper". Spe-
cific localities in which specimens have been collected include :- Entebbe, Bussu
(near Jinja), Mabira Forest, Serere (Teso), Katwe (Lake Edward), Budongo Forest



I. Atractaspis irregularis.
2. Atractaspis aterrima.
S. Aparallaclus

3. Chlorophis heterodermus.
4. Naja goldii.

and Wadelai (Albert Nile). Extralimitally, in the adjacent Belgian Congo it has
been obtained at the northern end of Lake Kivu, as well as in the Haut Uele Dis-
trict in the extreme north-east, and in the vicinity of Mahagi Port at thenorth-western
end of Lake Albert.
Description-Atractaspis is a difficult genus which at present contains a number
of named species some of which appear to be so closely related that they are not
easily separable. In the absence of an adequate quantity of the necessary material
for critical examination it is impossible to form any conclusions or even offer sug-
gestions, though it seems fairly definite that the form of the subcaudals is not a
constant character, and in certain species the subcaudals are often both paired and
single in the same individual, while in one published record of A. microlepidota there
is a combination of "first 23 subcaudals single then 8 paired followed by the last 3
The geographical range and variation in scale counts of certain East and
Central African species of Atractaspis are as follow :-

Forest border,
from Liberia and
Togo to the Congo
mouth and Uganda.

No. of


Atractaspis corpulenta Forest, from
Liberia, Cameroon (2+)
and Gaboon to N.E.
Atractaspis bibronii East Africa to S.E.,
=-rostrata=-conradsi E. and N.E. Congo. (16)

Atractaspis aterrima West Africa, from
Portuguese Guinea (9)
to Cameroon, Congo
and Uganda.

Atractaspis heterochilus Eastern portion
of Rain Forest
(Albertville, N'Goma
and Medje)and
Lower Congo. (2)
Atractaspis reticulata Gaboon-Cameroon (?)
Atractaspis schoutedeni Unique: N'Goma, (i)
north end of Lake Kivu.
Atractaspis microlepi- East and Central
dota= phillipsi Africa, Sudan and (i4+)

Scale-rows. Ventrals.

23-27 220-257

23-27 178-193

21-25 212-260

19-23 251-300

23 .341-353
21-23 308-330

23 245

25-37 g21-258



Mainly single;
a few paired.

Mainly single;
a few paired.





Atractaspis irregtu-

diagnostic characters of the various species in the above table are marked
in heavy type. A. irregularis has a very short, rounded snout which is the princi-
pal feature distinguishing it from A. bibronii, which has a prominent, sharp snout
and the anal entire, in the former the anal is divided.
From the data available it seems that A. schoutedeni is doubtfully distinct from
A. irregularis; while A. heterochilus and A. reticulata are so closely related to
each other that they appear to represent respectively the eastern and western ex-
tremes of one Rain Forest species.
The prominent features of A. irregularis are detailed in the descriptive note
on the genus and in the foregoing table and remarks.
The colour varies from uniform dark brown or black to plumbeous above
with purplish brown or reddish -purple below : the tiny eye is black. There is a
small though noticeable spine at the end of the tail. The largest of four specimens
obtained by the American Museum Congo Expedition measures 578 mm. in which
the tail of 34 mm. represents .06 of the length. Boulenger's maximum (also out of
four examples) is 560 (tail 35) mm.
Loveridge (1935), referring to an Entebbe specimen, mentions:- "is very young
with umbilical scutes still unhealed, it measures 247 (230+ 17) mm."; and the same
author (1937) records 293 (273+20) mm. as the total length of juvenile collected
in the Belgian Congo by the George Vanderbilt African Expedition. A female
obtained by the writer in the Budongo Forest measures 495 mm.; and three Mabi-
ra Forest examples measure respectively, 6o0 (565+35) mm., a female; 521 (482+
39) mm., a male; and 216 (203+ 13) mm., a juvenile.
The scale counts in fifteen Uganda specimens are as follow:-
Locality. Sex. Scale-rows. Ventrals. Subcaudals.
Katwe ? 23 243 24
Budongo Forest ? 25 236 22
Budongo Forest female 25 236 23
Serere (Teso) male 23 236 25
Serere (Teso) half-grown 23 237 25
Serere (Teso) half-grown 23 224 26
Serere (Teso) half-grown 23 240 26
Port Alice (Entebbe) female 25 253 22
Entebbe female 25 228 24
Entebbe female 25 234 24
Bussu (near Jinja) female 23 234 23
Bussu (near Jinja) female 23 237 22
Mabira Forest female 23 243 24
Mabira Forest male 23 222 25
Mabira Forest juvenile 23 242 23
Moyo, West Madi ? 27 236 .i

be Witte ([933) unfortunately gives no data in respect of twenty examples
collected by Schouteden in the Belgian Congo.
Habits -Very little is known about the mode of life of this interesting species.
Previous reference has been made to the curious attitude adopted when travelling
over the ground. These snakes are rarely met with on account of their burrowing
Three specimens, two adults and a juvenile caught in the Mabira Forest, were
handled freely and never once attempted to bite. They looked harmless enough
and it was only when they were touched near the head or alarmed that they re-
vealed their identity by at once depressing the head and snout into the typical
"prepare to burrow" attitude.
According to Loveridge (1928) in a general reference to the members of this
genus one was "unearthed by natives digging a pit," another was found "beneath
a mass of earth and stones representing a demolished building," and another dug
out of an old termite colony, about a foot below the surface. The same authority
(1933) found 127 mm. of a Leplotyphlops sp. (the head completely digested away) in
the stomach of the previously mentioned, Entebbe juvenile (247 mm).
A pair were found mating in the Mabira Forest on 17th September and the
female contained six small undeveloped ova.
Fangs and Venom-The exaggerated development of the fangs in all the
species of Atractaspis is remarkable, and it is difficult to suggest a sensible reason
to account for what one can only describe as over-development.
Loveridge (1928) mentions:- "It has been suggested by one writer that the
fangs were so long that the snake could not gape widely enough to bring them
into play, but in one (Atractaspis rostrata) I captured at Morogoro the fangs were
brought down on either side of the lower jaw (like the teeth of a musk deer) in the
act of striking"................."A small boy, about eight years of age, was bitten on
the finger by a young burrowing viper which was only 323 mm. long. When I
arrived at the place two days afterwards, he had a very swollen hand and wrist
but it subsided on the third day," it was poulticed "to ease his mind" !...............
It squirmed convulsively like a blind snake, or worm, when disturbed. When pick-
ed up with a pair of forceps it struck out viciously discharging a quantity ofstraw-
coloured venom upon the instrument."
Wherever any species of Atractaspis occurs, it will be found to be a snake-
when recognised--which inspires great fear among the natives who regard it as par-
ticularly deadly, and yet, in Uganda, vernacular names are almost impossible to
According to E. G. Boulenger writing in Chapter XC, Some Mysteries of Ani-
mal Life (pages 994 and oo1000) in Wonders of Animal Life edited by J. A. Hammer-
ton: "The reptile world offers one very striking instance of over-developed arma-
ture. This is seen in the Atractaspis viper, a small snake with such dispropor-
tionate fangs that it is unable to use the two in unison. The fangs exceed the head
in length and when out for blood the snake extends only one at a time at an obtuse

angle to the roof of the mouth." This is a truly remarkable exposition, and there
appear to be no published records in confirmation of his claim; moreover inde-
pendence in action in the case of the fangs though perhaps not impossible is, as far
as the writer is aware, something unique.


Black Burrowing Viper, Burrowing Viper or Mole Viper.


(Plate XVII, Fig. 2: Coloured Plate (S), Fig. 4).

Native names-Equally, as in the case of A. irregularis there seem to be no
vernacular names applicable exclusively to this species, which is strange, for
although relatively small and inconspicuous, it is on account of the unexpected
agility in attack that all members of the genus Atractaspis where recognized are
evidently objects of considerable dread to the natives.
Distribution-According to Schmidt (1923, p. 139) this species "illustrates a
consistent Sudanese (subprovince) range" (see Botanical Map). It occurs from
Portuguese Guinea, in the extreme west, easterly to the Gold Coast, Togo, Nigeria,
Cameroon, N.E. Belgian Congo and Uganda. One would expect it to be found in
the Anglo-Egyptian Sudan, from which territory, however, so far there are no
Occurrence in Uganda-The sole Uganda record appears to be the specimen
collected by Emin Pasha at Wadelai, where the same scientist also obtained
an example of A. irregularis.
Description-There is not a great deal left to be said about A. aterrima in view
of the comprehensive general remarks incorporated respectively in the descriptive
notes on A. irregularis and on the genus. A. aterrima, however, unlike A. irregu-
laris which is grouped with those species in which the anal is divided, has the anal
entire, which, combined with the low dorsal scale count, constitute good diagnostics.
Boulenger gives a maximum length of 650 (tail 30) mm. Schmidt (1923, p.
139)refers to a specimen from Niangara in the N. E. Belgian Congo measuring 569
(tail 32) mm. Loveridge (933, p. 281) records a male measuring 559(525+34) mm.
from the Uluguru Mountains in Tanganyika Territory, an extension of range not
only further eastward than previously recorded, but also a long way south of the
Equator to the north of which its recognized range had previously been restricted.
Scales in 19-23 rows: ventrals 251-300: subcaudals 18-24. In the Wadelai spe-
cimen there are 19 scale-rows, 300 ventrals and 19 subcaudals.
As in A. irregularis this species has a rounded snout helping to distinguish it
from A. bibronii which is sharp-snouted. The variation in dorsal scale-rows was
believed to be 19-21 until Loveridge (1933, p.281) identified his Uluguru specimen
which had 23 with A. aterrima.

habits-There is little to add to the general notes on the genus or to those
included in the details of A. irregularis; but there is worth reproducing a note by
Herbert Lang quoted by Schmidt (1923, p.139) i.e. "Was tied to a stick and appear-
ed to be lifeless. Intending to take it behind the head, I was surprised at the ex-
traordinary rapidity of its movements. It succeeded in stabbing me with one of its
fangs near the finger nail; but immediate application of permanganate of potash
probably prevented serious developments. The natives regard them as very poison-
ous." Its diet is unlikely to differ from that of others of this genus.
Fangs and Venom-There is nothing to add to the note recorded in connection
with A. irregularis.
Other species-It is possible that both A. bibronii and A. microlepidota may
eventually be found within the limits of present-day Uganda. The former (under
the name of A. rostrata) was recorded from "Uganda" of the early days, now ex-
tralimital, and has also been obtained in Tanganyika Territory as close as Bukoba:
the latter is common in the Anglo-Egyptian Sudan, and has been collected at
Lodwar, in Turkana, in the region west of Lake Rudolf.
In both these species the anal is entire and the subcaudals all or part entire;
but A. bibronii is prominently sharp-snouted and has been referred to as the Snouted
Burrowing Viper, while in A. microlepidota the pointed snout, though prominent,
is very short.
In colour A. bibronii is uniform dark brown or blackish; or dark purplish brown
above, yellowish or pale brown beneath: A. microlepidota is uniform dark brown.
Barbour's (Proc. Biol. Soc. Washington, Vol. XXVI, p. 148, June 3oth, 1913)
description of A. philipsi (-microlepidota) is :- "Scales in 31 rows. Ventrals 232;
anal entire; subcaudals 24, all single. Body solid black above, head very dark iron
gray; belly deep plum colour, almost black; vent white. Length, 290 mm.; tail 22
Loveridge (1936) records finding in Kenya Colony two males of A. bibronii:-
"beneath a pile of rotting palm fronds and the palm thatch of a collapsed hut," one
of which contained "A skink (Riopa sundevalli) :" and in the same publication re-
ferring to three female A. microlepidota collected at Voi (Kenya) he states:- "Both
the adult females were taken together under the rotting grass roof of a collapsed
native hut" ...... "Though taken at the end of a long dry season and the stomachs
of all these snakes were empty, they possessed considerable deposits of fat."
A. bibroniimeasures up to 600 (tail 37)mm., and 600 (tail 25) mm.: A. microlepi-
dota up to 770 (705+65) mm., a female.
Corkill (1935, pp. 30-31) records some interesting information concerning A.
microlepidota (also called the Black Burrowing Viper) which he finds:- "is the second
commonest poisonous snake in the Sudan and the second most frequent cause of
accidents and deaths from snake bite. The symptoms of poisoning from its bite re-
semble those in Echis cases. The writer has notes of three cases of bite in which
the snake was secured and identified as A. microlepidota. One of them, a healthy
adult, was bitten in a sagia at Kassala in 1927, he died in six hours. The other two

cases, also adult men, were bitten by immature snakes; one was very seriously affect-
ed for over a fortnight. This snake wherever it occurs with Echis (the Carpet
or Saw-scaled Viper) is very much the more feared of the two"......"It has no neck, a
short stumpy tail, a spike on the end of the tail, minute eyes and, of course, the
usual erectile fangs." It is probable that the foregoing remarks in respect of the
high toxicity of the venom are equally applicable to the other representatives of the
genus. Corkill further states:- "It appears to occur mostly in cotton soil areas
and has been noted in diverse habitats such as houses, gardens, walls, rocky hill-
sides, cultivation, in trees, in building debris, and in sagias. It is said to eat
According to Corkill this snake has a variety of names in the vernacular two
of which are particularly entertaining-one means the "father of prodding or
jabbing", and the other, presumably in allusion to the supposed duration of life
in a person who has been bitten by it, is the "father of ten minutes." The latter
name is especially interesting in view of its connection videe Corkill) with :- "a
small black snake in Eritrea which is very much feared and which is said to have
an Abyssinian name meaning 'the snake of seven steps', because, after being bitten
a person only takes seven steps before falling to the ground and dying. This
rather suggests A. microlepidota or a kindred species."





Green Tree Snake, Green Forest Snake or Mottled Green Snake.


(Plate XVII, Fig. 3: Coloured Plate (T), Fig. 2).

Native names-In Western Kigezi, the only part of Uganda in which I have
come across this species, the Bakiga refer to it by the name "Mukangaeeni" by
which Chlorophis irregularis is commonly known. The Baganda call it "Nawanda-
gerra" or "Nawandagala" which is the Luganda term for all species of Chlorophis
and Philothamnus.
Distribution-Chlorophis heterodermus is a Rain Forest species which ranges
from Sierra Leone, the Gold Coast and Nigeria in the west to Cameroon and the
Belgian Congo, the eastern limit being reached in South-western Uganda and in
the neighbourhood of Faradje in the North-eastern Belgian Congo. It is evidently



O.Frc c1a -

1. Philothamnus sem/var/egata dorsal/s.
2. Chlorophis heterodermus.
3. Rhamnophl's iturfensis.
4. Psanmmophis brev/rostris.

Presented by the Uganda Government

la. Lateral Section.
2a. Lateral Section.
3a. Lateral Section.
4a. Lateral Section.



' 1


JohnBaLeSIns & Cur ow Lt London.

lb. Ventral Section.
2b. Ventral Section.
3b. Ventral Section.
4b. Ventral Section.

plentiful in the Western Congo close to, and just south, of Leopoldville, and else-
where in the Congo has been obtained at Albertville-the most southerly record,
amongst the Birunga volcanoes near Lake Kivu, in the neighbourhood of Rutshuru,
and in the vicinity of Stanleyville.
Occurrence in Uganda-The only Uganda records are a couple of half-grown
examples collected respectively in bamboo forest (personally), at an altitude of 7500
feet, fringing the highland bog near Behungi, and on the outskirts of the "Impene-
trable Forest" in the Kayonsa region at a similar altitude-both these localities
being situated in Western Kigezi. It is possible, however, that this species will
turn up in other Western Uganda forests which retain Rain Forest associations.
Description -There appear to be few published data available concerning this
species. De Witte (I933) records the several localities in the Belgian Congo in
which Schouteden collected a series of twenty-five, but unfortunately omits their
data. Boulenger out of a series of seven (one male, four females and two juveniles)
listed in the Catalogue of Snakes quotes a maximum length of 740 (tail 200) mm.
Loveridge (1936) when describing material in the Field Museum of Natural
History at Chicago refers to a Cameroon specimen which measures 581 (428+ 153)
mm., and a Lake Kivu specimen of 725 (55+210o) mm. One of the Uganda
juveniles is 304 (23+ 73) mm., the other (damaged) is approximately 270 mm.
Scales in 15 rows: ventrals 147-181: subcaudals 78-99. One of the half-grown
Uganda examples has 181 ventrals and 85 subcaudals; the other 164 (+) ventrals
and 86(+) subcaudals; both the body and the tail being damaged. Females
usually have a greater number of ventrals (152-i65) than the males (141-156); in the
females there are 79-94 subcaudals and in the males 83-95.
Prominent characteristics are detailed in the descriptive note on the genus; in
addition the anal is entire, the ventrals keeled and there are two anterior temporals.
The undivided anal constitutes a good distinguishing feature, for although it is also
whole in Chlorophis carinatus the 13 scale-rows of carinatus should readily identify it.
Boulenger describes the coloration as:-"Green or olive above, the skin
between the scales black; young with black cross bars; lower parts greenish yellow."
He also refers to two females from Sierra Leone which are "black above, blackish ol-
ive beneath." The Uganda juveniles are somewhat nondescript in colour, above dark
generally, either dark green, dark olive or blackish, sprinkled plentifully though not
very conspicuously with bluish, greenish, yellowish and grey, in fact they resemble
very dark, small examples ofPhilothamnus s. semivariegata. Below dull grey-green.
Habits-Besides the fact that C. heterodermus is a forest species which in Ugan-
da is associated with the maximum altitudes at which snake life is found there seems
to be practically nothing on record concerning its mode of life. Its diet is unlikely
to differ from that of other members of the genus. One of the Uganda juveniles
had eaten a small frog.
Note- Chlorophis heterodermus is not included in the Systematic List on page
39* as the Uganda specimens were not obtained until after its compilation.

* See Vol. III, No. 2, p. 136,


Black-collared Snake.

(Mildly Venomous).

(Plate XVII, Fig. 5: Coloured Plate (W), Fig. 6).

At the time of going to press a species of Aparallactus collected by Mr. W. J.
Eggeling on Rom Mountain in north-eastern Acholi in December, 1935, has been
determined through the courtesy of Mr. H. W. Parker of the British Museum
(Natural History) as Aparallactus nigrocollaris.
According to Parker (in lit., 6- 1-37) :- "I have just heard from Angel of
the Paris Museum about that Aparallactus. He has compared it with the type of
A. nigrocollaris and is of the opinion that it belongs to that species. This is con-
sequently a new record for Uganda."
Native names-None known.
Distribution-Owing to a lack of the relevant literature for reference all that
at present can be recorded here is the fact that the type and paratype were
described by Chabanaud from the French Congo (the precise locality is not men-
tioned) and that single specimens are further recorded respectively from the Semliki
Valley in Western Uganda north of the Ruwenzori Range, and from Rom Mountain
as previously mentioned.
Occurrence in Uganda-One example, the specimen under reference, from
Rom Mountain in north-eastern Uganda, and another collected by the Lord
Howard de Walden expedition in the Semliki Valley in 1930.
Description-Measurements and a precise description of the coloration unfor-
tunately were not taken prior to the despatch of this spirit specimen to the
British Museum (Natural History) for critical examination.
Recorded length 267-299 mm., in which the tail varies from 45-67 mm.
Scales, smooth, in 15 rows: ventrals 148-156: subcaudals 33-51. This is a small,
slender snake of a general light brown colour, paler below, with a conspicuous,
black, nuchal collar. A translation of Chabanaud's description of the coloration is:-
"Upper surfaces brown, more or less deep or slightly olive, lighter on the flanks,
with the edges of the scales appearing a little yellowish by transparency. Upper
surface of the head to the occiput, the ocular region to the edge of the lip and
a transverse bar on the upper surface and sides of the neck brownish black. Upper
lip, except the 3rd and 4th labials, the whole lower lip and under sides of the head,
neck and the rest of the body uniform yellowish white."
Habits-There is nothing to record about the habits of this species except
that it appears to be associated with semi-arid savanna at the forest edge and
seems to be equally at home at altitudes of 3000 feet and 5ooo feet,




The generic name in Lusoga for all species of Typhlops is "Mugoya" the same
as in Luganda: in Lunyoro it is "Kinaboo".
According to Corkill (1935, p. ):- "Although these snakes possess a power-
fully poisonous salivary secretion, it is infinitesimal in quantity and, therefore, the
family is not considered dangerous."


Midbody scale-rows 24 the diameter at midbody being included in the total
length 30 times.


Loveridge, in Publication 360, Vol. XXII, No. i, Field Museum of Natural His-
tory, 1936, p. 20, with reference to Typhlops p.punctatus mentions:- "the color form
congestus, which cannot be considered either as a race or as a distinct species." It
has been previously suggested that congestus is based on the larger examples of
punctatus, and it now seems preferable to follow Boulenger and Loveridge in recog-
nising only punctatus and relegating to its synonymy congestus which accordingly
must be removed from the Uganda List.
Measurements and variation recorded separately for T. punctatus and T.
congestus must therefore be united.
Additional recorded localities include Lake Nabugabo, Bukalasa (30 miles
north of Kampala) and Katunguru (Ankole) on the Kazinga Channel. Extralimi-
tally, not far from the Uganda border, this species has been obtained at Bukoba in
Loveridge (1936) when detailing certain characters states :- "A more rounded
snout is characteristic (generic) of the young, the obtusely horizontal edge being a
secondary character developed to facilitate burrowing." In T.punctatus the mouth
is not hooked.
Examples measuring 655 (645+ o) mm. and 600(585 + 15)mm., a female, were
collected by the George Vanderbilt African Expedition of 1934 respectively in the
Belgian Congo and French Equatorial Africa; the latter on i Ith November contained
numerous developing ova in size 20 x r imm. A specimen from Kampala measures
555 mm., and another from the Budongo Forest, 582 mm.
Midbody scale-rows 20-32: diameter at midbody 7-30 mm.: diameter included
in the length from 20-41 times.

According to Loveridge (Bull. Mus. Comp. Zool., Vol. LXXIX, No. 5, 1936) the
specimen collected by Dr. Hugo Granvik on Mt. Elgon and referred to Typhlops
boulengeri by L6nnberg (1922) is actually an example of T. punctatus.


In the British Museum (Natural History) there are six "Uganda" examples of
T. schlegelii (collected by Simons) which were purchased from Rosenberg. These
may be extralimital as far as present day Uganda is concerned.
Midbody scale-rows 30-38 and diameter at midbody 5-23 mm., the diameter
being contained in the length 22-35 times.
Kenya (coastal) examples are evidently extremely variable in coloration, and
according to Loveridge (1936) a juvenile is "Above, blue gray, each scale edged
laterally with black which results in a lineolate appearance. Below, pinkish white":
an adult "Above, black, the centre of each scale with a small buff spot. Below,
buff': two adults about to slough "Above, pale silvery gray or white. Below pink-
ish white"; and another adult in life "White or slightly bluish white."


Midbody scale-rows 24 and diameter at midbody 3-8 mm., the diameter being
contained in the length 54-81 times.


It is probable that the record previously quoted of the occurrence of Typhlops
blanfordii at Mushongero (Kigezi) in South-western Uganda is based on a wrongly
identified specimen and it is possible that this Mushongero example is closely
allied to or the same as Typhlops lestradei described by de Witte from adjacent
Ruanda (Belgian Congo) in 1933.
Also, another specimen of Typhlops collected in 1935 in the Budongo Forest
in Bunyoro has been determined by Parker at the British Museum (Natural
History) as Typhlops near blanfordii.
It is necessary therefore for further critical examination of these two examples
of Typhlops from Uganda before they can be satisfactorily identified, though it is
practically certain that neither can be identified with typical blanfordii.
In typical blanfordii there are 30 midbody scale-rows and the diameter is
included in the total length 40 times.


The generic name in Lusoga for all species of Leplotyphlops (as well as for
Typhlops) is "Mugoya" the same as in Luganda: in Lunyoro it is "Kinaboo."


__^M- ffl i9 -mwwww w wwww9

x2 '4

- w S ;

0 FThssart

1. Typh/ops punctatus punctatus.
2. Typh/ops congestus.
X 4. Typh/ops sudanensls.
6. Typhlops near blandfordi,.
8. Leptotyph/ops conjuncta.
I Not from actual specimen, but composite.

JohnBle, Sons & Curnow LtdLondor

la. Lateral Section. lb. Ventral Section.
3. Typh/ops sch/ege/ii mucroso.
5. Typhlops b/andford/i.
7. Leptotyph/ofs emin/.

8a. Lateral Section.

Presented by the Uganda Government.

8b. Ventral Section.

----- -- ---re

.. - ------ -----ulsn~


An additional locality previously referred to in a footnote is Ruwenzori : also,
this species is recorded extralimitally from Kissenyi at the northern end of Lake
Kivu by de Witte (1933).
Midbody scale-rows 14 and diameter at midbody 2 mm., the diameter being in-
cluded in the length 45-57 times, and the tail in the total length 8-1u times.
The type was described from Karagwe by Boulenger and not from Bukoba as
previously stated : two examples were collected there by Emin Pasha, the maximum
length being io mm.


Additional recorded localities include Kampala, Bugungu (near Ripon Falls),
Jinja, Budongo Forest and Kichwamba (W. Ankole). The specimen from Kich-
wamba measures just under five inches.
Midbody scale-rows 14 and diameter at midbody 1.5-2.5 mm., the diameter
being included in the length 32-72 times, and the tail in the total length 10-13 times.
Loveridge (1936)records that soon after death the diameter of an example he
collected in Kenya shrunk from 2.5 to 2.25 mm.: its length was 127 (115+ 12) mm.

Family BOID/E.



In the pythons and boas the ventral scales do not stretch completely across the
belly, but only extend over the central portion.
An additional name in Luganda is "Enziramire": in Lugishu the Python is
called "Nzatu", by the Karamojong "Emorotot", and in Kiswahili, "Satu."
Scales in 81-95 rows: ventrals 269-286: and subcaudals 63-77.
According to Major S. S. Flower (P. Z. S., Vol. 107, Ser. A, 1937, P. 32)
a female Python purchased by the Zoological Society of London on "2 September,
1849, which deposited eggs 13 January 1862, was still alive i January, 1863, after
13 years 4 months. And another female, deposited 13 February, 1923 ...............
died 24 July, 1936, after 13 years 5 months 11 days". The same author also
quotes an extract from a letter written by Dr. Burgess Barnett "a note of the his-
tory of a collection of Python sebae ........ caught in February 1929 by the side of
their empty egg "shells" and were presumably just hatched. They were hand-
reared (in the London Zoological Gardens)........ ...Two of them were first seen to

Copulate inter se in December, 1934, and eggs were laid in the following January.
The female then measured more than eight and less than nine feet. This is the only
evidence I know of, of the age at which they become sexually mature"-i. e. about
5 years zo months.
During 1935 a python's egg was hatched in the London Zoo for the first time.
The newly-born reptile was a little more than a foot long.
According to Christy (Big Game and Pygmies, p. 279) referring to the length
of time pythons can exist without feeding:- "Seven months is in fact quite a
short period............two pythons in the (London Zoological) Gardens went
without food for thirteen months and two and a half years respectively; but one
in Paris fasted for over three years."
Tsetse flies will feed freely on pythons, and on other snakes, when oppor-
tunity offers.
The cestode Bothridium ovatum is recorded from a Kenya example.
In Northern Rhodesia the natives of the region east of Lake Bangweulu are
most brutal in their treatment of pythons. When they find one of these big
snakes in a hole in an ant-hill they stick the point of a spear through its head
and then forcibly pull out the tail far enough to reach the claws, which are
carved out of the wriggling serpent. The snake is then hauled out and treated
with contempt, for these natives say that the life of the Python is in its claws!
According to Fitzsimons :- "These giant snakes (pythons) are frequently
afflicted with a reptilian disease called canker of the mouth, or in other words,
nasty scabby sores which spread and eventually kill the snake. This is a
difficult disease to eradicate, and requires persistent and prolonged treatment...
......If damaged teeth are not removed, and the mouth sterilized canker will soon
Further notes on the ailments to which pythons are subject will be found in
"Addenda-General Notes" in the section dealing with Diseases in Captivity.


Scales in 53-63 rows: ventrals 196-207 and subcaudals 28-37.

Subfamily BOINiE.


East African Sand Boa.

If the Sand Boa is eventually found within Uganda limits it is likely to be
referable to the East African, and not to the typical, race.
Scales in 46-59 rows; ventrals 162-182; subcaudals 20-27.


2. Typhlops near blanfordii.
(South-west Kigezi).

3. Typhlops near blanfordii.
(Budongo Forest).

I. Typhlops blanfordii.

The distinguishing characteristic of the East African race appears to be the
lower number of ventrals.
Loveridge (1936), in a series of fourteen examples from Kenya, records the
maximum measurement "634 (584+ 47) mm.," a female (presumably the total should
read 631 mm. C.R.S.P.).
Although the slashing bite of this Sand Boa is inocuous, actually the salivas
of certain sand boas contain powerful poisons though in minute quantity.
Also, referring to E. thebaicus, according to Corkill (1935, p. 13):- "It has no
fangs, but its genus has been shown to possess a poison-secreting gland, and its
saliva, as a result, must be considered poisonous. It is very doubtful whether it
should be regarded as a dangerous snake."


According to Corkill (1935):- "Certain of the species have been shown to
possess poisonous salivas."



Additional recorded localities include Bukakata, Kaianja (Lake Edward), Nan-
sere (Masaka), Mjanji, Lalle (Lake Kioga) and Kome Island.
An adult from Bukakata has a very abnormal tail measuringno more than
o1 mm., it is evidently the result of an injury inflicted when extremely young, though
superficially the tail looks free from damage.
The coloured plate unfortunately is a poor representation of a faded spirit
specimen and not of a "a brightly hued specimen" as stated in the text.
Scales in 18-19 rows: ventrals 131-151: subcaudals 44-87.
A fairly large Kenya male measures 464 (332+ 132) mm. and a female from
Kaimosi 491 (400+91) mm.
According to Loveridge (1936):- "the dark olive band or dorsal stripe (is) char-
acteristic of snakes from the Central African Lake Region. This stripe does occur,
though rarely, in the coastal (Kenya) series."
The same authority also records finding two females at Kaimosi on February
14 and 19 respectively, each of which held 6 well-developed eggs measuring approx-
imately 23 x 7 mm.; and six females at Ngatana (Kenya) between June i i and 20
which contained 6 small eggs: 6 eggs 15 x 6 mm.: 6 eggs 19 x 7 mm.: 6 eggs 19 x
iimm.: 6 eggs 22 x iomm.: and 6 eggs 23 x i r mm., these last quite ready for
The diet of this species includes toads,


According to Loveridge (in lit.) the letter "Z"(Zanzibar) included in the 1924
East African Check List is a misprint for "U" (Uganda), an error which he never
had the opportunity of correcting; in consequence Zanzibar is not included in the
range of this species.
This unmistakable snake has been described to the writer as occurring in
some of the Kyagwe forests besides the Mabira Forest.
The George Vanderbilt African Expedition of 1934 collected in French Equa-
torial Africa a female measuring 898 (733+165) mm., which is apparently an
"undescribed color variant" as "Instead of the five light, longitudinal dorsal lines,
it has only a red, hair-like, vertebral line on the dorsum with traces of a lateral one
on either side of the tail."


Lined House Snake.

In Lugishu this snake is called "Namage."
Additional recorded localities include Katebo, Kome Island, Kasiriye (Kyagwe),
Kagera River, Katunguru, Katwe, Bulisa (Bagungu), Kisoro (Kigezi), Bugishu,
Ruwenzori, Mt. Elgon (2,400 metres) and Rhino Camp (West Nile).
Loveridge (1936) draws attention to the fact that examples from the East
African interior plateau (2,000 to 7,000 feet) have from 29-34 midbody scale-
rows and 201-238 ventrals, while in those from the coastal plain (under 500 feet)
the variation is 25-27 and 186-213 respectively: he is also much impressed by the
general smaller size of a coastal series:- "The average total length of thirteen
males was 480 mm. as against an average total length of thirteen unselected males
from the more tropical central region (Sipi, Butandiga, Kaimosi) of 657 mm."
An Entebbe female, measuring 1041 (921+120) mm., on 7th November con-
tained seven eggs ready for deposition measuring 33 x 17 mm. A specimen from
the Ruwenzori measures 934 (8oi+ 133) mm.; a female from Kaimosi (Kenya) 944
(837+ 107) mm. and another female from Bukalasa 818 mm.
Scales in 2o-34rows:ventrals 186-240: subcaudals 41-71 : this variation refers
to material both from the East African interior plateau and the coastal plain.
The coloration of the eye varies from bright to dark brown, the pupil is
.An Entebbe snake in mid-August contained five eggs measuring 30 x I mm.
Loveridge (1936) records finding ova in four Mt. Elgon and Kaimosi specimens
between 16th December and ist March, i.e. one egg measuring 37 x 18 mm., one
egg measuring 28 x 13 mm., 7 eggs 29 x 15 mm., and io eggs ready for

Included in the diet of this species are shrews, geckoes, lizards and skinks.
A species of cestode of the genus Proteocephalus has been recorded from
B. lineatum, and the frequent occurrence of ticks, often in considerable quantity,
on this snake is particularly noteworthy.

Peters and Loveridge (1936) record an example of B. lineatum being found
in the stomach of a specimen of the rare Banded Harrier Eagle Circaetus
fasciolatus which was killed at Ngatana in Kenya.
According to Corkill (1935):- A young specimen.........that had a damaged
back, squeaked like a mouse when handled with forceps."


By the Bukedi this species is called "Musaga."
An additional locality which though extralimital is not many miles from the
Uganda border is Rutshuru in the Belgian Congo: this species is also recorded by
de Witte (1933) from Mahagi Port adjacent to the West Nile District and from
the Haut Uele in the north-eastern Congo.
Scales in 25-29 rows: ventrals 191-214: subcaudals 38-59.

The eye in a juvenile female about one-third grown is hazel, in a fully adult
female it is rich hazel.


In Lugishu the Wolf Snake is called "Wahobi."
Additional recorded localities include Katunguru, Katwe, Katebo, Kasiriye
(Kyagwe) and Bukalasa: also Rutshuru, which although in the Belgian Congo and
extralimital is not many miles from the S.W. Uganda border, and the Haut Uele in
the north-eastern Belgian Congo.
Loveridge (1936) records two females from Mt. Elgon measuring respectively
558 (495+63) mm. and 505 mm. (snout to anus) with an injured tail, the latter on
9th December containing six eggs in size 17 x 12 mm. The same authority also
refers to a small example from Kaimosi (Kenya) in length "267 mm. which held
but a single egg measuring 20 x 6 mm. on March 2, 1934."
The coloration of the eye in examples from Bukalasa has been described as
"red-brown, pupil black", and in another specimen which was evidently about to
slough "white with black lines and spots, pupil white."
Scales in 17 rows: ventrals 164-214: subcaudals 27-54.

According to Hewitt (in lit):-"I am a little doubtful as to Lycophidion capensis
capensis in Uganda. I believe that on the colour Uganda and Cape specimens are
always distinguishable. Colour alone is in my opinion sufficient for subspecific
or race separation."
A Kampala specimen has generally a blue-grey appearance: above it varies
from greyish-brown to slaty-brown with a greyish lustre in certain lights. Below it
is a delicate to bright grey-blue, the pale coloration invading the flanks for two to
four scale-rows, the scales above being edged darker imparting a pronounced
grey-blue effect. The tail is very short and is terminated by a sharp spine.
Included in the diet of this species are the small skinks of the genus Ablepharus.
Lycophidion lateral is recorded by de Witte (1933) from Lulenga (near Rut-
shuru) not far from the Kigezi border of S.W. Uganda.


An additional recorded locality is Rhino Camp, in the West Nile District.
A review of this species by Loveridge will be found in Publication 360, Vol.
XXII, No. I, Field Museum of Natural History, Aug. 1936, on pp. 25-27.
"These snakes are apparently ophiophagous", so a previous reference to the
absence of cannibal snakes from Uganda is incorrect. Also, according to Fitz-
simons :- "Most snakes are cannibals according to circumstances."


An additional recorded locality is Sipi (Bugishu) on Mt. Elgon. Although
Loveridge's male example obtained at Sipi on i8th December, 1933 is the first pub-
lished (1936) record of the occurrence of this species in Uganda, the writer actually
obtained one in Western Kigezi in November of the same year, the identification
being confirmed at the British Museum (Natural History). Extralimitally this species
is recorded by de Witte (1933) from the north-eastern portion of the Haut Uele
adjacent to the Uganda border.
In Lugishu this snake is called "Kangasira," and, according to Loveridge (i 936):-
"The Bagishu recognize the distinctiveness of this species from C. hoplogaster.
which occurs alongside it both at Sipi and Kaimosi, but which is far more abundant."
Scales in 13 rows: ventrals 146-165: subcaudals 74-91.
Three Uganda and Belgian Congo specimens in the British Museum collection
have scale counts as follow:-
Locality. Sex. Ventrals. Subcaudals.
Kayonsa, S. W. Uganda female 164 77
Ituri Forest ? 153 78
Irumu (Avakubi) ? 150 83

According to Loveridge (936) the coloration of a male from Sipi (Bugishu) is:-
"Above, dark olive with 104 deep black, irregular crossbands between head and
anus, these are represented on the tail by black flecks; the olive scales between
the crossbands are edged with pale blue on the anterior two-thirds of the body;
upper lip brownish-olive anteriorly, white below the eye shading off into olive
posteriorly. Below, throat pure white, anterior ventrals tinged with yellow,
remainder of the under surface dark green with its anterior third heavily suffused
with yellow; on the anterior two-thirds edged with yellow laterally, on the
posterior third with bluish white, on the tail with dusky.........The centre of the
eye is black surrounded by a light area, then by a fine orange line, then by an
olivaceus area flecked with black; outermost ring, black."
The 13 midbody scale-rows appears to be the only character whereby
C. carinatus can be distinguished from C. heterodernmus.


Additional recorded localities include Mt. Debasien, Sipi (Mt. Elgon) and
Butandiga (Bugishu).
It has been previously stated in error that this species is absent from the
Belgian Congo, for de Witte (1933) gives several Congo records including Faradje
in the Haut Uele in the extreme north-east.
According to Loveridge (1936) this snake is called :- "Emun (Karamojong):
Naranyase (Lugishu), but neither specific."
Scales in 15 rows: ventrals 141-178: subcaudals 82-109.
Loveridge (1936) out of a series collected in Uganda and in Kenya (just
south of Mt. Elgon) records the largest male as measuring 930 (700+ 230) mm.
but the tip of the tail is lacking; and the largest female 1045 (762+283) mm.
The same authority (1936) records several Uganda females collected between
i ith November and 5th January which contained from 4 to 7 eggs varying in size
from 13 x 5.5 mm. and 17 x 5 mm. to 31 x 13 mm.; and also gives a list of
stomach contents which include chameleons, skinks, a lizard, a gecko, toads and


An additional recorded locality is Mt. Debasien.
Previous mention of the non-occurrence of this species in the Belgian Congo
is evidently incorrect as de Witte (1933) records single specimens respectively
from Rungu and Dramba in the extreme north-eastern portion of the Haut Uele
adjacent to the north-western corner of the West Nile District.
Scales in 15 rows: ventrals 149-171 : subcaudals 77-120.
A Kenya example measures 917 (648+269) mm.

Three females collected by Loveridge (1936) in Kenya in May and June con-
tained from 5 to 7 eggs varying in size from 17 x 5 mm. to 23 x 8 mm. (and
18 x 9 mm.).
Peters and Loveridge (1936) record finding a specimen of this snake in the
stomach of an example of the Harrier Eagle Circaelus g. gallicus killed at the foot
of Mt. Debasien on 2nd December.


Synonomy-According to Loveridge (1937) in Zoological Results of the George
Vanderbilt African Expedition of 1934, Part VII-Reptiles and Amphibians, p
273 :- "I fail to see sufficient grounds for regarding heterolepidotus (Guinther) as dis-
tinct (from irregularis C. R. S. P.), nor even maintaining the latter as a race in view
of their ranges being practically coextensive". And, further, this authority draws
attention to the fact that in the key to the genus Chlorophis in the Catalogue
of Snakes, Boulenger (1894,2, p. 92) gives the subcaudals as 115-r35, the ventrals
as 175-190, whereas owing to an error in transcription 190 (subcaudals) has been
substituted for 135 at some later date, the error being "faithfully copied by Bou-
lenger, Schmidt and others though unsupported by the literature."
It is not proposed here to discuss whether or no Loveridge is correct in
his diagnosis, though attention can be conveniently drawn to the fact that in the
drawings of the lateral sections of these two species depicted in Coloured Plate (E)
the arrangement of the scales appears to differ considerably. The apparent dif-
ference, however, may be due to age variation in the two specimens used respect-
ively as models.


An additional recorded locality is Mt. Debasien. According to Loveridge
(1936) this snake is called "Emun" by the Karamojong though the name is not
specific and is applied equally to C. hoplogaster.
Scales in 15 rows: ventrals 148-191: subcaudals 91-137.


Hewitt (in lit.) mentions :- "The strongly marked keeled ventrals of Philotham-
nus remind me much of those found in Chrysopelea of the Malay region : as you
know that genus is known to "fly", or at least to make flying leaps fi om the tops
of the trees downward. Have you any local stories of similar behaviour in Philo-
thamnus? or in any other tree snake?"
In this connection on page 189 of A Naturalist on Lake Victoria by Hale Car-
penter will be found a most interesting note referring to snakes on the Sese Is-
lands:- "A similar bright green species might sometimes be seen high up among
the branches of trees, sometimes travelling rapidly and almost leaping across gaps

too big to be spanned. The manner in which it seemed to spring across from ohi
tree to another is very interesting, seeing that in the Malay Archipelago certain
snakes have carried the process further, and take gliding flights from a high to a
lower branch."
The snake to which Carpenter refers is probably either a species of Chlorophis,
Dispholidus typus or Dendroaspisjamesoni kaimosae: Philothammnus s. semivariegata
only occurs rarely on the Sese Islands.
Dendroaspis jamesoni kaimosae, the Uganda forest (green) mamba, evidently
indulges in conspicuous leaping antics when in the trees as its Luganda name is
"Bukizi" derived from the verb "buka" to leap or jump.


Additional recorded localities include Mt. Debasien (5000 ft.), Butiaba (Lake
Albert), Budongo Forest and Kome Island (Sese).
Scales in 13-15 rows (Loveridge, 1936, records a Kenya example which has
13): ventrals 149-207: subcaudals 12-160.
Loveridge (1936) refers to a Kenya female which measures 1233 (785+448)
mm.; and the same authority found three eggs measuring 32 x o1 mm. and three
others 25 x 6 mm. in females obtained on the Kenya coast respectively in early
May and at the end of June. A female from Kome Island measures 775 mm.
The Mt. Debasien example, a juvenile, had swallowed a gecko, Lygodactylus
p. gutturalis; Loveridge further records geckos in the stomachs of five Kenya
specimens and a frog in another; and also "One (of these snakes) was recovered
from the stomach of the rare Banded Harrier Eagle (Circadtus fasciolatus) at
Nagatana" (Kenya).


According to Loveridge (1936, in lit.):- "I do not think for a moment that
P. s. dorsalis from Lubwa's was anything but a P. s semivariegatus. Couldn't you
get a series from there and settle the point? I regard my Mwaya material as
intermediates agreeing structurally with dorsalis but in colour with semivarie-
gatus". In consequence it appears likely that dorsalis will eventually have to be
deleted from the Uganda List, though for the present it is being retained.
De Witte (1933) records dorsalis from Faradje in the Haut Uele in the
extreme north-eastern Belgian Congo, not very far from the Uganda border.


Cameroon examples measure 990 (602+388) mm., and 870 (540+330) mm., a
A female from the Budongo Forest measures 997 mm., and a juvenile from the
same locality 395 mm.

Loveridge (1936) quotes 974 (700+274) mm. as the largest ieasureient of
three females obtained at Kaimosi in Kenya, this specimen (on February 19th)
containing four eggs measuring 16 x 5 mm., in its stomach was a frog.

It is probable, almost certain, that this snake ought to be included in the
Uganda List.
At the time of going to press it has not been possible, owing to the lack of
comparative material, to examine critically an unsexed example of a species of
Rhamnophis obtained in the Kajansi Forest at mile 8 on the Kampala-Entebbe
In consequence this specimen has been submitted to the British Museum
(Natural History) for specific determination. It measures 1183 (815+368) mm. It
agrees with both Rhamnophis ituriensis and Rhaninophis aethiopissa elgonensis in
having 15 scale-rows at midbody: the ventral count is 157 which is lower than the
lowest available (159 in each case) record for either species: and the subcaudal
count I19 is markedly lower than Schmidt's (1923) 134 for ituriensis, but falls within
the range, i.e. 117-133, given by Loveridge (1929) for R. aethiopissa elgonensis.
Moreover, it agrees with aethiopissa elgonensis in having 7 upper labials, whereas
in ituriensis there are 8. The anal is divided.
According to Loveridge (1929) referring to aethiopissa elgonensis:- "Differs
from (typical) aethiopissa in lacking the spotting and edging of black on the head
shields." In the specimen under reference the two large occipital shields are broadly
edged blackish posteriorly. The coloration generally does not differ from that
previously quoted of the Budongo forest example of ituriensis.
It would appear that the Kajansi Forest snake is referable to R. aethiopissa
elgonensis and not to R. ituriensis. The stomach was empty.


In Lugishu, according to Loveridge (1936), black adults are called "Yakobe,"
and the dark olive adults and half-grown juveniles "Isilukanga."
Additional recorded localities include Kitala (near Entebbe), Sipi (Mt. Elgon),
Butandiga (foot of Mt. Elgon), Fort Portal and Toro: this species has been obtained
on Mt. Elgon up to an altitude of 8,ooo feet.
Scales in 17-21 rows: ventrals 188-203: subcaudals 130-155. The head shields
exhibit extraordinary variation; a Mt. Elgon adult personally examined was quite
unrecognizable from the head shields.
A large Mt. Elgon example measures 195 (tail 56) cm., the tail lacking the tip.
Its diet includes mice, tree rats, birds and chameleons. Developing ova were found
in 4 Mt. Elgon specimens between i4th December and I ith January, i.e. 12 eggs
measuring 19 x 8 mm.: 10-31 x 16 mm.: 7-35 x 8 mm. and 8-23 x 9 mm.




I. Thrasops jacksoni// jacksonii (juv.)
2. Gray/a smythi/ (juv.)
3. Dasype/tl's scaber rufouss form)
4. Dispho//dus typus (brown form)
5. Boi/a blandingl/ (juv.)

la. Lateral Section.
2a. Lateral Section.
3a. Lateral Section.
4a. Lateral Section.
5a, Lateral Section

Ib. Ventral Section.
2b. Ventral Section.
3b. Ventral Section.
4b. Ventral Section.
5b. Ventral Section


r 1:

1~------------ ----------

biinsie Bons 2 Curnow. I'd Lni~ton.

According to de Witte (1933) the closely-allied Thrasopsflavigularis has been
obtained in the highland region of the Belgian Congo west of Lake Edward.
Scales in 21 rows: ventrals 176-204: subcaudals 63-92. On 29th March a female
collected by Loveridge in Kenya contained two eggs measuring 40 x 8 mm.
Loveridge in the Bulletin of the Museum of Comparative Zodlogy, Vol. LXXIX,
No. 5, V, Reptiles, November, 1936, gives convincing reasons for relegating Coro-
nella regulars to the synonymy of C. coronata.
Scales in 19 rows: ventrals 168-205: subcaudals 63-77. A Kenya example
measures 520 (410+1 io) mm.
C. coronata at times closely resembles Natrix o. olivacea from which, however,
the former is at once distinguished by the absence of dark pigmentation on the
labial sutures and the much more numerous ventrals.
Additional recorded localities include Kitala (near Entebbe), Kome Island (Sese)
and the River Nile (near Jinja).
Scales in 17 rows: ventrals 149-165: subcaudals 90-o10. A very large female
from Mjanji exceeding 5 feet 6 inches in length, collected in December contained
eighteen fully-formed ova the size of pigeons' eggs. Another big female from Enteb-
be weighed 6 lbs. (stomach empty), and, in spite of several inches of the tail being
missing, measured 63 (52+ I I +tip) inches: the sides of the head, i.e. all upper and
lower labials, and below head, are deep yellow. This species is sometimes referred
to as a "ringed" snake.
A Kitala female measuring 1365 (1310+55) mm. on August i4th contained
numerous developing eggs still only about 20 x ro mm. It is curious to have to
record that ticks are occasionally found on this water-frequenting species.
As de Witte (1933) records Grayia ornata from Faradje in the Haut Uele in
the extreme north-eastern Belgian Congo it is possible that this species may turn
up in the north-western portion of the West Nile District, still more likely is the
possible occurrence there of Grayia caesar which the same authority has recorded
from Arebi which is considerably closer than Faradje to the Uganda border.
Scales in 15 rows: ventrals 134-145: subcaudals nio.
An additional recorded locality is Kisoro in S.W. Kigezi: de Witte also re-
cords it from the vicinity of Rutshuru in the neighboring Belgian Congo. A mea-
surement of 434 (384+50) mm. from S.W. Uganda is unusually large.

be Witte (1933) records this species from Mahagi Port in the Belgian Cong6
at the north-western end of Lake Albert adjacent to the West Nile District.
Scales in 15 rows: ventrals 133-152: subcaudals 19-34. According to Love-
ridge (1936) :- "Sexes of this species as judged by the counts of eight males and
seven females, may be readily distinguished by the ventral and subcaudal range, viz.
Males have 133-140 ventrals and 30-34 subcaudals.
Females have 140-152 ventrals and 19-24 subcaudals.
Major S. S. Flower (P. Z. S., 1933):- "obtained a specimen (of the western
Prosymna meleagris) at Singa on the Blue Nile."
On 28th May a Kenya female held three eggs measuring 20 x 6 mm. Measure-
ments of 238 (200+38) mm., male; and 254 (232+22) mm., female; are recorded
from Kenya.
Scales in 21-25 rows : ventrals 185-240: subcaudals 50-73.
According to Loveridge (1936):- "Pale gray flecked and spotted with white, this
juvenile colouring is very different from that of the adult."
A Belgian Congo adult female measures 1195 (1o2o+175)mm.
In Lugishu called Namagi" the name by which the House Snake Boae-
don lineatum is also (possibly erroneously) known.
Additional recorded localities include Bukalasa (30 miles north of Kampala),
Sipi (Mt. Elgon) and Butandiga (foot of Mt. Elgon).
A Cameroon female measures 898 (763+135) mm. A male from Kampala, col-
oured brown with darker bars or chevrons dorsally, measures 568 mm.
Scales in 21-27 rows: ventrals 185-263: subcaudals 33-96. A large brown Mt.
Elgon male measures 615 (518+97) mm.; and a large black female, also from Mt.
Elgon, 825 (725+100oo) mm.
At Sipi on 23rd December, a female contained numerous ova measuring 16 x 5
mm.; and on January i6th one at Kaimosi held 14 ova measuring 17 x 8 mm.
Loveridge (1936) records finding the stomach of a Sipi female distended with
yolk, in bulk equal to that of three hen's eggs!; it was also full of encysted nematodes.
It is just possible that Dasypeltis macrops may turn up in S. W. Kigezi as it
has been recorded by de Witte (1933) from N'Goma and Kissenyi at the northern
end of Lake Kivu, as well as from Lulenga near Rutshuru in all of which localities
D. scaber occurs.
In the reptile collection at the British Museum (Natural History) there is an
example of a species of Tarbophis from Wad Medani on the Blue Nile which has
not yet been specifically determined although superficially it resembles T. semi-

With reference to this specimen, according to Parker (in lit., 25.9.36):- "The
Sudanese Tarbophis is not semiannulatus. It might be either an abnormal obtusus,
an aberrant guidimakaensis or a new species; but with only a single specimen
I cannot decide and prefer to leave it un-named until further specimens can be
obtained". This specimen is mentioned as there is a remote possibility of examples
of a species of Tarbophis being found in northern and north-eastern Uganda.

One more example, a half-grown juvenile, was collected in the Mabira Forest
during September, 1937. A tiny mouse was found in its stomach : it also harboured
numerous ticks.
Additional recorded localities include Bugala Island of the Sese group in the
Victoria Nyanza and the Mabira Forest.
A female from the Budongo Forest measures 1745 mm. A Kampala example
(not sexed) with the tip of the tail missing measured 7 feet 9 inches: its colour is
described as :-"Darkish gunmetal with a purplish mottling towards the tail, belly
pale yellow from head to tail but with slight brown markings on the posterior half
increasing in occurrence towards and on the tail. Eye round and black."
A female from the Mabira Forest measures 1387 (1082+305) mm. i.e., 5 feet.
The coloration of this adult is strikingly handsome. Generally, it is rich, deep
chestnut-above and below. On each side laterally and alternately handsome
chocolate blotches I x I inch, and ; inch apart, each blotch containing a small white
spot on part of one, two or three scales.
b Posteriorly on the body, and along most of the tail the blotches are con-
fluent, though on the posterior portion of the tail the markings alternate again.
All blotches are conspicuously edged paler. Above, between blotches, the scales
are very pallid whitish brown, peppered or dusted darker. Anterior fifty ventrals
and below head, pale yellowish tinged green. Head above, dark brown. Posterior
three upper labials dull greenish grey. Upper labials broadly edged black, pos-
teriorly, except the last. Behind the large eye, two elongate dark brown blotches.
The snout is curiously truncated.
The whole effect is really lovely. Eye, iridescent hazel: the pupil appears
more round than vertically elliptical. This large adult came down to the ground
from a big fig tree by a kitchen hut after dark at 8.30 p.m.
Loveridge (1937) records a female from the Belgian Congo which measures
1990 (1540+450) mm., and which had a tick on the scales.
Additional recorded localities include Mt. Debasien in Karamoja, Katunguru
on the Kazinga Channel and Kome Island in the Victoria Nyanza.
Scales in 19-21 rows: ventrals 141-180: subcaudals 29-65. Eye, brownish or
greyish olive.

Loveridge (1936), referring to a series of thirty-four (one from Uganda, the
remainder from Kenya), cannot find the sex determinable from scale counts:- "thus
fifteen males have a ventral count of 157-166 and subcaudals of 45-65; twelve
females from 159-174 ventrals and 40-54 subcaudals."
According to Loveridge (1936) an example in the Field Museum of Natural
History, Chicago, measures 613 (528+85) mm.
During June five females collected by Loveridge in the coastal region of
Kenya contained from three to six eggs varying in size from 13 x 4 mm. to 27 x
12 mm. to 32 x 8 mm. to 35 x 1 mm.
A tapeworm (Ophiotaenia crotaphopeltis) was identified from a specimen collect-
ed (by Loveridge) on the Kenya coast.
According to H. W. Parker (in lit.) the "coarse creeping water grasses" pre-
viously described as associated with this snake's habitat have been :- "identified as
Panicum repens Linnaeus. It is a species widely distributed on the tropical and
subtropical coasts of both hemispheres and in India and Africa".

In the reptile collection at the Field Museum of Natural History, Chicago,
Loveridge (1936) records:- "the largest example of this race (from Ethiopia) which
I have ever seen, surpassing by over 150 mm. the biggest in a series of thirty-five
of the typical race which I collected in 1930 (in southern Tanganyika); it measures
1159 (995+64+tipof tail which is missing) mm."
Scales in 17 rows: ventrals 148-192: subcaudals 90-118. Loveridge (1936) re-
cords large Kenya male which measures 1025 (715+310) mm., anda large Kenya
female of 1334 (922+412) mm.
An additional recorded locality is Ongino, Teso (south-east end of Lake Salis-
bury), which extends the range of this species a further 180 miles to the east than
was previously known (i.e. Bagungu, on north-east shore of Lake Albert).
In Teso this snake is called "Enyeneropi." The Ongino specimen is a female
which, on 27th October, contained several small, elongate eggs in course of develop-
Scales in 17 rows: ventrals 177-196: subcaudals 118-178. Loveridge (1937)
refers to a Kenya female measuring 1532 (972+560) mm. collected by the George
Vanderbilt African Expedition of 1934 which contained to eggs measuring ap-
proximately 40 x 15 mm.

Lizards, including the large species Latastia longicauda, are favourite food of
this snake.
Additional recorded localities include Mt. Debasien (6,000 feet) in Karamoja
and Katwe (Lake Edward). It is impossible at present to amplify or amend the
remarks on distribution and habitat which have been previously recorded.
Loveridge (1936) records a Kenya male which measures I182 (792+390) mm.
The pair of black lines along the belly, sharply defined and clear, constitute a
good distinguishing character; in Dromophis lineatus, a species which closely
resembles P. subtaeniatus, especially when preserved, there are transverse lateral
streaks on the ventrals which is never the case in the latter.
The Debasien specimen was collected by Mr. W. J. Eggeling who found it
under a rock "where I disturbed it (in January) when digging up a plant."
Frogs are included in the diet of this species, which also is recorded as har-
bouring nematodes. The cestode Oochoristica crassiceps has been recorded from a
Kenya specimen.
In Lugishu this snake is called "Namasanurugi", and in Teso "Enyeneropi"
the name by which Rhamphiophis oxyrhynchus is known.
Additional recorded localities include Sipi (Mt. Elgon), Butandiga (foot of Mt.
Elgon), Ongino in Teso (south-east of Lake Salisbury) where it is evidently plen-
tiful, Katwe (Lake Edward), Katunguru on the Kazinga Channel, Kitala near
Entebbe and Katebo (north-eastern Victoria Nyanza).
A male from Sipi is recorded measuring 1095 mm. from snout to anus, and a
female from the same locality is 1303 (935+368) mm.: a female from the Budongo
Forest region is 162 mm.
According to Loveridge (1936):- "The ventrals of the forest-edge specimens
from the Central African material (Sipi, Butandiga, Kaimosi) exhibit well-defined,
though dusky, lateral lines". Psammophis sibilans and P. subtaeniatus appear fre-
quently to be almost inseparable where Uganda material is concerned.
Loveridge (1936) obtained three females on, or in the vicinity of, Mt. Elgon
(in Uganda and Kenya) between 13th December and i8th January, which held res-
pectively 4eggs measuring 13 x 6 mm., io eggs measuring 27 x 10 mm., and 7 eggs
-measuring 38 x 19 mm.; the last were quite ready for laying. Developing ova were
found in an Entebbe female in late October.
Included in the diet of this species is the very angular, rough-scaled lizard
Agama atricollis, and the ill-flavoured, strong smelling shrews.
This species is sometimes very heavily infested with the nematode Physaloptera
paradoxa: it is also likely to harbour nematodes of a species of Kalicephahts. In an
Entebbe snake was found the blood parasite Haemogregarina brendae.


According to Loveridge (in lit., 1937):- "Psammophis brevirostris is at most a
southern form, Transvaal southwards, of sibilans." It can apparently be confused
with Trimerorhinus t. multisquamis. De Witte's record (1933) of a series of twenty-
three examples of P. brevirostris from the Belgian Congo in consequence requires
critical investigation. It is evident from the existing confusion that a revision of
the status of certain species of the genus Psammophis is necessary. P. brevirostris
must be deleted from the Uganda List.


Scales in 15 rows: ventrals 142-165: subcaudals 92-131. Loveridge (1936)
records from the coastal region of Kenya a male which measures 872 (541+331)
mm., and a female measuring 1o2o (660+360) mm. According to the same authority
the large lizards of the species Latastia longicauda are preyed upon frequently by
this snake.

Additional recorded localities (1936) are Lutoto hill (Western Ankole), the
vicinity of Fort Portal, and the Budongo Forest. A large male from Kenya mea-
sures 1422 (821+601) mm. In the Field Museum of Natural History, Chicago, the
largest example measures 1470 (935+535) mm.
The heads of three specimens examined from Western Uganda are all uniform
green above. This snake has been found to harbour cestodes of a species of

Additional recorded localities include Butandiga at the foot of Mt. Elgon,
where Loveridge obtained a brownish example, and Bukalasa (30 miles north
of Kampala).
Scales in 19-21 rows: ventrals 161-201 : subcaudals 87-131. A large male
from Mt. Mbololo in Kenya measures 1298 (962+336) mm., and a large female from
the same locality 1391 (1073+318) mm. : these constitute a most interesting pair,
for according to Loveridge (1936) their coloration is, in the male :- "Top of head
bright brick red like the red soil of the region, back nut brown but the keel of each
scale partly white. Below, buffy white"; and in the female :- "Top of head as in the
male but the back reddish brown, tail bright pink. Below, pink".
The coloration of an Entebbe male is:- "Deep grass-green, concealed edge
of each scale bright blue: interstitial skin, black. Sides of ventrals deep yellow-
green. Below, yellowish with a greenish tinge".
A Bukalasa female is described as:- "Pinkish-brown on yellowish. Eye, pale
green; pupil, black".




C~ 7 ~

Naja goldii
me/ano/euca (juv.)
Dispholidus typus (juv.)
Dasype/fis scaber (black form)
Psammophis subtaen/atus (juv.)
Apara//actus nigroco//aris



JohnBaieSons & Crnow. Lt London.

Presented by Mr. R.M.Grant

la. Lateral
2a Lateral
3a. Lateral
4a. Lateral
5a. Lateral
6a. Lateral


lb. Ventral
2b. Ventral
3b. Ventral
4b. Ventral
5b. Ventral
6b. Ventral



Scales in 17-21 rows: ventrals 163-208: subcaudals 16-38. According to
Loveridge (1936):- "The rostral apparently develops with age as in Prosymna,
it is not distinguishable from that of a Rhinocalanius of similar size so should be
avoided as a key character."
A large Kenya female measures 434 (410+24) mm. Quoting further from
Loveridge (1936):- "Shortly after capture, the Changamwe (Kenya) snake dis-
gorged a wolf snake (Lycoplidion c. acutirostre) only 20 mm. shorter than itself.
The similarity in the parallel development of these two blackish, burrowing
snakes was striking. The head of the wolf snake was too digested for a labial
count, but it was a male with midbody scale-rows 17: ventrals 154: anal single:
subcaudals 31, having 9 ventrals less, and 4 subcaudals more than its vanquisher.'

An additional recorded locality is Katebo on the north-west coast of the
Victoria Nyanza, where a specimen was obtained in October, 1936.

This species is not included in the Uganda list though it is possible that it may
eventually be found in south-west Kigezi as it has been recorded by de Witte (1933)
from N'Goma at the northern end of Lake Kivu.
Loveridge (Bull. Mus. Conp. Zool., Vol. LXXIV, No. 7, Oct. 1933) mentions
"The only record in Africa east of Uganda of which I am cognisant", referring to
a record which is not available to the writer and which may concern present-day
A male and a female collected by the American Museum Congo Expedition
measure respectively 722 (tail 39) mm. and 652 (tail 27) mm.; another female from
the Congo is 661 (617+44) mm., and a Tanganyika female 479 (448+31) mm.
Scales in 15 rows:ventrals 215-252: subcaudals 18-27 : anal divided.
The normal coloration is uniform black above and below: according to Love-
ridge (1933) evidently referring to a specimen which had recently sloughed it is:-
"Uniformly iridescent blue-black above and below": in one colour phase longitudinal
lines are apparent.
Schmidt (1923, P. 119) describes the coloration of two Congo examples compre-
hensively :- "Dorsal color-dark grayish brown, darker posteriorly, head grey.
Two scale-rows next ventrals lighter and more reddish brown. Venter uniform
light yellow. Yellow of venter often extends on to sides of neck, suggesting ap-
proach to nuchal collar often present in species of this genus."

Scales in 15 rows: ventrals 143-166 : subcaudals 50-71. A large Kenya female
measures 520 (420+ 00oo) mm.


A specimen (not sexed) from the Budongo Forest measures 478 mm., another
from Cameroon 409 (335+74) mm., and a female from the Belgian Congo 401 (350+
51) mm.
The tail in male snakes is usually markedly longer than is the case in the

In Lugishu this snake is called "Mugoya", a name which in Lusoga and
Luganda refers to any snakes of the genera Typhlops and Leptotyphlops.
Additional recorded localities include Bukalasa (30 miles north of Kampala),
Sipi (Mt. Elgon) and the region at the mouth of the Kagera River.
In two females from Bukalasa the coloration of the eye is respectively, translu-
cent with silver pupil, and black with invisible pupil : one of these snakes has a
white tip to the tail.
The cestode Proteocephalus elapsoidea has been recorded from a Tanganyika

In Lugishu this cobra is called "Wahobi", the name by which the wolf snake
Lycophidion c. capensis is also known.
An adult male collected at Entebbe has the eye blackish with a narrow pale
yellow inner ring and the pupil black.


A species of cestode of the genus Taenia is recorded from an example of the
"spitting" cobra.

According to Loveridge (in lit., 9th September, 1937):- "It 'spits' its venom as
I have seen it do in the Bronx."


Specimens collected at Entebbe have sometimes harboured a few ticks.


This species is often referred to as the Forest Puff Adder.