Ube journal OF THE East Africa and Uganda Natural History Society / / November 1918 No. 13. CONTENTS PAGE . 279 “PITTA ANGOLENSIS LON GIPENNIS. ’ ’ Dr. V. G. L. yan Someben THE LUMBWA AND ELGON CAVES. C. W. Hobley . . . .280 DESCRIPTION OP CAVES ON NANDI ESCARPMENT. T. L. Deacon . 292 NOTES ON SALT CAVES IN LUMBWA DISTRICT. A. Knight-Bruce 295 NOTE ON THE FORMER LEVEL OF THE VICTORIA NYANZA. Oswald ......... GAME AND DISEASE. A- Blayney Percival . ON SNAKES IN EAST AFRICA. A. Loyebidge A VOLCANIC ERUPTION IN EAST AFRICA. C. W. Hobley . NOTES DOLPHIN-FISHING OFF MOMBASA. Capo?. G. W. Woodhquse JACKALS ATTACKING NATIVES. C. M. Dobbs . FISHING IN LAKE RUDOLF. A. Blayney Percival A SHOOTING INCIDENT. H, C. Babnes . ABNORMAL TIDES. Editor ..... EARTHQUAKE SHOCKS IN EAST AFRICA. Editor REPORT OF SOCIETY. The Hon. Secretary . BALANCE SHEET OF SOCIETY. The Hon. Treasurer LIST OF MEMBERS ...... PLATES AND ILLUSTRATIONS LUMBWA AND ELGON CAVES — Diagram . MAP OF PORTION OF LAKE VICTORIA BASIN . NANDI ESCARPMENT AND CAVES . LUMBWA SALT CAVES-Sketch Map . BAGAU SALT CAVES-Plan ..... BAGAU SALT CAVE ...... MAP OF LAKE NATRON AND VOLCANIC REGION NILE PERCH-FISHING IN LAKE RUDOLF . Felix 300 315 343 347 347 34S 348 349 350 352 354 281 Plate to face 298 . 341 Plate to face 347 EDITOR C. W. HOBLEY, C.M.G. LONGMANS, GREEN, AND CO. 39 PATERNOSTER ROW, LONDON. NEW YORK. BOMBAY, CALCUTTA AND MADRAS All rights reserved Additional copies: to member s, Bs, 3 (4j-) ; to non-members. FEB 3 joio ^a/ iWu*^ ' •* THE JOURNAL EAST AFRICA AND UGANDA NATURAL HISTORY SOCIETY HIS EXCELLENCY R. T. CORYNDON, C.M.G. SIR H. C. BELFIELD, K.C.M.G. SIR E. P. C. GIROUARD, KC.M.G., R.E., D.S.O. SIR JAMES HAYES SADLER, K.C.M.G., C.B. president HON. C. C. BOWRING, C.M.G. Executive Committee E. BATTISCOMBE, Assoc. Arbor. Society J. P. COOK, M.B.O.U. R. W. BURKITT, F.R.C.S.Ire. F. G. HAMILTON J. H. H. PIRIE, F.R.C.P., F.R.S.Ed. W. A. PAILTHORPE A. W. KEMPE W. KENNEDY, M.R.C.Y.S. Ibonorarg {Treasurer W. McGREGOR ROSS, B.A., M.Sc., B.E. IbonorarE Secretary . van SOMEREN, L.R.C.P. & S., L.R.F.] L.D.S .Ed., M.B.O.U. Curator Vacant. ^Editor of Journal C. W. HOBLEY, C.M.G. 1918 OF THE November 1918 No. 13 patrons SIR H. HESKETH BELL, K.C.M.G. No. 13. U CONTENTS PAGE PITTA ANGOLENSIS LONGIPENNIS. Dr. Y. G. L. van SOMEREN .......... 279 THE LUMBWA AND ELGON CAVES. C. W. Hobley . . .280 DESCRIPTION OF CAVES ON NANDI ESCARPMENT T. L. Deacon 292 NOTES ON SALT CAVES IN LUMBWA AND DISTRICT. A. Knight- Bruce 295 NOTE ON THE FORMER LEVEL OF THE VICTORIA NYANZA. Felix Oswald 300 GAME AND DISEASE. A. Blayney Percival .... 302 ON SNAKES IN EAST AFRICA. A. Loveridge .... 315 A VOLCANIC ERUPTION IN EAST AFRICA. C. W. Hobley . 339 NOTES g DOLPHIN FISHING OFF MOMBASA. Capt. C. W. Woodhouse . 343 JACKALS ATTACKING NATIVES. C. M. Dobbs .... 347 FISHING IN LAKE RUDOLF. A. Blayney Percival . . .347 A SHOOTING INCIDENT. H. C. Barnes 348 ABNORMAL TIDES. Editor 348 EARTHQUAKE SHOCKS IN EAST AFRICA. Editor . . .349 REPORT OF SOCIETY. The Hon. Secretary .... 350 BALANCE SHEET OF SOCIETY. The Hon. Treasurer . . 352 LIST OF MEMBERS 354 PLATES AND ILLUSTRATIONS LUMBWA AND ELGON CAVES, DIAGRAM . . . .281 MAP OF PORTION OF LAKE VICTORIA BASIN . . .287 NANDI ESCARPMENT AND CAVES 293 LUMBWA SALT CAVES, SKETCH MAP 296 BAGAU SALT CAVES, PLAN 297 BAGAU SALT CAVE Plate to face 298 MAP OF LAKE NATRON AND VOLCANIC REGION . . .341 NILE PERCH— FISHING IN LAKE RUDOLF . Plate to face 347 PITTA ANGOLENSIS LONGIPENNIS 279 PITTA ANGOLENSIS LONGIPENNIS (Reichenow) By V. G. L. van Someren I wish in this short note to draw attention to this interesting and somewhat rare bird, in the hope that some member may be able to obtain specimens and forward them to the Museum. The true home of the Pittas is the Malay Archipelago, including the Islands of Borneo, New Guinea, and Sumatra, while certain other species have been found in India and Australia. Considerable interest was aroused when an example of a true Pitta was discovered in West Africa (Angola) in 1816, and named Pitta angolensis. Later on, a second species was discovered in Central Congo, and although referred to P. angolensis was afterwards proved to be a distinct species and called P. Beichenowi. In 1899, Alexander procured specimens of a Pitta, which he referred to P. angolensis, from the Nyassaland district and Zambezi ; but Reichenow, in working over this group, showed that the Eastern birds belonged to a distinct sub-species, and named them P. angolensis longijpennis. Apparently no other specimens of this Eastern form were collected until my collectors procured a single example in the great Mabira Forest of Uganda, and another specimen was captured in the Grand Hotel in Nairobi in 1912 ! This later specimen flew into the hotel one evening, and was caught ; unfortunately the skin was by some mistake sent to the British Museum. These birds are no doubt more plentiful than supposed ; but, owing to the fact that they inhabit the thick dense forests, they are easily overlooked. Netting them would no doubt give better results than shooting — as was proved to be the case in New Guinea. They lead, more or less, a terrestrial life : feeding on the insects which they obtain from amongst the fallen leaves. 280 THE LUMBWA AND ELGON CAVES So far as my information goes, the nest of this Eastern form has not been found ; but that of other species is described as being a domed structure, placed on the ground, and con- structed of roots, leaves, and twigs, and lined with fine rootlets and grass. The eggs are white speckled with purplish grey. A characteristic feature of these birds is their long tarsi which enable them to hop and run with great rapidity. These birds, though conspicuously coloured, are difficult to see in the gloom of the great forests. THE LUMBWA AND ELGON CAVES, WITH SOME REMARKS ON THEIR ORIGIN AND THE GEOLOGY OF THE REGION By C. W. Hobley In the Lumbwa highlands there are to be found an ex- tensive series of caves, the occurrence of which has been the subject of considerable discussion, and it is only of late that systematic examination has been possible. It is now proposed to record the information available up to date. The exis- tence of these caves was first brought to my notice by my friend the late H. B. Partington about eleven years ago. He was then the Commissioner of Lumbwa District. He informed me of the existence of certain caves called ‘ Kipchebos ’ on the Kiptiget River, and stated that at this place there was an extensive chain of caves which, according to the natives, ran due south for a long distance, and even extended under the bed of the Sondu River. He recorded the fact that the natives excavated a kind of salt earth from these places, but pronounced no opinion as to their actual origin. During the last year these caves have formed the subject of investigation by Mr. Dobbs and Mr. Knight-Bruce, and some notes by the latter are given below, with a map of the area and a plan and section of one of the caves. 282 THE LUMBWA AND ELGON CAVES Both Mr. Dobbs and Mr. Knight-Bruce appear to be of the opinion that these caves are the work of the Lumbwa natives excavating in search of salt earth for the use of their stock. Until further evidence is available this view must stand, for the reason that the natives are digging in the caves and carrying away this earth at the present time. I only recently had an opportunity of examining a number of these caves, and I am able to support the view that they are of artificial origin. The only other series of caves of such magnitude occurs on Mount Elgon, the inhabitants of which are the Lako, El Gonyi, and Savei, and are, by the way, of the same race as the Lumbwa or Kipsikis. I visited a number of these caves many years ago and tentatively formed the opinion that some of them were the work of wave-action operating upon soft layers of volcanic ash during a period when the waters of an inland sea attacked the flanks of the mountain, the mountain mass being later on raised through tectonic folding of the earth’s crust. There is little doubt that the waters of the inland sea now known as Lake Victoria once covered a greater extent than at present ; in fact, there is clear evidence that they ex- tended up the Nyando valley certainly as far as Muhoroni, but that only premises a comparatively small rise in level. Now we know that in Miocene times the land comparatively near to the present shore of the lake was the habitat of beasts such as the Dinotherium, and the bones found in the beds near Karungu could only have been deposited in swampy flats of no great depth, and the greater extent of the lake at that period corresponds quite well with its extension up the Nyando valley and over the Kitosh plain. Dr. Oswald has recorded beds of Miocene age near Karungu about 140 feet thick. In addition to the facts just recorded, there is a theory that at a later date than Miocene times the lake extended far away to the north-east past Mount Elgon, and certainly an examination of the map produces the impression that originally the waters of the lake may have included Lake Kioga and also the chain of lakes called after Salisbury and Gedge in Kumama country ; if this is so, a careful examination of the area would probably disclose traces of old lake-terraces, THE LUMBWA AND ELGON CAVES 283 but nothing of this nature has as yet come to light. The above theory also premises that the outlet of the lake sea- wards was to the north-east, possibly through the Rudolph trough, or if not, into the Pibor flats and so on into the Sobat valley. The lowest line of caves on Mount Elgon must be at a height of nearly 6000 feet, and some, I think, occur up to the 7000 feet level, and the caves in Lumbwa are stated by Mr. Knight-Bruce to occur at the following heights : Bagau cave 7000 feet, Gitoi cave 6000 feet, Kibipiten cave 6000 feet, Kiptoit 6900 feet, Sausit 6100 feet, a total range in altitude of 1000 feet. Now if the caves on Elgon are attributable to excavation by water action we must adopt one of two hypotheses : either that there was a vast inland sea extending up to the Mau Escarpment on the east, and (if at an altitude of 6000 feet to 7000 feet) bounded by mountain ranges of still greater altitude on all other sides, which is somewhat inconceivable ; or we must fall back on the theory that in comparatively recent times the mountain- mass of Elgon was vastly lower, and at that time in contact with the inland sea, and that it has since been raised by earth movements to its present level. With regard to the second proposition, viz. the elevation of the mountain : this problem is not an easy one, for great mountain masses like Elgon would not, as a rule, be likely to be raised to any great extent, for after the occurrence of a vent in the earth’s crust through which volcanic activity is manifested and a vast amount of material ejected, there comes a period of quiescence, and the tendency would be for the mass of ejecta to settle and arrive at a position of equilibrium as regards the surrounding area, for the piling up of an enormous mass of material cannot fail to produce a local stress, which has to be slowly adjusted. If this adjustment occurs very gradually nothing happens ; if it occurs near a great line of weakness, such as a big fault plane, and abundant water is present, further eruptions probably take place. With regard to Elgon our knowledge of the sequence of events is faulty, but, as before remarked, there appears to be little doubt that at one period early in the history of the moun- 284 THE LUMBWA AND ELGON CAVES tain the waters of Lake Victoria extended far to the north- east and covered what is now Bnsoga and Bukedi country and extended into the Karamojo plains north of the mountain. If this is correct we have the water necessary to excavate caves, but at first sight it seems unlikely that the waters of this extended sea were more than a few hundred feet above their present level, and we have caves on Elgon some 2000 feet above the ordinary level of the surrounding terrain, so that the question then becomes ‘ Can the mountain have risen ? * It may have risen to some extent, for the following reason : At one period in its history there was evidently a great convulsion, for the eastern wall of the mountain is seen to be torn asunder, and a mighty stream of lava issued from the breach and flowed eastward for many miles over what is now the northern part of the Uasin Gishu Plateau, and formed a ridge which now separates the watershed of the Turkwell and^Nzoia rivers. Judging by the way the region in the vicinity of the Lake is faulted, there can be no doubt that the crust of this portion of the earth is in a condition of stress, and it is quite possible that the removal of several million tons of material by the eruption above referred to so lightened the mountain mass that it rose owing to reduction of local stress. The drying up of what may be termed the north-east arm of the lake would also further relieve the pressure on the crust in this area. These causes would, however, be unlikely to produce movement to the required extent. On the north side of the alluvial plain between Muhoroni and Kibos there is a gigantic escarpment running east and west composed of ancient gneisses and schists, and rising nearly 2000 feet above the plain. High up on this escarpment there are a number of caves : one was the scene of a fight during the Nandi expedition about 1901, and it was said to be big enough to hold 200 head of cattle. These caves have recently been explored by Mr. P. L. Deacon, District Commissioner, Nandi, and his description leaves little doubt that they are what is called fault-caves. We have on the Nandi Escarpment a mass of metamorphic rocks which are traversed by granitic dykes ; a fracture in the earth’s crust which runs from the Maragoli Hills north THE LUMBWA AND ELGON CAYES 285 of Kisumu due east to opposite Muhoroni, and this produced a gigantic fault which was the cause of the Nandi Plateau. At a later date a north and south fracture' occurred which was the origin of the fault scarp which runs from Kabras to Kibos. The reason for the occurrence of this fault on an east to west line is obscure : the folds of the continent are from east to west, and therefore most of the fault scarps run north and south as in Taita, the Ukamba country, the Rift Valley, &c., and one cannot but suspect that it is in some wry con- nected with the intense volcanic activity which is centred in Tinderet. This problem will, however, not be clearly solved until a detailed geological survey * of the region is made. There, however, appears to be little doubt that the caves on the Nandi Escarpment are due to the falling together of masses of the granitic dykes, fractured and dislocated, during the formation of the fault scarp. A certain amount of water erosion has doubtless increased the cavities, for we hear of the occurrence of small streams in some of them. On the south side of the Nyando valley we have what is probably a similar fault scarp, but it is not so sharply defined, and the upthrow on the face is not so great ; in fact, it appears to have been masked to some extent by volcanic deposits, but the foci of eruption have not been identified. It is, however, probable that the Lumbwa highlands have been elevated by faulting in Tertiary times, but the information regarding the area south of the Nyando valley is at present too scanty to warrant any dogmatic conclusions. To refer again to the east and west fault scarp parallel with the railway between Muhoroni and Kibos (as a matter of fact, the bearing or strike of this fault is about W.N.W. and E.S.E.), The highest portion of the Nandi Plateau is close to the fault plane, and the terrain of the Nandi country falls steadily to the north until the valley of the Nzoia is reached ; north of the Nzoia valley the country gradually rises again towards Mount Elgon. It would therefore not appear that any causes which produced the elevation of the Nandi and Lumbwa plateaux would have appreciably affected Mount Elgon and raised that mountain to a similar extent. As previously mentioned, there is a theory of considerable 286 THE LUMWBA AND ELGON CAVES probability to the effect that the lake once had a great ex- tension to the north-east, and it is quite likely that this Nandi fault line, extending in a W.N.W. direction along the southern face of the Maragoli Hills and on past the Samia Hills, slightly raised the northern shore of the lake between Samia and the Nile, and cut off the ancient north-east arm, thus forming the present lake shore. The upthrow of the fault would naturally decrease to the westward. A section and map showing the faulting of this region is given. To come back to the Elgon caves and the possible modes of formation. Caves can be classified under various heads. There are the caves of limestone countries due to the excava- tion of the rock by a process of solution by water charged with carbonic acid gas, which finds its way down the joint planes and along the bedding. There are the caves formed by the action of waves beating against a rocky shore and detaching a bed of softer rock. Examples of both these classes are common in Europe, and examples of the latter class are to be found along the shores of East Africa. Now what do we find in the way of caves in volcanic countries ? We find caves due to the occurrence of huge bubbles in lava, which are caused by accumulations of water met with during the course of the lava stream and which generate steam ; the lava cools rapidly on the surface and forms a crust over such pockets of water -vapour. Ages afterwards the lava weathers away and exposes such cavities. Other caves in lava occur owing to the rapid cooling of the crust and the molten mass under- neath breaking out in a weak place and leaving behind it a cavity covered by a thin crust. These caves are more common than those caused by steam, and many examples are to be found in Hawaii, Iceland, &c. Caves are also sometimes formed upon lines of faults ; the space between the two rock walls is often filled with broken debris, and this may be removed by the agency of water, leaving curious caverns. Since examination of the Lumbwa caves, I must confess to being prejudiced against my former view — that the Elgon caves were caused by the action of water, especially as it necessitate the foundations of a satisfactory theory explain- THE LUMWBA AND ELGON CAVES 287 *1% ft. Ilkwi JSJYANZA 57£0 Map showing North-east Portion’of the Lake Victoria Basin. Scale approximately 1 inch = 35'5 miles. 288 THE LUMBWA AND ELGON CAVES ing the elevation of the mountain mass by some 2000 to 8000 feet. We should, however, consider all the facts available. Sir Frederick Jackson recently informed me that when he made the first ascent of the mountain in 1889 he saw at one spot a cave half-way up a cliff about 200 ft. in height, and in a position quite inaccessible by man, but of course the floor of the valley in which it occurred may have been very rapidly eroded. I remember visiting one about 1899, which had a great vaulted hall nearly 30 ft. in height, with a spring dripping from the roof and with a colony of natives living inside ; this cave tapered away in its recesses to a height of about 2 ft. I asked the natives if it was excavated by man, but they ridiculed the idea. In some of the caves, however, marks were seen where fragments of the soft ash had been chipped off with the butt end of spears. A hurried examination of the floor of a cave to a depth of 4 feet demonstrated that it had been inhabited at intervals for a considerable period, for charred wood, ashes, and frag- ments of bones were found at varying depths. I should like to add that none of the caves visited by me were in lava, so they could hardly have been caused by cavities in eruptive masses or by steam bubbles ; they were invariably in porous volcanic ash or agglomerate. In one or two cases there was a waterfall over the mouth, but no evidence that the stream once flowed through the cave instead of over the top. The majority of the caves were very dry and dusty inside, and some swarmed with fleas. The present population on the mountain is scanty, and it is hard to believe that they could account for the numerous caves on the mountain. Some of them are high up in the bamboo zone, and far above the limit of human habitation ; others again are on the east flanks, which are now uninhabited. The present inhabitants, moreover, do not appear to be excavating anything from the caves. Of course if it could be shown that the mountain is, or has been, rising with comparative rapidity, it might just be possible that what is now the high forest zone was in Palaeolithic times a zone habitable by man — that is to say, if one entirely adopts the view that they are excavated by man. Primitive man, THE LUMBWA AND ELGON CAVES 289 as far as we know, rarely, however, excavated caves ; his implements were of too rude a character for extensive work of this kind, and he generally occupied such caverns s s nature provided. We have no evidence of the period which has elapsed since Elgon and the Nandi and Lumbwa plateaux were first occupied by man ; the present tribes have not been in occupation more than a few hundred years at the outside : they came in from the country north-west of Elgon, viz. from the hilly country east of the Nile. They were also probably the first people to bring domestic animals, such as cattle, sheep, and goats. If this premise is correct, it tends to confine the period during which natives would have dug into the earth for salt. When they arrived in this area from the north we do not know if they found the country already occupied by agriculturists, but some think that this may be the case on account of the existence of what appear to be ancient dams for conserving water for irrigation. The real ancient inhabitants were, however, the Dorobo or Oggiek or Okiek, and these were the people who fashioned the stone implements, arrow-heads, and bowls which are now being brought to light as the country is opened up. They are generally classed as Neolithic in type, but this diagnosis fails to give us a date, for there is no reason to believe that stone implements in Africa are contemporaneous with implements of a similar type found in Europe, and which were gradually superseded by those of the bronze age, and later on of the iron age, and so on until we can obtain a grip upon a basis of historical dates. In order to render some of the previous references to earth movements, &c., clearer to readers, the tabulated statement on p. 290, given by Professor Gregory (vide his book, ‘ Great Eift Valley,’ p. 285), is republished. It is an attempt at the correlation of the volcanic rocks and earth movements in British East Africa, and is a marvellous example of geological insight, especially when one considers the limited time that scientist had in the country and the scanty information at his disposal. So far as I am aware, the sequence he sets forth generally stands, but I am venturing to suggest the following amend- ments : East African Sequence of Volcanic Series according to Professor Gregory. 290 THE LUMBWA AND ELGON CAVES dn a a° FH © a C .2 s 8,1 a g ^ a a£ o.o... W). s 0'43 o (S ° * 02 © as d a a 02 a a _o ^ .2 1> .2 43 § ap 3 P< a FH © J ^3 o © •a **3 43 a 02 © £ ^ a a, a © g1 oB 02 13 m a m a © 4^> r2j P 5 eS i-rt +3 CS OQj © c, ^ « S S o -S g a « *4H 0 s-1 •h 4i a ^ a ■© -a as (D ^ as ■£ bo 02 t> K c3 Sr§ . OB* ® "osS a a * ® a © .© «+H >> a O r— , 02 +) IS M £>.S M [X 8®* Fh >» CD CD o 15 © bO T3 s 14 |I d3 '43 P 2-2 £ 2 o a ° d 5P-2. ^•5 S hct.2 a o x a: a © -a g> tsM ^ a^ § rf o P 6 > M & c3 i~i. pq S4 Ph Ph Ph a a m a a a a a © o Q PH a W £ 0^. CB a © © o 43 a © © a © a o © © a m "© S © .2 H © .2 © © o W 43 a THE LUMBWA AND ELGON CAVES 291 1 . Pleistocene Period. — Elgon series in eruption : this is doubtless correct, but what may be called the birth of Elgon will probably be found to date back to late Miocene times. 2. Cretaceous Period. — Kaptian. Referring to Professor Gregory’s ‘ Great Rift Valley,’ p. 230, it will be noted that he places the formation of the lava sheets of the Kapiti Plains as far back as the Cretaceous era, and considers these plateau eruptions to be chronologically coincident with certain denuded volcanic cores which protrude through the Jurassic rocks near the coast north of Vanga at Jombo and Kiruku. He unfortunately does not give the data which led him to the conclusion that the Kapiti lava sheets are as old as he states ; personally I am inclined to essay the opinion that they are not older than the Pliocene era. The insignificant hills of Jombo and Kiruku are, however, I consider, the oldest traces of volcanic activity in British East Africa (excluding, of course, granitic dykes in the old metamorphic series), and quite dis- tinct from that of the high plateaux of the interior. Between these two hills, for instance, lies another hill called Mrima, and this is composed of Triassic sandstones, and an unaltered exposure of these beds was observed close to Kiruku peak, which shows that at that period Kiruku at all events had ceased its active life, and had been eroded down to a mere core. The rocks from both Jombo and Kiruku are of an older type than anything I have seen up country, but I am unable to accurately describe them until a section can be made for microscopic examination. At any rate these old volcanoes should, pending a more detailed survey, be tentatively considered to date back to Jurassic times. The full consideration of these problems has, I fear, brought under review a number of questions which at first sight may have appeared somewhat irrelevant, but if any of the caves are of natural origin the causes which produced them are inextri- cably bound up with the geological history of the Lake Victoria basin. It is to be regretted that fuller information is not available, and I would have then ventured to take up a more dogmatic view. After considering the evidence, and seeing that in Lumbwa the natives of the same race to those living on Elgon are actively excavating material 292 DESCRIPTION OF CAVES from caves in very similar rock, and which they consider to be of great fattening value for their live-stock, the presumption at present is undoubtedly to the effect that the Elgon caves were excavated by man for a similar object. Anyone who realises the size of East Africa and the difficulty and cost of exploration work will readily understand that the full solution of this and many other kindred questions may have to wait until the country is in a position to bear the cost of a properly equipped geological survey. A great deal can, however, be done by the accumulation of data by officers resident in the districts, and by planters and farmers who reside for years in one area, and attain a more detailed knowledge of a particular locality than it is generally possible for a district officer to obtain. I desire to express a grateful acknowledgment of the data supplied by Messrs. Dobbs, Deacon, and Knight-Bruce. DESCRIPTION OF CAVES ON NANDI ESCARPMENT By P. L. Deacon The largest of the caves called Kiptile is situated a short distance from the top of the escarpment beneath the hill on which is a beacon. This hill is the next one to the west of the Gordi Juok, which is the name of a large and prominent rock, easily visible from Kisumu or the Uganda Railway. This cave is situated on a ledge some two or three hundred feet from the summit of the escarpment, and there are huts near by. It must be some 2000 feet or more above the level of the lake. The cave is about 20 yards long and 10 broad, and has a curved roof over 7 feet high in the middle. It is hollowed out from beneath a gigantic boulder or possibly in the middle of the boulder. The boulder gives the impression of having been eaten away to form the curve. The floor is solid stone, and is probably part of the same boulder. It is covered with earth to a depth of a foot. Flakes 1 of rock 1 These turn out to be of a pink granite. ON NANDI ESCAEPMENT 293 chipped off the roof are sent. They are coated with soot, for herdsmen with their cattle still make use of the cave and build fires there. The cave is perfectly straight, and does not bend round. Nothing of interest was found by digging in the floor of the cave. The formation of the cave is somewhat different from any of the others about to be described. The other caverns are apparently originally formed by the falling together of two or more large boulders which have come to rest, leaving a space beneath. This space forms a cave, and has been apparently hollowed out gradually by fresh- water springs which still exist in some caves. The large cave above described, however, is formed by a single boulder, and is not the result GORDtJUOK PEAK of other boulders falling together. From the ledge on which this cave is situated a magnificent view is obtained of the whole of the country bordering on the Kavirondo gulf, as far as Uyoma. Ascending to the top of the escarpment and skirting the beacon hill, one is guided down the valley of Nderum, on the further side of which are two of the most interesting caves of all. Climbing a short steep rise and cutting your way through bush, you approach two large boulders closely adjacent to each other. This place is called Kakamaket, the place of hyaenas : leopards and jackals are said to live here. It is an eerie place, and the entrance to each cave is exceedingly unattractive. Nandi natives do not use these caverns. You can only enter on hands and knees, but after entering you find you can stand up. The first cavity is formed by a large x 294 DESCRIPTION OF CAVES boulder which is held up and supported by several smaller boulders, something after this position : You go round the boulder by a curved passage and find a centre room under the boulder with a stream running through it. This is where a herd of cattle was extracted by the troops during the Nandi expedition, but the herdsmen escaped. There are three or four shafts by which an exit is possible. Emerging from this cave, you approach the other. The en- trance to this is even more uninviting, being a very small hole through which it is just possible to squeeze. As it is also the reputed home of leopards and jackals, hesitation is natural. However, a Nandi moran crawls in first, and you hand him the lamp and follow. You meet nothing worse than a few bats, which brush past your face, startled by the lamp. After crawling a few yards, you find a small chamber on your right in which you can stand up. There are other open spaces inside, and the whole thing resembles an underground dungeon or oubliette. Passing along, you see light ahead, and emerge from the other side of the boulder. At the top of the valley in the hill called Mogon, behind Arap Kossabey’s camp, is another small cave. This, as in the case of all the caverns except the large one, is difficult of approach. It is a long, narrow passage under a big over- hanging rock, and can only be entered on hands and knees. It is about 15 yards long, and is formed apparently by two large boulders having fallen together. There is another typical cavern at the back of Chebanon hill. This does not face the escarpment, but is situated in a valley separated from it by a hill. It is exceedingly difficult of approach, and is, like the others, formed by two large boulders having come together, and leaving a cavity beneath. The cavern inside is curved. You can proceed for about 20 yards, and in the inmost recesses there is a running stream. The ON NANDI ESCARPMENT 295 floor is solid rock covered with soil a foot deep thickly mixed with cattle manure. This’ is a cave in which some Nandi took refuge during the expedition and lived for a month with their cattle. They placed a barricade of small boulders at the entrance, and these can still be seen lying where they were when pulled aside. It is said that the Masai levies, who accompanied the forces, advanced behind a wall of shields. The roof and side of the cave, as in all these caves, is smooth and hard, and it was impossible to obtain a specimen of the stone, and digging in the soil produced no results. In the neighbouring hills many small cavities were observed under overhanging rocks, in many cases with small streams running through them. In general, to the lay mind, these caverns appear to have been caused by a severe earthquake powerful enough to shake together the huge granite boulders, leaving open spaces beneath them. Many of them have been increased by the water action due to streams. The whole country-side is full of these boulders, and presents a rugged appearance as if on the morrow of some huge upheaval. NOTES ON SALT CAVES IN LUMBWA DISTRICT By A. Knight -Bruce There are, it is believed, some nine caves of considerable size in the above district, and their names are as follows : 1. Gitoi. 4. Kibarabara. 7. Sausit. 2. Kibrisin. 5. Kiptoi. 8. Riptoit. 3. Kibipiten. 6. Choma. 9. Bagau. The approximate altitude of the principal caves above sea- level is as follows : Gitoi, 6000 ft. Riptoit, 6900 ft. Kibipiten, 6500 ft. Sausit, 6100 ft. They are used at the present day by the Lumbwa natives, who dig a saline earth in them which they remove in baskets and feed to their cattle and sheep outside the caves. 296 NOTES ON SALT CAVES IN LUMBWA DISTBICT Sketch Map showing Position of Salt Caves, Lnmbwa. Caves underlined. Tracks Scale approximately 13 miles = one inch. The rock forming the walls and roof is very soft, and frequent falls occur. In some of the caves it is considered safer to dig in them by night. In one cave in Sotik a stone bowl of Neolithic type was found. No stone implements or bones have been discovered in any of the caves, and the softness of the walls and the con- tinual digging preclude any chance of finding any prehistoric mural decoration. The rock appears to be a decomposed fine volcanic ash or mud. NOTES ON SALT CAVES IN LUMBWA DISTRICT 297 A description of two caves which were explored is now given : 298 NOTES ON SALT CAVES IN LUMBWA DISTRICT Bagau Salt Cave . This cave is situated near the top of the escarpment which rises from the Nyando River on its left bank between Lumbwa and Fort Ternan stations. Its entrance is under a waterfall on the Sambula stream. It is difficult to arrive at even an approximate idea of its age, though the natives agree that it is not of any great antiquity. Its origin is attributed to game being noticed nibbling at the face of the rock ; probably cattle followed their example, until men, realising that they gained some benefit from the practice, began to dig the stuff and to feed their stock with it. Till some four years ago, the cave was free from rocks falling ; now, owing to the number of fatal accidents which have occurred, digging is confined to the entrance only. The accompanying diagram will probably give a clearer idea of the formation of the cave than a description. The tree near the entrance should be noticed — it is evidently of no great age yet, as evidently since it was of mature growth the action of the waterfall has removed all the soil from its base which was above the level of the roof of the entrance to the cave. The height of the entrance is only 4 feet, but rises to over 8 feet almost at once, and a cave of very irregular shape, of an average width of about 40 feet, is disclosed. This continues for 150 yards, when the floor suddenly rises, and the cave bifurcates into two branches. About 50 yards further along each of these branches the passage was stopped by huge slabs of rocks which had fallen down from above. A considerable fall of rock had occurred the night before I entered, and many more slabs appeared to be on the verge of falling, and as a space of only some 18 inches was left, further progress was impossible. The nature of the walls varies — at the entrance the rock strata are not well defined, and fine-grained white and red ash, called Ngelel and Ngebirir by the Lumbwa, is found in patches. Further on, however, the strata become uniform — there are thin bands of black basalt lying in horizontal lines with Ngebirir between each strata at distances of about a foot. BAGAU SALT CAVE. NOTES ON SALT CAVES IN LUMBWA DISTRICT 299 Another variety of ash called Ngetoi by the Lumbwa is found in patches. The roof is formed of rock lying in strata all the way. Two supports have been left. No water was noticed in this cave. To reach the end is a reputed journey of two to three hours. Kibipiten Salt Cave This cave is situated on the Kiptiget River, about five miles upstream from the bridge over which the old bridle-path to Kericho crosses — and is about 6500 feet above sea-level. Its situation is practically the same as Bagau, but the waterfall is higher. The entrance is also low, but inside the height rises to about 12 feet, with an average width of over 60 feet. After running for 60 yards, this cave also forms into two branches, and the height decreases to about 8 feet ; after following these for about 80 yards I was in each case stopped by deep pools which had been formed in pits formed by digging for salt. No regular strata are found here as in Bagau. The rocks in this cave are more indurated, which is either due to heat or the infiltration of water charged with silica. One of the speci- mens from this cave is a typical agglomerate. I was able to obtain more definite information as to the age of this cave. Elders I questioned stated that within their memory it only entered the rock for a distance of 25 yards — their grandfathers started the excavation when they first came to that district. Now it is supposed to take two hours to reach the end, but it should be remembered that progress is very slow, espe- cially as the native only sees his way by lighting torches of twigs, and most of the way it is necessary to crawl on hands and knees. General Notes . Both the other caves visited (Gitoi and Riptoit) present almost similar features as those described above. (The former has already been described by Mr. C. M. Dobbs in the Journal of the E.A.N.H.S., vol. No. 10, p. 125.) 300 NOTE ON THE FORMER LEVEL From inquiries made, the same remark appears to be true of those which I have been unable to visit. In absence of expert investigation I formed the following conclusions : These caves have, at any rate, been mainly formed by the labour of man, and, though resulting from an industry some- what alien to the Lumbwa character, do not represent such a heavy task as might be imagined. For instance, Kibipiten, say fifty years ago, was only 25 yards long, now it is reported to take two and a half hours to search the end, though I was unable to verify this fact. I should say, however, that this represents a mile at the outside — this therefore means a progress of about 35 yards a year, but personally I should not think it likely that this cave is a mile long, since the task of removing a basket filled with rock this distance would occupy nearly a day. While crawling about in caves like these without any means of measurement one naturally forms totally erroneous ideas of distance — apart from the usual exaggeration of native reports. Again, the fact that these caves appear to be universally suffering from their roofs falling in, in the absence of any marked disturbance, hardly seems to point to their being of any great antiquity. It was most unfortunate that these phenomena coincided with the investigation. NOTE ON THE FORMER LEVEL OF THE VICTORIA NYANZA By Felix Oswald, D.Sc., F.G.S. In compliance with the Editor’s request, I am appending to his article the results of my observations indicating a higher level of the Victoria Nyanza within a recent geological period, together with their effect upon the question of the origin of the caves on Mount Elgon. OF THE VICTORIA NYANZA 801 My personal knowledge is confined to the area south of the Kavirondo Gulf, between Karungu and Kendu ; and the evidence which I was able to collect did not disclose a higher level of the lake in former times than about 800 ft. above the present water-line. At this altitude, viz. approximately 4000 ft., I found quantities of well-rounded pebbles of quartzite and gneiss, especially at Kikongo, near the lake, and farther up the valley of the Kuja River, e.g. on the south side of its left tributary, the Olasi River, between Nangena and Nyaroya; and also along the low ridge above the Ogo Ford over the Kuja, viz. at about 16 miles in a direct line inland from the lake. I have already suggested 1 that this zone of pebbles at the 4000-foot level marks an old beach-line of the Victoria Nyanza, and it is an extremely noticeable feature of the landscape that in the whole of this area the hills below this altitude are gentle and rounded, with particularly broad valleys, whilst abruptly above this level I noticed that the hills are pointed and terminate in rough and rugged crags. The same striking contrast is evident in looking northward across the Kavirondo Gulf from Kendu. It seems to me also to be more than a coincidence that on the western coast of the Victoria Nyanza gravel deposits and caves occur likewise at a height about 300 ft. above the lake in the cliffs of the coast of Buddu, north of the Kagera River.2 This uniformity in level of the old beach-line on both the eastern and western shores would seem to indicate that a similar uniformity persists all round the lake. But the caves on Mount Elgon occur at heights varying from 6100 to 7000 ft., i.e. as much as 3000 ft. above the present level ; whilst I was unable to find evidence of a beach-line higher than 300 ft. I find it difficult therefore to conceive that the waters of the Victoria Nyanza could ever have stood as high as 3000 ft. above the present level so as to be able to erode the caves of Mount Elgon, although I speak with great diffidence since I was not able to visit this mountain. On the other hand, it is equally difficult to understand how this vast volcano could 1 ‘ The Miocene Beds of the Victoria Nyanza, &c.,’ Quart. Journ. Geol. Soc.t vol. lxx., 1914, p. 146. 2 Sir William Garstin, Report on the Upper Nile, Cairo, 1904, pp. 32-39. 302 GAME AND DISEASE have been elevated several thousand feet, for it is more probable that (like other volcanoes) it owes its present altitude (14,200 ft.) merely to the long-continued accumulation of its lavas and tuffs. Evidence of downward sagging of the surrounding strata is indeed likely to be forthcoming, and a local analogous instance of such sagging is furnished a little farther south by the northerly tilting of the Miocene beds in the Karungu district towards the centre of the volcanic mass of Gwasi (6384 ft.), owing to the enormous weight of its successive lava-flows and tuffs, which are revealed so graphically in the lofty cliff-sections between the mouth of the Kavirondo Gulf and Karungu. The fact that the caves of Mount Elgon occur at varying altitudes, and not at the same level, would seem also to militate against their having been excavated by wave- action of the Victoria Nyanza ; and I should feel inclined to ascribe the formation of these caves rather to springs arising from water percolating through the friable tuffs in which the caves occur, and issuing at the junction with an underlying impervious flow of lava. It is noteworthy that Mr. Hobley mentions the dripping of a spring from the roof of one of the Elgon caves, and he also states that in the case of the caves of the Nandi Escarpment small streams still occur in some of them. It seems to me, therefore, not unreasonable to suppose that both the Elgon and Nandi caves may owe their origin to the erosion of underground water at a period when the annual rainfall was heavier than at the present day and before the progressive desiccation had set in to which Mr. Hobley has called attention in the pages of this Journal (No. 9, p. 4). GAME AND DISEASE By A. Blayney Percival When talking from personal observation of disease amongst game, one must of necessity be vague : for it is at the best but conjecture. One cannot, as with domestic stock, use a GAME AND DISEASE 803 thermometer ; nor can one take a blood smear, unless one kills the animal to do so. Under these circumstances there are certain rules of nature that must be taken into consideration, and the first and most important point, to which I wish to draw attention, is the way nature takes her own steps to prevent the spread of disease, yet at the same time using disease as one of her methods of providing that none but the ‘ fit * shall increase in numbers and stock the land. Nature is a cruel mistress and has no use for weaklings, nor has she safeguards for them. In a true state of nature the meat-eating animals are all the time at work taking their toll from the herds of game, and killing the sick or weak animal which gives them the easiest chance. It is often put forward as an argument against game reserves, that, if game is protected, it will increase until it is so numerous that disease will break out. This is true if one is strictly preserving — as one preserves pheasants at home, and where all the meat-eaters are killed off ; but in a reserve, where nature is allowed full swray, there is little danger, for a sick animal is not allowed to live many hours. A point which I wish to make very clear is this : ‘ So long as the balance of nature is kept, there is little or no disease.’ Should an animal be a weakling, nature sees that it does not live long : it may die of some sickness, or in all probability it is killed by the ‘ meat-eaters,’ but it will not propagate the species and so carry on a weak strain. An odd sick animal is one of the rarest sights, except when a new disease arises, against which there is no immunity, or when the animals are poor, and therefore more liable to a disease like anthrax, which is largely disseminated by dust. I can say that in twenty odd years amongst the game in South Africa and British East Africa, I have not seen more than a few sick animals : either those that had been wounded or when an epidemic of some disease was at work. It must be understood that disease is seldom heard of, unless the animals die in such large numbers that the hyaenas and other scavengers are unable to dispose of them quickly, or in a district which is closely under observation — as Nairobi was during the recent outbreak of anthrax. Here the ten or 804 GAME AND DISEASE twelve animals, which during the height of the outbreak died in one night, all vanished but a few broken bones within a few hours of daylight. The instinct of a sick (or wounded) animal is to leave the herd and seek solitude ; it is therefore deprived of all assis- tance in the way of warning, and, in some cases, of protection. The case goes even farther, for it is the instinct of the herd to drive out the sick beast. Such an animal — alone, sick, and dull — naturally falls an easier prey to the carnivora ; for not only are the real killers — such as lion and leopard — a danger, but the scavengers — such as hyaenas, jackal, and the like — will readily pull down a sick animal. For instance, in the case of wounded animals, one knows the large numbers that are hit and get away ; yet how seldom a wounded beast can be found the next day ! Often when having wounded an animal, knowing that if it is followed up at once it will go for miles, I have sat down to give it time to stiffen or to die. Then while waiting, I have either seen with my glasses a jackal pick up the blood-spoor and follow it up, or else found the jackal near the animal when it was reached. Should, however, that animal get away and live till nightfall, then every hyaena, jackal, hunting-dog — to say nothing of lions — crossing the blood, will at once follow up ; and though jackal and hyaena will seldom chase or attack a hale animal, they will not hesitate to pull down a wounded one, or even a sick one. In a real game reserve the sick animals are killed before they have time to die, for there is little doubt that the scent left by a sick beast would be at once recognised by such animals as live on the dead and dying. Vultures are possibly the quickest to spot the sick animal, and alight near to wait for its death, thus acting as guides to the four-footed scavengers, who all make for the spot where they see the vultures dropping. Taking this into considera- tion, it can be easily seen that the only occasions on which disease in game is recognised is when an epidemic breaks out. The gathering of the scavengers — * those who come for the broken meat ’ — is a most wonderful sight. If an animal dies in the open, it is only a case of minutes GAME AND DISEASE 805 before the first vulture appears. A few minutes later the air is full of them, and then the four-footed scavengers begin to appear — the jackals first, usually in pairs. In an out-of-the- way district, or game reserve, where the sportsman is rare, they come openly trotting up, showing little fear ; and if the vultures are not too numerous, they will at once attack the carcass. If the vultures are in large numbers, they will not venture near, but will sit and await their chance of a titbit, or wait until only the bones are left, and then they will take what remains. Possibly an old hyaena, very hungry, will be tempted to move in daylight, but his slinking approach is very different from that of the jackal : he seems to feel that he is a creature of the night, for he comes along, taking all the cover he can, till he sees the feast, when he will hurry up to it, and, driving off the vultures, set to work ; even then he is very nervous, and constantly looks round for a possible enemy. Should he manage to get a large piece of meat, he will even retire to the nearest cover with it ; one can see it is against his instincts to be abroad in the daytime. The quick arrival of the vultures on the scene is explained by the way in which they quarter the skies so high that the human eye cannot see them ; yet with their No. 12 Zeiss-power eyesight they can see all that goes on over a big area. The sick or wounded buck is seen by one vulture, and he drops towards it. From north, south, east, and west, other birds sail in towards the same point, and so the circle spreads until hundreds of birds are at the kill. The collection of hyaenas at the feast must work out something on the same lines. Every hyaena has its own regular round, which is followed night after night, and, as he travels, his howl notifies his neighbours as to his movements. Should he find a kill his howls cease, and I have no doubt that this is quite enough to bid his friends to the feast. This is of course if a kill is fresh. If a kill is a day or two old, which only happens when a beast has died in thick bush, and out of sight of the vultures, it will of its own accord notify everything down wind for a big distance ; or the track of a sick or wounded beast will attract every meat- eater that crosses it. 806 GAME AND DISEASE The carcass of a rhino or elephant will mean good feeding for days, and this is the time to see hyaenas. When an epidemic is at work, all the fisi for miles congregate for the broken meat. When the anthrax was bad outside Nairobi, in this year, there were many of them about, and I frequently saw twenty or more in a day’s ride, and on one occasion saw in broad daylight twelve fisi by one kongoni carcass. During the war in East Africa, a point in favour of the hyaena was the fact that he greatly helped to keep the camps healthy. In an ordinary country the fact that ten or twenty animals a day were dying, would have meant endless fatigues to bury or burn ; but here it only meant putting the carcass out a hundred yards or so, and the hyaenas saw to the burial. Kipling remarked, with reference to the South African Cam- paign, ‘ One horse will move a camp, if it has been dead long enough.’ British East Africa has seldom had such experience, though, on the coast, it nearly happened, only it was not a horse but an overlooked Arab. I am told that during the movements of our mounted columns the lions followed close to get the abandoned horses, and in many cases became very troublesome. As shooting was forbidden, the lions became bold, and often came up to the outskirts of the camp. Whilst talking about scavengers, it is interesting to note that Africa is wonderfully free from blow-flies and scavenger beetles. We recently had a visit from a Mr. Le Soeuf, from the Sydney Zoo, who was particularly interested in our scavengers, as in Australia they suffer from a veritable plague of blow-flies. Here, a carcass is disposed of in a few hours, or even minutes ; for where vultures are thick, it does not take more than a few minutes to dispose of all the soft parts. On the other hand, Australia has practically no carnivora, and the carcass of a sheep lies till the blow-flies have laid their eggs, and so increase their breed. When in 1905 the game died in great numbers on the Athi, the carcasses were fly-blown, with the extraordinary result that the upper side of the hide, exposed to the sun, became hard and dry, whilst everything inside was eaten away. One therefore found a dry skin, covering a complete skeleton, GAME AND DISEASE 807 covered with the cast pupa-cases of the blow-flies ; yet even then I do not remember that they seemed very thick. It was not necessary to carefully cover one’s blankets during the day to keep them from getting fly-blown, as is the case in both Australia and New Zealand. In the southern game reserve, where I do not think it is any exaggeration to say that thirty thousand head of game — chiefly hartebeest and wildebeeste — were killed as meat for troops during military operations, we might have had a serious upset of nature, and the lions and other meat-eaters, defeated by man in the eternal struggle to keep alive, might have produced a possible reign of man-eaters and cattle- killers amongst the more or less defenceless natives. To a very small extent this has happened ; but the zebra, which did not interest the man shooting meat for the camp, remained to feed the lion, while the deaths amongst the horses and cattle fed and drew the carrion feeders. Hundreds of hyaenas followed the columns, and added to the horrors of a bush war : it was 4 Heaven help ’ the wounded man who was not found by dark. As one who went through some of this fighting, I can say that, had I been left wounded, it would not have been long after dark before I fired my last shot, and at such close quarters that there would have been no possibility of a miss. Since the military operations passed southward and left the game reserve, there have been more complaints from the Masai of cattle-killers amongst the lion than I have ever heard before. However, these gentry are being dealt with, and I hope that we will soon have things quite quiet again. It is always recognised that where game is plentiful the lions seldom attack man or his cattle. In districts where at times game is plentiful, coming from another district, returning lions are often bad. They follow and feed on the game, settling down until, later, they suddenly find that the game has gone. As they are unable to get food, they become man-eaters, or elso rob the stock kraals. Practically all our man-eating lions of any fame have been in such districts, but there are of course exceptions to this. 808 GAME AND DISEASE Ngar Narua The commonest disease known amongst the game, par- ticularly on the Athi Plains and the southern game reserve, is ngar narua. For fifteen years I have known the disease, but it is only wdthin the last year that it has definitely been proved to be anthrax. Having something to go on, I have made careful inquiries amongst the Masai — who are without question extremely clever on the subject of stock diseases — and from them I learn that two forms of ngar narua are recognised : mbarua, the bad form, which is the more general, and a milder form, likened to plague — in that it attacks the glands of the leg. This I take to be black quarter. In the case of mbarua, the Masai recognise that the disease is one which man may contract, and treat it accordingly. They say that contact with the meat, blood, or skin, causes sores, and, if treatment is not given to these, they often result in death. The treatment given by the Masai is to sweat the patient thoroughly, by wrapping him in the freshly stripped, warm hide of an animal — bullock or sheep, according to his size — depriving him of all milk or water, but giving him a decoction of some root. The milder form is not looked upon as dangerous to man, and the meat is considered eatable, except the affected parts. The disease (which is recognised as one which chiefly affects young stock) can in most cases be cured by copious bleeding, and cauterisation of the affected glands. This year we had an outbreak of ngar narua amongst the kongoni, near Nairobi, which was proved by specimens and swabs to be anthrax. In this case we have a clear history of an outbreak of the disease amongst the cattle at the camp of the King’s African Rifles, at Mbagathi, which was followed a week or ten days later by deaths amongst the game in the neighbourhood. Upon finding, from the specimens which I submitted to the bacteriologist, that this was anthrax, I at once put men on to carefully watch the spread of the disease. This was by no means difficult, as by watching the vultures a dead animal could be at once located. Much to my surprise the disease did not spread, but was confined to an area some six miles by GAME AND DISEASE 809 five, and here in two months about three hundred kongoni died. Other game was plentiful, but no other animal was reported as dying in the area. I kept men on the area for a month after the disease had apparently stopped, and also sent others with prepared swabs to work right through the reserve, but with negative result. The Masai in my employ insist that this was the same disease which every few years breaks out in one district or another amongst the game ; but it is interesting to note that this outbreak was during the wet season, while ‘ ngar narua ’ is usually met with in the dry season, or, to quote the Masai, * when the grass is white ’ (dry and bleached by the sun). I think in 1905 there was the worst outbreak of ngar narua that I can remember, the game dying on the Athi Plains till they lay so thick that hyaenas and vultures could not deal with them. The mortality was worst near Eldonyo Sabuk and Lucania Hills, where thousands died. These districts have now been settled for many years without any further outbreak of anthrax — a disease which, according to all authorities, infects for years the soil where an anthrax carcass has lain. Ol Ki Piri During the present outbreak of pleuro, no report of the disease amongst the game has been received. I have had men down in the districts where this disease exists, and they report that eland and buffalo are feeding over the same ground as the affected cattle, without any apparent ill-effect. Pleuro, so far as my limited knowledge goes, is a disease the transmission of which needs immediate contact between a sick beast and a healthy one, and therefore is less likely to be passed from cattle to the game. Even if passed, it may not spread, owing to the absolute ‘ law ’ amongst wild * animals that a sick beast at once leaves the herd, the solitary sick animal having but small chance of living for many hours. Nature takes her own steps to prevent the spread of disease. The old Masai tell me, however, that when the first bad outbreak of pleuro took place — which was some six years before Y 810 GAME AND DISEASE the rinderpest struck the country — buffalo and eland died, but not in large numbers. The Masai have a legend that ol ki piri (pleuro) was intro- duced by a bull which arose from the waters of Naivasha or Elmenteita, and proceeded to cover the female stock in a herd of cattle grazing near by. The Moru who was in charge of the herd speared the bull, but to his surprise no blood came, but a fluid like the discharge from an unhealthy wound. The animal, however, died from the wound, but the herd developed ol ki piri, and from them it spread all over the Masai country. Other stories say it came from the west, and that the El Burgu first suffered from it. Rinderpest This is the worst disease that has ever struck Africa. It swept from Abyssinia to South Africa in about six years, killing practically all the cattle and decimating the game. So far as I have been able to make out from the sources of information available, the following appears to be the history of this dreaded disease : Somewhere in the eighties, the cattle- plague, as it was then called, reached both Abyssinia and Egypt — the former from India, the latter from the Black Sea. For some years there does not seem to have been any notice- able spread of the disease, but in 1891 it appeared in British East Africa. I have been making most careful inquiries, and the general view amongst the Masai is that the first out- breaks in Masailand were amongst the Loi-tok-i-tok, on the slopes of Mt. Kilimanjaro. Here it had been brought by Masai raiders, who visited the coast about the end of 1890, and brought back a quantity of looted cattle. These cattle developed the dread pest, and in a short time the Loi-tok-i-tok had no cattle left. They, being without stock, came to the Malabalo, who lived on the Athi, and with them made a joint raid upon the Wakamba, to find that the Wakamba cattle were dying. However, they brought some looted stock back with them, only to start disease amongst their own cattle, which so far had been clean. An old Mkamba told me they contracted it from cattle brought from near Basso, or Lake Rudolph : this is a more probable story. At the GAME AND DISEASE 311 same time, I have heard stories to the effect that rinderpest came from the Wenyemwezi cattle which were looted from the country to the south-east of Lake Victoria. It is possible, therefore, that a second source of infection was from the west — probably from Egypt. There are some points difficult to follow in the way the disease spread. Coming from the north, it seems first to have got into the Wakamba cattle, to have spread from them through the neighbouring tribes, and to have again worked north-west to Laikipia and the country to the west. A few isolated spots — such as Mt. Marsabit and Mt. Nyiro in the northern frontier district, also some parts of the coast, seem to have escaped ; but although these places were few and far between, still they helped to re-stock the country. It is interesting to note that, in 1893, Arthur Neumann, writing of the country near Lake Rudolph, speaks of rinderpest as a thing of the past ; while Swayne, writing of Somaliland, about the same time, speaks of the numbers of buffalo and other game, and never mentions the disease. By this it appears as if it did not reach the east of Somaliland till after 1893. Travelling south, from Lake Victoria and Kilimanjaro, it spread rapidly along the eastern side of Tanganyika ; but Nyassaland does not at this time seem to have suffered. It apparently passed between Lakes Tanganyika and Nyassa into the Angoni country, and from there south, across the Zambezi, appearing north of Bulawayo in October 1895. By March 1896 the Rhodesian Government had given up all hopes of doing anything, and in October or November 1896 it was in the Eastern Transvaal, where I personally met it, and saw the greater kudu wiped out. The same year it reached Cape Colony : the pace at which it spread being extraordinary, going, in one year, from the Zambezi to the Cape — a distance of 1000 miles. In Rhodesia, it had great assistance from the transport roads — all the transport at that time being by waggon and oxen. This, however, was quickly stopped, but without result ; and in March 1896, all hope of checking it having been given up, the restrictions were removed in order to allow as much food as possible to be brought up before all the cattle died. When , in 1 896, the Matabele Rebellion took place, the fact that waggon transport had to all intents and purposes finished, badly hampered the troops. 312 GAME AND DISEASE It has been estimated that in Rhodesia alone 100,000 cattle died ; and that to the south, Khama and his people lost 800,000 head. What the total for Africa must have been is in- conceivable ; yet the wonderful recovery shows what a country it is for stock-rearing. That great hunter and observer, the late F. C. Selous, states that in Rhodesia there was an extraordinary absence of vultures, and he attributes it to their having died of a surfeit of rinderpest meat in the early stages of the disease. This, however, does not appear to have been the case in this country. Rinderpest in British East Africa. My chief informants on the subject are the Masai and other stock-owning people, who are naturally more observant than the agricultural tribes. When rinderpest first reached the country, only the cattle suffered ; but it was not long before the disease was apparent amongst the then huge herds of buffalo. Eland also showed it early ; and subsequently all the game, except the gazelle, zebra, elephant, rhino, and hippo, became infected. Here an interesting point is to be noted, for the wildebeeste was said to be the last animal to become infected, none dying until after all the cattle were dead. Giraffe, always subject to disease, died in large numbers. Those species which suffered worst were buffalo, eland, greater kudu, roan, lesser kudu, and bush- buck ; of these, all but the greater kudu have made practically complete recovery. The buffalo, perhaps, showed their losses more than any- thing else, for the whole of the animals inhabiting the open country vanished : either they were dead, or had taken to the bush, where they spread the infection to the bush- buffalo. These forest- and bush-loving buffalo were seldom in such large herds as the animals of the open, and some of these herds appeared to escape, for, when I first began to investigate the game of this country in 1901, I found small isolated herds here and there. During a recent small outbreak of rinderpest amongst the buffalo, I have, on visiting the scene of the outbreak, been GAME AND DISEASE 318 alarmed to find that there were no living buffalo, where only a month or so previously there were numbers. However, I was reassured by the Wandorobo, who informed me that should an outbreak occur, it is the invariable custom of the buffalo to move at once. They dive deeper and deeper into the forest, away from the other buffalo, who in an ordinary way haunt the edges and open glades of the forest. This habit may have helped the forest-buffalo to escape the fate of his brothers in the open. When I first visited the Athi Plains, sixteen years ago, the remnants of the buffalo and wildebeeste heads could still be seen, only the boss of the horns and part of the skull remaining. They lay mostly along the river banks ; for, as usual, a sick beast goes near the water to die. At this time there were not more than forty buffalo on Eldonyo Sabuk, and rather less in the Kamiti swamp ; now, in spite of shooting them, there must be nearer 400 in the former and 200 in the latter haunt, although many have been shot there. For many years there seemed to be little or no increase in the eland and buffalo ; but suddenly the change came, and the increase was rapid. It always happens in nature that, should any species get really low in numbers — as in South Africa and in America — then the strictest preservation will only, at the best, enable the animal to increase slowly until it reaches a certain stage. After that, better results may be expected. I only wish I could give as good a report on the greater kudu as on eland and buffalo ; but the kudu seems still in the preliminary stage of recovery. They have increased somewhat, but very slowly, so that one almost despairs. They undoubtedly have more enemies than buffalo or eland, the hunting-dog being one of the worst. The kudu, however, is in comparison with a horse a wonderful performer amongst the rocky foot-hills, which he usually inhabits, but is really easily pulled down by such a remarkable tracker as the hunting-dog. The recovery of the eland from the rinderpest is wonderful. Sixteen — and even ten — years ago, eland were extraordinarily scarce. To-day there are hundreds on the Athi ; and from the Rift Valley and the Uasin Gishu Plateau so many complaints 314 GAME AND DISEASE have been received of damage done by eland, that the special protection given to them in these areas has been taken away. Since, however, they have increased just as much or more in the outside districts, I have no fear of any ill effects : in fact, I consider that eland and buffalo have now reached a stage where, under fair sporting shooting, there should be no danger of their being much reduced in numbers. Another animal which suffered, and was slow in its recovery, was the bongo. This animal is always looked upon as rare, but is really common, and is found in quite large herds in its own particular haunts — the high dense forests. To-day the bongo is no easier to shoot than he was ; but he is fifty times more common than he was fifteen to twenty years ago. The other game never suffered to an extent sufficient to throw them back more than a year or two. Wild pigs of all sorts suffer from rinderpest. In South Africa, I saw my first bush-pig, dead — not one, but many ; and in British East Africa the outbreaks amongst warthog have been numerous. In fact, it is almost certain that if rinderpest breaks out, the -warthog will get it — even when there are no records of the disease amongst other game — and I look upon them as a distinct cause in spreading the disease. Without doubt the game, and also the local cattle, have developed to some extent an immunity against the disease ; yet the warthog does not appear to have done so. If game were still as susceptible to rinderpest as they were at one time, the disease would not be in isolated areas as it is to-day, but would, after the first outbreak, in a few days, be all over the country. This would also apply where double inoculation was being practised ; yet I have not been able to hear of the spread of the disease from any such centre. On the point of the distribution of rinderpest, I personally think that to-day the tick-bird, as well as the vulture, is an obvious danger ; yet it must be admitted that Mr. F. C. Selous’s statement that the vultures vanished during the outbreak in Rhodesia, and the fact that the disease travelled as fast in South Africa, where vultures had been practically exter- minated. shows that they are not the chief carriers. NOTES ON SNAKES IN EAST AFRICA 315 Jackal. — Hardly game in the full meaning of the word, but still a game little beast and, so, worthy of a note. The first outbreak of any disease that I noticed amongst the jackal was in 1906, when distemper broke out, and swept the whole country east of the Kikuyu Escarpment. The jackal died by the hundred and soon became a rare animal ; while even the ‘ pie-dogs ’ of Nairobi suffered, and most of them died. The old warriors of all ages went as quickly as the pups. This was, I think, the first outbreak of distemper in British East Africa. I could get no proof that the hunting-dogs or hyenas died, but for some few years they certainly appeared to be much scarcer. A year or two later the disease spread into the Rift Valley, where again the jackal died and the Wandorobo lost all their little hunting- dogs. Since then several outbreaks of disease have occurred — the latest being in Laikipia, where the jackal seemed to die off. In this case no signs of distemper were seen amongst dogs, nor did any of the jackal show signs of rabies — a disease which apparently affects them. For, every few years, one hears of the jackal either attacking man in a strange way, or inviting their own death by wandering into farms unconcernedly in broad daylight. Though I know of several cases of natives and dogs having been bitten, I have never heard of any further development. NOTES ON SNAKES IN EAST AFRICA By A. Lqveridge Pleasant memories of chasing grass-snakes along English hedgerows, of smooth snakes on the Dorset heaths, or adders midst bracken and gorse of Welsh mountains are not effaced by three years’ gadding about in East Africa with its greater variety and more dangerous serpent-life. Indeed, one is prone to think that the former provided better sport. At any rate one could pounce more whole- heartedly upon a grass-snake ; probably, however, it is only 816 NOTES ON SNAKES IN EAST AFRICA the associations that show it in this rosy light. It is interesting to note that all the three genera of British snakes ( Tropidonotus , Coronella, Viper a) are represented out here by two species of each. The two typical vipers are both very local, one indeed having only been discovered in the Aberdare Mountains, British East Africa, during the last decade. The distribution of the other is confined to Delgado Bay as far as German East Africa is concerned. The writer has not been fortunate to visit either of these places. The commonest viper is undoubtedly the Puff Adder (Bitis arietans). In length it is usually about 89 inches, but in girth it is as thick as a man’s arm ; a female in young is a still more unwieldy creature, and in view of their prolific nature it is strange that the creature is not more commonly met with. One such specimen which I chloroformed to death at Morogoro, and afterwards dissected, had 16 young in the left oviduct and 18 in the right, making a total of 84 young averaging 7J inches in length (November 16, 1917). This was quite eclipsed by another puff adder killed on November 28, which had no less than 71 eggs in the oviducts. Another female, caught on the western slopes of Mount Kenia in November 1915, had 12 and 18 eggs in the respective oviducts. The snake-catcher can hold in contempt this snake, with its enormous fangs often J inch long. It is so loath to move that it is easy to capture with a forked stick ; this same slug- gishness also constitutes a danger, as it is overfond of lying immovable across paths or in little clearings. A fine male the writer met with a fortnight ago did not even move when its neck was pressed to the ground with the forked stick, and remained quiet until picked up, when it commenced to squirm and struggle. This same snake on being placed in a wire- netted cage escaped about 4 p.m. and was killed two hours later 500 yards further up the hill. Another closely-related creature is the handsome Nose- horned Yiper ( Bitis nasicornis), which owes its name to a pair of horn-like points formed by erected scales. One would naturally suppose from the carpet-like, multi-coloured skins of these big snakes that they must be very conspicuous objects NOTES ON SNAKES IN EAST AFRICA 817 in their native haunts. It is far otherwise, however, for on one occasion, whilst cleaning out the case in which some were confined, three of these creatures were released on a piece of ground covered with short grass. The two larger ones measuring 8 feet 8 inches were secured readily enough, but three of us searched for nearly ten minutes before the writer discovered that he was standing right over the missing snake, which was twisted in and out of the grass and lying motion- less right between his shoes. All the four species of Night Adders ( Causus ) are found in East Africa ; they are much the same size as an English adder, but differ somewhat in habits, as they become most active towards dusk. The Rhombic Night Adder ((7. rhombeatus) is extremely common in Nairobi, and when disturbed crossing the road, as is often the case, it coils itself up and displays a vicious disposition ; if unmolested it sails off with the neck flattened out and the head raised about five inches from the ground. Both in a wild state and in captivity it feeds well on young Square-marked Toads ( Bujo regularis) and mice. The day-time is spent in rubbish-heaps, piles of old stones, or among the litter of outbuildings, whither it has gone in search of its prey. A specimen of Causus resimus was killed on Makindu platform (Uganda Railway) just as our troop-train came in. It is uncommon in the Mombasa coastal region and about Mount Elgon, from which places the writer has examined many specimens. The Snouted Night Adder ( C . de Filippi) seems to be uncommon, though no less than ten specimens were found preserved in a German house near Morogoro ; the writer has only captured two examples during twelve months’ residence in the neighbourhood. The still scarcer Lichten- stein’s Night Adder ( C . Lichtensteinii) was found at Yala river, British East Africa, but the writer has not as yet had any living specimens in captivity. A very interesting group of vipers are the tree-haunting species ( Athens ) ; if it were not for their triangular heads and waspish ways one would never suppose these active and brilliant-green snakes to be vipers at all. The writer kept a specimen of Athens squamiger in captivity for some months, 318 NOTES ON SNAKES IN EAST AFRICA but it did not feed, possibly on account of its quarters being somewhat confined. A still more specialised and unviperish-looking genus is found in the burrowing vipers (Atractaspis) , of which there are no less than seven species in East Africa. Owing to their subterranean habits they are rarely met with, and the only living examples that came into the writer’s possession were taken near his tent at Morogoro ; one (A. rostrata) while digging a pit, another under a mass of earth and stones con- stituting the remains of a demolished building, and a third w*as seen by one of the natives near a log ; it wriggled down a termite hole about four feet from where it had been lying. On digging down it was found about a foot below the surface and squirmed convulsively like a Typhlops or worm when disturbed. When travelling over the ground its snout was held downwards in an unusual manner. It has been said that these snakes are probably not dangerous to man, as their fangs are so enormously developed that it seemed impossible to erect them in its small mouth. This one struck out viciously, but instead of opening the lower jaw, the fangs came down on either side and a little to the front of it ; a quantity of pale straw-coloured venom was discharged on to the forceps with which the writer was holding it. The next group of venomous snakes constitute a division of the family Colubridae known as the Proteroglypha, and contain some of the most dangerous species. Unlike the vipers which have movable poison fangs, the Proteroglypha are characterised by fixed grooved fangs set well forward in the upper jaws. The viperine tooth is hollow like the needle of a hypodermic syringe ; the colubrine tooth is grooved on the anterior surface. Should one of these snakes under con- sideration bite a man through the wrappings of a puttee or other garment a considerable amount of the venom will be absorbed by the doth, enhancing his prospects of recovery. The Mambas ( Bendraspis ) belong to this group and are notorious for the fact that their venom is more toxic than that of any other African snake ; moreover, the Mambas are the only African snakes that will attack man without provocation, NOTES ON SNAKES IN EAST AFRICA 319 but it must by no means be assumed that they will always do so, for generally they attempt to escape. In one case, of which the writer heard recently, a European entered a banda and, finding it unoccupied, was turning to leave, when a mamba struck him in the back just above the kidneys. Though preventive measures were undertaken im- mediately he died twenty-four hours later. A native, whilst bringing a tea-tray on to the verandah where his master and a friend were sitting, trod on a mamba which was lying on the mat ; dropping the tray he attempted to reach his master, but fell and expired within the minute. My informant (who was the visitor) vouched for this. A comrade in the E.A.M.R. recounted how on one occasion he was cycling down a hill when a mamba darted across the road. He swerved to avoid it, and then pedalled for his life, but, overtaking him, the snake struck him on the leg ; after spending many weeks in Bulawayo Hospital he recovered. A native who has caught many snakes for the writer got bitten by a 5-foot mamba. The bite was given on the back of the left hand in the fleshy portion between the base of the forefinger and thumb. A week after this occurred there only remained an ugly sore, but Mushairi stated he had been very sick for several days. On January 31 near Morogoro station the writer captured a 5-foot mamba. The day before this occurred, whilst standing under a mango tree watching a carpenter-bee, his attention was suddenly attracted by a movement amongst the foliage of a branch about a yard from his head, and not more than 8 feet from the ground. A snake was slipping quietly away, when his boy spotting it gave a yell which caused it to quicken its pace, climbing straight up through the dense foliage. Throwing up the only available thing — his net — it tumbled the snake down 4 feet, but before he could pick up the net it had made up the lost ground and was well out of reach. All this time he thought it was the less dangerous Boomslang. The only ‘ spot ’ difference between these two tree-haunting reptiles is the larger eye and more oblique scales of the Boom- slang ; as onty the body of this mamba could be seen it was difficult to be certain of its identity. 820 NOTES ON SNAKES IN EAST AFRICA Returning to the mango tree the following day, and cautiously approaching the bough on which it had been lying, it was easily seen by reason of its having selected some darker foliage against which its bright green coils were conspicuous. At the first attempt the snake was dislodged to a lower branch, and at the second hooked out of the tree on to the ground. It immediately started off, but was overtaken and captured. Half a mile away on another tree no les3 than four young mambas were seen, evidently only just born, for the ventral scutes in the umbilical region had not as yet healed up. Three of these were captured and might easily be mistaken for one of the six species of Green Snakes ( Chloropliis ) which are found in East Africa, or for the Spotted Wood Snake (Pholothamnus semivariegatus) which is common enough about Morogoro. They were very fierce, and attempted to use their fangs. For nearly six months the writer had two fine female mambas in captivity ; these measured 8 feet 1 J inches, and 7 feet 8-J inches respectively. They were very active, and when at first confined struck at the glass when anyone approached their cage ; possibly realising the futility of this procedure they abandoned it, but continued for a couple of months to draw themselves up and threaten the observer with open jaws. In time even this menace was given up, but to the last they continued to follow every movement of the onlooker with their bright eyes. They fed readily enough, and it was rather a difficulty to keep up the supply of mice ; for though they killed large rats for food they refused to swallow them except on a few unique occasions. Several times they ate dead mice, and once one of them seized upon a dead weaver bird which was put into the cage. Finding itself watched, however, it dropped the bird, and during the period of its captivity never touched another. On December 7 the smaller mamba killed and swallowed two very large rats, one at 8 a.m., the other at 11 a.m. ; owing to carelessness the case containing these reptiles was left in the sun from 4 p.m. till 5.80 p.m., with the result that both creatures died from the heat. On opening the stomach of the one that had recently fed, it was found that the flesh and fur NOTES ON SNAKES IN EAST AFKICA 321 had already been digested from the skull of the rat which had been swallowed eight hours previously. During their six months’ captivity they only once cast their skins and appeared in fresh clean olive-green coats. All the young mambas under 6 feet which the writer has kept have been of a bright leaf-green colour, those over 6 feet of an olive tone. A reasonable theory is that as they increase in size and give up their arboreal habits their colour becomes black ; certain it is that there is no detectable specific difference between the so-called Green and Black Mamba. The species just referred to is the common South African Mamba ( Dendraspis angusticeps), which ranges over Central and East Africa, having been taken at Tanga and Mombasa ; two other species are known in East Africa, one of which is doubtful, since it rests upon a single specimen collected in the Kilimanjaro region and called Dendraspis Sjostedi. Jameson’s Mamba ( Dendraspis Jamesoni), on the other hand, is as common as the South African species ranging over the whole of Tropical Africa and Angola. The writer has examined a number of specimens which were shot in trees in the Yala river district. It is a matter of surprise to many persons to learn that there are cobras in East Africa, and in many places — Nairobi in particular — the common Black-necked Cobra ( Naia nigri- collis) is erroneously called the Black Mamba. Whereas mambas have only a very limited power of flattening their necks, cobras when startled or enraged will spread a hood two to three times the normal diameter of the neck. The colour of the Black-necked Cobra at Nairobi is frequently olive- green or brown, with yellow bars on the throat; sometimes, however, as in all the Morogoro specimens yet seen, it is a lead-colour or slaty-black, in which case the throat is barred white or pink. At Longido, however, a beautiful salmon-pink is the normal colour of this snake with or without black spots. Anyone unacquainted with the variable colour- ation of reptiles could be forgiven for supposing it to be an entirely different species. On one occasion at Longido a corporal asked the writer if he could tell him the name of a terra-cotta snake. He said ‘ It is harmless, I think, because when I found it under 822 NOTES ON SNAKES IN EAST AFRICA my pillow I jabbed a jack-knife into it, and it glided into the wall without attempting to bite.’ He and another corporal had made themselves a bivouac by fixing two ground-sheets against the loose-built wall. Nearest the wall this improvised tent was not more than four feet from the ground, and to enter one had to go on hands and knees. Riding into camp about 11 a.m. the writer was hailed by the other corporal who was lying reading, but called out casually 4 The snake is here lying on the blankets now if you want it,’ and he rolled back to continue his reading. Picking up a short stick the writer asked him to vacate so that he might come in. On hands and knees he waited on the one bed, whilst the pillow was overturned revealing nothing. With his face to the ground he was able to see the coils of a pink cobra in the interstices of the wall. Cautioning the others as to the snake’s propensities for spitting, he pinned the nearest visible piece of snake down with a stick, placing his foot on the other end of the stick, and at the same time rising to a stooping posture. Almost simultaneously both ends of the snake appeared, the anterior end chawing at his ammunition boots in a very thorough manner. Seeing it was so occupied, he seized the tail which was coiling and squirming, and releasing the stick pulled the cobra off his boot, tossed it out into the open, and before it had recovered from its surprise, pinned it down by the neck. With a thread of cotton wound round its jaws it sailed about the horse-lines, swaying to and fro with spread hood to the interest of all observers. A specimen alive at the present time was caught three months ago within ten feet of the writer’s tent. It was wriggling along in the ditch at the time, and owing to the hardness of the ground, when pinned by the neck, it turned over and spat at its would-be captor. Fortunately he turned his cheek to the enemy at the critical moment and received the charge of venom on his neck and arm from wrist to shoulder. Unless the venom chances to enter a cut or abrasion, such as anyone might have after shaving, there is no danger. The exception to this of course is the eyes, and it is always at the eyes the cobra ejects its venom, with remarkable precision. NOTES ON SNAKES IN EAST AFRICA 828 The immediate result is intense pain and blindness caused by the superficial capillaries absorbing the venom. This conjunctivitis subsides in a few days if remedial measures are taken, such as bathing the eyes in a weak solution of permanganate of potash, boric powder, or milk. One settler, who promised to collect snakes for the writer, disturbed a cobra in the rickyard one day, the reptile taking refuge behind a disused door leaning against a shed. As it disappeared, my friend struck at it with a panga, cutting off a portion of its tail. He then foolishly looked behind the door, with the result that for a moment he saw the cobra with spread hood facing him at close quarters. His brother led him back to the house and then unwittingly bathed his eyes in a solution of formalin, which he admitted to me was as painful as the venom itself. Nevertheless in about three days he recovered. Another friend was crawling through long grass after a Kongoni. A cobra suddenly rose in front of him, and my friend had the presence of mind to close his eyes tight as the venom was ejected on to his eyelids. Lying back, he waited the arrival of his syce, whom he told to take his water-bottle and empty its contents on his face. After a careful sponging away with a wet handkerchief he opened his eyes, and was none the worse. The cobra is very nervous in captivity, and for this reason nearly always spits when disturbed. On one occasion it spat on a passing native’s neck through the ventilating gauze when the boy was fully four feet from the cage. This happened several times, and for safety the gauze had to be removed from the cage. Toads seem to be their favourite diet, and the half-grown cobra at present in the writer’s possession eats about one per week. On December 7 it killed a very large one, and its deglutition was one of the most difficult and laboured I have ever seen ; the head lost all shape, resembling a circular band of skin in which shone two beady eyes, the quadrate bones stuck up against the distended skin like horns about to bud. In conformity with its diet it is nocturnal in habit, though frequently to be found basking by day. One of the most 324 NOTES ON SNAKES IN EAST AFRICA exciting captures occurred one night last August, when the writer was sitting with a friend and a boy rushed up to the tent gasping 4 Nyoka Bwana,’ and was away before he could be interrogated. Seizing the lamp we followed him through the camp till, halting near his banda, he waited for the light by which we were able to see a fine cobra streaking away at full speed in the direction of a marquee, around which some forty boys were sitting on empty boxes and upturned petrol tins. They also caught sight of it, and a fine commotion ensued with the upsetting of tins and the falling over tent ropes as the alarmed natives endeavoured to escape. Shielding his eyes with his helmet the writer headed it off, and when it turned he again got in front of it when it stopped. The night was pitchy- dark, and when the snake was on the move it continually got beyond the rays of the solitary lamp. When a second light was brought the writer pinned the snake down by the neck, but the ground being hard it withdrew its head and spat, although in the darkness one could not see where the venom went. A second attempt was more successful, and the reptile was picked up by the neck. In the western portions of East Africa the Black-lipped Cobra ( Naia melanoleuca) is found, and the writer received a number of specimens which were ingeniously caught by means of a bent stick and a snare set over their holes. As the snake emerged from its retreat the spring was released, and the reptile was suspended in the air by its neck. This species is readily distinguished from its near relatives by the vertical black lines bordering the upper lip-scales. A fine Egyptian cobra ( Naia haie), probably eight feet long, was shot at Longido West. This snake ranges from Palestine and Arabia to Zululand and the Transvaal, but does not appear to be common in these parts. Like the black-necked species it also has the power of discharging the venom from its fangs to a considerable distance. Gunther’s Garter Snake ( Elapechis Guentheri) is a short viperish-looking relative of the cobras. It is beautifully ringed in coral-pink and black or white and black. It is quite common in Nairobi, where it is frequently to be found lying NOTES ON SNAKES IN EAST AFEICA 825 among the fallen leaves on the edge of the forest reserve. The writer met with it at Kagiado, but nowhere in German East Africa, though it has been recorded by others from Tanga and elsewhere. The widely-distributed Yellow-bellied Sea-snake (Hydrus platurus), which inhabits the Indian and Pacific Oceans from Australia and India to Panama probably also frequents the East African coasts. A comrade while bathing at Dar-es- Salaam saw what he took to be an eel, but from his description it appears to be more probably this Sea-snake. All sea-snakes are highly specialised for their aquatic life, having vertically flattened tails with which they propel and steer themselves. Their nostrils also are provided with valves enabling them to be closed when the creature dives. All the species are very poisonous. The next division of the Colubridae comprises the back- fanged ( Opisthoglypha ) venomous snakes, and, by reason of their poison-conducting teeth being situated so far back on the upper jaw, it is difficult for them to inflict a dangerous bite upon man unless the reptile is of large size. Their venom is often highly toxic, in some species approximating to cobra venom in its neurotoxic nature ; in others the haemotoxin preponderates as in the vipers. Whilst searching beneath stones at Longido one frequently met with a terra-cotta-coloured snake with a black head ; in bulk they were much the same as a lead pencil and did not exceed a foot in length; their narrow gape rendered them harm- less. At Arusha one night the writer was called up by some- one who said that there was a snake in one of the tents in the maxim lines. On arrival we found the tent in darkness, and no one willing to move. One complained of a snake crossing his face, having waked him up, and he absolutely refused to move till a light was brought. When an inch of candle was at last obtained, and being assured the snake was not to be seen, he sprang up quickly, and after a brief search, discovered the snake beneath his blanket. It was one of these Jackson’s black-headed snakes ( Apparallactus Jachsoni), — there are seven species of the genus in East Africa. The common South African Boomslang ( Dispholidus z 826 NOTES ON SNAKES IN EAST AFRICA tyjous) occurs here in even a greater variety of colouration than it does in the south. At Handeni a fine green female was killed by natives when clearing the ground for camping. At Morogoro the writer has obtained the vivid green variety, the black variety, and the brown variety. In the Yala river district alone, not only did the all-green and all-black varieties occur, but the intermediate ones in which each green scale had a small black speck giving the appearance of a black- spotted green snake, and a fifth variety in which the black speck is so enlarged as to almost oust the green from the scale, which gives the appearance of a green-speckled black snake. The following extract from the writer’s diary will show how easy it is to confuse snakes when guided by colour alone : ‘ “ Has the snake-catcher any time to spare ? ” inquired a Canadian motor-driver putting his head in at my banda. “ Ready if it’s not too far,” I replied, preparing to follow him. His ambulance was stopped some sixty feet from my quarters. u As I pulled up,” said he, “ the beggar came down the tree, it appeared to form a circle and slid down like a ring in a moment. On reaching the ground it made straight for the car and there it is.” So speaking, he pointed to one of the front wheels, around which lay twisted a bright enamel-green snake with its head resting on the tyre under the mud-guard. ‘ “ Oh,” said I, confidently, “ less than a month ago I took a snake of the same species from the back of a car which was standing not thirty feet from where yours is now.” Now, having only seen a brown boomslang during the whole time I had been in Morogoro I was unprepared for it, and mistook this specimen for the harmless Spotted Wood Snake, which is so common in the trees forming the avenue. Approaching the snake I made a diversion with the fingers of my right hand whilst slowly approaching those of the left to the snake’s neck. It remained motionless till my fingers were within six inches of it, when, slipping from the wheel, it would have made off had I not hastily grabbed it by the tail and swung it round till its body was round my hand, and pressed my thumb on the back of its neck just as its head reached my hand. 1 Not till then did I notice the oblique nature of its scales, which immediately showed it was not the harmless species NOTES ON SNAKES IN EAST AFRICA 827 I took it to be. It was very gentle, making no effort to escape until placed in a vivarium, when it dashed about in the wildest manner, inflating its neck vertically and striking at the glass when anyone came near. A brown boomslang was bitten by a puff adder, which would doubtless have eaten it had it not been removed from the case. On eviscerating it, the heart was found to be still beating quite ten minutes after it had been bitten ; the site of the bite some three inches anterior to the vent was in a fearful state, and, had one not known the circumstances, one would have supposed the creature to be in the last stages of putrefaction. The scales were loose and falling off, pinkish blood oozed from between them and from the vent, the fat was speckled with blood, and the intestines were full and oozing with the same. The whole region was blown up as if with gas and had the puffy appearance of decomposition. In captivity these boomslang fed well on living or dead birds. The Bird Snake ( Thelotornis Kirtlandii) of Tropical and South Africa seems to be uncommon at Morogoro, as only two specimens have been brought in during two months. The angular and elongated head is marked with green, the body is brown, slender and whip-like. It frequents trees. The next genus is that of the Sand Snakes (. Psammophis ) with six East African species, specimens of all being in the writer’s possession. One recognisable feature that they possess in common is a fine speckling of the lip-scales which are usually white. Three species are to be found at Morogoro, namely the Angola Sand Snake (P. angolensis), the Hissing Sand Snake (P. sibilans), and the Underlined Sand Snake (P. subtaeniatus) . Specimens of the last named, which is by far the commonest, were taken at Lukigura and Matomondo, whilst at Morogoro scarcely a week passes without one being seen. It may be crudely described as a brown slender snake with a bright yellow underside which is marked along its length with a pair of parallel black lines. Their favourite food is the Striped Skink ( Mabuia striata), which they pursue with inconceivable swiftness and hold until the slow-flowing venom has paralysed 328 NOTES ON SNAKES IN EAST AFRICA the prey, when it is swallowed in a half-dead condition. The writer has been bitten by this species without any unpleasant effects. On October 20 two geckos ( Hemidactylus ) were chasing each other about the roof of a grass hut ; the pursuer was pale grey, the pursued brown-black ; before the pursued quite realised it, the writer had seized and popped him into the Underlined Sand Snake’s case. To see the snake chase it was very interesting ; being midday and the weather hot, the snake streaked and doubled about the case, and the gecko dropped its tail ; the snake undeceived by the wriggling of this red herring shortly afterwards seized and chewed the gecko as they usually do ; the gecko became very limp, was worked round in its mouth, and swallowed head first. On September 23 10 eggs were found in a 45-inch female. „ October 22 8 „ „ 39 „ >> >) 22 7 „ „ 27 ,, „ „ 22 6 „ „ 28 From this it may be safely assumed that the number of eggs produced varies in direct ratio with the size of the parent. On the same date, October 23, another snake laid six eggs measuring 1J inches by 7 inches, and on December 16 and January 1 sixteen newly hatched young ones were taken. The Hissing Sand Snake (P. sibilans), seen at Voi and Bissel and common at Morogoro, is an olive-green, brown or straw-coloured reptile ; underneath it is dirty white or yellow. The largest specimen taken locally by the writer measured over five feet. They feed readily in captivity, so eagerly in fact that they will take their food practically from one’s fingers. The writer has watched one kill and swallow three striped skinks in rapid succession. On another occasion one of these snakes, having ineffectually struck at a skink several times, seized the lizard in its coil with the head downwards. After failing to find the head it bit savagely at the base of the tail ; finally, loosening the top coil, it bit the skink in the abdominal region, and as soon as it became limp proceeded to swallow it in the usual manner. NOTES ON SNAKES IN EAST AFKICA 329 I put a live mouse ( Mus belli) into the Hissing Sand Snake’s cage on November 28 ; it was seized almost immediately and chewed and swallowed. A naked nestling mouse was also swallowed. Then followed two skinks, one of which was taken almost from my hand by the larger snake ; the other lizard jumped into the biscuit tin which served as a water tank. After watching it for a moment in the water, the other sand snake, turning, slid up to the snake which already had a skink in its jaw and watched it intently. Gradually approach- ing, it suddenly made a grab at the skink as snake ‘ A ’ was changing its hold. The jaws of 4 A ’ enveloped the upper jaw and head of ‘ B,’ but 4 B’s’ mouth closed on 4 A’s ’ lower jaw and both parties worked their fangs vigorously. Seizing the combatants I attempted to extricate the tangle, but their small hooked teeth were so firmly embedded that I left them in order to get a pair of forceps. On returning I found they had both dropped the skink and 4 A ’ had seized ‘ B ’ by the neck — a great commotion ensued, ‘ B ’ attempting to get its coils round * A’s ’ neck. Finally 4 B ’ became exhausted, and 4 A,’ working 4 B’s ’ head round till the snout was in its own mouth, began to swallow. After a foot was out of sight 4 A ’ lost heart and, gaping widely, reversed gear and disgorged poor 4 B,’ who made off with a gory head and slimy skin. After taking a couple of turns round the case ‘ A ’ commenced to poke about till it found the skink, which it very soon swallowed. On December 7 a second fight began but did not go to such lengths as the first, and 4 A ’ was given a Gherrosaurus major lizard, which was a big mouthful for the snake, and served to keep it quiet for a time. On December 9, however, it attempted to take a mouse out of the coils of a Rhamphiophis oxyrhynchus snake, and not wishing for further trouble, the writer removed the quarrelsome snake to a cage with an Underlined Sand Snake. On the 30th the latter died, probably because the former had eaten all the food, so after feeding the Hissing Sand Snake with a skink the writer put the Under- lined Snake’s head into its mouth and it continued swallowing it until it was gone. The digestion of this big meal took precisely eight days. 880 NOTES ON SNAKES IN EAST AFRICA This snake is preyed upon by eagles, for on June 4, whilst stalking a martial hawk eagle, the bird disappeared for a moment and the next minute rose from a mealie patch bearing one of these snakes in its talons. Whilst circling low over the writer’s head it seized the snake in its beak, and still holding it in its talons swallowed it on the wing. On June 8 another species of hawk eagle was shot with a Hissing Sand Snake in its stomach. There were twelve eggs in a female which was killed on October 22. These eggs measured 1J inches by J inch. The Short-snouted Sand Snake ( Psammophis brevirostris) is very common in Nairobi, and the writer has often captured them when crossing paths and roads both there and at Arusha. Its habits of feeding and dietary in captivity are similar to the preceding species. An unusually coloured specimen of the Two-lined Sand Snake (P. biseriatus) was taken basking on a heap of stones on the rifle range at Yoi. A handsome and gentle snake is Rhamjphiophis oxyrhynchus, which owns no English name, but is called by the natives Earth Snake, probably because its reddish colour is somewhat similar to the soil. When first caught it is naturally wild, but becomes so tame that the writer has moved eight or nine from case to case as they lay in a twisted heap. After a few days in captivity it is safe to handle them freely without any risk of being bitten. They probably eat mammals, birds, and reptiles. A live mouse being put into the cage on December 9 at 10.30 a.m. was seized and constricted with three coils around it ; at 10.31 a.m. a Hissing Sand Snake came up and seized its hind-leg ; at 10.33 a.m. the Earth Snake took hold of the mouse’s tail at the root and loosening its coils endeavoured to swallow the creature tail first ; the hind-legs proving a serious obstacle, however, it moved its jaws round to mid- body by 10.35, when it got the mouse’s head into its mouth ; at 10.39 a.m. only hind legs and tail were left. At 10.41 — 50 seconds — the last vestige of tail had disappeared. On July 12 a live and newly-fledged young sun-bird which I put into the snake-case, where there was a small python NOTES ON SNAKES IN EAST AFRICA 881 and an Earth Snake, was found dead two minutes later on my return. Its head was swollen tremendously, having been bitten in the throat by the Earth Snake, which, how- ever, was not to be seen ; possibly it had withdrawn into the grass at the writer’s approach or had found the scimitar- shaped bill too awkward for his small mouth. On December 81 one was seen to eat a striped skink. The biggest specimens measured five feet ; a small female laid thirteen eggs on October 10 ; young specimens were taken early in January. The Herald Snake (. Lejptodira Jiotamboeia) , which is a Tropical and South African species, is common both at Nairobi and Morogoro. Black above, it is white-lipped and white beneath ; the upper surface is freckled with minute white spots, which serve to distinguish it from the otherwise very similar but non- poisonous House Snake which is also very common. The Herald Snake is usually under two feet in length, but occasional specimens are to be found over three feet long, in which cases the colour tends to olive. They frequent tents, outbuildings, or rubbish heaps in search of the mice or toads which constitute their food. Nocturnal in their habits, when molested they squirm like a worm. Another snake which rambles abroad after dark is the Half-ringed Cat Snake ( Tarbophis semiannulatus) , which the writer never met with till he came to Morogoro, and even here it seems scarce. The third and last division of the Colubridae embraces all the solid-toothed and therefore harmless Aglypha. Of these by far the most interesting is the almost toothless Egg-eater (Dasypeltis scaber). This extraordinary reptile has a most curious development of the vertebral hypophyses ; the lower spines of some thirty-two of its vertebras not only pierce the gullet, but are tipped with enamel and are used by the snake to crush the shells of eggs which it swallows whole. After the contents of the egg are safely received into the stomach, by some curious mechanism the two halves of the shell are fitted into one another and disgorged. The snake can be readily recognised by the large gular scales which enable it to distend its throat to the capacity of 882 NOTES ON SNAKES IN EAST AFRICA the egg. The body scales are also exceptionally strongly keeled. The variations in colour are many. The variety found at Morogoro rather resembles the Rhombic Night Adder in colour and markings. Egg-eaters of this type were also collected in a trench at Mbunyi, and were brought into Makindo camp in a hollow log, in which was also a Zonure lizard. At Nairobi both all-black and all-brown varieties are to be obtained. An insignificant little bronze-olive snake called Homalosoma lutrix was taken in the hospital banda at Arusha. The Spotted Wood Snake ( Philothamnus semivariegatus), a single specimen of which was found dead in the road at Mombo, is also abundant at Morogoro, a favourite haunt of these reptiles being the acacia trees which form an avenue up to the Secretariat. The method of capturing this snake may be best illustrated by the following extract from the writer’s note under date of October 4 : ‘ A Spotted Wood Snake was seen in the acacia trees near my tent. In a very short time I was after it, and pursued it to the topmost twigs, which I shook violently. With wonderful swiftness it travelled into the next tree, into which my toto climbed, and when it had got to the extreme end of a branch he also began to shake it. The snake wound itself tightly round, but by and by was worked loose and dislodged ; on reaching the ground — a fall of twenty feet — it made off without a moment’s hesitation, and when I seized it, it bit a couple of times drawing blood. ’ On one occasion, whilst standing under a tree in the native camp, a Wood Snake was blown down by the violence of the wind ; another snake was captured in the hood of a motor standing under the aforementioned avenue of trees. A female taken on November 25, which was 52 inches in length, contained six undeveloped eggs. The common East African Green Snake ( Chlorophis neglec- tus) was a common object lying on sprays of foliage of the bushes which overhung Nairobi river, and when disturbed they slipped quietly into the water. One was seen crossing the lawn of a house at Kerogwe. The writer found that they would feed on small frogs in captivity. From the Yala river NOTES ON SNAKES IN EAST AFRICA 88B district nearly a hundred green snakes were sent to the writer for examination, and strangely enough they belonged to no less than four different species of this genera ( Chlorophis emini, hoplogaster, heterolepidotus, and irregularis). The Mole Snake ( Pseudaspis cana) was met with in various parts of British East Africa, the farthest north being on the West Kenia plains. Two chequered young ones were un- earthed whilst digging trenches at Arusha, and when walking up to camp at 5 p.m. one day a fine female over five and a half feet in length shot across the path, and was at once captured by the writer. This snake was kept alive in a pillow case in his bivouac during the three weeks at that camp. It was always gentle and docile. The brown or olive adult Mole Snake is extraordinarily similar to a black-necked cobra, and many persons who were familiar with the cobra could not be convinced that the speci- men just referred to was not the more dangerous snake. At Makindo some natives were felling a very big but hollow tree, and as it fell a fine mole snake issued from the hollow trunk, only to meet with death from a native’s panga. These beautiful and harmless snakes by reason of their diet of rats and mice are most useful to the agriculturist, for they are able to pursue the rodents down their burrows and devour the young in their nests. The Wolf Snake (. Lycophidium capense) was taken by the writer at Nairobi, Longido, and Morogoro. It is a small snake under eighteen inches in length and not often met with owing to its retiring and nocturnal habits. A specimen alleged to have been killed on Government Farm was brought to the writer, and was left among some papers on his table, where some hours later it started to move about. On being picked up it appeared quite lively, and was temporarily dropped into a biscuit tin containing a small mouse, but during the two hours that elapsed the mouse had not only killed, but eaten all the flesh off the anterior half of the snake’s backbone. The Brown House Snake ( Boodon lineatus) is abundant at Nairobi and Morogoro, but, whereas in the former place speci- mens commonly met with are frequently over four feet in length, in Morogoro they are generally under eighteen inches ; from 884 NOTES ON SNAKES IN EAST AFRICA thisut may be argued that some natural enemy prevents their reaching maturity. In captivity they feed well on mice, which they are able to swallow though greater in diameter than their own bodies. The outline of the mouse can clearly be traced by the distended skin in the region of the stomach. The pursuit of their favourite food leads them to frequent the abodes of man, where they meet with little mercy though so perfectly harmless. One apparently resided under the writer’s bed for a long time where it was found in a box too gorged to offer resistance. Two cast skins were found under the floor, which looked as if the snake had frequented the spot for several months. They will bite readily enough, and their many pin- like teeth draw blood. On one occasion some time after hand- ling one of these house snakes, whilst drying his hands on a towel, the writer found a tooth embedded in his hand. A young specimen of the Olive Grass Snake ( Tropidonotus olivaceus), the East African representative of the English Grass Snake, was found down by the stream at Makindo when taking horses to water ; it was unable to get away and had probably been trodden upon. Its eviscerated body when placed in a tin of water overnight was carried off by a rat, whose foot- prints were clearly traceable in the dust. The only other fresh specimens seen were taken in the Yala river district. The next family under consideration is the Boidae, and here let me say that no Boa- constrictors are found in Africa; the true Boa- constrictor is found nowhere outside South America. A small stumpy Sand Boa ( Eryx thebaicus), which is more likely to be mistaken for a viper, is found in East Africa, and specimens were collected at Mbunyi and Kahe. The colouration is a mixture of pink and chocolate. These snakes feed on mice and young birds, which they constrict before swallowing. The large Python (Python sebae), which ranges over the whole of Tropical and South Africa, where it is variously called Rock Snake and West African Python, reaches a considerable size. A headless skin 26 feet long was brought in by natives from Ngeri-Ngeri. The largest of five specimens brought in by Mushairi was 12 feet in length, weighed 80 lbs., and has been in the writer’s possession since September 12, when it was NOTES ON SNAKES IN EAST AFBICA 835 made as comfortable as possible in a netted cage 12 feet long, with a stream of water running through, and a tank sunk in the middle. However, it was not content, and absconded in the night, having found a weak spot or broken strand in the wire through which it forced its way. A broad track led past the tent door and down through the camp amongst the marquees ; one old native admitted having seen it about midnight, but * thought it was a Swahili,’ so hurried into bed. This idea, that the souls of certain distinguished people enter into the bodies of big snakes, is not an uncommon belief amongst the natives. On September 18 about 8 a.m. there were cries of ‘ Nyoka,’ and presently a boy came running to say that a big snake had been seen near the Askari village 120 yards from here. As the snake stick had been broken the previous evening, it was quite ten minutes later before the writer reached the spot where the snake had been seen, just behind the village in thick bush not more than 200 yards from the cage from which it had escaped six days before. Under the brushwood I immediately recognised my escaped python, as it lay half-concealed beneath a fallen tree. A native woman was hopping about like a restless sparrow on the trunk of the tree and shouting information to three of her sisters, who had withdrawn to what they considered a reasonable distance thirty feet away. All were armed with pangas, with which they had been chopping off the branches of this tree when they had discovered the snake. Despite the fact that the woman was shrieking information about him back to the village, the reptile never moved, not even when a boy lifted off the brambles and thornbush immediately above him. The first movement was when the writer tried to seize its neck, when he gave a lunge with open jaws, not necessarily with the intention of seizing, but rather to intimidate, after the fashion of a dog showing his teeth. Almost simultaneously he commenced sliding backwards, but this was circumvented by the writer’s boot being placed lightly on his neck, which was then grasped with both hands. For a few moments a tug of war ensued, and considerable strength had to be used to get it clear before the beast could be crowded into a sack. 836 NOTES ON SNAKES IN EAST AFRICA Quoting from my diary of September 20 : 4 At 1 a.m. this morning I was roused by an orderly “ Did I want a fine big snake ? ” Dressing hurriedly I accompanied him to the Mess tent of the sick officers’ wards. We saw its track in the road as we went. It lay extended beneath the table, and as we entered with a feeble lantern, it struck forward with open jaws a couple of times. Obviously it was excited and more lively than in the daytime. ‘ I made a detour round the table as it started to retire and grabbed it by the neck just as the orderly disappeared to answer a patient’s call. For full two minutes I struggled with the creature, slipping off its coils as fast as it put them on my arm ; my foot upon its tail gave me an advantage and I managed to get its head into a sack, after which it went quietly enough. Within a quarter of an hour from the time I was disturbed I was back in bed, having first ascertained that it had squeezed its way out through the wire mesh through sheer force.’ Its feeding possibilities proved very interesting. From September 12 to November 27 it only ate three rats, after which it cast its skin. On November 28 it took a beautiful little malachite king- fisher (. Ispidiana ). On December 1 it caught and killed a young goshawk which was just able to fly. On December 2, hearing the monkeys making a great to-do, I hastened out to find the python sliding silently towards the bush in which was the goshawk’s nest ; the remaining goshawk was only removed just in time, for a few minutes later the python had climbed up and was nosing round in the nest. On December 9 it ate a dead fowl. On December 12 a dead rat and also two headless blue rollers that were smelling badly ; it defecated a large amount of hair and feathers and apparently disgorged the quills of the fowl swallowed two days ago. On December 14 it distinguished itself by not only taking a dead wingless fruit pigeon, but the skinned body of a little egret. Having got the body crosswise in its mouth it was in great difficulties, but immediately on the writer taking hold of its neck it dropped it easily enough. It then took it again and swallowed it without further trouble. After NOTES ON SNAKES IN EAST AFKICA 887 this it had several rats, but refused a white-shouldered crow on December 26. On January 2 it took fully twenty minutes to swallow a large duck, on the 4th it swallowed a fowl which was green and stinking, on the 5th a crow, on the 6th a fowl, on the 7th a fowl and a duck. On November 8 a 6-foot and an 8-foot python, which were temporarily placed in a sack at the foot of the writer’s bed, escaped, and as there had been heavy rain during the night, tracking them was out of the question. During the same afternoon, whilst endeavouring to dig out a monitor lizard, which had retreated into a clump of grass, the boys had a surprise. Having burnt off the tall sedges which overhung and trailed into the water, one native then began to dig with a trenching tool whilst the writer waited with a forked stick for the moment when it should break cover. Seeing the bank was cracked and that the dislodgement of the loose portion would facilitate matters, he reached down with his left foot and calling to the boys to stand clear, pushed it off and it went rolling down into the stream in which the natives were standing knee deep. Almost simultaneously a handsome piece of scaling slid through the remaining undergrowth and into the water like a flash. Eealising it was a python, the writer slid down the bank after it, but all he got for his pains was a smear of mud and burnt grass up his side and a wetting, the snake making good its escape in the water. On December 25 a 7-foot long python was to be seen lying by the path up the mountain, having been killed by a native on the river the previous day. The five species of East African Worm Snakes ( Glaucoma ) belonging to the family Glauconidae are curious little creatures, not much thicker than the lead of an ordinary pencil and jet black in colour. Their highly polished scales render them very slippery and difficult to hold when they squirm about. They live underground and feed upon ants. The writer once put a specimen of Glaucoma conjuncta into a box containing two bucketsful of comb of the white ant or termite, with the result that next morning the snake was dead with two of the 388 NOTES ON SNAKES IN EAST AFRICA soldier ants biting it. The writer has taken quite a number of specimens of Glauconia conjuncta at Longido and Morogoro in the following situations : under a wood pile ; among roots of a tree that was being dug up ; in holes being excavated for a banda pole ; crossing paths, roads, &c. The last and possibly most primitive family of living snakes is the Typhlopidse or Blind Snakes, which are adapted to a subterranean existence like the Worm Snakes, but come above ground after showers of rain. The prevailing variety of the Spotted Blind Snake in German East Africa seems to be a mottled blue-grey with white tongue. At Garagua one was taken under a stone ; at Mbunyi in a tent trench ; at Handeni one came up through the soil during a shower and another was caught in the road ; at Morogoro in a salvage pit and another in the road. They are uninteresting objects to keep in captivity, as they immediately bore into the soil with their well-adapted head, obtaining a purchase with the spiny-armed tail. They feed on termites, and a large Blind Snake being placed in the same box as the unfortunate Worm Snake just referred to, throve well. Another somewhat uncommon snake is Schlegel’s Blind Snake (Typhlops Schlegelii), a dirty- white or flesh-coloured reptile, of which the writer obtained three specimens at Moro- goro. Eleven other species of Blind Snake are to be found in East Africa, but owing to their very local distribution are difficult to obtain. From the foregoing it will be seen that there is no easy criterion whereby one may readily distinguish between the poisonous and harmless species. The variations in colour are legion, and the shape of the head is no guide. Some perfectly harmless snakes flatten their head until they look very wicked, whilst such a dangerous creature as the Mamba has a head as shapely as those of any of the harmless colubrines to which it is nearly related. Only by constant examination and handling of dead specimens is one likely to be able to identify a live snake as it flashes past across one’s path. A VOLCANIC ERUPTION IN EAST AFRICA 889 A VOLCANIC ERUPTION IN EAST AFRICA By C. W. Hob ley About forty miles south of the Anglo-German boundary, and twelve miles south of Lake Natron in the southern extension of the Rift Valley, there is a mountain, called by the Masai — Donyo L’Engai (‘ The mountain of God ’). This mountain is of volcanic origin, and in shape is a very perfect cone, and various travellers have recorded the fact that considerable quantities of steam issued from its crater. Its altitude is, according to German authorities, 9480 feet. It is now bare and devoid of vegetation. According to the Masai tradition, it is the fountain from which their original live-stock came forth ; and they will tell one that they can at times still hear the lowing of cattle inside the mountain ; and they occasionally send offerings in the shape of black sheep and goats, which are sacrificed on the mountain- slopes — especially in times of famine, or disease among themselves or their stock. Subterranean noises are, however, not uncommon in volcanoes which are only partially extinct. In January 1917, however, this mountain suddenly burst into active eruption, and continued in eruption until about June. It is said that flames and smoke could be seen as far away as the slopes of Meru Mountain, which is a great extinct volcano about fifty miles east of Donyo L’Engai. A vast amount of material was ejected from the mountain, and the country for many miles round is covered with a pall of grey volcanic ash and soda. The Masai state that the layer of volcanic ash was deposited over a large area to the west and as far south as Ngaruka. When the eruption commenced, the Masai were extremely exercised, and sent women to sacrifice animals and pour out calabashes of milk on the slopes of the volcano ; the women 840 A VOLCANIC ERUPTION IN EAST AFRICA even milked their own breasts to appease the anger of the spirits believed to reside within the mountain. The lava has flowed for a long distance down the valleys which score the flanks of the mountain, and, in cooling, it has cracked into irregular masses, having the appearance of cakes of grey cement. Farther from the mountain, the country is covered with a powdery ash, and this extends, mixed with soda, to a distance of twenty-five to thirty miles ; and large areas of grazing-land have been temporarily destroyed. In some places the deposit of ash resembles black sand. The western flanks of Gelei Mountain, which rises to a height of nearly 10,000 feet on the east side of Lake Natron, were covered with ash and soda, and the water-holes on the mountain were so fouled with soda that they have become unusable ; the springs on the east and south-east side were, however, unaffected. The river, known as Engare Sero, or Mito miwile, rising in the high land to the west of Donyo L’Engai, is unaffected, but its banks are covered with a thick deposit of volcanic ejecta. The water in both areas of this stream is still sweet, because the springs rise high up in the western plateau. Heavy rain occurred in this region early in June 1917, and it is recorded that for a period of four to five days the rain- water was strongly alkaline and undrinkable by cattle. The Masai state that many herds of cattle, located in places where the only water was in pools, have died through drinking the alkaline water. The mountain is now quiescent and only slightly smoking. The ejection of vast quantities of soda from this volcano appears to be evidence that sodium carbonate is one of the principal constituents of the magma from which the eruptions of the Rift Valley are derived. The persistence of soda in so many of the lakes in that valley, from Rudolf down to Natron and Eyassi, is a marvellous phenomenon. Mr. Parkinson quotes in one of his papers a theory of Gautin to explain the genesis of thermal waters. It is assumed that the subterranean magma is largely composed of sodium silicate, and that plutonic water charged A VOLCANIC ERUPTION IN EAST AFRICA 341 A A 842 A VOLCANIC ERUPTION IN EAST AFRICA with C02 breaks up the sodium silicate into amorphous silica and sodium carbonate. This would of course satisfactorily explain the existence of the ridges of flint-like silica which occur on the flats near Magadi Lake. It may be, however, that the magma contains vast amounts of metallic sodium, and that the oxidation of this has provided the motive power for the vast eruptions which have taken place in this area. However, be that as it may, there now appears to be little doubt that the soda is what may be termed inter-telluric, and is not produced by the decomposition of soda felspars or other rocks containing the alkali in question. There is no record of any previous eruption of this mountain, but it was always remarkable for large white patches on the sides of its cone and a pinnacle on its summit — the latter was, however, blown away during the recent eruption. Herr Kasin, a German scientist, who accompanied Dr. M. Scholler’s expedi- tion, informed me that he had ascended Donyo L’Engai and found that the white deposit on the upper portion of the cone was soda. The slopes are now covered with a grey whitish deposit, which is said to give the impression of snow that has fallen on a rough surface. An examination of the German geological map discloses the fact that, south of Lake Natron, the Rift Valley continues to run southwards through Lake Mweri and Dalanga ; another branch runs south-west through Lake Eyassi, and in between the two there is a fault valley which the Germans named the Hohenlohe Graben. From the direction of the two latter, it may be surmised that the complicated scheme of faulting, which caused them, originated in the vicinity of the volcano Donyo L’Engai. This junction has, however, been masked by an extraordinary number of large volcanoes commencing at the north end of Eyassi and running in a north-east direction : these are Oldiani, Ngorongo, Ololmoti, Ololmasin, Elanairobi, Kerimassi, and finally, Donyo L’Engai. Whether this terrific amount of volcanic action produced the split in the Rift, or vice versa, it is impossible at present to conjecture ; but it is safe to say that there must have been, and probably still exist, very severe stresses in the earth’s crust in this region. DOLPHIN (FALOOSI) FISHING OFF MOMBASA 848 The recent recurrence of eruption in Donyo L’Engai is, to my mind, clear evidence that this area has not reached a state of equilibrium. I am indebted to Major E. D. Browne for the particulars of this eruption. DOLPHIN (FALOOSI) FISHING OFF MOMBASA. By Capt. C. W. Woodhouse The name ‘ Dolphin ’ probably conveys to most people a picture of a large porpoise-like animal, which leaps repeatedly out of the water alongside some ocean-going ship ; but the dolphin described here is the dolphin of poetry — Coryphcena haletus — a true fish, Swahili name ‘ Faloosi,’ well known since very early times, owing to the successive waves of colour which play over its body while dying. These colours include gold, green, green- blue, silver, and, finally, grey when dead. The colours, when alive and free, swimming, are a silvery body, with rich orange on the belly, and a sort of golden sheen over the silver, in some lights ; blue spots on the side (when seen from above ; a sideways view, when leaping, is silver). Green, mottled with black culmen to the head ; spring back-fin, blue and green ; eye, bright yellow. The weight of this fish runs from 10 to 80 lb. It is entirely a deep-water fish and, accord- ing to fishermen’s reports and personal observation, never comes into shallow water. It is an exceedingly swift-swimming fish, and has the habit of repeatedly leaping out of the water, both when swimming free and when hooked. Its food seems to be cuttle- fish (squid), sardines (seemu and dagaa), and possibly other small fish. With regard to these latter, seemu appears in Mombasa, anyhow, to be the sardine or sprat caught in the harbour ; while dagaa seems to be pelagic. The writer could not distinguish any difference in the two species. The season for faloosi is from January to June, but small numbers are caught all the year round. The fishermen state the faloosi 344 DOLPHIN (FALOOSI) FISHING OFF MOMBASA first appear at Mombasa in January, coming down from the north ; and these fish are not in shoals, only two or three, or perhaps single ones, being encountered, which, however, are always the largest fish, running from 20 to 30 lb. In March and April, the large schools arrive, and these average some 15 lb., running from 10 to 20 lb. apiece. The bait always employed is squid (Swahili name, 4 Ngisi ’), and to make catches of faloosi it is necessary to paraphrase Mrs. Beeton’s advice, * First catch your squid.’ The local squid at Mombasa is a hungry yet secretive animal, and has to be approached with caution. Its habit is to rest on the reef, hidden, or imitating the seaweed which covers the floor of the lagoon inside the outer reef. The procedure is as follows : The canoes are paddled in the early morning out over the reef , and in the still water allowed to drift slowly in from J to 1 fathom of water. The occupants start casting a slightly weighted dead fish, some 5 inches long, well secured by the tail to a fish-line. This is allowed to sink to near the bottom, and then slowly drawn in. If this dead fish passes near a concealed squid it is practically certain to be seized. The squid leaves its hold and seizes the fish with almost incredible swiftness. There is no doubt that the ‘ wings ’ on each side of the squid materially help it in this rapid move- ment. The dead fish having been seized by the squid (average length 2 feet), the beast is cautiously coaxed within 15 to 20 feet of the canoe, when a leaded triangle hook attached to a stout line is cast just beyond the squid, and a sharp jerk hooks it. After discharging its sepia, and with a lot of splashing, it is pulled on board and immediately ‘ broken, ’ where the head is inserted into the body just above the eyes. It be- comes quiet, dies almost at once, and turns from mottled umber colour to a pearly white. A sufficiency having been caught, the canoe puts out to sea through a ‘ mlango ’ or a gap in the reef. Great care must be exercised, especially in the mon- soon, in passing through the reef, and it is as well to ascertain beforehand that the fisherman is fully cognizant of the passage and is not merely trusting to luck, as otherwise a nasty accident may occur. The mlango is often crooked, and at times it is necessary to turn the canoe broadside to the waves. Passing DOLPHIN (FALOOSI) FISHING OFF MOMBASA 345 the reef, preparations are being made to fish. The first bait, and that which is considered most deadly, is the inner parts of the squid carefully lashed on to the top of the hook with a fibre of palm-leaf. Next the hook is passed several times through the detached squid’s head and lashed, the tentacles waving free ; and lastly the two long arms are detached, and the hook passed once through the ends, so as to wave about behind the hook. If the weather is fairly smooth, the canoe, after clearing the breakers, is sailed, if not rowed, out to sea. Some 70 yards of line are let out behind the boat, and if shoals of faloosi are expected, more lines are prepared and baited with strips of squid, which is scored traversely to make it pliable. Presently, after perhaps having gone some three or four miles out to sea, a violent snatch is felt on the line, and 70 yards astern a gleaming fish shoots up 6 feet clear of the water. In the early season, when the fish run large, the sportsman will have all his work cut out to keep the canoe in a seaway, and at the same time handle a species of small torpedo gone mad. If the fish is a big one, and no gaff is handy, it is a matter of some skill to persuade it into the canoe ; but by taking advantage of the roll of the sea it can usually be hauled over the side without getting more than a fair share of the Indian Ocean given in as a free gift. As soon as the fish is hooked, another baited line is at once cast out in case more fish are about. If, however, schools of fish are expected, the tired-out fish is cleat ed to the side of the canoe with about 10 yards of line between the fish and the boat. Almost immediately will be seen the rest of the shoal round their tethered comrade, showing little fear of the canoe, but all intent on finding out what their friend had found to eat. These greedily seize the next baited line thrown into their midst, and as many as ten may be captured before the shoal takes alarm. The captive is then brought on board, and the canoe heads in the direction they were last seen swimming, towing a baited line as before. On overtaking or encountering another school the process is repeated. 346 DOLPHIN (FALOOSI) FISHING OFF MOMBASA After midday the fish seem to cease feeding, and it may be taken that the earlier hours of the day are the best. Tide does not appear to affect their feeding ; but on certain days the fish may be found within a mile of the outer reef, while on other days it may be necessary to proceed as far as ten miles out to sea. The danger of the latter is that, should it come on to blow badly, some difficulty may be experienced in making harbour. It may be noted that the native canoe is a far safer boat in a bad sea than the usual shore-boat or motor- boat. With regard to the latter, some auxiliary propellant should be on board — such as oars or sweeps — as a breakdown of the motor at sea might be serious. The writer killed his faloosi on a hand-line, not wishing to risk a valuable rod and reel in case of a capsize, for this tackle might be unreplaceable in war time in East Africa ; but the experiment of this form of fishing was made to gain knowledge of how these fish might be caught and killed on light tackle. The suggestion is made that, as in the native manner, a portion of cuttle-fish should be ‘ spun,’ i.e. drawn behind the boat, mounted on a small lip-hook and two triangles on a wire trace. If fishing from a suitable boat, two rods with the baits at different distances astern might be used; and on a strike on one rod the other to immediately reel up, and cast out as soon as the fish was near the boat under control, in the hopes of hooking its mate. This would apply to the larger and more solitary fish in January and February. For the smaller school fish it is suggested the native line be used first, and, having secured the captive fish, short casts might be made on both sides of the boat by the rod-armed sportsman. It is noteworthy that if on these occasions a fish is seen hesitating at a bait, a couple of jerks which make the bait jump in the water will usually overcome the fish’s suspicion. Finally, it may be noted that many factors enter into this form of sport : the catching of the squid, the sailing of the boat out to sea, the passage of the reef, and the constant watch on the weather — added to the fine capabilities of the fish — all combine, at Mombasa, anyhow, to render a successful morning a most enjoyable one. Perhaps at Malindi the same sport could be had in more sheltered water. NILE PERCH. FISHING IN LAKE RUDOLF 347 JACKALS ATTACKING NATIVES By C. M. Dobbs In June this year, while standing in the grounds round my house at Kericho, one afternoon about 5 p.m., I was surprised to see two jackals come right through the garden. One of them passed me about ten yards away, and the other a little farther. I subsequently mentioned this fact to some natives, and was told that they had recently become very bold and had attacked several people in broad daylight. One man had met a jackal on the road, and it refused to budge ; he threw his stick at it, whereupon the jackal promptly attacked him, and he had some difficulty in driving it away. On another occasion, a native woman, while working in her shamba, was attacked by one of these animals, and, after a hard tussle, succeeded in throwing round it one of the skins that she was wearing, strangling it in this way. Note. — Similar incidents to the above occurred in Kiambu district some years ago. The natives reported several cases of attacks by jackals, and, on one occasion, a jackal came up to the kitchen of the police-officer’s house and bit a native in broad daylight. The explanation prevalent at that time was that the jackals were suffering from an epidemic of rabies ; I am, however, not aware that this theory was ever proved. — Editor. FISHING IN LAKE RUDOLF Note by A. Blayney Percival The Nile perch, or baggera, depicted in the plate in this number (weight 47 lb.), was killed in Lake Rudolf on a light rod and tackle, after two hours’ fight, by an officer of the K.A.R. A number of these fish were killed, weighing between 15 and 30 lb. ; a dead fish of very much larger size was seen. Other fish taken were tiger-fish up to 10 lb., and two species of Silurians (scaleless barbel). Small fish of many species were very plentiful. 348 A SHOOTING INCIDENT A SHOOTING INCIDENT By H. C. Barnes I was out snipe-shooting with a friend, near Nairobi, on May 22 last, when a curious incident happened, which might interest the readers of the Natural History Journal. My friend’s dog pointed a snipe about 50 yards away from me. When I got up, the dog flushed the snipe, and caught it in his mouth. I told him to drop it, and stooped down to pick it up, but it flew off, pursued by the dog. I could not fire for fear of killing the latter, and was watching the chase, when I heard a rush of wings, and two falcons swooped past me. One devoted its attention to the dog by getting between it and the snipe. They then soared up, and we could plainly see the falcon away up with the snipe in its claws. It was impossible to identify the kind of bird with any certainty, but my friend said they were Jagga falcons. ABNORMAL TIDES By Editor A report has been received that exceptionally high tides were experienced at Yanga in March and April last, on the 14th and 15th of each month, that of April being the highest. The water came over the sea-wall and reached the Customs Superintendent’s house, and the path leading from there to the main road. The water also made its way on to the rice-fields from south of the Yanga- Jasin road (part of which was under water as far as half a mile away from the township), and by the small watercourse which crosses the track from the township to the log-bridge over the Umba River. The results were disastrous, practically all the rice reached by the water being spoilt. This was the more unfortunate, as EARTHQUAKE SHOCKS IN EAST AFRICA 849 the rice in question was nearly ready for reaping, and com- prised most of the plantings which survived the drought. In the opinion of the Liwali, these tides were the highest since the days of Seyyid Burgash. It is interesting to note that in April exceptionally high tides were also experienced at Gazi. A report has also been received from Mr. W. McClelland from Kismayu, as follows : There is a large flat piece of ground, known as the Parade Ground, between the Wireless Station and the town. During the equinox, September and March, this land became flooded, and last September there was a very high tide which washed away the gates and fence on the sea front of the Custom House. In March there was a still higher tide ; but an extraordinary thing happened on the 12th, 18th, and 14th of April, when high tides were not expected. On these dates the sea came over the whole sea-front from the Custom House to the Pier ; the water was at least four feet deep on the Parade Ground, and several streets in Kismayu were flooded. This has never happened before within the memory of anyone here, and this high tide came a month after the equinox. EARTHQUAKE SHOCKS IN BRITISH EAST AFRICA By Editor It may be of interest to record the following information regarding some recent earth tremors which occurred on May 16, 1918, at 8.55 p.m. Two shocks were felt at Voi in quick succes- sion and lasting about half a minute. These were followed by a slighter quiver about two minutes later. The stationmaster’s house and dak bungalow were slightly damaged. May 17, 8.5 p.m. — Another slight shock at Voi, and again another during the night, time not recorded. May 17 at about 8.0 p.m. — Shock felt at Masongoleni ; the spring forming the source of the Masongoleni River has been affected, and it has ceased to flow. 350 ANNUAL REPORT, 1917 One of the above earthquakes has closed a spring on the Taita Mountains, near Kedai, and caused it to reopen in another position. Masongoleni is about seventy miles north north-west of Yoi, and Kedai is about twenty miles west -north-west of Yoi. ANNUAL REPORT OF THE EAST AFRICA AND UGANDA NATURAL HISTORY SOCIETY, 1917 By Honorary Secretary The Society has now completed its ninth year, which, owing to the continuation of hostilities, has been marked by no special occurrence and no great additions to the collections. Membership. — We are able to report that, owing to the energy of a few individuals, the membership for this year has increased from 125 in 1916 to 211 in 1917, and this in spite of the fact that, during this year, an entrance fee has been charged in addition to the first annual subscription, and that the associate class has been abolished. Museum. — Interest in the Museum is sustained, as evidenced by the numbers of visitors, and its collections have proved of use to members and the public. Collections. — As already stated, no great additions have been made, except to the series of mounted birds — especially the duck and the water-bird groups. A temporary rearrangement of some of the show-cases has been carried out, and two new cases installed to receive the series of game-birds and water-fowl. Curator. — Mr. Turner, who acted as curator during 1916, resigned in order to go on a collecting expedition, but was re- engaged in April and continued in the service of the Society until the beginning of June, when he asked to be allowed to accept a situation which would allow one of our members to join the military. The Society tenders its thanks to Mr. Turner for his services during his period of office. Although without a curator, we have been able to keep the ANNUAL REPORT, 1917 851 Museum open to the public daily, having engaged a caretaker to attend to the exhibits. New Museum. — Owing to lack of support, both from mem- bers and the Government, the Society has been unable to com- mence the building of the New Museum and so make use of the excellent plot allotted for this purpose by Government. Scientific Expeditions. — At the kind invitation of General Llewellyn, G.O.C., K.A.R., the Society arranged for a collector to accompany a party proceeding to North Turkana, towards the latter part of the year. A report on the progress so far is encouraging, and the Society ought to derive considerable benefit to its collections from this expedition. Publications. — Two Journals have been issued during the year, containing considerable information on scientific matters and items of general interest. A special Supplement has been published by private sub- scription, containing a ‘ Check List of the Birds of East Africa and Uganda.’ This should be of considerable interest to those studying the ornithology of the two countries. Members' Meetings. — Seven evening meetings have been held during the year, at which papers have been read dealing with scientific matters of general interest. We take this opportunity of thanking those members who helped in this way to sustain the interest in the Society. The attendance at these meetings has only been fair. Special Funds. — The Illustration Fund and Building Fund have received scant support, and members are reminded that these are of prime importance to the welfare of the Society. Financial. — The Balance Sheet, as submitted by the Honorary Treasurer, shows a balance in favour of R3. 8194-42 as compared to Rs. 6945*20 of last year — an increase of Rs. 1249-22. The Committee takes this opportunity of publicly thanking Sir Northrup McMillan for his continued and generous support during this year, in having again contributed the curator’s salary. The thanks of the Society are also due to those gentlemen who have held office during the year. EAST AFRICA AND UGANDA NATURAL HISTORY SOCIETY. FINANCIAL STATEMENT, 1917. 852 BALANCE SHEETS 00 o 05 d P EH 3 J25 d PH X *r3 £ © bD rs ‘pJ CD S >-5 Ph O ©=> o3 PH d o O -H-S CM O O CO o o O O CO o o o CM CD CO o CM '*Tl 05 CO ip 7—1 CO 05 Tin • • • d • • Ph r3 O Tb L— -P> © d T—i 05 • r© H # 1 — ! O .3 d c5 'd' c3 e5 © 1 'T3 Ph f— H rr^ d d © 02 iP »P o ip pq o o t— IP tr- CM Ip 00 00 o o o CO ip d-i do cb T—i pq o 7—1 cb ip CO 00 GO a* CO o 00 7—( TJH 7—1 T—i o 7—1 O C3 H GQ S>5 c3 » Ph 63 S PS c3 O w Ph ^ ^ ^ *1- g ^ PH 03 ^ O r-H Ph ~H? '£ g 3 H S ° ^ £ ^ O c3 ^ O I ® g © PS H ,H ^ c3 rO 3 w pH d H © ph 53 EH S c3 CD d ^ S Eh g «+-H 03 o «+H ° § 02 *i-i CD *H-^ UD « *1 PH c8 * P-H 2 02 Ph CD 3 S? £ g rd CD © rd '+3 c3 «H-H Ph O • i— 1 1 o H=> *3 d g CD d Ph ^ ® c3 Pp Ph rd rs (—1 05 3 ° CO O 02 ' 1 db p d •- pq pq ib ip ib iH o a:’ o ip L— OO OO op OO t- pq CO T—i 00 Pi GO CO t- T— 1 CO CO CO pq 7 1 7 1 pq 02 EH Ph 3 O w Ph CO 05 00 Ph © P 3 CD O CD ft d o rd d cO PQ rd 02 fH c3 pqO CO CO CO tH f-1 Q5 CD rH rd t-T ^ r; ® cq CO © o Sft & ■Safi a s ^, § ^ . §2 1a> rj 5-4 ^ ft .S ^ ’© £ jj § CD © ft Ph Ph O oo ph t- CO th 05 02 Ph Eh fc P O o O P P ft H fc H o £-1 , ■ '»’ )>■'' \ m ' ,n ■■ ,S‘m wOl^';J I p:« '? ’ > I ff imm k^^mmjm ■ BM ip iw^'i gfe gs II I . ... .'i 11 . %i ) i i t{ ■ ,■; i ' * > : ' V* ,r 1$?<& ter tiiip Mm Its , k ,*> ‘ -: >’U f 4 > h V 'V^'4W Ube journal OF THE East Africa and Uganda Natural History Society JULY 1919 No. 14, CONTENTS PAGE SOME POINTS OP GENERAL INTEREST ON PLAGUE. Major C. E. Southon, I.M.S. .......... 361 INSECTS AND THEIR RELATION TO DISEASES OF STOCK. R. E Montgomery, Vet. Pathologist ........ 381 ON LEMURS AT MOROGORO, G.E.A. Arthur Loveridgb . . . 401 GAME-NETTING IN UZIGUA. D. K. S. Grant . . . . .403 ON CROCODILES. C. W. Hobley .407 EBURRU STEAM-JETS. J. H. Hervey Pirie, M.D. . . . . .410 NOTES ZIZYPEUS JUJUBE. C. W. Hobley NESTING HABITS OP A SWIFT. A. Loveridgb . . . , .412 NOTES OF INTEREST ON SOME EAST AFRICAN BIRDS, V. G. L. van SOMEREN ........... 414 CONCERNING BABOONS. A. Blayney Percival . . . . .415 PRECOCITY IN ORANGE-SEEDLINGS. C. M. Dobbs . . . .415 CURIOUS GROWTH OF WILD RUBBER-VINE. C. M. Dobbs . 416 EUROPEAN CRANE. V- G. L. van Someren . . . ... 416 FIGHTING LIZARDS. C. M. Dobbs 417 SUPERNATURAL BELIEFS OF THE LUMBWA. I. O. Orchardbon . . 417 ALBINISM. H. G. B. Hesse ......... 420 ALBINO CHILD. A. M.-H. 420 ANT-BEAR OR AARD VARK. V. G. L. van Someren .... 421 ANT-BEAR. Capt. R. O- Crewe-Bead . . . . . . .421 ANNUAL REPORT. Hon. Secretary . ... ... 422 FINANCIAL REPORT. Hon. Treasurer . . . . . . * . 414 ILLUSTRATIONS SACRED IBIS, CORMORANTS, AND STONE CURLEW Frontispiece ANATOMY OF PLAGUE FLEAS . ..... pp. 374-377 STEAM-JETS, EBURRU (TWO PHOTOS) .... To face p. 411 ANT-BEAR OR AARD VARK ...... „ p. 421 EDITOR C. W. HOBLEY, C.M.G. LONGMANS, GREEN, AND CO. 39 PATERNOSTER ROW, LONDON, NEW YORK, BOMBAY, CALCUTTA AND MADRAS All rights reserved Additional copies : to members , Bs. d(4j-); to non-members , Bs. 4 (5/4) r t * l Island on Victoria Nyanza. By Dr. R. van Someren, THE JOURNAL OF THE EAST AFRICA AND UGANDA NATURAL HISTORY SOCIETY JULY 1919 No. 14 patrons SIR EDWARD NORTHEY, K.C.M.G. SIR F. J. JACKSON, K.C.M.G., C.B, F.Z.S., M.B.O.IL SIR H. C. BELFIELD, K.C.M.G. SIR E. P. C. GIROUARD, K.C.M.G., R.E., D.S.O. SIR JAMES HAYES SADLER, K.C.M.G., C.B. SIR HESKETH BELL, K.C.M.G. president THE HON. C. C. BOWRING, C.M.G. I0ice=presibents C. W. HOBLEY, C.M.G. , M.R.Anthrop.Inst., C.M.Z.S. A. B. PERCIYAL, F.Z.S., M.B.O.IL lEjecutiue Committee E. BATTISCOMBE, Assoc. Arbor. Society D. S. K. GRANT H. C. E. BARNES, O.B.E. J. P. COOK, M.B.O.U. F. L. HATELEY R. DAWSON W. A. PAILTHORPE Y. K. KIRKHAM, B.Sc., F.I.C.AgL IbonorarE {Treasurer W. McGREGOR ROSS, B.A., M.Sc., B.E. Ibonorarg Secretary Y. G. L. van SOMEREN, L.R.C.P. & S., L.R.F.P. & S., M.B.O.U., &c. Curator A. LOYERIDGE, C.M.Z.S. (Away on Active Service) Bbitor of Journal C. W. HOBLEY, C.M.G. 1919 BB fa No. 14. CONTENTS PAGE SOME POINTS OF GENERAL INTEREST ON PLAGUE. Major C. E. Southon, I.M.S 361 INSECTS AND THEIR RELATION TO DISEASES OF STOCK. R. E. Montgomery, Vet. Pathologist ..... 381 ON LEMURS AT MOROGORO, G.E.A. Arthur Loveridge . . 401 GAME-NETTING IN UZIGUA. D. K. S. Grant . . . .403 ON CROCODILES. C. W. Hobley 407 EBURRU STEAM-JETS. J. H. Hervey Pirie, M.D. . . .410 NOTES ZIZYPHUS JUJUBE. C. W. Hobley 411 NESTING HABITS OF A SWIFT. A. Loveridge . . .412 NOTES OF INTEREST ON SOME EAST AFRICAN BIRDS. V. G. L. VAN SOMEREN ........ 414 CONCERNING BABOONS. A. Blayney Percival . . .415 PRECOCITY IN ORANGE-SEEDLINGS. C. M. Dobbs . .415 CURIOUS GROWTH OF WILD RUBBER-VINE. C. M. Dobbs . 416 EUROPEAN CRANE. V. G. L. van Someren . . . .416 FIGHTING LIZARDS. C. M. Dobbs 417 SUPERNATURAL BELIEFS OF THE LUMBWA. I. 0. Orchardson 417 ALBINISM. H. G. B. Hesse 420 ALBINO CHILD. A. M.-H 420 ANT-BEAR OR AARD VARK. V. G. L. van Someren . . 421 ANT-BEAR. Capt. R. O. Crewe-Read 421 ANNUAL REPORT. Hon. Secretary ..... 422 FINANCIAL REPORT. Hon. Treasurer 424 ILLUSTRATIONS SACRED IBIS, CORMORANTS, AND STONE CURLEW Frontispiece ANATOMY OF PLAGUE FLEAS .... pp. 374-377 STEAM- JETS, EBURRU (TWO PHOTOS) . . . To face p. 411 ANT-BEAR OR AARD VARK p. 421 861 SOME POINTS OF GENERAL INTEREST ON THE SUBJECT OF PLAGUE By Major Ci E. Southon, I.M.S. The endeavour in this paper is to place before members some points of general interest on the subject of plague. The history of plague will first be briefly traced, then the root-cause of the disease, viz. Bacillus pestis, will be dealt with by tracing it through the laboratory and noting its vitality both inside and outside the living organisms of various animals, with particular reference to the principal host — the rat. The next step will be to examine the life history, habits, and morphology of the connecting link between the rat and human beings, viz. the flea. The final stage is to consider plague in its relation to man. To summarise, the various steps are : the Bacillus pestis, the rat, the flea, and man. Definition. — Let us first define our subject. Plague is an acute, infective, febrile disease, accompanied by inflammation of the lymphatic glands, and caused by a micro-organism, the Bacillus pestis. History In thinking of plague, some of us are apt to regard it as a comparatively new disease, or at any rate our thoughts do not carry us farther back than the Great Plague of London, in the year 1665. But plague has been with us from time immemorial. Thus we find a reference to it in the writings of a physician, Rufus of Ephesus, about a.d. 100, and he was referring to an outbreak in the third century before Christ. The Greeks also wrote of ‘ pestilential buboes,’ and we meet it again in the Great Plague of Justinian, which started from Egypt in a.d. 542, and spread over a large part of Europe. At the end of the seventh century, bubonic plague was recorded 862 SOME POINTS OF GENERAL INTEREST in Italy and in England. These epidemics were traceable for the most part to Egypt ; but in the fourteenth century a new epidemic invaded Europe from Asia by way of the Crimea and the Black Sea, its origin being referred to Cathay or China. This terrible pestilence appeared in Southern Italy in 1846, and made its way over the whole of Europe. It affected England in 1848, and Scotland and Ireland did not escape. A second epidemic occurred in 1361, and a third in 1368. Many historians have dwelt on this great calamity, and Hecker calculates that 25,000,000 persons — i.e. about one-fourth of the population of Europe at that time — died of this disease. The first epidemic, known as the Black Death, and to which I have referred as starting in Southern Italy in 1346, was evidently one of what is now called pneumonic plague. Successive epidemics occurred in Great Britain through the fifteenth, sixteenth, and seventeenth centuries. Finally they culminated in the Great Plague of London in 1665, in which about 70,000 persons died, and which extended widely over the country. Those of you who remember your history dates by the aid of rhymes will recall the couplet : In sixteen hundred and sixty-five. There was scarcely a man in London alive. Soon after that date the disease vanished, helped no doubt by the cleansing action of the Great Fire of London, when In sixteen hundred and sixty-six. The Fire of London burnt the bricks, though of course the fire was merely a local factor. It did not revisit Britain till a localised epidemic occurred in 1900 in Glasgow. During the first half of the nineteenth century, plague prevailed in Turkey, and made occasional advances into the countries round about the Danube. It finally left Europe in 1841, receding to the regions round about Persia, Kurdistan, and Mesopotamia. Let us now try to trace its history to India. There were epidemics of plague in India from the eleventh to the end of the seventeenth century. In 1615 a great epidemic, which ON THE SUBJECT OF PLAGUE 868 lasted for eight years, affected the Punjab and Mohammedan India generally. From 1684 to 1707 an epidemic ravaged Bombay, Surat, and a great part of Western India. In 1836 an epidemic broke out in Bajputana, known as the ‘ Pali Plague,’ from the name of the town in which it first occurred ; and ‘ Maha Murrie,’ which is undoubtedly bubonic plague, has recurred several times in the district of Gharwal in the south-west of the Himalayas since 1823. In 1896 plague was introduced into Bombay from China, and has ravaged the country since that time. During the first ten years of the present outbreak in India, it accounted for a mortality of over 4,000,000 people. China has probably had plague since time was. The first definitely known epidemic was in Yunnan in 1860 ; and from there by way of Pakhoi and Canton it went in 1894 to Hong- Kong, and from this great seaport it was carried by sea traffic to India, Australia, Japan, Europe, Africa, and America — North and South. In 1899 the plague crossed the Equator and appeared at Tamatavi in Madagascar, probably imported from Bombay. From Tamatavi it went across to Lorenzo Marques, and in the same year appeared at Fort Louis in Mauritius. In 1900 it appeared in Capetown, having been imported, in spite of quarantine precautions, from Bosario in South America. It spread to the city and to Port Elizabeth. In 1902 it was recorded for the first time in Nairobi in this Protectorate. There were sixty-five cases and twenty-one deaths. Where this came from is not accurately known. Possible sources were Bombay, South Africa, or Uganda, which last is held to be an endemic centre of plague. Bacteriology Now let me introduce you to the Bacillus pestis. It was first discovered in the year 1894 by two independent investi- gators, viz. : Kitasako, who had been sent by the Japanese Government to investigate the plague outbreak at Hong- Kong, and Yersin, sent by the French Government for the same purpose. Plague belongs to the group of haemorrhagic septicaemias, 364 SOME POINTS OF GENEEAL INTEEEST which are diseases due to microbes which invade the blood stream acutely, and are found in large numbers over the whole body. The plague bacillus, as seen under the microscope in plague lesions, shows a considerable polymorphism. Three forms predominate : short oval rods, long rods, and large oval, pear-shaped or round involution or degenerate forms, which only take on stains faintly. In pure cultures all these forms are also met with. In staining the B. pestis, we find a peculiarity which separates it from most other bacilli, though not from all, and that is the phenomenon of what is known as ‘ polar staining.’ The micro-organism appears stained at the ends, but not at all, or only slightly so, in the middle. This, for diagnostic purposes, is a striking and important point. Now let us see how the bacteriologist would proceed to grow this bacillus. He would employ several methods. First he would take a little material from, say, the bubo, and streak it on a gelatinous substance made from seaweed, known as agar. If he added 2 to 3 per cent, of salt to this agar, he would find that the characteristic involution forms I have just described would be readily obtained ; this feature dis- tinguishing it from other bacilli. He would next proceed to grow the bacillus in broth, especially in broth on which oil-drops were floating. After a time, he would find a growth as a powdery thread, in regular stalactite forms, hanging from the under surfaces of the oil-drops. This, again, to his mind, would form a characteristic and diagnostic feature. (I shall remind you later on of these stalactite forms when I am speaking on plague vaccines.) You will wonder why the bacteriologist takes all this trouble. Are there, then, other bacilli so like plague in their forms, and in the way they grow on these cultures, as to be mistaken for it ? There are several. For instance, the bacilli that causes swine plague ; but in particular, one with a name far longer than itself, the Bacillus pseudotuberculosis rodentium, which is moderately pathogenic to rats, but not to human beings. This so closely resembles B. pestis, not only in micro- scopical but in cultural tests, that we have to resort to the cutaneous inoculation of white rats to differentiate it. We ON THE SUBJECT OF PLAGUE 365 find, then, that where B. pestis kills, the B. pseudotuberculosis rodentium is harmless to these animals. Vitality of the Plague Bacillus outside the Body. — I have described to you now the form of the bacillus and how it is grown. Let us turn to the important point as to the degree of resistance to destruction which it manifests when outside the living body. In general the resistance is slight ; but pure cultures, kept in the dark and prevented from drying, can live for months or even years, depending a good deal on the temperature at which they are kept. Otto was able to demon- strate that the plague bacillus taken from putrid cadavers of plague rats were alive up to sixty-one days, provided they were kept at as low a temperature as 6° C. In bubonic pus it has been found to survive for twenty days, and also for that length of time in fresh water, and in sea water for forty-seven days. In cow-dung it has been found to live for months when kept moist. You may perhaps be aware that the common custom in India is to plaster the sides of the walls and the floors of the houses with moist cow-dung, which is then allowed to dry. On such floors plague bacilli were found to remain infective for forty-eight hours. The resistance of B. pestis to drying is not great. Thus, plague matter dried on wool, silk, linen, glass, wood, &c., died within six days. In dust it rapidly dies out ; and in Hankin’s experiments it was found to die out within thirteen days in grain and meal. Exposure to light causes rapid sterilisation of plague cultures — three to four hours sufficing. Dry heat at 160° C. kills in a minute. Cold, however, has very little effect on plague bacillus. It was found alive after forty days, after the severe treatment of freezing and thawing it daily. Chemical disinfectants rapidly and easily kill it. Thus, 1 per cent, carbolic acid or lysol will kill in ten minutes. Pathogenic Action of B. pestis on Animals. — What happens when this bacillus gets inoculated into animals ? And which are susceptible to its action ? The most susceptible are all kinds of rats, also guinea-pigs, mice, monkeys, mongooses, squirrels, bats, jerboas, and marmots. Dogs, jackals, and hyenas are mostly unsusceptible. Cats are not highly so, and the bird family is immune. 866 SOME POINTS OF GENERAL INTEREST All varieties of rats can be infected with the greatest ease, and Albrecht and Ghon, of the Austrian Plague Commission, have shown that by smearing plague material upon the intact shaven skin of a guinea-pig or rat, infection occurs. This is an important crucial test. Rats can become infected when fed on plague cultures or on plague cadavers, and the German Commission gave it as their opinion that, under natural conditions, rats frequently infect themselves by gnawing the infected cadavers of other rats, and that they can also become infected through the unbroken nasal or conjunctival mucous membrane. Workers in India have also been able to infect rats by feeding them on plague material or on the cadavers of plague rats. It would be natural to conclude that, as the German Commission stated, rats in nature often become infected in this manner. But this is probably not the case, for the common site in naturally infected plague rats of the primary bubo is in the neck, where the fleas usually are most numerous ; whereas in experimentally infected animals, fed on the afore-mentioned material, buboes in the intestinal or mesenteric site are in the great majority. Thus out of 5000 post mortems on naturally infected plague rats not a single mesenteric bubo was seen. Post Mortem Appearances. — Supposing we found a dead rat and suspected it of having died of plague. Apart from the microscope, what post mortem appearances would we expect to find ? Before examining, we would of course take precau- tions to rid the animal of fleas, if indeed these had not already deserted its cold carcass. In dissecting the rat, we would find dark-red, subcutaneous injection of the flaps of the abdominal walls. There would be fluid in the pleural cavity, haemorrhagic swelling in the outer layers of the glands, and swelling of the neck-glands, and, in particular, a creamy mottled appearance of the liver. The spleen also would be found swollen, congested, and granular in appearance. Kinds of Bats. — The kinds of rats that concern us mainly are those which we might describe as the domestic breed as opposed to the field variety. These are Mus rattus , the common house-rat, and Mus decumanus, otherwise known as ON THE SUBJECT OF PLAGUE 867 Mus norvegicus, the sewer -rat. The distinguishing features between these two are that in the sewer-rat, M. norvegicus , we have the ears barely reaching the eyes when laid forward, and the tail rather shorter than the length of the head and body together ; whereas with M. rattus, the house-rat, the extended ear covers or reaches beyond the middle of the eye, it has a sharper nose, thinner ears, and a tail which is longer than the length of the head and body combined by 25 per cent. Mus alexandrinus is a variety of M. rattus, and the common mouse is known as M. musculus. In the United States the ground-squirrel, Cytellus beechyi, acts as a reservoir of plague, and has as its flea Hoplopsyllus anomalus. Insects which have been implicated in the Spread of Plague. — ■ The insects which have been specially studied in connection with plague are flies, bugs, mosquitos, pediculi, ants, cock- roaches, and fleas. They are of more or less importance in this connection in relation to their habits of sucking blood and of transferring themselves from one host to another, as, for example, the flea, biting first the rat and, later, man. Great interest also attaches to the problem as to what becomes of the plague bacillus in the alimentary canal of such insects, and especially whether it is capable of multiplying there, and either increasing or decreasing in virulence. Flies have been proved to be capable of conveying the bacilli on their legs and thus infecting food material. Plague bacilli may remain virulent in the intestines of a fly for forty- eight hours. Bugs were found to have plague bacilli in their intestines after sucking the blood of rats suffering from plague ; but it was found impossible to transmit plague through the agency of these insects. Mosquitos also have not been implicated. Pediculi have been shown by the experiments of de Raadt, in 1915, to be capable of transmitting plague infection. Both the body-louse and the head-louse have been implicated. Ants of a certain species ( Monomorium vastator ) can trans- mit plague, and they often succumb after feeding on plague rats. This is on the authority of Dr. Hankin. 868 SOME POINTS OF GENERAL INTEREST Cockroaches may harbour plague bacilli and carry them for a considerable distance in their intestines. Dr. Hunter was able to recover cultures from faeces of these insects. Fleas, however, are the most important of them all, and on them we shall dwell at somewhat greater length. Life History of Fleas. — These undergo a complete meta- morphosis. The adult female lays eggs, about the size of a small pin’s head, on the ground. These are not attached to the fur of the host. One to five eggs are laid at a time, and they hatch in about two days in such a climate as that of Bombay, and it would probably be about the same time here. A larva issues from the egg, and in this stage the insect is very vulnerable. It has chewing or biting mouth-parts, and lives on any kind of animal or vegetable refuse. These larvae are remarkably long in the case of Pulex cheojpis, the common rat flea, and comparatively short in P. irritans, the human flea. After about a week, the larva becomes sluggish, ceases to eat, and spins a cocoon of white silk-like fibres, and, enclosing itself in this, turns into a pupa. The surface of the cocoon is frequently covered over with dust or other small particles of rubbish which adhere to the fibres. In from seven to fourteen days the pupa turns into the perfect flea — the imago, and escapes from the cocoon. The mouth-parts of the flea are so constructed that it can only live on liquid food. Young fleas, which have just escaped from the cocoon, can live with- out food for one to two weeks, whereas, after having taken their first meal, older fleas die within a week if deprived of food : at least, this was the case in the Bombay experiments, and I will just pause here to warn you that we must be cautious in our generalisations. Things which occur in one country, and under a certain set of conditions, will not necessarily be the same in another land where the conditions are different. The time taken from the egg to the imago or perfect flea is about twenty-one to twenty-two days, under favourable circumstances ; but under unfavourable ones this period may be greatly prolonged. Climatic conditions affect the develop- ment of some species of fleas. Dampness in the surroundings of a breeding-place hinders the development, and adult fleas ON THE SUBJECT OF PLAGUE 869 dislike wetness. The breeding-places differ with the habits of the host. Thus, in the case of Mus decumanus, they are found in the burrows, while in the case of M. rattus in all sorts of situations which afford a shelter to these animals. Thus, on grain or gunny-bags. This, as a source of distribution of the flea, should be kept in mind. Temperature has a decided effect on the breeding of fleas. A high mean temperature of about 90° F. not only seems to restrain the imago from depositing eggs, but is deleterious to the development of the eggs into larvae. There seems to be an optimum temperature at which breeding takes place more vigorously than at other temperatures. In Bombay this appeared to be from 75° to 80° F. A French worker in Annam states that P. cheopis eggs do not hatch there under 55° F., and at 90° F. 75 per cent, of them remain sterile. He states that for the active larvae a certain amount of moisture and a still atmosphere are necessary. Badly ventilated houses, therefore, favour their develop- ment. He also states that the adult flea can live thirty-eight days without food, and that it lives for preference on the body of its host rather than in its bed, which makes it a far more dangerous agent in plague transaction. Fleas have Particular Hosts. — Different kinds of fleas have their own hosts in particular. Thus P. irritans is natural to man as P. cheopis is to the rat. This does not mean that fleas will not attack other animals than their normal hosts : for instance, P. felis has been found on the dog, cat, tiger, panther, goat, horse, rat, hedgehog, kangaroo, deer, guinea- pig, rabbit, monkey, and on man. But, nevertheless, they prefer their natural hosts. P. irritans, the human flea, does not thrive on the rat, and, when artificially placed on such a host, rapidly dies off. Yet it can be infective for at least four days after feeding on a plague rat. In some experiments, however, which were carried out in Bombay, it was found that the guinea-pig is as readily chosen by P. cheopis as its true host the rat. When many rat fleas are present, some of them will attack man, even when a rat is available for their food supply. Conversely, when the number is small, they will not desert their true host 370 SOME POINTS OF GENERAL INTEREST for man. Rat fleas, when starved, will readily attack all animals, not then being particular in their choice of host. Such fleas, deprived of food from seventy-two to ninety-six hours, attack and feed on man more readily than at other times ; but even so, they prefer their true host to man. A point to be remembered is that rat fleas may be attracted to man, jump on him, but take some time to feed ; and thus a man may carry fleas from one place to another which when brought near a rat will attack it in preference to the man. Mode of Dispersal of Fleas. — I have already indicated one way in which fleas may be carried from one vicinity to another. Other modes are : (a) By means of the host in its natural wanderings. The flea does not stay continually on its host, but often drops off ; thus rats are continually shedding fleas wherever they go. Also remember that sick rats harbour a greater number of fleas than healthy ones. (b) When the rat is carried about in merchandise, fleas accompany him. (c) But in the absence of the rat, fleas themselves may travel in merchandise, grain, or clothing. However, it must be remembered that the adult flea, in the absence of any host, will rapidly die, generally in about four or five days. Still, the larvae, which can feed on any kind of organic rubbish, and the pupae, which require no food, can be carried considerable distances in merchandise, i.e. for periods as long as one or two months. In whatever way transported, they will select, when reaching their new surroundings, their true host or the next best available animal. In numbers rat fleas seem to have a distinct seasonal variation. Moreover, this appears to coincide with the height of the plague epidemic more or less closely. Also during the epidemic season of plague, fleas remain infective for a period of from ten to fifteen days ; whereas during the non-epidemic season they are only infective for seven days, and then not so highly. The Collection and Examination of Fleas. — How are we to ON THE SUBJECT OF PLAGUE 371 catch our fleas for examination ? The simplest way is by means of chloroform or Keating’s powder. If collecting from a rat, put the rat into a large glass bottle, on the bottom of which a little chloroform has been dropped. The fleas become anaesthetised and fall to the bottom of the jar, or remain entangled in the hairs. As soon as the rat becomes anaesthetised it may be removed from the bottle and the fleas picked off it and placed in a test-tube. They rapidly recover from the anaesthesia, and can be used for experimental purposes. I would like to warn you here, if you wish to collect fleas from a caged rat, not to expose the cage to sun- light ; for fleas dislike the sun intensely, and will leave their hosts to seek the shade. This is a point also which I would call to the notice of those who are catching rats on a large scale in their endeavours to prevent the spread of plague. If these cages are carried for any distance in the sunlight through a town, they will inevitably be shedding their fleas on the way. Verbum sapienti. Another method is by the use of animal traps. Thus, should we desire to catch P. cheopis in a room, a guinea-pig or man would serve as a suitable trap. The guinea-pig is allowed to wander about the floor of the room and soon gathers up all the fleas present. If the experiment takes place at night in a rat-infested room, the bag will be a large one. Still another method might be by the use of fly-paper ; the ordinary Tanglefoot paper is very suitable. Put a suit- able animal in a cage, and all around it for a depth of at least 6 inches, place Tanglefoot paper. The flea’s maximum hop is about 4J inches. Attracted by the animal, the flea will endeavour to reach it, and be caught in the process. For examination, pick it off and wash in alcohol. Examination of Fleas. — Now, having caught our fleas, we proceed to examine them. Living fleas may best be examined while under the influence of chloroform, and J-inch objective of an ordinary microscope is of sufficient power for identifica- tion. For examining dead specimens, dissection is employed, and, to facilitate this, fleas may be boiled for a short time in caustic soda or potash, or allowed to soak in water at 80° F. for twenty-four to forty-eight hours. The soft parts of the 872 SOME POINTS OP GENERAL INTEREST body disintegrate, allowing the harder shell, which is composed of chitin, to be separated into segments. Microscopical speci- mens can be prepared by boiling the fleas in glycerine and mounting them in that medium, or they may be boiled in alcohol, cleared in cedar -wood oil, and mounted in Canada balsam. The Process by which the Flea transfers Infection. — Now let us introduce the Bacillus pestis into the flea — or rather he will do that himself, given the chance. The average capacity of a rat flea’s stomach is about 0-5 c.mm., and on this basis a flea may take in as many as 5000 plague germs into his stomach after imbibing the blood of a plague rat. Moreover, multiplication of plague bacilli takes place in the stomach of the rat flea. This multiplication varies with the season of the year. In the epidemic season it is six times greater than in the non-epidemic. Plague bacilli are present in the rectum and faeces of fleas taken from plague rats, and such faeces are infective to guinea-pigs, both by the cutaneous and sub- cutaneous methods of inoculation. Plague bacilli have been found in the oesophagus of the flea, but never in any other region of the body, such as the body cavity or salivary glands. A single flea may transmit the disease, and both male and female fleas can do so. Two modes of transmission have been proved. (1) The flea defaecates while he is yet sucking blood, and the rubbing of this material into the puncture brings about infection. (2) Regurgitation, as the result of masses of plague bacilli in the oesophagus of the flea, causes injection of plague bacilli into the rat or man in the act of biting, and this is the more important mode. Experiments showing that the Flea is the Transmitting Agent of Infection. — 1. Certain experiments which wTere carried out in godowns in Bombay went to prove that close and continuous contact of plague-infected animals with healthy animals, if fleas are excluded, does not give rise to an epizootic among the latter. As the godowns were never cleaned out during the experiments, close contact included contact with faeces and urine, and the eating of food contaminated with these materials. ON THE SUBJECT OF PLAGUE 873 2. When fleas are present, the epizootic, if it does start, varies in severity and rate of progress according to the season, and this season corresponds to that of the plague epidemic. 3. An epizootic of plague may occur in a godown con- taining infected fleas without direct contact of healthy animals and infected animals. 4. In an infected godown, the infection is effective in proportion as the test animals are accessible to fleas. 5. Infection can take place without any contact with contaminated soil. 6. The experiments exclude aerial infection. 7. The general conclusion was that fleas, and fleas alone, were the transmitting agents of infection. Construction of Godowns. — These were six in number, and were built in a row. The walls were of brick and mortar and the floor of cement, and so were rat proof. The rooms were of the same size, and the essential difference was in the construction of the roofs. In the case of rooms Nos. 1 and 2, the roof was of ordinary, semi-cylindrical country tiles, which afford a very good harbourage to rats. Immediately underneath this roof was a wire netting, carried on a wooden framework, which was let into the walls with cement ; the idea being to allow a free passage to the rat flea, but not to the rat. The roofs of rooms Nos. 3 and 4 were similar, but built with the flat Mangalore tiles. These, though they afford harbourage to rats, do not do so to anything like the same extent as the country tiles. The third set of two rooms had corrugated iron roofing, which should have been rat and flea proof. As a matter of fact they had to make these last rooms with concrete, as they found that somehow or other fleas did actually find their way through the corrugated iron roof — probably by way of cracks in the cement. The difference in the construction of the roofs is of such a nature that the natural supply of fleas, depending as it does on the number of rats in the roofs, varies in the different 874 SOME POINTS OF GENEEAL INTEREST go-downs. In Nos. 1 and 2 they are abundant and regular, in Nos. 8 and 4 they are scanty, whereas in Nos. 5 and 6 there should be none at all. Differentiation of Fleas. — The fleas which are of main, interest to us are those wTiich are met with in association with P. cheojpis on rats, and from which we might require to differentiate them. These are : — 1. Pulex irritans, the human flea. 2. Ceratophyllus fasciatus, the European rat flea. 3. Pulex felis, the cat and dog flea. 4. Ctenojpsylla musculi, found on mice and rats. 5. Sarcojpsylla gallinacea, a common bird flea. The varieties of fleas are very numerous. Thus, Loemopsylla, to which genus Pulex cheopis belongs, has alone twenty-four known varieties. Still, as P. cheopis comprises 98 per cent, of the flea population on rats in India, and as it was found to comprise 66 per cent, of them during the epidemic of 1909, in the Muanza district, other kinds of fleas on rats, though possibly capable of transmitting plague, are not of such practical importance. ON THE SUBJECT OF PLAGUE 875 In Europe, the common rat-flea is the Ceratophyllusfasciatus , which occupies in point of numbers a similar position to P. cheopis in tropical climates. Now to point out the differences between those I have mentioned. They have all well-developed eyes, with the exception of Ctenopsylla musculi. Of the remainder, Cerato- 'phyllus fasciatus and P. felts have a prothoracic comb of bristles and the latter also a perioral comb, which are absent in P. cheopis, P. irritans, and Sarcopsylla gallinacea. The last named can readily be distinguished from the other two by its angular fish-shaped head and its largely developed mandibles. We are left now with the two important ones, P. cheo'pis and P. irritans. The plague flea has been given various names by workers who thought they were dealing with new varieties. Thus Taraboschi in Italy called it P. murinus, Tidswell in Sydney P. jpallidus, and Herzog in Manila P. jphilijppinensis. Bothschild identified all these as P. cheojpis and P. irritans. These may be differentiated from each other as follows : — cc 876 SOME POINTS OF GENERAL INTEREST 1 . P. cheopis is small and more yellow than brown in colour when compared with P. irritans. It was this feature that led Tidswell in Sydney to call it P. pallidus, thinking he was dealing with a new variety. 2. The number of bristles on the head is greater in P. cheopis than in P. irritans , and the ocular bristle in cheo'pis is situated nearly on a level with the upper border of the eye ; while in P. irritans it arises nearer to the lower margin. 8. The antipygidial bristle in P. cheopis is longer than that of P. irritans . ON THE SUBJECT OF PLAGUE 4. The claws in P. cheopis are small in comparison with those of P. irritans. 5. Two mandibles slenderer and longer than in P. irritans. Another flea which might be noticed in passing is the Pygiosylla ahalae, which is the outdoor flea, commonly found on field-rats. It is said to be unimportant in regard to the spread of plague from one native dwelling to another ; but as the house- and field-rat come into fairly intimate connection 878 SOME POINTS OE GENERAL INTEREST in the neighbourhood of dwellings, this statement needs confirmation. Distinguishing Characters of Genus Loemopsylla 1. Pour segmented labial palps. 2. Closed antennal groove. 8. Anteriorly solid antennal club. 4. Position of dorsal apical bristle of the seventh abdominal tergite remote from the edge of the segment. 5. Presence of short spines on the inner surface of the hind coxa. 6. The division of the rodlike incrassation inside the mid coxa taking place near the base. 7. The structure of the modified abdominal segments. 8. Two mandibles slenderer and longer than in P. irritans. Genus Pulex. Species — one only. P. irritans 1. Chief character, greatly reduced thorax. Mesosternite (i.e. sternum) highly specialised, narrow, strongly oblique, and lacks the internal cariniform incrassation found in other fleas in compensation. The anterior ventral portion of the mesosternite is much strengthened inside. 2. Male genitalia like Sarcopsyllidae, and not in other species, i.e. second and third processes of the claspers form a kind of claw. 8. Bristle in front of eye, in most other species, is absent in P. irritans and is replaced by one below the eye. 4. Hind coxa is distinguished from that of any known flea by bearing a number of hairs on the inner surface of the posterior (meral) portion. 5. Has a small tooth (often absent) at the genal edge of the head slightly behind the lower oral corner. Genal portion — below eye from oral edge to antennal groove. Human Plague We will now shortly glance at plague as it affects man. There are two main forms of plague : Pestis minor , which is a ON THE SUBJECT OF PLAGUE 879 mild form of bubonic plague, and Pestis major , which is a severe form. The latter is further subdivided into a number of varieties. Where the plague bacilli are found chiefly in the glands, we call it bubonic plague ; when in the lungs, pneumonic plague ; when as a general septicaemia, septicaemia plague. But it must be remembered that all forms have a common cause — the Bacillus pestis. Pestis major. — The incubation period in Pestis major is generally from three to five days. It may be more. The onset is sudden, and the patient may suffer from headache, giddiness, and a staggering gait. He soon gets a thick, slow mode of speech. The temperature rapidly rises from 102° to 104° F., and even higher. The pulse also is rapid, from 90 to 180 ; the normal rate being about 75. The temperature lasts two to five days. The lymphatic glands are enlarged, and in the majority of cases those in the groins are affected. The en- largement of these may indeed be the first symptom noted ; but usually they occur after the commencement of the fever. They are very painful, and at first very hard. If they go on to suppuration, it is considered a favourable sign ; but if, on the other hand, rapid softening, flattening, or disappearance of the bubo takes place during the height of the fever, it generally means that the patient will succumb. In severe cases haemorrhage from various organs may be observed : thus bleeding from the nose or lungs may be seen ; and the latter is characteristic of the pneumonia variety of the disease. The duration of the attack is usually from three to five days, in fatal cases ; but it may be very much shorter than this, even to sudden death. The mortality in plague varies. During the height of the epidemic it is often 80 to 90 per cent. In India the general case mortality has been 70 to 85 per cent. In Hong-Kong it was 89 to 96 per cent, in Chinese ; but was less among the Indians there, 77 per cent., and still less among the Japanese, 60 per cent. In South Africa it was 66 per cent, for the coloured population. Among Europeans it appears to be much less : thus in Hong-Kong it was only 34 per cent., in Bombay 30 to 40 per cent., and in Capetown 33 per cent. The percentage of deaths may tail off at the end of the epidemic season. 880 SOME POINTS OF GENERAL INTEREST The pneumonic and septicaemic varieties of plague are extremely fatal forms. Unlike the bubonic form, pneumonic plague can easily be conveyed from one person to another without the intervention of fleas. In coughing, a pneumonic plague patient is showering his neighbourhood with millions of germs, and an attendant accidentally receiving some of this material on the mucous membrane of the mouth, or the conjunctiva of the eyes, may contract plague. The treatment of plague, so far, has not produced any specific. Most drugs have been tried, but with disappointing results. Anti-plague serum has also been disappointing in producing curative results. On the other hand, from a prophylactic point of view, we can produce a very considerable degree of active immunisation by the inoculation of cultures of plague bacilli or its products. The chief of these is Haffkine’s Vaccine, of which all of you have personal acquaintance. Plague bacilli are prepared in broth cultures, in the stalactite forms, to which I called your attention in the earlier part of this paper. These are culti- vated for five to six weeks, and, when pure cultures are obtained they are killed by raising the temperature to 65° C. for an hour, and, as an additional precaution, a small quantity of lysol is added. Of this preparation, from 0*5 to 4 c.c. are injected, according to the age and size of the individual treated. Sus- ceptibility to plague is reduced by about one-fourth, and of those attacked after previous inoculation the mortality is only about one-fourth of what it is among the non- inoculated. We have not yet got accurate knowledge as to the length of time this inoculation confers protection : it probably is for a year, and possibly even longer. Preventive Measures Now that you have learnt what plague is and how it is spread, you will readily see that preventive measures must be directed against the rat and the rat-flea. The United States Public Health Service have outlined such measures in a very succinct form. % ON THE SUBJECT OF PLAGUE 881 Scheme based on the Experience of the U.S. Public Health Service CD a CD o !>s CD S-t 3 TJ1 o 5-i s \Ph (a) Plague laboratory (b) Delineation of in- fected areas fRodent destruction Rat-proofing Fumigation S02, CO, HCN Rat proofing xn