ALVMNVS BOOK FYND BIOLOGY LIBRARY i TSETSE -FLIES THEIR CHARACTERISTICS DISTRIBUTION AND BIONOMICS WITH SOME ACCOUNT OF POSSIBLE METHODS FOR THEIR CONTROL BY MAJOR E. E. AUSTEN, D.S.O., ASSISTANT - KEEPER IN THE DEPARTMENT OF ENTOMOLOGY, BRITISH MUSEUM (NATURAL HISTORY) AND EMILE HEGH, CHEF DE BUREAU AU MINISTERE DES COLONIES DE BELGIQUE WITH A PREFACE BY THE EARL BUXTON, G.C.M.G., CHAIRMAN, HONORARY COMMITTEE OF MANAGEMENT, IMPERIAL BUREAU OF ENTOMOLOGY LONDON : THE IMPERIAL BUREAU OF ENTOMOLOGY, 41, QUEEN'S GATE, LONDON, S.W.7. 1922 Price 7s. Gd. net. All rights reserved. TSETSE -FLIES THEIR CHARACTERISTICS DISTRIBUTION AND BIONOMICS WITH SOME ACCOUNT OF POSSIBLE METHODS FOR THEIR CONTROL BY MAJOR E. E. AUSTEN, D.S.O., ASSISTANT - KEEPER IN THE DEPARTMENT OF ENTOMOLOGY, BRITISH MUSEUM (NATURAL HISTORY) AND EIY1ILE HEGH, DE BUREAU AU MINISTERE DES COLONIES DE BELGIQUE WITH A PREFACE BY THE EARL BUXTON, G.C.M.G., CHAIRMAN, HONORARY COMMITTEE OF MANAGEMENT, IMPERIAL BUREAU OF ENTOMOLOGY LONDON : THE IMPERIAL BUREAU OF ENTOMOLOGY, 41, QUEEN'S GATF, LONDON, S.W.7. 1922 All rights reserved. +, * VH Ill TABLE OF CONTENTS PAGE PREFACE . . . . . . . . . . . . . . . . . . iv LIST OF PLATES . . . . . . . . . . . . . . . . vi LIST OF FIGURES IN THE TEXT . . . . . . . . . . vii INTRODUCTION — Importance of the Study of Tsetse-flies . . . . . . . . 1 Chapter I. GENERAL CHARACTERS OF TSETSE-FLIES — GEO- GRAPHICAL DISTRIBUTION OF TSETSE-FLIES IN GENERAL — CLASSIFICATION OF TSETSE-FLIES . . 4 ,, II. DISTINCTIVE CHARACTERS AND GEOGRAPHICAL DISTRIBUTION OF THE VARIOUS SPECIES OF TSETSE-FLIES 8 III. REPRODUCTION OF TSETSE-FLIES. — Copulation — Gestation — Birth or Extrusion of the Larva — Duration of the Period of Gestation — Larva — Pupation — Pupa — Emergence of the Perfect Insect or Imago — Parthenogenesis — Duration of the Pupal Stage 23 IV. BREEDING PLACES — BREEDING SEASON . . . . 34 ,, V. " FLY-BELTS " OR AREAS IN WHICH THE DIFFERENT SPECIES OF TSETSE ARE FOUND . . 44 VI. DISTRIBUTION OF TSETSE-FLIES IN BELGIAN CONGO 60 ,, VII. POINTS IN BIONOMICS. — Seasonal Migrations — Range of Flight — Travelling Habits — Resting Surfaces — Pro- portion of the Sexes 64 ,, VIII. TSETSE-FLIES AND THEIR FOOD. — Times when Tsetse-flies are most active : the Search for Food — Experiments in the Taking of Food by Tsetse-flies — Feeding and Diges- tion— Study of the Natural Food of Tsetse- flies by Examination of the Contents of the Alimentary Canal — The Drinking of Water by Tsetse-flies — Experiments in Feeding Tsetse-flies with Extravasated Blood . . 78 IX. RELATION OF TSETSE-FLIES TO THE SMALLER ANIMALS . . . . . . . . . . . . 90 X. TSETSE-FLIES AND BIG GAME .. .. .. 94 686686 IV CONTENTS— continued. PAGE Chapter XI. THE EFFECT OF EXTERNAL FACTORS UPON TSETSE-FLIES. — The Effect of a Lowering of the Temperature — The Effect of Raising the Temperature — Mode of Death — Influence of Colours — Influence of Odours . . . . . . . . 107 ,, XII. MEANS OF LIMITING AND DESTROYING TSETSE- FLIES. — Enemies of Tsetse-flies — Predatory Enemies of Tsetse-flies — Predacious Arachnids — Pre- dacious Insects . . . . . . . . 110 ,, XIII. MEANS OF LIMITING AND DESTROYING TSETSE- FLIES (continued}. Parasites of Tsetse-flies — Parasites of Tsetse Pupae — Parasites of the Adult Flies . . 119 „ XIV. PROPHYLACTIC MEASURES. — Clearing — Late Burning — Catching and Systematically Trapping Tsetse-flies — Simple Catching — Trapping and Catching by means of Bird-lime or " Tanglefoot " — Other Methods of Catching and Destroying Tsetse- flies — Artificial Breeding Places — Employ- ment of Tsetse-repellents for the Protection of Stock — Other Precautions for the Pro- tection of Stock — Personal Protection of Human Beings — Destruction of Big Game. . 136 XV. MODE OF COLLECTING, PRESERVING AND STUDY- ING TSETSE-FLIES. — Materials needed for Collecting Tsetse-flies — Catching Tsetse-flies — Method of Collecting Tsetse-flies — Killing Tsetse-flies after being Caught — Pinning Tsetse-flies : their Trans- mission to England — Preservation of Speci- mens in Alcohol — Preservation of Larvae and Pupae — Field Notes — Number of Specimens of Each Species Required — Attacks by Ants or Termites : the Effects of Mould. . . . 152 XVI. KEEPING AND BREEDING TSETSE-FLIES IN CAPTIVITY. — Rearing Tsetse-flies in Laboratories in Europe . . . . . . . . . . . . 158 XVII. DISSECTION OF THE SALIVARY GLANDS OF TSETSE- FLIES 159 ,, XVIII. COLOUR CONVENTIONS TO INDICATE THE DIS- TRIBUTION OF TSETSE-FLIES 161 ,, XIX. SPECIAL POINTS FOR CONSIDERATION IN STUDYING THE BIONOMICS OF ANY SPECIES OF Glossina. — Suggestions for Future Work . . . . 162 XX. BIBLIOGRAPHY OF THE PRINCIPAL PUBLICATIONS CONSULTED . . . . . . . . . . 165 INDEX 179 PREFACE In connection with the economic development of the vast potential resources of Tropical and Subtropical Africa, one of the most important problems to be solved is the control of the commoner species of Tsetse-flies. Their menace to human life as carriers of sleeping sickness is very grave, and requires to be met ; but apart from this their serious adverse effect on the agricultural development of very large areas of fertile land is of even greater immediate importance. The presence of these insects, most of which transmit various forms of trypanosomiasis, a disease that is fatal to domestic animals, constitutes at present an insuperable bar to their use in these regions. These flies are widely distributed throughout Tropical Africa, and when it is realised that probably one-half of Tanganyika Territory is infested by one or other species, and that in Northern Rhodesia and Sierra Leone nearly three-fourths of these countries are similarly infested, it becomes clear that investigations directed towards the discovery of practical means of extirpating this scourge are urgently needed. It is perhaps of interest to recall that the earlier Portuguese explorers were continually baulked in their endeavours to settle on the healthier inland plateaux of East Africa, owing to the fact that their transport animals were inevitably killed off by Tsetse-flies : and, indeed, but for the presence of these insects, the subsequent history of Tropical East Africa would have been very different from what it has been. It is not possible to speak too highly of the amount and quality of the preliminary work on the habits and life-history of Tsetse-flies that has already been accomplished by various observers, frequently conducted at no little personal risk. It is evidently desirable that these investigators should have at their disposal a full summary of our present knowledge concerning these dangerous flies to aid them in their work. But, at present, the results attained are widely scattered throughout various scientific and medical journals of different countries, and are not readily available to workers in the field. Such a resume was published in 1915, in French, by M. Emile Hegh, of the Belgian Colonial Office ; and the Honorary Managing Committee of the Imperial Bureau of Entomology recently decided to publish an English edition of this pamphlet with such additional matter as might be necessary to bring the information up to date. The work was placed in the capable hands of Major E. E. Austen, D.S.O., of the British Museum (Natural History), who is well known as one of the leading authorities on the flies of the genus Glossina, and the present work, with its illuminating Introduction, is the outcome of his labours. VI It is fitting that this publication should be issued under the aegis of the Imperial Bureau of Entomology, for the Committee of that Body have always taken a special interest in furthering research in connection with Tsetse-flies. A few years ago they submitted to the Colonial Office a Report emphasising the need for more experimental work on the life-history of Glossina ; and, indeed, it is only by an intimate knowledge of its life-history that we can ascertain with any certainty how best to attack and defeat its deadly activities. This Report undoubtedly led to an increased interest being taken in the problem by the various African Governments concerned ; and special investigators are now at work in Nigeria, Uganda, Tanganyika Territory, Nyasaland, Rhodesia, and Zululand. It is to be hoped that the present publication may prove to be of real use to those who directly or indirectly are engaged in investigation by supplying them with an epitome of the work that has already been done elsewhere in various directions, by suggesting new lines of investigation, and by stimulating them to test and to amplify the results obtained by others. BUXTON (Chairman of the Committee of the Imperial Bureau of Entomology}. V.I LIST OF PLATES I. Fig. 1. Glossina palpalis, Robineau-Desvoidy, £ (x about 4J). ,, 2. Glossina tachinoides, Westwood, .$ (x about 4|). II. ,, 1. Glossina morsitans, Westwood, $ (x about 41). ,, 2. Glossina pallidipes, Austen, $ (x about 4 J). III. Glossina brevipalpis, Newstead, 'flie celebrated discoveries of Surgeon-Major (now Major-General Sir David) Bruce in Zululand, in 1895-96, resulted in the dissemination of these two diseases by Tsetse-flies being placed beyond doubt. Since the present work is limited to a study of the insects themselves, we cannot here enter into a detailed description of the mechanism of the transmission of trypanosomes by Tsetse-flies. Suffice it to say that the trypanosome that is the cause of nagana exists in the blood of many species of African big game, without apparently doing these animals much harm, and that when a Tsetse-fly bites one of these creatures it sucks up the micro-organisms in question with the blood. The trypanosomes then undergo a developmental cycle in the body of the insect, and when, after the lapse of a certain time, the fly bites a healthy domestic animal, such as a horse, ox, dog, etc., the parasites escape into its blood, multiply there, and set up the disease. In a similar manner the trypanosome of sleeping sickness is transmitted from man to man by certain species of Tsetse-flies. The trypanosomes that are conveyed by Tsetse-flies belong to a number of different species. Their study, which belongs to the domain of medical and veterinary science and cannot be touched upon here, is a most complicated subject, to which many observers are devoting their attention. Excluding races or varieties, nineteen species of Glossina— all confined to the African continent, with the exception of one which also occurs in Arabia (in the Aden Hinterland) — are at present known. Some of these species are very widely distributed and very abundant in many countries ; others are more local, and yet others are very rare and but little known. These nineteen species of Tsetse, while agreeing in essential characters,, differ in details of structure, habitat and habits, as also, which is more important, in their ability to transmit trypanosomes. Until some ten years ago it was believed that Glossina palpalisf Rob.-Desv., and Glossina morsitans, Westw., the two most widely distributed and best known species of their genus, were alone respon- sible, the former for the transmission of sleeping sickness, the latter for that of nagana or Tsetse-fly disease of domestic animals. In reality, however, the matter is much more complicated, since it has been discovered that in Nyasaland and Northern Rhodesia, where G. palpalis. does not occur, G. morsitans transmits a trypanosome known as Trypanosoma rhodesiense, which produces a form of sleeping sickness in man and is very closely allied to T. gambiense, the cause of sleeping sickness in other parts of Africa. Similarly it appears that nagana, or other closely related trypanosomiases of domestic animals, are disseminated by the bites of several kinds of Tsetse-flies, including G. palpalis, so that it may be said that at the present time all the species of Glossina are suspected of conveying trypanosomes.1 The study of Tsetse-flies is of the utmost importance to the future of Tropical Africa, and the discovery of effective means of combating these insects, especially G. palpalis and G. morsitans, would render an immense service to its development. Within the last twenty years entomological science has made great progress as regards knowledge of the specific characters of the various 1 On the other hand it is by no means proved that Tsetse-flies are the only insects capable of infecting an animal with trypanosomiasis (see p. 106). species ot Tsetse, as also of their mode of reproduction and feeding, their habitat, geographical distribution, the extension or contraction of areas in which they are found, etc. A large number of specialists — entomologists, medical men and others — are devoting their energies to the problems presented by these insects, and data, published in technical journals, continue to accumulate. But the greater the progress in scientific knowledge of Tsetse-flies, the clearer it becomes that what is already known amounts to little indeed in comparison with what we have yet to learn, and we are confronted with many questions, on the solution of which depends in some degree the economic future of large regions of Tropical Africa. Thus we know but little at present on the subject of the predacious enemies and parasites of Tsetse-flies, although in undertaking a campaign against these formid- able Diptera a knowledge of such enemies would be of importance. A profound study of the zones infested by the different species of Tsetse at the various seasons of the year ; of the migrations and other movements of these insects ; of the conditions governing their breeding ; of the spots in which they deposit their larvae, etc., might likewise suggest effective methods of destruction or limitation, by modifying the environment of the flies, or by means of other prophylactic measures. Lastly, a study of the various methods of capture and of the influence of external agents — heat, light, humidity, colours and odours — might also be expected to lead to results of practical value. The present little work has been written for the benefit of those who desire to study Tsetse-flies in various parts of Tropical Africa, and,, as has already been indicated, its object is simply to furnish them with information as to the biological facts already ascertained and the points that still require investigation. Since it may be that an apparently insignificant detail will provide the solution so keenly looked forward to, our knowledge of Tsetse- flies should be as exhaustive as possible, and everything pertaining to these Diptera should be observed and noted down. Thus, collections should be made of specimens of male and female flies, and of larvae and pupae ; while investigators should record observations on habitats, extension or contraction in limits of occurrence, method and time of feeding, animals upon which the insects feed, influence of atmospheric factors, local and general geographical distribution, copulation, deposition of larvae, breeding-places, duration of the pupal stage, proportion of the sexes, predacious enemies and parasites, diseases, etc. With regard to the apparent diffuseness of certain of the following chapters, it should be explained that this is due to the fact that state- ' ments by different observers, working in widely separated regions, are often contradictory, and that, in order to be impartial, we have felt it necessary to give a resume of the various opinions. The chapters on the areas occupied by the different species of Tsetse, and on the influence of big game, are cases in point. The attention of those desirous of making investigations in Belgian Congo is specially directed to the question of the distribution of the [various species of Glossina in the different regions of that vast terri- tory. It is important that a map should be prepared without delay, showing as accurately as possible the distribution of these insects in Belgian Congo, since such a map would be of great assistance in connection with the campaign against sleeping sickness and the establishment of cattle-breeding centres. CHAPTER I. GENERAL CHARACTERS OF TSETSE-FLIES.— GEOGRAPHICAL DISTRIBUTION OF TSETSE-FLIES IN GENERAL. — THE CLASSIFICATION OF TSETSE-FLIES. The species of the genus Glossina are dark brown, blackish, yellowish- brown, or yellowish flies, ranging in length (excluding the proboscis and wings) from about 6 or 8 mm. in the case of G. tachinoides, to as much as 13 or 13 -5 mm. in that of G. brevipalpis or G. longipennis. In all species there is a prominent proboscis ; the abdomen in the G. morsitans Group and in G. tachinoides is of a paler colour and marked with sharply denned dark brown bands, interrupted in the middle line, but this part of the body is invisible when the insect is at rest, since it is then concealed beneath the wings (Austen). The blood-sucking habit in Tsetse-flies is common to both sexes.1 " The sexes of Tsetse-flies can readily be distinguished when specimens can be examined, since in the male the external genitalia form a con- spicuous knob-like protuberance (hypopygium) beneath the end of the abdomen, which is absent in the female." (Austen.) How to Distinguish Tsetse from other Flies. ^ It is probable that only those who have suffered from the attacks -of Tsetse-flies can recognise them when on the wing, but in the resting position their identification is easy. In this attitude they can be Fig. 1. — A Tsetse-fly, Glossina longipennis, Corti (Somaliland), in resting: attitude, showing the position of the wings. (Partly diagrammatic, x 3. — Aften Austen. Reproduced by permission of the Trustees of the British Museum.) 1 In most other phlebotomic Diptera the females alone suck blood, while the males are harmless. distinguished from all other blood-sucking Diptera with which confusion could possibly take place (especially from those belonging to the genera Stomoxys, Fig. 2, and Haematopota, Fig. 3, which are most likely to be mistaken for them), by the fact that the wings lie closed Fig. 2. — Stomoxys calcitrans, L., in resting attitude, showing the position of the wings. (Partly diagrammatic. x3. — After Austen. Reproduced by per- mission of the Trustees of the British Museum.) flat over one another down the back, like the blades of a pair of scissors, while the proboscis (i.e., the proboscis ensheathed in the palpi) projects horizontally in front of the head (see Fig. 1)" (Austen, 3).1 "As already mentioned, species of Stomoxys and Haematopota are most likely to be mistaken for Glossina, and apart from these confusion Fig. 3. — Haematopota vittata, Lw. (Tropical Africa), in resting attitude, showing the position of the wings, the markings on which are omitted. (Partly diagrammatic, x 3. — After Austen. Reproduced by permission of the Trustees of the British Museum.) can hardly take place. The members of both genera are greedy blood-suckers, and often torment human beings and domestic animals very greatly. Although Stomoxys also has a prominent proboscis, it 1 With the object of avoiding a multiplicity of footnotes, and constant repetitions, throughout this work references to authorities are indicated by numbers in round brackets, which refer to the bibliography at the end (pp. 165- 178). is not ensheathed in the palpi, and is consequently much more slender in appearance than the proboscis of Glossina. The species of the former genus are usually greyish, greyish-black, or blackish flies, and since their wings, when in the resting position, instead of closing one over the other, diverge at an angle (see Fig. 2), like those of Musca domestica, it is easy to distinguish them from Tsetse-flies. Haematopota, on the other hand, a genus of small horse-flies (Family Tabanidae) abundantly represented in Tropical Africa, resembles Glossina some- what closely when at rest. Although the species of this genus are of much the same elongate shape as Tsetse-flies, their wings in the resting position do not close one over the other, but diverge slightly at the tips and are also somewhat tectiform, i.e., they meet together at the base like the roof of a house (Fig. 3). The antennae, too, afford at further means of distinction. While the antennae of Tsetse- flies, as 'of all other Muscidae, are drooping, those of Haematopota project horizontally in front of the head, and being of some length are readily seen " (Austen). Geographical Distribution of Tsetse- Flies in General. " Although, since the discovery of Glossina tachinoides, Westw., in Southern Arabia, the generalisation that existing Tsetse-flies are con- fined to Africa no longer holds good, it is nevertheless true that the genus Glossina at the present day occurs only in the Ethiopian region,1 inasmuch as the southern portion of Arabia, belongs zoo-geographically to the region in question. In the African continent Tsetse-flies have a very wide distribution in the tropical and sub-tropical zones. Roughly speaking, so far as our present knowledge goes, the northern boundary of the genus may be represented by a line drawn from the mouth of the Senegal River across the middle of Lake Chad to the Nile, just south of the twelfth parallel of north latitude, and thence to the east coast at about 4° N. ; while its southern limit may similarly be shown by tracing a line from the mouth of the Cunene River, the southern boundary of Angola, to the north-eastern extremity of St. Lucia Lake, in Zululand. Within this area Tsetse-flies .... are not found continuously, but are restricted to ' belts ' or ' patches ' of forest, 1 There is, however, scientific proof that, at a remote epoch, Tsetse-flies also existed in North America, for within the last thirty years the fossil remains of no fewer than four species of Glossina have been found in the Miocene shales of Colorado. Professor T. D. A. Cockerell (31), to whom our knowledge of these species is chiefly due, has drawn attention to the interesting theory — originally advanced by Professor H. F. Osborn — that the disappearance of many large mammals from America may have been directly connected with the former presence of Tsetse-flies in that continent. Nowadays, with the exception already mentioned, outside the African continent Tsetse-flies are found only in two islands in the Gulf of Guinea — Fernando P6 and Principe, both of which are, or were until recently, infested with G. palpalis. The occurrence of the fly in the former may well be the result of natural immigration, since this island is only 25 miles from the African coast, In the case of Principe, which is separated from the nearest point on the main- land by a distance of 125 miles, according to Da Costa and his colleagues (41) the insect is said to have been imported at the beginning of the nineteenth century, " at a time when the slave trade and the cattle trade kept alive a frequent movement of shipping between the island and different points on the coast of the Gulf of Guinea, especially the Gaboon." The remarkably successful measures fay means of which this island has practically been freed from the scourge are referred to below (pp. 142-143). bush, banana-plantation, or even reeds " (Austen). It may be added that the cover haunted by the majority of Tsetse-flies " is mainly found in the valleys of rivers and watercourses and on the shores of lakes " (Austen). The Classification of Tsetse- Flies. Two systems of classification have been proposed for the genus Glossina — that of Major E. E. Austen, of the British Museum (Natural History), which is founded upon external characters ; and that of Professor Robert Newstead, F.R.S., of the Liverpool School of Tropical Medicine, based upon the morphological characters of the male genital armature. Austen divides the genus into four groups, which, as originally constituted, are as follows : — The GLOSSINA PALPALIS Group, consisting of G. palpalis, Rob.-Desv., G. caliginea, Austen, G. pallicera, Bigot, and G. tachinoides, Westw. ; the GLOSSINA MORSITANS Group, com- prising G. morsitans, Westw., G. pallidipes, Austen, and G. longipalpis, Wied. ; the GLOSSINA FUSCA Group, formed by G. fusca, Walk., G. fuscipleuris , Austen, G. nigrofusca, Newst., and G. tabaniformis , Westw. ; and, lastly, the GLOSSINA BREVIPALPIS Group, the components of which are G. brevipalpis, Newst., G. medicorum, Austen, and G longipennis, Corti.1 In Newstead's scheme the genus Glossina is divided into only three groups, the original arrangement being : — The GLOSSINA FUSCA Group, comprising G. fusca, Walk., G. nigrofusca, Newst., G. brevipalpis, Newst., and G. longipennis, Corti ; the GLOSSINA PALPALIS Group, composed of G. palpalis, Rob.-Desv., G. maculata, Newst.,2 G. caliginea, Austen, G. tachinoides, Westw., G. fuscipes, Newst.,3 and G. pallicera, Bigot ; and the GLOSSINA MORSITANS GROUP, which includes G. morsitans, Westw., G. submorsitans, Newst.,4 G. pallidipes, Austen, and G. longipalpis, Wied. For the distinctive features of these two classifications, into which we cannot enter here, the reader is referred to Austen's " Handbook " (3) and Newstead's able paper (105), both of which were published in 1911. In 1919 an account of the genital armature of female Tsetse-flies was given by Miss A. M. Evans (51). The structures in question, which had not previously been studied, were found to afford few characters of definite specific value in the case of the G. palpalis and G. morsitans Groups ; but in the remaining two divisions of the genus the conditions met with confirmed the conclusions as to validity of species already based upon other characters. 1 This classification dates from 1911; during the eleven years that have elapsed since its publication, five new species, G. austeni, Newst., G. ziemanni, Griinb., which are additions to the Glossina palpalis Group ; G. severini, Newst., G. schwetzi, Newst. & Evans, and G. haningtoni, Newst. & Evans, have been described, while the name G. fuscipes, Newst., has been found to denote simply the eastern race of G. palpalis. 2 Not now regarded as a valid species. 3 Now considered to be merely the eastern race of G. palpalis, Rob.-Desv. (vide supra, Note J). 4 Held by Austen to be a f or m or race of G. morsitans, Westw. (vide infra, p. 15). CHAPTER II. DISTINCTIVE CHARACTERS AND GEOGRAPHICAL DISTRIBUTION OF THE VARIOUS SPECIES OF TSETSE-FLIES. In this chapter we shall confine ourselves to a consideration of the different species of Glossina, describing the external characters by means of which they may most easily be distinguished, and furnishing some particulars as to their geographical distribution. For the purpose of this short account Austen's work, referred to above, has been followed, with modifications necessitated by later discoveries. GROUP I. Glossina palpalis, Rob.-Desv. (Plate I, Fig. 1). — This species is one of the most important of its genus, since it is the disseminator of Trypanosoma gambiense, the cause of sleeping sickness in West and Central Africa.1 The abdomen in G. palpalis is "clove-brown or blackish-brown ; 2 the thorax usually paler, with dark brown markings on a greyish ground .... first segment " of the abdomen, " and a median triangular area on the second .... buff-coloured or cinereous, the pale triangle continued backwards as a narrow, more or less well- defined median stripe, usually reaching at least as far as the hind margin of the fifth segment ; lateral margins of the segments from the second onwards grey, expanded on the apical angles into triangular markings ; extreme hind margins of the segments from the second to the sixth usually narrowly pale or grey ; seventh segment .... entirely grey .... Femora in typical race more or less mouse- grey, greyish-brown, or dark slate-coloured ; tibiae, extreme tips of femora, and first three joints of front and middle tarsi buff or ochraceous- buff, hind (or middle and hind) tibiae sometimes infuscated in $ ; hind tarsi blackish-brown or clove-brown above ; wings strongly tinged with sepia-brown, but not quite so dark as in G. caliginea, Austen " (Austen. — See Plate I, Fig. 1). " The area occupied by Glossina palpalis," writes Austen, " includes the whole of West Africa, from the mouth of the Senegal River (about 16° N.) to Angola (where the variety wellmani apparently predominates), and extends eastward into the southern Bahr-el-Ghazal," thence reaching the River Omo, which falls into the northern end of Lake Rudolf. " Proceeding southward, the eastern boundary of the species follows the valley of the Nile and includes the eastern shores of Lakes Victoria " (in Kenya Colony and Tanganyika Territory) " and Tan- ganyika "3 (in Tanganyika Territory), " and their affluents ; from the southern end of the latter lake the boundary trends south-west, approxi- mately following the frontier between north-eastern Rhodesia and Belgian Congo, and passing through the Katanga district of the latter country into Angola. G. palpalis appears not to occur on Lake 1 Recent investigations carried out in Nyasaland and Northern Rhodesia have shown that in those regions sleeping sickness, caused by Trypanosoma rhodesiense, is transmitted by Glossina morsitans. 2 In the case of many species of Tsetse-flies the abdomen, if containing blood, becomes greatly darkened in colour. 8 On the western shore of Lake Tanganyika, in the Kasongo district of Belgian Congo, the species also occurs in abundance (Rodhain. — 115). 9 Nyasa," though it is met with in the north-western corner of Rhodesia. It likewise exists in the islands of Fernando Po (Spanish) and Principe (Portuguese), in the Gulf of Guinea. Several races or varieties of G. palpalis, differing but slightly from the typical form, are known : Glossina palpalis var. wellmani, Austen, is " distinguishable from typical G. palpalis, Rob.-Desv., by a peculiar reduction in the markings of the dorsum of the thorax " (Austen). These differences are merely superficial, for G. palpalis wellmani is met with in various localities, in company with the typical form, and in an extensive series of specimens all intermediate forms may be found. Glossina fuscipes, Newstead, at first regarded as a distinct species, is in reality like the variety wellmani, Austen, nothing more than a race of G. palpalis, and is in fact the eastern representative of the typical form. In his classification of 1911 Austen admitted it as a species whose claim to recognition rested on a solitary specimen " taken by Dr. Shircore at Nimule, Nile Province, Uganda." Since that time, in view of the fact that all the specimens of G. palpalis collected in Uganda exhibit the characters attributed to G. fuscipes, it has been concluded that the insects in question represent not a species but a race, presenting no outward differences from the typical form, and distinguishable only by means of minor divergences in the shape of the male genital appendages.1 It is stated by Dr. J. W. Scott Macfie (94) that by far the most abundant Glossina in the Ilorin Province of Northern Nigeria " differs markedly," as regards external characters, from all known Tsetse- flies, but has nevertheless been identified as G. palpalis, Rob.-Desv., by both Newstead and Austen. The insect referred to is a small Glossina, " with abdominal markings practically identical with those of G. tachinoides," the only difference being that the ochraceous-buff or buff coloration of the pale areas " is replaced by a very characteristic grey-blue tint." Dr. Marshall, Director of the Imperial Bureau of Entomology, after examining specimens of this fly, was also of the opinion that there is no specific difference between it and typical G. palpalis, the male genitalia being " practically inseparable." Dr. Scott Macfie, who gave an excellent coloured illustration of this small variety, collected examples of it in all parts of the Ilorin Province, where it was often found associated with typical G. palpalis and with G. tachinoides. It rarely happened that typical G. palpalis alone was met with. In 1918 Dr. J. J. Simpson (144), who first met with this variety "in the upper reaches of the River Gambia," in 1911, and subsequently found " large numbers " of it at Yapi, in the Northern Territories of the Gold Coast, where there were " very few typical G. palpalis," proposed for it the name G. palpalis var. pallida. On the other hand, according to Dr. Roubaud, there are several well- differentiated forms of G. palpalis. The flies of this species found in the damp forest regions of French West Africa (Gaboon, Middle Congo, Lower Ivory Coast) are usually of large size and dark in colour, certain forms on the Lower Sangha being almost black and exhibiting marked differences in the male genitalia. The western forms of G. palpalis occurring in the vicinity of French Sudan are usually smaller and paler, 1 According to Dr. R. E. McConnell (92) all Tsetse-flies regarded as typical G. palpalis to the east of the watershed between the great lakes and the basin of the Nile on the one hand, and the Congo fluvial system on the other, are in reality G. palpalis var. fuscipes, Newst. 10 and Dr. Roubaud, who collected small and very light-coloured specimens at Bani, states that the same forms may be met with to the west of the Niger between Bumba and Kirtashi (13°-14° N. Lat.). The form of G. palpalis found in the valley of the Lower Casamance is inter- mediate between the two extreme types. Roubaud mentions that G. fuscipes — which, as we have already stated, is now recognised as the eastern form of G. palpalis — is represented in the Museum National D'Histoire Naturelle in Paris by a large number of specimens, collected by Cronier on the volcanoes of the Kivu, in Belgian Cone:o, at an altitude of 5,000 to 5,500 ft. Dr. Roubaud considers that great importance attaches to the question of geographical races of Tsetse-flies in connection with the transmission of trypanosomiases. Glossina caliginea, Austen. — The distinctive characters of this species, which in general appearance is very like G. palpalis, are summed up by its author as follows. " Allied to and resembling Glossina palpalis, Rob.-Desv., but browner, and usually somewhat larger ; head distinctly broader, front if anything slightly narrower, at least in <£ ; dark portion of dorsal surface of abdomen browner (dark sepia-brown, instead of clove-brown or blackish-brown), paler area on second segment, instead of being confined to a usually narrow, median triangle, broad and more or less quadrate or irregular in outline, its lateral margins being generally ill-defined ; hypopygium of <£ buff or ochraceous-buff, instead of grey as in typical G. palpalis ; pollinose dorsum of seventh abdominal segment in both sexes often cream-buff, and consequently yellower than in G. palpalis ; extreme hind margins of preceding segments (except posterior angles) never lighter. " Up to the present G. caliginea, which is very rare, has been met with only in Southern Nigeria, Gold Coast (two or three specimens) and Cameroon (H. Gldser). Glossina pallicera, Bigot. — The following diagnosis is given by Austen : " Ground colour of dorsum .of thorax olivaceous or olive- grey, with well-developed, dark sepia-brown, longitudinal markings of usual type ; dorsum of abdomen mummy-brown or dark sepia- brown, first segment and a narrow, median, cuneate area on second (not always extending to hind margin) buff ; in $, portion of second segment immediately adjacent to and on each side of cuneate area, and a corresponding region on third and fourth, or third, fourth and fifth segments, sometimes with exception of hind border in each case, usually cinnamon-coloured, therefore paler than lateral portions of these segments ; third joint of antennae pale (cream-buff to ochraceous buff), clothed with long and fine, pale yellowish hair, forming a con- spicuous fringe on front and hind margins ; hind tarsi dark brown above, last two joints clove-brown." G. pallicera, which exhibits " an exceedingly close resemblance in general appearance to an unusually light-coloured G. caliginea," is " apparently one of the rarest of the Tsetse-flies at present known " (Austen). The " area of distribution " of this West African species " extends, at any rate, from the Sierra Leone Protectorate to French Congo " (Austen). Professor Laveran has stated that in the latter country a specimen of G. pallicera was taken in January or February, 1907, on the Benue River (Mayo-Kabbi region), by Dr. Ducasse. " In Ashanti, according to Dr. Graham, the species appears to be 'extremely local ' ' (Austen). G. pallicera has also been recorded from Cameroon (H. Gldser ; Newstead and Evans, 109a), Liberia, 11 French Ivory Coast and Southern Nigeria, while in Belgian Congo (North Katanga), it was met with on one or two occasions by Dr. Schwetz (132). Glossina tachinoides, Westw. (Plate I, Fig. 2).— " One of the smallest of the known Tsetse-flies. Dorsum of thorax olive-grey or smoke-grey, with dark brown longitudinal markings of usual type, though often indistinct ; ground colour of abdomen ochraceous-buff or buff, dorsum with sharply denned clove-brown or dark brown markings .... second segment with a very conspicuous, square or oblong, pale (ochraceous-buff or buff) area in centre ; hind tarsi dark brown, i.e., all the joints more or less dark, extreme tips of first three joints and extreme bases of second and third joints buff, a larger or smaller area at base of first joint also paler than remainder of joint, especially in $ " (Austen). " G. tachinoides," writes Austen, " has a very wide range in West Africa, where, as for instance in Northern Nigeria, it is often locally abundant. Besides being found from Senegal to French Congo, the species also occurs in the French Sudan (on the shores of Lake Chad and the banks of the Bani and Shari rivers) [and likewise in Cameroon] ; and, although not yet recorded from Belgian Congo or the Uganda or East Africa Protectorates [Kenya Colony], its range evidently extends right across Africa, since G. tachinoides .... has also been met with in Southern Arabia (26«) , thus being, so far as is known, the only existing species of Glossina found outside the African continent and •certain of the islands off the coast. " As regards the French possessions in West Africa, G. tachinoides has been identified by Professor Laveran from Casamance, Marigot de Bayla, Bignona, and Carabane in Senegal ; the Lower Rio Nunez, Boke, and the Tinkisso River in French Guinea ; the banks of the Bani and Shari rivers in the French Sudan, and the Mayo-Kabbi region in French Congo. " The species under consideration is evidently very widely distri- buted along the rivers and streams of Northern Nigeria [Katagum and Bornu Provinces]. " In Nigeria it was found a few years ago by Mr. G. C. Dudgeon that the areas of occurrence of G. tachinoides appeared to alternate with those of G. palpalis .... In Togoland, however, Dr. Zupitza met with both species together on the Oti River (a large tributary of the Volta) , although ' on the German bank of the Volta north of 8° Glossina tachinoides was found and no palpalis.' In the Western Province of Ashanti, in 1910, Dr. A. Kinghorn always found G. tachinoides together with G. palpalis on the Tain and Black Volta rivers, the former species being the commoner ; Dr. Kinghorn believes that both occur on the Volta, ' from its mouth to the French border/ " (Austen). Glossina austeni, Newst. (syn. G. brandoni, Chubb, 1915). — This species, described since the appearance of Austen's " Handbook " and placed by its author, Professor Robert Newstead, F.R.S., in the Glossina morsitans Group owing to the characters exhibited by the male genitalia, presents more external resemblance to G. palpalis and its allies, with which it therefore seems better to associate it. Among the more noticeable characters of G. austeni are its small size and somewhat attenuated shape, the relative narrowness of the head, the presence of an interrupted, but more or less sharply marked, black, longitudinal stripe on each side of the dorsum of the thorax, 12 the shining ochreous or reddish-ochreous coloration of the abdomen, the upper side of which generally exhibits traces of widely interrupted, dark, transverse bands, the dusky tint of the posterior tarsi, and, in the male, certain peculiarities of the genital appendages. The first example of this species (a $) submitted to Professor New- stead for identification was collected by District Commissioner R. P. Filleul, on 17th March 1912, at Alexandra, Gorha, Jubaland, in Kenya Colony. A little later 22 additional specimens (8 #£, 14 $?) of this fly, taken in the same district, were forwarded by Mr. Filleul. In February of the same year (1912), five specimens (2 <$<$, 3 $$) of G. austeni, easily recognisable, in spite of their small size, by the shining strawberry colour of the upper surface of the abdomen, were also captured by Dr. S. A. Neave at Voi and in the vicinity of Witu, Kenya Colony. In the colony in question, according to Dr. Neave, G. austeni appears to be confined to the coastal belt, where it is fairly widely distributed. In 1913 Professor Newstead recorded the receipt of 2 &£ and 10 ?? of- G. austeni, Newst., forwarded by Dr. Morstatt, of Amani, German East Africa (now Tanganyika Territory), and collected in January and February of that year, at Nyussi, Tanga District, being the first examples of this species to be recognised from Tanganyika Territory.1 Since then G. austeni has been reported as likewise occurring in Portu- guese East Africa and (under the synonym G. brandoni, Chubb — 30) in Zululand. G. ziemanni, Griinberg (600). — The following is a resume of the description of this species, published by its author in 1912. Total length (including proboscis) 10-5-11 -5 mm. ; including wings, 14 mm. Ground colour deep blackish-brown, the pollinose covering dirty ashen-grey, tinged with brown. The head shows the same brownish ashen-grey pollinosity as does the entire body, this pollinose covering being visible everywhere except on the deep blackish-brown frontal stripe. The clypeus and antennae are likewise dark brown, except that the third segment of the latter has a light reddish tint ; the fringe of black hair on this segment is short, its length being equal at most to one-sixth of the breadth of the segment. The arista is of the same colour as the third segment, and the branched hairs are black. The proboscis is dark brown, the palpi, in consequence of the thick covering of bristles, appearing black. The dorsum of the thorax has a dense, brownish ashen-grey, pollinose covering, the pleurae exhibiting paler ashen-grey spots, or being definitely ashen-grey without a brownish tint ; the blackish-brown stripes and spots on the dorsum are arranged as in G. palpalis. The scutellum has a fairly intense metallic sheen of a violet tint, very distinctly visible through the dark pollinose covering, a similar sheen being also seen on the basal portion of the wings and between the veins ; the two lateral black scutellar spots are very indistinct. On the abdomen the dark ashen-grey pollinose covering is not so pronounced as that on the thorax. The third to the sixth segments inclusive each exhibit a sharply defined, dead black, interrupted transverse band, these markings being distinctly separated from each other as in G. tachinoides or G. pallidipes, but owing to the dark ground colour much less conspicuous. Viewed at a low angle from 1 We now know that all Tsetse-flies recorded from what was formerly German East Africa under the name of G. tachinoides, Westw., were in reality G. austeni, Newst. 13 behind the entire abdomen displays a faint metallic sheen, showing through the pollinose covering and similar to that on the scutellum. The wings, which are deep blackish-brown, are much darker than in G. palpalis or G. fusca, with more or less distinct dark brown borders to the veins. The legs are entirely black throughout, while the femora and tibae have a pollinose covering of dark ashen-grey ; the tarsi of all three pairs of legs without distinction are uniformly black. In concluding his description, the author of this species writes : — " Glossina ziemanni is at once distinguishable from all its known congeners by its uniformly dark coloration, intensely infuscated wings, and, above all, by its entirely black tarsi. Owing to the latter characteristic, it cannot be included in any of the known groups, but must be regarded as belonging to a group of its own." For the present, however, until the acquisition of further specimens enables an examina- tion of the genital armature to be made, it would seem preferable to treat G. ziemanni as an additional member of the G. palpalis Group, the original diagnostic characters of which are in no way invalidated by its inclusion. The typical series of specimens of this species was received in alcohol, and while still preserved in this way the insects " at once attracted attention owing to their uniformly dark coloration, 'and especially owing to the deep blackish tint of their wings. They were strongly reminiscent," continues Griinberg, " of Glossina palpalis, and at first looked not unlike somewhat dirty examples of this species, although serious doubts as to their true character arose forthwith. After the flies had been dried, it was evident at the first glance that they belonged to a distinct species, essentially different from G. palpalis, as from all others." The only individuals of G. ziemanni so far discovered are 1 £, 5 $?, taken at Mina, Mbam R., Cameroon, 15.i.l912, and forwarded to Berlin by Dr. H. Ziemann. No information as to the bionomics or habitat of the species is yet available. GROUP II. All the Tsetse-flies still to be described under Group headings exhibit the following characters in common : — " Hind tarsi not entirely dark brown, blackish-brown, or black above, but either last two joints alone so coloured, and thus forming a conspicuous contrast with remaining joints, which are entirely pale, or last two joints, viewed from above, conspicuously darker than first two or first three joints " (Austen) . The summary of characters given by Austen for the Glossina morsitans Group, or Group II, consisting of G. morsitans, Westw., G. pallidipes, Austen, and G. longipalpis, Wied., is as follows :— " Upper surface of abdomen distinctly banded, i.e., ground colour pale (drab-grey, buff, or ochraceous-buff), marked with very con- spicuous, dark brown or clove-brown, transverse bands, interrupted in the median line." Glossina morsitans, Westw. (Plate II, Fig. 1). — The following is Austen's diagnosis : — " Dorsum of thorax light grey, olivaceous-grey, or smoke-grey in $, drab-grey in $, the thoracic markings in both sexes incompletely developed and reduced to brownish or mouse- grey longitudinal streaks ; dorsum of abdomen buff to ochraceous- buff (in pinned specimens sometimes mouse-grey or olivaceous owing 14 to post-mortem changes), with a larger or smaller clove-brown blotch (sometimes indistinct or almost wanting) near each basal angle of the second segment, and the third to the sixth segments, inclusive, each with a very conspicuous clove-brown transverse band, interrupted in the median line, not reaching the lateral margins, and not extending beyond the basal three-fourths of the segment, if so far ; legs buff, last two joints of hind tarsi clove-brown or black, last two joints of front tarsi and penultimate joint of middle tarsi conspicuously tipped with clove-brown or dark brown, last joint of middle tarsi entirely dark brown above in typical race, otherwise distal half or third of last joint of middle tarsi alone dark brown or clove-brown, remainder of joint merely brownish or even entirely pale." From the economic point of view Glossina morsitans is, after G. palpalis, the most important of the Tsetse-flies, since not only is it chiefly responsible for the dissemination of Nagana, or Tsetse-fly disease of domestic animals, but, as has already been stated,1 in Nyasaland and Rhodesia it likewise transmits Trypanosoma rhodesiense, the cause of sleeping sickness in those regions, where G. palpalis does not occur. " Glossina morsitans, Westw., once regarded as an exclusively South African species," writes Austen, " is unquestionably the most widely distributed of existing Tsetse-flies, since its range extends from Sene- gambia (about 16° N.) in the north-west, to southern Kordofan (about 12° N.) and southern Abyssinia in the north-east, and thence south- wards to the Bechuanaland Protectorate (Lake Ngami district), the north-eastern Transvaal and Zululand. Although present knowledge is not sufficient to enable us to state whether G. morsitans exists in every country and Protectorate within the limits indicated, the species (besides occurring in the countries already mentioned) is certainly found in : the Gambia, French Guinea, the Gold Coast (Northern Territories), Togoland, Dahomey, Northern Nigeria, Belgian Congo (Katanga District),2 the Bahr-el-Ghazal Province of the Egyptian Sudan, the Uganda Protectorate,3 German East Africa [Tanganyika Territory], the Nyasaland Protectorate, Rhodesia (N.-E., N.-W., and S.), and Portuguese East Africa." According to H. Glaser, no species of the Glossina morsitans Group exists in Cameroon. G. morsitans itself, however, occurs in French Congo, though, according to Jamot (76), there are but few places in the Ubangi-Shari district where the fly is very numerous. Glossina morsitans shares with G. palpalis the distinction of being the Glossina upon which most observations have been made during recent years, and of which the largest number of specimens have been collected. Moreover, although not the first member of its genus to be discovered, " it was to G. morsitans that the name ' Tsetse ' was originally, and indeed until a few years ago exclusively, applied" (Austen), while it is the only species mentioned in the accounts of the earlier African explorers of the nineteenth century, such as Livingstone and others. 1 P. 8, note !. 2 In the Welle district of North-eastern Belgian Congo, where, according to Rodhain (113), G. morsitans is also met with, it occurs only in the extreme north-east, its area of distribution representing the south-western limit of the Sudanese extension of the species. 3 It is stated by Dr. S. A. Neave that Glossina morsitans, Westw., has not yet been shown to occur in Kenya Colony ; in Uganda, it appears to be confined to a little-visited zone, of comparatively limited extent. 15 Several varieties or races of Glossina morsitans, Westw., have been described. " Relying chiefly on certain differences in the internal genital appendages of the male " (Austen), Newstead in 1910 described a form which he termed Glossina submorsitans, regarding it "as a shccics allied to the true G. morsilans, Westw." Austen, however, " after making a minute external examination of .... typical and other specimens of G. morsitans, Westw., .... from many different localities in Africa, and comparing with them the types of the supposed new species .... " considers that " it is impossible I to regard G. submorsitans, Newst., as specifically distinct from G. morsitans, Westw., and that it is in reality nothing more than a form or race of the latter." Austen adds that " G. morsitans is the most j widely distributed of all the Tsetse-flies, and it is, therefore, not j surprising to find that even morphological characters are not absolutely I identical in the case of specimens from localities situated at the opposite i extremities of its area, and separated from one another by some thousands of miles." ' The form submorsitans," continues Austen, | " has hitherto been found in West Africa and the Anglo-Egyptian i Sudan, and it is of course possible that, e.g., a male from Northern | Nigeria belonging to this form might be incapable of mating with a female of the typical race of G. morsitans from the Bechuanaland I Protectorate. Still, if the area of G. morsitans could be traversed from end to end, . . . . it is not unreasonable to suppose that it would be found that the form submorsitans constitutes, as it were, one end of a chain of groups of interbreeding Tsetse-flies, of which the other extremity would be represented by typical Glossina morsitans, Westw."1 Mention may also be made of two varieties of G. morsitans found by Dr. J. O. Shircore among some hundreds of typical representatives of the species collected in the Dowa district, Nyasaland, and described by him (138) under the names G. morsitans, Westw., var. pallida, and G. morsitans, Westw., var. par ado xa. The question of the races or varieties of G. morsitans is perhaps an indication that there is still work to be done before the limits of species within the genus Glossina can be regarded as finally established. Glossina pallidipes, Austen (Plate II, Fig, 2). — The author of this species summarizes its distinctive characters as follows : — " Medium- sized or rather large species, closely resembling the West African G. longipalpis, Wied., but distinguishable owing to the coloration of the last two joints of the front and middle tarsi. Front in <$ very narrow ; dorsum of thorax olive-grey, smoke-grey, or drab-grey, with the usual dark brown, longitudinal markings generally much reduced, broken up, and but little conspicuous ; abdomen buff, dorsum with dark brown or clove-brown bands . . . , ; legs buff, last two joints of hind tarsi clove-brown, front and middle tarsi either entirely pale or, at most, last two joints of front tarsi faintly brownish at the tips, and last joint and distal half of penultimate joint of middle tarsi light brown — never so dark as to form a sharp contrast with the remaining joints." " The area of distribution of G. paUidipes, which is the eastern representative of the West African G longipalpis, Wied., includes the south-eastern and eastern portions of the African continent, and extends 1 All West African specimens of Glossina morsitans really belong to the form submorsitans, though it is only by examining the male genital appendages that the latter can be distinguished from the typical race. 16 from Zululand at any rate to the northern boundary of the East Africa Protectorate " [Kenya Colony] (Austen) ; the species also " occurs in Portuguese East Africa, the Nyasaland Protectorate, north-eastern Rhodesia, German East Africa [Tanganyika Territory], and the Uganda Protectorate," besides being found in the south-eastern portion of Belgian Congo. Its existence in Italian Somaliland, on the middle and lower Shebeli River, was reported by Croveri (35) in 1919. " The occurrence of G. pallidipes in the Mozambique district of Portuguese East Africa is recorded by Professor Laveran, who states, on the authority of M. G. Vasse, that in the region in question G. pallidipes is found at higher altitudes than G. morsitans. In the East Africa Protectorate [Kenya Colony], however, where it is said to be more widely distributed than any of its congeners, G. pallidipes has not been observed at an altitude of more than 5,000 ft. In the Uganda Protectorate, Dr. R. van Someren .... states that the present species occurs, in company with G. morsitans, Westw., in several localities in Toro " (Austen). It may be added that, according to Dr. S. A. Neave, G. pallidipes, Austen, is met with chiefly in the littoral region of East Africa, whence it ascends the valleys of the principal rivers, more especially near the equator. In the west and north-west of Uganda it is found in some numbers, particularly in the Semliki Valley (eastern frontier of Belgian Congo) . Glossina longipalpis, Wied. — Of this Tsetse-fly Austen writes : — " Precisely similar to Glossina pallidipes, Austen, in general appearance, but distinguished by the last two joints of the front and middle tarsi having sharply denned and conspicuous clove-brown tips, and in the male sex by the front not being distinctly narrower at the vertex." According to the same authority, " the authenticated area of distri- bution of Glossina longipalpis," which is " essentially a West African species," extends from Senegal to the Katanga district of Belgian Congo. In 1917 the species was recorded by Bouet and Roubaud (13) as occurring in the Ivory Coast, and Roubaud (1230) has since stated that at the present time G longipalpis is found all along the Ivory Coast Railway, between Dimbokro and Bingerville. He thinks, however, that the existing distribution of the fly in this region is the result of the construction of the line and of the traffic upon it. GROUP III. In the two remaining groups of Tsetse-flies, the upper surface of the abdomen does not exhibit the markings characteristic of the Glossina morsitans Group, that is to say, the abdominal markings do not consist of " interrupted dark bands on a pale ground." The distinctive characters of the Glossina fusca Group (Group III) are given by Austen as : — " Wings fairly dark (dull sepia-coloured), palpi (except in G. tabaniformis, Westw.) long and slender." Glossina fusca, Walk. (Newsteadina fusca, Towns.). — Of this Tsetse-fly, Austen writes : — " Large, dark brown, dark-winged species, with long and slender palpi, and without distinct dark bands on the abdomen, which is lighter at the base ; in general appearance very closely resembling Glossina nigrofusca, Newst., but distinguishable in both sexes by the third joint of the antennae not being clothed with long and fine pale hair, forming a conspicuous fringe down the anterior and posterior edges." 17 According to the author quoted, the area of distribution of G. fusca, Walk., appears to extend " at any rate from the Sierra Leone to the Uganda Protectorates." The species is known to occur in the Sierra Leone Protectorate, Liberia, Gold Coast Protectorate (Ashanti), and Northern and Southern Nigeria, as well as in Uganda ; but up to the date of Austen's "Handbook" (1911) no examples of it had been received at the British Museum (Natural History) from " French Guinea, the French Ivory Coast, Togoland, Dahomey, Cameroon, French Congo, or the Congo Free State " [Belgian Congo] (Austen). Since then, however, its existence in Togoland has been recorded by Glaser ; Roubaud, who also took a specimen on the Upper Sangha, has reported its occurrence on the Lower Ivory Coast ; Kleine (80) has stated that it is fairly frequently met with in Cameroon, where (at Kumbe in 1914) Dr. Fischer captured 1,011 specimens in four weeks ; Jamot (76) has found it abundant on small rivers in French Congo ; and Schwetz (132) has studied its habits and breeding-places in various localities in the Katanga district of Belgian Congo. Dr. S. A. Neave states that G. fusca is not rare in the Toro and Semliki Valley forests (Uganda), while it is said also to be found in the Budongo Forest, in Unyoro. Under the name G. fusca var. congolensis, a new variety of this species was described in 1921 by Newstead and Evans (109). The variety in question, of which fourteen examples, belonging to both sexes, were taken by Dr. Schwetz between Kisengwa and Katompe (situated respectively in the Lomami and Tanganyika Districts of Belgian Congo), while agreeing generally with the typical form in coloration and markings, is stated usually to have " more heavily infuscated " wings and to exhibit certain structural differences in the genital armature of the male and female. Glossina haningtoni, Newst. and Evans. — Described so recently as the spring of the present year (1922), this species is stated by its authors, Professor Robert Newstead and Miss A. M. Evans (109#), to be " Closely related to Glossina fusca," while " differing externally by the relatively much shorter palpi (proboscis), and the slightly more robust appearance. But the most marked morphological differences," we are told, " can be seen only in the genital armature of both sexes. A careful study of these organs at once reveals the strikingly distinctive features of this species, and its affinities with other members of the " Fusca Group ' of tsetse-flies." The authors' diagnosis of G. haningtoni is as follows : — " A large dark-coloured species, with infuscated wings, belonging to the ' Fusca Group.' Hairs of the third antennal segment relatively short. Pro- boscis (palpi) 0.7 to 0.9 mm. shorter than in G. fusca. Width of front in both sexes similar. Harpes of male each with three processes, the distal one angular and emarginate in front. Signum of female with height slightly exceeding width and paired crescentic folds almost continuous behind." The only specimens of this species at present known are the typical series of 2^, 2?$, taken 14th December 1921 in South Cameroon, " in the neighbourhood of Basho, where the ground begins to rise to the north into fly-free mountain-plateaux " (Dr. J. Hanington). Glossina fuscipleuris, Austen. — The describer of this Glossina characterised it as a : — " Dusky species, allied to and resembling G. fusca, Walk., but distinguished by the pleurae being dark grey instead of drab-grey or isabella-coloured, by the hind coxae being mouse-grey instead of buff or greyish-buff, and by the black hair fringing (5979) B 18 the posterior margin of the hind coxae below being distinctly longer and finer." At the time of its description, in 1911, the only known specimen of this species was a solitary <$, taken in the Ituri Forest, north-eastern Belgian Congo. From subsequent investigations, however, it would appear that, although by no means common, the species is very widely distributed, and besides being found in Belgian Congo it is now known also to occur in Uganda and Kenya Colony. In Uganda it was taken on the north-east side of Lake Albert, in 1906 (Dr. A. Hodges), and subsequently in the following localities : — Durini River, Toro, 6.iii.l911, "in forest " (Dr. R. van Someren) ; Daro or Durro Forest Toro, alt. 4,000-4,500 ft., 25-29 x.1911 (Dr. S. A. Neave) ; Buamba Forest, Semliki Valley, alt. 2,300-2,800 ft. (Dr. S. A. Neave). G. fuscipleuris also occurs in Upper Lukuga, Belgian Congo, between the River Lualaba and Lake Tanganyika (55). In Kenya Colony, the species was met with " in some numbers " on the Amala River, Southern Masai Reserve, alt. 5,000 ft., 27.vii. and 2.viii,1913, by the late Captain R. B. Woosnam. It is possible that Newham's record (104) of the occurrence of Glossina fusca in south-east Tanganyika Territory really refers to the present species. Glossina nigrofusca, Newst. — Of this Tsetse Austen writes : — " Resembling Glossina fusca, Walk., especially in the $ sex, but distinguishable inter alia by the third joint of the antennae being clothed with long and fine pale hair, forming a conspicuous fringe down the anterior and posterior edges, the length of the fringe on the anterior edge being equal to from one-half to three-fourths of the width of the third joint/' As regards its distribution, the same authority remarks : — " The few specimens of this species received or recorded up to the present time [1911] show that the range of G. nigrofusca extends at any rate from Ashanti (where, according to Dr. Graham, it appears to be of local rather than general occurrence) to the Congo Free State " (Belgian Congo — Kasongo). The species has since been stated to occur in the Ivory Coast (Bouet and Roubaud. — 13). Glossina tabaniformis, Westw. — " Large, dark brown, dusky- winged species, closely resembling G. nigrofusca, Newst., in general appearance, but distinguished by the palpi (and proboscis) being much shorter, and by the length of the fringe of fine hair on the anterior edge of the third joint of the antennae not exceeding one- third (instead of being equal to from one-half to three-fourths) of the width of the third joint " (Austen). The area of distribution of this species, as at present known, includes the Ivory and Gold Coasts, Southern Nigeria, Cameroon (H. Closer), and French and Belgian Congo. In French Congo, G. tabaniformis was met with by Dr. Roubaud in Sangha, Ogove, Lefini and Alima ; in Belgian Congo a $ was taken at Leopoldville in 1903, and subse- quently two or three other specimens were met with by Dr. Schwetz (132) in north Katanga (see below, p. 56). In October, 1920, Schwetz encountered the species in large numbers in south-western Belgian Congo, in the Kikwit Forest on the Kwilu River. GROUP IV. The special characters of this, the Glossina brevipalpis Group are given by Austen as : — " Wings pale (pale Isabella-coloured, or light drab), palpi short." 19 Glossina brevipalpis, Newst. (Austenina brevipalpis, Towns.) (Plate III). — Austen's diagnosis of this Tsetse-fly is as follows : — " Large species, somewhat resembling G. longipennis, Corti, but distinguished inter alia by the front being narrower, and by the absence of a dark brown ocellar spot, of a brown tip to the proboscis bulb, and of the characteristic dark brown spots on the dorsum of the thorax ; dis- tinguished from G. fusca, Walk., inter alia by the proboscis and palpi being much shorter, the head (especially in the <^) distinctly wider and in both sexes closer to the thorax, the front in the $ relatively narrower, the frontal stripe and general coloration of the dorsum of the thorax and of the wings paler, the usual longitudinal markings on the dorsum of the thorax generally less strongly developed, the size being often larger and the general appearance usually bulkier." Of the abdomen of G. brevipalpis Austen writes: — "Dorsum dark brown, first segment, and also second segment to a greater or lesser extent, paler, ochraceous-buff or cinnamon-coloured, extreme hind margins of second to sixth segments, inclusive, usually pale (cream- buff), posterior angles of these segments drab-grey or smoke-grey, dark brown area occupying third to sixth segments inclusive, some- times distinctly broken up into a series of interrupted transverse bands ; seventh segment in both sexes yellowish-grey or fawn-coloured pollinose ; . . . . third, fourth and fifth segments very short (shorter than corresponding segments in G. fusca)." As to the distribution of this species, Austen wrote in. 1911 : — " Glossina brevipalpis, to which the name G. fusca has hitherto been applied in error; is the common large Tsetse-fly in many parts of south- central and East Africa, but, although found in the Congo Free State [Belgian Congo], it does not, so far as our present knowledge goes, occur in West Africa proper. Reliable records and the data attached to the specimens available for examination show that G. brevipalpis exists in Portuguese East Africa, the Nyasaland Protectorate (where it has a wide distribution, being abundant at certain seasons in and near the Songwe Valley, at the north end of Lake Nyasa, besides occurring in and on the edge of the Elephant Marsh, near the Lower Shire River, to the south of the lake), North-Eastern Rhodesia (where it has hitherto been met with only in the Luangwa Valley, in the vicinity of Hargreaves (Chutika's), and in very small numbers), German East Africa [Tanganyika Territory] (where it is found in large numbers in the vicinity of Amani — at the foot of the Usambara Mountains — and also elsewhere), the (British) East Africa Protectorate [Kenya Colony], the Katanga district of the Congo Free State [Belgian Congo],1 and in Angola." Glossina medicorum, Austen. — The following is the original diagnosis of this species, as published in Austen's " Handbook " (1911) : — " Medium-sized, or in the ? sex (which, so far as may be judged from the three specimens of each sex available for examination, considerably exceeds the $ in size), fairly large species, allied and presenting a superficial resemblance to G. brevipalpis, Newst., but distinguished in both sexes by the wings having the appearance of being uniformly coloured (the upper portion of the anterior transverse vein, the posterior transverse vein, and the portions of the fourth longitudinal vein, that in G. brevipalpis are conspicuously infuscated, 1 According to Gerard (55), it is likewise found in Upper Lukuga, between the R. Lualaba and Lake Tanganyika. 20 are not or scarcely at all infuscated in the present species), and in the $ by well-marked differences in the hypopygium and hectors." " Up to the present time this species of Tsetse has been found only in West Africa (Liberia, Gold Coast and Southern Nigeria) " (Austen, op. cit.). G. medicorum is now also known to occur in the Ivory Coast (Bouet and Roubaud. — 13) and Ashanti. Glossina longipennis, Corti. — This species, originally described in 1895, is characterised by Austen as : — " Resembling G. brevipalpis, Newst., in size and general appearance, but distinguished by the greater width of the front in both sexes, by the ocellar spot being •dark brown or clove-brown, and conspicuous, instead of merely light brownish and inconspicuous, by the proboscis bulb having a sharply denned brown or dark brown tip instead of being uniformly pale yellow (at any rate except on the upper margins), and by the dorsum of the thorax exhibiting, in addition to a pair of admedian spots on the suture itself, four sharply denned, dark brown, more or less oval or elongate spots, arranged in a parallelogram, two in front of and two behind the transverse suture." Of the abdomen, Austen writes : — " Dorsum ochraceous-buff, ochraceous, or tawny-ochraceous, longer hair at base of second segment entirely golden-yellow, third to sixth segments inclusive, each . . . with a dark brown more or less lunate mark on each side at base, widely distant from median line but not always extending quite into basal angle, seventh segment and posterior angles of third to sixth segments, inclusive, yellowish-grey pollinose." According to Austen, " Glossina longipennis is an East African species, the range of which is apparently somewhat restricted, since up to the present time [1911] this Tsetse-fly has been recorded only from Somaliland and the East Africa Protectorate [Kenya Colony]. Since, however, it is found on the line of the Uganda Railway, it is possible that its area also extends into German East Africa [Tanganyika Territory], while to the north it perhaps occurs, as Dr. Andrew Balfour has pointed out, in Southern Abyssinia. Towards its southern boundary the range of G. longipennis overlaps that of G. brevipalpis, Newst." Dr. S. A. Neave states that G. longipennis is the characteristic Tsetse of dry and desert countries, and is confined to north-eastern Africa. It is widely distributed in the low-lying and arid regions in the east and north of Kenya Colony, and very probably also exists in that part of the Uganda Protectorate which is situate to the south-west and west of Lake Rudolf. It appears to be absent from the littoral belt, where the climate is doubtless too humid. The known range of this Tsetse-fly has been defined more precisely by H. H. King and the late Dr. A. J. Chalmers, who state that it extends from about 6° N. Lat. to 4° S. Lat., and from 33° to 47° E. Long. The area comprised within these limits includes Kenya Colony, the south and west of Italian Somaliland, the southern region of Abyssinia, and the south-eastern portion of the Egyptian Sudan. Additional Species. The two following species, both of which have been described since the appearance of the " Handbook," cannot as yet be classified according to Austen's system, and it therefore seems better to characterise them briefly, without attempting in either case to assign the species to a definite Group. 21 Glossina severini, Newst. (\06a. — 1913). — The only specimens of this Tsetse-fly hitherto discovered are two males from Belgian Congo. Besides giving details of the genital armature, accompanied by a figure, Newstead characterises the species as follows :— " Male. — Hind tarsi either uniformly dark brown, or with the first and second segments slightly paler than the rest ; pleurae and hind coxae dusky grey, harpes very small, narrow and irregularly serrated distall y. "Length (two specimens), 10-3 to 10-5 mm. Length of wing, 10-5 mm. " Palpi relatively long and stout. Third segment of the antennae clothed with short hair. Thorax very robust, markings well defined and of the usual type or similar to those in Glossina palpalis ; pleurae dusky grey. Legs with the hind coxae grey or greyish-buff ; tips of front and middle tarsi black ; hind tarsi all dark (almost uniformly so in one example), or with the first and second segment slightly paler than the rest. Abdomen almost unicolorous." The typical specimen, which, Newstead states, " had evidently been preserved in alcohol, so that it is highly probable that the colours may have been slightly changed," was taken on Lake Moero (R. Luvua), in May 1907; the precise locality where the second example was obtained is unknown. Although simple reliance on the coloration of the hind tarsi might lead us to assign G. severini to the G. palpalis Group, it would appear that other characters exhibited by the species afford no warrant for such procedure, and Professor Newstead writes : — "As to the affinities and characteristics of this new Tsetse-fly, one gathers from the two examples that, so far as the colour of the hind coxae and pleurae are concerned, it resembles G. fuscipleuris ; but it has relatively longer and stouter palpi ; and, furthermore, is dis- tinguishable from the latter by the darker and more uniformly coloured hind tarsi. " If we take the structural characters of the male genital armature, we find that it is very closely related to G. medicorum, Aust., from which it may, however, be readily separated by the curious rudimentary harpes, the greater length of the marginal hairs of the editum, and, furthermore, by the greater length of the lateral branches of the hairs of the arista." Nothing is yet known as to the bionomics or habitat of this species, so that it is to be hoped that readers of this work who may visit or be stationed in south-eastern Belgian Congo will be able to supply the necessary information, in addition to obtaining further specimens for study. Glossina schwetzi, Newstead and Evans (109. — 1921). — Apart from details of the "strikingly characteristic" genital armature of both sexes, which is described and figured by the authors, the following are the chief diagnostic characters of this species, as given by Professor Newstead and Miss Evans. " Hairs of the third antennal segment about one-sixth to one- seventh the width of the segment. Wings of the female with the thickened portion of the anterior transverse vein darker in colour than the rest. Harpes of the male divided into three processes, the proximal 22 process short and spine-like. Female with signum of the uterus consisting of a single chitinous plate, the long axis transverse and widest in the distal third. " Male. — Length, 10-11 -9 mm. ; proboscis, 2-3-2-8 mm. ; width of head, 3 -2-3 -3mm.; front at vertex, 0-6-0-7 mm. ; wing, 11-12 -4 mm. " Female. — Length, 12 -5 mm. ; proboscis, 3-1 mm. ; width of head, 3 -5-3 -6 mm. ; front at vertex, 0-75-0 -8 mm. ; wing, 13 -2-13 -6 mm. "Male. — Head with the posterior surface 'mouse-grey' (Austen), antennal cavity pearl-grey, sometimes with a pale vinous tinge ; ocellar spot and frontal stripe unicolorous pale brown ; antennae with the distal two-thirds of the third segment infuscated, the rest pale buff ; outstanding hairs on third segment short, about one-sixth to one- seventh the width of the segment. Proboscis bulb pale translucent buff-yellow, the upper lateral margins brownish or orange-brown, ventral median suture proximally, dusky to orange-brown. Thorax with the usual distinctive colour and markings ; sterno-pleurae more or less infuscated ; scutellar bristles long. A bdomen : Dorsum of first and second segments light brown ; the remaining segments, together with the lateral margins, dark brown (' mummy-brown/ Austen), the distal angles either unicolorous with the rest of the abdomen or slightly paler. Legs : Light or dusky ochraceous-buff ; leg I. with the femora infuscated on the upper half of the inner surface, tips of the last two segments dark brown or black ; leg II. with the tips of the last two segments generally more strongly marked than in I. ; leg III. with the last two segments entirely dark brown or black, paler beneath, proximally. Wings with the thickened portion of the anterior transverse vein scarcely darker in colour than the rest. " Female. — Colour and pattern similar to that of the male ; but the distal segment of the abdomen usually paler." This species, like the foregoing, occurs in Belgian Congo, the original series of specimens having been taken by Dr. J. Schwetz on the River Kwango (on the frontier of Angola), at Kasongo Lunda and in the neighbourhood of Kundi, 24-25. x. and l.xi.1920. According to its discoverer, on descending the river by boat, he " came to a region abounding in this species," and also in G. palpalis. The only other field-note recorded is to the effect that some of the examples of G. schwetzi, of both sexes, were captured on the ground (in the case of two males, on a path) in the evening. After concluding their description of G. schwetzi, the authors write : — " In its general external facies the male of this species bears a striking resemblance to that of Gl. tabaniformis, West. ; the female, on the other hand, owing to the darker colour of the thickened portion of the anterior transverse vein, might easily pass as a specimen of Gl. brevi- palpis, Newst. Both sexes may, however, be readily distinguished from any other members of the ' Fusca Group ' by the strikingly characteristic genital armature. Furthermore, the antenna of Gl. schwetzi can be distinguished from that of Gl. tabaniformis by the shorter fringe of fine hairs on the anterior edge of the third segment." In assigning G. schwetzi, as they do, to Newstead's " Fusca Group," the authors of this species are, of course, perfectly correct ; but since this Group, as has been shown above (see p. 7), is coextensive with the G.fusca and G. brevipalpis Groups (Groups III. and IV.) of Austen's system, which is followed in this book, the species is perforce dealt with here. 23 CHAPTER III. REPRODUCTION OF TSETSE-FLIES. " The mode of reproduction in Tsetse-flies — first discovered by Sir David Bruce, during his classical investigations in Zululand in 1895-96 — is extremely remarkable, since, instead of laying eggs as do the majority of Diptera, the pregnant female produces at each birth a single full-grown larva, which, while retained within the oviduct of the mother, is nourished by the secretion of special glands, and on being born crawls away into some hiding place and immediately changes into a pupa" (Austen). In the following sections we shall pass briefly in review the different stages of reproduction in Tsetse-flies, in so far as it is possible to describe them at the present time, in the light of recent discoveries, made by various investigators, chiefly in Rhodesia. COPULATION.— The act of copulation, in the case of captive G. morsitans. has been described by Dr. Allan Kinghorn (79). " Copu- lation," writes Dr. Kinghorn, " was frequently observed between the captive flies, more often after feeding. During this process the mates are firmly locked together, and refuse to separate even on violent shaking of the tube. It continues for several hours, as a rule, and on a few occasions two flies which coupled in the afternoon were found the following morning still fastened together. Coupling may occur more than once between the same two flies." As regards G. brevi- palpis, Dr. J. B. Davey, writing of the habits of this species in Nyasaland, says : — " One pair was observed in coitu on a tree at midday, the terminal abdominal segment of the male, which was on the back of the female, being curved round the terminal segment of the latter ; at intervals the male made a rapid vibratory movement of the wings : the wings of the female were slightly abducted from the body. They were so firmly locked that one was able to capture and kill them without causing separation." As to the period at which copulation occurs, it is stated by Dr. Roubaud (1190), who observed the behaviour in captivity of G. tachinoides in Dahomey, that, just as in the case of G. palpalis under similar conditions, coupling in these flies takes place immediately after emergence from the pupa. Dr. Roubaud, on the other hand, never witnessed the copulation of G. longipalpis in captivity. It will be gathered from the foregoing that present knowledge as to the act of copulation is still very imperfect, while with regard to certain species we know nothing about the matter at all. GESTATION. — As has already been stated, a pregnant female Tsetse- fly produces at each birth a single fully developed larva, which, while retained within the oviduct of the mother, is nourished by the secretion of special glands. Towards the close of the period of intra-uterine life, the larva almost completely fills the abdomen of the mother, and is visible by transparency through the abdominal wall, as a dirty yellowish-white, oval body, with prominent black lips. The latter already show very distinctly, several days before birth. BIRTH OR EXTRUSION OF THE LARVA. — Dr. Kinghorn (79) states that, in the case of G. morsitans, " The extrusion of the larva is accomplished very quickly, and the maggot commences to crawl about as soon as it is born." 24 On one occasion the deposition of a larva by a female of G. morsitans, form submorsitans , under natural conditions, at Larabanga, in the Northern Territories of the Gold Coast, was witnessed by Dr. J. J. Simpson (144). According to the author in question, the insect " took up her position on the under side of the trunk [of a felled tree], about an inch from the ground, and thence let the larva drop. Within five minutes the larva had penetrated the soil and had disappeared." " According to Stuhlmann's observations on female specimens of G. brevipalpis, kept by him in captivity at Amani, German East Africa [Tanganyika Territory], the act of birth, though occasionally lasting for a quarter of an hour or more, generally takes place very quickly " (Austen). Dr. B. Moiser (100), writing of the behaviour of captive specimens of G. tachinoides in Bornu Province, Northern Nigeria, says : — " While watching a female at rest on the floor of the bottle, I noticed a violent contraction of the abdomen, starting from the thorax, and immediately a white larva was extruded. The fly remained on the spot for a few minutes, and then flew away." DURATION OF THE PERIOD OF GESTATION. — " Observations show that, under normal conditions as regards temperature and humidity, the extrusion of larvae takes place at intervals of from about eight to eleven days " (Austen). The foregoing, however, scarcely holds good with regard to G. morsitans in captivity, for Kinghorn (79) states that : — " Considerable irregularity has been displayed by the females in extruding larvae. After the first one had been born, many of them did not produce a second for a considerable length of time, and this may have been dependent upon the meteorological conditions to some extent. A number of the flies, however, have produced a second larva at an interval of from 14 to 15 days after the birth of the first, and this may be taken to represent the normal period of gestation under the laboratory conditions." According to Croveri (35), in Italian Somaliland females of G. pallidipes, presumably under laboratory conditions, deposit from seven to ten larvae at an average interval of ten to eleven days. Dr. Roubaud (1190), from his observations on G. tachinoides in cap- tivity in Dahomey, states that, at a temperature of 25° C.(77° F.), the larvae of this species are deposited at intervals of eight days. In the case of females of G. longipalpis, likewise observed by Dr. Roubaud, the extrusion of larvae at the same temperature occurred at intervals of about ten days. Reverting to Austen's account (already referred to) of Stuhlmann's observations on the breeding of G. brevipalpis at Amani, Tanganyika Territory, we read : — " The larvae produced by a female kept at a temperature of from 23° to 25° C. (73-4° to 77° F.) were extruded at intervals, which varied with the temperature from ten to twenty-two days, the mean interval being about twelve days ; thus in three months and a half a single female gave birth to eight larvae, two of which, however, were not viable. Throughout the period during which females of G. brevipalpis were kept at Amani (i.e., from the beginning of September, 1905, until the middle of May, 1906), the extrusion of larvae proceeded uninterruptedly at approximately regular intervals, though these were somewhat shorter in the hot than in the cold weather. Stuhlmann consequently sees no reason why in nature the production 25 of larvae should not continue throughout the year, though he suggests that the height of both the wet and dry seasons may be unfavourable to the development of the pupae." LARVA. — We give below a description of the larva of three species of Tsetse-flies, G. palpalis, G. morsitans and G. brevipalpis. An excellent illustrated account of the external characters and internal anatomy of the larva of G. palpalis has been furnished by Dr. E. Roubaud (96). The following details, from which, owing to considerations of space, all reference to internal anatomy is necessarily omitted, are taken from this description. The new-born larva, which measures from 7 to 8 mm. in length by 2 -8 mm. to 3- 5 mm. in breadth, is a yellowish-white, cylindrical maggot, with a deep black posterior extremity bearing the rounded terminal protuberances [i.e., the "tumid lips" (Austen) or "poly- pneustic lobes" (Newstead. — 108)1]. Crawling very actively about, the larva undergoes constant changes of shape as the result of undu- latory contractions ; in this way the anterior region of the body some- times swells out suddenly, so as to act like a battering-ram. Thanks to these special movements, coupled with great plasticity, the creature is able not only to burrow rapidly, but also to insinuate itself into the smallest fissures beneath the actual surface of the soil. According to the author referred to, this mode of progression is a special characteristic of the Glossina maggot, and is not found in other Dipterous larvae. " Including the pseudo-cephalic region and the caudal callosities," continues the French author, " the number of visible segments in the larva of G. palpalis is thirteen. The twelfth or anal segment is scarcely distinguishable on the upper surface, where it is represented by a narrow band, squeezed between the chitinised border of the thirteenth segment and the posterior margin of the eleventh ; on the ventral side it is much more conspicuous, and exhibits in the median line, towards the posterior margin of the segment, a small and very inconspicuous punctiform depression, with a blackish, chitinised margin, which represents the vestigial anus. This orifice, which occupies the same position and exhibits the same degree of reduction as in the larvae of the Diptera Pupipara, was misunderstood by Stuhl- mann, the only author who has hitherto studied the organisation of Glossina larvae in any detail. " The cephalic region, in fully extended larvae, narrows perceptibly from the fourth segment onwards. The head, as in all Dipterous larvae of the acephalous type, is excessively inconspicuous. Lying concealed towards the centre of the second segment, its limits are difficult to distinguish, but it is nevertheless recognisable as being divided into two lobes by a slight anterior and ventral furrow, each of 1 Nine years after the publication of the work of M. Roubaud it was shown by Newstead (108), in a paper of remarkable interest, that these terminal pro- tuberances are respiratory organs, which function during the intra-uterine life of the larva ; and that the "innumerable papillae," which cover the outer surface of the lobes and give them their characteristic appearance, are really minute stigmata or openings communicating with the larval tracheal system. Newstead, who therefore terms these protuberances " polypneustic lobes," shows that closely similar structures are found in the larvae of Hippoboscidae, as well as apparently in other Dipterous larvae, whose identity and systematic position have yet to be determined. In the case of Tsetse-fly larvae, the number of stigmata in each polypneustic lobe is very large — in the larvae of G. palpalis, according to Newstead, probably more than five hundred. 26 these lobes giving rise towards the outer side to a small appendage consisting of a single segment, in the shape of a short and slender, squarely truncate, rigid and chitinised process, of a blackish-brown colour. "The pseudo-cephalic aspect of adult Glossina larvae is therefore closely comparable to that of the larvae of Thrixion studied by Pantel (1898, 1909), and consists in a more advanced reduction of the condition normally met with in the acephalous Muscid larvae. In the young larva the head is much more prominent, and its external conformation is more easily perceptible." " The larva of Glossina morsitans," writes Dr. Kinghorn (79), " resembles those of other species of the genus which have been described. " It is a small, elongated maggot, characterised by the black tumid lips which spring from the last segment. The body is composed of twelve segments,1 and is widest and thickest at the junction of the posterior and middle thirds, whence it tapers more noticeably towards the pointed anterior end. The tumid lips are placed slightly dorsally and rather obliquely to the long axis of the body, so that they are directed backwards and upwards as the larva moves about. In colour it is a dull yellowish-white with a slightly dusky tinge, which becomes more pronounced during the wave-like contractions of the body. When examined under a pocket lens, the integument is seen to be very finely mammillated, and these mammillae have a dark tint, which imparts the general appearance of duskiness mentioned above. " From the first segment two minute black bristles, about 0-5 mm. in length, project. These are parallel to one another, and are situated laterally on either side of the middle line. On magnification they are seen to be club-shaped, and are the representatives of the antennae. Between them lies the mouth, the lips of which are directed dorso- ventrally, and are blackish in colour. A trachea can be traced, on either side, as a retractile tube extending from the base of the tumid lips to about the central portion of the body, while on the mid-ventral line, at the junction of the eleventh and twelfth segments, the anus is represented by a small, round, black spot. "The tumid lips are characterised by a row of short longitudinal striations along the basal portion, from which spring two knob-like prominences, rounded externally, and separated by a comparatively deep sulcus. Internally they are hollowed out to form a small cavity, on either side of which lies one of the spiracles. The prominences are beset with small tubercles, and are marked off into three more or less equal portions by two shallow furrows. In colour the tumid lips are jet-black." As regards the larval stage of Glossina brevipalpis, Austen (3) writes : — " The newly extruded larva, which is of the usual Glossina type and shape, and, as stated by Stuhlmann, measures from 9 to 10 mm. in length by 2 to 3 mm. in width, is, with the exception of the hindmost segment, pale yellowish-white in colour, with the surface of the integument of the first eleven segments of the usual finely granular and shagreen-like texture ; the last or twelfth segment,2 which bears the tumid lips or anal protuberances, is, as in other Glossina larvae, deep black. The anterior portion of this segment is encircled by a xThe actual number of segments is thirteen (vide supra, p. 25). 2 Really the thirteenth segment (cf. supra] . 27 band of longitudinal ridges, and the outer surface of each lip or pro- tuberance is, as usual, granular (though the granulations are not nearly so coarse or sharply denned as in the case of G. palpalis), and bears two more or less clearly marked grooves or furrows, which thus divide it into three prominences. The lips are actually smaller than in G. palpalis and different in shape (not so full and rounded). A dried larva of G. brevipalpis from Amani (Dr. F. Vosseler), measuring 9-5 in length, by 4 -25 mm. in greatest width, shows that the inner surfaces of the tumid lips, instead of being close together and separated by a deep and narrow notch, as in G. palpalis, are divided by a relatively wide and shallow space, something like a wide V." PUPATION. — In his " Notes on the Preliminary Stages of Glossina morsitans, Westw.," from which we have already quoted, Dr. Kinghorn (79) writes : — " The behaviour of the larva differs somewhat with its surroundings. If left in a glass dish it may crawl around for from one to two hours, but sooner or later becomes stationary, retracts its head, assumes the oval puparial form, and proceeds to pupate. The move- ments are worm-like, the contractions passing along the body in waves. From time to time the larva ceases to progress, and after retracting the anterior end slightly, moves its head from side to side and up and down, as though trying to push aside some obstacle. The reason for this becomes apparent as soon as it is placed on dry, powdered earth. In this situation the larva at once commences to burrow down and is lost to sight in a minute or two. When it has gone from one to two centimetres it becomes quiet, and proceeds to pupate as before. During this process the skin darkens gradually, and at the same time hardens to form the pupa-case. These changes are usually completed in from four to five hours. " Apparently the larva is capable of secreting a slightly viscid fluid, for in glass tubes they often adhere to the sides. When this has occurred, pupation soon follows. The purpose of this fluid may perhaps be to gather the earth around the pupa, and may possibly account for its not having been found earlier/' A different explanation of the function of the fluid in question was, however, subsequently given by Lamborn (83). In an account of observations on the larva of G. morsitans, made by him in Nyasaland, this author says : — " As has been already noted by previous observers, the newly born larva is covered with a clear slimy secretion, and a theory has been advanced that the purpose of this is to cause earth to adhere so that the pupae are less readily recognised by scratching animals. This supposition has not seemed to me to be tenable, seeing that one never finds such a covering, however carefully one digs up the pupae ; and, moreover, though a certain amount of dust does cling to the pupa, it is invariably in the form of minute particles only in the neighbourhood of the intersegmental rings. " A more likely supposition has seemed to me that the secretion might have a protective function, the period between the birth of the larva and its burying itself in the ground, during which it crawls on the surface, being probably the most critical in the life-history of the fly. Ants are invariably found in the breeding places, and as they depend so largely on insect food, a series of experiments have been conducted with a view to finding out the attitude of various species towards the larvae." After describing the behaviour of three different species of ants when confronted with Glossina larvae, Dr. Lamborn writes :— " From these results I think that a reasonable inference mav be drawn 28 that the secretion is protective against such ants as are usually found in breeding places. It is probable also that it is of material advantage in protecting the delicate cuticle from injury as the larva makes its way down to its future resting place in the soil." Our faith in Dr. Lamborn's conclusion as to the protective value of the secretion is, however, somewhat shaken by his admission that a small Ponerine ant, Euponera senaarensis, Mayr, " on several occasions uncompromisingly seized the larva and ran off with it." The possi- bility that the true or main purpose of the secretion is to assist partu- rition does not seem to have occurred to any observer ; should this be the case, however, the viscid fluid would naturally be secreted by the oviduct of the mother, and not, as Kinghorn supposes, by the larva itself. It should be added that Simpson, who, in the Gold Coast, tested the behaviour of eight different species of ants towards pupae of G. morsitans form submorsitans, accepts Lamborn's inference quoted above. " After prolonged observation," he writes (144), " I agree with Lamborn that the viscid fluid around the larva is a protection against carnivorous enemies. The ants approached them, but backed away quite hurriedly before even touching them." Similar details to those furnished by Kinghorn, in the case of the larva, of G. morsitans, are given by Stuhlmann, with reference to the larva of G. brevipalpis. " If," writes the German observer, as quoted by Austen, " the new-born larva be placed in a glass dish or on blotting- paper, it crawls about for a time exactly like an ordinary fly-maggot, after which it becomes stationary and soon contracts, its chitinous integument thickens and darkens, and in about three-quarters of an hour it has assumed the appearance of a ' coarctate ' pupa. If, however, the larva be transferred to moderately damp sand, it at once burrows into it, making a straight tunnel ; thus in one case a larva penetrated to a depth of 8- 5 cm. (3Jin.). Under such conditions, from an hour and a quarter to an hour and a half elapsed before the change to the pupal stage was completed. In dry sand a larva did not burrow so deeply, since, as it burrowed, the sand continually fell in, but nevertheless it reached a depth of from 2 to 3 cm. (f to 1£ in.). We may assume that in nature the larvae behave in a similar way ; the fly will deposit its offspring on a spot which is sheltered and slightly damp, and the larva will at once burrow beneath the surface." According to Croveri (35), whose observations were made in Italian Somaliland, the larva of G. pallidipes, on being extruded, immediately seeks shelter, and in the laboratory was found to burrow into sand to a depth of about three-fifths of an inch. PUPA. — The pupa- case, or puparium, of a Tsetse-fly, which is com- monly spoken of as the pupa, although the actual pupa lies within, is more or less regularly ovoid in shape, and varies in colour from dark reddish-brown to clove-brown, or dull blackish-brown, with the exception of the tumid lips, which are jet-black. The integument of the pupa (puparium), like that of the larva, is finely mammillated, or in other words, of a shagreen-like texture. " The tumid lips seen in the larva are equally conspicuous in the pupa, and the shape of the notch between them, in conjunction with the size and shape of the lips themselves, affords a valuable means of identifying and distinguishing pupae belonging to different species " (Austen. — See Fig. 5, p. 30). 29 " The pupa-case of Glossina morsitans," writes Austen, " is smaller than that of G. palpalis1 and neater in appearance ; the tumid lips are also conspicuously smaller, and the notch between them lacks the characteristic keyhole-like constriction exhibited by the notch between the lips of the puparium of the latter species. The tumid lips are smaller than those of any other Glossina -puparium at present known " (cf. Figs. 4 and 5, C and F). Further notes on the pupa of G. morsitans are supplied by Kinghorn (79), who writes : — " The means of the measurements of twenty pupae are as follows, in millimetres, the extremes being given in brackets :— Total length, 5-53 (5 to 6) ; width, 3-2 (3 to 3-75) ; depth, 2-97 (2-75 to 3-25) ; width across tumid lips, 1 -45 (1 to 1 -5). It will be seen from these data that the pupa is slightly compressed in the dorso- ventral diameter. " The larval mouth is represented on the first segment of the pupa as a small black dot, while the anus is similarly marked on the mid- ventral line at the junction of the eleventh and twelfth segments, and is enclosed in a small eye-shaped area marked off by two small furrows, the anterior of which is formed by the dividing line between the two segments. •! 1 [•] B ffl I) Fig. 4. — Pupae (puparia) of six species of Tsetse-flies, dorsal aspect, natural size. — A, G. brevipalpis, Newst. ; B, G. fusca, Walk, (cephalic (lower) extremity missing) ; C, G. morsitans, Westw. ; D, G. pallidipes, Austen ; E, G. austeni, Newst. ; F, G. palpalis, Rob.-Desv. (From photographs of actual specimens.) (Reproduced, by permission of the Trustees of the British Museum, from "A Handbook of the Tsetse-Flies [Genus Glossina]," by E. E. Austen. — London : British Museum (Natural History), 1911.) " The line along which the pupa-case splits to permit of the emergence of the imago is indicated on the anterior portion of the puparium as a small ridge directed laterally. This ridge passes dorsally around the mouth region as a slight curve." We append a few notes on the pupae of certain other species of Glossina (cf. Figs. 4 and 5, p. 30). Writing with reference to G. brevipalpis, Austen says : — " The pupa, which is of the usual black or clove-brown colour, measures about 7-5 to 8mm. in length, by 4 to 4 -5 mm. in greatest breadth ; the shape and dimensions of the tumid lips and the shape of the notch 1 The following are the dimensions and weight of the puparium or pupa-case of G. palpalis, as given by Roubaud (96) : — Length 6-5 to 6-6 mm. ; average width, 3 • 5 mm. ; weight, 0 • 025 grin, to 0 • 028 grm. 30 mm • r B D E F Fig. 5. — Pupae (puparia) of six species of Tsetse-flies, dorsal aspect, x 6 : the same specimens as those represented in Fig. 4. — A, G. brevipalpis, Newst. ; B, G. fusca, Walk, (cephalic (lower) extremity missing) ; C, G. morsitans, Westw. ; D, G. pallidipes, Austen ; E, G. austeni, Newst. ; F, G. palpalis, Rob.-Desv. (From enlargements of photographs of actual specimens.) (Reproduced, by permission of the Trustees of the British Museum, from " A Handbook of the Tsetse-Flies [Genus Glossina]," by E. E. Austen. — London : British Museum (Natural History) , 1911.) 31 between them are, of course, the same as in the larva " (see description on p. 27). The measurements given are based upon an examination of seven specimens from Amani, and since these were " from larvae deposited in captivity," Austen adds "it is possible that in certain cases the dimensions are slightly subnormal." According to Austen, " the pupa of G. fusca measures some 8mm. in length, by about 4-5 mm. in greatest breadth, and consequently agrees in its dimensions with the pupa of G. brevipalpis, Newst. ' The pupae of these two species can, however, at once be distinguished by the shape of the notch between the tumid lips or anal protuberances [op. A and B, Figs. 4 and 5]. If a pupa of each species be placed side by side in the position shown in the figures, that is to say, with the tumid lips directed upwards, and the more convex surface of what is now the upper extremity of the pupa towards the observer, in such a way that the tumid lips are viewed exactly in profile, it will be seen that whereas the notch between the lips of the pupa of G. brevipalpis is relatively wide and shallow, looking something like a wide V, the corresponding notch in the case of the pupa of G. fusca is narrower, deeper, and distinctly U-shaped. The edges of the tumid lips are also sharper in the latter than in the former species, in which the lips have a decidedly blunter and less cleanly cut appearance. The breadth of the posterior extremity of the pupa (as measured by a horizontal line drawn through the tumid lips when the pupa is in the position indicated) is greater in G. fusca than in G. brevipalpis, and in the same position the outer margins of the lips in the former are more convex." " The pupa of G. pallidipes," writes Austen, who examined four specimens from Zululand, " .... is from 6-4 to 7 mm. in length, by 3-4 to 3 -75 mm. in greatest width. The tumid lips, which are from 0-8 to 1 mm. long, and separated by a U-shaped notch (somewhat resembling but with straighter sides and fuller at the bottom than the notch in the case of the pupa of G. fusca, Walk.), are much larger than the lips of the pupa of G. morsitans, while the notch between them is about twice as wide as in the latter. The notch is quite different in shape from that exhibited by the pupa of G. brevipalpis, Newst. (cp. D and A, Fig. 5], and the inner edges of the distal portions of the lips are more rounded and less sharp than in the case of the pupa of either G. brevipalpis or G. fusca " (cp. D, A and B, Fig. 5). Of the pupa of G. austeni, a specimen from Amani (believed at the time to belong to G. tachinoides) was examined by Austen, who states that it is " 5- 5 mm. long, by 3mm. in greatest width." According to Austen, " the most striking characteristic " of the pupa, as also of the larva, of G. austeni " is the relatively large size of the tumid lips, which are actually larger than in the case of G. pallidipes [cp. E and D, Fig. 5], though the perfect insect of the latter is, of course, a much bigger fly . . . ." "In the present species " [G. austeni], adds Austen, " the tumid lips are just under 1 mm. in length, and the notch between them is similar in shape to that between the lips of the pupa of G. morsitans, although not so narrow [cp. E and C, Fig. 5]. The notch is much wider than, and has not the characteristic shape of that exhibited by the pupa of G. palpalis " (cp. E and F, Fig. 5). It is unnecessary to describe the well-known pupa of Glossina palpalis, the characteristic shape of the tumid lips in which, and of the keyhole-like notch between them, is clearly shown in Fig. 5, F ; in the case of all the other species of Tsetse-flies, the pupa has not yet, so far as we are aware, been discovered, so that there are still investiga- tions to be made. When Glossina pupae are found, they should be 32 forwarded for identification to the British Museum (Natural History), and should invariably be accompanied by all information necessary for their determination, including a detailed description of the places where they were unearthed. EMERGENCE OF THE PERFECT INSECT, OR IMAGO. — On this subject Austen remarks : — " Although a Glossina escapes from its pupa-case in precisely the same manner as any other Muscid fly, it may be of interest to quote Stuhlmann's account of the emergence of the imago of G. brevipalpis. ' The fly,' writes Stuhlmann, ' bursts off the cap of the pupa-case, and protruding from the opening its cephalic vesicle,' termed by Dipterists the ptilinum, ' forces its body-fluid into the latter, so that its body becomes sufficiently thin to be able to creep out of the puparium. Immediately after the insect has made its escape the cephalic vesicle is retracted into the interior of the head and never used again. In a short time the young fly draws air into its tracheae, and, as a result of so doing, expands greatly ; the wings, which until now have been folded, become spread out, and the abdomen swells to twice or thrice its original size. The proboscis, which in the pupa was directed backwards, is now stretched out in front, and the chitin everywhere hardens. After from three to five hours it would be impossible to tell from the appearance of the fly that it is still quite young, and although on the first day reluctant to feed, it sucks blood greedily on the second.' ' PARTHENOGENESIS. — " It may be noted that there is some evidence that parthenogenesis may in exceptional circumstances occur in Glossina, since Stuhlmann states that at Amani on two occasions virgin females of G. brevipalpis, bred in captivity, produced fully •developed larvae " (Austen). DURATION OF THE PUPAL STAGE. — Thanks to investigations made in Northern Rhodesia by Dr. Allan Kinghorn and Mr. LI. Lloyd, both •of the Luangwa Sleeping Sickness Commission, we are in possession of a. certain amount of information as to the duration of the pupal period in Glossina morsitans. The first observations covered only " a very short period, from the middle of June to the middle of August." " The pupae," writes Kinghorn (79), " as they were obtained, were placed in glass tubes and left exposed to the laboratory air without any protection. In these circumstances the time which has elapsed from the birth of the larva to the escape of the imago has varied from 47 to 53 days." This, it should be noted, was during the height of the dry season in Northern Rhodesia. According to Lloyd (87), " Flies which emerged from pupae obtained in October did so after a pupation period of 23 days, on an average. The approximate mean temperature to which these pupae were exposed •was 85° F. The shortest pupation period observed was 21 days, whilst the longest at this temperature was 25 days. Of the apparently healthy pupae obtained during this period 49 per cent, died." In a subsequent paper (88), recording the results of his experience with Glossina morsitans in the Luangwa Valley, Northern Rhodesia, Lloyd writes : — " The hot dry season has had a very deleterious effect on the breeding of the flies, as shown not only by a reduction in the numbers of pupae produced, but also by the subsequent death of a 33 large proportion of these. In the hotter months great difficulty was experienced in keeping the stock flies alive." Below is reproduced a table, in which Mr. Lloyd summarises the observations made on all the pupae produced, in the course of his investigations in the Luangwa Valley, from 27th June 1911 to 15th March 1912. Table showing the duration of the pupation period of G. morsitans at various temperatures (altitude 2,000 ft.). Period during which the pupae were obtained. Number of appar- ently healthy pupae. Number of pupae from which flies emerged. Average duration of the pupation period in days. Mean tempera- ture in labora- tory. Approxi- mate mean tempera- ture to which the pupae were exposed. Relative humidity of the outside air. June 27-30 3 3 51-0 67° 48-6 July 1-15 6 5 46-6 64-1° 70° > 45 • 7 „ 16-31 24 18 38-7 68-3° 71° " TrO / Aug. 1-15 21 9 35-5 69-4° 73° > 35 -8 16-31 5(?) 1 33-0 73-0° 76° " OO O Sept. 1-15 2 2 25-0 72-7° 82° 01 . C* 16-30 14 12 24-5 80-7° 83° > 0 1 O Oct. 1-15 22 15 22-9 82-9° 85° 16-31 20 10 23-3 86-4° 85° - 31-8* Nov. 1-15 17 6 24-2 86-0° 83° „ 16-30 25 16 25-9 81-5° 80° • 41-1 Dec. 1-15 27 23 27-1 80-6° 79° /"»n -i 16-31 13 12 27-6 78-8° 78° • 69- 1 Jan. 1-15 20 16 27-6 79-0° 78° 77 .7 16-31 31 28 29-1 78-0° 77° r Feb. 1-15 48 32 31-0 77-2° 77° 16-29 40 30 30-1 76-9° 77° > 73-8 Mar. 1-15 — — — 78-8° — 62-5 * Approximate. Under the heading " Breeding," Mr. Lloyd writes : — " Abortions were most numerous during September and October, while in the cooler months there were not many. The number of stock flies was roughly constant, the small number of pupae obtained during September being due to the raid of ants . . . [an incursion of driver-ants, which raided the laboratory and killed every fly in the bottles]. Flies freshly caught do not breed so well as those which are accustomed to captivity The table shows that as the temperature rose from 67° to 85° the duration of the pupal life was reduced from ol to 23 days Without direct experiment it would be impossible to estimate the relative influence of temperature and humidity. Since, however, the humidity of the air to which the pupae of November 1-15 were exposed was greater than that during the pupation period of the October groups, it would appear to be the high temperature rather than the low relative humidity that caused the death of the pupae." Under laboratory conditions in Italian Somaliland, the average length of the pupal stage of G. pallidipes, according to Croveri (35), is 28 days. Some pupae of this species, kept in a room in which the afternoon temperature exceeded 85° F., gave rise to adults in a shorter time. (5979) c 34 As regards species of Glossina other than the foregoing, Austen, writing of G. brevipalpis, remarks : — " At Amani, the duration of the pupal stage was found by Stuhlmann to be from 30 to 65 days, according to the temperature ; when pupae were kept in the breeding-cage at^30°C. (86° F.) the flies emerged on the average in about 36 days." In Stuhlmann's opinion the pupal stage lasts somewhat longer in G. brevipalpis than in G. austeni. On the other hand, Dr. Roubaud (119) in Dahomey, studying G. longipalpis in captivity, found that the duration of the pupal stage in this Tsetse-fly,. at a mean temperature of 24°-25° C. (75-2°-77° F.) varied from 26 to 35 days. CHAPTER IV. BREEDING PLACES — BREEDING SEASON. BREEDING PLACES. — Existing knowledge on the subject of the breeding places of Tsetse-flies relates chiefly to the two best-known species : Glossina palpalis, the disseminator of sleeping sickness in West and Central Africa, and G. morsitans, the carrier of the Rhodesian form of sleeping sickness and of nagana of domestic animals. G. palpalis.— The larva of this species is deposited by the female in a shady place, usually, though not invariably, near water, * where the soil is light and sandy, moderately dry, and often composed of vegetable debris. According to Simpson (144), in the Gold Coast the breeding places of G. palpalis are very similar to those of G. tachinoides, the most common situation in both cases being in the decaying humus beneath overhanging trees, where the sun seldom or never penetrates, and the ground is never really dry. In Uganda and the Katanga district of Belgian Congo, pupae of G. palpalis have been found in large numbers in sandy spots ; and Dr. H. L. Duke (47), from investigations made by himself and others on islands in Lake Victoria, concludes that the chief characteristics of the favourite breeding sites of this species are shade and free air circulation, accompanied by dry and loose soil, generally gravel or coarse sand. Fiske (54), however, while stating that the latter are frequently chosen, has shown that fine, dry vegetable debris serves as well. In the Congo pupae of this species have also been met with on trees (in the forks of branches beneath accumulations of earth or moss, in clefts in the bark and elsewhere), up to a height of three and a half metres from the ground. After a number of unsuccessful attempts, Dr. Bequaert (116), a member of the Mission Scientifique du Katanga, succeeded in dis- covering in 1911 two breeding places of G. palpalis near Bukama, on the Lualaba. The first of these, at the Kalengwe rapids on the river just referred to, 12J miles north-west of Sankisia, yielded at the initial examination on 4th August 1911, 21 empty Glossina pupa-cases and three living pupae ; the latter produced two G. palpalis and one ^ G. morsitans. The pupae were found shallowly buried in light sand (a sand-bank submerged during floods, but high and dry in the dry season) collected between large roots at the base of a tree trunk, and very dry JIn Uganda, Fiske (54) found a breeding place of G. palpalis at a distance of 0 yards from the margin of Lake Victoria; and quite recently Dr. Schwetz (136 a) has recorded his discovery of a living pupa and four emptv puparia of this species in a thicket 1,500 metres (1,640 yards) from water in 'the Rwanda district of Belgian Congo. 35 and well shaded. A second visit to the same spot procured a good number of empty puparia, and three living pupae (G. palpalis). The other breeding place, which was visited more frequently, was situated on a small island in a bend of the Lualaba, above Bukama. The islet in question is almost entirely covered with dense forest growth, ! including all the trees characteristic of the belt of forest fringing this portion of the Lualaba, with, in addition, numerous lianas and luxuriant bushes, the interlacing of which produces a very dense shade. Rather more than 400 living pupae of G. palpalis were collected on I this island, and yielded 265 flies— 135 $ $, and 130 $ & so that the ! two sexes were represented by an equal number of individuals. In all cases the pupae were found at a very slight depth (three to four , centimetres beneath the surface), at high-water mark, in light, dry sand, which was well shaded and usually covered with a layer of dead leaves. At each visit to the spot in question there were unearthed, in the space of about a square metre, from 50 to 80 living pupae, besides a considerable number of empty puparia, showing that the flies continued > to deposit their larvae regularly throughout the dry season, while , during the same period the hatching of batches of adults proceeded in regular succession in the laboratory. Of G. austeni, Swynnerton (145), in North Mossurise, Portuguese East Africa, found some sixty puparia in all under four prostrate trees or logs " in two localities on the Mtshanedzi, near its junction with the | Buzi." With reference to one find, this author writes that " beside the Buzi, below the junction of the Mtshanedzi, we came on a great up- i rooted tree 3J feet thick, lying head downwards in a donga, and beside it the rootlings of pigs. Under it at a point at which it was seven inches from the ground were found nine empty puparia of Glossina austeni, in sandy soil." As to the other three logs under which puparia of G. austeni were discovered, Swynnerton says : — " One, more or less rotten — with only two puparia — was beside a game path used by bushbucks and waterbucks. Another with 30, showed many blue duiker ' forms ' about it, but these were too recent ; leaf-carpet does not preserve traces well. The third, with 23 puparia, was also in dense shade, and exactly at the junction of the primary-type forest and some heavy Markhamia bush (mubfeya). Right up to the log, in the latter bush, were the old tramplings and lying places of elephants. A similar log in similar shade lay 18 yards away ; it was away from the tramplings, etc., and covered no puparia." G. morsitans. — Between 1910 and 1914, search for pupae of (z. morsitans under natural conditions was carried on with success, both by Mr. Rupert W. Jack, Government Entomologist for Southern Rhodesia, and by Mr. LI. Lloyd, Entomologist to the British South Africa Company in Northern Rhodesia, and to the Luangwa Sleeping Sickness Commission. Since the females of G. morsitans, like those of all other Tsetse-flies, produce but a single larva at a time, they are by no means prolific as compared with the vast majority of other insects, and, as we have already seen in the case of G. palpalis, in nature their pupae are so well concealed that it was not until after minute search had been made that specimens were discovered. The honour of having found the first puparium of Glossina morsitans under natural conditions rests with Mr. R. W. Jack, to whom reference has been made above. According to Mr. Jack (67), the puparium in question " was taken in November, 1910, from the collec- tion of mould and humus between the roots of a large tree (Ficus sp.) 36 near the Gorai River below the escarpment in the Lomagundi district, Southern Rhodesia. The tree is situated about 60 to 70 yards from the actual bank of the river, as it runs at present At the time of the visit ' fly ' was abundant in the shade of the trees bordering the river, the sandy bed of which then contained no water on the surface, the nearest open water being 12 to 15 miles away in the Hunyani River. It was thought at the time that this was the only pupa-case secured, and a search all along the river bank produced no others. A number of empty pupa-cases of other Diptera and Lepidoptera were, however, placed in a corked phial and labelled, and on going over these later in the laboratory, the anal end of a second tsetse pupa- case was found amongst them, showing the characteristic protuber- ances. This spot was again visited in April, 191 1, and under the same tree five more empty puparia were found, but, as before, the most diligent search elsewhere failed to produce any more." The second discovery of pupae of G. morsitans was made by the Mission Scientifique du Katanga ; as already stated, a pupa was found almost on the surface of the ground, near the Kalengwe Rapids, on the Lualaba River. In his paper, from which we have just given an extract, Mr. Jack continues: — "During August, 1911, the fly-area in the Sebungwe district [Southern Rhodesia] was visited, and here the search for pupae met with greater success. Altogether 87 empty puparia were found, and four living pupae, or at least puparia from which the fly had not emerged. These puparia were found in two different localities, namely, at a river .... called Manzituba, lying some 35 miles east of Kariyangwe (Sebungwe) .... and again, near a kraal belonging to the induna Sinombi, situated on the Sinyama River, about 23 miles east of Kariyangwe. " At the point where the investigations took place at Manzituba there is a sandy escarpment, a hundred feet or so in height, which suddenly descends to a wet boggy vlei,1 lying east and west, through which the course of the river runs. In August the river consists of a series of pools. This vlei is bordered by shady trees, and in this respect differs from the surrounding bush, in which the trees are almost entirely leafless at this time of the year. From the drier portions of the vlei old termite-mounds rise in places and are surmounted by tall, shady trees. About the shady borders of this vlei tsetse occur in the greatest numbers in August, and at the bases of the trees, where conditions were favourable, the puparia were found. In every case the puparia were in the soil, either sheltered by a hollow in the trunk of a tree, or, when there was a very sharp incline, as on an ' ant-heap/ under the exposed roots. Generally the soil was sandy and soft, and often much humus, dead leaves and vegetable debris were present. One or two puparia were, however, found in hard soil even as deep as 1 J in. As it is not to be thought that the larvae could have penetrated baked ' ant-heap ' to this depth, one can only conjecture that the cases were very old, and that the larvae entered by a crack or division between the soil and the trunk of the tree, not apparent at the time of digging. " At Sinombi. the Sinyama River is met by a tributary stream, and the valleys run north, south and west. Along these valleys and on 1 According to Dr. G. A. K. Marshall, C.M.G., vlei is : — " A term used in South Africa for open, low-lying land, which is always marshy during the rains, but which may or may not be dry during the dry season." 37 the actual banks of the streams a number of fine shady trees are found which are in full leaf in August, and here the tsetses congregate. On the hills around, the forest is leafless and affords but little shade. Puparia, for the most part empty, were found at the bases of the shady trees, mostly on sharply sloping ground, but not invariably. " Searching in the leaf mould, etc., beneath bushes on the banks of the river, near the trees where the pupae were found, again proved fruitless at Sinombi, and this, coupled with the results at Manzituba and the very prolonged searches made in such situations near the Gorai River, Lomagundi, in November and April, leads me to the opinion , that such spots are rarely, if ever, selected for the deposition of the 'larvae. The tsetse-fly is such a comparatively slow breeder that it can scarcely afford to expose its pupae to the scratchings of the game birds which frequent exactly the same haunts as the fly during the dry weather, and often in amazing numbers. Along the Gorai River in .November the numbers of guinea fowl, 'pheasant' (Pternistes) and ^'redwing' (Francolinus) were astonishing. They rose at almost every step along the banks, and all the ground under the bushes had been scratched over and over again." On the other hand, Professor Newstead, F.R.S. (107), writing on the bionomics of Glossina morsitans, as observed by himself and i Dr. J. B. Davey in 191 1 in Nyasaland, in a district near the Shire River, about four miles south of Lake Malombe, states : — " The pupae of Glossina morsitans were found in four different parts of the forest, but all of them occurred in ' sanya country '* . . . . The first pupa was found at the foot of a sanya tree (Copaifera mopani). It was buried in the soil about three-quarters of an inch below the surface and was lying so that it almost touched the bark of the tree. The soil was decidedly hard, so hard, in fact, that it was not easily broken into small fragments with a trowel .... The second pupa of G. morsitans was found on the same day ; and, like the first, was unfortunately empty. In this instance it was lying at the foot of a large tree (apparently a species of Erythrina) , or, rather, between the tree trunk and a very small termites' nest, being buried in the earth which had accumulated between the tree and the nest to a depth of about four inches from the surface The third pupa, a living one, was found just below the surface in a small quantity of earth which had accumulated in the cavity of a large ebony tree on a level with the surrounding surface soil. Here the soil, though rather hard, was of a much more friable nature than that in which the pupae were found on the previous occasions The fourth and also perfect pupa was found in loose earth, which had accumulated in a deep recess formed by the buttressed roots of a baobab tree (Adansonia digitata), facing and exposed to the south-west. Sunlight would not, however, be able to enter the recess until the late afternoon, if at all From these observations we are of opinion that the breeding grounds of Glossina morsitans in this region are thinly scattered over the whole of the tsetse country, and it is quite evident that the pupae, whether living or dead, do not occur in large numbers in any given spot. In localities where the flies are much more numerous than they are in 1 " The tree which preponderates in this region is the Sanya or Iron- wood tree (Copaifera mopani, Kirk) ; in fact it is so abundant that those portions of the forest in which it abounds are referred to as ' Sanya country,' both by natives and Europeans " (Newstead, loc. cit.}. 38 the Upper Shire Valley, larger numbers of pupae would probably exist in similar habitats." More extensive investigations upon the distribution of breeding places of Glossina morsitans were made in Northern Rhodesia ("at Ngoa, in the Mpika division, on the high ground of the Congo-Zambesi watershed ") between January and August 1913, by Mr. LI. Lloyd (89), who summarises his very important results as follows :— " Searches for the pupae in nature have resulted in the finding of 735 living pupae and 1,500 empty cases in 189 positions. A summary of these gives the following results : — " (a) In 40 instances the pupae were found in hollows in trees, 75 pupae and 350 cases being taken. The hollows are at varying heights from the ground up to six feet, and are sometimes filled with very hard clay and sometimes with soft soil, dead leaves, the droppings of insects, and the stomach-castings of birds. When the surface is hard the living pupae are found in crevices or quite exposed on the surface. " (6) In 30 instances pupae were found below trees or branches which slope at an angle or run parallel to the ground before rising. In such positions 129 pupae and 197 cases were taken. The ground was usually very hard, and the pupae were taken on or near the surface or in cracks. " (c) Beneath fallen dead trees or branches 100 positions yielded 493 pupae and 759 cases [see Plate V, Fig. 1] ; these include the most important breeding places. In such places accumulations of dead leaves and twigs usually occur, and the pupae are found among these. When they are absent, the effort to burrow is great, and pupae have been taken at a distance of 18 inches from the shelter of the tree. The trunks were often raised one or two feet above the ground, so there was ample room for any insectivorous animal to search beneath them. " (d) Pupae were found in six instances in the chambers of termite nests in rotten upright stumps ; 19 pupae and 61 cases were found in these positions. In several cases the termites were still living in the nests. " (e) In ten instances they were found in the burrows of various animals, 14 pupae and 99 cases being taken These are usually the holes of bush-pig or wart-hog, but pupae have also been found in smaller burrows. " (/) Pupae were taken three times in the hollows in old termite mounds excavated by the large mammals for the sake of the salt ; five pupae and one case were found in these. " The larvae are thus not always dropped in places where they can hide in the earth or under debris, nor are they always placed in positions where scratching animals could not find them. It is a common feature to find the pupae in places where they are daily warmed by the sun for some hours. This shortens the pupation period, which is dependent on temperature. The one feature which is common to all the breeding places found is that above them there is always some relatively dark spot in which the female fly may rest concealed during pregnancy. It is believed that this, rather than any care for the offspring, is what guides the mother fly in the selection of breeding spots." Mr. Lloyd states that, in the investigations the results of which are summarised in the foregoing paragraphs, " special attention " was paid "to an area of about two square miles," which was " repeatedly and systematically searched for pupae." In this area 174 breeding places were found, and the plan published with the author's paper indicates their distribution in relation to native and game paths, fords, drinking 39 places and salt licks. While 123 of the breeding places, yielding " less than ten pupae each," were " scattered generally through the bush with no special relation to the paths," the spots in which pupae were found in by far the greatest numbers showed unmistakable relationship to the latter. " Over 50 and under 150 pupae," writes Mr. Lloyd, " were found in 13 positions which yielded 892 pupae, an average of 68 • 5 per position ; these positions bear a very close relation to the paths, nine of them being within a yard, and the remaining four being less than 100 yards from a path." The following is an extract from a summary of a report (50) by another entomologist (the late 2nd Lt. R. A. F. Eminson, K.R.R.) upon work done by him in investigating the bionomics of Glossina morsitans in Northern Rhodesia, from May to July 1914 :: — " In the course of his remarks on breeding places Mr. Eminson states that he has not yet succeeded in ascertaining definitely why certain spots are specially preferred by G. morsitans for breeding purposes ; but two negative points are noted, namely, that in country otherwise suitable, no favoured breeding places have been found in any localities in which there was any depth of sandy soil, or in which there was a dense growth of long grass. Although a search was made for pupae over a considerable area, the great majority were found within a comparatively restricted space. This favoured area is described as being ' uniformly covered with forest trees ; very little grass grows amongst the trees, and that little is short ; the soil is of a sandy, gravelly nature and very thin, merely covering the underlying granite which crops out in places. The surface of the ground shows a gentle slope in a north-westerly direction to a vlei [open, moist, low-lying land] in which a little water stands in the wet season, but which is completely dry at this season [mid- winter] . I cannot say that I have noticed much spoor of game, except some impala and wart-hog ; certainly I should not say that there was any more game than in the surrounding country, if as much. The breeding places which yielded the greatest number of pupae and empty pupa-cases were situated near the path ; the fly had evidently been feeding on game, or more probably human beings, preparatory to depositing their larvae/ " It is noted that many of the logs under which large numbers of pupae were found were devoid of bark, and, in Mr. Eminson's opinion, the female tsetses prefer a smooth barkless log on which to rest before depositing their larvae. They also show a preference for a log which in some part of its length is raised a few inches above the ground, thus affording a shady resting place ; a point which has already been emphasised by Mr. Llewellyn Lloyd." Writing in 1915 on G. morsitans breeding places in Nyasaland, Dr. Lamborn (83) says : — " Until April the search for breeding places had been almost without result, five pupae only having been found in hard clayey soil. But since then they have been obtained in large numbers, as the result of realising that the condition of the soil is one of the all- important factors in determining whether or not a particular spot will be selected by the female. " As pointed out by Mr. Lloyd, the fly seeks some relatively dark and shady spot. Many such are to be found in a tsetse area, mostly under fallen trees ; but in my experience, which is now considerable, being based on an examination of 512 breeding places, few, if any, pupae will be found unless, in conjunction with the shelter, the soil is likely to be dry at all seasons, having no hard top crust, such as is formed under the influence of moisture and sun, and containing sand 40 and decaying wood or other vegetable matter so as to make it light. In nearly every instance the breeding places have been situated beneath a fallen and well-decaying tree, which has been prevented by some of its limbs from actually touching the ground, and is of such girth as to keep an area beneath it shaded and free from moisture. The soil in such a place, being leavened by humus, becomes light and friable, so that the larva can have little difficulty in making its way into it. " Very few pupae have been found in hard clayey soil beneath trees, and as it has been found experimentally that the larvae have very weak boring powers, being unprovided with bristles, their occurrence in such situations is probably purely accidental, the parent fly having possibly failed to find a suitable place in time for the birth of its offspring. Such soil, moreover, in the dry season dries almost as hard as stone, needing such considerable force to break the top crust as would render it almost impossible for the newly emerged fly to break its way to the surface. " Some few pupae have been found under trees in accumulations of dead leaves on the surface of the ground, but so few, that this, again, is probably purely accidental." " As has already been pointed out by workers in Rhodesia, pupae may be deposited in cavities in dead limbs of trees, a certain amount of soil, the result of the admixture of earth originally brought by termites with wood debris, being invariably found in such situations. " The question of soil in the various breeding places has been studied, but no special sort seems to be favoured, the chemical constitution being immaterial, so long as the soil is friable. All the trees also, in relation to which pupae have been found, have been carefully examined with a view to ascertaining whether one species more than another is favoured by the fly. In the majority of cases decay has been so far advanced that the determination has been out of the question, but in 48 instances specimens of the foliage, representing at least fourteen species, have been obtained and are available. In five cases a few pupae have been found under the shelter of a fallen Borassus palm. " Further study of the question has shown that the presence of a dead tree is by no means essential, for in certain parts of the Monkey Bay district breeding grounds sheltered by overhanging rocks have been found, each yielding from two or three to as many as ten pupae. " Attention has been paid to the orientation of all these places, the conclusion arrived at being that this is immaterial, all that is apparently required being that the breeding places should be sheltered from the overhead sun. " The conclusions arrived at from my study of the question are that the only essentials inducing a female tsetse to select a particular spot are looseness of the earth and shelter. In Nyasaland, as in Rhodesia, the situations most favoured are near game and native paths, and near water-holes, whereby the newly emerged fly is in the most favourable situation for obtaining food in the shortest possible time after emergence. Beyond this, the choice of a site by the parent fly is not, in my opinion, influenced by any special type of soil or of vegetation. The insect fauna in breeding places has been studied as digging has proceeded. It is never very numerous, the only frequent occupants of such places being small blackish Myrmicine ants [Pheidole mega- cephala, F.] and the larvae of ant-lions. Small cockroaches and their oothecae are not uncommonly found, and also the pupa-cases of the large Ponerine ant, Paltothyreus tarsatus, F. . . . . 41 " It is by no means .uncommon also to find small lizards and geckos,, which doubtless take some toll of pregnant and newly emerged flies. " No evidence has been seen that any scratching animals habitually seek food in such places, the various Grallatores — francolin, partridge,, and guinea fowl — which abound in the tsetse areas, preferring more open places, especially where game is in the habit of resting." According to Mr. LI. Lloyd (91), all the G. morsitans breeding places examined by him in Northern Rhodesia, in 1915, showed the presence of a dark hiding place for the flies, and almost all were of the type met with in other fly areas. Exceptions occurred near the Lukanga River, in the thin mopani forest, and in the bed of the Burbwa stream in Chutika. These three breeding places were such as to show the improbability of pregnant females being influenced by any special odour, such as that of humus. It appears that pupae are most numerous in districts in which game is plentiful. This conclusion is strongly supported by the more recent experience of Swynnerton (145), who states that, when investigating the breeding habits of G. morsitans in North Mossurise, Portuguese East Africa,, he " took many hundreds of full and evacuated puparia in all from more than a hundred breeding places The vast majority of the pupae were in the nearest suitable hiding-place to a spot in which big animals had been lying, down — buffalos (mainly), wart- hogs, hartebeests, etc." Swynnerton, who thinks that the female flies generally deposit their larvae when hiding during intervals between meals, and that the spots in which they do so are such as are convenient to themselves, " but by no means necessarily " to their offspring, obtained " indirect evidence " indicating that the maggots are some- times dropped from a height of several feet above the ground. On the subject of the breeding places discovered, this author writes : — ;< Most of my puparia were found under prostrate trunks and branches, raised little or much from the ground, but batches were also found in the angles of root-buttresses, under mere leaning trees, under fallen palm leaves, and below and between the outleaning dry leaf stems at the bases of palms (Hyphaene ventricosa), in holes in trees and in holes made by animals in banks, and where the hoof-marks of passing animals had broken the slight crust of a sandy stream-bed immediately under light overhanging grasses. I failed, in spite of much search, to find tsetse pupae in the leaf-sheaths of palms, though I took Lepidopterous and often other Dipterous pupae there. The situation was usually dry, but some of my sand-stream pupae were in a moist situation, and I took a pupa-case that had emerged successfully under a fallen Eugenia log in very damp ground on the edge of a vlei." " Provided that the hiding-place itself afforded shade to the mother,, overhead shade (as an analysis showed) was a matter of complete indifference, and the ground in which puparia lay was commonly reached by the sun during part of the day. " The puparia were usually close beneath the surface of the earth, and the evidence suggested that the maggot had wandered but little from where it was dropped. Under each log was usually harder ground,, interspersed with softer pockets where babblers (Crateropus) or small mammals had previously scratched. The puparia tended to be congregated in these pockets, but they appeared in some cases to give evidence also of gregarious settling by their mothers." As regards G. pallidipes, in Italian Somaliland, according to Croveri (35) , the pupae of this species are found in nature under the surface soil 42 beneath bushes in places where the earth is dry. They also occur among the roots of trees, banana plants, etc. In the North Katanga district of Belgian Congo, Dr. Schwetz (132) on several occasions found pupae or empty puparia of G. pallidipes, together with those of G. brevipalpis, in very light, sandy soil beneath a layer of dead leaves, in spots shaded by branches of trees and thinly covered with low shrubs or herbage. Swynnerton (145), in North Mossurise, Portuguese East Africa, took puparia of G. pallidipes " with those of G. morsitans and under the same logs, but in very small numbers indeed." Of the larger Tsetse-flies, the only species of the breeding places of which we have as yet any knowledge are G. fusca and G. brevipalpis. With regard to the former, it is stated by Austen (3) that on 5th December 1906 a number of empty puparia and one living pupa of this species were found by Dr. A. G. Bagshawe " in a dry gully in the vicinity of the Mzizi River, near the south-east corner of Lake Albert, Uganda Protectorate." For later information as to breeding places of G. fusca we are chiefly indebted to the painstaking investigations of Dr. Schwetz (132), in the North Katanga district of Belgian Congo. After preliminary failures, during which on one occasion this observer, assisted by ten or a dozen natives, examined places of all kinds for three weeks without finding a single living pupa of this species, better success was encountered. Empty puparia of G. fusca, with an occasional living pupa, sometimes associated with puparia or pupae of G. brevipalpis, were found in small numbers in various spots of the usual character, chiefly beneath fallen tree-trunks. Finally, in a belt of forest near Kisenga, not traversed by any stream, Dr. Schwetz met with a very large fallen tree lying in a thicket so dense as to be well-nigh impenetrable. In the hollow at the base of the roots of this tree, in light, dry soil underneath a layer of dead leaves, twigs and humus, 95 empty puparia of G. fusca and five of G. brevipalpis were found. More puparia and a few living pupae were discovered beneath the tree itself, where it lay on the ground or was raised a few inches above it ; altogether this single tree yielded 205 empty pupa-cases and nine living pupae of G. fusca, and nine puparia and one living pupa of G. brevipalpis, so that Dr. Schwetz is amply justified in speaking of it as a " breeding place." It would seem, then, that, in evincing a preference for depositing its larvae beneath fallen trees, G. fusca displays a predilection similar to that oi\ G. morsitans. In the case of G. brevipalpis, the attraction of fallen trees would appear to be not always manifest. Thus, in the account of his observations upon this species conducted in Nyasaland, on the Lingadzi estate, " which is situated on either bank of the river of that name, \ about eight miles from Lake Nyasa," Dr. Lamborn (83) writes :— " When endeavouring to discover the breeding grounds of this fly, my attention was at first naturally concentrated on places similar to those utilised by morsitans, and in some of these, under the shelter of fallen dead trees, I found pupa-cases both of morsitans and of brevipalpis, never more than three or four of each. But on removing the low, thickly growing vegetation at a spot well sheltered by the foliage of the large trees, I obtained in the course of a two days' search over an area about 25 yards square, no less than 507 pupa-cases and seven living pupae. This spot differed in regard to soil and vegetation in nowise, so far as I could ascertain, from the rest of the neighbourhood, but it 43 was traversed by a path along which game is evidently in the habit of passing to and from the water, and moreover it is close to the edge of a sandy bank (the original edge of the lake, which is gradually receding), and I imagine it to be probable that beasts may loiter there as a measure of caution before descending into more open country, as a result of which replete female flies, incapable of prolonged flight, would shelter there. More pupae have been found in similar situations — 109 pupa- cases in one place, and four pupae and 54 cases in another." On the other hand, Mr. C. F. M. Swynnerton (145), in the course of a preliminary investigation of the habits and distribution of Tsetse- flies in North Mossurise, Portuguese East Africa, carried out in 1918, although failing to discover any living pupae of G. brevipalpis, " found a few batches of pupa-cases in the Brachystegia bush on the Buzi east of Spungabera," and on one occasion unearthed nearly 100 puparia from beneath a fallen tree (Piptadenia buchanani. — See Plate IV, Fig. 2) . "In the case of this fly," writes Mr. Swynnerton, " the batches were beside lying places of buffalos and wild pigs, and the flies at the moment of dropping the maggot had in the biggest find evidently been resting for the most part on the under sides of the coils of large rough lianas (Cissus, Landolphia and Bauhinia), apparently favourite resting- places. Some of the puparia were found under logs, and some of the replete gravid females taken during the bush-pig incident were resting on the lower surface of fallen Uapaca sansibarica stems, under which the pigs had been lying. The ground (shale) was in two cases exceed- ingly hard and compact, but humus was present in the situations in which the pupae were found. There seemed to be no such localisation of the breeding centres as I found with morsitans." Reference to the finding by Dr. Schwetz (132) of puparia and pupae of G. brevipalpis in small numbers, in various situations in North Katanga, has already been made in connection with breeding places of G. fusca (see p. 42). This author thinks that, when depositing their larvae, both these species, like G. palpalis, avail themselves of any spot suitable for the emergence of the adult insect. BREEDING SEASON. — So far as we are aware, this question has not yet been studied in the case of any species other than Glossina morsitans. As regards the latter, it would- seem from the investigations of Mr. R. W. Jack, and those of the entomologists who have worked in Northern Rhodesia, that it is probable that the places where the larvae are deposited vary according to the season. Limited during the dry season to the borders of rivers and similar shady places, in the wet season and immediately thereafter breeding places may occur wherever bush exists. It appears from Mr. LI. Lloyd's account (89) of his later work at Ngoa, Northern Rhodesia, " on .the high ground of the Congo- Zambezi watershed," from which extracts have already been given, that in the district referred to " breeding is almost confined to the warmer part of the dry season." The conclusion at which Lloyd arrives is that : — " On the high plateau of Northern Rhodesia G. morsilans begins to breed freely about the second month of the dry season (July), and almost or entirely ceases to do so in the rainy season." Nevertheless, the experiments in rearing G. morsitans made in Nyasaland in 1911, by Newstead and Davey, seem to show that the breeding season in that country continues throughout the year, though it is probable that it is more intense at the commencement of the rains and at the end of the wet season than during the dry months. 44 CHAPTER V. " FLY-BELTS," OR AREAS IN WHICH THE DIFFERENT SPECIES OF TSETSE ARE FOUND. " In those parts of Africa in which Tsetse-flies occur they are not found everywhere, but are generally confined to particular tracts, which are known as ' Fly-belts/ and are often of very limited extent. We are still somewhat in the dark as to the factors that determine the limits of these ' belts/1 but, although Tsetse are undoubtedly dependent upon the blood of vertebrates for their continued existence, all recent evidence goes to show that the most important element is the physical character of the locality, and, probably, its suitability as a breeding ground. As a general rule it may be said that Tsetse-flies are usually met with in damp, hot localities, either on the borders of rivers or lakes, or at any rate not far from water ; this, however, applies more especially to the species belonging to the Glossina palpalis group ; other Tsetse-flies, especially those of the Glossina morsitans group, are often found at a considerable distance from water. Cover, in the shape of more or less thick bush or forest, shady trees, or reeds, is essential, and Tsetse-flies are absent from open grass plains " (Austen, 3) „ We will now give, as concisely as possible, a rapid review of the essential facts already ascertained as to the habitats of the different species of Glossina. For fuller information upon this subject the reader should consult the Bibliography at the end of this volume, (see pp. 165-178). Glossina palpalis, Rob.-Desv. — As is well known, this Tsetse-fly haunts the margins, overgrown with vegetation, of rivers, streams, watercourses, pools, lakes and marshes in the low-lying, humid countries of Tropical Africa ; and, unless it has followed human beings or animals, it is rarely met with more than some 30 yards from water. Fiske, however (54), writing after many years' study of G. palpalis in the islands in, and on the mainland shore of Lake Victoria, Uganda, considers it " pure coincidence that the insect is never found far from water," and asserts his confident belief " that it would exist, and that it will be found eventually to exist, in any inland localities wiiere host animals of favoured species occur in well sheltered areas provided with suitable breeding places." On another page of the same paper this authority writes : — " Every bit of evidence that has been gathered supports this final conclusion concerning the range of fly inland from the lake shore ; it is primarily controlled by the distribution of host animals, and there is nothing to indicate that this tsetse requires open water or humid conditions, or that water is even attractive to it. Its favoured hosts are amphibious in habit ; the most attractive types of shelter occur more commonly near the shore than inland ; and it rarely finds good breeding places elsewhere." As regards the character of the shelter needed by this species Fiske remarks : — " Two kinds of shelter are requisite, (a) light, such as serves at breeding grounds and 1 In this connection Mr. R. W. Jack (74), writing with reference to G. morsitans. in Southern Rhodesia, in 1918, stated that observations show that the dispersing range of this species does not exceed two or three miles, but that there is no apparent reason for this limited migration unless it be a definite instinct >to avoid wandering. The males may be carried in any direction up to ten miles and more- by the movements of animals and human beings, but there is a marked tendency to return to their place of origin. 45 for the active flies ; and (b) massive, or forest-like, which is required by the inactive flies." Dr. S. A. Neave (103) states that he is not aware of " any authentic record of G. palpalis occurring at over 4,000 ft. even at the equator, and the further from the equator, the lower the limit seems to be, as might be expected." Shade is an absolute necessity for this species, which, however, on Lake Tanganyika has been found among reeds, although the shores of this lake are usually wooded. In West Africa G. palpalis 'commonly occurs in mangrove swamps. The published statements as to " fly-areas " in the case of this species, are however, so numerous that we cannot here attempt to give, in a few lines, a resume of all that is at present known on the subject. The literature relating to the biology of the Tsetse-fly chiefly concerned in the dissemination of sleeping sickness is considerable, and is continually increasing ; the latest information thereon is to be found in the Bulletins of the Tropical Diseases Bureau, the Review of Applied Entomology (Series B), and other special publications. " One point in respect of clearing operations," writes Neave, " is of interest, namely, that this and probably all other species of Glossina habitually live very near the ground, and probably never settle more than a few feet above it. Hodges and others have demonstrated that clearing operations need not involve the removal of large trees with clean trunks free from low-growing branches. Tt seems to be the low growth of shrubs, etc., which provides the necessary cover for G. palpalis, and this will probably be found true for all species of Glossina." As regards its habitat, Glossina palpalis var. juscipes, the form of G. palpalis occurring in the eastern region of Central Africa, differs in certain noteworthy respects from the typical race found in West Africa. With reference to G. palpalis var. fuscipes, as observed by him in the Nile Province of the Uganda Protectorate, Dr. R. E. McConnell writes (92) : — " The most favourable environment for the fly in this Province consists of the conjunction of a tree- or scrub-shade, water, and a rocky clean bank. They are often very numerous on rocky promontories, the vegetation on which offers the very modest shade lent by creepers, tufts of grass, and scattered, very low scrub. They are also very common on bare rocks, but these have always been so situated that access to typical tree-shade was available, and doubtless the flies retired there. They seem quite as numerous near rushing rapids as near the more peaceful waters." Glossina caliginea, Austen. — Very little is yet known as to the habitat of this Southern Nigerian Tsetse-fly, which is a vicious blood- sucker and, according to Austen, " prone to attack travellers in canoes." Mr. G. Garden (quoted by Austen, 3) states that, at a spot on a creek between Agbabu and Akitipupu, where the fly attacked him in his canoe " during heavy rain/' there was " dense cover," consisting of " long grass and water-plants." Glossina pallicera, Bigot. — Almost all that can be stated with reference to this rare species of Tsetse is due to " observations made in Ashanti, in 1907, by Dr. W. M. Graham," and recorded by Austen, who writes that Graham " always found the fly in the ' bush,' where it ' does not settle on the ground, but sits on the upper side of leaves.' " On the N'Kawi Road, Kumasi, according to the same observer, where it " ' was found in a narrow bush path, bordered by short bushes, but free from tall trees,' " G. pallicera appears to be " ' strictly localised. . . . . It did not associate with G. palpalis upon this path, nor was it to be found associated with G. palpalis upon the main road into 46 which the path opens/ ' In French West Africa, according to* Roubaud (1230), G. pallicera is, like the members of the G.fusca Group, extremely local and, with occasional exceptions, scarcely to be found outside the limits of the true forest. Glossina tachinoides, Westw. — In Africa, at any rate (it will be remembered that G. tachinoides has also been found in Southern Arabia), according to various observers, this species appears to share the preference of G. palpalis for the vicinity of water. " Writing in the Sleeping Sickness Bureau Bulletin No. 6 (7th May 1909), p. 210, Dr. A. G. Bagshawe remarks that Glossina tachinoides, ' at least in Nigeria and on the Lake Chad waterways, is confined to river banks, as is palpalis and occurs in large numbers, biting man. It is therefore allied in its habits to palpalis/ .... Dr. Dalziel, writing from Yola Province, Northern Nigeria, states that ' Glossina tachinoides seems to be found only close to the river bank, in sight of water almost, though it may follow along the path. The longest distance to which it was found to follow away from water was 1 ,200 yards (one fly) . . . / In the Western and Northern Provinces of Ashanti, on the Tain and Black Volta Rivers, Dr. A. Kinghorn finds that in habitat and habits G. tachinoides closely resembles G. palpalis, in company with which it always occurs. According to Kinghorn, the present species ' will feed on man with avidity/ and is ' quite active in dull weather and in the very early hours of the morning, both times when Glossina palpalis is usually quiescent/ According to Dr. Alexander, on a river near Amar, in South Bornu, ' Glossina tachinoides bites after dark. . . / In Central Dahomey, according to Roubaud, G. tachi- noides is a migratory species, which comes from the north, and ' is present in abundance on the banks of the large watercourses only from May to August, when the rains are at their height and con- siderable changes in the river-systems take place. Its habitat is the same as that of G. palpalis, though it displays a preference for more open wooded tracts, and is not found in the groves of wild paims along the small streams/ Roubaud states that in Dahomey during the wet season, when the air is saturated with moisture, G. tachinoides, palpalis, and longipalpis are all found beyond the limits of their usual haunts, and isolated specimens may be met with in the savannah zone, far from any watercourse" (Austen). Dr. Bernard Moiser (101), writing in September 1913, after experience gained during rather more than two years spent in Bornu, Northern Nigeria, states that in that region Glossina tachinoides is " confined to small patches of dense jungle situated along the course of small rivers." He adds that the species is " found in quite localised areas, in close proximity to rivers or marshes, where there is water all the year round, and where the ground is covered with tall shady trees, mostly tamarinds, and thick undergrowth of thorns and creepers, with some ebony trees." From an earlier paper by Dr. Moiser (100), dealing with the same locality and species, we extract the following " conclusions." " (1) Deep shade and proximity to water appear to be the chief factors determining the localisation of the flies. " (2) The natural resting-place of G. tachinoides is on the lower side of twigs and branches of undergrowth, under the shade of large trees, at a height usually not greater than a foot from the ground. " (3) They are very restless, and do not usually remain long in one position. 47 " (4) The flies do not usually travel higher than 4 or 5 ft. from the ground, and probably never ascend as high as 10 ft. " (5) They probably do not feed on monkeys or birds, but on ground animals, e.g., wart-hog, duiker, or bush-buck. " (6) The flies require a meal fairly frequently, and cannot withstand starvation (without water) for longer than 24 to 30 hours. In cap- tivity, at any rate, they will feed on the dead bodies of other Tsetse- flies. I am inclined to think that they may feed naturally on other insects, ticks, grasshoppers, etc. They certainly feed voraciously on human beings. " (7) I am of opinion that, during the day, the flies are constantly moving about from place to place within the fly-belt, i.e., in deep shade, and only for short periods rest on the under side of twigs and small branches, and perhaps on the ground." On the other hand, Dr. J. J. Simpson (142 III.), writing on the distri- bution of Glossina in Southern Nigeria, remarks : — " The conditions which are most favourable for G. palpalis are, generally speaking, most unfavourable for G. tachinoides. Where the country is open, the vegetation sparse, the dry season well defined and the rainfall slight, there G. tachinoides is most abundant. Consequently, this species is the predominant one in the region bordering the Sahara in Northern Nigeria ; it is the only member of the palpalis group found in the Lake Chad area." Simpson's statements are corroborated by Roubaud (I23a), who>, after stating that the partiality for the immediate proximity of water displayed by G. palpalis is exhibited in equal degree by G. tachinoides, continues : — " This species, however, as I have mentioned elsewhere, possesses greater resistance to heat than does G. palpalis. This resistance, which may exceed a mean of 35° C. [95° F.], enables it to exist in a more northerly habitat. Thus it is G. tachinoides rather than G. palpalis that is met with in the belts of thin forest fringing the Soudanese rivers (forest belts of the Soudanese zone of A. Chevalier), where the temperature is subject to much greater variations than is the case in true forest belts. It is not uncommon, in the shade of strips of woodland infested by G. tachinoides, to find the temperature of the air during the day exceeding 30° C. [86° F.]. " This fly, which is the smallest of the Tsetses, is also the most active,, and the quickest in its movements and biological manifestations. It is likewise an insect which, more exclusively than G. palpalis, derives its sustenance from wild animals. Unlike the species men- tioned, it is scarcely ever found permanently established in the vicinity of villages ; on the contrary, it displays a predilection for sparsely inhabited regions full of game, where there is abundant animal food at the water's edge." " In Southern Arabia, G. tachinoides was found by Captain R. M. Carter sparsely and locally in thick belts of euphorbia, babal thorn and tamarisk, and also in cactus belts. It was never seen ' in the date groves or along patches of cultivation,' and was not always met with near the edge of water " (Austen). Glossina austeni, Newst. — Writing of this species in his " Notes on the Blood-Sucking Insects of Eastern Tropical Africa" (1912), Dr. S. A. Neave (103) says : — " It occurs in company with G. pallidipes and G brevipalpis, though it seems to require more heavily forested country than that in which those species sometimes are found. It would 48 appear to be confined to the coast belt in British East Africa [Kenya . Colony], where it evidently has a fairly wide distribution , Up to the present it has not been captured at a greater elevation than about 1,500ft. above sea-level." G. austeni, which, according to Newstead appears to be especially harmful to stock, likewise occuis in Tanganyika Territory and Portuguese East Africa. Glossina morsitans, Westw. — " As indicated by the term ' fly-belt/ ' wrote Austen (3) in 1911, "originally applied to G. morsitans, this species is usually confined to quite definite tracts, often of very limited extent. Within these ' belts ' the fly sometimes attacks human intruders in such numbers as to have been compared to a swarm of bees. Cover, usually in the shape of large trees with thick under- .growth, open thickets, or scattered, shady trees, is essential to the existence of G. morsitans, which, like other species of Tsetse, is never found on the open, sun-scorched veld.1 Authorities like Mr. F. C. Selous, Sir John Kirk, and others (whose experiences, it should be pointed out, relate to the period before the buffalo was driven north- ward, or almost exterminated by rinderpest), write of G. morsitans occurring in swarms on the banks and ' along the water's edge ' of the -Zambesi, Chobe, Rovuma and other rivers. Recent observers, on the other hand, such as Dr. L. Sander (in German East Africa), Sir Alfred Sharpe (in Nyasaland), and Mr. S. A. Neave (on the south-west shore of Lake Nyasa, and on the Luangwa River in north-eastern Rhodesia), lay stress on the fact that, in their experience, G. morsitans evinces a dislike to, rather than a preference for, the immediate vicinity of water, whether river or lake, and is not infrequently met with at a considerable distance (half a mile or so) from any water.2 In this connection it is not uninteresting to remember that the original specimens of G. morsitans, found by Vardon and Oswell in 1845, were obtained on the Siloquana Hills in the Northern Transvaal, between the Magalaqueen or Nylstroom River and the Limpopo. Glossina morsitans, indeed, is by no means confined to low-lying districts. Some three years ago Mr. E. A. Copeman, at that time District Com- missioner at Kasempa, forwarded to the [British] Museum a number of specimens of this species taken by him in January 1908, on the Congo-Zambesi watershed, Kasempa district, North-Western Rhodesia, at an altitude of from 5,000 to 5,500 ft. In Nyasaland, however, 1 This refers more especially to conditions such as those in the Transvaal. Dr. S. A. Neave (103), writing in 1912 of his " own personal experience and observations of Glossina morsitans " in Eastern Tropical Africa, says : — " I am of opinion that among the essential factors which determine the distribution of this species, are a combination of the presence of such vegetation as will provide moderate but not excessive cover, coupled with a hot and mpderately or even very dry climate I have on several occasions, particularly in Northern Rhodesia, noted that when the grass is long (not less than four feet) G. morsitans is inclined to be more numerous in grassy areas of limited size than in the woodland or bush-covered country surrounding them. I have more than •once seen this species swarming in long grass on plains of some size, from half a mile to a mile or more from the line of bush at the edge In large short -grass plains, such as those to the east of Lake Bangweolo, this fly does not occur, nor from the paucity of the cover would it be expected to do so. G. morsitans also appears to avoid the other extreme, viz., dense forests where the atmosphere is cool and damp." 2 Writing in 1912 with reference to G. morsitans, Dr. S. A. Neave (103) remarks : " Compared with most species of Glossina, the complete independence of water exhibited by this species is remarkable. In the Luangwa Valley I have seen it swarming, in intensely hot weather, at least five miles from any known water. Indeed, the drier the atmosphere the greater seems to be the activity of this -fly." 49 according to Sir Alfred Sharpe, G. morsitans is ' seldom found above 3,000 ft.'1 " The idea, formerly generally entertained, that G. morsitans ' appears to avoid the presence of man, and is rarely found in the vicinity of human habitations, or within the confines of a town or other settle- ment,' is only partially true, since native villages in areas in which this species occurs are often infested by the fly."2 Since the publication of Austen's " Handbook " in 1911, a consider- able number of investigations have been made with reference to the areas in which Glossina morsitans is found. It is impossible for us to summarise them all, but we give below a short epitome of the conclusions that may be drawn from two series of important researches carried out respectively in Northern and Southern Rhodesia by Messrs. LI. Lloyd and R. W. Jack. It is clear that these conclusions have special reference to Rhodesia. 1. In the course of the dry season, water remaining in certain spots (rivers and pools) favours the growth of species of trees different from those in the surrounding bush, and enables others to retain their leaves. G. morsitans, therefore, being especially partial to cover, assembles in winter (the dry season) in these spots, which then form well-defined tracts, or, in other words, "fly-belts." This concentration appears to be in no way due to the attraction of water, since, as we have already seen, it has been found in many localities that the fly shows, if anything, a repugnance to the immediate vicinity of rivers. In Southern Rhodesia, according to Jack (75), the type of haunt where G. morsitans is most numerous in the dry season consists of isolated areas of low lying grassland, with evergreen trees along at least some portion of its margin ; here the grass-feeding animals concentrate in large numbers and provide an ample food -supply. 2. Before or after the dry season fly-belts are much less sharply defined. During and immediately subsequent to the rains, G. morsitans is much more generally distributed throughout the surrounding bush, the trees in which are then covered with leaves and provide the necessary shade. 3. G. morsitans is concentrated in certain tracts, and its extension towards fresh districts appears to take place very slowly, and to be due to pressure of numbers Thus it is by no means found in every suitable locality. 4. Fly-belts do not appear to be characterised by geological forma- tion, a peculiar physical aspect of the soil, or a special type of vegetation. Nevertheless G. morsitans avoids places where the vegetation is either impenetrable or insufficient to ensure the shade that it requires. As regards Belgian Congo, according to Dr. Schwetz (132), the north-western boundary of the great G. morsitans zone coincides more 1 Dr. S. A. Neave (103) writes : — " The limit of elevation at which G. morsitans occurs in Nyasaland would appear to be about 3,000 ft. In Northern Rhodesia "I have never met with it myself at over about 4,200 ft., though it is said to have been taken at somewhat greater elevations. This difference may perhaps be accounted for by the fact that Northern Rhodesia has on the whole a somewhat drier climate than Nyasaland." 2 " With regard to the relations of this species with man, though, as is well known, it bites man readily, I think it is doubtful whether it has any preference for him as compared with other mammals I am in complete agreement with the views expressed by Sir Alfred Sharpe and others that it is the cleared and cultivated area that usually surrounds native villages which is the deterrent, and not the actual human habitations or their accompaniments " (Dr. S. A. Neave, 103). (5979) D 50 or less everywhere with that of the wide expanse of Katanga " orchard bush/' Bush fires, however, modify for the time being the limits of the G. morsitans area, by causing a true temporary migration of the insect. Thus, after the bush to the east of Katompe is burnt in the middle of the dry season, the fly invades the station in fairly large numbers, and may even be seen from two to five kilometres to the west of it, although its ordinary boundary lies a few kilometres to the east of the settlement. Some recently published observations by Dr. Roubaud (I23a) on the habitats of G. morsitans in French West Africa are of sufficient importance to be worth quoting in extenso. "In G. morsitans," writes the distinguished French authority, " we have a true savannah fly, which is most often to be found among tall grass or bush far from sheets of water. It is a species that loves dry places (espece xerophile)i and is met with in the sparsely covered savannah-park, sometimes even amidst the thorn-scrub of the zone bordering on the Sahara (zone sahelienne). The presence of watercourses, at least in the period following the winter season, appears to be a matter of indifference to it, and it may frequently be observed resting on the ground or on herbage a long distance from water of any kind. When, however, the drought becomes excessive, the fly returns to the vicinity of streams, which it henceforth does not leave to any great extent. " Of all species of Tsetse-flies, G. morsitans is the one most readily noticed on entering a zone frequented by it. Here it is generally in great abundance, and, flying openly along the tracks and paths, it assails the traveller in dense swarms, which accompany him as he proceeds. It is more especially this habit of collecting together in large numbers, which is so entirely characteristic of the species, that gave rise to the, nowadays classic, English term fly-belts. " The study of the physiological peculiarities of this Tsetse-fly serves to explain very clearly the peculiarities of its habitat. A high degree of atmospheric humidity is inimical to it ; when subjected to the action of air completely saturated with moisture it speedily succumbs, whereas G. palpalis and G. tachinoides are capable of resisting similar hygrometric conditions for a long time, even in a confined space. Conversely, G. morsitans can be kept for a long period in a medium of which the hygrometric index is below 35 per cent., although such a medium is rapidly fatal to the two moisture-loving species referred to. Lastly, though the absolute capacity for resistance to heat displayed by G. morsitans is perhaps not quite so great as that of G. tachinoides, the former is able to support a daily mean temperature exceeding 32° C. [89-6° F.] ; and, as has been observed in the scantily covered savannahs frequented by this species, it resists thermic variations which may reach 40° C. [104° F.] at extreme temperatures. " More even than G. tachinoides and G. longipalpis, which are ' wild ' species having no habitual connection with man, G. morsitans is, par excellence, the big game fly. All the savannahs in which it is met with are sparsely populated but remarkably rich in game : e.g., the contiguous plains of Upper Gambia and Guinea, the Niom region in Saloum, the north of Dahomey in the vicinity of Western Nigeria, etc. The presence of the large wild mammals in abundance seems indeed to be the principal factor upon which the occurrence of this fly in a given region depends.1 Many tracts of savannah in the Guinea 1 Cf. G. Bouet and E. Roubaud, " Trypanosomiases et Glossines de la Haute- Gambie et de la Casamance." Bull. Soc. Path. Exot., Paris, t. v. (1912). 51 and French Sudanese zones, though identical in appearance with those in which G. morsitans occurs, are totally wanting in this species of Tsetse Now, the only differences that seem to us to explain this absence of the fly consist in the fact that game is more or less scarce. G. morsitans always swarms in French West Africa in very scantily populated regions, where human activity is of little intensity, while the soil is scored by the spoor of wild animals, testifying to their incessant movements to and fro in complete freedom from all restriction. In these regions, again, man is bitten but occasionally by this fly, when he ventures into the infested zones. G, morsitans is never met with as G. palpalis is encountered, in small foci at a distance from the main zones, and in proximity to man alone." " Generally speaking, G. morsitans zones are met with much further to the north than those infested by G. longipalpis, as was only to be expected in view of what has already been stated as to the fondness for dry places exhibited by the former species. In French West Africa there are at least four great morsitans zones, in all probability continuous one with another at least during a portion of the year, and lying between 10° and 14-5°N. Lat., from Nigeria to Senegal. Of these zones the most important is that which comprises the hinterland of Guinea, Upper Casamance, Upper Gambia and the corresponding regions of Thies-Kayes. I have already pointed out elsewhere (121) that, as the furthest outpost towards the west, this zone still exhibits in the fly-belts of the Niom-Bato a proof of its former continuous extent. " It is, moreover, probable that, as regards their geographical limits at the present time, these morsitans zones are not absolutely fixed ; indeed, among the principal factors producing corresponding variations in the morsitans fly-belts we must include the migrations of game and the possible variations in the abundance and size of the herds of wild animals. " All causes capable of producing variation in the distribution of big game — hunting forays, epidemics, railway extensions, etc. — will necessarily also have a repercussion on the dispersion of G. morsitans, which, in view of this, must be considered much less stable than that of the other Tsetse-flies." Instructive statements as to the habitat of G. morsitans in the North Mossurise district, Portuguese East Africa, are given by Mr. C. F. M. Swynnerton (145), as the result of three months' personal observation, from June to September 1918. After describing the haunts of G. brevipalpis, the author in question continues:—" Glossina morsitans was found in a very different kind of country. It was very obviously far less dependent on good shade than G. brevipalpis, and I failed to iind even stray individuals in the densest forest types It occurs both on the granite-gneiss and the basalt of the lowlands, the latter usually with very poor bush indeed ( Combretum, etc. . .), and the former with savannah forest of a poorer type than that of the sedimentary area, though in each of these two it is a Brachystegia. On the basalt and on most of the gneiss this fly occurred during my visits only as very rare individuals, and on the basalt and away from vleis only males were taken. The granite-gneiss, however, carries very numerous vleis, many of them with permanent water, and in spite of much painstaking search elsewhere, it was only at these vleis (and then only at some of them) that morsitans was found 52 in great numbers together and its breeding going on with some vigour in the dry season. " Here the male flies occurred in the usual bands, containing in some cases at least a few hundred flies at a time on the short or shortish grass and sedge. They were sometimes out in the sunlight amongst or beyond the scattered chidsgwati shrubs and stunted Parinariums that separate the open vleis from the Brachystegia bush surrounding them, sometimes a little way in the Brachystegia bush itself, never. very far from the vleis The flies were found in rather special '' association with a low, heavily-headed sedge (Fuirena), with which I found the local natives had also learned to associate them. This remained green after the grasses generally were dry and, with one or two low-growing associates, covered considerable areas' at the edges of certain vleis. " I did not find that G. morsitans appeared to be at all dependent on the presence of undergrowth. The bush in which it occurred was for the most part devoid of such growth, though some of it was itself low." On the other hand, Swynnerton states that he "obtained! no evidence in favour of the view, but much against it," that either G. morsitans, G. pallidipes or G. brevipalpis "will live in open grass country devoid of bush." " Even the attacks of morsitans on the basalt," he continues, " were always in or beside shade, were it only that of a large shrub or a semi-leafless tree, and when noted definitely resting, either in my experimental net or in the field, this fly was always in the shade — under a log, on the shady side of a trunk, etc. A replete resting morsitans female that I disturbed repeatedly always settled again on the shady side of trees. I have seen waiting male clusters furthest from bush in large vleis, but there was some reason to suppose that the individual flies did not stay with the cluster indefinitely." Finally, with reference to the habitat of G. morsitans form sub- morsitans, it may be added that Dr. J. J. Simpson has shown that, in Northern Nigeria, this form occurs in the driest districts, where the " savannah forest " predominates. In the Gold Coast, according to the same author (144), G. morsitans form submorsitans is seldom found near rivers, but is more abundant in the savannah forest and more open country, where small water-holes exist, probably because game comes to water there, this form undoubtedly migrating by following herds of game for long distances. Simpson states also that this fly will follow human beings and attack in the open in the hottest sun, though rarely at night. Glossina pallidipes, Austen. — "The late Dr. W. A. Densham, writing from Nimule on 24th July 1906, with reference to Glossina pallidipes, as observed and collected by him in the Nile Province, Uganda Pro- tectorate, said : — ' Of thirteen specimens taken, nine were males and four females. The flies were found near and in a narrow belt of true forest at Kibero, between Nimule and Wadelai, in June 1906. They were numerous along the native path, in long grass with scattered trees, for a quarter of a mile before reaching the forest ' " In the East Africa Protectorate [Kenya Colony], where, according to Dr. P. H. Ross, the species is most numerous in August and Sep- tember, Dr. A. D. Milne, P.M.O., states that G. pallidipes, in company with G. brevipalpis and longipennis, attracted by lamplight, frequently enters railway carriages when a train is waiting during the night at' | 53 one of the stations in the fly-belt on the Uganda Railway, and thus may be carried for a distance of 150 miles or more " (Austen). Writing from personal experience of Glossina pallidipes, Dr. S. A. Neave (103) says: — "This species somewhat resembles G. morsitans in its habits, at least as regards the type of country in which it occurs and the conditions under which it feeds, but it would appear not to be so completely independent of water. So far as my experience goes, though not at all confined to river banks, it is always associated with a fairly considerable amount of bush in rather low-lying river valleys ; but, given sufficient cover, it would no doubt be found a mile or so from water.1 There would appear to be some ground for thinking that, in British South Africa, at any rate, this tsetse has a more marked seasonal prevalence than G. morsitans, being much more numerous at the end of the wet season and the months immediately following it than during the dry. . . There would seem to be some evidence that this species, like G. morsitans, is kept at bay by cultivation." According to A. S. Leese, during the hot season G. pallidipes is most •' active by daylight, before 9 a.m. and after 4 p.m. ; but it sometimes bites in the dark. In the wet season it is probably also active in the middle of the day. In Italian Somaliland, according to Croveri (35), the situations pre- ferred by G. pallidipes are wooded spots covered with low, dense bush, | and near stagnant or slow-flowing water. Such localities are limited i in area during the dry season, but extend considerably during the wet | season, so that the infested zone is then increased. G. pallidipes avoids open and cultivated areas in general, though it is found in banana and rubber plantations, and among low trees. On sunny days it bites from early dawn up to about 8 a.m., reappearing about 4 p.m. and disappearing again at nightfall. On cloudy or rainy days it bites during the whole day, but on bright moonlit nights only if disturbed ; it does not bite on dark nights. The average length of life of this fly is given by Croveri as three months. In North Katanga, Belgian Congo, according to Dr. Schwetz (132), G. pallidipes frequents orchard bush or well-wooded savannah, but, as a rule, is not found in forest ; whether at rest or on the wing, this species, like G. brevipalpis and G. fusca, prefers and is found almost exclusively along roads and paths. Like the two other species men- tioned, it spends the day, or in any case the greater part of it, sitting motionless on the trunks and branches of trees, or on creepers, the attitude in its case, as in that of G. palpalis and G. morsitans, usually being head upwards. G. pallidipes, according to this author, occurs, where it exists, in very large numbers, sometimes almost equalling those of G. morsitans, and it appears to have the same habits as G. brevi- palpis, being active in the evening, when it flits low over roads and paths. It is, however, on the wing earlier and disappears sooner than the larger species, its time of maximum activity being given by Dr. Schwetz as about 4 p.m. or 4.30 p.m., while towards 5 p.m., or an hour before sunset, it vanishes completely. Writing of this species as observed by him in North Mossurise, Portuguese East Africa, in 1918, Swynnerton (145) remarks : — Glossina pallidipes is distinctly more catholic here in its general 1 T. J. Anderson (la), writing of G. pallidipes as observed by him in the Southern Masai Reserve, Kenya Colony, in 1917-18, says that this species was found generally in the bush and fairly close to water, though four individuals were taken at least two miles from water. 54 tastes than either of the other two [G. morsitans and G. brevipalpis}. It occurs both in the country favoured by brevipalpis and avoided by morsitans and in that favoured by morsitans and avoided by brevi- palpis, as well as in wooding of its own ; nor do the stray individuals, at any rate, avoid the extremes. Males were taken on the sparsely shaded basalt .... in dry windy weather and with the leaf falling freely, and occasional individuals of both sexes were taken in primary forest dominated by Khaya nyasica in full leaf " I do not feel that I disentangled its habits sufficiently from those of G. morsitans on the granite-gneiss. Here, in the Brachystegia bush, it appeared to occur in every place in which we found morsitans It was in great numbers only where morsitans was also numerous — namely, at particular vleis and glades ; it was very sparse elsewhere, but less so than morsitans, and both flies (and once brevipalpis, too) were on us or the cattle together. When pallidipes puparia were taken at all, they were under the same log as those of morsitans. " West of the Sitatongas, where morsitans does not appear to occur at all, stray individuals of pallidipes were taken in very open bush— and in every type of bush — but it was never taken even two or three together except in Brachystegia and the less tall of the ' dense secondary ' types. . . . ".Generally speaking, brevipalpis needs coppice with overwood, while pallidipes prefers coppice without it, but in the right types — such as rather poor Brachystegia — it can apparently dispense with yet lower growth .... It may possibly best be described as a ' low- wooding fly/ the low wooding varying from mere coppice and bush savannah to poor Brachystegia, etc., though better wooding is to some considerable extent utilised." In Zululand, where G. pallidipes is found in Acacia thickets, it would seem that this species is extending its range. According to a recent statement by Mr. R. H. Harris (62a), the extension of the infested area is a secondary result of irregularity in burning off the grass, since tracts which once were open country are gradually becoming covered with A cacia, and consequently favourable to the fly. The germination of the seeds of this thorn is greatly accelerated by heat, and after a bush fire young trees soon appear. If the ground were burnt regularly every year, the seedlings would have but a remote chance of survival, but each year that they escape a fire they become more resistant to it. Neglect of burning for a few years, therefore, causes a gradual transformation from open grass to thorn scrub ; the change occurs almost imperceptibly, but its resultant influence on the numbers of G. pallidipes is very marked (compare Swynnerton's recommendation (145) of regulated, annual, late burning as a method of Tsetse control in North Mossurise, Portuguese East Africa.— See pp. 140-141). Glossina longipalpis, Wied. — " Like other members of the Glossina morsitans group, to which it belongs," writes Austen, " G. longipalpis would appear not to be so closely restricted to the immediate vicinity of water as the species belonging to the Glossina palpalis group of ten, although not invariably, are .... At a certain spot on the Oueme River, in Dahomey, where both G. longipalpis and palpalis occur, the latter species, according to Roubaud, is for the most part confined to the immediate vicinity of the water and to the belt of forest fringing the banks ; G. longipalpis, on the other hand, which is scarcely 55 to be found in the actual palpalis zone, predominates in the bush bounding the forest on the outer side. " In Central Dahomey, where, according to Roubaud, the sexes show a well-marked separation or localisation, G. longipalpis occurs in the vicinity of streams and large rivers. ' The males are found only in the clumps of brushwood along the inner edge of the forest belt near streams ; the females are to be met with in open clearings where there are acacias and mimosas/ Roubaud states that in Dahomey G. longipalpis is abundant during the rains, but seems to disappear almost completely in the dry season, especially after the bush is burnt ; its habitat is more restricted than that of either G. tachinoides or G. palpalis, both of which are also found in the same district, and it mingles with its congeners less than does either of the two species mentioned. " In the Western Province of Ashanti, according to Kinghorn, G. longipalpis ' is essentially an open country fly and is not found in the forest belt/ ' The same fact is expressed in somewhat different terms by Dr. J. J. Simpson, who, writing of the Gold Coast in 1914, says that G. longipalpis "is found in the denser parts of the savannah forest, and so far has not been seen south or west of the line dividing the monsoon forest from the savannah, while it occurs along the coast in the savannah forest area." In a recently published memoir (1230), Roubaud states that, in French West Africa, the habitat of G. longipalpis is " closely and necessarily " connected with watercourses. " One meets with this fly/' he continues, " in or on the margin of bush which, though of little density, is always green and damp, forming, as it were, the outer edge of the strips of forest, bordering the savannah. Without absolutely entering the edging of forest fringing the stream itself, this species scarcely forsakes it, remaining always in proximity to the permanently moist zone." Glossina fusca, Walk. — " Dr. A. Kinghorn, who has observed G. fusca in Ashanti, states that as regards its habitat this species resembles G. longipalpis much more closely than G. palpalis. ' It was usually found just at the fringe of patches of bush, and it was rare to catch more than one specimen, or at most a couple. It feeds not uncommonly at night, and may be met with in dull or rainy weather ' " (Austen). The similarity of G. fusca to G. longipalpis in habitat was also noticed by Dr. J. J. Simpson (142 in), who, writing with reference to Southern Nigeria, remarks : — " The distribution of G. fusca is almost coterminous with that of G. longipalpis and its habitats are similar. It favours dense vegetation and a moderately moist climate." In a subsequent paper Simpson says :— " Glossina fusca . . . . is to be found in Sierra Leone in the regions of densest forest growth ; in fact .... the delimitation of the forests is at the same time a delimitation of the areas where Glossina fusca occurs." This opinion is confirmed by Dr. S. A. Neave (103), who met with G. fusca " on the western boundaries of Uganda," and writes : — " From a limited experience of this insect it would appear to be essentially a dense forest, rather than a riverine, species, and evidently delights in very deep shade. Though occurring in many places with Glossina palpalis, the distribution of the two species is by no means coincident, since G. fusca seems to be able to exist at considerably greater elevations and in much cooler localities. In Uganda the limit for this species seems to be about 4,500 ft., as compared with rather under 4,000 ft. for G. palpalis in the same region. It is of course 56 also numerous at much lower elevations, e.g., the forested portions of the Semliki Valley at under 2,500 ft." According to Dr. H. L. Duke, in Western Uganda G. jusca preys to a considerable extent upon the hippopotamus. In the North Katanga district of Belgian Congo, according to Dr. Schwetz (132), G. fusca is restricted to forest, and usually, though not always, to the vicinity of water. Like G. brevipalpis and G. pallidipes, whether at rest or on the wing it prefers and is found almost exclusively along roads and paths. Again, like the other two species mentioned, it spends the day, or at any rate the greater part of it, sitting motionless on the trunks or branches of trees, or on creepers, the resting attitude, as in the case of G. brevipalpis, being head downwards. G. jusca is stated by Dr. Schwetz not to occur in swarms like G. brevipalpis, although where it exists it is often present in numbers. It is a more bloodthirsty species than G. brevipalpis, and much more silent on the wing ; although it often tries to bite at any hour of the day, the night is its real period of activity, and the time when it is most lively is about two hours after sunset. Glossina Juscipleuris, Austen. — Beyond the fact that the type was found in the Ituri Forest, North-Eastern Belgian Congo, and that it has been taken in forests in Uganda by both Dr. R. Van Someren and Dr. S. A. Neave (see p. 18), little or nothing is yet known as to the biology and habitat of this species, which, as we have already remarked, appears to be very widely distributed, though by no means common. Glossina nigrofusca, Newst. — Of this Tsetse-fly, again, we know practically nothing. According to Austen, in Ashanti Dr. W. M. Graham met with a specimen in a bush path. Austen adds that G. nigrofusca " sometimes enters houses and will even attack the inmates," appearing "to settle by preference on the legs." Glossina tabanijormis, Westw. — Not much more can be said under the present heading in the case of this species. In February 1917, and in March of the following year, three specimens of G. tabaniformis — all males — were taken by Dr. Schwetz (132) and his fly-boys in North Katanga, Belgian Congo, in a narrow strip of forest some twenty yards in breadth on the Buitshi stream, between the Rivers Lomami and Sankuru (Long. 24°-25° E., Lat. about 5°S.). In each instance the insect was resting head downwards on a tree-trunk ; of the two specimens captured 20-23.iii.1918, one was caught about 5 ft. from the ground shortly after sunrise, and the other in the same position at sunset on the following day. Glossina brevipalpis, Newst. — In his resume of what had been recorded as to the bionomics of this Tsetse-fly down to 1911, Austen writes : — " In view of the possibility that G. brevipalpis may ere long be proved to be a carrier of human trypanosomiasis, and the fact that in the German East African littoral, at any rate, it is stated to be the chief disseminator of Tsetse-fly disease among domestic animals,1 special importance attaches to a knowledge of the present species. Although many important details have doubtless still to be recorded, 1 Dr. S. A. Neave (103), however, writing on G. brevipalpis a year later remarks : ' The evidence as to whether this species is a carrier of trypanosomiasis is at present very conflicting. It must be remembered that it frequently occurs in company with G. pallidipes on the east coast of Africa, or with G. morsitans in Nyasaland and Northern Rhodesia. On the other hand, where it is the only known species of Glossina, as in the country to the north and north-west of Lake Nyasa, which is full of cattle, there is no definite evidence of trypanosomiasis among the stock." 57 thanks to the observations and writings of Drs. Franz Stuhlmann,. J. B. Davey, and Meredith Sanderson, more is already known con- cerning the habits of G. brevipalpis than about those of any other Tsetse, with the exception of G. palpalis and morsitans. Stuhlmann . . . besides furnishing a detailed description of the anatomy and histology of the internal organs, has also provided us with an interesting account1 of G. brevipalpis as observed by him in German East Africa [now Tanganyika Territory], both under natural conditions and in captivity, and the following resume is based upon his statements and those of the English writers mentioned above, whose observations were made in the Nyasaland Protectorate. " Glossina brevipalpis, which is often met with in very large numbers, is found, at altitudes varying from sea-level to some 1,700 ft., as a rule only where there is abundant shade and protection from wind, its favourite haunts consisting of bush mixed with creepers and young forest trees, either close to watercourses (dry or otherwise), or at any rate within a few hundred yards of water. Stuhlmann has encountered the fly close to the sea, near Dar es Salam, and Davey records the occurrence of a single specimen amongst ' bango ' 'reeds a few yards from the edge of Lake Nyasa. Sanderson has been informed by natives that during the rains (i.e. about January), ' at which time practically the whole country is under water/ the species is very prevalent in North Nyasa all over the grassy plain lying between the shore of the lake and a line of foot-hills some ten miles awa}^, but this statement requires confirmation. Like G. palpalis and other species of Tsetse, G. brevipalpis will occasionally follow cattle and other animals to some distance from its usual haunts ; Stuhlmann states that in this way, especially during the hot weather, isolated specimens were frequently found for a time among the mountains of East Usambara, at an altitude of from 800 to 1,000 metres (2,600 to 3,250 ft.), while, during the period from December to April, others have often been met with in settlements such as Kwamkoro and Amani. Davey records the capture of a single individual ' on the finger of a native standing under a tree in the middle of a small village/ and in another village Sanderson once caught two specimens in a hut occupied by natives, and also containing a cow infected with trypanosomiasis." From his observations "in several localities in Nyasaland," Davey concludes that G. brevipalpis occurs either "in considerable numbers over a limited area," or "in very small numbers," but he admits that the visits on which these conclusions are based were made in the wet and the dry seasons, respectively. The fly was " most numerous under the largest trees." Writing in 1912 on G. brevipalpis as observed by him in Eastern Tropical Africa, Dr. S. A. Neave (103) remarks :— " This species seems to be, partially at least, dependent on the presence of water or a moderate degree of atmospheric humidity. It is therefore only found in river valleys and amongst a considerable amount of cover and shade, much more than would be necessary for G. pallidipes, but less than for G. palpalis. In the height of the dry season it seems to be found only in the immediate proximity of water or cool and damp river beds. It prefers comparatively low country, and I do not know any record of its occurrence at much over 3,000 ft." 1 Austen's version of this has already been drawn upon more than once in the compilation of the present volume. 58 As regards G. brevipalpis in the North Katanga district of Belgian Congo, Dr. Schwetz (132) confirms and amplifies most of the observa- tions by Sanderson and Davey in Nyasaland recorded above, though, unlike Sanderson, he never found the fly resting " under the leaves of bushes, or in the grass." Whether at rest or on the wing, G. brevipalpis, like G. pallidipes and G. fusca, according to Schwetz, prefers and is found almost exclusively along roads and paths. Where it exists, this species occurs indiscriminately in forest, orchard bush, and even well-wooded savannah, and, like the other two species mentioned, it spends the day, or in any case the greater part of it, sitting motionless on the trunks or branches of trees, or on creepers, its resting attitude, like that of G. fusca, being head downwards. G. brevipalpis is stated by Dr. Schwetz, who was the first to meet with the species to the west of the Lualaba, to be less ravenous than its congeners, although the females are much more bloodthirsty than the males. Specimens taken on the wing were almost invariably of the latter sex ; on tree trunks females were very often caught, though never in greater numbers than 10 per cent. G. brevipalpis, which often occurs in swarms, does not fly high, but flits low over the ground ; though it may be seen on the wing in the early morning, its real period of activity is from about half an hour before to half an hour after sunset. It is interesting to note that on several occasions in the afternoon Dr. Schwetz saw a male G. brevipalpis settle upon a leaf of some low-growing plant (especially Amomum sp., a very widely distributed member of the Zingiberaceae, known to the natives as " Mantungulu "), bury its proboscis therein, and suck. Lastly, with reference to the habitat of this species as observed by him in North Mossurise, Portuguese East Africa, in 1918, Swynnerton (145) writes :— Glossina brevipalpis, so far as I have seen (and I was with it for some weeks in all), relies very greatly indeed on shade, and is rarely found away from fairly heavy shade. The requisite degree of shading is provided by wooding with leafy undergrowth. This may be either primary forest (in which the undergrowth is sometimes such as to give sufficient shelter alone . . . .) or secondary bush in leaf with thickets and sapling clumps below. Of these types the primary forest . . . and, in many places and most seasons, heavy Brachystegia bush are the best capable of carrying this tsetse through the dry season. It is not at all dependent on the presence of vleis, and I have found it waiting in all the thickets at some distance from water of any kind and in hot weather in September with the ground baked. It is true that it was in greater numbers near certain little streams than in the Brachystegia thickets, but (as results from the same type of forest away from streams appeared to show conclusively) this was only because those streams were lined with fringing forest of primary type. " So dependent does the fly appear to be on good shade that, except in the early morning, after sunset, and on dull days, it will leave animals it is on as soon as they emerge from the shady bush into the sunlight. On dull and rainy days it will follow freely into the most open country and at high noon, so that the term ' crepuscular/ which has been applied to this fly, is not altogether justified by my observations. Thermo- meter readings taken at the same time in primary forest, primary forest with its undergrowth cleared, and, thirdly, in a sapling thicket in Uapaca wooding alongside, were identical, yet brevipalpis had deserted the cleared piece. This, with a failure to take brevipalpis in some 59 high, dark secondary wooding on the Sitatonga base, excepting at the occasional thickets, suggests that extra shade may not be the only advantage gained from thickets. Protection from drying winds, additional protection from the eyes of enemies (such as the thickets very definitely afford) and some advantage to the pupae are the three that occur to me, but I am convinced that shade is the chief consideration." Glossina medicorum, Austen. — Of this species we still know nothing more than we did eleven years ago, when Austen (3) wrote that " it would seem that it sometimes moves about after dark, at any rate, when attracted by light " Glossina longipennis, Corti. — Few observations have yet been made upon this Tsetse-fly, which, as has already been stated, is in the habit of entering carriages on the Uganda Railway at night, in company with G. brevipalpis and G. pallidipes. Dr. S. A. Neave, to whose statements on the subject we have previously referred, writes (103) : — " This is a desert-haunting species, confined, so far as at present known, to North-Eastern Africa. It is widely spread over the lower-lying and drier regions to the east and north in the British East Africa Protectorate [Kenya Colony], and will very probably be found extending into the Uganda Protectorate in the country to the south-west and west of Lake Rudolph. It appears to be absent from the sea-coast, where the climate is probably too humid for it. It would seem to be entirely independent of water, and indeed rather to avoid it.1 I found it most striking, when travelling from station to station on the railway between Voi and Makindu, to find numbers of this species in the dry, semi-desert, thorn-scrub country between the rivers, while on the river banks it was replaced by G. brevipalpis. " Like the other large species of Glossina, it is chiefly on the wing .and inclined to feed in the early morning and late evening. It is probably the species which most frequently enters the railway carriages on the Uganda Railway at night, that being the time when the principal trains traverse the region between Voi and Makindu." In elevated districts G. longipennis is met with between the rivers and not upon their banks (Chalmers and King) . To conclude the foregoing sketch of what is known as to the habitats •of the different species of Tsetse-flies, which, though brief, is necessarily somewhat diffuse owing to divergences of view among the authorities cited, we reproduce the following " classification according to environ- ment," drawn up by Dr. S. A. Neave (103). Although restricted by its author to East African species of Glossina, it seems to us that the scheme might also be made to include in a general way the entire genus. "A. Requiring a great degree of atmospheric humidity :— A.I. Requiring a high temperature. G. palpalis, R.-D. A. 2. Not requiring a high temperature. G. fusca, Walk. " B. Requiring only a moderate degree of humidity :— B.I. Requiring comparatively little cover. G. pallidipes, Aust. B.2. Requiring fairly heavy timber and bush. G. brevipalpis, Newst. B.3. Requiring more or less dense forest. G. austeni, Newst. "C. Independent of water and most active in a dry atmosphere. G. morsitans, Westw., and G. longipennis, Corti. 1 Curiously enough, in the Southern Masai Reserve, Kenya Colony, in 1917-18, •G. longipennis was found by Anderson (la) unexpectedly abundant on moist, swampy flats and on the bank of a river. 60 " The three large species, G. brevipalpis, G. longipennis and G. jusca, may also be separated from the others by their being mainly crepuscular or nocturnal in their habits, instead of being most active in sunshine." CHAPTER VI. DISTRIBUTION OF TSETSE-FLIES IN BELGIAN CONGO. Since Belgian Congo has been comparatively little visited by British entomologists, we think it may prove useful to devote a few pages to the distribution of Tsetse-flies in that country. For a considerable portion of the statements printed below we are indebted to the remark- able report on the Travaux de la Mission Scientifique du Katanga, by Drs. J. Rodhain, C. Pons, F. Van den Branden and J. Bequaert (116-118), published at Brussels, in 1912, though we have added certain data derived from other publications. DISTRIBUTION OF G. palpalis. — As is well known, G. palpalis abounds, throughout the entire Congo basin, with the exception of Upper Katanga. Lower Congo. — According to Dr. J. Bequaert, G. palpalis is the only representative of its genus found in Lower Congo, where it haunts the numerous narrow strips of forest which traverse the savannah, fringing the watercourses and decking even the smaller valleys. In these belts of leafy and luxuriant vegetation, the fly finds a humid atmosphere and kindly shade, and, thanks to these biological conditions, it pene- trates into the interior to a considerable distance from the banks of the Congo. At Kisantu on the Inkisi, 37J miles from its mouth, Dr. Bequaert met with G. palpalis in abundance, and he thinks that it would be interesting to determine the distribution of the insect along the various rivers with which the Lower Congo region is seamed, taking into account the influence of the seasons. If this were done,, we might perhaps succeed in discovering some biological peculiarity in the distribution of the fly, which could serve as a starting-point in an effective campaign against it. Central Congo. — Dr. Bequaert has recorded observations made by him upon G. palpalis, in the course of a journey from Leopoldville to* Stanleyville. The greater part of the route traverses the equatorial forest, in which the hot and humid atmosphere, coupled with the dense shade on the banks of the river, constitute the ideal biological conditions required by this species of Glossina. The result is that the fly is extremely abundant everywhere, and not a day passes on which it does not come on board the steamers in numbers. At Ponthierville and in its immediate environs, however, Dr. Bequaert was surprised to find that G. palpalis apparently did not exist. It may therefore happen, even in the midst of the equatorial region, that there are places from which G. palpalis is absent, although they seem to include all the biological conditions necessary for the life of the adult fly. This is doubtless due to the nature of the soil, which, around Ponthierville, consists of red clay, and is therefore incompatible with the existence of breeding places, since the latter appear always to be established in spots which are sandy, dry, and sufficiently shaded. Above Ponthierville, as far as Kindu, G. palpalis is much more rare ; the biological conditions on this portion of the banks of the Lualaba are apparently less favourable than those on the banks beginning at Bangala. 61 Manyema. — In Manyema the banks of the Lualaba are at the present time to a large extent denuded of arborescent growth ; conse- quently in this section G. palpalis is relatively rare, or even in many cases absent (e.g., at the Nyangwe post). This is in no way due to the climatic conditions of the country, since wherever there is a dense and luxuriant growth of trees along the watercourses Tsetse-flies reappear. Katanga. — In Katanga, so far as regards Belgian Congo, the range of G. palpalis attains its most southerly limit ; the latter has been determined with great exactness by Dr. S. A. Neave, who mentions the following localities as the southernmost points at which this species is found : — Kapepwe's Village, on the Luapula (11° 30' S.) ; Gombela's Village, on the Kafira (10° 50' S.) ; Tsinika's Village, on the Lufira (10° 50' S.) ; Shara's Village, on the Dikuluwe (10° 30' S.) ; Koni's Village, on the Kalule (10° S.) (Dr. Goessens) ; Nzilo Gorges, on the Lualaba (10° 30' S.) ; Ndzimina's Village, on the Lufupa (10° 25' S.) ; and Chianda's Village, on the Lubudi (10° 40' S.). It will be noticed that the most southerly points at which G. palpalis occurs are usually situated about 10° 30' S., except on the Luapula, where the fly reaches 11° 30' S. Dr. Bequaert was able to satisfy himself by personal observation that the limit mentioned is in accord- ance with facts ; nevertheless he states that there is no reason why G. palpalis should not one day invade Upper Katanga, and even penetrate into the basin of the Zambesi. In the valley of the Dikuluwe and its affluents, M. M. Vermeesch, who in 1914 was entrusted by the Government of Belgian Congo with the task of surveying and examining the region in question, ascertained the presence of G. palpalis on the Lower Pande, the Dikuluwe, the Lower Kabango, the Luvilombo and the Lufira, where it is found especially in the scrub, as well as beneath certain thorny trees and bushes with very dense foliage. Valley of the Lukuga and the shores of Lake Tanganyika. — According to Dr. J. Schwetz, all the western affluents of Lake Tanganyika come from the mountains, with the result that in this region Tsetse-flies are not met with until within quite a short distance of the lake shore. This, however, does not apply to the valley of the Lukuga ; the river contains much water, and its banks are covered with luxuriant vegetation, affording excellent cover. Consequently G. palpalis is to be found all along the valley, albeit close to the river and its banks, while G. morsitans occupies the remainder of the valley and the margins of Lake Tanganyika. Between the western shore of the latter, where G. palpalis exists, and the rivers Lualaba and Luvira, where it is likewise found, the intervening district is very mountainous and free from human trypanosomiasis. DISTRIBUTION OF G. morsitans. — According to Dr. Bequaert, little is yet known as to the northern limits of G. morsitans, which is very common in Katanga. Dr. Bequaert found that this species does not exist at Kongolo (Lat. 5° 30' S.), but he met with a number of specimens of it at Kabalo (Lat. 6° S.), further up the Lualaba. On the other hand, we know from Dr. Schwetz's observations that G. morsitans occurs in abundance in the region of the Lukuga near Lake Tanganyika, but we are absolutely ignorant as to how far the range of this Tsetse- fly extends towards the north, in the country lying between Lake Tanganyika and the Lualaba. 62 We are somewhat better informed as regards the western limit of G. morsitans, which between the sixth and eighth parallels, in Dr. Bequaert's opinion, is represented approximately by the course of the Lualaba, on the left bank of which the fly would appear to exist only in sporadic colonies. Further south conditions are different, and beyond the eighth parallel G. morsitans has invaded the left bank of the river for fairly long distances ; it occurs in abundance at Kikondja (between Lake Kisale and the Lovoi River), and throughout the region comprised between the Hakansson Mountains and the Lualaba. Below the ninth parallel this Tsetse even extends slightly beyond the Lubudi ; its western limit in this region seems to lie somewhere about Long. 25° E. Information is lacking as regards the most southerly part of Katanga, between the Lualaba and Dilolo ; all that is known is that G. morsitans does not exist at Kayoyo, or in the intervening district between this post and Kinda post (Dr. Rodhain). Dr. Bequaert adds that it is a very remarkable fact that, while G. morsitans is generally absent from the wide tracts of country lying between the Lubudi and the Lubilash (Sankuru), Dr. Rodhain found an isolated colony of this species near the Nzie (a small affluent of the Lubilash, about Lat. 9° S. and Long. 24° E.), covering a space of 30 kilometres (18| miles). The Nzie colony represents the most westerly point at which G. morsitans has been met with in Katanga. It is probable that, beyond the regular area of distribution of the species, there are other analogous spots at which the fly makes a sporadic appearance, and it is eminently desirable that the entire south-western region of Katanga should be explored in this connection by a pro- fessional entomologist, capable of recognising the biological conditions suited to G. morsitans in this country. In Katanga, G. morsitans frequents by preference the savannah forest, where it is found even in the driest places and far from water, although it is also actually met with on the rivers. It may be added that, as a general rule, Dr. Bequaert finds the habits of Tsetse-flies much more elastic than they are represented to be by the majority of authors. In Katanga G. morsitans may be said to be fairly uniformly distributed everywhere, while in a given place the insect may vary in numbers from one season to another, and even doubtless from one year to another. It would be difficult to formulate a rule on the subject, or to delimit fly-belts properly so-called. On the other hand it may be asserted — and the fact is of considerable practical importance — that the highest parts of the tablelands of Katanga are entirely free from G. morsitans. The fly is not met with above 5,250 ft. in Manika (Biano), Kundelungu and Marungu ; in Biano, to be precise, Dr. Pons found the last G. morsitans at an altitude of 5,216 ft. According to statements published a few years ago by M. Bovone,1 a veterinary officer, if G. morsitans does not exist in Marungu, a region of lofty, dry and denuded plateaus, unsuitable for Tsetse-flies, it nevertheless occurs in the regions adjoining this massif, especially to the north, at Mulonde, along the Baudoinville road, and to the south- west, where it has been encountered on the margins of the Lukifva, seven and a half miles from the Marungu boundary. Lastly, it has also been observed to the west of Lusaka village and along the Lufuko, while M. Bovone has even met with it on two occasions at Mutambala's 1 Bovone, " L' Agriculture et Felevage au Marungu " : Bull, agric. du Congo beige, Vol. v., no. 3, p. 457 (Sept., 1914). 63 farm, on the extreme edge of Marungu. This, however, was certainly an accidental occurrence, coinciding with the arrival at the farm of native caravans. During his agricultural prospecting expedition, in 1914, M. M. Vermeesch observed that Glossina morsitans infests the entire eastern slope of the Bianos, and is very widely distributed throughout the basin of the Dikuluwe and its affluents, the only exceptions to this rule [ being the Upper Pande (Kapiri-Kansenia), the Upper Loami and the [ Upper Movia. Throughout the entire northern portion, including the (• basins of the Kabengo-Lukenshi, Luvilombo-Kabinde, Lisele-Kanianga, Dikuluwe and Lufira, G. morsitans is even very abundant, and M. :, Vermeesch does not consider that its disappearance in the immediate ! future is at all likely.1 It is much to be wished that the attention of medical officers, veterinary surgeons and agricultural officials may be drawn to the study of the distribution of Glossina morsitans in Katanga. Those i who desire to contribute to the elucidation of this subject should never ( omit to collect, and send to Europe for identification, specimens of the ' fly from each locality where it is observed, since by so doing possible confusion with other species of Tsetse-flies will be avoided. DISTRIBUTION OF THE OTHER SPECIES OF TSETSE. — G. pallidipes, Austen. — We have seen, in Chapter II. , dealing with the general distribution of Tsetse-flies, that the range of G. pallidipes includes the south-eastern and eastern parts of the African continent. Dr. J. Bequaert, in the course of his own investigations, never met with this fly, but a number of specimens of it, taken on the Lualaba (Eastern Province) a few hours north of Kongolo, were sent to him by Dr. Russo. G. pallidipes appears to be fairly widely distributed throughout North- Eastern Katanga, and, since in behaviour and general appearance it closely resembles G. morsitans, it is possible to mistake it for the latter species. For more detailed information concerning the occurrence of G. pallidipes in North Katanga, the reader should consult the valuable series of papers by Dr. Schwetz (132). G. Jusca, Walk. — This species was taken by Dr. Schwetz in the valley of the Lukuga, at a spot near Kanhogo, which lies on the bank of the river ; a little later, with the assistance of his boys, he caught a large number of flies, and found that the bag consisted of a mixture of G.fusca and G. morsitans, in the proportion of 1 to 3. When he drew near to Mikato's village, he found that G. fusca became gradually scarcer, disappearing completely when the place was reached. There was nothing distinctive in this portion of the Lukuga Valley, and yet this was the only part of it in which G. Jusca was discovered by Dr. Schwetz. G. brevipalpis, Newst. — It is stated by Dr. Bequaert that all the large Tsetse-flies collected by the Mission Scientifique du Katanga, at Kibombo, Kassongo (Manyema) (Dr. Pons), Kongolo (Dr. Rodhain), and in the Luena Plain, one hour from Likongo (Sankisia) (Dr. Rodhain), have been identified by Prof. Newstead as belonging to this species. In the vicinity of Kibombo, G. brevipalpis seems to occur more or less everywhere, but it is always extremely rare, and one may live in a district for a long time without even suspecting its presence. 1 M. Vermeesch, " Les Vallees de la Dikuluwe et de ses affluents, au point de vue de leur valeur agricole " : Bull Agric. du Congo beige, Vol. v, no. 3, p. 516 Sept., 1914). 64 Thus, it was only after the Mission had been established in the Bukama-Sankisia region (Katanga) for seventeen months, during which a very large number of Tsetse-flies had been collected, that Dr. Rodhain was fortunate enough to capture a solitary specimen (a male) of G. brevipalpis, enabling him to establish the existence of the species in the centre of Katanga. As regards the distribution of this species in North Katanga, much information will be found in Schwetz's papers (132). The habits of G. brevipalpis, as observed by Dr. Bequaert at Kibombo, were very different from those of G. palpalis. The insects were met •with in the forest, a long distance from any water, and usually made their appearance about 5 or 6 o'clock in the evening, when they were to be seen on leaves or on sandy paths. In these places they were never met with in the morning, or about 2 or 3 o'clock in the afternoon. It likewise appeared to Dr(. Bequaert that this species is little inclined i to attack man (vide also Swynnerton's statements (145) with reference to G. brevipalpis in Portuguese East Africa, p. 76), which perhaps to some extent explains the difficulty found in proving its existence in a district. With reference to the other species of Tsetse, namely, G. pallicera, Big., G. longipalpis, Wied., G. fuscipleuris, Austen, G. nigrofusca, Newst., G. tabanijormis, Westw., G. severini, Newst., and G. schwetzi, Newst., reported as existing in one or other part of Belgian Congo, we possess no records of occurrence other than those already given in : Chapter II., p. 11 ci seq. CHAPTER VII. POINTS IN BIONOMICS. Seasonal Migrations — Range of Flight — Travelling Habits — Resting Surfaces — Proportion of the Sexes. We have included the subjects mentioned above in the same chapter, since the majority of them are more or less intimately related, while in some cases they are still far from being elucidated. The seeker after knowledge concerning Glossina will find himself -confronted with a number of questions such as the following : — Upon what do the seasonal variations in the numbers of Tsetse-flies depend, .and what are the exact relations between these variations and the meteorological conditions ? Is the diminution in numbers in the course of the dry season due to the scarcity of the animals whose blood serves fo/food, to destruction by birds or other enemies, to lessened production of larvae, or to the prolongation of the duration of the pupal period ? In the event of a prolonged drought, is it possible that all the adult flies perish, and that the perpetuation of the species is ensured only by the hatching of old pupae at the commencement of the rainy season ? What is the reason why Tsetse-flies are sometimes met with on one of the banks of a river, and not on the other ? Is this due to the soil in the one case being more favourable for the deposition of larvae, or to other causes ? On the other hand, why do male flies usually appear to be in the majority ? Does their number vary with the locality, as in the case of G. palpalis, and does it depend upon the situation of the breeding > 65 places ? Is there an actual predominance or disproportion, or are the males simply easier to catch than the females ? It is impossible here to provide answers to all these conundrums, some of which, moreover, as already stated, are still far from being solved. We shall therefore confine ourselves in the present chapter to dealing with the principal ones among them, and certain kindred questions. SEASONAL MIGRATIONS. Statements by Messrs. LI. Lloyd and R. W. Jack, with reference to the dry season concentrations of G. morsitans in Northern and Southern Rhodesia, have been dealt with in Chapter V. (see p. 49). Similar seasonal migrations occur in the case of other species, such as G. brevipalpis, and, as regards West African forms, we give below some interesting observations upon this subject by Dr. E. Roubaud (I23a), taken from a recent publication by this authority dealing with the Tsetse-flies of French West Africa. " The climatic and hydrological modifications that supervene in the course of the seasons," writes Roubaud, " exert considerable influence upon the geographical distribution of Tsetse-flies, and upon their relative abundance in a given locality. Thus, should it have been impossible to discover the presence of any of these insects at the crossing of a watercourse in the dry season, even though the stream had not entirely disappeared, it would not necessarily follow that Tsetse are never to be found at that spot ; indeed, it is even probable that in the winter season the same watercourse might be infested with Glossina to a serious extent. " Generally speaking, at the commencement of the rains a sudden apparition of the different species of Tsetse-flies is seen, and the insects all at once become numerous in many localities whence they had more or less completely disappeared in the course of the dry season. The phenomenon occurs not only in the case of riparian species, but also, though not so markedly, among the Tsetses of the savannahs. In each instance it is dependent upon the seasonal hygrometric modifications. " The fact is that the onset of the rains, with the profound modifica- tions in the hygrometric state of the air occasioned thereby, and their repercussion upon the hydrographic condition of the watercourses — which is, as we know, very great in French West Africa — produce migrations of Tsetse-flies. These migrations, when determined by the pluvial factor, are generally directed from the north towards the south in the case of the Sudanese species, and from the south towards the north in that of the species of the coastal belt. The flies established beside the watercourses of the Sudanese zone in the dry season are drawn towards the south before the coming rains, at the epoch of the rising of the waters, which is influenced by the earlier rains of the southern regions. Conversely, when the period known as ' the little dry season ' supervenes in the coastal regions, the Tsetse-flies infesting the latter tend to move towards the north in pursuit of hygrometric conditions characteristic of the height of the winter season. Thus it is, for instance, that G. tachinoides, which is absent almost throughout the year in Central Dahomey, nevertheless puts in an appearance there at the commencement of the winter season, during the months of April and May, when it may be met with down to the environs of Abomey (7° N. Lat.), and perhaps even lower still. (5979) E 66 When the rains have definitely set in, the fly leaves this zone of transi- tory dispersion and travels northwards, into its normal zone ; it is found as a permanently resident species in Upper Dahomey, to the north of the llth parallel. " In the case of G. longipalpis, we have found that the opposite phenomenon occurs : established in January-February in the coastal region of Dahomey, it is not until May that this species appears in Central Dahomey, towards the 8th parallel ; there it becomes very abundant at the height of the winter season, in July-August, dis- appearing for good in October and regaining its southern haunts, where it then finds the humidity necessary to it. " The species that definitely prefer dry places, like G. morsitans, appear to react in quite a different manner, since their period of maximum dispersion seems to coincide with the end of the rains and the commencement of the dry season. " In conjunction with the adjustments occasioned by pluvial condi- tions, there is another meteorological factor which has a predominant influence upon the relative dispersion of Tsetse-flies, namely, winds. The east wind, known as the harmattan, which blows in the dry season over a vast portion of West Africa, notoriously accentuates the desiccating effects of the cessation of the rains, and, in consequence, has a considerable influence upon the bionomics of Glossina. Under the dehydrating action of this extremely dry wind, the mean hygro- metric degree falls to an exceedingly low level, and Tsetse-flies more or less completely forsake the haunts in which they are liable to be exposed to its influence. Moisture-loving species, such as G. palpalis, naturally react much more than the drought-loving species, but the latter are likewise not indifferent. " G. palpalis is extremely sensitive to the harmattan, and under its influence it forsakes all the small watercourses where the vegetation is of little density, even when they are not entirely dried up, and : concentrates in haunts affording the best cover, in the vicinity of! large sheets of water. Thus it is that during a large portion of the year this species is scarcely to be found throughout the northern part of its zone, and it exists there only in areas of extremely limited extent. The typical drought-loving species, G. morsitans, however, is also very markedly subject to the influence of the desiccating wind. In the zones in which it is met with, its dispersion tends to become much more restricted from the commencement of March, when the harmattan is in full blast, than towards the close of the winter season, and the fly also shows a greater tendency to take refuge in the vicinity of the streams. " In the case of each species of Glossina, there are thus zones of- permanency (permanent haunts) in which the species persists throughout the year, and zones of dispersion (temporary haunts), in which its. presence is more dependent upon seasonal, meteorological circum- stances. Migrations take place from the zones of permanency to the. zones of dispersion, when the season in the latter becomes favourable to the life of the flies, and conversely .... " Far from being merely a matter of curious interest, the seasonal migrations of Tsetse-flies are of very great practical importance. All ; the causes that inhibit, even temporarily, the dispersion of these insects exert upon the regions concerned an economic influence of primary value, by liberating them more or less completely from the: 67 oppressive domination of this winged scourge. In Sudanese regions, it is during the dry season that the shiftings of flocks and herds, and the movements of caravans and trading expeditions, that give a special character to life in these parts, are rendered possible. The effect of the east wind may consequently be considered as eminently favourable. The harmattan combats the dispersing influences of the winter season, drives back the Tsetses into their permanent haunts, and to some extent cleanses a large portion of the country from the flies that infest it. " The intimate relations existing between the relative dispersion of Tsetse-flies and seasonal meteorological influences serve to demonstrate, once again, how important it is for us thoroughly to understand, before evervthing, the determinism and the nature of these influences, as well as the laws that govern them in the different regions of French West Africa ; and, starting from these data, to follow out the biological manifestations of the species of Glossina." In concluding his remarks upon this topic, the author urges, on these grounds, a closer study of the meteorology of French West Africa. RANGE OF FLIGHT. Dr. J. J. Simpson, who, in the Colony of the Gambia in 1911, studied the bionomics of Glossina palpalis and G. morsitans form submorsitans, writes as follows (142.1.) with reference to the range of flight of the former species. " As is well known, G. palpalis prefers to remain in the shade and near the ground, preferably in low scrub, and darts out into the open in the full sunlight only when in search of a meal. The maximum distance to which a Tsetse will venture into an area with no shade is of great importance in delimiting clearings. Various estimates have been given, but although one has seen Tsetse in camps separated from the bush by over 200 yds. of clearing, it is not fair to assume that this distance was actually and deliberately traversed in. one flight. It is quite possible that these isolated examples followed the natives or horses, a mode of procedure which they often adopt. On one occasion, however, when on board a launch more than 100 yds. from the beach, I saw a G. palpalis fly on board in the cool of the afternoon, and in view of this and similar observations, 150 yds. may be put down as the lowest estimate for clearing, though 200 yds., if possible, should be aimed at." In December 1914 Dr. W. A. Lamborn (82) made some very interesting " preliminary experiments " in Nyasaland, " with a view to determining the range of flight of Glossina morsitans." " Between 4th and 22nd December," writes Dr. Lamborn, " 1,810 males and 312 females were captured in the Lingadzi district, where the elevation is about 1,700 ft., and between 4th and 17th December 854 males and 66 females had been marked and released at two different points — at Chunzi, at an elevation of 2,420 ft., 10 miles due south of the Lingadzi fly area ; and at a spot which has an elevation of 1,950 ft. and is situated two miles south of the Lipimbi River and five miles south of the same fly area. " The flies released at Chunzi were marked by snipping off the first right foot through the middle of the metatarsus, an injury unlikely to occur in nature, and those released at the second spot were marked by snipping through the second foot on the same side. Their fitness seemed to be little impaired by the operation, for test specimens fed 68 on a goat with avidity a few minutes after it ; and it is usual to be much assailed by such flies immediately they are released." No female flies were recaptured, but, between 7th and 22nd December, inclusive, 14 marked males were taken, four of them near or at Chunzi, the remainder at Lingadzi. "Evidence has therefore been obtained," continues Dr. Lamborn, "of 10 flights of five miles and of one flight of no less than 10 miles back to the locality from which the flies were originally taken, and, in the case of three other flies, of flights tending to show that they were on their way back also. The flies recaptured were all taken at the fringe of the fly area, where, on account of the greater probability of meeting with them, the collectors had instructions to work. The country which they necessarily traversed is all well-wooded and without open spaces, and though careful search was made for released flies in directions away from the main fly area, none were obtained. No flies were fed previous to release." In three subsequent series of experiments by Dr. Lamborn in con- tinuation of the foregoing, the system adopted was the same as before. Dr. Lamborn writes (83) : — "In passing from a morsitans area into country apparently free from them, whether bush or open country, it has always seemed to me that the same tsetses which have been hovering round continue to follow for a considerable distance. With the object of deciding whether such flies do really attend one and are not chance new-comers, a series of experiments have been conducted, consisting in liberating marked flies at a definite point, and then after a walk to various distances capturing all those in the immediate neighbourhood/ ' In the first series of experiments, " conducted in the neighbourhood of Domira Bay," 120 male flies were released, and 14 were recaptured at a distance of two and a half miles from the starting-point. In the second series, " carried out at Monkey Bay," 600 males were released, and 83 recaptured at distances varying from half to one and a half miles. The third series of flight experiments was made in the proclaimed area, and consisted in " releasing marked male flies, as before, at a spot two miles south of the Lipimbi river, at an altitude of 1,950 ft., and recapturing as many as possible in the region from which they had been removed, namely Lingadzi, about five miles distant in direct line, with an altitude of 1,700 ft." " In making these experiments," writes Dr. Lamborn, " every precaution was taken to guard against possible fallacy. The flies were liberated by placing the cage in a tuft of grass before opening the door, by which means it is possible for the person conducting the experiment to get away to a distance before any flies are able to follow. In no case did the person releasing them return to the tsetse area, and there was very little probability of any flies returning on the backs of chance passers-by." In all, 1,776 male flies were released at Lipimbi River, on various dates from 3rd February to 16th February 1915, inclusive, and from 4th February to 17th February 17 flies were recaptured. Flight experiments conducted by Dr. J. J. Simpson (144) in the Gold Coast, with marked individuals of G. tachinoides, yielded the following results : — The greatest distance covered by a single fly was four miles ; all seemed to return to water, as none were caught at a greater distance from the river than the points where they were liberated ; if streams intervened between the place of liberation and the river, the flies made their way down stream to the river ; if the stream 69 bed had shady banks, but no water, the flies congregated there and did not attempt to reach the river ; not a single tsetse was found in the open bush bordering the road at a greater distance than 200 yards from it, and there was practically no discrepancy in the proportion of the sexes recaptured. TRAVELLING HABITS. It has been shown by Fiske (54) that individuals of G. palpalis, far from remaining always at the same spot, are in constant movement along lake shores and similar places. One of the experiments con- ducted by the author in question on Lake Victoria, Uganda, consisted in catching G. palpalis during five days, four of which were consecu- tive, in November 1913, on a narrow spit of land (" Crocodile Point ") at one extremity of Bulago Island. In the four consecutive days " more than 2,000 flies were caught," but " neither density nor female percentage underwent notable change." " Notwithstanding the semi-isolated position of Crocodile Point," writes Fiske, " as many flies penetrated its area each day as were caught on it, leaving no other conclusion possible than that, if no flies had been caught, as many would have moved away -from it. Otherwise stated, the fly population of this region was so far from permanently fixed there that hardly any individuals sojourned there for more than a single day. ' The experiment of the 5th to 13th November 1913 was repeated on the 17th to 22nd January 1914 . . . . and completely confirmed it. The conclusions were subsequently confirmed in various other ways, and there is no doubt that the flies of this species move freely about from place to place, forming continuous streams of fly traffic along the shores of lakes, banks of streams, and, it was subse- quently ascertained, along the borders of woodland, game trails or human pathways, etc." . Data furnished by the author show that by catching fly for consecu- tive hours at points on the Lake shore the density of males was reduced, but not that of active females, and Fiske observes : — ' The only conclusion that can be drawn is that the active females habitually move along shore much more rapidly than the males, or than many of the males. This is entirely in accord with the conclusions reached through experiments .... that active females are hungry, and actively seeking food, and that degree of activity is correlated with abundance or scarcity of food. Their movements along shore are stimulated by hunger. The movements of the males are in part stimulated by hunger, but also in part . . . by sex instinct. The object of the females — and of such males as require it — is to seek food ; the object of many of the males (forming a majority when females are inactive and the female percentage low) is merely to seek the females, and this is accomplished by loitering along the routes most freely followed by the food-hunting flies." On two other occasions, on each of which catches of fly were made at two different points close together, the character of the adjacent shelter or type of vegetation, which differed in each case, was found to have an extraordinary effect upon the percentage of females in the catch. After giving statistics, Fiske observes that " in both cases the high percentage of females is associated with a type of vegetation known to be especially repugnant as shelter and the low percentage with a type of vegetation known to be attractive. 70 " A hypothetical explanation for this phenomenon," continues the author, " which has withstood all tests applied to it, is as follows : — " (a) The body or mass of active flies is continually in movement, and streams of flies are continually passing points along shore . . . I " (b) These streams of moving flies are made up of (1) food-hunting flies of both sexes, which compose a variable proportion dependent upon abundance of food and which move rapidly ; and (2) male flies which are not seeking food, but which frequent the routes followed by food-hunting flies, and which move much more slowly " (c) The relatively idle and lingering males tend to prolong their sojourns at points where sheltering vegetation is of the most attractive type, and to pass quickly, or not at all, by points where the vegetation is of an unattractive type. It follows that (1) density of active males will be greatest where shelter is most attractive, and least where it is least attractive, and (2) that the percentage of females amongst caught flies will be greatest where shelter is least, and least where shelter is most attractive to the lingering males (because food-hunting flies must, for several reasons,1 consist principally of females)." It is perhaps advisable to point out that the object of the experiments described above under the previous heading (RANGE OF FLIGHT) was to determine the distance which Tsetse-flies are capable of covering by their own unaided efforts. Without here entering into a discussion of the distance to which G. morsitans may accompany big game, such as buffalos,2 it may be mentioned that male flies in search of females often travel for several miles on the backs of porters or the bellies of domestic cattle. Thus Swynnerton (145), in Portuguese East Africa, twice found G. morsitans " following" man in this way for "rather more than six miles, and many times various less but still long distances. On one occasion, " he adds, " a journey of over six miles was done in three instalments by a marked fly on three successive days." The same writer refers to a record by another observer " in which tsetses were still on his cattle 25 miles outside the fly-belt." Swynnerton also " measured one of these rides [on cattle] by five male brevipalpis in dull weather, first marking each of them with a dab of white paint. The flies had come to us," he writes, " just before and were following, not feeding. The greatest distance travelled was 5 J miles by one fly that stayed to the end. All five flies were still present at 4 miles and 700 yards. Every kind of country was traversed, brevipalpis bush, simple coppice and open grass. The flies deserted us on our getting back into their native bush." It should be added that Swynnerton states that he has met with " no evidence at all that either sex travels appreciably on its own initiative." 1 In a foot-note Fiske writes : — " These reasons include the following : — " 1st. Because the females in any district or region appear to outnumber the males " 2nd. Because the females must nourish their young as well as themselves, and most probably require food somewhat more frequently, in nature, than males. " 3rd. Because the males are normally active at all times during good weather, and are more apt than the females to encounter host animals without specifically seeking for them." 2 That big game, especially the buffalo, is mainly responsible for the annual spread of G. morsitans and other species of Glossina each rainy season is stoutly maintained by certain writers, e.g., Swynnerton (145); others, however, such as Christy (29) and Jack (75), while admitting the movement of the fly, are strongly inclined to doubt its alleged habit of migrating with the game (see pp. 100, 104). 71 RESTING SURFACES. In Nyasaland it was found by Lamborn that G. morsitans displays a preference for baobab trees as resting places, although large trees of any kind are utilised. The predilections of the species mentioned, as well as of G. brevipalpis and G. pallidipes, as regards actual resting surfaces were tested experimentally in Portuguese East Africa by Swynnerton (145), who writes : — " When studying morsitans, and later brevipalpis, I erected a large mosquito net, 9 ft. high, over cut-back tree-trunks, shrubs, etc., and furnished it with stems having different kinds of bark, stones, etc., in order to study the resting habits of the lies. These I turned into the net in some numbers — well over a hundred in the case of brevipalpis. Some pallidipes were also included in each experiment, and each was continued for from three to four days. None of the three species used confined themselves to the few feet next the ground ; they settled, colour conditions being correct, up to the full height available. I have noticed unconfined morsitans resting up to at least 6 ft. from the ground. " The rough-barked stems were selected in preference to the smooth, and large and small holes in the trunk and grooves in the bark were freely utilised for hiding in. A completely smooth-barked trunk was entirely neglected. Diplorhynchus mossambicensis was a favourite with the large tsetse, and a distinct colour-harmonisation took place also, the blacker tsetses (morsitans] choosing blacker bark, the greyer grey bark, the brown ones (brevipalpis) brown bark and the underside of rough or knobby lianas, on which they easily passed as one of the knobs. "For the morsitans experiment, in pyrophytic wooding, in which the trees always show, on one side especially, blackening by the grass fires, the Diplorhynchus trunks were specially selected on account of the strong contrast between the colours of their two sides. Shade con- ditions were about equal (as the result of overhead shade) and, the position of the branch being also suitable, the female flies in particular always tended on settling to select the blackened side. On this side, furthermore, they settled chiefly where the black of the raised cork ridges alternated with pale brown grooves, and this was at 3-4 ft. from the ground. Below this area the bark was uniformly black, above it rather more uniformly pale and uncharred. The females also settled, the colours being right, mainly on the lower side of a branch or log, the males settling as frequently on the upper sides. A raised stone and clods of earth were also used for resting under in the brevipalpis experiment. Leaves and thin twigs were used by active, not resting flies ; these flies were readily disturbed . . . The hiding flies were mostly, but by no means entirely, females. From my observations in the field also, it seemed clear that in Glossina, as in so many other animals, the female trusts mainly to concealment for defence against enemies, the male more largely to activity, and that the difference in the requirements of the sexes and their methods of meeting them is the chief reason for the female's special seclusion, though Lamborn's factor (avoidance of males) may be operative also." PROPORTION OF THE SEXES. As regards the apparent disproportion of the sexes in the case of G. palpalis, reference may be made to certain observations and experi- ments by Dr. J. W. Scott Macfie (93) at Zungeru in Northern Nigeria, during December 1911 and January 1912. " These months," writes Dr. Macfie, " cover the height of the dry season, when the Tsetse-flies 72 (G. palpalis) are restricted to a few shady patches of bush along the river banks, probably owing .... to their intolerance of the; dryness of the atmosphere and the burnt-up condition of the country. Of the flies brought to me, the proportion of males to females was as three to one, but whether this was due to the females being actually less numerous than the males, or to the fact that they were shyer feeders or more cunning at avoiding capture, it is impossible to say, The fact that in some of the temperature experiments the females seemed to be more sensitive than the males to a raising of the tempera- ture may have had something to do with it, as they may have been less eager to fly out into the sunshine to feed, and may thus have escaped capture. The converse, namely, that the female is apparently less affected by cold, may account for the fact that on one occasion only, on a singularly cool and sunless day, the proportion was reversed, namely, three females being taken for every male. The flies were also found to be susceptible to wind, and on the days when the harmattan wind was blowing strongly few or no flies were caught." Fiske (54), who quotes Carpenter to a like effect, has shown as a result of his work on islands in Lake Victoria, Uganda, that " When any considerable number of flies of this species of Glossina are caught, it is unusual to find the sexes evenly represented. Although they are produced in equal or approximately equal numbers, they are caught in unequal numbers." " This disparity between the sexes of Glossina palpalis," continues Fiske, " is an extremely variable quantity, ranging in different localities (in catches of 100 flies or more) from 1 -9 per cent, to 85-0 per cent, of females, or from 15-0 per cent, to 98 • 1 per cent, of males. Females never predominate to quite the extent of males, and most frequently the males are in excess." Such variations in the sex ratio of G. palpalis are, according to Fiske due to " purely local differences in conditions of life," and more espec- ially to food supply and its nature. Man is not naturally a favoured host of G. palpalis, consequently the destruction or expulsion of favoured hosts such as crocodiles and Varanus (monitors) causes the fly, impelled by hunger, to attack man to a greater extent than usual, the increase of attack, and therefore of liability to capture, being most marked in the case of the female. According to Fiske, " the most plausible explanation of this phenomenon " is that " the male flies are normally active and easily caught at all times during good weather, whether they are hungry or not, but that the females are normally inactive and not to be caught except when hungry and seeking food." This hypothesis, supported by an observation made in the course of a previous experiment, was directly tested " by banishing the host animals from a small islet without catching off any of the flies, and observing the effect on the sex ratio. :' This experiment," proceeds Fiske, " was made on the small island of Lugazi, in the following manner and with the following results. " 19th and 20th December 1913. A total of 197 flies were caught on the islet, the sex was determined and they were then liberated (in order that catching off of males should not affect the sex ratio) . The ratio was, males : females : : 166 : 31 = 15' 5 per cent, females. " 22nd to 27th December 1913. All host animals known to be fed upon by Glossina, consisting of several Varanus and crocodiles, were systematically hunted from the islet. " 26th and 27th December 1913. A total of 208 flies were caught, showing sex ratio, males : females : : 89 : 119 = 57-2 per cent, females. ' 73 " This experiment was carefully conducted, and careful notes were kept upon the behaviour of the flies on the islet towards man and also towards certain domestic animals which were tethered there .... They were so strikingly affected by the banishment of their reptilian hosts as to leave no doubt that they had been principally dependent on them, and that the increase in the percentage of females from 15-5 to 57-2 was the direct result of food shortage." After giving precise statistics as to the attacks of the flies upon the domestic animals (two goats and two pigs) and upon the human personnel employed in the foregoing test, Fiske writes :— " This experiment of hunting the wild hosts of tsetse was accidentally repeated on the peninsula of Neozi on the island of Bugalla (Sesse) where camp was pitched in November 1914. There was a not heavy or noticeable infestation by tsetse, and the principal host of it was situtunga. But no sooner was the camp occupied than these animals evacuated the peninsula and were seen crossing the isthmus connecting it with the mainland one or two miles distant only a few hours after the men began work on the tent and huts. On the following day (Sunday) the behaviour of the flies was not notably changed, but on the third day they became so unbearably persistent in their attack that (in view of the possibility of human infection from their bites) the camp was abandoned." A high female percentage in catches of G. palpalis, which, as has already been shown, indicates a dearth of favoured hosts, and con- sequent hunger and extreme pertinacity in attack on the part of the females, is, in fact, as is explained by Fiske, an important index of danger of transmission of human trypanosomiasis, and therefore a phenomenon of far more than mere academic importance. " Making allowances for error at every point," writes the author, "it is clear that density of infestation being equal, the fly is several hundred times more likely to feed upon man where wild hosts are very few and female percentage very high than when they are very many and female percentage low. " Perhaps the most pertinent point in this connection is that the chances favouring transmission of the virus of sleeping sickness from man to man are vastly less proportionately when few flies feed on man than when many do so. The same fly must feed on or bite the human host twice in order to transmit disease from an infected to a healthy man. If only one fly in 500 or 1,000 actually bites man, the chances that that same fly will attack man a second time are absurdly small ; if every second or third fly feeds upon or bites man the chances that that same fly will attack man a second time are stupendous in com- parison. ' The female percentage may thus be a very valuable index to the chances favouring transmission of human disease." In order that sex percentage data obtained in different localities and on different dates shall be inter-comparable, it is important that the flies shall be caught in a definite manner. On this subject Fiske remarks : — " The most convenient method for measuring variations in the density of Glossina palpalis is that of employing expert ' fly boys ' and of counting the number of flies which can be caught per boy per hour under standardised conditions. This method had already been used by Dr. G. D. H. Carpenter and others, and when care was taken to eliminate sources of error very reliable figures were secured." 74 In a foot-note to the foregoing passage the author adds a series of directions, which are here reproduced for the guidance of future investigators. " Proper attention to the following suggestions will serve to eliminate various possibilities of error : — " (1) Boys should be trained at least one month ; all new boys in a corps without one or two experts as teachers would require longer training. " (2) Nets must have a standard-sized ring and a standard length of handle ; an 8 in. ring and an 18 in. handle were used. " (3) Catches should not be made before 8.30 or 9 a.m., nor later than 2.30 or 3 p.m. " (4) Boys will make more even catches if provided each with a dark-coloured umbrella, upon which the flies will settle, and from which they are easily caught. " (5) Unless density is very low — less than 1*0 — it is better not to spend more than 2 to 3 ' boy-hours ' at the same point on the same day ; density is easily reduced, temporarily, by catching. " (6) Boys should be stationed along the routes most likely to be followed by moving flies, and always, if there is shadow, at the "edge of it. " (7) Estimates of density must be based on catch of males, on account of the variability of female activity. " (8) Good results cannot be secured on cool, cloudy or windy days." With reference to the proportion of the sexes in G. morsitans taken by him in Nyasaland, Dr. Lamborn (83) says : — " My experience in this matter is the same as that of other observers, namely, that when the flies are bred out from pupae, the sexes emerge in almost equal proportions ; and when the flies are captured, the males far outnumber the females .... There are very definite reasons why the female flies should to some extent shun the society of the males. As I have before remarked, and have since repeatedly observed, coitus takes place as the result of capture without preliminary courtship . . . In captivity even females in an advanced state of pregnancy are not secure from the violence of the males, and as abortion is so frequent with captive females, it seems possible that this may conduce to it . . . . These facts undoubtedly indicate the necessity for the pregnant female to seclude herself, when once fertilised, from the further atten- tions of the males ; hence the unequal proportion of the sexes among captured flies." Further statements on this subject, in connection with G. morsitans in Nyasaland, are contained in a subsequent paper by the same author (84). In studying the proportion of sexes among captured flies, it was observed that male flies become attached to possible hosts in order to secure females which come to feed. When the possible hosts rest, the females are less attracted and most of the males fry off. Pairing is attempted by the male whatever may be the stage of development of the female. Collectors can readily obtain flies by catching them off each other, and these captures always show a small proportion of females. If flies which have settled near the possible host are included, the results show a much higher percentage of females. The number of active G. morsitans decreases during the hottest part of the day. Examination of possible sheltering places during this period showed that large numbers of females are to be found in recesses in the trunk and bark of large trees, especially baobabs, while the males occur in more exposed positions. The proportion of females taken in the 75 ordinary way in morning and evening was 12 and 10 per cent., respec- tively ; when taken at midday, mostly off trees, the proportion was 43-5 per cent. The following explanation is suggested for the varia- tions observed in the proportion of males and females captured at different times of the day : — Males which have recently fed attend man to await the females which come to feed. The female may be secured by a male, or, if pregnant, may be driven off, and attempt to feed later. After feeding, the female retires under cover of large trees or to breeding places to protect itself from the males. It is almost certain that the sexes exist in the same area in equal numbers. Writing with reference to the Gold Coast, Dr. J. J. Simpson (144) agrees that a marked disparity exists in the proportion of the sexes of Glossina, and states that, in explanation of the fact, there have been advanced various hypotheses, such as the size of the river on which the flies are caught, the time of capture, the number of inhabitants in a given area, the season of the year, and meteorological conditions. To these factors, in the opinion of this author, must be added the abundance of game, and consequently food, which, by causing the retirement of the gorged females, gives rise to an apparent disparity. Swynnerton's experience in Portuguese East Africa, in connection with G. brevipalpis, G. morsitans and G. pallidipes, which is in harmony with that of the authorities already referred to, strongly supports the view that the numerical predominance of the males seen in captured specimens is apparent rather than real, and is due to marked differences in the habits of the two sexes. " G. brevipalpis," writes Swynnerton (145), " may be found in the daytime scattered through all the little thickets in the bush it frequents, two or three or more to each. Just at sunset the males emerge from the thickets in the neighbourhood of game and other paths and (where plentiful) distribute themselves at short intervals along them, sometimes for a considerable distance. As one walks each male moves on in front of one for at most a few feet, evidently scanning the walker for any females that may be with him, then falls behind. At one time, when a domestic matter called me home from the Buzi, my police boy, left in charge of the work, captured no less than 88 of these flies, every one a male, on three successive evenings without the aid of cattle along the same stretch of path." " The males of G. morsitans, though occasionally seen like this on game paths, were found mostly crowded together in the short grass beside the path — some of them on the path — and were also (as events each time fully proved) watching for females or carriers to take them to the females. G. morsitans usually waited in a crowd, brevipalpis always in a queue." On a subsequent page, when dealing specifically with the subject of the present section, Swynnerton writes :— " The following statements represent the result of day-to-day observation and analysis in the field — the only reliable method. " (1) Wherever we had to deal with male crowds or queues awaiting females, or, in the case of brevipalpis, the bush immediately bordering on the path on which the male flies would line up at sunset, we took, as might be expected, practically nothing but males. " (2) Elsewhere, using cattle bait, we found usually either an approxi- mate equality of the sexes or a preponderance of females. (3) Many more females proportionally came to cattle than to man and goats. ' The cattle were with me only during my last four days in the morsitans area .... but the proportion of females rose abruptly 76 on their arrival. It amounted now to from 35 to 40 per cent., in spite- of the fact that the cattle were used very largely at a strong male cluster in order to obtain the material for certain experiments. Previously 7 per cent, had constituted a good day . . . ." " Bagshawe's result (more male palpalis where crocodiles were, present, more females where not) and McConnell's (more palpalis females on the small tributaries — where food is doubtless scarcer — more males on the main Nile) appear to be completely explained by observation No. 3 above. Bequaert's observation (more males in uninhabited regions, where non-human food would be more abundant) would appear to be the same, had not Graham (quoted by Bagshawe) noted that, in Ashanti, palpalis prefers human blood to that of cattle. But both Graham and Zupitza (also quoted by Bagshawe) observed that when flies bite readily males are not in excess. " In spite of a heavy preponderance of males in my total takings — the inevitable result of following paths and of special attention to all male crowds or queues .... I came across no evidence at all which, properly analysed on the spot, suggested an actual preponder- ance of the male sex. Rather the reverse ; and it was likely, both in view of the systematic attack by birds which I found taking place on a conspicuous male crowd and from a consideration of the breeding habits (which are such as to allow for much male wastage), that, at a given moment and apart even from the freer dispersal of the males,, the females are more numerous than the latter, though they will show themselves in their true proportions only when hungry." Lastly, after stating that G. brevipalpis is " very loth to feed on man," and that " the female sex comes primarily to feed," Swynnerton adds :— " My results would seem .... to show clearly that the apparent scarcity in the presence of game of the females of the three flies chiefly investigated is probably due to a relative dislike of man." In North Katanga, Belgian Congo, it was found by Schwetz (132) that, " contrary to what happens in the case of G. brevipalpis, and even G. pallidipes, among captured specimens of G.jusca, whether taken on the wing or resting upon trees and lianas, females occur in large numbers, and constitute nearly one-half of the total." To conclude this short statement, we append a few of the observa- tions made by various entomologists with reference to the relative proportion of the sexes in G. palpalis, G. morsitans and G. brevipalpis. In the report on his investigations in Uganda upon the biology of G. palpalis vax.fnscipes, Dr. G. D. H. Carpenter (25) notes the propor- tion of the sexes of this Tsetse-fly in three localities. At Jinja, on the shore of Lake Victoria, 12,773 flies yielded 55 • 7 per cent, of females ; in Damba Island, Lake Victoria, among 6,000 flies, females formed 21 -6 per cent., and in Bugalla Island 20 • 6 per cent, among 56,775 flies. Since in breeding experiments he obtained the two sexes in approximately equal numbers, Dr. Carpenter thinks that it is possible that in nature female pupae die in consequence of variations in temperature, humidity, etc., which do not occur under laboratory conditions. On the other hand, Mr. LI. Lloyd, Entomologist to the Luangwa Sleeping Sickness Commission, in describing his observations on G. morsitans in the Luangwa Valley, Northern Rhodesia, writes (88) :-— " Much difficulty was experienced at first in obtaining sufficient females for breeding purposes. Frequently only three or four females were brought in amongst two or three hundred males. Later, the . boys were shown the differences between the males and females, and 77 there was then no difficulty in obtaining a sufficiency of the latter. The relatively small proportion of females that is normally taken is a matter of great interest. In the laboratory the sexes are bred in equal numbers, so that it is fair to assume that they are really equal in nature. Several writers have drawn attention to the fact that the swarms of Tsetse which accompany a moving object leave it as soon as motion ceases .... The majority of the swarm, however, are not desirous of feeding .... If the flies which actually bite are caught, they are found to be females as frequently as males, but the small proportion of females that is caught normally shows that the swarm consists almost entirely of males. An analogous case may be quoted. The vast swarms of midges so often seen on fine evenings are known to consist entirely of males, and directly a female joins the swarm it is mated and the couple drop from the dance. With Glossina, as long as an animal is moving there is a probability of females rising to feed and of the accompanying males thus finding mates. When the animal stands there is little chance of females rising to it, and this may be the reason why the males leave." Writing in 1915, with reference to observations made by him in Nyasaland towards the close of the previous year, Dr. W. A. Lamborn (82) states that the reason why "Tsetses are especially attracted by moving objects," and " that several flies will often travel on a person for considerable distances without attempting to bite, is not yet clear, but an examination of 43 individuals taken off the ground on various occasions as they followed showed that all were males, and it seems a reasonable supposition that such flies may be lurking in anticipation of the advent of a female. No courtship takes place, several males often making a simultaneous rush at the female on sight, and one seizing her, pairing takes place in the air, the couple then settling. It is by no means uncommon to take off natives in one's company a couple which have been seen to pair, and several times paired couples have been taken by means of bird-lime smeared on material borne on a native's back." It may be added that, out of a total of 2,460 G. morsitans caught by Dr. Lamborn's fly boys in the course of his investigations, 2,040 were males and 420 females, thus confirming the disproportion of the sexes (predominance of males) met with in many fly-belts. As regards G. brevipalpis in Nyasaland, Dr. J. B. Davey notes that out of 117 specimens " captured at Kaporo in the evenings in August (dry season), all were males. Seven were captured at the same place and time of year about midday, and of these four were males and three were females." Sir David and Lady Bruce, and their collaborators, writing on G. brevipalpis as vector of trypanosomiases in Nyasaland, mention that among 500 flies caught at the mouth of the River Lingadzi, on the western shore of Lake Nyasa, there was not a single female,, and that among several thousand specimens brought to their laboratory, there were only four females. It is suggested that the females remain hidden in dense bush and do not fly outside it as do the males. Lastly, Austen (3. — 1911) writes with reference to the same species that, out of 359 specimens examined by him, " only 42 were females, and the enormous preponderance of males over females usually seen among captured specimens has been noted both by Stuhlmann and Davey. It is pointed out, however, by the former author that, as in the case of other flies, it is shown by breeding G. brevipalpis in captivity that the two sexes are produced in approximately equal numbers. 78 In order to catch females, Stuhlmann states that it is necessary to use animals such as cattle and mules as decoys, but even so he observes that they are much more difficult to capture than males. He thinks it probable that ' the gravid females are much more wary and move about less than the males, and are consequently more rarely caught.' ' CHAPTER VIII. TSETSE-FLIES AND THEIR FOOD. The study of Tsetse-flies in relation to feeding includes various matters, which it is as well to examine separately. All the observations, investigations and experiments made upon this subject are of great practical interest, since they throw light upon the transmission of trypanosomiases by Glossina. TIMES WHEN TSETSE-FLIES ARE MOST ACTIVE. — THE SEARCH FOR FOOD. — It is well known that Tsetse-flies are much more active and pertinacious when they attack moving objects than when the latter are stationary, and it therefore seems logical to conclude that, in singling out an involuntary host, they trust more to the sense of sight than to any other. In the majority of cases, if one remains absolutely motionless in a haunt of G. palpalis not a single insect is seen, but immediately one moves the flies commence to attack, and the same fact is observed in the case of horses and cattle. Similarly, when a river steamer approaches the banks in a fly-area in Belgian Congo, swarms of Tsetse fly on board, but if the boat stops the insects cease their attacks, resuming them so soon as it begins to move again.1 The fact should be borne in mind when clearing places on shore for the purpose of landing-stages, ferries, etc., where movements of some importance in river traffic take place. Dr. W. A. Lamborn (82), writing in 1915 of G. morsitans in Nyasaland, says : — " Tsetses are especially attracted by moving objects, and it has been my frequent experience that the faster one moves, the more one is beset, for which reason I now invariably walk rather than cycle in a fly area. If one stands still, such flies as may have been hovering around come to rest on objects near by, frequently on a blade of grass, or they settle on the ground, resuming their attentions immediately one is in motion again ; though, if one remains still long enough, as I have often found during a rest for lunch in the open, the majority gradually fly away." Dr. Lamborn adds that, as has already been mentioned, " several flies will often travel on a person for considerable distances without attempting to bite." As is well known, Glossina morsitans feeds with avidity on the blood of almost all large animals with which it meets ; according to certain observers it is more aggressive during the hottest hours of the day, while others state that it attacks human beings not only throughout the day, but also during part of the night, especially in warm weather and when the moon is bright. 1 In the course of his journey on the Congo from Leopold ville to Stanleyville, Dr. J. Bequaert made some very interesting experiments in connection with the pursuit of fast river steamers by Tsetse-flies, by catching flies which flew on board, marking the insects by snipping off the tarsus of a particular leg (varying according to the day of capture), and then releasing them. In this way it was found that a fly had been carried for a distance of some 87 miles, which was doubled on the downward journey. The steamers may therefore be responsible ^ for the dissemination of G. palpalis and trypanosomiasis. 79 The investigations upon G. morsitans made in Rhodesia by Messrs. R. W. Jack and LI. Lloyd show that this Tsetse-fly is most active during mild weather ; mists, strong winds and cold weather, as also, too, intense heat, do not suit it at all. As a general rule, except when there is a mist, G. morsitans is ready to feed immediately after sunrise, retiring to cover during the hottest hours of the day and recommencing to bite in the afternoon, when it is cooler. The fly may sometimes continue to bite after sunset when the nights are warm ; during showers of rain it usually disappears.1 As regards G. longipalpis, Dr. Roubaud (H9a) says that in Dahomey this species, like G. palpalis and G. tachinoides, lives chiefly at the expense of wild animals, and accompanies big game in its movements ; paths recently trodden by hippopotami and elephants are favourite haunts of this Tsetse-fly. With regard to times when greatest activity is displayed by G. pallidipes, observations by various observers (Leese, Croveri, Schwetz) in different parts of Africa have already been recorded (see p. 53). Writing of this species as observed by him in the coastal regions of Kenya Colony, the late Captain R. B. Woosnam remarks (165) :— " From personal experience I can testify to the fact that G. pallidipes feeds greedily during the whole night, but, in my experience, not during the daytime." This statement is not borne out by the other writers referred to, though Leese says that the fly " sometimes bites in the dark." "G. brevipalpis," writes Austen, " is active and desirous of feeding only at two periods of the day — early in the morning before 8 o'clock, and in the evening from about 4 p.m. onwards. The intervening hours, from 8 a.m. to 4 p.m., are passed by G. brevipalpis in concealment, 'under leaves of bushes or in the grass, always near the ground' (Sanderson), or low down on the trunks of trees, two or three feet from the base. While thus resting motionless the flies are difficult to discover and capture, and ' their presence would be entirely unsuspected ' (Sander- son) ; at Kaporo, near the north end of Lake Nyasa, Davey found that they preferred to rest on ' trees surrounded by creepers and undergrowth, and hid away in crevices in the bark or under the origin of branches.' Both Davey and Sanderson observed a pair in coitu on a tree-trunk, in the one case at midday, in the other at about 5 p.m. ; attempts to find the pupae of this fly by digging round the bases of various trees were, however, unsuccessful. Davey states that about 4 p.m. G. brevipalpis ' seems to emerge from its hiding place and settle on dried leaves, sticks or dust on paths, apparently lying in wait for a meal ' ; the same writer adds that he has noticed that ' game, on its way to water in the evening, often stands for a time on emerging from the forest on to a path, and would thus afford the fly an opportunity for feeding.' According to Davey, G. brevipalpis as a rule ' seems ready to bite human beings in the evening, but it does not set to work with the rapidity and voracity that G. morsitans often exhibits.' When feeding captive flies, Stuhlmann finds that the insects generally wait for a while before deciding to suck ; they then, however, drive in the proboscis with great quickness, and the abdomen usually becomes 1 On several occasions Dr. Bequaert was bitten by G. morsitans at night, when it was quite dark ; on the other hand he states that, in Katanga, the general experience is that rain, even when fairly heavy, never prevents this insect from coming out and gorging itself with blood. 80 swollen out like a balloon with blood in from thirty to forty seconds. In •captivity flies of this species show very little inclination to feed in very hot and sultry weather, and this is also the case when the temperature is abnormally low. Although continuing to bite and fly about ' for some time after dark ' (Sanderson), G. brevipalpis appears not. to bite during the actual night hours, and specimens kept in captivity in glass cylinders were found in complete darkness to be perfectly still (Stuhlmann). Like G. palpalis and blood-sucking flies in general, the present species displays a predilection for dark-coloured animals and surfaces. Stuhlmann states that he has frequently noticed that, when a light- and a dark-coloured mule were moving side by side, only the latter was attacked, and Sanderson writes : — ' The flies bite through dark clothes, but have never been seen to settle on white surfaces.'1 "As to the relation of G. brevipalpis to game, Davey observes that wherever he met with this fly in Nyasaland ' game (antelope, warthog, etc.) was abundant.' ' Writing of his experience of G. brevipalpis, G. morsitans and G. pallidipes in Portuguese East Africa, Swynnerton (145) remarks :— " G. brevipalpis, as I found, attacks in the shade of the bush at any hour, but does not fly appreciably outside its own thicket to do so except in the early morning, after sunset and on dull or rainy days. Having attacked, even in sunny weather, it completes its object of feeding or riding, unless its victim emerges into the open. In this event it usually leaves. In a special experiment that I carried out with the aid of a lantern occasionally lit and turned on I found that both this fly and pallidipes attacked right up to dark. Then every fly suddenly dis- appeared. On moonlight nights they continue their attacks, and all- three flies come to light, including firelight, and, once there, bite freely. G. morsitans, I found, as have other observers, to be most active in the rather warmer hours and least on cold or dewy mornings, but it too attacked up to dark. It is probably dependent on shade to retire to occasionally when blood and water are not available, for I captured several male flies that had been resting under logs in the wooding adjoining vlei edges frequented by male clusters. The frayed wings of some showed that they had not just emerged." EXPERIMENTS ON THE TAKING OF FOOD BY TSETSE-FLIES. — Investi- gations and experiments upon the feeding of G. morsitans and G. palpalis in captivity have been made respectively by Dr. Allan Kinghorn and Mr. LI. Lloyd in Northern Rhodesia (1911-13), and by Dr. J. W. Scott Macfie, West African Medical Service, at Zungeru, Northern Nigeria (1911-12). While, as has just been stated, the subject of Dr. Macfie's experiments for the purpose of studying the phenomena of feeding and digestion was G. palpalis, the data obtained will apply, in their broad outlines, to all the species of the genus. We give below excerpts from Dr. Macfie's paper (93). FEEDING AND DIGESTION. — " In considering the processes of feeding and digestion of G. palpalis," writes Dr. Macfie, " it was found to be 1 Swynnerton (145) says : — " I was myself bitten by morsitans far less than my natives, but this was presumably a matter of skin-colour and clothes ; Mabuzana, living near the Mtshanedzi south of Gogoyo's, volunteered the information that tsetses were specially attracted by a black coat. I said ' How do you know that ? ' ' Because I have one ! '" 81 essential that the flies should be naturally fed." After a number of unsatisfactory attempts with other sources of blood, the experimenter in every case permitted the flies to feed on his own fingers. Feeding. — On this subject the author writes : — " The attitude during feeding is of course well known, the fly standing firmly on all six feet and thrusting its honey-brown proboscis vertically downwards into the skin, whilst the ensheathing palpi project forwards and upwards. As pointed out by Dr. Moiser, the fly sometimes partly withdraws its proboscis and thrusts it in again and again, but once a satisfactory well of blood has been tapped, the abdomen fills rapidly, and in from one to two minutes the act of feeding terminates and the fly, withdrawing its proboscis, moves to one side and presently flies heavily away. ig. 6. — A Tsetse-fly (Glossina morsitans, Westw., O), before feeding. " In my 'experiments I have usually found the bite painless. Some- i! times, however, it was decidedly painful, and I can only account for | this by supposing that on these occasions the proboscis entered I sufficiently near to one of the sensory spots of the skin to stimulate it. On the other hand there was generally some degree of smarting at the site of puncture after the feed was over, and it is probably this which usually attracts the attention of the victim to the presence of the fly, which having already fed is able to escape on the first indications of uneasiness. Fig. 7. — A Tsetse-fly (Glossina morsitans, Westw., $), after feeding, showing abdomen distended with blood. ( x 6.) From a drawing from life by Lady Brace, R.R.C. (Both the above figures reproduced, by permission of the Trustees of the British Museum, from " A Handbook of the Tsetse-Flies [Genus Glossina]," by E. E. Austen. — London : British Museum (Natural History) . 1911.) " If undisturbed, the fly feeds until the abdomen is relatively greatly distended and of a bright red colour, due to the blood showing through the tense and semi-transparent tissues. The dorsal plates of the abdomen are widely separated, so that the body loses its characteristic dark colour. The appearance is therefore somewhat different from (5979) F 82 that given in the figure of 'A Tsetse-fly (Glossina morsitans), after feeding,' in Austen's ' Handbook of the Tsetse-Flies ' on page 93. TSee p. 81]. To determine the amount of blood actually drawn up, the following experiments were carried out. Two or three tsetse-flies (G. palpalis), that from the almost linear dimensions of their abdomens had evidently not been fed for some time, were stupefied with chloro- form and in this condition were weighed. They were then allowed to recover consciousness, and when quite restored were fed undisturbed on my fingers. .... They were once again stupefied with chloroform and reweighed. The difference between the two weighings represented the weight of the blood taken up, for sufficient time was not allowed to elapse after the feeding for any exudations to be lost, and in the condition of stupefaction produced by chloroform the process of digestion appeared to be arrested As the result of a number of such experiments the average weight of blood ingested by the male was found to be -027 gm., which, reckoning the specific gravity of normal blood as 1-06, corresponds to 2-5 cmm The weight of the flies varies considerably with the length of time that has elapsed since the last meal, but nevertheless there is an appreciable difference between the weights of the unfed males and females. As a rough average the unfed male weighs -020 gm., and the female, which is somewhat larger, -028 gm., The same applies to the fed flies — they do not always engorge themselves to the same extent, but the average weight of blood ingested is somewhat greater in the case of the female. On one occasion as much as -0485 gm. (or 4-5 cmm.) of blood being drawn up by a female, and -030 gm. by a male. The proportion is perhaps better stated in comparison with the body weight — a male G. palpalis is capable of sucking up blood weighing 1 • 3 times his body weight, and a female 1 -6 times her body weight. " With regard to the frequency with which the flies feed it was generally found that once fed they refused to bite again until about 30 hours had elapsed. But here again the intensity of the last meali has to be taken into account, for on one occasion a fly which had fed very fully refused to feed again until 70 hours later. As the flies were all confined in glass jars containing no water this shows that when fully and naturally fed they are able to live unfed much longer than the 30 hours that is sometimes given as the limit of their endurance under such conditions. " Digestion. — Having fed, the Tsetse-fly settles quietly to digest. At first the under surface of the distended abdomen is uniformly red' in the case of the male, but the female always has a pale opaque spot at the distal end. In from two to five minutes, however, a pale bubble appears at the proximal pole, situated generally just to the left of the mid line. This bubble shows slight movements of contraction and expansion, as though it were influenced by some peristaltic-like action of the bowel, and moves gradually over towards the left side, keeping all the time at the proximal extremity of the abdomen. On dissecting a fly at this stage the bubble was found to be gaseous, and from the fact that the red tint of the blood lingers along its edge long after it has faded from the rest of the abdomen, we may surmise that it contains oxygen as a constituent at any rate. Two other processes are mean- while observed to be going on, namely, the exudation of drops of fluid from the anus and a darkening of the ingested blood. " It is not until two or three minutes after the fly has settled down to digest its meal that drops of fluid begin to be extruded from the anus. The first drop is often, though not invariably, of a buff colour and 83 turbid, but the succeeding drops are of clear water-like fluid and are of the size of a pin's head. For about twenty minutes they are exuded at intervals varying from 15 to 75 seconds, the intervals being somewhat longer towards the end of this period. Thereafter no more drops are ejected for several hours, and those that follow are of a thick chocolate brown excrement .... Coincident with the extrusion of these drops, which are presumably the fluid constituents of the blood, there is a marked shrinking in the size of the abdomen. " Soon after digestion has begun the appearance of the blood in the distended abdomen of the fly begins to change. Commencing at the distal end and spreading upwards the abdomen loses its bright red colour and gradually darkens, so that in from twenty minutes to half an hour the last tint of red has gone. The last part to darken is a narrow zone around the gaseous bubble, and as has been suggested above, this may be due to the presence of free oxygen in it. " Slight differences were observed in the case of the female G. palpalis. When fully distended with blood her abdomen appeared to 1 be deeper, or less dilated laterally, than in the case of the male, and i the process of darkening of the ingested blood was considerably more rapid. In one case the last trace of red colour had disappeared from the abdomen in ten minutes." As regards the feeding of Glossina morsitans, Dr. Allan Kinghorn, of the Luangwa Sleeping Sickness Commission, Northern Rhodesia, | wrote (79) in August 1911 : — " The method observed in preserving and feeding the captive flies was essentially that described by Kleine. " Freshly caught flies, two females and one male, were placed in i wide-mouthed glass tubes, measuring 5 by 2 in., and the mouth was I then closed over with mosquito-netting held on by an elastic band. i The flies were fed daily on native fowls during the warmer hours of i the day, when they readily gorged themselves. Roubaud has recorded, i in the case of Glossina palpalis, that the females do not feed as often as the males, but this has not been found to be so with G. morsitans. Both sexes have fed with equal avidity. After the completed meal j the insects were changed into clean, unlined tubes, and under these i conditions it has been possible to keep some of them alive for over two months, at the time of writing, and to obtain larvae. " When first introduced into the tubes the flies are very active, making wild efforts to escape, but in a very short time they become habituated to their new surroundings and remain quiescent. The males have been found to die much sooner than the females, and this may possibly indicate that the natural duration of their life is shorter." Mr. LI. Lloyd, Entomologist to the Luangwa Sleeping Sickness ; Commission, wrote (88) in the following year : — " Owing to the difn- ! culty of obtaining suitable mammals, native fowls were used almost i entirely as blood donors for the captive flies. The small numbers of • pupae obtained would seem to indicate that fowls' blood is not a very ; suitable food. This is perhaps because the blood forms large firm clots in the sucking stomachs of many of the flies. When the blood has attained this condition it cannot apparently be utilised by the flies, as the clots persist for some weeks after the meal on fowls, though monkeys may have been used as food in the interim. When such a clot is formed it would be impossible for the fly to retain a full-grown larva in the uterus. The phenomenon has not been observed with mammals' blood, nor has it been observed in nature. " No evidence has been obtained that the flies take any food other than blood. Out of 67 freshly caught flies that were dissected, 84 mammalian blood was recognised in 15, nucleated red cells in four; one fly contained both nucleated and non-nucleated red cells ; in 32, blood in a more or less digested condition was seen ; while 15 had either not fed or the detritus was not recognisable. " It seems possible that a little water may be taken, as the flies have been observed at the edges of puddles and apparently drinking. Kinghorn has observed the same thing with G. palpalis on the shores of Lake Tanganyika. If a moistened sponge or blotting paper is placed in a bottle with captive flies, they will settle on it and thrust in their proboscides. I have not been able to prove by supplying coloured fluids and subsequently dissecting the flies that fluid is actually absorbed. Exactly the same thing was observed with slices of ripe water-melon, but again it could not be demonstrated that food was taken up. Flies placed in bottles with Sphingid larvae died of starvation, and were never seen to take anything but the most perfunctory interest in the caterpillars. " A gecko which was placed in a cage of flies caught and devoured the insects. Experiments were also made with a monitor, which was placed in the double bottom of a cage made of gauze in such a manner that it was incapable of movement. The flies in the cage attempted to feed, but only one succeeded in doing so, thrusting in its proboscis in the region of the head. The animal was then killed and a cut section i was offered to the flies, when several readily gorged themselves/' In the Gold Coast, it was found by Simpson (144) that the proportion of mammalian to non-mammalian blood imbibed varies widely according to the species of Tsetse-fly. G. morsitans seemed to be much more dependent on mammalian blood than did G. palpalis, while the chief food of G. tachinoides appeared to be reptilian blood, and that ; of a large bat which abounds on the banks of the River Volta. The small percentage of reptilian blood found in G. morsitans form submorsitans may be due to lack of opportunity and not to preference, since in the open country where this form abounds a reptilian fauna : is practically non-existent. The work of the Luangwa Sleeping Sickness Commission terminated in 1912, but in the following year the investigation into the bionomics of Glossina morsitans was continued by Mr LI. Lloyd at Ngoa, " in the Mpika Division, on the high ground of the Congo-Zambezi watershed." Some details — extracted from a paper (89) published by Mr. Lloyd in 1914 — of an important series of experiments carried out in the course of this later work are given below. " INFLUENCE OF VARIOUS BLOODS ON BREEDING CAPABILITIES. — In the method which was adopted a definite number of freshly caught; female flies were fed on goats, monkeys, fowls, ducks and chameleons. In most of the experiments the flies were kept in the usual type of fly bottle, four females and two males in each. A few of the series were kept in large wooden cages with fronts of mosquito muslin. Records were taken of the length of life of each fly in captivity and the numbers of pupae and aborted larvae which were deposited ; each pupa was • also measured. Nineteen experiments were commenced, of which 13 were completed." Details of the results of these experiments are shown in tabular form, and in the "Summary and Conclusions" at the end of the paper: the author writes :— " Reptilian blood is not suitable to G. morsitans as a continued diet. Mammalian blood has a slight advantage over: avian as a diet, and this is shown by the larger average size of the pupae. 85 produced in the laboratory." Elsewhere, Mr. Lloyd says : — " On the average the pupae produced on mammalian blood are larger than those bred on avian blood. This advantage is 4 per cent, in length and 5 per cent, in breadth .... " In the avian series there was a large proportion of very small pupae. Of these 22 (36 per cent.) measured 5 • 5 mm., or less, in length, while in the mammalian series 11 (13 per cent.) only were of these dimensions. The avian series, however, also included a few pupae which were equal in size to the largest bred on mammals' blood. " The small size of these pupae is considered to be due to the pressure of the clots of blood which form in the sucking stomachs of some of the flies when fed on avian blood." A tendency to overfeeding on the part of the captive flies was observed in the course of these experiments. " On three occasions," writes Mr. Lloyd, " they have been known to feed till an internal rupture of the gut was caused." Passing on to consider the relation of G. morsitans to the smaller animals, the author remarks with reference to the foregoing series of investigations : — " These experiments point to the conclusion that the advantages of a mammalian diet as opposed to an avian one are not great and are due to the fact that mammalian blood is more easily digested than avian While the fly is apparently unable to carry on its species on a reptilian diet, occasional meals of such blood would assist the individual in prolonging life." As to this, Dr. G. A. K. Marshall writes in a footnote : — " Mr. F. W. Fiske has already pointed out (Bull. Ent. Res., iv., p. 103) that the results of laboratory experiments upon the food value of reptilian blood should be accepted with some reserve, as they are not borne out by field observations. Dr. H. L. Duke and Dr. G. D. H. Carpenter have noted that G. palpalis feeds freely on monitor lizards, under natural conditions on the islands in Lake Victoria, and as a result of his own recent observations in the same locality Mr. Fiske concludes that these lizards constitute the most favoured food of that fly." STUDY OF THE NATURAL FOOD OF TSETSE-FLIES, BY EXAMINATION OF THE CONTENTS OF THE ALIMENTARY CANAL. — A better method of determining the normal food of Tsetse-flies consists in the examination of the intestinal contents of flies caught under various conditions of environment. We give below a resume of some of the investigations that have been made in this direction upon G. palpalis and G. morsitans. The Proceedings of the Royal Society of London, Series B., Vol. LXXXIL, 1910, contain a series of important reports on the work of the Sleeping Sickness Commission of the Royal Society, in Uganda, 1908-10, by Colonel (now Major-General) Sir David Bruce, C.B., F.R.S., A. M.S., Captain (now Lieut. -Colonel) A. E. Hamerton, D.S.O., R.A.M.C., Captain H. R. Bateman, R.A.M.C., and Captain F. P. Mackie, I. M.S. Writing on " The Natural Food of Glossina palpalis," the authors state : — " In the laboratory it was found that the flies fed with far more avidity on birds than on monkeys, while they could hardly be tempted to feed on young crocodiles, iguanas [monitors], or lizards." The microscopical examination of the intestinal contents of 403 captured flies, however, yielded the following results. Of 220 G. palpalis, " caught at various parts of the Lake-shore," and dissected for the most part " during October, November and December, 1908," generally about 24 hours after capture, 27-3 per cent, contained blood in the intestine. As regards character of the 86 blood, this percentage was made up as follows : — mammalian, 9-1 per cent. ; non-mammalian, 4 • 1 percent. ; non-recognisable, 14-1 percent. In a second experiment, 183 G. palpalis were caught at a spot on the Lake-shore where the insects " swarmed," and both crocodiles and birds were abundant. Immediate " naked-eye " examination showed that " blood in a more or less digested state " was present in 59 per cent, of the captured flies. On subsequent microscopical examination, it was found that " out of the original 183 flies examined, 52 (28 per cent.) contained blood which was recent enough to show the red corpuscles. Out of these 52, 7 (13 per cent.) were from mammalian blood, whilst non-mammalian blood was present in 35 flies (67 per cent.)." In the latter " reptilian blood was twice as frequent as the blood of birds." In the Proceedings of the Royal Society of 6th August 1914 we find the results of other investigations, made by Sir David Bruce, Major A. E. Hamerton, Captain D. P. Watson and Lady Bruce, on the food of Glossina morsitans in Nyasaland. The most important paragraphs in this short paper (20) are given below. " Five hundred flies, freshly caught in the Proclaimed Area, were killed by chloroform and the gut of each was roughly dissected out, smeared on a slide, fixed by osmic vapour and alcohol, and subsequently stained by Giemsa. The flies were all caught in the bush, away from the paths, the fly-boys proceeding in single file and catching the flies with gauze nets as they circled round, or settled on the boys or the grass. " The proportion of male flies to females caught was roughly two to one. But only 30 females were used in the present experiment, the majority being sent to the breeding-station at Chunzi. " Of the 500 flies examined, 288, or 57-6 per cent., were found to contain mammalian blood in a recognisable state. No measurements were made of corpuscles, which in most cases were much altered by the digestive processes, but the small type of cell appeared to predominate, such as occurs in the hartebeeste, waterbuck and other antelope. " In only three cases were nucleated red corpuscles found, and in two of these there was only a small proportion of nucleated blood mixed with a large amount of mammalian. In the third case the blood was all nucleated. Thus, of those flies which contained recognisable blood, only 1 '0 per cent, contained nucleated blood. From measure- ments, it seems highly probable that in all three cases the blood was avian, not reptilian " In no case was vegetable matter noted in the intestinal contents. " From experiments with flies in the laboratory, it was found that blood is recognisable in stained specimens for two to three days after a feed, but not beyond the third day. Hence it may be inferred that, roughly, half the flies examined had fed within, at most, three days of their capture, and that therefore the flies feed naturally at least once every six days." Another series of investigations upon the bionomics of Glossina morsitans was made in Northern Rhodesia, from May to July 1914, by the late Mr. R. A. F. Eminson. The following particulars, with reference to "an examination of the gut contents of captured flies in various conditions of environment," are extracted from a summary of a report by Mr. Eminson to the British South Africa Company (50). " During May 1914, 300 flies were examined .... and recognisable blood was found in 43 ; of these, 41 contained mammalian and 2 non-mammalian blood, so that the latter amounted to 5 per 87 cent. It is stated that a similar investigation during the wet season months of February and March yielded only 1 per cent, of flies con- taining non-mammalian blood, but the actual numbers are not given. As a possible explanation of this difference, it is suggested that the flies are probably much more voracious during May than in the wet season, as it is believed that the greatest number of larvae are deposited in May. It is clear, however, that a seasonal change in the available food supply is probably also an important factor." Finally we may note a few of the data furnished by Dr. G. D. Hale Carpenter, in his " Second Report on Bionomics of Glossina fuscipes (palpalis) of Uganda" (25). With reference to the percentages of flies in which mammalian (M.) or non-mammalian (N.) blood was found at three different localities, Dr. Carpenter gives the following figures :— Jinja (shore of Lake Victoria), M. 31-5 per cent., N. 68-5 per cent. ; Damba Isle, M. 15-2 per cent., N. 84-8 per cent. ; Bugalla Isle, M. 68-7 per cent., N. 31-3 per cent. "If one examines flies daily," writes Dr. Carpenter, " and notes the proportions of M. and N., one is at once struck by the fact that from day to day the proportions are very different. If, for instance, there are fresh tracks of Varanus (the large monitor lizard known in Uganda as ' Nswa-swa ') on the sand when flies were caught, the proportion of N. to M. of that day is much greater than the average. This shows with what eagerness the flies avail themselves of any source of food which presents itself ; and also, that it is quite impossible to gauge correctly the proportions of M. to N. from one or two casual examinations in any particular locality. " The proportions of M. to N. were found to vary considerably from one month to another, but when the results were plotted out in a curve, and compared with the curves of temperature, humidity, number of flies, etc., no correlation whatever could be seen, and the variations appeared purely fortuitous." As regards the origin of the non-mammalian blood, it was found from examination of a large number of flies that this works out as follows: — Avian, 4-3 per cent. ; reptilian, 95-7 per cent., the chief source of the reptilian blood being undoubtedly Varanus. " Amphibian or ophidian blood," says Dr. Carpenter, " has never been found." Dr. Carpenter adduces facts showing that, in spite of the difficulty in deciding whether G. palpalis imbibes vegetable juices in addition to its normal diet of blood, " strong evidence has been obtained," by microscopic examination of the gut contents of no fewer than 1,239 flies, " that it feeds from plant tissues." In a similar manner " definite microscopical evidence " has been secured that, " on occasion," the fly sucks up water. THE DRINKING OF WATER BY TSETSE-FLIES. — On this subject, besides the statement by Carpenter to which reference has just been made, and Lloyd's remarks quoted on p. 84, additional information has been furnished both by Lamborn and Swynnerton. The former, writing with reference to G. morsitans and G. brevipalpis in Nyasaland (83), observes .:— " I have repeatedly seen flies settled on wet sand, for the purpose, I imagine, of drinking, though as the fine proboscis only is lowered, while the palps remain in the usual position in front of the head while the insect is feeding, and as it does not permit one to approach very near, it has been impossible to settle the point. I have endeavoured to study the matter further by starving flies in captivity, providing them only with wet sand, but even though they survived longer than a similar number kept entirely without water, I was not 88 quite sure even then that they did drink, though I thought I saw one or two do so. " I have endeavoured to test also whether flies ever drink dew, and whether they will feed on various fruit juices, saline solution, etc., but have not obtained any positive results." After experience gained by studying G. brevipalpis, G. morsitans and G. pallipides in Portuguese East Africa, Swynnerton writes (145) : " My flies frequently applied their proboscides to grass-stems put into the tube wet (to simulate the effect of dew), never to those inserted dry. My native informants on the Umvuazi agreed in asserting that they frequently saw tsetses in numbers ' drinking ' on the wet sand of the. river in very hot weather. I have watched Tabanus on the wet mud of pig wallows at the Amanzimhlope head-waters with its proboscis in definite contact with the moisture, as were those of non-biting Muscid flies of several species that were also present, and this habit of drinking is a common one in hot weather on the part of both Diptera and butterflies. Kinghorn and Lloyd have both noticed tsetses at the edges of puddles ; Moiser obtained records of it from his natives, and saw his captive flies insert their proboscides into the wet soil of their bottles ; and Lloyd obtained the same result from moistened sponge and moistened blotting paper that I obtained from wet grass. Carpenter has gone further, for he has traced the presence of the liquid inside the fly. The point is not merely of importance in relation to the hot- weather needs of the fly ; it may also be important, it seems to me, in relation to the possibility of poisoning the highly localised male swarms of G. morsitans by spraying the grass they rest on. " Lloyd's positive results from slices of water-melon go to confirm Maugham's observations (under natural conditions) as to the sucking of vegetable juices [cf. also Schwetz's observation of the sucking of leaves of plants by males of G. brevipalpis in North Katanga — p. 58]. It is probably moisture rather than real nourishment that would thus be sought and obtained." EXPERIMENTS IN FEEDING TSETSE-FLIES WITH EXTRAVASATED BLOOD. — In order to feed Tsetse-flies with extravasated blood, Dr. Rodhain and his colleagues of the Mission Scientifique du Katanga (October 1910 to September 1912) devised a small piece of apparatus, very simple yet very ingenious, by aid of which the insects were fed without the slightest difficulty. The object was to cause famished Glossina morsitans to bite through a pervious membrane, into blood kept under pressure by force of gravity. The following is a description of the apparatus, extracted from the report (116) on the Travaux de la Mission Scientifique du Katanga (Drs. Rodhain, Pons, Van den Branden and Bequaert. — See Fig. 8). Over the lower orifice of a small glass cylinder, 3 cm. in height by 1 cm. in diameter, is stretched a piece of fresh skin of some animal (preferably a mouse or rat, in which the skin is very fine and supple), with the hair outwards. Into the upper orifice is fitted, by means of a rubber joint, an ordinary graduated pipette, of the capacity of 1 c.c. and marked in divisions of 0-1 c.c. The lower cylinder being completely filled with citrated blood, on fitting in the pipette the liquid rises into the latter, to a height which is easily controlled. 89 The surface of the blood in contact with the layer of skin is then subject to the pressure of the entire column above it, which it is easy to measure in millimetres. This arrangement enabled the Mission Scientifique not only to estimate the tension that liquids must possess in order that Tsetse-flies may gorge themselves, but also to measure the volume of blood imbibed by the insects when feeding. In the course of a number of experiments, during which more than 100 G. morsitans fed from the apparatus, it was found by the investi- gators, who employed citrated sheep's blood and fresh skin of the common rat, that, in order that the flies may suck blood, the latter needs to be subject to a pressure which is in reality very slight, amounting to almost nothing. Fig. 8. — Apparatus for feeding Tsetse-flies with extravasated blood, devised by Drs. Rodhain, Pons, Van den Branden and Bequaert (Mission Scientifique du Katanga. — 116). A. — Apparatus complete (half natural size). B- — Lower part of apparatus (twice natural size), showing piece of skin stretched over a rubber ring (b), which is fitted on to the lower end of a short glass tube ; (a) rubber joint in upper end of short tube, carrying a graduated pipette. Subsequently Dr. Rodhain and his colleagues sought to determine yet more precisely the limit of pressure to which a liquid must be subject, in order that G. morsitans may suck its fill. With this end in view, another little device, likewise extremely simple and highly ingenious, was constructed, for the description of which the reader 90 must be referred to the original work. The conclusions deduced from these experiments are stated as follows : — " In order that G. morsitans may gorge itself, it is not necessary that the liquid upon which it is feeding should be subject to positive pressure. " It may be asked why these flies do not suck up extravasated blood, offered to them in that condition, in a receptacle not covered by a cutaneous membrane. We believe that the role of the latter is purely mechanical ; that it serves solely to guide the effort of the fly, which is accustomed to 'bite through a wall penetrable by its proboscis. " As regards the quantity of blood that Tsetse-flies imbibe at a meal, the average amount that appears to satisfy their appetite and causes the abdomen to become greatly swollen is, according to our experiments (in the course of which the pressure of the liquid absorbed has varied from 11 cm. to 4-5 cm.), from 0-056 c.c. to 0-06 c.c. Certain flies are capable of imbibing as much as 0-091 c.c., but this is the maximum observed by us. " We have succeeded in causing specimens of G. morsitans to imbibe blood serum devoid of red corpuscles, but it is only when greatly famished that they gorge themselves with this fluid. Similarly, they have been given blood containing neutral red in a solution of 1 in 3,000, and tartrate of soda in a solution of 1 in 20,000. The latter appeared to have an injurious effect upon their digestive functions. " It has also been observed, on several occasions, that Tsetse-flies imbibe blood more readily when its temperature is raised to 36° C. or 37° C. (96-8° F. or 98-6° F.) than when it is cooler. In the same way, if it be desired to feed these insects in vitro several times in succession, it is necessary to take certain aseptic precautions, for should the flies imbibe blood already contaminated with a certain number of bacteria, the multiplication of the latter continues in the alimentary canal, and the insects soon die." CHAPTER IX. RELATION OF TSETSE-FLIES TO THE SMALLER ANIMALS. Mr. LI. Lloyd (89), writing on the bionomics of Glossina morsitans in Northern Rhodesia, remarks that: — "The dependence of the tsetse- fly on the larger mammals . . . depends on the ability of the smaller mammals, birds and reptiles to avoid the insect." " The behaviour of some of the smaller animals when placed in a cage with tsetse," continues Mr. Lloyd, " was studied in the laboratory. The flies used were freshly caught, and most of the experiments were carried out in a glass cylinder 10 in. high and 6 in. diameter." The following animals were employed as subjects. Mammals : " An adult wild rat " ; "A young wild rat, just weaned " ; "A burrowing rodent (a species of MURIDAE) " ; "A very young specimen of another species of MURIDAE " ; "A young wild mouse " ; "A mole rat (Georychus) " ; " A species of dormouse " ; "A shrew (Crocidura sp.) " ; and " A young banded mongoose (Crossarchus sp.)." Birds: "A fowl." Reptiles : "A skink lizard " ; "A young gecko " ; "A chameleon." Amphibians : " Two species of toad." After detailing results, Mr. Lloyd observes : " The conditions under which the experiments were carried out were very artificial, the animals, and still more the flies, being eager to escape. It is apparent, 91 however, that the tsetse is willing to feed on the smaller animals. On the more active ones it would have little chance of obtaining a meal when they are alert. Many of the smaller mammals are nocturnal, and spend their days sleeping in hiding-places which are the same in many cases as those which the tsetse haunts. This applies also to many of the nocturnal birds. It is a very common experience to see tsetses fly out of a burrow in the ground or a hollow in a tree, while the numbers of pupae which are taken in such positions show that they are much frequented by the female flies. When asleep, such animals would probably form a ready prey to the fly, as is the case with man and the mosquito .... It is therefore possible that these small animals supply a larger proportion of the food of the tsetse-fly than is generally supposed." In Nyasaland, Lamborn (84) considers that G. morsitans derives part of its food from baboons, which occur in all districts in which this species has been studied. Bare-necked birds, such as vultures, according to this author, may furnish food material to a slight extent,1 but in trie case of other forms Lamborn believes that the presence of feathers renders feeding very difficult. The scarcity of reptiles appears to negative the supposition that these animals may furnish a blood supply, and small lizards are probably too active to be of material value as a source of food. Tests with toads and tree-frogs made by Lamborn yielded negative results. It should, however, be noted that the evidence that G. morsitans does not normally feed on non- mammalian blood is less conclusive than would appear from Lamborn's statements. The preference of G. palpalis, at any rate, for the blood of reptiles, such as crocodiles and monitors, under natural conditions is well known,2 and even in the laboratory a similar catholicity of taste is displayed by this Tsetse fly. " Minchin, Gray and Tulloch (1906)," writes Roubaud (96), " fed Glossina palpalis in captivity on lizards, chameleons and birds ; Zupitza (1908) induced this fly to bite fish of the genus Periophthalmus ; and in the laboratory at Brazzaville we ourselves have fed it with ease on toads (Bufo regularis Reuss), frogs, young crocodiles, chameleons, fishes of the Silurus group living in streams which are non-perennial, small nocturnal lemurs (Galago demidoffi Fisch.), etc. G. palpalis also bites birds, although not so readily." 1 At Kashitu, Northern Rhodesia, Lloyd (91) saw two G. morsitans on the bare skin of the neck of a ground hornbill, one being in the act of sucking blood. - Fiske (54) states that on Lake Victoria, in Uganda, the crocodile and the monitor (Varanus}, both of which are amphibious in habit, are the two favourite hosts of G. palpalis. Not only are they the most attractive to the fly, but the most favoured breeding grounds of the insect are frequently identical with spots selected by the crocodile for breeding places, or by Varanus as a basking ground. G. palpalis therefore becomes almost a specific parasite of these reptiles. It is considered by Fiske that complete extermination of the four principal hosts (crocodile, Varanus, situtunga and hippopotamus) on Lake Victoria would cause a reduction in density of infestation amounting to 95 to 99 per cent. Many years ago the late Robert Koch stoutly proclaimed the paramount import- ance of the crocodile as the principal host of G. palpalis in what was then German East Africa. With reference to G. palpalis Fiske writes : — "There are two tests for distinguishing between attractive and repulsive hosts : — "(1) An attractive host is apt to collect a 'following swarm of fly.' This curious phenomenon is associated with the assembling of the sexes. . . . " (2) The range of fly — especially its range inland from the shore — is likely to be notably extended through the presence of favoured hosts." 92 Lamborn's statements as regards G. morsitans and baboons are endorsed by Swynnerton (145), who remarks with reference to Tsetse- flies and the smaller mammals in North Mossurise, Portuguese East Africa : — " The natives of the morsitans area were unanimous in their statement that tsetses feed freely on baboons, and that ' wherever you find baboons you will also find fly.' They gave me many instances. in which baboons driven from their gardens had left numerous replete flies behind, and others in which flies were attracted in numbers to baboons that were killed. Simpson's similar personal observations and native statements on the Gambia and Gold Coast will be recalled, as will Lamborn's indirect evidence. Monkeys ( Cercopithecus pygerythrus,. which lives in lower bush than C. albogularis beirensis and raids natives' gardens) were stated more rarely to have done the same, and — an important statement if true — it was stated that replete tsetses were occasionally found where cane-rats (abundant in much of this country and very largely diurnal) had been many together." On the subject of pigs as an important source of food-supply for Tsetse-flies, Swynnerton writes in the same paper: — "Pigs were as universally incriminated as baboons, and here I obtained a quite excellent instance myself. In the ' Oblong ' (east end) in cloudy weather we walked right on to four bush-pigs sleeping. In their hasty rush they left the flies behind, and these streamed after them in great numbers and with quite a hum. We captured nearly 20 that through repletion could scarcely fly ; all but one were brevipalpis, the exception being a pallidipes, and more than half were females. Austen records both morsitans and brevipalpis as feeding freely on wild pigs, and quotes Dr. Hearsey's statement that G. morsitans was seen to settle literally in hundreds on the carcase of a wart-hog, behind which animal I also took these flies. " My indirect evidence of the value of such animals as pigs to the fly was also interesting. In a mile-wide patch of primary forest east of the Sitatongas, in which I saw much brevipalpis, both native information and a careful search for spoor showed that it could have been feeding on nothing but pigs, baboons and smaller fry. The same applies to a piece of high, dense, secondary forest on the western foot of the hills, into which, according to the owner of a kraal on the spot (confirmed by the usual search for spoor), no big game had entered for some months. It also applied, I am certain, to much of the rubber forest area, in which pigs and little blue duikers — and these only — are abundant . . . " Finally, working for over a fortnight round my camp on the Buzi east of Spungabera, in an area in which game is relatively abundant, I made a special, point of studying the daily spoor in relation to the distribution of G. brevipalpis. To sum up the result, there was a considerable area that I am certain was not entered by big game during my stay, or for some days before it. There was a smaller, inner area, immediately round certain kraals, that had probably not been visited by such animals for many weeks or months. In general the spoor showed much less movement of the big game now than in the wet season, yet the fly was equally present throughout, lurking in all the thickets to attack passing animals. " The only ' passing animal ' that showed a similar ubiquity was the bush-pig, and I was convinced from the evidence that the fly was living practically entirely on bush-pigs at the time of my visit. Man was not being attacked. My friend Mr. G. D. Otterson spent a few days with me here, and declared on leaving that the evidence of the 93 fly's independence of the bigger game in the matter of food was a ' complete eye-opener ' to him. " I conclude from these observations that any attempt in north-west Mossurise to destroy the fly by starving it in its permanent haunts is doomed to failure if the bush-pigs, and perhaps the baboons also, are not destroyed ; and the destruction of the pigs in this type of country is not easy." For various reasons Swynnerton is not apparently inclined to attach great importance to lemurs, certain bats, and various small, " largely diurnal " rodents and insect! vores. He thinks that " the part played by the small nocturnal mammals, including bats, needs far fuller testing than it has received," but adds that : — " Dassies or rock- rabbits (Procavia) may be of much use to the fly, but very locally and only in rocky places."1 In certain localities not only bush-pigs, but domesticated pigs play an important part in the sustenance of Tsetse-flies. Thus, according to Greggio (56), in the Inkissi Valley, Central Belgian Congo, while there is no direct correlation between the existence of the human trypanosome and that of the pig, indirectly the keeping of pigs con- stitutes a serious danger to the inhabitants, the animals affording an abundant and favourite food-supply for Glossina and encouraging its increase. The pigs roam freely about the country in search of food, which is often to be found in the most fly-infested spots, and return to the villages bearing flies on their bodies. These include several species, G. palpalis being the most common. In a subsequent paper (58) the same author states that white pigs are extremely rare in the Inkissi Valley, but are attacked by Glossina to the same extent as are black or brown animals. Though the fly certainly prefers a dark colour, when a herd is composed entirely of white individuals it attacks them readily. Animals kept in the villages, which never go down to the river, are equally affected, because other pigs that are running free bring back palpalis to the village, and animals tied up close to the huts are much worried by the flies. Again, in the Kwilu district, Western Belgian Congo, according to Schwetz (132), G. palpalis is established permanently in small isolated thickets near or actually in villages on hills, and sometimes at a con- siderable distance from any water, and the presence of pigs in the thickets referred to is said to be the reason for this strange phenomenon. Christy (29), while attaching special importance from our present point of view to the bush-pig and wart-hog, considers the semi-domesticated pigs frequently seen in native villages to be even more, dangerous. Lastly it may be noted that, in the remarkably successful measures adopted some few years ago for the ridding of the Island of Principe from G. palpalis and sleeping sickness, the destruction of the wild (feral) pigs, which swarmed in the island, occupied a prominent place (see p. 142). The extent to which Tsetse-flies feed upon the smaller animals — mammals, birds and reptiles — is a matter of great importance, and all observations upon this subject that can be collected will be extremely useful. This brings us to one of the most hotly debated questions, namely, that of big game. 1 Lloyd (91) states that in three instances in Northern Rhodesia, in which G. morsitans was feeding on rabbits, the flies were observed to be distended with a clear fluid instead of blood ; this may have been serous fluid, obtained by piercing the body wall. 94 CHAPTER X. TSETSE-FLIES AND BIG GAME. OUTBREAKS OF CATTLE TRYPANOSOMIASIS APPARENTLY DUE TO BITING FLIES OTHER THAN Glossina. In the whole subject of Tsetse-flies, the two matters most in dispute are undoubtedly the dependence or otherwise of these insects upon big game for their food-supply, and its destruction as a means of combating them. Below will be found a resume of some of the arguments put forward, by the opponents as well as by the defenders of big game living in regions infested by Glossina. There can be no doubt, as is clearly shown by Bruce, Foa and many other writers, that G. morsitans has long lived at the expense of African big game ; but since the latter is still very abundant in many localities where the fly exists, it seems evident that these wild animals now exhibit, if not complete immunity, at least a very high degree of tolerance towards the trypanosomes that exist in their blood in small numbers without causing serious ill effects.1 Yet it is from this blood that Tsetse-flies obtain the germs of the disease that they inoculate into the blood of domestic animals. In the Katanga district of Belgian Congo, the simultaneous abundance of big game and Tsetse in certain zones has been reported on several occasions by prospectors looking for land suitable for stock-raising. The more plentiful the game along the course of a stream, the greater would be the numbers of G. morsitans, while tracts where there was no fly would also be the poorest in game. Since the discovery of the transmission by G. morsitans of the trypanosome known as Trypanosoma rhodesiense, which is very closely related to T. gambiense and likewise produces a form of sleeping sickness in man, efforts have been made to determine whether big game harbours this parasite. Kinghorn and Yorke, after seeking for the reservoir of T. rhodesiense in Northern Rhodesia, stated that they met with the organism in the blood of the following animals : — Cobus ellipsiprymnus (waterbuck), Bubalis lichtensteini (Lichtenstein's harte- beeste), Aepycerosmelampus (impala), Tragelaphus scriptus (bushbuck), and lastly Phacochoerus ethiopicus (wart-hog) . Apart, however, from T. rhodesiense, which, as has just been stated, is a cause of human trypanosomiasis, the same authors have shown that big game may be the source of animal trypanosomiases. Out of 127 wild mammals, belonging to 19 genera, examined by Kinghorn and Yorke at Nawalia (N. Rhodesia), 33 contained trypanosomes, the following species being found to be infected : — Cobus ellipsiprymnus, 1 According to Dr. Andrew Balfour, C.B., C.M.G., who has studied the animal trypanosomiases in the Lado (Upper Nile), in Southern Kordofan there is a small race of black cattle, which, exactly like big game, appears to be immunised to trypanosome diseases. These cattle seem to be the only ones capable of continued existence in the infested Koalib zone, in which G. morsitans abounds and transmits a malady which is apparently due to Trypanosoma brucei. This observation may be compared with another by Dr. J. Pollard, W.A.M.S., who, in his " Notes on the Tsetse-Flies of Muri Province, Northern Nigeria" (112), remarks : — " There is in the Munshi Division, and in the northern part of the Province, a small black breed of cattle which is apparently immune to tsetse. At any rate these cattle can be kept in the Munshi district where no horses can live and where imported Fulani cattle all die." 95 Tragelaphus scriptus, and Strepsiceros kudu (koodoo). At Ngoa, also in Northern Rhodesia, out of 124 mammals, trypanosomes were detected in 21. x In Nyasaland, the plain extending inland from the western shore of Lake Nyasa is for the most part " covered with thorn scrub," which is " the home of the Tsetse-fly [G. morsitans] and also of numerous wild animals." In studying the trypanosomes found in the blood of wild animals living in this area, the members of the Sleeping Sickness Commission of the Royal Society (Surgeon-General Sir David Bruce, C.B., F.R.S., A. M.S. ; Majors David Harvey and A. E. Hamerton, D.S.O., R.A.M.C. ; Dr. J. B. Davey, Nyasaland Medical Staff ; and Lady Bruce, R.R.C.) adopted the following procedure. When an animal was shot in the fly-country by one of the Commission, a specimen of the blood was " taken in a bottle containing citrate of soda solution," and despatched by a motor-cyclist to the camp, where it was " at once injected into a goat, a monkey, and a dog." Thick and thin films of the blood were also " spread on glass slides for microscopical exami- nation." In this way, among "180 wild animals shot in the fly-area adjoining the Commission's camp at Kasu .... 57 were found to harbour pathogenic trypanosomes — 31-7 per cent."2 The animals proved to be infected were the waterbuck, hartebeeste, reedbuck, duiker, koodoo, eland, bushbuck, oribi, buffalo, hyaena, and wart-hog. In the opinion of the Sleeping Sickness Commission, in the fly-district in question the first four are " dangerous neighbours to man ; the eland, koodoo, bushbuck and buffalo to cattle, goats and sheep." It is also pointed out that " the wart-hog is the only animal which harbours T. simiae, the lightning destroyer of the domestic pig." " We see, then," writes Dr. Bouet (11), " the important part played by the large antelopes and other big game in the propagation of trypano- somiasis. These animals constitute the reservoir of virus whence Tsetse-flies derive the trypanosomes, the development of which takes place in the alimentary canals of their insect hosts. Trypanosomes have no influence upon big game, and do not seem to affect their health in any way." Several proofs, obtained chiefly in French West Africa, of the fact that big game is the sole reservoir of trypanosome virus are adduced by Dr. Bouet, who adds that, " without wishing for the total destruction of our African fauna, it is desirable that the restrictive measures that must be taken for the protection of the game should not be too pro- hibitive, since there can be no doubt that, from the day when the native is prevented from hunting or forced to pay a tax for the right to do so, there will be a recrudescence of trypanosomiasis in consequence of the swarming of game." Another great argument of the partisans of the destruction, or at least the driving back of the big game in Central Africa is based upon the fact that, as a sequel to the rinderpest that ravaged the Transvaal and Rhodesia some five and twenty years ago, much of the big game, and with it the Tsetse-flies in large measure, disappeared from those regions. It may be urged, however, by way of objection to this, that 1 Out of 256 monkeys examined not a single one was infected, which, in the opinion of Kinghorn and Yorke, is due to the fact that these animals catch Tsetse-flies before the latter have time to bite them. 2 " The species of trypanosomes found are T. brucei vel rhodesiense 7-8 per cent., T. pecorum 14-4, T. simiae 1-7, T. caprae 11-1, and T. ingens 1-7." The last- mentioned species is not at present known to be pathogenic either to domestic animals or to man. 96 it is by no means certain that the disappearance of the fly at the time of the rinderpest was due to the diminution of its food-supply. While on the one hand we have no information as to the climatic conditions prevailing at the period in question, experiments have shown that low or high temperatures have a considerable influence upon the life of Tsetse-flies, and consequently the disappearance of Glossina morsitans at the time of the rinderpest may have been nothing but a simple coincidence. It would seem that trypanosomes have never been recorded as occurring in the blood of the elephant or hippopotamus. It has indeed been asserted by certain big game hunters and travellers that G. morsitans evinces a special predilection for the latter animal, and it is said that wherever there are buffalo these flies are found to be infected with trypanosomes. In 1914 a study of the relation between Glossina morsitans and big game in Southern Rhodesia was published by Mr. R. W. Jack (68), to whose interesting investigations repeated reference has been made in the course of the present work. We give below some extracts from Mr. Jack's paper. After mentioning the sources of his information, which was derived not only from personal observation but also from the testimony of other Europeans especially qualified to express an opinion, Mr. Jack writes : — " It may be stated at once that the case built up is considered strongly in favour of a vital association between the prevalence of big game and the continuance and increase of the fly .... During the last four and a half years the writer has visited the great majority of the fly-belts in the territory, and the main belts repeatedly, and, broadly expressed, the results of observations on game and fly are to the effect that in most cases game is more or less abundant all the year round in fly-infested country, and that in no instance is the larger animal life altogether absent, even during a portion of the year. " Turning now to broader considerations, the evidence in favour of the necessity of big game to the tsetse in Southern Rhodesia and adjacent territory may be summed up under four heads :— " (1) Tsetse retired before the advance of civilisation in the Trans- vaal, the only known modification of conditions being the destruction of the game. " (2) Tsetse disappeared from large tracts of country immediately after the rinderpest epizootic in 1896. " (3)r Tsetse has increased and spread since the rinderpest only in those parts of Southern Rhodesia where big game has increased. " (4) Tsetse has greatly decreased of late years in the Hartley district in those parts where the big game has been most effectively destroyed or driven away." With regard to the possibility that big game is not necessarily the only source of the requisite blood-supply, and dealing in the first place with reptiles and amphibians, the author remarks : — " As a matter of fact, in the case of G. morsitans such a supply is rarely available, as the fly is not often found on the banks- of the larger rivers where crocodiles abound, and the belts in the dry season are frequently far removed from water of any sort, in which case water-loving reptiles, such as Varanus and fresh-water Chelonia, as well as most amphibians, are not available. The smaller lizards, including chameleons, are 97 more likely to make a meal of tsetse thanwc^ versa, whilst the terrestrial tortoises are too scarce to be of any account. Amongst the mammals, certain orders, namely the Insectivora, Chiroptera and Edentata, are obviously of no service on account of their nocturnal habits and the seclusion of their diurnal retreats. Certain rodents, such as hares, may serve the fly for a meal on occasions, but the order can be of but little importance on account of the small size and activity of the majority of its members and the fact that they are largely nocturnal in habit. Of the larger species the porcupine (Hystrix) is entirely nocturnal, but certain squirrels attain a size not greatly inferior to that of a rabbit and are of diurnal habit. The larger forms of the latter family are, however, rarely seen in country suited to the tsetse. No doubt the larger Carnivora are attacked by tsetse when they enter its haunts, but their numbers are relatively so small that they are practically negligible .... The hunting dog (Lycaon) is always migratory, and leopards and jackals, in addition to their scarcity, lie up during the day. The smaller Carnivora — FELIDAE, VIVERRIDAE, MUSTELIDAE — are also practically nocturnal, and from their alert and active habits are unlikely to submit to be fed upon .... It would seem, therefore, that the whole class of Mammalia, with the exception of ungulates and certain primates, are little fitted to be of service to tsetse even in the aggregate." As regards birds, Mr. Jack considers that there is " some ground for belief that for some constitutional cause, such as their conformation, their- armature of feathers, their activity, their habit of pecking at Insects, or such causes combined, birds as a class do not form a very suitable source of sustenance to tsetse." Finally, with reference to G. morsitans and the buffalo, Mr. Jack asserts that " it is quite certain that in Southern Rhodesia to-day the fly is not in any way dependent upon this species of animal," seeing that as a rule the two do not occur together in the same districts ; and he remarks in conclusion that " no one having travelled through the fly-belts in Southern Rhodesia .... could entertain the idea that, except in a few localities, the blood of the buffalo is, even at long intervals, a regular food for the tsetse, much less that it is an essential one." The defenders of big game commence by insisting that it is by no means proved that the blood of such animals is the sole or even the chief food of Tsetse-flies, as is shown by the conditions in the islands in Lake Victoria, which are excellent breeding grounds for G. palpalis, in spite of the non-occurrence of any large wild animal. The simul- taneous presence of Tsetse-flies and big game in certain areas is stated to be due to the fact that, during the dry season, G. morsitans elects to live and breed in shady, damp places, which at this season of the year are likewise frequented by both big and small game. On the ground that from big game Tsetse-flies derive the food necessary for the perpetuation of their species, the extermination of the animals in question in all countries ravaged by Nagana is advocated by some, even to the extent of demanding that all protective enact- ments should be abolished, and that the big game still existing in African colonies and protectorates should be destroyed wholesale. Apart from the excessive nature of. such demands, it may be asked what results would be obtained by methods of this kind. If the whole of the game in a given region were destroyed, would stock-raising in countries infested by Tsetse be possible ? We venture to doubt it. The result of such a destruction of life would probably be (5979) G 98 to compel the flies to feed, in so far as they do not do so already, upon small mammals, such as rodents and the like, and upon birds, reptiles, batrachians, etc. According to M. Trouessart, of the Paris- Museum, Tsetse would perhaps diminish in number, but would not become extinct, and even though the keeping of domestic animals in the district concerned were to cease for ten years, at the end of this- period the deadly flies would again begin to swarm and to produce the same results as before. Moreover, many of those who have had great practical experience of Tsetse-flies are by no means of the opinion that the distribution of the insects is governed by the presence of big game. We shall here confine ourselves to quoting the conclusions of papers published in recent years by certain observers, including Sir Alfred Sharpe,, Major J. Stevenson Hamilton, Dr. S. A. Neave and Mr. C. F. M. Swyn- nerton, all of whom are entitled to speak with authority upon the subject. During six or eight years Sir Alfred Sharpe, K.C.M.G., C.B., formerly Governor of the Nyasaland Protectorate, made very careful observa- tions in the areas infested by Glossina morsitans in Nyasaland, North- Eastern Rhodesia, and the portions of Tanganyika Territory and Portu- guese South-East Africa bordering on Lake Nyasa, all of which regions he frequently traversed. In a paper (137) which appeared some eleven years ago he writes : — " My own opinion, which I have expressed on various occasions before now, is that the existence of wild game has very little, if anything, to do with the existence of any species of tsetse- fly. At the time when rinderpest visited Nyasaland, there was a noticeable decrease in the quantity of big game in some of the districts where tsetse-fly exists, especially as regards buffalo, but I have never noticed that this had any effect on the presence of Glossina." Sir Alfred Sharpe does not share the opinion of those who think that G. morsitans is " spreading," and he considers that fly areas do not " alter their limits to any appreciable extent." In concluding his remarks, he observes :— " There are various causes for fly being now more noticed : — more attention than formerly has been called to the subject ; cattle are now kept by natives in districts where formerly „ owing to wars and raids, there were none ; large native settlements have altered their locations (notably at Fort Johnston, at the south end of Lake Nyasa), and as the old clearings grew up fly appeared. Increased transport by ox-wagons has revealed the existence of fly also in places where it was not noticed before." Major J. Stevenson Hamilton, Warden, Transvaal Government Game Reserves, in a paper on "The Relation between Game and Tsetse- Flies" (62), writes:— "The theory that the larger wild mammals, commonly called 'big game/ are solely and entirely responsible for the presence of tsetse-flies has become so firmly fixed in the mind of the average ' man in the street ' in South Africa, that no amount of contrary argument or even proof, were such forthcoming, could ever have much effect in altering his opinion. The newcomer quickly assimilates the same idea, and after a time begins to voice it as asser- tively as his mentor. It is thus most difficult to obtain really reliable data. An investigator, beginning an enquiry with an open mind, finds himself flooded with such a mass of apparently well-substantiated statements, provided by experienced and obviously sincere persons, that he can with difficulty keep his mind free from a certain amount of bias." Nevertheless, from " personal experience," gained during a journey through the northern part of Portuguese East Africa in. . 99 1908, " of the existence of G. morsitans in large numbers where there was little or no sign of the larger mammals/' Major Stevenson Hamilton feels " fairly convinced that/' within the area in question, " the fly .... exists to a great extent independently of the blood of large quadrupeds." In his paper on the " Blood-sucking Insects of Eastern Tropical Africa " (103), Dr. S. A. Neave, who has closely studied the East African Tsetse-flies, expresses himself as follows : — " With regard to the rela- tions of G. morsitans with game, I must unhesitatingly group myself with those who consider that the presence or absence of big game is not the primary factor in determining the distribution of the fly. That the presence of a large quantity of big game within a morsitans area might influence the numbers of the fly by increasing the food supply is probably true, but that the game materially affects the distribution of the insect I do not for one moment believe. The majority of those who hold the opposite view have acquired their experience in Northern Rhodesia and Nyasaland and .... have probably been misled by peculiar and abnormal circumstances. It is now, I think, pretty generally admitted that, north of the Zambezi, at any rate, there is no special relation between G. morsitans and buffalo. Certainly in this part of Africa their distribution by no means coincides. Where it does so, these animals are certainly very attractive to the fly, probably because they are slow-moving and are usually in herds of some size." In concluding his remarks upon G. morsitans the author writes : — " In view of the strong probability that this species can feed on a great variety of vertebrate life, the mere destruction of the large game would seem at first sight absurd, unless the said game can be proved to be the sole reservoir of a noxious trypanosome. If a serious attempt is to be made to starve out the insect, the only logical step would involve the removal or destruction of all such classes of animal life, including man and his domestic animals." As regards the efficacy of game destruction in reducing the numbers of G. morsitans, it was found in Southern Rhodesia, in 1916 (17), in the area between the Sengwe and Umniati Rivers, that the fly appeared to be extending its range ; and that, although the game laws had been suspended, there was no appreciable decrease in the numbers of game, except in the case of the elephant, to which to some extent hunters were devoting their attention in preference to other game. It also appeared that the opening of certain areas to free shooting had had the effect of driving game thence into inhabited country. Nevertheless the Director of Agriculture considered that, in a clearly defined zone, the extermination of game with the aid of natives was feasible and likely to have the desired result. Later in the same year Mr. R. W. Jack reported that G. morsitans was continuing to spread rapidly in certain parts of the Sebungwe district, Southern Rhodesia. In the south-west corner of the great belt in that region, the fly had advanced in a southerly direction from the Mzola to the Kana River, and in a westerly direction to the Shangani River. At the head-waters of the Mzola River, although no actual advance was recorded, G. morsitans was certainly increasing rapidly. An advance of seven or eight miles had been made up the Sengwe River since 1914, and the fly was reported also to have spread north- ward near the Sengwe and Sasame Rivers ; at other points visited by Mr. Jack no advance since 1914 was apparent. In some of the 100 . cases mentioned the spread of the fly was believed by Jack to have been due to the movements of game. In a paper published three and a half years later (74), Jack stated that all data collected by him tended to confirm the dependence of G. morsitans on the presence of wild ungulates ; at least there was every reason to believe that the fly feeds upon them by preference and thrives best in their company. According to Dr. H. L. Duke (49), whose experience has been gained in Uganda, in no case has prolonged existence of G. morsitans been discovered in an area where game is entirely and permanently absent. The fly is generally found associated with game in considerable numbers and variety, though it may occur in localities where wild ungulates are apparently scarce. The theory regarding buffalo as essential to* the subsistence of the fly has been found untenable. Lloyd (91), in the Ngoa area, Northern Rhodesia, in 1915, where game was plentiful, collected four times as many pupae of G. morsitans •as in the Kashitu area, where game was relatively scarce (see above, p. 41). He therefore thinks that the flies themselves are more numerous in districts in which game is abundant. Man was frequently attacked where game was scarce or non-resident, while the flies were rarely seen where game was present in large numbers. Lloyd considers that the reduction of game in a fly area would lead at first to an apparent increase in the numbers of the fly, owing to increased hunger. The number of females — which approach man with the deliberate intent of feeding, unlike the males, which chiefly do so as a means of meeting and mating with the opposite sex — in a capture would increase to about 50 per cent., while pupae would occur in fewer numbers, thus showing that an actual reduction had taken place. On the other hand Christy (29) states that large areas of country may swarm with Tsetse, and yet contain scarcely any game ; such areas, according to this author, are found in the Upper Bahr-el-Ghazal, where tall spear-grass is a conspicuous feature. In some districts, both there and in the Eastern Welle basin, G. morsitans occurs in millions, apparently irrespective of whether game or human beings are present at all. The presence or absence of game depends upon grazing facilities, but these have no relation to the presence or absence of Tsetse-flies ; nor has the prevalence of sleeping sickness any relation to the number of G. morsitans, or to the number of game in any given area. Wild animals may act as hosts for the trypanosome of sleeping sickness, but Christy considers it a dangerous assumption to conclude that they are the chief reservoir of the disease. Even if this were the case, he regards the extermination of all the wild animals in any part of Tropical Africa as quite impracticable, nor does he think that such a measure would have the effect of eliminating the disease. Though excluding the majority of wild animals as a danger in the transmission of sleeping sickness to man, this author is inclined to suspect one or two. Of these, pigs are considered the most dangerous, and not only the wild bush-pig and wart-hog, but, as has already been stated (see p. 93), more especially the semi-domesticated pigs frequently seen in native villages. In an attempt to collect evidence in the Bahr-el-Ghazal and the Congo for or against the theory that wild animals are an important reservoir of sleeping sickness, Christy made a microscopic examination of the blood of many animals as soon as possible after each was shot. Of 160 animals examined in this way, only five were found to have trypanosomes in their blood, and of these only one, a wart-hog, 101 ;;,,,, harboured a trypanosome that might be mistaken f o^ 'that , of, disease in man. On no occasion were trypanosonfes. in buffalos. Swynnerton (145), writing with reference to the North Mossurise district, Portuguese East Africa, says : — " My finds of pupae in the morsitans area seemed rather to suggest a preference for the buffalo on the part of the fly. I found this definitely asserted by some of the natives and denied by others. My guide in the morsitans area, a very observant native, was particularly convinced of it. Questioned as to a connection between the two animals, he replied that one may find fly where there is no buffalo, but that where there is a choice it follows the buffalo. ' The buck/ he said, ' are much more restless under its attentions than the buffalo, the hartebeests especially keeping up their dance when tsetse are about them ; so that the fly can feed more easily on the buffalo ' . . . . Elephants were stated by the natives to be attended by tsetses when the latter are numerous, and I took a male brevipalpis waiting on a much-used elephant path. Roubaud and Bouet (referring to G. longipalpis) are both quoted as speaking of a special association between tsetses and the elephant and hippopotamus." Under the heading " Man — and the Fly's Preferences," Swynnerton remarks : — " Concerning the practical question, ' If other sources of food were eliminated, could the tsetse still keep going with the aid of man ? ' it is certain from the observations of Lloyd, Maugham, Steven- son Hamilton, myself and others that the tsetses would then attack man much more. Tsetses (as I have seen) constantly obtain full feeds from man and escape unscathed and, where the bush comes up to a village, so far from avoiding it, morsitans and pallidipes — but not brevipalpis — appear to become rather a nuisance." With regard to what follows, it is perhaps as well to point out that, in the area dealt with by Swynnerton, nagana rather than sleeping sickness is the all-important problem. Writing under the heading " Distributors of the Fly," Swynnerton observes : — " Game not only helps (a) to feed the fly, and (b) to provide the trypanosome, but (c) it helps to distribute the fly, carrying it back each summer into the areas from which the fall of leaf had driven it." He further says, after discussing various kinds of game : — " I believe, then, that these animals, and the buffalo in particular, are mainly responsible for the annual spread of the fly, and that so far as we are concerned, in and near Mossurise, our main grievance against the game is not so much that it feeds the fly (which would be fed and contami- nated in any case by the pigs) but that it carries it far and wide in the rainy season and so brings it into contact with the cattle." However, after considering the effect of the rinderpest outbreak in 1896 upon the game and upon G. morsitans and G. pallidipes in North Mossurise, Swynnerton comes to the conclusion that : — " The failure of the rinderpest to destroy the fly here to any appreciable extent .... tells also against the hope that we may exterminate the fly in the Mossurise district by destroying the bigger game only." Later in the same paper, in a section headed " Game Destruction," Swynnerton writes : — " In view of the evidence I have already alluded to, we are bound to keep an open mind with regard to the possibility that even G. morsitans might survive the destruction of all large mammalian life. Yet the fact that under present conditions game-paths are the regular rendezvous of the sexes, and that the connection with game appears generally to be an essential point in 102 ^ ,«Jth$ir r&tker" 'Complicated social scheme, makes one hesitate to reject '1 «.4K£ ^tferriative view. Lamborn (Bull. Ent. Res. vii, 1916, p. 37) has argued well for the view that it need not be a large game population that will support fry . . . . " In a relatively clean-stemmed area like that on the granite-gneiss, in which also bush-pigs are not over-abundant, it is to me very con- ceivable that wholesale game destruction might banish the fly. An obstacle, however, would be that the whole territory is one vast game area, so that the game would pour again into a given section of it as soon as the persecution was relaxed — unless an effective barrier were created such as could probably only be made a permanency with settle- ment behind it. Whether the returning game would bring the fly in again would depend on (1) whether the whole continuous fly area had been cleared of fly, or (2) whether the portion cleared had been split off from the uncleared portion by an effective barrier against the fly itself. " West of the Sitatongas the country is much more jungly, and wholesale game destruction is proportionately more difficult In addition this country abounds in bush-pigs, which are difficult to destroy, and which in anything approaching their present numbers can probably alone support the fly, with baboons, abundant cane-rats and other animals, which may all contribute to its sustenance. I consider that it will be impossible to starve the fly at all generally by ordinary game destruction here, at any rate before the country is very fully settled, though buffalos and elephants might be banished by adequate and persistent shooting. " What can be done is to protect particular places, like Spungabera and the British border, that are outside the fly, and are threatened only by the wanderings thence of the bigger game. Fencing, the judicious placing of native kraals, and shooting are amongst the possible measures, and organised and repeated drives might be undertaken locally. I have suggested similar means of keeping the fly from being carried into any areas that may in the future be settled. " Again, it seems to me that game destruction should check the advance of a fly-belt, whatever its effects on the ground already infested. No problems are at present better worthy of study than these — (1) the extent to which tsetses will travel independently of game, and (2) the distances that the female will travel on game." In 1918, a number of suggestions for checking the spread of G. morsitans and trypanosomiasis in the Wankie district were made by the Government Entomologist of Southern Rhodesia (18). While some of these proposals were, for various reasons, admittedly either imprac- ticable or inadvisable, it was thought that a properly organised attack on the whole area invaded or threatened by the fly would have an excellent chance of success. It was stated that this idea would involve the wholesale extermination of game, first around the boundaries of the affected area and gradually towards the centre, and that such an undertaking appeared both feasible and economically sound. Destruc- tion of winter haunts in threatened areas was likewise recommended, it being suggested that since the fly is dependent during the time that the trees are leafless upon evergreen trees in the vicinity of watercourses, the removal of such trees would probably render the locality unsuitable as a permanent fly-belt. Subsequently (17) it was resolved by the Government of Southern Rhodesia to give effect, at least experimentally, to both of the foregoing proposals. As regards the former of these, itVas decided to take steps 103 for the destruction of game within a prescribed area in the southern part of the Wankie district, within the limits of which antelopes and other mammals were to be removed as thoroughly and rapidly as possible by a series of hunts or battues, while the cleared area was, so far as practicable, to be maintained free from such animals, by means of patrols, for a sufficiently long period to test the effect on the fly. If results within the area treated should warrant a continuance of this method, operations would probably be extended to another area. As an experiment in the destruction of the haunts of the fly, an isolated patch of bush was selected, and it was proposed, after a preliminary inspection to determine the degree of Tsetse infestation of this area, to effect the complete removal of all evergreen or deciduous heavy- foliaged trees by felling, coupled with burning of the grass. Such clearing was to be repeated in subsequent seasons, with periodical inspection by the entomologist to note results. About the time when the measures mentioned in the previous paragraph were sanctioned, the Government Entomologist of Southern Rhodesia reported (18) that, in the Sebungwe and Umniati fly areas, as existing in June-Juty, 1918, records of extreme abundance of G. morsitans in certain spots coincided, with one exception, with localities where game was particularly plentiful. In the case of the exception (on the west bank of the Umniati River) it was thought that the fly might be feeding on the troops of baboons (cf. supra, pp. 91, 92), which were extraordinarily plentiful there. The fact that game in this neighbourhood had been checked- by professional hunters for the previous five or six years, without a corresponding reduction in the numbers of Tsetse, was considered disappointing, but not a proof that this method would not succeed elsewhere, the unusual prevalence of baboons being a complicating factor. A report by a committee appointed by the Natal Province, published in 1918, recorded a very great spread of nagana in Zululand owing to the presence of Tsetse-fly (G. pallidipes). In some localities natives had not been able to keep cattle for several years, and others had sustained heavy losses in their stock. Whatever arguments might be brought forward with a view to disassociating trypanosomiasis from game, it was considered that actual demonstration in Zululand had clearly proved that, once the reservoir of infection in the shape of certain species of game was removed from a locality, losses from trypanosomiasis cease forthwith. In the opinion of the committee it was imperative in the interests of a closer settlement of Zululand by both Europeans and natives, that all game known to act, or suspected of acting, as carriers of trypanosomiasis should be strictly confined within the limits of their reserve. With regard to G. morsitans and big game in Nyasaland, a recent paper by Dr. W. A. Murray (1010) states that, in the narrow forest belt extending along the western shore of Lake Nyasa, from Chintechi in the north to Fort Johnston and the Murchison Cataracts in the south, the numbers and range of the fly have both diminished and increased in correspondence with those of big game. " Enquiries from old residents at Mvera," writes Dr. Murray, " elicited the fact that prior to 1896, when the rinderpest swept through the land, and decimated alike native cattle and all kinds of large game, the latter had roamed freely over practically the whole of the forest, and that wherever game was found there was also tsetse-fly. This information was confirmed by Sir Alfred Sharpe, then Governor of the Nyasaland Protectorate, during a conversation with the writer at Dowa, in 1902." For some six or eight years after the rinderpest, according to Dr. Murray, the 104 small surviving herds of game well-nigh confined themselves to one locality — " the Tsanjoka Marsh, some ten miles south of Domira Bay " — and during this period this was the only spot where G. morsitans was to be found. From about 1904, however, when the animals began again to increase in numbers, the game " commenced to scatter over a wider area of forest, and gradually returned to their old haunts over the greater part of the forest. The fly, which had been strictly limited for six or eight years to the Tsanjoka Marsh, spread concurrently with the game." Dr. Murray adds : — " The foregoing account of the diminution and spread of tsetse-fly over a comparatively small section of the lake shore forest serves as an example of the close connection between the movements of game and tsetse-fly over the whole Nyasaland fly area. This, of course, only applies to the low-lying parts of that forest, as beyond an altitude of 3,000 ft. tsetse-fly was rarely found in Nyasaland, even though game was abundant in otherwise suitable country." However, Dr. George Prentice, of Kasunga, N'gara, Nyasaland, in a letter dated 7th May 1921 — an extract from which is published on p. 139 of the number of the journal containing Dr. Murray's communication — reports an outbreak of sleeping sickness in two villages on the Lingadzi River, " at an altitude of just 3,500 ft." Dr. Prentice also states that " with the cessation of free shooting in this area there is a very rapid advance of tsetse, and herds of cattle are going down before it, kraals being cleared." The agency of big game in distributing and extending the range of G. morsitans, though affirmed, as we have seen, by writers such as Swynnerton and Murray (see p. 101 and supra), is disputed in a recent paper by Jack (75), who, as the result of experience gained in Southern Rhodesia, expresses himself as extremely sceptical concerning this fly's alleged habit of migrating with game. During ten years' investi- gations, Jack has found no tendency on the part of the fly to migrate under the stimulus of hunger or in company with game. There is, however, always the seasonal scattering during the wet season, and sometimes there are forced movements induced by the destruction of the forest ; in the latter case the fly naturally follows the retreating shade. There has been a fairly generally accepted idea that G. morsitans locates game, and follows its movements more or less continuously, at least within the limits of infested country, but Jack attempts to show that this is not its general habit. Indications against the theory are that female flies are apparently not known to follow to any great distance at any time ; that gravid females seek seclusion and are not likely to attempt to follow a moving herd ; that hungry flies tend to feed fully and then abandon their hosts, which are unlikely to remain in the vicinity until the flies recover ; and that the fly is diurnal in habit, while game moves largely at night. It would seem, therefore, that the only individuals capable of following game, even for a few hours, are those that do not desire to feed. According to Jack, the probability is that the fly neither ranges the forest in search of its host, nor follows it when encountered for any length of time, but that it lies in ambush, waiting for the animal to come within range of its perceptive powers. The maximum distance at which a hungry fly readily detects its hosts is a short one, possibly less than 100 yds., and following on the part of females has only been proved up to about 400 yds. ; the maximum following distance in the case of this sex remains to be determined. Jack is not yet convinced that the females seek animals and human beings only for the purpose of feeding, though this appears probable. The reader, who has taken the trouble to study carefully the whole cf the statements printed in this chapter with reference to relations 105 between Tsetse-flies and big game, cannot fail to have noticed in certain instances a marked conflict of evidence and opinion among some of those best qualified to furnish both. This fact alone should make us hesitate before adopting what to some may appear obvious con- clusions. A further word of warning also seems necessary. The close association admittedly existing between certain species of Tsetse-fly and big game, in parts of the African continent, must on no account be assumed to be those obtaining between game and all species of Glossina, everywhere throughout their range. Thus, writing with regard to conditions in the islands in Lake Victoria, Uganda, between Entebbe and Jinja, Carpenter (26) considers that the extermination of large mammals would have no effect upon G. palpalis, which prefers reptilian blood (see above, p. 91), and that the destruction of all crocodiles and other large reptiles would be a difficult matter. We may conclude from what has been stated above that, though there be definite grounds for considering big game as constituting a reserve of food enabling Tsetse-flies such as G. morsitans, G. pallidipes and G. brevipalpis to subsist in a given region, there is as yet no scientific proof that the presence of these animals is absolutely indispensable to the continued existence of the insects. Under these conditions the utmost caution would seem to be necessary before definitely advising the taking of measures which would have irreparable results. " The disappearance of species of animals which adorn our globe, and are important assets in the maintenance of native populations," says M. Perrier, " besides being regrettable in the highest degree from a scientific standpoint, would have repercussions that it is impossible to foresee. By destroying the equilibrium established in nature, we may lay ourselves open to disasters of the most serious description." Before we decide to take action in the sense indicated, science must have expressed its opinion with no uncertain voice, and even though it should definitely pronounce judgment against big game, we must see that the measures that we are compelled to take do not degenerate into abuse, and are limited to zones in which the introduction of stock- raising is not merely possible but also realisable under present condi- tions. To permit the destruction of important natural resources, on the pretext that they may be inimical to more or less uncertain colonising projects, would be tantamount to sacrificing substance for shadow. A further preliminary condition is that we shall have devoted serious study to the other possible methods of destroying or banishing Tsetse-flies, and that no effective treatment for trypanosomiases has been discovered.1 1 In this connection the following extract from a paper by the late Captain R. B. Woosnam (164) is worth quoting : — " From the point of view of game preservation there are six questions which need answers, and until these questions are answered it is impossible to decide upon a definite and practically useful plan of campaign. Briefly these six questions are : — "1. Are game animals the only wild animals which are acting" as ' reservoirs ' for trypanosomes ? " 2. Are the trypanosomes found in the blood of game animals pathogenic for man and domestic animals ? And if so, are not the trypanosomes found in the blood of other animals also pathogenic ? "3. Are tsetse flies the only transmitting agents of these trypanosomes in the infected areas ? "4. Are game animals the only source from which the tsetse flies or other transmitting agents draw their blood supply ? And if not, what are the other sources of supply ? "5. Can tsetse flies live and breed upon food other than blood, such as plant juices ? " 6. Are the distribution, increase and spread of tsetse flies, if this latter occurs, dependent upon game alone ? And if not, what are the governing factors ? " 108 But science has not yet pronounced an opinion, while the views of entomological specialists show great divergence. What is needed at the moment is not hasty and dangerous measures, but an exhaustive •study of the question. The experiment indicated in the subjoined extract from a paper by Mr. LI. Lloyd (89) seems well worth' trying. " Further evidence of the relation of the fly to the larger mammals," writes Mr. Lloyd, " could be obtained by compelling it, under as natural conditions as possible, to attempt to support itself on a diet of the smaller mammals, birds and reptiles. The following experiment is therefore suggested. A large fly-proof cage, of some such dimensions as 100 yds. long by 50 yds. wide and 7 ft. high, would be constructed on a piece of country favoured by tsetse-fly and in which breeding places were known to exist. Into this cage would be introduced a number of small mammals and birds, the insectivorous species being excluded. A large number of tsetse would then be set free in the cage and daily observations as to their increase or decrease would be made by a well-veiled observer. At the end of twelve months it should be known if the fly is able to continue its species on such a fauna. In this event, increase should occur, since there would be few enemies in the cage. If the numbers of the fly decreased, the experiment would be repeated with the introduction of a few young antelopes, goats or sheep into the cage, and similar observations would be made over the same period. If the increase occurred under these new conditions, the dependence of the fly on the larger mammals would be made clear. The first of these experiments would also yield evidence as to whether the smaller mammals could act as the reservoir of the pathogenic trypanosomes of man and domestic stock." OUTBREAKS OF CATTLE TRYPANOSOMIASIS APPARENTLY DUE TO BITING FLIES OTHER. THAN Glossina. — More than a decade ago, Bruce and his collaborators in Uganda observed seasonal outbreaks of trypano- somiasis among cattle, under circumstances pointing to the agency of Tabanidae (Horse-flies) as disseminators. The subject is obviously one of no little importance to stock-breeders and cattle-owning natives in areas not directly threatened by Tsetse-fly, and in recent years several writers — notably Mr. R. W. Jack in Southern Rhodesia — have devoted some attention to it. The more important statements and conclusions contained in these later papers are summarised in the following paragraphs. Jack (69), who gives details of an outbreak of trypanosomiasis in 1916, on a farm on the Sikombella River, Southern Rhodesia, considers the case in question as practically amounting to proof that trypano- somes of the type of T. pecorum can, under natural conditions, be transmitted from an infected to a healthy ox by some agency apart from a Tsetse-fly. In a subsequent paper (70) this author furnishes detailed accounts of a number of similar outbreaks, three being among cattle, and two among pigs on isolated farms in the same region, some miles away from a belt of G. morsitans. From these he concludes that, given the necessary conditions, transmission of the disease may take place, and has occurred more often than has generally been recognised, in the absence of Tsetse-fly. Such transmission, which is mechanical and direct — not cyclical, is considered to be most probably effected by Tabanidae, Stomoxys cakitrans, Lypewsia and mosquitos, and it is not necessary to infer that ability to disseminate the disease in this way is 107 confined to one species or family. These agents have not, however, led to the establishment of trypanosomiasis in areas away from the fly- belts, and in no instance so far recorded in Southern Rhodesia can Glossina be definitely disassociated from the inception of an outbreak. In the following year (1918) Jack (72) again expressed his conviction that the infection of cattle with trypanosomiasis in the absence of Tsetse-fly is of frequent occurrence in Rhodesia, and urged that the greatest care should be taken not to allow infected cattle to come into contact with healthy herds, especially during the spring and summsr months. In 1919 Jack stated (74) that, though direct experimental proof was still lacking, the evidence that in Southern Rhodesia Trypanosoma pecorum is commonly spread among domestic animals in the absence of Glossina had still further accumulated. Other testimony to the same effect, likewise relating to Rhodesia has been brought forward by Hornby (63), who mentions a case in which 1,000 village cattle, examined just at the commencement of the rains, showed less than a dozen cases of infection. Six months later, in the same village, 500 of the cattle previously examined were found to be either dead or very ill from trypanosomiasis. On neither occasion was a Tsetse-fly seen in the vicinity, although occasional individuals were observed in neighbouring areas. The author thinks this seasonal occurrence of the disease, which is coincident with that of biting flies other than Glossina, to be significant, when it is considered that out of 300 cattle kept for one month in a thin fly-belt, only 20 animals became infected. He adds, further, that on estates where the danger from mechanical infection was realised, and measures were taken to isolate all infected animals, no fresh cases occurred during the following rainy season, although severe outbreaks had previously been the rule. It .should be stated that, in an annotation to Hornby's paper, the Chief Veterinary Surgeon, while admitting that in some areas trypanosomiasis in domestic animals may be transmitted by flies other than Tsetse, re- marks that he believes such cases to be very rare in Southern Rhodesia. In 1917 Chambers (28), while recording no fresh evidence of his own, summarised the observations and experiments of a number of authors Avho have reported outbreaks of cattle trypanosomiasis in Northern Rhodesia and adjoining territories, in areas definitely known to be free from Tsetse-flies. Quite recently (November, 1921) Dry (450) has given details of an outbreak of trypanosomiasis in cattle on a farm in the Kericho district, Kenya Colony, under circumstances pointing to species of Stom'oxys as being the probable transmitters. The only other blood-sucking flies found on and in the vicinity of the farm, the altitude of which is about 6,300 ft., belonged to the genus Haematopota (Tabanidae). No Tsetse-flies were met with, and there were no records of any having been observed in the neighbourhood. It is clear that the whole question calls for careful experiment and investigation. CHAPTER XI THE EFFECT OF EXTERNAL FACTORS UPON TSETSE-FLIES. Great interest attaches to the study of the influence of external factors upon the duration of life of Tsetse-flies, since it may suggest means for the destruction of these insects. In particular, it has been observed that G. palpalis, more than any other insect, is easily killed 103 by heat, and cannot therefore remain long exposed to the direct rays of the sun. Tsetse-flies after being caught, if carried in pill-boxes with other insects, are almost invariably the first to succumb to the effects of heat ; the only method of keeping such flies alive consists in boring air-holes in the lid of the box and keeping the latter in a cool place. A series of very interesting experiments upon the influence of external factors — cold, heat and moisture — on Glossina palpalis was made in 1910 at Zungeru, Northern Nigeria, by Dr. J. W. Scott Macfie,. West African Medical Service (93). THE EFFECT OF A LOWERING OF THE TEMPERATURE. — " In order to determine the effect of lowering the temperature on G. palpalis/' writes Dr. Macfie, " a simple cold chamber was constructed by filling a large glass jar with ice and salt, or ice and water, in which a smaller glass jar fitted with a thermometer was completely immersed. Into the smaller jar the fly was introduced, together with some leafy twigs on which it might alight. In such a chamber it was possible to watch simultaneously the fluctuations of the temperature and the behaviour of the imprisoned fly." By this means it was found that the female G. palpalis is " able to withstand exposure for an hour and twenty minutes to a temperature ranging from 10°-7° C. (50°-44 • 5° F.) without harm." "This is a degree of cold to which -she would never be exposed naturally at Zungeru, where the minimum temperature recorded in 1910 was 13-8° C." " The immediate effect of a lowering of the temperature on G. palpalis is to cause a reduction in activity. Should the temperature fall to a point varying from 10° C. to 6-5° C. the fly becomes inert, and falls to the ground as if dead. Partial recovery was found to occur even after half an hour's exposure in this condition to a temperature ranging from 2-5°C. to 4° C. Recovery from less severe exposures was complete, and accompanied by violent buzzing." THE EFFECT OF RAISING THE TEMPERATURE. — " In considering the effect of a rise in temperature upon G. palpalis/' says Dr. Macfie, " account must be taken of the highest temperatures to which the fly may naturally be exposed ... It was also recognised at the outset that a distinction must be made between shade temperatures, and those attained by direct exposure to sunshine." In the experiments performed by Dr. Macfie, " tsetse-flies were exposed to the various temperatures in glass jars fitted with a thermo- meter. Twigs on which the flies might settle were always present. It may be stated at once that the flies appeared to be much more sus- ceptible to a raising of the temperature than they were found to be to a lowering." In one of the previous experiments " the fly withstood for some time a lowering of the temperature of 23-5°C. (viz. from 26° C. to 2-5° C.), whereas in experiment No. 8 a raising of the tem- perature only 10° C. (viz. from 34° C., the temperature of the external air at the time of the experiment, to 43°-44° C.) was more than could be endured." ^ Six experiments in all were made. In the first two the insects " were exposed in glass jars to the direct rays of the sun. Small twigs were included in the jars, but no water." As to the remaining tests, Dr. Macfie writes : — " The jars containing the flies in these experiments were protected from the direct rays of the sun by a covering of thick 109 brown paper. In some experiments water was present in the jars, in others it was absent." The conclusions deduced by the author from the whole series of experiments are stated as follows : — "1. Glossina palpalis is able to withstand a lowering of the tem- perature to 10°-7° C. " 2 Direct sunlight is rapidly fatal to G. palpalis. " 3. When shaded, exposure for one hour to a temperature of 40°-41°C. is fatal to the male G. palpalis. Higher temperatures are more rapidly fatal. "4. The presence of water enables G. palpalis to withstand for a longer period the action of direct sunlight, and perhaps of a shade temperature of 40°-41° C. "It is probable, therefore, that the diminution in the number of tsetse-flies during the dry season, and their limitation to the bush along the river banks, is due not so much to the low temperatures registered at nights during this season, nor to the high temperatures occurring during the days, as to the dryness of the atmosphere and the burnt-up condition of the country." MODE OF DEATH. — On this subject Dr. Macfie writes : — " A note should perhaps be included here on the manner of death of the flies used in the experiments. When an exposure was nearing a fatal termination the flies became less active, and would presently settle on a twig, or on the floor of the jars in which they were contained, in a lethargic condition, being with difficulty persuaded to move, even when touched with the end of a probe. A little later they would fall over on to their backs with their legs pointing up into the air, and their probosces projecting forwards. In this state they might remain for a considerable time. At first, when disturbed, they would be able to struggle to their feet, only to fall over on to their backs again in a few moments. Later, they were only able to respond by movements, more or less free, of the legs and proboscis. In this condition, which in the notes of my experiments I have described by the expression ' as if dead/ they remained until death took place, the movements gradually becoming less perceptible. In exposures to an increased temperature the flies sometimes spread out their wings convulsively j ust before death. After death the proboscis was generally depressed. THE INFLUENCE OF COLOURS. — The following statement is made by Dr. Macfie with reference to the influence of colour on Glossina palpalis. " Having previously ascertained that, in the case of the male G. palpalis confined in a glass jar without water at a temperature of 40°-41°C. (104°-105-8° F.), whereas exposure to direct sunlight was rapidly fatal, a similar result was only produced after an hour's exposure if the jar was shaded from the direct rays by a covering of thick brown paper, some experiments were carried out to determine whether under similar conditions different colours would produce different results. Male G. palpalis flies were therefore exposed to a temperature of 40°-41° C. in glass jars shaded respectively with red, green, blue and yellow paper covers. No water was present in the jars. In the case of the red, yellow and blue covers, after an hour's exposure the flies seemed to be none the worse, and remained very active throughout the experiments. In the case of green, however, after an hour's exposure the fly was showing evident signs of distress, and by continuing the experiment for a further 18 minutes he was killed .... Bearing 110 in mind the rapidly fatal result of exposure to direct sunlight, it is rather remarkable that in the semi-darkness of the jars shaded by thick brown paper a temperature of 40°-41°C. should have had a more serious effect on the tsetse-flies than in jars shaded by the various coloured papers. In one typical experiment with a fly in a jar shaded by thick brown paper signs of distress were noticed after 35 minutes, and the fly was dead after a further exposure of 25 minutes " As regards the colour preference exhibited by Glossina morsitans, Professor Newstead and Dr. J. B. Davey (107) have recorded the results of experiments carried out in Nyasaland in 191 1 . " During the month of October/' they write, " one of us made a series of observations on ten different days to test the colour preference of this tsetse-fly. For this purpose natives were provided with vests of the following colours : (1) white (clean) ; (2) bright canary-yellow ; (3) red ; (4) khaki ; (5) dark blue ; the bare back of a native being used as a control. The results were that khaki headed the list for attractiveness, and in a diminishing sequence red, blue, bare back of a native, and white ; no flies were seen upon the yellow garment. Dirty white vests were subsequently found to be much more attractive than clean ones. Furthermore, the buff-green canvas such as used for machilla coverings, bags, tents, etc., was found to be quite as attractive as the khaki." It was also noted by Dr. Warrington Yorke that khaki-coloured clothing, though excellent as a shooting kit, appears to attract G. morsitans in large numbers. INFLUENCE OF ODOURS. — Bagshawe and Austen (5) have pointed out that it would be highly interesting to discover whether, in their selection of a breeding place, Tsetse-flies are in any way influenced by odours, or whether, as Mr. LI. Lloyd believes in the case of G. morsitans, the principal factor in the choice is the existence of a relatively dark hiding-place, to which the female insect resorts during the period of gestation. The odours that attract insects are not necessarily per- ceptible to the human nostril ; if, therefore, the breeding places of G. palpalis and G. morsitans do not appear to us to give off a definite smell, it does not follow that such an odour cannot be detected by the pregnant Tsetse-fly. In India some years ago it was found by the late Mr. F. M. Howlett that Stomoxys calcitrans will oviposit upon cotton-wool impregnated with valerianic acid, one of the acids occurring in fermenting vegetable matter, in which the eggs of this fly are natur- ally deposited. It may be that on analysing the humus in Tsetse-fly breeding places some substance or essence having specially attractive properties would be found in it, and such a discovery might be turned to account in establishing artificial breeding places — a matter that we shall have to discuss later on (see p. 146). CHAPTER XII. MEANS OF LIMITING AND DESTROYING TSETSE-FLIES. PREDATORY ENEMIES OF TSETSE-FLIES. The search for means of limiting or destroying Tsetse-flies is a matter of exceptional importance, since the discovery of such means would at least enable us to check the progress of human and animal trypano- somiases. It would also render cattle-breeding possible in immense tracts of Africa (as in Northern Rhodesia, parts of Nyasaland and Ill elsewhere) which, though suitable for colonisation, are at present paralysed in the absence of animals for transport purposes, cultivation and the other necessities of agriculture and commerce. The abolition of Tsetse-flies would likewise be of especial benefit to the Katanga District of Belgian Congo, where, in consequence of the impossibility of keeping alive the oxen, horses or mules necessary for clearing, ploughing, weeding and harvesting, settlers have been obliged to import machines from Europe at great expense, and to employ skilled white labour to keep them going. As yet we know very little with regard to means of limiting and destroying Tsetse-flies, and it is to these points especially that the attention of investigators should be directed, for the discovery of an effective method of campaign against these terrible Diptera would render an immense service to all Tropical African colonies and protectorates. The results that have accrued from a precise knowledge of the bio- nomics of mosquitos belonging to the genera Anopheles and Stegomyia, the disseminators of malaria and yellow fever, should not be overlooked. Thus, to quote a single instance, the Americans, by taking effective measures for the destruction of these insects, and especially their aquatic larvae, have succeeded in ridding the Southern States of the Union and the neighbouring island of Cuba from a scourge that was retarding the economic development of some of the most fertile portions of the globe. Measures of this kind have also been of great assistance to the American authorities in the sanitation of the Panama Canal Zone. The very low death-rate among the personnel engaged upon the work of construction, and the fact that the gigantic feat represented by the completion of the Canal was performed in a minimum of time, were largely due to the progress made by medical entomology.1 In view of the fragmentary nature of the available data, all that can be done here is to indicate possible aids in conducting a campaign against Tsetse-flies. For the sake of convenience these will be dealt with under different headings. ENEMIES OF TSETSE-FLIES. — Under this heading, before con- sidering species directly hostile to Glossina, it may be of interest to say a word or two as to indirect enemies. These may be cither carnivorous mammals or biting flies belonging to other genera, both 1 The antimalarial measures taken by the Americans in the Panama Canal Zone, covering 10 miles in breadth and 45 miles in length, entailed the employ- ment of a numerous and highly trained staff. No fewer than a quarter of a million human beings, introduced into this zone during the five years covered by the American occupation, had to be protected against the terrible malarial fever, which had proved so formidable an obstacle to the first attempt at piercing the Isthmus made by the French. The measures adopted consisted essentially in : — (1) The destruction of Anopheles breeding places to a distance of 100 metres from habitations ; (2) the abolition, over the same area, of all cover for the adult mosquito ; (3) the screening of all openings in dwellings, so as to prevent the entrance of mosquitos ; (4) The application of crude oil, or another larvicide of special composition, to all breeding places which it was impossible to drain, in order to destroy the larvae. This procedure was based on the facts that Anopheline larvae are accustomed to live in fresh, clear water, containing an abundance of algae and other vegetation, and that the adult mosquito is not supposed to travel far, and needs herbage and bushes in which to shelter from the wind. 112 of which by attacking and preying upon " favoured hosts " of Tsetse- flies may, in certain localities and under certain conditions, have a distinct controlling effect on the range and increase of the latter. Thus Fiske (54), writing with reference to G. palpalis on islands in and also on the mainland of Lake Victoria, Uganda, states that the " Leopard is of very great importance in the bionomics of tsetse as a deprivative enemy, for it is a major factor in controlling the range and density of bush-buck, situtunga, bush-pig and perhaps other hosts of the fly." In a subsequent section of his paper the author furnishes detailed evidence in support of this statement, concluding by remarking with regard to the mainland : — " There would be inland extension of infestation to approximately five times its present depth at many points, and along extensive reaches of mainland shore, if it were not for the deprivative effect of the activities of leopards. They are certainly a more efficacious natural enemy of the tsetse than spiders, or than any other destructive enemy that has been identified/' Fiske also shows that onslaughts upon host animals by species of Tabanus, of whose bloodthirsty attacks the larger African antelopes are well-known to be intolerant, are inimical to G. palpalis. "It is not necessary," writes the author, " that the host animals shall be destroyed in order to injure the fly; but it is quite sufficient that they should move a little outside the range or reach of flies from either shelter or breeding places. " Therefore if the animal is induced by Tabanus, or any other biting fly, to move a little farther away from the specifically protected precincts, Glossina is forced to follow, and in following is exposed to all the risks, and enemies of a destructive nature, that make specific protection a requisite for its existence. The effect of the rival is precisely equivalent to either of the following : — (a) reduction in quantity of available food, (b) reduction in quantity or degree of protection, or (c) increase in number or destructiveness of enemies. " This is if the host is induced to leave the locality. It is the same if retaliatory activities on the part of the host induce the flies to desist from attack and to move on in search of another, more submissive animal. To do this the flies must leave protected precincts and undergo greater risk of being destroyed themselves, or of being unable to find protection for their young. " The activities of Tabanus have the direct effect of reducing the quantity of available food in protected precincts, which is the equiva- lent of either reducing the amount of degree of protection, or increasing the number or destructiveness of enemies." The species of Glossina are subject to the attacks of natural enemies, which serve to prevent excessive increase ; these enemies appear to be of various kinds and fairly numerous, and it is as well that we should be familiar with them. We have collected together certain statements relating thereto, scattered through papers by different authors, which have been published in recent years. The enemies of Tsetse-flies — like those of any other harmful species of animal — may be grouped into two main categories : predatory animals and parasitic organisms. The former do not confine their attacks to any particular species, but prey indifferently upon a very large number. Being themselves relatively far from numerous, all that they can do is to keep the increase of noxious animals within certain limits. Parasitic organisms, on the contrary, are for the most part highly specialised, having each a single 113 species as host. Their reproductive power, moreover, is very great, and when, as the result of exceptionally favourable circumstances, their host multiplies to an abnormal extent, they are capable of completely arresting its invasion. PREDATORY ENEMIES OF TSETSE-FLIES. — The predatory enemies of Tsetse-flies include vertebrates (mammals, birds and reptiles), arachnids, and certain insects . We will begin by taking the role of the vertebrates . Reference has already been made (see p. 90) to the experiments carried out by Mr. LI. Lloyd at Ngoa in order to determine the relation of G. morsitans to the smaller animals — mammals, birds, reptiles and amphibians In the course of these experiments it was found that, in the majority of instances, small animals of this kind in captivity, when attacked by captive Tsetse, succeeded in catching and devouring them, provided, of course, that the animals were awake. It is therefore important to observe and note carefully everything relating the pursuit of Tsetse-flies by small mammals, birds, reptiles and amphibians, whether insectivorous or omnivorous. Practically nothing is known at present as to the habits and food of tropical insectivorous bats, particularly those of Central Africa. This also applies to the part played by these animals in the destruction of noxious insects, particularly blood-sucking Diptera. We do not know whether they are capable of devouring Tsetse-flies, and since bats are nocturnal animals and fly fairly high in the air, while Tsetse-flies are seldom- on the wing by night and rest close to the ground, there are doubts on the subject. The role of bats in the destruction of mosquitos is by some writers regarded as important, and Dr. C. A. Q. Campbell, of San Antonio (Texas), who made a special study of bats as destroyers of these insects, of which, he states, they are the most inveterate enemies, concludes from analysis of the droppings that a single bat devours about 500 mosquitos per day.1 It would be well therefore to learn more of the natural history of Central African bats. What has been said with regard to bats may apply equally well to all the other insectivorous mammals, as well as to insectivorous birds, 1 At the beginning of 1911 Dr. C. A. Q. Campbell established a roosting- place or shelter for bats, constructed according to his own designs, in a place near San Antonio where mosquitos were abundant. In August the number of bats -entering the roost was estimated at several hundred, and the creatures took twenty minutes to enter; in 1912 the number of bats inhabiting the roost was .so great (500,000), that they were several hours in leaving it ; as a result, the local mosquito population soon showed a great reduction. The breeding of bats is therefore useful from a hygienic standpoint, and its commercial side is not to be despised, either, since the roosts are capable of furnishing a considerable quantity of bat-guano. ' It should, however, be stated that Dr. Campbell's belief in the value of bats as mosquito destroyers is not shared by Dr. L. O. Howard. Writing on " Mosquitoes and Bats " (Public Health Repts., Washing- ion, D.C., xxxv, no. 31, pp. 1789-1795, 30 July 1920), Dr. Howard cites evidence showing that mosquitos do not form a large proportion of the diet of bats, that only a very few species of bat are gregarious, and that no diminution in the numbers of mosquitos or in the amount of malaria was observed in certain localities in the United States where bats were swarming. In connection with mosquito reduction, Dr. Howard does not think that bat roosts are sufficiently promising to justify the expenditure of public money upon them. In all probability bats do not suffer from the bites of mosquitos, Tsetse-flies and other blood-sucking Diptera, owing to the peculiar conformation of the coat of hair with which their bodies are covered, and their special odour. They also appear to be little subject to disease, since, in the caves in which they live in enormous numbers, packed one against the other or even hanging in clusters, the collectors of guano very rarely find a dead bat. (5979) H 114 reptiles and amphibians.1 Above all, search should be made for vertebrates — if such there be — which devote themselves especially to the pursuit of Diptera, since such alone are really of interest from our present point of view. It is true that in Tropical Africa there are plenty of insectivorous animals that may occasionally devour Glossina, but only those belonging to the former category could be protected and bred with advantage. As regards birds, a certain amount of evidence was obtained by Prof. Newstead and Dr. J. B. Davey (107) during their investigations in Nyasaland in 1911. "Of the many birds which frequented the tsetse-fly area and its immediate neighbourhood," they write, " post- mortem examinations of 36 different species were made in order to ascertain to what extent, if any, they fed upon Glossina morsitans. Out of the total of 61 birds examined, only two were found to contain the remains of tsetse-flies .... It would seem, therefore, that tsetse-flies do not enter very largely into the dietary of insectivorous birds, but as such birds were everywhere in evidence, they must at least be considered of some potential importance, and may hereafter be found to feed upon tsetse more extensively than these records would lead us to believe. Those which had fed upon Glossina morsitans were the common black drongo of the country, Dicmrus afer, and the small black and yellow throated bee-eater, Melittophagus meridionalis. The other birds which are thought most likely to prey upon tsetses are the crow-shrikes (Prionops and Sigmodus), the grey babbler (Crateropus), and the roller (Coracias caudatus), all the more so seeing that such representatives of the avian fauna were always in evidence throughout the fly-infested areas." With regard to Glossina palpalis, Dr. G. D. Hale Carpenter (25), writing after upwards of two years' study of the bionomics of G. palpalis var. fuscipes, on Lake Victoria, Uganda, remarks : — " I do not think it likely that Glossina fuscipes is devoured by birds. It may well be that other species such as G. morsitans, frequenting more open country, are devoured by bee-eaters, drongos, &c. But the species which is the subject of this paper, frequenting as it does bush or forest with thick undergrowth, is out of the way of the larger insectivorous birds of powerful flight, which would alone be capable of catching such an active insect." An examination made by Dr. Carpenter " of the stomach contents of 64 insectivorous birds shot while feeding in, or on 1 Insectivorous fishes need not be considered, since the larvae of Tsetse-flies are terrestrial, but it is of interest to note that at the present time we are acquainted with a number of species of fish that feed upon mosquito larvae. The following, among others, may be mentioned : — (1) "Millions" (Lebistes reticulatus, Peters,, otherwise known as Girardinus poeciloides, G. guppyi and Poecilia reliculata), tiny, viviparous fishes, which are very prolific, and occur in fresh and brackish waters in Venezuela, Guiana, Trinidad, and the Windward Islands. They have been exported into various tropical countries for the purpose of reducing the numbers of mosquitos, but in some cases those introduced into Africa are said to have been devoured by frogs. Of three consignments despatched to ^ South Africa, however, one was successful, and has made excellent progress. (2) " Top- Minnows " (Gambusia affinis, Girard), likewise small, viviparous fishes, which inhabit the Mississippi, and fresh and brackish waters from Florida to Texas.. (3) Haplochilus pumilus, Blgr., an oviparous fish occurring in Lakes Tanganyika and Victoria. Although known to devour mosquito larvae, little has been recorded as to the natural history of this species. (4) Haplochilus grahami, Blgr., an active little fish, living in small pools a few yards in extent, in Southern Nigeria. In times of flood, fishes of this species rea'lily migrate from one pool to another. 115 the borders of, the fly areas of Damba and Bugalla Islands," showed no trace of Tsetse-fly remains in any instance. In order to determine the nature of the food of vertebrates, and especially of birds, investigators have a choice of several methods. Among these the following may be mentioned : — simple observation, in the open, of the food of the adults or their young ; experiments and observations in captivity ; microscopical examination of dejecta found in burrows, holes in treas, caves, nests, &c. ; and, finally, microscopical analysis of the stomach contents (remains of insects and vegetable matter) of the subjects of investigation after being shot.1 Attention should be paid, not only to vertebrates which prey upon perfect insects, but also to those which may feed upon pupae, and especially to scratching birds (Gallinaceae), such as guinea fowl, " pheasants," francolins, &c., which are frequently found in great abundance in the fly-belts. Mr. LI. Lloyd, who examined the crop contents of ten guinea fowl in the Luangwa Valley, Northern Rhodesia, writes (90) : — ' The crop of each was filled with vegetable matter, small bulbs, roots and flower buds. In three only were insects found, and these were in very small numbers. The following insects were observed in them : small Staphyline beetles, coleopterous larvae, and lepidopterous larvae. No pupae of any kind were recorded. From this it is clear that the guinea fowl is a vegetable feeder, eating insects when it finds them in its search for vegetable food. In the case of G. morsitans the majority of the pupae are placed in positions which would be inaccessible to these birds, though a few could be easily found. Doubtless a few of the pupae are eaten by guinea fowl, but the number must be so small that the bird cannot be considered to act as a control." Similar investigations into the crop contents of the different francolins (Francolinus and Pternistes) would be of interest.2 Predacious Arachnids. — A study of the bionomics and food of African arachnids would possibly lead to interesting discoveries. It appears that spiders do not prey upon any species in particular, but capture and suck the juices of all arthropods that they are able to overpower. In the case of certain species that have been observed to prey upon Tsetse-flies, practically all belong to the Family Salticidae (jumping spiders), the species of which do not spin webs, but lie in 1 Systematic investigations of this kind have already been carried out in different countries. In the United States, where the Bureau of Biological Survey of the Department of Agriculture has, for years past, been enquiring into the nature of the food of American birds, the collecting of stomachs and the examina- tion of their contents are conducted in the most systematic manner. Between 1885 and 1915 some 150 memoirs/dealing with the economic value of birds, were published by the Biological Survey. In England, too, Mr. Walter E. Collinge, a well-known biologist, has made a scientific study — chiefly by examination of the stomach contents — of the nature of the food of a large number of British birds. The results of this investigation have been published in various works and brochures, among which the volume entitled The Food of some British Wild Birds (1913) deserves special mention. Lastly, in India, the nature of the food of 1 10 Indian birds has been investigated by Messrs. C. W. Mason and H. M. Lefroy, and the results of the examination of 1,330 stomachs have been recorded in a report published in 1912 by the Indian Department of Agriculture, under the title The Food of Birds in India. 2 It seems to us that it would be well to continue these investigations ; the examination of a larger number of guinea-fowl crops, in other fly-belts, would perhaps yield different results. 116 wait for their prey and then leap upon it.1 Of course Tsetse-flies are not the only insects attacked by these spiders. Dr. J. J. Simpson (142.1.), in his account of entomological researches made by him in the Gambia in 1911, writes : — " As already noted, I observed in the Government steamer an Attid spider [Plexippus paykulli, And.] attack and kill numerous tsetse, and this species of spider seemed to show a distinct predilection for G. palpalis, although several other species of Diptera were almost equally abundant. It is very doubtful if this is of actual importance, but, as a fact in the bionomics of tsetse, may be recorded here." Carpenter (26), writing with reference to G. palpalis in islands in Lake Victoria, Uganda, thinks it probable that certain spiders have been responsible for considerable diminution in the numbers of flies, where conditions are apparently favourable for their increase. Swynnerton (145), in his paper relating to conditions in North Mossurise, Portuguese East Africa, says : — " Spiders of various bark- haunting species that hide in crannies and leap on their prey accounted for a number of the flies I had under observation in the net [vide supra, p. 71], and it is probable that these are the female tsetse's most important enemy. One tsetse (morsitans) was seen caught in the net of a web-making spider." Fiske (54) notes " the incredible number of spiders that occur locally at points along the mainland shore, or on small islets " in Lake Victoria, Uganda. This peculiarly competent investigator, in addition to recording the catches of various insects entangled in the snares of a gigantic web-spinning spider (Nephele pilipes), which is particularly abundant in certain of the localities mentioned, also furnishes com- parative statistics showing the average infestation by G. palpalis of islands with no spiders and with many spiders respectively. From the evidence thus obtained, Fiske thinks.it "extremely probable that where infestation by spider is excessively heavy it must have some effect in reducing local density of tsetse " ; he also considers that certain islands " noted .... as very lightly or not at all infested, may possibly owe freedom from fly to infestation by spiders, which was fairly heavy in each case, as well as to absence of breeding grounds." After writing of dragonflies (see p. 118), the probable effect of which in reducing the numbers of G. palpalis in various islands or districts is also acknowledged, Fiske adds : — " Apart from spiders and dragonflies, no destructive enemies of tsetse have been observed that appeared to be specifically responsible for appreciable reduction in density of Glossina palpalis." Predacious Insects. — The small number of insects recorded up to the present as preying upon Tsetse-flies belong to the Orders Diptera, Hymenoptera and Neuroptera. Writing with reference to G. morsitans in the Luangwa Valley, Northern Rhodesia, Mr. LI. Lloyd says (88) :— " An Asilid fly was taken on the wall of the laboratory devouring a tsetse-fly ; it had made a wound in the side of the thorax. The favourite food of these rapacious insects seems to be Tabanidae, especially the larger species. One was taken eating a dragonfly. 1 The Family Salticidae includes a large number of species, which are almost always of small size ; they are roving creatures, spinning no webs, but lying in wait and springing suddenly upon their prey. In the majority of cases the tropical species exhibit magnificent colouring ; the males are often more brightly coloured than the females, and the quaint behaviour of the former, in search of the opposite sex with a view to pairing, is frequently very curious. 117 " During the latter part of August the laboratory was raided during the night by hordes of a red driver-ant. These climbed the bottles in which the tsetse-flies were kept, bit through the gauze, and, entering, destroyed every fly. The bottles in which the pupae were kept were also entered, but the pupae were not damaged. A still active larva also, curiously enough, escaped injury. This incident is not recorded as suggesting that driver-ants are enemies of the flies in nature, but as a record that the larvae and pupae are not destroyed by these insects." As regards natural enemies of G. morsitans in Nyasaland, the following is extracted from a paper by Dr. W. A. Lamborn (82), published in 1915 : — " Considerable attention has been devoted to the fossorial wasps of the genus Bembex as being possibly natural enemies of morsitans, one species in particular being especially numerous in the fly area. A long series of specimens, each taken on the wing carrying its prey, shows that this species attacks in particular flies of the families Asilidae and Bombyliidae. Several other species taking various Muscidae and Orthoptera have also been secured. " In mid-December, however, a Bembex was actually seen to seize a tsetse-fly and to carry it away. The particular species had been repeatedly observed buzzing round as I walked, but little attention was paid to it at first, seeing that an inquisitive disposition seems to characterise many Sphegid wasps. Then one of the insects was seen to make a rush at a tsetse-fly on my leg, which it failed to secure. When a goat was led through the district three more were seen in the course of half an hour, after flying round and round the animal and hovering a few feet away, to rush at tsetses on it, two unsuccessfully, but the third was captured on the wing, having seized one of the flies. So far, all attempts to find out the other habits of the insect have failed, though the matter is receiving attention." It is worth noting that it was likewise stated by Dr. E. Roubaud (1190) in 1911 that in Dahomey G. longipalpis is attacked by large Bembex, which are found exclusively in the zone infested by this species. When introduced into a long glass tube with a live tsetse, one of these wasps quickly attacks the fly and paralyses it with its sting. On the Congo Dr. Roubaud was informed by certain Europeans that they had seen wasps, doubtless belonging to the genus Bembex, darting at tsetse-flies and carrying them off. Reference may also be made to remarks by Dr. G. D. Hale Carpenter (25) on insect enemies of G. palpalis var. fuscipes in Bugalla I., Lake Victoria, Uganda. Writing of the dragonfly known as Cacergates leucosticta, Dr. Carpenter says : — " This is an extremely abundant species, and if one walks along the shore one is at once surrounded by a crowd of these insects, some flying backwards in front of one as one walks, and all apparently on the look-out for Glossina, which they frequently attempt to catch ; and on several occasions they have been seen to do so. " On one occasion in the early morning a young hippo was on the fly beach, and I was able to approach closely enough to see with glasses numbers of Glossinae on its flanks and buzzing round it. Very many dragonflies were in attendance, and could be seen darting at the Glossinae, and I have no doubt that they account for a certain number of full-fed flies which are relatively heavy on the wing." As to robber flies (Asilidae) Dr. Carpenter says : — "The powerful and very active predaceous flies of the family Asilidae might be expected to be important enemies of Glossina, since their appetites are apparently insatiable and they prey upon insects of all kinds, even the most active 118 falling victims. But although I have spent many hours watching in suitable places at the edge of the forest or in open sunny spots in tha forest (localities frequented by Asilidae) where Glossina abounded, an Asilid has only once been seen preying upon Glossina." Carpenter's statements as to dragonflies on Lake Victoria are sup- ported by Fiske (54), who writes : — " A dragonfly of a rather small and undetermined species occurs commonly, but irregularly distributed, throughout the islands and riparian belt. There are other species, but this one is conspicuous for its numbers and for a well-marked habit of following moving animals and man and feeding off the flies which are attracted to them. Large numbers — in exceptional cases perhaps as many as 40 or 50 — of these dragonflies may follow a man along the open shore, and they have been observed to capture tsetse many times. " The species varies greatly in local density from one island or district to another. It was observed most abundantly along the western shore of Bukone Island in September 1914. A fairly complete survey of the island was made, and the infestation indicated (male density 8-6 ; female ratio 23-4 per cent.) was considerably lighter than conditions, apart from the very exceptional number of dragonflies, would have led one to expect. It seemed quite probable that the destructor was responsible for a considerably less heavy infestation than would otherwise have occurred. " There is this difference between spider and dragonfly — that sheltering vegetation is of little or no protective value against spiders, but of much protective value against dragonflies. They will not follow a moving animal into cover, and the flies are not likely to be captured unless they hunt in the open " (cf. also p. 117 above). Information on the subject of dragonflies as enemies of G. morsitans • in Nyasaland is contained in a paper by Dr. Lamborn (83), which appeared in December 1915. "A species of dragonfly (Orthetrum chrysostigma, Burrn.), the male dull blue in colour, the female greenish blue/' writes Lamborn, " has now been found to prey on morsitans. In February, in the Lingadzi district, I saw one of these dragonflies, which had been following and hovering round the party of six boys with me, suddenly swoop down and take a tsetse from the back of one of the boys who was stooping at a pool to drink, its movements being extremely rapid. It settled on the grass near by and commenced to devour its prey. Later on I saw a dragonfly of the same species take a morsitans which had rested on a blade of grass, though again I was unable to see whether the capture was effected on the wing or not ; and on the following day I saw another capture a tsetse off a boy. A fact of importance in all three cases was that each dragonfly accompanied our party for some little distance, obviously expecting to find its prey in our vicinity. " I have further studied these dragonflies in relation to morsitans in the Monkey Bay district. They have been observed repeatedly to flit round passers-by as if in search of prey, settling near by if unsuccessful. When a number of people are walking in Indian file, it is quite common for the insects to make a search round each, often following on a short distance behind, and I have without any difficulty caught a number of specimens with their prey, in most cases tsetses, taken in the neighbourhood of natives accompanying me." An annotation by Dr. G. A. K. Marshall states that " Dr. Lamborn has sent notes of 21 cases in which he saw these dragonflies take tsetses on 23rd and 24th April and 3rd to 12th May 1915." 119 " When a dragonfly has appeared at such times as there happened to be no tsetses about," continues Dr. Lamborn, " further light as to the object of its presence has been obtained from time to time by flicking off one's hand a tsetse with one wing clipped. So fleet are they on the wing that in a number of cases the tsetse was seized before reaching the ground. By this method it was ascertained that tsetse-flies-, whether half starved or replete with freshly ingested blood (either from a goat or fowl), were equally acceptable to the dragonfly, and after eating a replete tsetse the dragonfly has been found to have the blood smeared abundantly over its face and jaws. The insect is indefatigable in its work, and both sexes are to be found on the move in glades, in open grass country, and on rocky barren soil far from water, whether early or late, in sunshine or bad weather. The females feed and oviposit readily in captivity, the eggs, which are numerous, being arranged in irregular masses. " I have kept a look-out to determine whether any other species of dragonflies habitually prey on tsetses. Only one other instance occurred, and in this case the insect (Crocothemis erythraea, Brulle) handled the tsetse-fly so clumsily, gradually slipping down and right round the grass stem to which it was clinging, in a vain endeavour to hold its prey and support itself at the same time, as to convince me that it was quite a novice with tsetses." According to Simpson (144), in the Gold Coast, although some spiders and dragonflies have been seen devouring Tsetse, " by far the most important insect enemies are Asilid flies and wasps of the genus Bembex," the former preying chiefly upon G. tachinoides, the latter on G. morsitans form submorsitans. The Asilids in most cases were observed to seize their victims on the wing. " The Bembex does not alight, but darts quickly down and either captures the tsetse on the wing or immediately it "has alighted/' To some extent the activities of Bembex may neutralise those of Asilids, for Barker (8), after studying a series of predacious insects and their prey in the Durban Museum, records a species of Bembex as preying, not only upon G. morsitans, but also upon seven species of Asilids ! Swynnerton (145), with reference to North Mossurise, Portuguese East Africa, writes : — " Dragonflies on several occasions in fly country inspected the backs and necks of my companions. A wasp once did the same. On one occasion, travelling from the Buzi to the Mtshanedzi, five brevipalpis were under the cattle, four being under the leading ox. We ran into some dragonflies, which at once began passing to and fro immediately under the belly of that ox. I saw no captures, but these four brevipalpis had disappeared when the dragonflies left us . ^ . Dragonflies were exceptionally and extraordinarily numerous in 1918, but the tsetse population remained large. I do not remember seeing any Asilidae." CHAPTER XIII. MEANS OF LIMITING AND DESTROYING TSETSE-FLIES — continued. PARASITES OF TSETSE-FLIES. The parasitic enemies of Tsetse-flies may be divided into two cate- gories— parasites of the pupae, and external and internal parasites of the adult flies. PARASITES OF TSETSE PUPAE.— In the following pages we give a statement of present knowledge as to parasites of Tsetse pupae, all of 120 which belong to the Orders of insects known as Hymenoptera and Diptera. Early in 1914, the Imperial Bureau of Entomology received from Mr. LI. Lloyd a small Dipteron belonging to the Family Bombyliidae, which had been bred by the sender at Ngoa, Northern Rhodesia, from a puparium of G. morsitans. The parasite, which measured 5-75 mm. in length, and emerged from the puparium instead of a Tsetse-fly, was found to represent a new species, which was described by Austen (4) under the name Villa lloydi (see Figs. 9 and 10.) Fig. 9.— Villa lloydi, Austen, $, x 9 : a small Dipteron of the Family Bombyliidae, parasitic in the larval stage in pupae of Glossina morsitans, in Northern Rhodesia. (After Austen, 4.) Fig. 10. — Early stages of Villa lloydi, Austen, a Dipterous parasite of Glossina morsitans pupae in Northern Rhodesia, a larva • b front view of head of larva; c, pupa. (After Lloyd, 91.) A second Dipterous parasite ( Thyridanthrax abruptus, Lw.), belonging to the same Family, was subsequently bred from G. morsitans puparia in Nyasaland by Dr. Lamborn. and was afterwards obtained in Southern 121 and Northern Rhodesia respectively by Messrs. R. W. Jack and LI. Lloyd (see below, p. 126). As regards Hymenoptera, at Bukama (Belgian Congo), in 1911, Dr. Bequaert, of the Mission Scientifique du Katanga, met with a small Hymenopteron which is a specific parasite of Glossina palpalis pupae. The insect, which is a member of the genus Conostigmus, Dahlbom (Family Proctotrupidae), received from its discoverer (Rev. Zool. Afric., Vol. ii., Pt. 2, p. 255, 1913), the name Conostigmus rodhaini : the female measures 18 to 20 mm. in length and the male 18 mm. On 24th June 1911, from a single pupa of Glossina palpalis, collected in a breeding place on an island in the Lualaba, near Bukama, Dr. Bequaert obtained six specimens (five females and one male) of this parasite, which must indeed be very rare, seeing that out of a large number of pupae found at the same spot no other was parasitised. Some interesting particulars with reference to parasites of Tsetse- fly pupae are to be found in the Bulletin of Entomological Research, Vol. V, Part 4, and Vol. VI, Part 1 (March and June 1915). Fig. 11. — Larva of Mutilla glossinae, Turner, a Hymenopterous parasite of Glcssina morsitans pupae, in Northern Rhodesia and Nyasaland. A, lateral view, X 10; B, front view of head, x 60; (a) palpi, (b) antennae, (c) mandibles, (d) internal chitinous supports. (After Lloyd, 91.) The following is extracted from a summary of a report by the late Mr. R. A. F. Eminson (50) upon work done by him from May to July 1914, in studying the bionomics of Glossina morsitans in Northern Rhodesia. " In a batch of 258 collected Glossina pupae, from one puparium there emerged, on the 21st June 1914, a wingless parasitic wasp of the genus Mutilla. It was observed that the wasp on emerging had broken open the puparium in precisely the same way as would the fly itself, so that a mere external examination of the case would not reveal the fact that the fly pupa had been parasitised. On investi- gating a number of pupa-cases collected in the field, four were found to contain remains of parasitic pupae which were probably referable to the same species of Mutilla. On 21st August, 84 of the tsetse pupae were still unhatched, and seven of them were therefore opened. 122 Two of these contained larvae of the parasite, and in the other five the fly pupae had died from other causes. From the 77 remaining puparia two males and eight females of the Mutilla were bred out between the 2nd and 6th September. " This Mutilla (see Fig. 11, p. 121), which measures 5-5 mm. in length, was described by Mr. R. E. Turner (148) as a new species under the name Mutilla glossinae. In a paper (160) entitled " Chalcidoidea bred from Glossina morsitans in Northern Rhodesia/' Dr. James Waterston, now of the Entomo- logical Department, British Museum (Natural History), dealt with other Hymenopterous pupal parasites of the species referred to. " In connection with investigations into the life-history, etc., of Glossina morsitans in Northern Rhodesia," writes the author, " special efforts have recently been made to secure parasites of the fly. As a result, a considerable number of Chalcidoids have been bred from puparia collected between August and December of last year at Kashitu (LI. Lloyd) and Mwengwa (R. A. F. Eminson). These interesting Hymenoptera have now been forwarded to the Imperial Bureau of Entomology by Mr. Lloyd, Chief Entomologist in Northern Rhodesia, with the parasitised puparia and some relevant notes .... The collection contains three species, representing as many widely separated groups in the super-family Chalcidoidea." The species in question were as follows : — Stomatoceras micans, sp. nov. (see Fig. 12, p. 123), Family Chalci- didae. Length of female, about 4 mm. ; wing expanse, 6 mm. " Mr. Eminson," writes Waters ton, " found the puparium from which the Stomatoceras was bred on 6th October 1914 . . . . No other specimen nor certain .evidence of the recurrence of this species was secured. The Stomatoceras emerged from the puparium by a large irregular hole." Anastatus viridiceps, sp. nov. (see Fig. 13, p. 124), Family Encyrtidae. Length of male, about 1 • 8 mm. ; wing expanse, 3-3 J mm. ; length of female, about 2| mm. ; wing expanse, over 4J mm. The author states : — " The Glossina puparium was taken on 11. xi. 14, and the parasites emerged four days later .... In his covering letter (30.xii.14) Mr. Lloyd remarks that ' in all, 9 £ $ and 6 £<$ emerged through a small round hole on the dorsal surface, a little in front of the anal cap ; . . . . copulation occurred shortly after emergence, and the males lived only a day or so.' " Syntomosphyrum glossinae, Wtst. (see Fig. 14, p. 125), Family Eulophidae. Length of male, 1-25 mm. ; wing expanse, 2-3 mm. ; length of female, 1 • 4-1 • 6 mm. ; wing expanse, 2 • 5-2 • 6 mm. Accord- ing to Mr. Eminson, " about thirty specimens " of this species, " which had emerged from a single pupa of G. morsitans," were " found on August 21st 1914." The collector adds :— " As will be seen, the parasites emerged through a minute hole in the pupa-case. Since that date three specimens of pupa-cases similarly attacked have been found. The pupa was collected on June 1st within a mile of the Kafue R., near Mwengwa." The following very interesting statements, with reference to parasites of G. morsitans pupae in Nyasaland, are taken from a paper (83) by Dr. W. A. Lamborn, published at the end of 1915. Mutilla. — Of Mutilla glossinae, Turner, a series of 6 males and 10 females was bred by the author from G. morsitans pupae collected in the neighbourhood of Monkey Bay. 123 " The appearance presented by a pupa-case from which a Mutillid parasite has emerged," says Dr. Lamborn, " seems to me characteristic and unmistakable, though a contrary opinion has recently been expressed (Eminson, quoted in Bull. Ent. Res., v, p. 382. — Vide supra, p. 121). On taking such a case in the fingers there is, owing to the presence of the Mutilla cocoon within, a sense of greater solidity than Fig. 12. — Stomatocerasmicans, Wtst., (j>, a small Hymsnopteron of the Family Chalcididae, parasitic in the larval stage in pupae of Glosswa morsitans in Northern Rhodesia and Nyasaland. (After Waterston, 160.) when a fly has emerged, and it is possible by gentle pressure to crumble away the wall of the puparium so as to obtain the cocoon, a light chestnut-coloured structure composed of several layers of a very tough silky-looking material. The orifice of exit is usually much smaller and has a serrate edge, instead of the larger clean-cut fracture produced by the emergence of the fly, owing to the parasite having nibbled out a 124 circular cap, and one can always see the walls of the cocoon within. In the course of time the cocoon tends to shrink, the result being that it draws in with it the margin of the hole of exit in the pupa-case, so that this is no longer circular but somewhat oval, a condition never seen in the case of a normal puparium. Fig. 13. — Anastatus viridiceps, Wtst., , a minute Hymenopteron of the Family Encyrtidae, parasitic in the larval stage in pupae of Glossina morsitans in Northern Rhodesia. (After Waterston, 160.) " There has been no difficulty at all in dealing with the Mutillids in captivity, for all, except two, one of which was accidentally drowned, are still alive now, one or two of the earliest specimens being several weeks old. The original male placed in a box with the female shortly after the emergence of the latter manifested the greatest excitement, running about with its antennae on the ground on the track of the female, which it overtook after a considerable chase and immediately seized, pairing taking place almost at once. It is noteworthy that in 125 the course of the chase it ran repeatedly very near to the female, but being off the fresh track did not detect it, the sense of sight being evidently of very little help to it in the matter. " Various pairs have been kept in captivity in jars containing a number of tsetse pupae buried in earth, in the hope that the females would sooner or later parasitise them. This expectation seems likely to be fulfilled, as on 30th May the first female, which emerged so long ago as the 3rd of that month, was actually witnessed ovipositing in one of the pupae .... At 5.45 p.m., on coming into camp from a day's trek, I removed from inside a box, which had been closed all day, a jar containing the Mutilla female, No. 1, and a number of tsetse pupae, mostly buried in earth, though one or two were on the surface. On the top of one of the latter the Mutilla was seen. It remained still a few seconds, then precipitately vanished beneath a lump of earth, as is their habit when alarmed. In a few minutes it came out into the open again cautiously, and after examining with its Fig. 14. — Syntomosphyrum glossinae, Wtst., Q, a minute Hymenopteron of the Family Eulophidae, a hypefparasite of the larva of Mutilla glossinae, therefore harmful. (After Waterston, 159.) antennae some pupae near the one on which I had first seen it, started, with its head facing the tail end of the pupa, to whittle away with its jaws at a point midway between the two poles with such energy as to rock it. Its antennae were crossed and below its head. After five minutes' work in the horizontal position it gradually raised itself vertically, with its head down on the pupa, supporting itself against the side of the jar, so that a full view of its movements with a lens could readily be obtained. Extremely fine movements of the jaws in and out took place with great rapidity, and with such delicacy that unless one's attention had been attracted by corresponding movements 126 of the labial palpi the operations in progress would have been undetected. From time to time a sharp turn of the head through? a quarter of a circle on each side also went on. " After half an hour's work the Mutilla retired a short way from the pupa and, resting on its side, cleaned its antennae and rubbed its legs together, but then manifested some uneasiness at the light, for at this point I had to watch it by lamplight, and concealed itself. But when the light was very much shaded it returned to the pupa, put its head to the site of its previous operations, where with a lens I could see a< small breach of surface, and then gradually raising the hinder part of! its body so as to rest again on its head, recommenced work, its antennae being this time in front of its head, but resting on the pupa. Occasional : movements of anteflexion of the abdomen then took place, as if the insect were preparing for oviposition, and finally, after at least an hour's work, the insect resumed the horizontal attitude on the pupa, , and having examined its work advanced so that its hinder end came somewhere over the breach. , It then moved to and fro, feeling for the exact spot, and then remained still, doubtless in the act of oviposition, running away a few seconds later. " Though the female worked so indefatigably, so tiny a puncture was produced as to be barely visible except in a certain light to the unaided eye, and it appeared moreover as if definitely sealed by the insect with some secretion after oviposition. This may possibly be a very important part of its final operations, as on several occasions I have found an exuberant fungus growth, brown in colour, sprouting at little fissures accidentally produced in tsetse pupae. By and by I hope to be able to study this further. " I should perhaps add that the night when the Mutilla was working was bitterly cold, a rather surprising fact, seeing that as a rule these insects are so lethargic except on bright sunny days. " I have now examined carefully all the pupae to. which this female has access, and seven out of the twenty show this evidence of attack. There is therefore every reason to believe that the Mutilla can be raised experimentally in some numbers and without any great difficulty in the laboratory. Their hardiness, their activity in finding food for themselves and their longevity make them singularly easy to deal with, so that I expect shortly to be able to submit a further report on the subject. " In regard to their general habits, they do not as a rule become active till the middle of the afternoon, remaining until then hidden beneath objects on the top of the ground, or buried beneath the superficial layers of the soil. The female is an adept at burrowing, and the male in pursuit of her does not hesitate to force his way into the soft earth. Both sexes run with extreme activity, the male being unusually loth to take flight for a winged insect. The females in captivity soon lose the quality of shyness, which is so marked during the first day or two after emergence, and will then run about uncon- cernedly, even though one is moving objects in the jar, almost as if they had learnt that they were unlikely to be molested." Other Parasites. — "A species of Bombyliid fly (Thyridanthrax abruptus, Lw.), a striking-looking insect, quite distinct from the Villa lloydi found also to be parasitic on morsitans pupae in Rhodesia, has been bred out from morsitans pupae here, three specimens having been obtained .... This Bombyliid is by no means uncommon in the fly area, especially at Lingadzi, so that I have been familiar with its appearance almost from the very first. I am endeavouring to proceed 127 further with an investigation into its habits." It may be added that a specimen of the Bombyliid in question was bred from a G. morsilans pupa in Southern Rhodesia, in November 1912 by Mr. R. W. Jack, and that the species is also known to occur in Kenya Colony and in Northern Nigeria. With reference to Hymenopterous parasites other than Mutilla, Dr. Lamborn writes : — " Within the last few days two nice-looking Chalcids [Stomatoceras micans, Wtrst.], large insects with black markings on the wings, have emerged from tsetse pupae through an opening situated in each case close to the tubercles at the posterior end of the puparium. These also are thriving in captivity, but unfortunately appear both to be of the same sex. " In connexion with the question of Chalcids, I found in April in cattle droppings a large number of pupae of a little Muscid [Musca sp. nov. ?] common in houses in this country, and having bred out numerous small Chalcids [Spalangia sp.] from them, endeavoured to parasitise tsetse pupae with them, but without result, so far as I have yet been able to judge. " A fourth species of parasite [Eupelmimis tarsatus, Wtrst.], an apterous insect, has also been bred out from a tsetse pupa." Proportion of Pupae Parasitised. — " Examination of the living pupae recently collected, 1,143 in all, affords evidence as to the probable parasitism of a small number, small marks like punctures being visible with a high power lens, and a study of the empty cases found at the same time, numbering 9,762, has afforded some statistics thereon, for owing to their chitinous nature and the sheltered positions in which they are placed it appears to be some time before the 'empty cases suffer by exposure. By far the greatest number of cases (no less than 8,543) had given exit to perfect flies ; for though there is no means of differentiating between normal pupae and those parasitised by Bomby- liids, the latter seem to be in so small a minority as to be almost negligible from the statistical point of view ; 351, or 3 J per cent., showed that they had been parasitised by Mutillids, the features characteristic of which have already been described, the parasite having in practically every case escaped at the cephalic end ; 107, roughly 1 per cent., were intact, except for a pin-point hole at one side, produced in all probability by the escape of some tiny parasite, possibly a Chalcid parasitic on the Mutillid, for a few of these which were broken open showed the remnant of the cocoon of the latter1 ; 264, roughly 2 per cent., show larger rounded holes produced by the escape of a parasite ; some at the posterior end, others to one side, the insect being in all probability one of the large Chalcids Some twenty, a few of which had contained Mutillids, showed evidence that the pupal contents had been eaten from outside ; and 477 were so damaged that it was impossible to form any opinion as to their history. " No parasites whatever have as yet been bred out from pupae obtained in the proclaimed area, all that have been obtained hitherto having emerged from pupae found in the vicinity of Monkey Bay, a fact which may explain the numerical superiority of morsitans in the former region." 1 In a note to this statement Dr. G. A. K. Marshall remarks : — " A large number of these Chalcids were bred subsequently by Dr. Lamborn from a tsetse pupa, and proved to be Syntomosphyrum glossinae, Wtst. There seems little doubt now that this species is harmful, being a hyperparasite of Mutilla glossinae."' 128 In a subsequent paper (84) the same author gives further details of his observations upon Hymenopterous parasites of G. morsitans in Nyasaland. Mutilla glossinae, Turn., was reared in large numbers from pupae of the latter species obtained near Monkey Bay. " The act of parasitism/' writes Dr. Lamborn, " has already been described in a previous report [vide supra, p. 125], and to my account I should add that in most cases, when the pupa attacked has been on the surface of the ground, the female is at particular pains to cover it with soil subsequent to oviposition, collecting earth together with its fore-limbs from various points, after the manner of a Bembex closing its burrow, and shovelling it back over the pupa with its hind limbs. Fig. 15. — Mutilla [benefactrix, Q, parasitic in the larval stage in pupae of Glossina morsitans, in Nyasaland. (After Turner1, 149.) " The site for puncture is invariably about midway between the two poles of the pupa. The delicacy with which the operation is effected seems to vary considerably, so that though in some cases the site is readily found on examination with a lens and may be marked with a little shining sticky patch to which a few grains of earth sometimes adhere, in others no sign of it can be detected." Dr. Lamborn states that he has obtained " indisputable evidence " that M. glossinae is a primary parasite, depositing eggs in Glossina pupae at varying stages of development, especially, he believes, the later ones. No indications that G. brevipalpis is attacked by this species of Mutilla under natural conditions were observed. The dura- tion of the life-cycle within the Tsetse-fly pupa varies very considerably, and may be influenced by climatic changes. In the case of one 129 individual reared under artificial conditions from a G. brevipalpis pupa, development occupied nearly 12J weeks, while in that of other specimens the developmental period was extended to 17 weeks. The numbers of M. glossinae and other parasites emerging from pupae increase rapidly in the dry season ; out of 762 pupae obtained from natural breeding places between 22nd August and 16th October, 97 flies and -131 Mutillids emerged. From 368 pupae collected within the pro- claimed area, however, only two parasites, a Mutillid and a Bombyliid, were reared. A new species of Mutillid parasite, M . benefactrix, Turn., was obtained from pupae collected at Monkey Bay. This species exhibited greater activity than M. glossinae, and differed in the method of pairing. Fig. 16. — Mutilla benefactrix, Turner, 3. (After Turner, 149.) xperimental evidence showed that the Chalcid, Syntomosphyrum glossinae, Wtst., is hyperparasitic on M. glossinae, which may be attacked in both larval and pupal stages. About 3 per cent, of the pupa cases of M. glossinae examined between 7th April and 2nd June showed evidence of parasitism, while between 22nd August and 16th October less than 1 per cent, were parasitised. Another parasitic Hymenopteron, Eupelminus tarsatus, Wtst., also regarded by Lamborn as a hyperparasite of M. glossinae, though there are grounds for con- , sidering it to be more probably a primary parasite, was obtained from (5979) I 130 pupae of G. morsitans collected at Monkey Bay. In captivity, pairing took place soon after emergence, and oviposition began about two hours later in pupae believed to be parasitised by M. glossinae. Nine females oviposited in pupae of G. morsitans, the cocoon of M. glossinae being present in every case except one ; 22 males and 58 females emerged. In those instances in which definite data were obtained, the period between oviposition and emergence varied from 28 to 32 days, the greatest number of parasites obtained from a single puparium being nine. In several cases Mutillids emerged from pupae into which E. tarsatus had inserted its ovipositor, and on two occasions Tsetse- flies were reared from such pupae. A few large Chalcids, Stomatoceras spp., were obtained from pupae found in the neighbourhood of Monkey Bay between 7th April and 2nd June, but did not emerge from pupae in the proclaimed area. These Chalcids may be hyperparasites of M. glossinae, since specimens of the latter were present in some cases, but more probably are primary parasites of Glossina, which develop more rapidly than M. glossinae and may incidentally attack the larva of the latter species. Six species of insect parasites of the pupae of G. morsitans were found by Lloyd (91) in Northern Rhodesia in 1915. According to this author, M. glossinae is generally distributed in G. morsitans areas in Northern Rhodesia, and the duration of the pupal period was observed to be 45 days. Adults in captivity fed readily on jam ; the females lived about three weeks, and the males from 10 to 14 days. Attempts to breed them were, however, unsuccessful. The percentage of parasitism by M. glossinae at Chutika (Hargreaves), in the Luangwa Valley, was on an average 7 per cent. ; at Mwengwa, on the Kafue River, in 1914, Eminson " found that about 10 per cent, of the 350 pupae he had collected were destroyed by this insect " ; at Nawalia it was observed by Lloyd that the parasitism reached 13 per cent., a quarter of the parasites being males. Parasites emerged between 15th August and 6th October from pupae collected during July ; emergence from pupae collected on 21st and 23rd August took place between 25th September and 26th October. Attempts to breed. Anastatus viridiceps, Wtst., in captivity failed. Stomatoceras micans, Wtst, discovered by Eminson at Mwengwa, was met with by Lloyd at Ngoa, on the plateau near Mpika. In one case an almost fully developed Tsetse emerged from a puparium parasitised by this species. The hyperparasite Syntomosphyrum glossinae, Wtst., obtained by Eminson at Mwengwa, was found by Lloyd at Ngoa and Kashitu, and probably also occurs in the Luangwa Valley. The insect, about 25 of which usually emerged from one puparium, was taken from September to November. The Dipterous parasites of G. morsitans pupae met with were the two Bombyliids, Villa lloydi, Austen, and (probably) Thyridanthrax abruptus, Lw., which were obtained at Ngoa and Chutika respectively. Puparia parasitised by the former were collected in September, and flies emerged during the same month ; two pupae attacked by the second species were taken in July, and the adults appeared in August. Up to the beginning of 1918 no parasites of Tsetse-flies had been recorded from West Africa, but in February of that year Simpson (144) reported that, among a large number of pupae of G. tachinoides collected by him in the Gold Coast, between September and December, 1915, he had found " a small proportion " parasitised by Chalcis amenocles, Walk. Simpson also stated that he had bred the same 131 species, in addition to two other parasitic Hymenoptera — Dirhinus inflexus, Wtst., and Coelalysia glossinophaga, Turn. — from pupae of G. morsitans form submorsitans obtained by him at Larabanga in the same Colony. Experiment showed that C. amenocles parasitised pupae of Sarcophaga much more readily than those of Glossina. " As species of Sarcopliaga are everywhere abundant around villages," writes Simpson, " and as they breed prolifically, it would be possible in a very short time to obtain very large numbers of these parasites for dissemination." Within the last few years several Hymenopterous parasites of the pupae of G. morsitans, in addition to those already mentioned, have been described by Messrs. Turner and Waterston. The following is a complete list of the parasites of Glossina pupae known in 1922, omitting Syntomosphyrum glossinae, Wtst., which, as already stated, has been proved to be a hyperparasite of Mutilla glossinae, and therefore harmful instead of beneficial. Parasite. Host. Locality. Collector. Hymenoptera. MUTILLIDAE Mufilla glossinae, Turn. . . G. morsitans N. Rhodesia R. A. F. Eminson ; H. C. Nyasaland Dollman ; LI. Lloyd. Dr. W. A. Lamborn. Mutillz benefactrix, Turn. , , auxiliaris, , , G. morsitans Portuguese E. Africa C. F. M. Swynnerton. PROCTOTRUPIDAE — Conostigmus rodhaini, Beq. G. pzlpalis .. Belgian Congo (Ka- J. Bequaert. tanga) BRACONIDAE — Coelalvsia gl3ssinophaga, G. morsitans form Gold Coast Dr. J. J. Simpson. Turn. submorsitans CHALCIDIDAE — Stomatoceras miains, Wtst. G. morsitans . . N. Rhodesia R. A F. Eminson ; LI. Lloyd. Nyasaland Dr. W. A. Lamborn. Stomatoceras (Centrochal- n n ti cis) exaratum, Wtst. Haltichella edax, Wtst. . . Chalcis amenocles, Walk. G. tachinoides, and G. Gold' Coast Dr. J. J. Simpson. morsitans form sub- * morsitans Dirhinus inflexus, Wtst. G. morsitans form (> tl submorsitans ENCYRTIDAE — Anastatus viridiceps, Wtst. G. morsitans N. Rhodesia LI. Lloyd. Eupelminus tarsatus, Wtst. Nyasaland Dr. W. A. Lamborn. BETHYLIDAE — Prolaelius glossinae, Turn. t> ( and Wtst. Dipteta. BOMBYLIIDAE Thyridanthrax abruptus. it S. Rhodesia R. W. Jack. Lw. N. ,, LI. Lloyd. Nyasaland Dr. W. A. Lamborn. Villa lloydi, Austen . . " N. Rhodesia LI. Lloyd. In view of the importance of the subject, investigators would do well to turn their attention to the parasites of Glossina pupae, and the breeding out of the pupae of the different species of Tsetse-flies would doubtless lead to interesting discoveries. For the propagation of these parasites it would perhaps be possible later on to employ the methods successfully adopted by the Americans, for the multiplication 132 or introduction into a given region of insect enemies of certain noxious insects.1 With the object of further emphasising this matter, we feel that we cannot do better than to reproduce some valuable and highly suggestive remarks extracted from a recent paper by Dr. W. A. Lamborn (85), who has done so much in Nyasaland to increase our knowledge of the parasites of Glossina pupae. Under the heading Suggestions in regard to Tsetse Control, Dr. Lamborn writes : — " One of the most pressing entomological problems at the present day in the British Empire is that of the control of the various species of tsetse-flies. As shown by a number of workers, the parasites destructive to their puparia are not few, at all events in the case of G. morsitans, and in some areas these parasites are known to exert a considerable influence in reducing the numbers of the fly. But, generally speaking, no very material reduction, from the point of view of man, would seem to be effected by these agencies, though at this early stage of research in Fig. 17. — Mutilla auxiliaris, Turner, $ <$» parasitic in the larval stage in pupae of Glossina morsitans in Portuguese East Afiica. (After Turner, 151.) regard to the fly it may perhaps be premature to make any unqualified statement as to their control value. Instances have been recorded where the tsetse-fly in a district has unaccountably diminished without there being any marked diminution in the numbers of the game animals therein. Major E. E. Austen in his ' Handbook of the Tsetse-flies ' (1911, p. 65) mentions the disappearance of G. morsitans from the Victoria Falls, where at one time it abounded. The late Captain F. C. Selous, in his book ' African Nature Notes/ speaks of the disappear- ance of the flies at the same time with buffalos from the valley of the central Limpopo and its tributaries, where other game — kudu, zebra, wildebeeste, hartebeeste, impala and bushbuck — continued to exist in considerable numbers, and he suggests that the flies died out because they were unable to maintain themselves on game other than buffalos. 1 In the United States the process of introducing parasites of destructive insects has been carried on for years past by the Bureau of Entomology of the Department of Agriculture, under the direction of the well-known entomologist, Dr. L. O. Howard, and the results obtained are most interesting. In particular, Dr. Howard has succeeded in establishing in the country a certain number of parasites of the eggs, larvae and chrysalises of the Gipsy Moth and the Brown-tail Moth, as also a predacious beetle, Calosoma sycophanta, L. 133 More recent study of the flies has shown that this is not the case, and the writer suggests that a more reasonable explanation is to be found in the local extermination of the flies by their parasites, a process, as he believes, now steadily proceeding in the area near Mzeze, Nyasaland. The question therefore naturally arises — how far is it possible to increase their influence either by the introduction of new parasites or by breeding on a larger scale those already known to science ? " The genus Glossina being now limited to the Ethiopian region, it is doubtful how far the parasites of other Diptera, brought in from other lands, would seek out and destroy its puparia. It is to be apprehended that those obtained from the puparia of other Muscids, breeding under more or less similar conditions, might do so. A more hopeful line of action could possibly be found in the search for other species naturally parasitic on Glossina but having a different geo- graphical distribution. For instance a Bombyliid fly, Villa lloydi, Austen, parasitic on G. morsitans, has so far only been discovered in Rhodesia, and a second species of parasitic Mutillid, M. benefactrix, Turner, in Nyasaland only. The parasites of the fly on the West Coast are probably different from those in the east and south, the insect fauna, generally speaking, being largely distinct. The various species of tsetse-flies may each have different parasites yet to be discovered, which might be interchangeable. a Fig. 18. — Puparia of Glossina morsitans, showing holes of emergence of: a, the hyperparasite Syntomosphyrum glossinae, Wtst. ; b, tho parasite. Stomatoceras micans,Wtst. (Alter Waters ton, 161). "It is probable, and some of the evidence on this point is very definite, that in the case of rapid extension of a fly area, parasites less endowed than their hosts with the power of ranging far and wide- strength on the wing having doubtless played a considerable part in the present-day success of the tsetse as an insect — have lagged more or less behind in the foci first inhabited by the flies. And this is particularly likely to be the case with Mutillids, the females of which are wingless. One of the present suggestions in connection with fly control is to endeavour to enhance the value of the parasites already known to science, the Chalcids in particular. They could unquestion- ably be bred readily enough in the laboratory on the puparia of then- natural host, though owing to the difficulty in obtaining these in any 134 numbers, the laboratory output would necessarily be very limited. It has occurred to the writer that it might be possible to employ as alternative hosts the puparia of some of the common Muscids, then liberating in great numbers the little insects, each with its dominant aim in life to seek out and destroy the greatest of the insect scourges of man and beast in the African Tropics. " The general upset of work as a consequence of war conditions made any attempt to put the idea to a practical test impossible, but an opportunity was made to carry out, during 1918 in East Africa and during the last six months of 1919 in Nyasaland, research work in regard to the question of alternative hosts for Chalcids normally parasitic on certain Muscids. The results (which will be written up later) showed some promise, though this work has again, unfortunately, been brought to a premature close." Some very interesting suggestions have been made by Dr. E. Rouband (122), as to the possibility of utilising for the control of Glossina the Chalcid Nasonia brevicornis, Ashm., which is known to parasitise the pupae of various Muscids throughout the world. In the course of an investigation conducted by him in Paris, Dr. Roubaud succeeded in inducing this parasite to oviposit in pupae of G. morsitans obtained from an imported stock of Tsetse-flies, which he had been able to breed in France for several years (see p. 159). Unfortunately the Tsetse puparia proved to be too tough to allow the adult parasites to emerge. Nevertheless it is considered by Roubaud that the destructive powers of N. brevicornis in connection with the pupae of various flies, and its perseverance in reaching and puncturing its victims, even when the latter are concealed under a light layer of earth, indicate the possibility of using this species in the campaign against Glossina, the pupae of which are hidden under bark, in holes in trees, under light sand, etc. As we have seen, a number of specific parasites of Tsetse-flies have been described during the last few years, but their biology is at present too little understood to allow of their being utilised to any great extent. This cannot be said of N. brevicornis, which can be obtained in large quantities from the pupae of common Sarcophagid flies. Under the conditions of Roubaud's laboratory experiments, the mode of action of the parasite towards the pupae of G. morsitans was the same as towards those of French Muscids. Young pupae were searched for eagerly, the old ones avoided. The development of the pupa of Glossina being slower (one month) than that of Calliphora (Blow-flies) , the possible limit of infestation of Glossina is naturally more extended. The impossibility of emergence limits the extent of parasitism, although it does not reduce the immediate destructiveness of the parasite. There naturally arises the question of the possibility of acclimatising the insect. N. brevicornis seems experimentally capable of enduring the usual temperature in which Tsetse-flies live ; at 77° to 82° F., which is the average temperature of Glossina haunts, its life is apparently normal, the activity of the adults being increased and the life-cycle appreciably shortened. A female reared at this temperature produced 105 individuals in 12 days. A greater difficulty is the maintenance of parasitism in Tsetse breeding grounds, if the continuity of parasitism at the expense of Glossina be impossible. Roubaud thinks it probable, however, that the Chalcidid will find species of Diptera other than Glossina in which successive generations can breed. Experiments in this connection would be interesting, and as regards disseminating the parasite among Glossina haunts no difficulty need be apprehended. Having obtained a quantity of parasitised pupae of 135 various Blow-flies, such as Calliphora, Liicilia, Sarcophaga, Chrysomyia, etc., these would be distributed a few days before emergence of the parasites, being either scattered by hand in the Glossina haunts, or , placed in small masses lightly covered with humus to protect them from predators. This method, applied to permanent breeding grounds of limited extent, would, in Roubaud's opinion, undoubtedly offer an • appreciable check to the increase of Glossina. Some years ago it was urged by Austen and Bagshawe (5) that it would be well worth while to endeavour to establish the various American species of the genus Spalangia (Family Chalcididae) in a Glossina morsitans area, and to watch the result. One of these minute , Hymenoptera (Spalangia muscidarum) " has been bred in large numbers from the puparia of Stomoxys calcitrans," and, with other species of the same genus, " also attacks the pupae of other flies, including Musca domestica and Lyperosia irritans." As to the possibility of obtaining assistance from Hymenopterous i parasites in reducing the numbers of G. palpalis, it must be admitted , that recent statements by Fiske (54) are not encouraging. Writing with reference to the islands and shores of Lake Victoria, in Uganda, this author says : — " The parasitism of puparia of Glossina palpalis in good breeding grounds by Hymenoptera was found to be absolutely negligible. The only instance encountered was on Wema Island, i where in a catch of 203 puparia, three were found attacked by a small gregarious Chalcid. A percentage of parasitism at this point of 1 • 5 ; was thus indicated. At other points near at hand enough more unbroken puparia were collected and examined to make a total of 1 ,775, and not another parasitic specimen was found. From other points, first and last, more than 3,000 living puparia were collected and examined, making some 5,000 in all, and no other case of parasitism was encountered. " Occasionally empty shells are found with small round holes such as are left by Hymenopterous parasites in emerging, but except at the one point noted above, examination of such shells has never disclosed the characteristic (and unmistakable) exuviae of such parasites. The selected breeding grounds of the species are practically absolute protection against this class of destructors. " The parasite discovered on Wema bred with the greatest freedom on puparia of Glossina in confinement. About one month was required for the generation. It could be an enemy if it would, but its instincts lead it elsewhere than in the breeding places of Glossina in search of prey." PARASITES OF THE ADULT FLIES. — The parasites that attack Tsetse- flies in the adult state may be divided into external parasites and internal parasites. So far as we are aware, the only external parasites hitherto recorded are those met with in Northern Nigeria by Dr. J. W. Scott Macfie. Writing in 1913 with reference to the Ilorin Province, Dr. Macfie says (94) : — " On a few specimens of both G. palpalis and G. tachinoides, taken during the months of May, June and August, minute red mites were found attached to the abdomen, thorax or legs. They were easily detached by the point of a brush, and in one instance the mite voluntarily left the tsetse-fly and was found crawling over the floor of a store-box. They were of two distinct types, and although at the same seasons other mites were common on such insects as mosquitos, 136 Chrysops, Stomoxys, etc., the forms found on the tsetse-flies were not observed on any other insects. Specimens of these mites were submitted to Mr. S. Hirst, of the British Museum, who . . . „ expressed the opinion that they were larval forms of Trombidiidae, and probably belonged to the genus Trombidium in its wide sense." As regards internal parasites of Tsetse-flies, the attention of investi- gators should likewise be directed to the parasitic worms1 (Nematodes), and to maladies of protozoal, bacterial or fungoid origin from which Glossina may suffer. Besides the various trypanosomes ( Trypanosoma gambiense, T. rhodesiense, T. brucei, T. congolense, T. angolense, T. cazalboui, etc.) of sleeping sickness and of Tsetse-fly disease of domestic animals, which pass a portion of their life-cycle in Tsetse-flies, we are already acquainted with others which so far appear to be peculiar to these Diptera, and to undergo their entire development within them. If we were to succeed in discovering a parasitic disease affecting Tsetse-flies, it might perhaps be possible to make pure cultures of the virus, to render it more virulent or more effective, to propagate it, and finally to arrive at results of practical value. Much of what has just been stated on the subject of enemies of Tsetse- flies, based as it is on a small number of facts, is evidently highly theoretical, and there can be little doubt that, among the investigations to which attention has been drawn, the majority will not lead to any result from the point of view of the destruction of Glossina. Never- theless it is well to point out that this line of research may yet repay serious attention. CHAPTER XIV. PROPHYLACTIC MEASURES. CLEARING. — CATCHING AND SYSTEMATICALLY TRAPPING TSETSE-FLIES. — OTHER METHODS OF CATCHING AND DESTROYING TSETSE- FLIES.— ARTIFICIAL BREEDING PLACES. — EMPLOYMENT OF TSETSE-REPELLENTS FOR THE PROTECTION OF STOCK. — OTHER PRECAUTIONS FOR THE PROTECTION OF STOCK. — PERSONAL PROTECTION OF HUMAN BEINGS. — DESTRUCTION OF BIG GAME. CLEARING. — In the campaign against Glossina at the present time efforts should especially be directed towards prophylactic measures, or, in the words of Prof. E. Trouessart, the hygiene of the soil. Of these measures, clearing is by far the most important, and its technique should be based on a study of the various questions relating 1 Between September 1911 and March 1912, in dissecting some 300 specimens of Glossina morsitans at Ngoa, in the Mpika Division, Northern Rhodesia, Mr. LI. Lloyd found in two flies four examples of a parasitic Nematode, which were identified by Dr. R. E. Leiper as immature forms of a species of Mermis, resembling fairly closely those met with some years previously by Prof. Minchin in G. palpalis, in Uganda. Dr. G. D. Hale Carpenter (25), on the other hand, in the course of his work upon G. palpalis var. fuscipes in Bugalla I., Lake Victoria, Uganda, found Nematodes in the abdominal cavity of 4 per 1,000 of the flies examined. Lastly, during the rainy season of 1911-12, the Mission Scientifique du Katanga (116) observed at Sankisia (Belgian Congo) four cases of intestinal parasitism of G. morsitans (two in wild flies caught in the savannah, and two others in laboratory-bred specimens) by small Nematodes, which were likewise identified by Prof. Gedoelst as larvae of Mermis. The largest of these worms measured from 38 to 40 mm. in length, by 0-25 mm. or 0'3 mm. in greatest breadth. The wild flies did not seem to be suffering any ill effects, but one of the others appeared to be somewhat affected. 137 to the biology of Tsetse-flies : fly-zones or fly-belts, range of flight, proportion of the sexes, seasonal variations, migrations, distribution of breeding-places, etc. Clearing is at present chiefly carried out in Tropical Africa as a means of protecting European settlements, native villages, cultivated areas, wells and springs, landing places, fords, etc., against Glossina palpalis, the principal disseminator of sleeping sickness. The work of clearing as a means of fighting G. palpalis belongs especially to the domain of tropical medicine and hygiene, but what is of service in the case of the species mentioned may likewise be applied to Glossina morsitans, the Tsetse-fly of stock, the distribution of which is much more extensive. " Since these Diptera are larviparous," writes Prof. E. Trouessart, of the Paris Museum, "it is impossible for us to attack the larvae themselves, and therefore in the vicinity of settlements we must systematically destroy the bush, which serves as shelter for the flies and provides them with spots in which to deposit their larvae. What has to be done is thoroughly to clear woods and thickets of all under- growth and scrub to a certain distance from farms, stables and stock- pens. Bush bordering roads, or surrounding or intersecting grazing- grounds, must likewise be grubbed up and destroyed ; it is also necessary to clear away scrub and tall grass for a considerable distance round watering-places, while the low vegetation growing under the shelter of bushes should be torn up and burned. If cattle remain in the open all night, it is well to surround them with fires producing an abundance of smoke, in order to keep away insects. Should it be necessary to make enclosures, instead of employing growing hedges it is always advisable to use simple fencing, or interlaced branches stripped of foliage, which might afford shelter to Tsetse-flies. Lastly, in the case of a locality free from Nagana, the utmost precautions should be taken against introducing animals coming from a district in which the disease is rife." With regard to G. palpalis in Uganda, Fiske (54), after prolonged study of the bionomics of this species on the islands and shores of Lake Victoria, expresses his conviction that : — " It is wholly impractical to consider any control measures involving artificial destruction of flies, and wholly 'necessary to rely upon measures designed to deprive the insects of either food or protection or to render food less available to them." For various reasons, however, this author considers it " inadvisable and even dangerous to contemplate control of the pest through depriving it of food. It is probable," he continues, " that if completely deprived of all favoured hosts it would be unable to exist on hosts favoured to a no greater degree than sheep, goats and man, but it is probable that it would continue to exist if cattle or pigs were provided. It is also certain that where favoured hosts are plentiful, man is almost immune to attack, but that when they are few man is freely attacked, and our object must always be to protect man rather than to destroy flies. " There is no such objection to the proposition of controlling fly through depriving it of protection, and it is on measures designed to this end that we must chiefly rely. They are the clearing measures already in use, and they have been proved efficacious on many occasions. The maximum of economy and efficiency is to be gained through clearing at precisely the right points — i.e., at the centres of infestation where natural increase of fly is most rapid. By clearing these the dispersion of flies into the surrounding zone is prevented, and the effect is general. 138 By clearing in the zone infested by immigrating flies the effect is local at best. A small amount of clearing at the centres of infestation may be much more effective than a large amount of clearing away from them." In discussing the degree of density that may be considered " ob- noxious/' Fiske expresses the belief that a density, " as measured by the males to be caught ' per boy per hour/ " of 6-0 " is ordinarily safe and sanitary, but that one of 15-0 or 20-0 would ordinarily be dan- gerous. Admittedly it would be desirable for purely sanitary reasons to reduce density to 0 everywhere, but the expense would be so enormous as to render it completely impracticable. Practical measures must combine efficiency with economy, and the most economical measures are such as do not carry reduction in density of the pest beyond what is necessary for sanitary reasons." In the course of some suggestions for the limitation and destruction of G. morsitans in Nyasaland, contained in a paper published in 1914, Dr. J. O. Shircore (139) writes as follows, after stating that, from his own personal observation, the " proclaimed Sleeping Sickness area/' in Dowa District, " contains at least four ' primary centres ' which harbour tsetse-flies." " The main reasons why these are defined as ' primary centres ' are : (1) Fly are found here abundantly throughout the year. (2) They are present in these situations when there are none or extremely few elsewhere, when the whole country is bare of grass, the trees without foliage, and the ground baked hard. (3) These are the only areas where in the dry season water is actually above the earth's surface or at no great depth below. For example, at the Lingadsi River the centre is almost definitely limited in length by the water and moisture in the river-bed. At Patsanjoka the water is present throughout the dry season. At Kuti and Chitawa it dries late, but here it is seen that although the marshes are superficially dry, yet during late October and early November, before any rain, fresh grass already springs up and trees begin to put on green foliage in the low-lying country near the marshes ; this is in marked contrast to other parts of the district, e.g., the portions between Nsadzu, Waya and Matumba to Mtalamanja, that being somewhat raised undulating broken country with a fairly rocky surface. (4) It is in these situations that the fly can best feed during the long season of drought ; there are herds of eland and buffalo at the Lingadsi ; waterbuck, hartebeeste and eland at Nyansato ; and various species of large and small game at Patsanjoka and Kuti dambos, and numbers of these can be seen at various times resting in the shade during the heat of the day. (5) In these places there is light forest with fairly short grass, here and there open glades, and water at no great distance, a combination which is ideal for both game and fly. The latter feed as the game come to and from water, and the light forest enables them to see and follow the game without difficulty. Heavy forest and thick high grass impede the flight of tsetses, and, moreover, they are unable to see any distance. They are sometimes found in such situations, but not in any numbers. " The fact that the fly are found at the above-mentioned centres all the year round, and that they still exist there at the height of the dry season, when there are few or none elsewhere, makes it quite reasonable to suppose that they breed at these places, and that it is from these centres that they extend into the surrounding country along connecting forest as soon as the conditions become suitable. This is actually observed to take place. During May, June, and early July, tsetse 139 are to be found constantly, though in small numbers, along routes where during August and September they absolutely disappear, or are present in almost negligible numbers, or only at odd times. This applies to almost all the main routes, except just at and opposite the four centres indicated. During late September, October and November a distinct but gradual increase becomes evident, and during November, before the rains, flies reappear in situations in which they were to be found early in May. Radiations from the primary centres begin about this period, so far as the physical character of the country, the large garden clearings and the seasonal conditions allow. With the onset of the rains the spread increases, as more suitable situations are created ; the rough stony country becomes a habitat, and other breeding centres — 4 secondary centres ' — are established. The fly thus continues to increase until the rains are over, the streams dry, and the grass withered, and then the hot season with its bush-fires once more performs a natural prophylactic measure. " On regarding the subject from this aspect, the measure which I advocate is that the forest connections along which radiations take place should be cut off, or otherwise dealt with, early in the year, about May, so that the actual ' primary centres,' which should have been delimited at the height of the previous dry season, are isolated. After this has been performed all the trees in these centres should be cut down and left to dry, and by the end of July or August, when the whole country is parched (grass-fires having been strictly prohibited during this period), wholesale and extensive burning should be carried out over these areas, commencing from the side of the ' extensions ' and progressing towards the main and isolated bodies of fly. At Kuti, Nyansato and Patsanjoka (Rifu) this could be done, and with especial ease at the last situation. At Lingadsi it would be difficult, but sufficient could be done here to prevent a southern extension which links up with the north arm of the Nyansato area during the rains. " If ' extensions ' depend on primary sources, as they undoubtedly do, then it must needs follow that their limitation or the lessening of their numbers must lessen the formation of ' secondary centres ' and their ' extensions/ and will therefore curtail the distribution and numbers of the fly. " Game plays a minor part in the actual habitat of G. morsitans. It is true that in the above-mentioned areas game is plentiful and that blood is necessary to the life of the fly, but there are numerous situations where game is present in this same prohibited area, yet fly are not found, and where fly though present during the wet season are absent during the dry." Unfortunately subsequent investigations by Dr. W. A. Lamborn, in the same " sleeping sickness area " as that studied by Dr. Shircore, do not appear to confirm the methods of clearing advocated by the latter. " Though the fly," writes Dr. Lamborn (82), " does occupy certain definite regions in the dry season from which, as I quite agree, they radiate out farther afield when the grass springs up everywhere and the game spreads abroad, the foci, in this district at all events, are in my opinion far too extensive to make any experimental attempt at clearing and reducing their number by systematic capture practicable. I have not been able to discover that the ' primary centres ' form localised breeding grounds, so that this very attractive hypothesis is up to the present unsupported by a shred of evidence." 140 It is clear that the entire question needs to be studied fully in other localities, and we have thought it worth while to reproduce Dr. Shircore's proposals above, since the observations that might be made in G. morsitans belts elsewhere, whether they proved to support his method or otherwise, could not fail to be of the greatest interest. We give below a summary of the " Preventive Measures " against Sleeping Sickness disseminated by G. morsitans, which were introduced some years ago in Nyasaland, under the supervision of Dr. H. Hearsey, Principal Medical Officer, and Dr. Conran.1 "1. Impressing upon the natives the necessity of avoiding being bitten by flies. " 2. Clearing for a distance of 300 yards or more round villages situated in close proximity to fly. Extending these clearings in many instances so as to embrace their food gardens also. " 3. Making clearings on each side of roads and main paths leading from one village to another, thus uniting groups of villages and enabling the natives to journey from one to another with commensurate safety. "4. .Instructing the natives to use this felled timber for firewood, and to avoid visiting the adjacent fly-infested woods for this purpose. " 5. Instituting public latrines in all the villages, to prevent natives from visiting the adjoining woods for the purpose of defaecation, " Prohibition as to the firing of grass till the month of October, when only it is fit for burning, so as to ensure as large and as effectively cleared areas as possible. " Removal, whenever feasible, of villages situated in danger zones> to fly-free areas." In a paper (84) published in 1916 Dr. Lamborn records some observations on the effect of bush fires on the distribution of G. morsitans in certain districts in the proclaimed area of Nyasaland. A marked decrease in numbers was noted in the burnt area, but pupae sufficient to repopulate the latter were found under logs untouched by the fires. Flies were abundant in a neighbouring district, and this abundance was undoubtedly the result of the fire and the driving away of the game. " The flames/' writes Dr. Lamborn, " are not to any extent instrumental in burning up the dead trees which form so large a proportion of the breeding places, the fact being that the fires are too fleeting as a rule to do more than just char most of the logs, which are also protected by dust and by earthy deposits of termites. " Systematic burning of the grass is not, in my opinion, likely to be of value for controlling the fly, because over much of the fly country the grass is patchy, so that the fires fail to spread, and in some regions it will not burn till late on in the dry season, when the fly is able to find refuge in areas burnt long before. " Where the fires have been extensive, most of the game seems to be driven off and to remain for the time being in other parts, with the exception of wart-hog, which then grubs up roots. This animal therefore probably provides food for the newly emerged flies." On the other hand, the probability that good effects will eventually result from clearing bush and woodland areas infested by Tsetse, especially G. brevipalpis, by means of regulated late annual burnings is strongly urged by Swynnerton, in his recent paper on " The Tsetse 1 Cf. " Nyasaland Protectorate. Sleeping Sickness. Diary." Part xxL By the Principal Medical Officer (Zomba. Printed by the Government Printer, 1913). Pp. 10, 11. 141 Problem in North Mossurise, Portuguese East Africa " (145). " Late burning/' writes this author, "is no emergency measure aganst fly. In some areas especially, where the grass is already well reduced, its full effects may take long to show. Again, its abandonment will mean the return of the old conditions, for the underground stumps and roots of pyrophytes are extraordinarily tenacious of life. It must be regarded as a piece of administrative policy and be kept up annually. That its adoption is necessary I am quite convinced, if only for the reason that the country generally [i.e. North Mossurise] is at present reverting more and more to wooding and becoming increasingly fitted for tsetse, even in those areas in which the cover has hitherto been sparse and light. The recommendation of the measure, in face of its having to be kept up indefinitely, must be that the fires take place in ! any case and that the annual cost of regulating them will be trivial. The latter may even be met by the fines imposed on unauthorised burners, and the actual burning, being done at a signal by the kraal natives themselves, will cost nothing. " Apart from the prophylactic .value of the measure just mentioned, its effect, grass conditions being equal (which they are not), should show , first and chiefly in relation to G. brevipalpis and (in the area in which ! this fly depends mainly on coppice) pallidipes. It will not affect the grown pyrophytic trees in any wholesale manner (unless in a very I exceptional season or unless they are weakened by ring-barking). Its ' effect in relation to morsitans might even be regarded as problematical, ! for that fly and pallidipes in the same wooding appear to be independent of undergrowth. The systematic burning back of smallish growth will assuredly tell as the established trees pass maturity and eventually go, with nothing to replace them, but the time involved will be long." In concluding this paper, Swynnerton writes : — •" I wish, therefore, to make it the one outstanding recommendation of this report that late burning — with, very frequently indeed, a year of no burning — should be given a trial over a considerable number of years ; also that, in the first season at any rate, an effort should be made to have the burning as simultaneous as possible over a large block." As regards the value of grass burning as a means of attack against G. morsitans in Southern Rhodesia, views recently expressed by R. W. Jack are similar to those of Dr. Lamborn quoted above (p. 140) with reference to conditions in Nyasaland. It is pointed out by Jack (75) that several investigators have emphasised the efficacy of grass fires in reducing the numbers of the fly. He himself, however, " has been unable to obtain the slightest evidence " in support of this theory ; "indeed the evidence to the contrary is almost conclusive." If the fly " could not avoid grass fires," it would seem impossible that the species could escape extermination " in parts of the country inhabited by natives." The mopani belts, where the grass remains short, are probably, this author thinks, a convenient refuge from grass fires in the dry season, and are also attractive to game after the onset of the rains. CATCHING AND SYSTEMATICALLY TRAPPING TSETSE-FLIES. — It is obvious that in the case of a fly like Glossina, with feeble powers of reproduction, catching the adults ought to produce a relatively much more satisfactory effect than in that of a prolific insect such as the House-fly, which is capable of producing from 600 to 900 eggs in the course of its life. During recent years various attempts at catching 142 Tsetse-flies have been made, with widely different results, as will be seen from the subjoined resume. Simple Catching. — A trial of simply catching Tsetse-flies, as a means of destroying them, was undertaken by H. Koch (81) in a well-defined infested area. The site chosen for the experiment was Mugassiro I., in Mara Bay, on the eastern shore of Lake Victoria, and the trial lasted from 29th January 1913 to 31st January 1914. The island, which is 1 1 mile in circumference and swarms with crocodiles, is inhabited and covered with dense bush. Two pairs of boys skilled in catching Tsetse worked there for nearly a year, each pair relieving the other every three months. The daily catch was placed in glass bottles, half full of spirit, the bottles being sent to camp at intervals of ten days for the flies caught to be counted. In 340 days the bag of G. palpalis amounted to 74,382—49,883 males and 24,499 females, but the flies on the island were not exterminated, an unsatisfactory result attributed by the author to insufficiency of personnel and the shape of the island. Trapping and Catching by means of Bird-lime or "Tanglefoot." — Thi2 subjoined description of a highly successful mode of catching G. palpalis is taken from the Sleeping Sickness Bulletin.1 " Mr. Maldonado, manager of one of the estates on the Island of Principe, has devised a method of destroying Glossina palpalis. Noticing that the flies attacked the backs of the labourers when they were occupied in mowing grass, and were consequently in a stooping posture, he ordered that such labourers should wear a black cloth covering their backs, coated with a glutinous substance [apparently bird-lime] on its outer surface. Between April 1906 and the end of 1907, 133,778 Tsetse were thus trapped on this plantation alone. While the Portuguese Commission was in the island there were not more than four persons who went about with these cloths. The Commission often asked Mr. Maldonado to send men with black cloths to places where they had seen a large number of flies. ' As a rule two men were enough in the short space of a week to make these places passable (practicables) .' On the first days the numbers taken would be 1,500-2,000 ; at the end of the week 15 or 20. The method has been tried on two other estates, with the same favourable results." Mr. Maldonado's method was subsequently adopted by the Medical Commission which, under the leadership of Surgeon-Captain B. F. Bruto Da Costa (36-40), was entrusted by the Portuguese Government with the task of deciding upon the measures to be taken to deal with and, if possible, free the Island of Principe of G. palpalis and sleeping sickness, which was introduced about the year 1890. In addition to prescribing the catching of Tsetse-flies by means of cloths coated with bird-lime, the following procedure was laid down by the Commission for the guidance of the planters : marshes were to be drained, and the woods affording shelter to G. palpalis were to be cut down ; the wild pigs, which swarmed in the island, were to be destroyed, and similar measures were to be applied to civet cats and stray dogs ; scrubby jungle was to be cleared, and all domestic animals suffering from trypanosomiasis were to be slaughtered. The medical measures adopted included the isolation of patients and the giving of atoxyl injections as soon as possible after the bite of the fly. Between the month of August 1912 and the end of May 1913, while the Commission was in existence, a brigade of labourers wearing black 1 Vol. ii, no. 13, p. 26 (Jan. 1,910). 143 cloths smeared with bird-lime caught 110,691 Tsetse-flies. The work was continued subsequently without intermission, and the diminution in the numbers of G. palpalis in all parts of the island where the sanitary measures detailed above were carried out is stated to have been beyond dispute. As a case in point Surgeon-Captain Da Costa mentions the Roca Sundy Plantation, where, at the commencement of the work of the Commission, from 9,000 to 10,000 Tsetse were caught every month. In October 1913 the number of had fallen to 988, and in November to 380, in spite of the fact that twenty-two men were employed for the purpose instead of the original seven. At Porto Real the initial monthly total of 7,000 fell in October 1913 to 314. In round numbers the catch for the entire island amounted in October 1912 to 15,970 flies, and in October 1913 to 2,300. Lastly the report of Surgeon-Captain Da Costa for February and March 1914, shows the complete success of the bird-lime method, in conjunction with the other measures of prophylaxis. While in February and March of the previous year 173 men still caught respectively 11,865 and 9,450 flies, 38 men, operating in the plantation that had been the worst infested, failed to catch a single fly during the month of March 1914. After 1 1th April 1914 the offer of a reward of 5 escudos1 for every fly failed to produce even one Tsetse, so that in the Island of Principe the total extermination of G. palpalis would seem to have been achieved. Experiments in trapping by aid of bird-lime, which met with less success, have also been made in Belgian Congo, German East Africa (Tanganyika Territory) and Nyasaland. The Mission Scientifique du Katanga (Drs. Rodhain, Pons, F. Van den Branden and J. Bequaert — 116), which worked in Katanga (Belgian Congo) from October 1910 to September 1912, made a few attempts at catching Glossina morsitans at Sankisia, by means of " tanglefoot " prepared from linseed oil obtained from Europe. Two natives, carrying on their backs sheets of tanglefoot measuring 7 by 11 in., set out at fixed times ; one of them was provided with a fly-net, which he was accustomed to use, and was told to catch G. morsitans. On the return of the ooys it was found that the total number of Tsetse caught by means of the tanglefoot was 134, while the bag made with the single net amounted to 145. According to Dr. Rodhain, the results of these experiments are in agreement with those obtained by the German investigators, who worked on a larger scale, and show that a single fly-boy catches more G. morsitans than two natives carrying sheets of tanglefoot on their backs. Dr. Rodhain therefore concludes that the latter method does not appear likely to yield appreciable results of practical importance. On the other hand it might be worth while to try an improved type of sticky trap, described by Dr. Zupitza in the Archiv fur Schiffs-und Tropen- Hygiene for June 1914 (No. 11). This trap, which is said to be just as effective as a butterfly-net, and to be capable of being used by one man, consists of a sheet of oxidised wire gauze, with meshes of one centimetre. The gauze is made into an oval tube, completely enveloping the body of the native carrying it ; openings are provided for the arms, and the contrivance is kept in position by straps. Under this armour the native wears a garment of porous black material which can readily be washed, the object of the meshes and of the loosely woven fabric being to allow the passage of emanations from the man's body. A coating of tanglefoot composition, suitably viscous, is applied 1 Equivalent in pre-war currency to about 15 shillings. 144 to the gauze ; this composition must not give off any odour repellent to Tsetse-flies, but the smell of resin does not appear to be of this character. Unfortunately the experiments that have been made with this trap have not hitherto proved satisfactory, since the Tsetse-flies that happened to be in the immediate vicinity of the man were the only ones that were caught. We quote below Dr. W. A. Lamborn's account (82) of experiments in trapping Glossina morsitans, carried out by him in Nyasaland, in December 1914. " In accordance with Sir Patrick Hanson's suggestion," writes Dr. Lamborn, " a cage of wire mosquito-proof netting was constructed in a part of the Lingadzi district where tsetses are especially numerous, and in this a goat was incarcerated, the outer side of the cage and the under sides of branches of trees near by being smeared with bird-lime in the hope of catching the flies attracted by the animal. The results were by no means satisfactory, for no more than eight flies were taken, though the goat remained there for seven days, a failure possibly to be explained by its not being called upon to move about with any degree of activity, and by the inability of the fly to see such movements as occurred, my experience in this connexion confirming that of other observers, namely, that the flies are attracted especially by moving objects. " Why, however, the odour of the goat, which was especially strong, the animal being a male, should have failed to attract them it is difficult to understand, seeing that, like some TABANIDAE, G. morsitans will after a little time discover a freshly killed animal and settle to feed, in which case the fly can only have been drawn to the scene by the sense of smell. When this particular goat was removed and paraded up and down in the vicinity it was much assailed by the flies. " A series of experiments was made in the use of English bird-lime spread on various materials carried on the backs of natives, in the hopes of catching morsitans in the same way as palpalis is caught in the Island of Principe. The results are tabulated below : — Date. Material used. Boys Duration o: experiment Tsetses captured. 8 Dec. Deal board, 3 ft. by 3 ft a 6 hours 18 9 Green cardboard, 3 ft. by 3 ft. a M 23 9 a b M 15 9 Tarred waterproof paper, 3 ft. by 3 ft. c ,, 39 10 » a a >i a tt 17 11 a 29 12 „ Board made of dried bamboos laced- a 7 ,, 12 15 ,, together, 3 ft. by 3 ft. Light brown paper, 3 ft. by 3 ft. « O a b " 7 53 59 16 a M 58 " " " b M 67 17 a H 31 " " " " b' it 54 18 a M 51 b " 47 " As morsitans is so frequently observed to settle on the under side of loads carried on the heads of natives, experiments were also made 145 in the use of material limed on the under side carried in a horizontal position, but the results were not so satisfactory as when the material was carried in a vertical position." The following extract from " Suggestions for Future Work," by Austen and Bagshawe (5), is perhaps also worth noting. " Correct conclusions as to the value of trapping and catching cannot be deduced from a trial lasting only two or three months, in which but a handful of natives are engaged ; the value of limed cloths and of nets in the hands of expert fly-boys should be tested in a selected locality for at least a year, and on a large scale. Some hundreds of natives should be employed in the work, and a corps of, say, a dozen keen and energetic fly-boys should be established in each village in a tsetse area. Payment should be by results, but small rewards would suffice, and the system should therefore prove inexpensive in working." It has been pointed out by Simpson (144) that in wide areas of Tsetse-infested country, such as occur in the West African colonies, systematic trapping of Glossina is impracticable. " Systematic trapping with bird-lime," when tried by this author in the Gold Coast, was " found to be less satisfactory than by means of nets. Four colours of paper were used — light brown, red, black waterproof, and green. These were most satisfactory in the order given." A trap suggested by Dr. J. O. Shircore (140) for use against G. morsitans, based on the known habits of this species, consists of a series of revolving canvas screens, smeared with adhesives and placed on routes known to be frequented by the fly. A later idea of the same author (141) is that logs, covered with some adhesive substance and propped on supports two or three inches off the ground, might act as effective traps. We are unaware whether either of these suggestions has yet been submitted to a practical test. OTHER METHODS OF CATCHING AND DESTROYING TSETSE-FLIES. A study of the tropisms, or physiological reactions, to which the genus Glossina, like all other insects, is subject, would perhaps lead to the discovery of an effective plan of campaign against these Diptera. In endeavouring to construct serviceable insect-traps, it is always well to bear in mind the special attraction exerted by light and by odours ; in the case of Tsetse-flies investigations are needed to discover some specially attractive substance. About ten years ago Dr. Andrew Balfour, C.B., C.M.G., had some large Tsetse-fly traps constructed by the Andres Maire Company, with a view to their being tested in Western Mongalla (Egyptian Sudan). Writing in 1913 of the experiments that it was intended to carry out with these traps, Dr. Balfour remarks 1:— " As regards methods of attraction, the following measures will be tried : — "1. Soaking the central bands or wicks in a mixture of water and either human or animal sweat. "2. Placing a live animal in the trap .... " 3. Soaking the central bands in citrated blood, and at the same time hanging up in the trap a piece of fresh meat from ox or buck with skin still adherent. 1 Annals of Tropical Medicine and Parasitology, Series T.M. Vol. VII. no. 1, p. 119 (March 31, 1913). (5979) K 146 "4. Soaking the bands as above, and having in addition a tube of citrated blood arranged according to the ingenious device of Rodhain and his colleagues [see pp. 88, 89]. If desired, the blood in this tube may be poisoned." It would be interesting to make an exhaustive trial of these various methods of trapping, in regions in which nagana is rife and Glossina morsitans abounds ; but it should be noted that in an experiment on the above lines carried out in the Sudan by the late Captain Ranken, R.A.M.C., the powerful odour of human (native) sweat entirely failed to attract this species. On the other hand, in attempting to devise some practical form of trap the attraction exerted upon Tsetse-flies by moving objects should not be forgotten. The continuance of the study of the influence upon Tsetse-flies of external agents, such as heat, cold, solar radiation, humidity, colours, etc., is likewise of great importance. Who knows whether some small discovery, insignificant though it be in appearance, may not enable us to deal effectively with a scourge which is at present devastating and hindering the development of Tropical Africa.1 As regards the pupae, the effect of certain insecticides, such as . compounds containing sulphate of copper, arsenical mixtures, sulphide of carbon, etc., should be investigated. ARTIFICIAL BREEDING PLACES. — The creation of artificial breeding places, leading to the concentration of the female flies and their pupae in chosen spots, thus facilitating their capture and destruction, would seem to be an effective method of attacking Glossina. In its simplest form, such a breeding place consists of a prostrate tree-trunk raised a few inches above the ground, so as to form a dark hiding-place to which the pregnant fly may retire during the period of gestation. The latter characteristic, as we have already learnt from the investigations of Mr. LI. Lloyd and Dr. W. A. Lamborn, is the only one common to all Tsetse-fly breeding places in nature (see Plate IV, Figs. 1 and 2, and Plate V, Fig. 1). Needless to say, in order that artificial breeding places may have a chance of being chosen by the female flies, all the natural breeding places in the fly-area concerned must, so far as possible, have previously been destroyed or rendered useless. Felled trees, being capable of forming breeding places, must be got rid of, and clearing operations must be carried out with care. The lower surfaces of materials employed in the construction of artificial breeding places may, as suggested by Dr. Shircore (see p. 145), be coated with bird- lime or tanglefoot, in order to catch the adult flies. All pupae should be collected at frequent intervals and allowed to hatch, the flies produced being destroyed, but any parasites being at once released. A large number of artificial breecting places for G. morsitans on the above lines were constructed in 1915 by Dr. Lamborn (84) in NyasalancU at Rifu in the proclaimed area, and near Monkey Bay, the object being to determine which trees are most favoured as resting places by the pregnant female flies. The work was planned on a comprehensive scale, and carried out in accordance with a well thought-out scheme, in such a way as to render it easy to find the breeding places again at any future time. In all cases the breeding places were formed of trees 1 Certain species of the Tsetse-flies appear as yet to have a very restricted distribution. A knowledge of the reasons for this natural limitation, as compared with the great extent of the areas occupied by other species, such as G. morsitans, G. palpalis, .and G. brevipalpis, would perhaps lead to the discovery of effective means for the destruction of these insects. 147 felled for the purpose, the trees selected belonging to a number of I different species, spaced as regularly as possible. Altogether more than one hundred breeding places were constructed in this way, some being in dense woodland and others in thorn-bush. The scene of operations near Monkey Bay covered two square miles of country, throughout which natural breeding places were, so far as practicable, eliminated. " In some of these [artificial] breeding places," writes Dr. Lamborn, " the natural soil has been replaced by soil brought from natural breeding places, and in others the earth has been well mixed with fragments of rotting wood and bark, with earth from termitaria, and with antelope droppings, the latter method being the only available means of imparting an odour of game to the places, with the idea of i; ascertaining whether the choice of the fly can be in any way influenced." Unfortunately the departure of Dr. Lamborn on active service in East Africa prevented him from following out this most interesting and important practical experiment, and no report on the results obtained has yet appeared. In Uganda much attention has in recent years been paid by Dr. G. D. H. Carpenter to the value of artificial bre'eding places in reducing the numbers of G. palpalis on Lake Victoria. According to Dr. Carpenter (26), the collection of pupae seems to offer the most promising results, owing to the proportion of the sexes being equal. It is therefore suggested that all breeding grounds should be cleared, and that artificial shelters, that would induce the flies to deposit their larvae in them, should be provided ; from these the pupae could be collected ; regularly. This method has been tried to some extent, and the results I undoubtedly warrant further experiment. Such shelters, Dr. Carpenter j thinks, should be in the form of a low lean-to shed, the base and I preferably the uprights being of metal, owing to the presence of j termites. The low front should face the direction of the prevailing i wind ; the roof should be thatched, with good eaves almost touching the ground in front ; the sides and back should be open, but protected by eaves and creepers. If the shelter can be made of fresh, green growth, this will add to its attractiveness, and creepers might be planted to ramble over it. Diagrams and photographs of such shelters are given by Dr. Carpenter. One of a suitable size has an area of 12 ft. by 4 ft., and is 4| ft. high at the back and H ft. in front. It is obvious that constant attention must be given to such shelters and the pupae regularly collected, otherwise they will tend to the increase of the fly rather than to its diminution. The roofs must be kept watertight. This method cannot be expected to exterminate the fly, though its numbers can be reduced immensely, and where possible the collection of. adult flies should be continued. Moreover, in Dr. Carpenter's opinion, this method is the simplest, the least costly and the most practicable of any yet proposed for dealing with Glossina, and should have the effect of keeping the fly at a minimum harmless number, below which no method, short of defoliating the whole country, could bring it. EMPLOYMENT OF TSETSE-REPELLENTS FOR THE PROTECTION OF STOCK. — A number of experiments with repellent substances, with a view to protecting stock from the attacks of Tsetse-flies, have been made in Africa, but without much success. In the United States of America a good deal of work has likewise been done, in studying the employment of substances of this kind for the protection of domestic animals against the attacks of biting flies. It seems highly important 148 to draw the attention of colonists in Tropical Africa to these investi- gations, in order that similar trials may be made in regions where stock is subject to the attacks of Tsetse-flies. While the destruction of the insects themselves would certainly be the most effective mode of attack in a campaign against Glossina, the discovery of a really serviceable repellent, costing little, easy to apply, and forming a barrier, temporary though it be, between the fly and its victim, would render an appreciable service to stock-raising and agriculture in Tropical Africa. With this end in view, we give below a few extracts from an extremely useful little brochure (59) published not many years ago in the United States. " There are almost innumerable homemade and proprietary external remedies for repelling flies. They contain various substances that are distasteful to the insects. Many of them contain strongly odoriferous ingredients that have a repelling influence on flies. The qualities to be sought in a satisfactory repellent are : Absence of toxic and other detrimental properties when applied externally to animals ; a marked repellent action on flies ; and a duration of this action for a reasonable length of time. A common defect of many otherwise rather good repellents is the very short period during which they are effective. Some repellents are undoubtedly toxic and must be used with care. " Repellents as a rule are in the form of liquids, and may be applied by means of a dipping vat, a pail spray pump, an atomizer such as that commonly used in gardens and greenhouses for applying insecticides to plants, or by means of a rag or a paint brush. The method employed necessarily depends to a very large extent on the number of animals to be treated, the physical character and toxicity of the preparation, its cost, and the individual preference of the farmer or stockman. Some preparations, either because of their cost or their toxicity, or for some other reason, are not adapted for use in a dipping vat or for application by means of a spray pump. Others may be applied by any one of the methods mentioned." " The powder remedies that have been used are pyrethrum powder and tobacco powder. " Various oils, emulsions of oils, and mixtures of oils are used in repelling flies. Crude petroleum, cottonseed oil, fish or train oil, and light coal-tar oil may be used pure. Jensen (1909) recommends for dairy cows an emulsion of crude petroleum containing an admixture of powdered naphthalin. " Fish oil is rated as one of the best repellents, and has been used alone and in combination with various other substances. Other substances that have repellent qualities and that have been used in various mixtures are pine tar, oil of tar, crude carbolic acid, oil of pennyroyal, and kerosene. " Jensen's formula1 is said to protect cows for a week. The pro- tective action of fish oil is stated to range from less than two days (Parrott, 1900) to six days. Moore's formula2 is said to protect for two days. This mixture is safe when applied lightly with a brush, but not when applied liberally with a pail spray pump. " Laurel oil is a very effective repellent. Mayer (1911) found that the protection lasted from 2 to 12 days. The oil when used pure has an irritating effect unless it is applied lightly. According to Mayer the irritating effect may be overcome by combining it with linseed oil in 1 Common laundry soap, 1 Ib. ; water, 4 gals. ; crude petroleum, 1 gal. ; powdered naphthalin, 4 oz. 2 Fish oil, 100 parts ; oil of tar, 50 parts ; crude carbolic acid, 1 part. 149 the proportion of 1 to 10. The present author found that 10 per cent, of laurel oil in cottonseed oil was active for less than a day." " In experimental tests carried out by the present author the following results were obtained : — "A 10 per cent, mixture of crude carbolic acid (21-8 per cent. phenols) in cottonseed oil has a very strong repellent action on flies, but this lasts less than a day, in consequence of which it is necessary to apply the mixture every day. The mixture should be applied lightly with a brush, since a heavy application with a spray pump is likely to cause phenol poisoning. " Mixtures consisting of 10, 20, and 50 per cent, of pine tar in cotton- seed oil have marked repellent qualities. They should be applied lightly, and it is necessary to apply them every day. A liberal applica- tion of a 10. per cent, mixture is deleterious to animals. This is also the case with a half-and-half mixture of pine tar and Beaumont oil when applied lightly with a brush. " A mixture of oil of tar (14 per cent, phenols, volatile with steam) in cottonseed oil and in Beaumont oil has a very marked repellent action. A 10 per cent, mixture of oil of tar in cottonseed oil is safe. A half-and-half mixture of oil of tar and cottonseed oil, when applied liberally with a spray pump and 50 per cent, oil of tar in Beaumont oil applied with a brush are not safe. Ten per cent, oil of tar in Beaumont oil is safe. When applied lightly it is necessary to apply 10 per cent, oil of tar in cottonseed oil or 10 per cent, oil of tar in Beaumont oil every day. " Mixtures of 10 per cent, oil of citronella, oil of sassafras, or oil of camphor in cottonseed oil are powerful repellents, but they are active for less than a day. " A heavy application of fish oil causes the hair to become sticky and fall out. A light application did not produce these results. " Pyrethrum powder is an effective repellent, but its action lasts only for about a day." In 1916, experiments were made in Canada by A. W. Baker (6) with a view to discovering a repellent which could be prepared cheaply by the farmer and would, without detrimental effects of any kind, protect cattle for at least two days from the attacks of Lyperosia and Stomoxys — well-known genera of blood-sucking Muscidae, of which several species are found in Africa. After several trials, it was found that the following mixture gave good results in the proportion of 1 to 2, 1 to 3, and 1 to 4 of water : 1 gal. fish oil, 1 gal. kerosene, 1 gal. slightly sour milk, 6 oz. oil of citronella. The cost of thoroughly spraying each animal from horns to hoofs is stated to work out at about one farthing. In India, according to Cross (33), citronella oil, used alone, prevents Tabanids and Stomoxys from attacking camels for a few hours, but has no repellent action after 17 hours. Castor oil was found to have no effect against either Stomoxys or Tabanids when only one pint per camel was used, but at the rate of four pints per camel prevents attacks by these flies for three days, and gives partial immunity for about two days longer. This treatment, however, is too expensive to be of practical use. It may be noted that in an experiment carried out by Simpson (144), in the Gold Coast, with various oils and essences smeared over the bodies of natives employed by him to collect Tsetse-flies,* the only substances that appeared to act as deterrents were beechwood creosote and beechwood oil. 150 Swynnerton (145), in Portuguese East Africa, was unable to make " any experiments in the matter of spraying or wiping animals with paraffin or other deterrent liquids," but it is pointed out by this author that " Against brevipalpis the animals' lower surface should be sprayed." He also makes the interesting suggestion that " an internal tsetsefuge," if such can be found, might possibly prove both effective in the case of cattle and of value as a preventive of sleeping sickness. " An internal tsetsefuge," writes Swynnerton, "might .... be experimented for and be used where the number of cattle was not too large for treatment. During our first epizootic of African Coast fever, in 1900, when I was already losing stock seriously, I tried dosing the surviving cattle each morning with a double handful each of garlic, crushed and given in a bottle of water — as we had then no dipping tanks. I am inclined to attribute the fact that I saved a much larger percentage of our small herd than did any of our neighbours (who lost nearly everything) in part to the deterrent effect of the garlic on the ticks. I do not know its effect in relation to biting flies, but it might be tested, and if it were successful, garlic-eating might become a useful if obnoxious habit in sleeping sickness areas." OTHER PRECAUTIONS FOR THE PROTECTION OF STOCK. — When cattle have to be transported by rail through a zone infested by Tsetse- fly, they should be conveyed in screened wagons, as is stated by Bouet (12) to be the case on the Ivory Coast railway. For the protection of a mob of cattle moving on foot through fly, Swynnerton (145) suggests the placing of animals of lesser value a short distance in front of the others. Experience gained by this author, when passing with cattle through country infested by G. brevipalpis, goes to show that by the adoption of this plan the leading animals draw off all the fly, those that come after getting few or none. PERSONAL PROTECTION OF HUMAN BEINGS. — " When shooting or working in a tsetse-fly country," writes Sir David Bruce (19), " some protection from the bites of these poisonous insects is necessary. It must be borne in mind that a single bite from an infected tsetse-fly will almost certainly give rise to sleeping sickness. It is evident, then, that shorts are not permissible. Their place must be taken by strong drill khaki breeches, with boots and gaiters or puttees. It is best to wear a coat over the shirt in order to protect the body and arms. The neck and face are favourite places of attack, and anything in the way of a net veil or covering attached to the brim of the helmet is apt to be hot and stuffy, and, further, is constantly being; caught and torn by the thorns with which all the trees in the fly-country are plentifully supplied. Moreover, the veil interferes with vision, and is in the way when shooting. " To remedy this a headpiece [figured by the author in his paper] was invented, which was found to be practical, fairly cool and safe. Knitting-needles, about one-third of an inch in thickness, were made of hard wood. With these a head-covering was knitted, similar in appearance to a Crusader's headpiece of chain-mail. Native string, which is stiff and hard like whipcord, was used. The result was a close-fitting cap, extending well down on the chest and back, and leaving only a small opening for the eyes, nose and mouth. It was found that the tsetse-flies seldom attacked this opening, and, if they did, were seen and driven away. The string made a layer about one- third of an inch thick like chain-mail, full of spaces and interstices, 151 which permitted free ventilation and evaporation. But a tsetse or other biting fly could get through these spaces. To prevent this a layer of coarse mosquito netting was sewn on the surface of the knitted string, so that a fly found its proboscis too short to reach the skin. As this fly-proof helmet lies close to the head, it is possible to creep or push through thorny scrub without any danger of being held up. Of course, as a protection against the sun, a large felt or terai hat must be worn over the helmet, but this can easily be arranged. " When cycling through fly-country, especially on a motor-cycle, the tsetse-flies often attack in great numbers, and very viciously. Sometimes they follow like a swarm of angry bees, and it is necessary to stop now and then and break off a branch in order by its aid to drive them away. On these occasions the cyclist is liable to be severely bitten unless he is adequately protected. A net is of little use, but the above-described headgear, with the addition of a pair of motor- goggles, gives excellent protection for the face and neck. " It is necessary to see that there is no undefended space between the goggles and the helmet. The only parts exposed are the nose and mouth ; if a fly attacks here it can be blown away. " To protect the hands, fingerless gloves with khaki gauntlets were made in the same way, except that the palm was made of leather, with a hole for the forefinger to go through." The failure of external applications to the skin, with the exception of beechwood creosote and beechwood oil, to afford protection from the bites of Tsetse-flies, and a suggestion as to the possibly protective effect of eating garlic have been noted above (see pp. 149, 150). DESTRUCTION OF BIG GAME. — This method of combating Tsetse- flies, which is advocated by certain writers, seems a radical one, but at present none can tell whether it would prove really effective. As we have already shown, the question is not yet settled, and what is actually needed is experiments scientifically planned, and carried out with every precaution necessary to ensure accurate results. The experimental abolition of the .wild fauna might, for instance, be tried in a definite region. It would thus be possible to determine the degree of influence exerted by big game on the bionomics of the fly, on its infectivity, and subsequently on the health of human beings and domestic animals. In this connection some pertinent recommendations are contained in a report by Sir Arnold Theiler (147), published in 1917. It is stated that an investigation had been carried out in Zululand, with the object of determining the extent of infection with trypanosomiasis, and with particular reference to the spread of the disease and the best means of prevention and eradication. The conclusions reached were that wandering game is responsible for primary outbreaks of the disease ; and that Glossina does not travel from the original fly-belt unless cover is continuous or it is in company with game. In the absence of Tsetse it was considered unlikely that outbreaks of nagana could originate from recovered or partly recovered cattle, or from game localised in the cattle-grazing area. As a result of the investigations, certain recommendations had been made with a view to localising game so far as possible, and for its efficient protection in the reserves. These include the issue of permits to Europeans and natives living in nagana areas, authorising them to destroy wandering game ; these permits to be valid during the summer months as well as during the open season. The destruction of game in 152 infected areas should be encouraged by the issue of licences at reduced fees to non-residents. In the low veld areas where there are few, if any, cattle, and in areas surrounding game reserves, shooting game under licence should be allowed. A European ranger should be appointed to supervise the game reserves in each district ; the existing open area on both sides of the roads should be maintained for some time, and the bush adjoining roads passing through Tsetse-fly areas should be burned repeatedly. The assistance of owners is stated to be necessary to control nagana, since matters such as insufficient herding, transport riding in summer in areas in which the disease is endemic, and lack of isolation measures are better dealt with by owners than by the State.1 CHAPTER XV. MODE OF COLLECTING, PRESERVING AND STUDYING TSETSE-FLIES MATERIALS NEEDED FOR COLLECTING TSETSE-FLIES. — The would-be collector of Tsetse-flies and other Diptera needs but little in the way of equipment : all that he requires is a net, a cyanide killing-bottle, glass-bottomed cardboard pill-boxes, entomological forceps, needles mounted in handles, entomological pins, cork-carpet or pith, one or two corkrlined entomological store-boxes, and a good platyscopic lens. The following notes on the foregoing articles are taken from " Instructions for Collectors : No. 7. — Blood-Sucking Flies, Ticks, etc." By Major E. E. Austen, D.S.O. Fourth Edition. 1914. 2 A suitable collecting-net can be obtained from any dealer in natural- history apparatus Any net used for collecting butterflies will serve, but, on the whole, perhaps an ordinary umbrella-net or kite-net will be found the most useful.3 One or two spare net-bags are advisable in case the one in use gets torn. The cyanide killing-bottle should be prepared as follows : Cover the bottom of a glass jar, which must have a wide mouth and a closely 1 According to a statement published at the end of 1920 (cf. Jl. Dept. Agric.t Union S. Africa, Pretoria, i, no. 9, p. 799, December 1920) owing to the serious losses of stock from trypanosomiasis experienced by recent settlers, arrangements were then being made by the Government of the Union of South Africa to carry out a special investigation of Tsetse-flies in the Empangeni area of Zululand. 2 London : British Museum (Natural History). — Obtainable from the Director, price 3d. 3 Should a ready-made net not be available, the reader can easily make a substitute for himself in either of the following ways, as advised by Prof. G. H. Nuttall, F.R.S., of the Quick Laboratory, Cambridge. (a) Bend a piece of strong wire (such as telegraph wire) into a circle about 12 in. in diameter, and bind the two ends to the end of a strong stick. To the ring thus formed attach a bag of muslin or fine mosquito netting, about 2 ft. in length ; the bottom of the bag should not be brought to a point, but should end bluntly, and the edges of the material should be se\\ n together on the outside, so as not to make an internal hem which would afford a lodgment for small insects. (b) To make a stronger and more practical type of net, proceed as follows. Procure a Y-shaped tube, with arms of smaller calibre than the stem, and bend the arms until they are at right angles to the iatter. Insert a flexible cane into one of the arms, and a strong stick into the stem, to serve as a handle. Bend the cane into a circle, and attach a net made as in (a) , but having the upper edge turned over to form a tube through which the cane is passed ; finally insert the free end of the cane into the other arm of the Y. A net of this kind is both light and cheap. 153 fitting stopper, with a layer of dry plaster of Paris to the depth of half an inch ; pour in above this a layer, equal in depth, consisting of crushed or powdered cyanide of potassium, mixed with rather more than its bulk of dry plaster of Paris ; cover this mixture with a layer of dry plaster of Paris to the depth of a quarter of an inch or so ; and pour in above the whole a layer, half an inch in depth, consisting of plaster of Paris mixed with water to the consistency of cream. So soon as the top layer of plaster is dry the killing-bottle is ready for use : the plaster, however, should be covered with several thicknesses of blotting-paper, to avoid risk of injury to specimens in case the surface should at any time become wet. The killing-bottle should never be allowed to remain open. Should the surface of the plaster become brown and lose its odour, the efficiency of the killing-bottle may be restored by slightly scratching the upper layer. Prolonged exposure to the fumes of cyanide is injurious to specimens, which should therefore be removed from the bottle so soon as they are dead ; if the killing-bottle is in proper working order it should not be necessary to allow Tsetse-flies to remain in it for more than five minutes at most. Glass-bottomed cardboard pill-boxes (obtainable in assorted sizes, up to 2J in. in diameter, packed in nests one inside another) should be protected from probable deterioration during the wet season by being first painted over with liquid glue, and then covered with strips of jaconet, cut on the cross. Entomological forceps are provided with curved ends for holding pins. Fine- (straight-) pointed forceps, preferably made of brass with very little spring, are also useful for arranging the legs and wings of Tsetse-flies when pinned. Needles mounted in handles may, if desired, be employed in place of fine-pointed forceps. Entomological pins made of silver wire have the great advantage of never becoming corroded, but, owing to their softness, need to be used with special care ; for Tsetse-flies the most suitable sizes are " 0 " and "3," costing in each case about 8s. 6d. per 1,000. Of entomological pins of the ordinary type, the most convenient sizes for pinning Tsetse- flies are No. 20 (a very fine, short pin, which, to prevent it from doubling up, should be held by the forceps near the point) and No. 7 (which „ though fine is considerably stouter and also longer than No. 20, of which an ounce contains about 3,750 pins). Both of these sizes, costing respectively 9s. 6d. and 3s. 3d. per ounce, are obtainable to order from the makers (Messrs. D. F. Tayler and Co., New Hall Works, Birmingham) in ounce or quarter-ounce boxes, as preferred. Two or three sheets of cork-carpet or pith, about 6 in. square, are useful as platforms on which to perform the operations of pinning, etc. As regards cork-lined entomological store-boxes, to contain specimens when pinned, perhaps the most convenient size for Tsetse-flies is the " postal " box measuring about 7 in. by 3| in. The platyscopic lens (obtainable, at a cost of about 22s., from either Messrs. Charles Baker, 244, High Holborn, London, W.C.I, or John Browning, 63, Strand, London, W.C.2) should have a magnifying power of from six to eight diameters. CATCHING TSETSE-FLIES. — Under this and some of the following headings extracts are given from the British Museum (Natural History) pamphlet " Instructions for Collectors : No. 7," to which reference has already been made. 154 Tsetse-flies, like all other Diptera, should, if possible, " always be brought home alive in the glass-bottomed pill-boxes1 .... to which they are to be transferred on being captured in the net, and should then be killed in the cyanide bottle or jar immediately before being pinned. As soon as a fly is taken in the net by a dexterous sweep, a sharp turn of the wrist must be given (following a smart downward or lateral stroke in order to bring the fly to the end of the net) in such a way that the end of the net containing the insect falls over the rim and so makes a closed bag from which it cannot escape. The end of the net can then be gathered up in the hand and the fly forced into a still smaller space, in which it will not be difficult to get it into a pill-box .... and then to slip on the lid .... Flies may also be transferred direct from the net to the killing-bottle, and so brought home dead ; but this method is not to be recommended, since prolonged exposure to the effects of cyanide of potassium is apt to injure the specimens, the ultimate condition of which, when thus treated, is rarely so satisfactory as if they had been brought home alive in pill-boxes. If, however, it is necessary for any reason to dispense with pill-boxes .... and to use the killing-bottle in the open, a little crumpled tissue-paper should be placed inside it ; this affords a lodgement for the specimens, and so lessens the risk of their being injured by rolling about .... It is always advisable when out collecting to carry a killing-bottle for use in case of need, in the event of the supply of pill-boxes .... running short." METHOD OF COLLECTING TSETSE-FLIES. — When it is desired to study the seasonal distribution of the Tsetse-flies in a given region, to discover what species of Glossina occur there or to learn the proportion of the sexes, or, again, when the object is to catch living females to be used in laboratory experiments, it is necessary to engage a certain number of natives, whom it is customary to term "fly-boys," and, after having trained them to recognise and catch these insects, to send them out as collectors into the different districts. Writing on this subject in connection with the Ilorin Province of Northern Nigeria, Dr. J. W. Scott Macfie observes (94) : — " Of the large number of natives who passed through my hands most were found to be of little use, because, although they were able to catch specimens where they were plentiful, they lacked the perseverance, or some peculiar personal allurement, necessary to procure them in less likely spots. Eventually, however, I was able to select three or four who showed some aptitude and a degree of discrimination, and by them were made most of the collections on which this report is based. I was also fortunate enough to obtain the co-operation of the Adminis- trative Officers, who interested themselves in the work and not only furnished me with valuable information, but also consented to take collectors with them when on tour in their divisions, thus enabling me to obtain insects even from the most remote corners of the province." In this way Dr. Macfie was able to obtain no fewer than 612 collec- tions, " from more than 500 different localities," of which he gives an alphabetical list at the end of his paper, besides so far as possible showing them on a map by means of Dr. J. J. Simpson's system of conventional signs (see below, p. 162). " The list," writes Dr. Macfie, 1 The lids of these should be pierced with a few small holes, to admit air. 155 "' has, of course, a positive value only ; and no doubt in many places the flies were overlooked, either because they were present in small numbers, or because the conditions were unfavourable — the day may have been wet or cloudy, or the locality may have been visited in the early morning before the flies were abroad." " My own rule," adds this author, " and the instructions I gave to my collectors, was to stop at every stream or closely wooded spot for half an hour. If in this time no tsetse-flies were seen, it was concluded either that they were not present, or that they were present in such small numbers that, for all practical purposes, they were not a menace to cattle or transport animals." KILLING TSETSE-FLIES AFTER BEING CAUGHT. — Tsetse-flies " brought back alive after a day's collecting should be killed by being placed for a few minutes in the closed cyanide bottle or jar. If the cyanide jar is sufficiently large, the pill-boxes themselves may be placed therein, first opening them a fraction of an inch on one side to allow the cyanide to take effect Diptera should never be allowed to remain exposed to the effects of the cyanide longer than is necessary to ensure their being quite dead. If the poison is of reasonable strength, four or five minutes should be sufficient to kill even the largest and strongest flies ; on the other hand it is necessary to make sure that the insects are really dead, since, if the exposure to cyanide has been too short, flies will often appear to be dead when in reality they are only stupefied. As soon as the insects are really dead, they should be removed from the killing- jar ; those in pill-boxes may be allowed to remain therein until one is ready to pin them, while the loose specimens should be turned out on to a sheet of cork-carpet or pith." PINNING TSETSE-FLIES ; THEIR TRANSMISSION TO ENGLAND. — Except when intended for dissection, Tsetse-flies should invariably be pinned, and this should be done as soon as they are dead. The operation of pinning is quite simple, and does not take long, but attention should be paid to the following points : — The specimen must never be allowed to remain close to the tip of the pin, but should be drawn two-thirds of the way up the latter, which should then be handled by grasping it with the entomological forceps below the fly. Wings and legs must be arranged, so far as possible, by means of the mounted needles or fine-pointed forceps ; wings must not be left closed over the body in such a way as to conceal the upper surface of the abdomen, nor must the legs be permitted to remain tightly flexed beneath it. Specimens must be pinned through the thorax. If but a single example of a species has been obtained, it should be pinned on its side. If two specimens of a species have been caught, pin one on its side and the other in the ordinary way, in the latter case thrusting the pin vertically downwards through the centre of the thorax. For purposes of identification, whenever possible at least six specimens, including examples of both sexes, of a species should be obtained. 156 Before forwarding specimens in a store-box1 through the post, pins- should be tested to see that they are inserted as tightly, as possible, and all should be " driven in to the same level, after which a sheet of soft paper (newspaper does very well) can be fixed into the box," resting on the heads of the pins, in order to minimise damage should a specimen happen to get loose. ' The box containing the specimens, should be well wrapped in cotton-wool or similar material, and firmly packed in an outer case for transmission (by parcel-post if possible and if the package is not too large) to England/' PRESERVATION OF SPECIMENS IN ALCOHOL. — " Unless intended for dissection," writes Austen,2 " Diptera must on no account be placed in spirit. Since all descriptions of Diptera have been based upon dried' specimens, flies in spirit have to be taken out, dried, and pinned before they can be examined and compared with descriptions. In drying, however, they invariably shrivel, while immersion in spirit often causes colour-markings to disappear, and inevitably results in hairs, scales, etc., being washed off or matted." " Pinned specimens," however, " should always, if possible, be accompanied by others preserved in alcohol. The following directions have kindly been supplied by Lieut. - Colonel G. M. Giles, who has found that the method indicated answers well in practice : — ' Flies intended for anatomical examination must be fixed by heat, since their chitinous envelope is penetrated so slowly by alcohol that, if this be not done, they decompose before the pre- servative can reach them. It is sufficient to bring them to boiling temperature in a test-tube of water, but for delicate histological work it is better to boil in a mixture of equal parts of 90 per cent, alcohol (ordinary rectified spirit) and aqueous solution of perchloride of mercury, 1 in 500. Even protozoal parasites are well preserved by this method. When fixed, preserve in 90 per cent, alcohol. Methylated spirit should not be used, since its acidity and other impurities may lead to embarrassment in the use of aniline stains.' The specimens should be preserved in small glass tubes, with a plug of soft paper placed inside on the top of the specimens to prevent them from being injured by washing about in transit. A slip of paper, on which all necessary data should be written clearly in pencil, should be placed inside each tube, and the pinned specimens of the same species should be labelled with a reference to those in spirit, so that the latter may be identified. The tubes containing the specimens should preferably be placed for transmission inside a larger bottle of spirit ; if this be done, it is only necessary to secure the mouth of each tube by tying muslin over it. Cotton -wool, or, better still, horsehair or shavings, should be inserted as packing between and especially on the top of the tubes " PRESERVATION OF LARVAE AND PUPAE. — Larvae and pupae of Tsetse-flies, which it is intended to preserve "should be killed by immersion for a moment or two in boiling water, and should then be placed in weak spirit (two- thirds spirit and one- third water) ; after being allowed to remain in this for two or three weeks until thoroughly hardened, they can be transferred to stronger spirit. The larvae and pupae of each species must, of course, be kept separate, and 1 " Should the collector run out of store-boxes, cigar-boxes, in the bottom of which is fixed a layer of cork-carpet or pith, make efficient substitutes ; but if pith is used, it should not be less than half-an-inch thick." 2 Instructions for Collectors : No. 5. — Diptera (Two-winged Flies). 4th Ed.. P. 8. London: British Museum (Natural History). 1919. 157 should be put up in a small corked glass tube full of spirit, the necessary particulars, with, if possible, a reference to pinned specimens of the perfect insect, being written in pencil on a scrap of paper and placed inside. The tubes should be packed for transmission inside a larger bottle of spirit, as described at the end of the previous paragraph." FIELD NOTES. — Specimens of Tsetse-flies, and of their larvae and pupae, should so far as possible be accompanied by clearly-written notes : — date, hour and place of collection ; animal upon which the fly was feeding ; name of collector ; colour of the living insect ; and an}T other biological detail such as those dealt with in the foregoing pages — distribution ; seasonal abundance ; proportion of the sexes ; migrations ; breeding places, etc., etc. When specimens are pinned, labels should be attached to the pins; in the case of specimens in small boxes, labels should be gummed on the lids, or placed inside. NUMBER OF SPECIMENS OF EACH SPECIES REQUIRED. — " At least half a dozen specimens of each sex of a species should, if possible, always be obtained, and a good look-out should be kept for specimens showing any abnormality in structure, coloration, or size. When the collector remains long enough in one spot, he should always endeavour to obtain specimens of a species on different dates, in such a way as to throw as much light as possible on the duration of its seasonal occurrence. Similarly the attempt should be made to illustrate the geographical range or local distribution of a species, by collecting specimens in as many different localities as possible. Specimens of species taken in new localities, though common elsewhere, will always be valuable." ATTACKS BY ANTS OR TERMITES. — THE EFFECTS OF MOULD. — " In the tropics and hot countries generally, boxes of pinned insects are very liable to the attacks of minute ants, which, if they once gain access to a box unobserved, will soon play havoc with its contents. In Brazil it has been found that ants can be prevented from entering insect-boxes by smearing the outside round the line where the box •opens, or any other possible place of entrance, with andiroba oil, repeating the process as often as necessary ; similar methods might be employed elsewhere in case of need. An even more certain mode of securing immunity from ants is to stand the legs of a table on which insect-boxes are kept in vessels (e.g., tobacco tins) containing water. The table legs must not touch the sides of the vessels, and the top must nowhere be in contact with a wall. To prevent mosquito-breeding, the water in the vessels, which must not be allowed to evaporate, should be kept covered with a thin film of paraffin." In damp tropical countries, another good mode of protecting collections from the attacks of ants and termites consists of the employ- ment of drying cages, suspended from the ceiling of a hut or roof of a tent. A cage of this kind can easily be made out of an empty packing- case, and a few sheets of fine-meshed wire gauze to form the front and back. The cords from which the cage is hung should pass through small calabashes containing oil, or, better still, filled with naphthaline, to prevent destructive insects from descending from the ceiling and attacking the specimens. Such a cage should be capable of being easily taken to pieces, and should be about 20 in. long, 12 in. high and 8 in. broad. " Mould, however, is an even greater enemy to collections of Tsetse- flies. Pinned specimens of Glossina, like those of other insects, rapidly 158 develop mould during the rainy season in tropical countries ; and since mouldy specimens are practically worthless for purposes of scientific determination, Tsetse-flies should always be sent home as soon as possible after being collected. The risk of mould may, however, be entirely prevented by thoroughly rubbing the inside of the box into which the specimens are pinned with a piece of sponge, or swab of cotton-wool, soaked in a saturated solution of naphthaline in chloroform, mixed with an equal quantity of ordinary medical (beechwood) creosote. The inside of the box should be rubbed with the mixture until the cork- lining shows through the paper. This method possesses the additional advantage of keeping away destructive insects. The fluid must not be allowed to touch the specimens, or they will be injured." CHAPTER XVI. KEEPING AND BREEDING TSETSE-FLIES IN CAPTIVITY. In the majority of the experiments to which reference has been made in this work, the receptacles employed for keeping Tsetse-flies alive were either wide-mouthed glass tubes, 5 in. in length by 2 in. in diameter, closed by a piece of muslin secured by an elastic band, or glass bottles. As a rule leafy twigs, for the flies to settle upon, were placed inside. In some cases wooden cages, closed in front by muslin, were employed. To study the behaviour of Tsetse-flies towards small animals — mammals, birds and reptiles — host and flies were kept together, in cages surrounded by gauze. In order to keep the flies alive, they must be given a meal of blood at regular intervals — of about a day in the case of G. palpalis and G. morsitans ; in this way it has been found possible to keep these insects for two months and more. According to Dr. Scott Macfie (93), the best way of keeping G. palpalis alive is to allow it to feed from one's fingers. For G. morsitans, small domestic mammals have been made use of, or, failing these, native fowls. As a general rule the method to be followed depends upon the special conditions of the experiment. . At Amani (Tanganyika Territory) Stuhlmann kept G. brevipalpis alive for a considerable time (four months in the case of certain females) by allowing the flies to gorge themselves with blood every four or five days. At the same place hundreds of Tsetse-flies were bred by Drs. E. Teichmann and H. Braun (146) in the manner advised by Kleine. The insects were brought to the laboratory in large glass vessels, covered with mosquito-netting, and having at the bottom a layer of earth in which were small plants. In this manner about 10,000 flies (G. brevipalpis for the most part, but also G. pallidipes and G. austeni) reached Amani alive, although they were a whole day on the way ; few died before arriving at their destination. On arrival, the flies were transferred to smaller jars containing from 5 to 10 specimens apiece, four females, destined exclusively for breeding purposes, being allotted to each male. Every three or four days the insects were fed, by turning the jar upside down and bringing the mosquito-netting into contact with the shaven skin of a warm-blooded animal ; for this purpose goats and sheep are more suitable than rats. The latter have to be stretched on their backs on a board, with their 159 feet tied, and since this causes suffering it is difficult to feed many flies from one of these animals. If a goat or sheep be tied up and hobbled, from six to nine jars can be placed upon it.1 REARING TSETSE-FLIES IN LABORATORIES IN EUROPE. — Some seven years ago the first results of an attempt at rearing Tsetse-flies in a European laboratory were described by Dr. E. Roubaud (1196). At the end of December 1913 Dr. Roubaud brought from Senegal a score of living pupae of G. palpalis and G. morsitans. Cold during the voyage proved fatal to the older pupae, but the younger ones hatched out in the laboratory of the Pasteur Institute, Paris. In the case of G. morsitans the resultant flies, placed in a Roux incubator kept at a mean temperature of 24°-25° C. (75°-77° F.), and fed daily on guinea- pigs and rabbits, lived and bred in a satisfactory manner. The G. palpalis pupae produced only two male flies, which were small but well- formed, and lived from 17th January to 26th February. From the G. morsitans pupae there were obtained two males and six females. One of the latter deposited four larvae, which were the origin of a stock which was still in existence three years later, being kept at a mean temperature of 24° to 25° C. (75°-77° F.), and in a mean humidity of of 50 to 55 per cent. In January 1915, 20 flies were alive, and it is stated by the author in a subsequent paper (121) that, from start to- finish, the experiment continued satisfactory for three years, the strain ultimately dying out naturally owing to an unusual preponderance of males and absence of females. The conclusion drawn by Dr. Roubaud is that it is possible to keep and breed Tsetse-flies in Europe, under conditions permitting of pro- longed experiments upon these insects. If the incubator be small, care must be taken to see that it is properly aerated, and the humidity must not be allowed to exceed 60 per cent. CHAPTER XVII. DISSECTION OF THE SALIVARY GLANDS OF TSETSE-FLIES. On this subject we cannot do better than reproduce the admirably clear and succinct directions given by Mr. LI. Lloyd (87). " Workers on the transmission of trypanosomes by various species of Glossina," writes Mr. Lloyd, " have hitherto been unable to agree as to the part played by the salivary glands in the process. It is of importance, therefore, that some method of removing the glands be adopted in which the risk of contamination by gut-contents would be reduced to a minimum. The usual method employed hitherto has been to remove the entire viscera by means of pressure on the abdomen, after either the removal of the last segment, or the severance of the abdomen from the thorax. These methods are open to the criticism that the gut, especially if it contains much blood, is liable to rupture ; that the glands are liable to lesion ; and that it is very difficult to dissect them free from fat body and gut. 1 Messrs. Baird and Tatlock, of Manchester, make little wooden cages, which are very light, portable and inexpensive. These cages, manufactured to the specification of Sir David Bruce, are of an oblong shape, about 6 in. long, 4 in. wide and 4 in. high ; two sides are covered with muslin or wire gauze, and a hole, fitted with a cork, enables flies to be introduced. There are two types of cages — one (with wire gauze sides) intended for carrying flies, the other (with muslin sides) for feeding purposes. 160 " The complete salivary glands may be obtained without risk of rupture of the gut by the following method. G. morsitans has been used in the dissections. The fly is held firmly in the hand and a longitudinal incision is made in the median dorsal line of the thorax from the neck to the abdomen. The insect is then immersed in normal salt solution and incisions are made along the transverse groove of the thorax from the median incision almost to the bases of the legs. The .strong muscles in the thorax which run in a longitudinal direction should also be severed. A needle is now placed in the anterior end of the longitudinal incision, and another in the posterior end. A gentle longitudinal pull applied to the fly by these needles causes the remainder of the thorax to break across. The alimentary canal breaks between the pharynx and proventriculus, while the salivary glands are drawn out of the abdomen quite free from fat body and with only the finer twigs of the tracheal system adhering. It is now a simple matter to dissect away the remainder of the thorax and head till a preparation is obtained consisting of proboscis, pharynx and salivary glands. The figure shows the appearance of such a preparation. The glands when p.b. Fig. 19. — Preparation of salivary glands and proboscis of Glossina morsitans •(semi-diagrammatic), x 10. s.g., salivary glands ; j.d., junction of salivary ducts ; oe, oesophagus ; ph., pharynx ; p.b., bulb of proboscis ; p.m., muscles of proboscis. (After Lloyd, 87.) first withdrawn are quite transparent and exhibit an active wriggling motion. They correspond exactly with the salivary glands of G. palpalis as described by Minchin.1 If the pull on the fly is not perfectly longitudinal, one of the glands may break off just posterior to the junction of the salivary ducts. Occasionally both the glands break away at this attenuated part of the ducts. In either event the remainder of the thorax attached to the abdomen is dissected off and the protruding glands are withdrawn by means of fine forceps. A little dissection of the anterior end of the abdomen is sometimes necessary to effect this. " Another method that has been adopted with some success is to draw the glands out of the fly by means of the proboscis. A longitudinal median incision is made in the thorax as before and the cut is opened •out by means of needles, with which the structures in the thorax are loosened. The dorsal surface of the head, the eyes and the jowls are 1 E. A. Minchin, Report on the Anatomy of the Tsetse-fly (Glossina palpalis) : Proc. Roy. Soc., London, Ser. B., vol. 76, p. 538 (1905)." 161 then cut through with needles till nothing remains externally connecting the anterior and posterior surfaces of the head except the thin soft integument immediately behind the proboscis. The proboscis is then seized with forceps and a needle is placed on the last segment of the abdomen. A slow pull on the proboscis causes the glands to be with- drawn as before. This method has been found to fail so frequently, however, that it cannot be recommended as a working dissection. With several species of Tabanus this latter method of dissecting the salivary glands has been found to be invariably successful." CHAPTER XVIII. COLOUR CONVENTIONS TO INDICATE THE DISTRIBUTION OF TSETSE-FLIES. The system described and illustrated below was proposed ten years ago by Dr. J. J. Simpson (143), and has since been generally adopted in maps of our Tropical African Colonies and Protectorates showing the ascertained occurrence of different species of Glossina. In introducing his scheme of conventional colours and symbols, Dr. Simpson wrote as follows : — " The necessity for graphic representation of the distribution of insects and the diseases which are transmitted by them is now so evident as to need no special demonstration. But it is further obviously important that some uniform method should be adopted in the preparation of such maps, in order to make them intercomparable, and the object of the present notes is to suggest a comprehensive system of notation for the whole genus Glossina — a system which may be readily adapted for use in other genera or families of blood-sucking insects/' " Symbols. — To achieve their aim these ought not to be excessive in number and should be sufficiently distinctive to be easily recognisable from one another at a glance. After a series of trials the following have been considered most satisfactory : — (1) a solid circle, (2) a solid square, (3) a circle enclosing a cross, (4) a square enclosing a cross, (5) a solid triangle, (6) a hollow triangle, (7) a hollow circle, and (8) a hollow square. Modifications of these could easily be multiplied [see below]. In the present system it is not necessary to adopt any more, as only in one group, namely the palpalis-group of Glossina, in which there are six species, do we require more than four symbols. " Coloiirs — Red has already been assigned to sleeping sickness, and it seems most advisable to retain its use for this disease. By means of the symbols already given it is easily possible to represent other diseases, if necessary, by the same colour." " The following colours have been selected after a series of tests as being sufficiently distinct to avoid unnecessary confusion and have been assigned to the various groups as follows : — palpalis-group — blue ; morsitans-group — green ; fusca-group — brown ; brevipalpis-group — orange . " Yellow has been purposely avoided as it is never very distinct, while in artificial light it is hardly recognisable." (5979) L 162 A symbol in black is suggested by Dr. Simpson " to indicate the presence of undetermined species of Glossina." The following, then, is the scheme : — Sleeping Sickness . . . . . . • (red) . Glossina. G. palpalis Group. G. palpalis G. palpalis var. fuscipes . . G. tachinoides G. caliginea . . . . G. pallicera G. austeni . . G. morsitans Group. G. morsitans . . G. morsitans var. submorsitans G. longipalpis G. pallidipes G. fusca Group. G. fusca . . . . G. tabani/ormis G. nigrofusca G. fuscipleuris G. brevipalpis Group. G. brevipalpis G. longipennis G. medicorum Undetermined species Symbols for additional species. . (blue). m ( m ( A ( (green) • (orange). • ( „ e ( „ ) 0 (black). O D © El A A CHAPTER XIX. SPECIAL POINTS FOR CONSIDERATION IN STUDYING THE BIONOMICS 01 ANY SPECIES OF Glossina. — SUGGESTIONS FOR FUTURE WORK. With a view to obtaining more accurate knowledge on the subj< of Glossina morsitans — the special enemy of the farmer and stock-rais in many parts of Rhodesia and elsewhere— the Glossina Sub-Committ( of the Entomological Research Committee (now the Imperial Burea\ of Entomology) drew up and disseminated in 1910 a series of carefulh considered questions (168), to which it was hoped to obtain precis answers from all those competent and willing to supply them. Thougl formulated more than a decade ago, this detailed interrogatory, witl modifications and additions suggested by facts ascertained in the interim, may usefully serve as a guide for future investigations upoi other species of Tsetse-flies besides G. morsitans, and in other parts < Africa. It is therefore reproduced below, together with the directioi by which the questions were accompanied. 163 " Those who answer these questions should observe the following precautions : — The answers should be concise and to the point, and should be distinguished by numbers corresponding with those of the questions to which they refer. Names and places which can be verified, and dates, should be given. Any information beyond the scope of the questions should be added in an Appendix. If the reporter is in doubt as to the species of tsetse, or whether more than one is present, specimens should be sent with the replies. He should in no case rely on native evidence. The longer the reporter has been resident in or near the fly-area the more weight will his evidence carry. The answers should be written on one side of the paper only. 1 . Are you resident in, or near, a fly-area ? In the latter case how far are you distant from the nearest point of the area ? 2. How long have you been resident in the vicinity ? 3. Can you state approximately the size of the fly-area ? (Give a sketch-map, if possible.) 4. Has this area appreciably increased or diminished in your own experience ? 5. If so, can you state the date at which these alterations took place, and the nature of the climatic conditions at the time ? 6. Was there any appreciable increase or diminution of wild animals about that time ? 7. Does the fly-area contain Glossina morsitans only, or have you reason to believe that another species of tsetse is present as well ? 8. Does the size of the fly-area vary at different times of the year, and, if so, is there a definite relation between these variations and the dry and rainy seasons respectively or the prevalence of wind ? 9. Are the limits of the fly-belt sharply denned, and, if so, what is the nature of the limiting boundaries ? 10. Are there any natural features which are specially characteristic of the fly-belt and which are not to be found in the surrounding fly-free areas (e.g., nature of soil, bush, herbage, presence or absence of water) ? 11. Are there any such features which you have specially noticed to be present in every fly-belt you have visited ? E.g., Does the presence of fly in large numbers seem to be associated with that of sand ? 12. Have you ever observed a tsetse-fly depositing its larva, and, if so, can you describe the circumstances ? 13. Have you observed any marked differences in the number of tsetses in a particular area at different seasons ? If so, describe these differences. 14. To what do you attribute such differences ? 15. Have you made any attempts to trap tsetses with pieces of black cloth or paper worn by men or animals and smeared with bird-lime, and, if so, with what success ? Have you tried any method of deterring the flies from biting animals while passing through a fly-belt ? 16. Do you know of any mammals, birds, reptiles, or predaceous or parasitic insects which actually destroy tsetse-flies ? 17. Are there, within your knowledge, any facts which lead you to believe that tsetse-flies are entirely dependent for their continued existence upon the blood of wild mammals ? 18. Have you any reason to believe that tsetses are specially dependent upon and associated with any particular species of mammals ? 19. Have you ever observed tsetses feeding upon birds, lizards or crocodiles ? 164 20. Have you any reason to suppose that tsetses may be capable of subsisting on the juices of plants or that they suck up water or dew ? 21. Do you personally know any facts tending to show that tsetses move about with big game either in the natural migrations of the latter or when they are driven away, and return when or soon after the game returns ? 22. Are you acquainted with any instance in which the appearance of tsetses in a new locality has been due to the movement of natives ? 23. Do you know personally of any cases of healthy domesticated animals of any kind living at a Kafir kraal within a fly-belt ? " To the foregoing, the following supplementary questions may be added : — 24. Have you found Tsetse-fly breeding places in the fly-belt studied ? What is the nature of the soil and vegetation in the places where you have found pupae ? What is the position of these spots in relation to water, shelter, etc., and what is their orientation ? 25. In the fly-belt, what is the ratio of the sexes at different seasons of the year ? Is this ratio a real one, or is it due to the fact that the females are less easy to catch or that they remain concealed ? 26. Have you ever observed a special migratory movement of females, tending to pass from localities favourable to reproduction to others less favourable ? 27. Have you observed that external agencies — cold, heat, humidity, solar radiation — have any influence whatever upon the activity and length of life of Tsetse-flies of either sex ? • 28. Have you observed the presence of parasites in Tsetse pupae ? In the pupae collected by you, what proportion was parasitised ? To what species did the parasite belong ? FURTHER SUGGESTIONS FOR FUTURE WORK. — This little volume may fittingly conclude with a brief reference to the " Report of the Glossina Sub-Committee of the Imperial Bureau of Entomology " (171), published about two years ago. Attention is drawn in this report to the necessity for further know- ledge of the life-histories of the various species of Tsetse-flies, as well as of their habits and the factors that favour their increase or inhibit their spread, before the wide areas over which they range can be reduced. The measures hitherto recommended for the eradication of Glossina cannot be carried out effectively over large stretches of country, and will only prove useful in restricted portions of such areas if persistently maintained over a considerable time. An essential line of investigation will be a thorough survey in each infested locality. The surveys will require checking at intervals throughout the year, and should include estimates, on a standard basis, of the density of the fly. The points suggested for observation and experiments are distribution of the flies ; effects of clearing ; breeding grounds ; food of the flies ; their parasites and other enemies ; destruction of adults ; and influence of odours. Zululand, Southern Rhodesia, Nyasaland, Tanganyika Territory, the Sudan and Northern Nigeria are suggested as the most suitable countries for the establish- ment of experimental stations to investigate these problems. The precise areas selected in these countries should represent different types of environment, and should, so far as possible, be situated in localities in which the presence of Tsetse-flies is actually obstructing settlement, or is likely to do so in the near future. Suggestions are made for the organisation of the staff for these stations, which, as the work progresses, will necessarily include a protozoologist. 165 CHAPTER XX. BIBLIOGRAPHY OF THE PRINCIPAL PUBLICATIONS CONSULTED. ANDERSON, T. J. : la. Insects affecting Stock : Brit. E. Africa Dept. Agric. Ann. Kept. 1917-1918, Nairobi, pp. 89-93 (1921). AUSTEN, Major E. E. : 1. A Monograph of the Tsetse-Flies : pp. ix -f 319, 9 plates, map, 16 figs, in text. London: British Museum (Natural History) (1903). 2. A new Species of Tsetse-Fly allied to Glossina palpalis, Rob.- Desv. : Bull. Entom. Res., London, Vol. I, Part 4, pp. 294-297, figs. 1-3 (January 1911). 3. A Handbook of the Tsetse-Flies (Genus Glossina) : pp. x -f 110, 10 coloured plates, map, 24 figs, in text. London : British Museum (Natural History). Price 5s. 6d. (1911). 4. A Dipterous Parasite of Glossina morsitans : Bull. Entom. Res., London, Vol. V, Part I, pp. 91-93, fig. 1 (April 1914). AUSTEN, E. E. and BAGSHAWE, A. G. : 5. Suggestions for Entomological Research in Connexion with Sleeping-Sickness : Minutes of Evidence taken by the Depart. Cttee. on Sleeping Sickness, London : Appendix D, p. 290 (1914). BAKER, A. W. : 6. Preliminary Notes on the Use of Repellents for Horn Flies and Stable Flies on Cattle : 47th Ann. Rept. Entom. Soc. Ontario for 1916, Toronto, pp. 52-56 (1917). BALFOUR, Dr. A., C.B., C.M.G. : 7. War against Tropical Disease : London, Wellcome Bureau of Scientific Research, and Bailliere, Tindall and Cox, 220 pp., 180 figs., 2 graphs in wallet. Price 12s. 6d. net (1920). BARKER, C. N. : 8. Some Records of Predaceous Insects and their Prey in the Durban Museum : Ann. Durban Museum, II, No. 2, pp. 94-96 (30th July 1918). BARRETT, H. T. : 9. Diary Notes on the Tsetse-Flies observed during a Journey down the Rovuma River, Portuguese East Africa : Bull. Entom. Res., London, Vol. II, Part I, pp. 43-46, and map (May 1911). BEVAN, L. E. W. : 10. Report of the Government Veterinary Bacteriologist for the Year 1916, Southern Rhodesia : Trop. Vet. Bull., London, Vol. VI, No. 1, pp. 48-63 (30th March 1918). BOUET, Dr. G. : 11. Les trypanosomiases et le gros gibier en Afrique occidentale frangaise : Bull. Soc. Nat. d'Acclim. de France, Paris, No. 24 (15th December 1913) ; 12. Contribution a 1'fitude des Zones a Glossines du Sendgal (Region du Chemin de Fer de Thies a Kayes) : Bull. Soc. Path, exot., Paris, T. IX, No. 10, pp. 802-813 (1916). BOUET, Dr. G. and ROUBAUD, Dr. E. : 13. Repartition des Glossines a la Cote d'lvoire : Bull. Soc. Path, exot., Paris, T. X, No. 1, pp. 37-39 (10th January 1917). 166 BOUILLEZ, M. : 14. Contribution a, 1'etude et a la repartition de quelques affections parasitaires au Moyen-Chari (Afrique centrale) : Bull Soc. Path, exot., Paris, T. IX, No. 3, pp. 143-167, 1 map (March 1916). BRANDT, FR. : 15. Report of the Veterinary Department for the Year 1918 : Ann. Rept. Agric. Dept. Northern Provinces, Nigeria, 1918, Lagos, pp. 7-11 (29th May 1919). BRITISH SOUTH AFRICA COY. : 16. Tsetse-Fly and Big Game in S. Rhodesia. — Correspondence with Colonial Office : Rev. Appl. Entom., London, Ser. B, Vol. IV, Part 5, pp. 77-78 (May 1916) ; 17. Tsetse-Fly in Southern Rhodesia. — Correspondence with Colonial Office, 29th October 1918, 8th November 1918, and 2nd January 1919 : Rev. Appl. Entom., London, Ser. B, Vol. VII, Part 2, pp. 36-37 (February 1919). BRITISH SOUTH AFRICA, DEPT. OF AGRICULTURE, SALISBURY : 18. Report of the Government Entomologist on the Spread of the Tsetse-Fly and Trypanosomiasis in the Wankie District. — Salisbury, 23rd June 1918: Rev: Appl. Ent., London, Ser. B, Vol. VII, Part 1, pp. 9-10 (January 1919). BRUCE, Sir DAVID : 19. Camping in the Tropics : Trans. Soc. Trop.Med. Hyg., London, X, No. 8, pp. 199-206, 6 figs. (July 1917). BRUCE, Surgeon-General Sir DAVID, C.B., F.R.S., A. M.S., HAMERTON, Major A. E., D.S.O., R.A.M.C., WATSON, Captain D. P., R.A.M.C., and Lady BRUCE, R.R.C. : 20. The Food of Glossina morsitans : Proc. Roy. Soc., London, B, Vol. 88, pp. 41-42 (6th August 1914). BRUMPT, Dr. E. : 21. Sur une nouvelle mouche Tsetse la Glossina decorsei n. sp. provenant de 1'Afrique centrale : C.-R. Soc. de Biologie, Paris, T. Ivi, p. 628 (1904) ; 22. Precis de Parasitologie, 2e ed., pp. 1012, 698 figs., 4 plates : Paris, Masson et Cie. (1913). CAMPBELL, Dr. CHAS. A. Q. : 23. The Eradication of Mosquitoes by the Cultivation of Bats ; Comm. to the Intern. Inst. of Agric., Rome : Monthly Bull, of Agric. Intell. and Plant Diseases, IV, No. 8 (August 1913). CAMPION, H. : 24. Some Dragonflies and their Prey. — ii. With Remarks on the Identity of the Species of Orthetrum involved : Ann. and Mag. Nat. Hist., London, Ser. 9, viii, No. 45, pp. 240-245, 3 figs. (September 1921). CARPENTER, Dr. G. D. H. : 25. Second Report on Bionomics of Glossina fuscipes (palpalis) of Uganda : Repts. Sleeping Sickness Comm. Roy. Soc., London, No. xiv, pp. 1-37 (1913) ; 26. Third, Fourth and Fifth Reports on the Bionomics of Glossina palpalis on Lake Victoria : Repts. Sleeping Sickness Comm. Roy. Soc., London, No. 17, 101 pp., 4 figs., 6 charts, 1 map (1919). 167 CARTER, Capt. R. M., I.M.S. : 26a. Tsetse-Fly in Arabia : Brit. Med. Jl., London, 17th November 1906, p. 1393 (1906). CASTELLANI, Dr. A. and CHALMERS, Dr. A. J. : 27. Manual of Tropical Medicine, 3rd Ed. London : Bailliere, Tindall and Cox (1919). CHAMBERS, F. : 28. Note on the Transmission of Animal Trypanosomiasis in Northern Rhodesia by Blood-sucking Flies other than Glossina : Trop. Vet. Bull, London, V, No. 4, p. 222 (1918). [Abstract from Vet. Bull., London, i, No. 3, pp. 222-227 (August 1917).] CHRISTY, C. : 29. Tsetse-Flies and Fly-Belts : Ann. Trop. Med. Parasit., Liverpool, Vol. XI, No. 3, pp. 279-282 (31st January 1918). CHUBB, E. C. : 30. Note on the New Zululand Tsetse-Fly, Glossina brandoni, Chubb : Ann. Durban Mus., Durban, I, No. 3, p. 253 (April 1916). COCKERELL, T. D. A. : 31. A Fossil Tsetse-Fly in Colorado : Nature, London, No. 1973, Vol. 76, p. 414 (22nd August 1907) ; 3la. Fossil Insects from Florissant, Colorado : Bull Amer. Mus. Nat. Hist., New York, xxiv, p. 65 (1908) ; 316. Another Fossil Tsetse-Fly : Nature, London, No. 2057, Vol. 80, p. 128 (1st April 1909) ; 32. New Species of North American Fossil Beetles, Cockroaches and Tsetse Flies: Proc. U.S. Nat. Mus., Washington D.C., Vol. 54, No. 2237, pp. 301-311, 2 plates, 5 figs. (July 1918). CROSS, H. E. : 33. Annual Report of the Camel Specialist for the Year 1915-16 : pp. 27. Lahore (1916) ; 34. Experiments with Emulsions for protecting Camels against the Attacks of Blood-sucking Flies : Agric. Res. Inst. Pusa, Calcutta, Bull. No. 76, pp. 12 (1917). CROVERI, Dr. PAOLO : 35. Osservazioni sulla Biologia della Glossina pallidipes della Somalia Italiana e sulla Trasmissione agli Animali domestici delle Tripanosi detta " Ghendi " : Ann. d' Igiene, Rome, XXIX, No. 1, pp. 432-447 (31st July 1919). DA COSTA, Surgeon Captain B. F. B. : 36. Sleeping Sickness in the Island of Principe : Sanitation Statistics, Hospital Services, and Work of Official Conservancy Brigade. — Translated by Lieut. -Colonel J. A. Wyllie, F.R.G.S., Indian Army (Retired). London : Bailliere, Tindall and Cox. Pp. viii + 90, 3 plates, 4 schedules (1913) ; 37. Report on the Sleeping Sickness Mission in Principe, October and November 1913, embodied in Report of Consul-General Hall to Sir Edward (now Viscount) Grey (1914) ; 38. Reports on the Sleeping Sickness Mission in Principe, December 1913, to March 1914, and April to June 1914, embodied in Report of Consul-General Hall to Sir Edward (now Viscount) Grey (1914) ; 168 39. Sleeping Sickness in the Island of Principe. — Dispatch from Consul-General Hall to Sir Edward (now Viscount) Grey, dated 19th April 1915 (1915) ; 40. [With Sant' Anna (J. F.), dos Santos (A. C.), and de Araujo Alvares (M. G.).] Sleeping Sickness: A Record of Four Years' War against it in the Island of Principe. — Translated by Lieut. -Colonel J. A. Wyllie. London : Bailliere, Tindall and Cox. Pp. xii -f 261*, 68 illustrations, 7 plates, 27 tables, 17 charts (1916). DA COSTA, B. F. B., SANT' ANNA, J. F., DOS SANTOS, A. C. and DE ARAUJO ALVARES, M. G. : 41. See (40) above. DAVEY, Dr. J. B. : 42. Notes on the Habits of Glossina fusca : Bull. Entom. Res., London, Vol. I, Part 2, pp. 143-146 (July 1910). DE GREEF, G. : 43. Monographic agricole de la region de FUrundi (ancienne province de 1'Afrique orientale allemande) : Chap. VII, Etat sanitaire des troupeaux : Bull, agric. du Congo Beige, Bruxelles, Vol. X, No. 1-4, pp. 65-69 (March-December 1919). Di DOMIZIO, G. : 44. Una Triponosomiasi del Dromedario eritreo (Gudho). Cenni sulle Mosche ematofage della Colonia Eritrea : Clinica Veterinaria, Milano, No. 16-17, pp. 391-413 (31st August-15th September 1918). DOLLMAN, H. C. : 45. Glossina morsitans, Westw. Some Notes on the Parasitisation of its Pupae : Trans. Entom. Soc. London, London, Nos. 3-4, 1915, pp. 394-396, Plate LXIV (2nd June 1916). DRY, F. W. : 45a. Trypanosomiasis in the Absence of Tsetses, and a Human Disease possibly Carried by Simulium in Kenya Colony : Bull. Entom. Res., London, XII, Part 3, pp. 233-238, Plates VII &VIII (November 1921). DUKE, Dr. H. L. : 46. Trypanosomiasis in Northern Uganda : Jl. of Hygiene, London, Vol. XV, No. 3, p. 372 (September 1916) ; 47. Some Observations on the Bionomics of Glossina palpalis on the Islands of Victoria Nyanza : Bull. Entom. Res., London, Vol. IX, Part 3, pp. 263-270 (March 1919) ; 48. Tsetse-Flies and Trypanosomiasis. — Some Questions suggested by the later History of the Sleeping-Sickness Epidemic in Uganda Protectorate : Parasitology, Cambridge, Vol. XI, No. 3-4, pp. 415-429 (October 1919) ; 49. An Enquiry into the Relations of Glossina morsitans and Ungulate Game, with special Reference to Rinderpest : Bull. Entom. Res., London, Vol. X, Part I, pp. 7-20, 2 charts (November 1919). EMINSON, R. A. F. : 50. Observations on Glossina morsitans in Northern Rhodesia : Bull. Entom. Res... London, Vol. V, Part 4, pp. 381-382 (March 1915). 169 EVANS, A. M. : 51. On the Genital Armature of the Female Tsetse-Flies (Glossina) Ann. Trop. Med. Parasit., Liverpool, Vol. XIII, No. 1, pp. 31-56, figs. 1-18 (12th May 1919). FELL, T. E. : 52. Notes on Tsetse-Flies and on Prophylactic Measures against Sleeping Sickness in the Western Province of Ashanti : Bull. Entom. Res., London, Vol. Ill, Part 3, pp. 227-231 (November 1912). FISKE, W. F. : 53. The Bionomics of Glossina : a Review with Hypothetical Con- clusions : Bull. Entom. Res., London, Vol. IV, Part 2, pp. 95-111 (September 1913) ; 54. Investigations into the Bionomics of Glossina palpalis : Bull. Entom. Res., London, Vol. X, Part 4, pp. 347-463 (May 1920). GERARD, P. : 55. Note sur la Distribution geographique du Genre Glossina dans la Region miniere de la Lukuga superieure : Rev. Zool. Afric., Bruxelles, T. VII, No. 3, pp. 229-235, 1 map (15th November 1919- 15th January 1920). GREGGIO, G. : 56. Trypanose des Pores ; Relations des Pores avec la Trypanose humaine dans la Vallee de 1'Inkissi (Moyen Congo Beige) : Bull, Soc. Path, exot., Paris, T. X, No. 2, pp. 113-117 (14th February 1917) ; 57. Note sur la lutte contre la trypanose a Kisantu (Congo Beige). Resultats et esperances : Bull Soc. Path, exot., Paris, T. X, No. 5, pp. 398-406 (9th May 1917) ; 58. A propos de la Trypanose des Pores dans la Vallee de 1'Inkissi : Bull, agric. Congo Belse, London, Vol. VIII, No 1-2, pp. 148-155 (March- June 1917). GRAYBILL, H. W. : 59. Repellents for Protecting Animals from the Attacks of Flies : U.S. Dept. of Agric., Bull. No. 131. Contrib. from the Bur. of Entom., L. 0. Howard, Chief. Washington : 10th September 1914. GRUNBERG, Dr. K. : 60. Die Blutsaugende Dipteren : pp. 188, 126 figs. Jena, G. Fischer (1907) ; SOa. Eine neue Tsetse-Fliege aus Kamerun : Sitzgsber. Ges. Naturf. Freunde zu Berlin. Jahrg. 1912, No. 4, p 246 (April 1912). HALL, P. E. : 61. Notes on the Movements of Glossina morsitans in the Lundazi District, North-Eastern Rhodesia : Bull Entom. Res., London, Vol. I, Part 3, pp. 183-184, and sketch-map (October 1910). HAMILTON, Major J. STEVENSON : 62. The Relation between Game and Tsetse-Flies : Bull Entom. Res., London, Vol. II, Part 2, pp. 113-118 (July 1911). HARRIS, R. H. : 62a. Departmental Activities : Entomology : Journ. Dept. Agric. Union S. Africa, Pretoria, Vol. Ill, No. 3, pp. 208-210 (September 1921). 170 HORNBY, H.E. : 63. Transmission of Cattle Trypanosomes by Flies other than Tsetse : Rhodesia Agric. JL, Salisbury, Vol. XIV, No. 2, pp. 168-176, 1 plate (April 1917) ; 64. Tsetse-Fly Disease in Domestic and Wild Animals : Rhodesia Agric. JL, Salisburv, Vol. XVI, No. 6, pp. 493-500, 3 figs. (December 1919). " HOWARD, C. W. : 65. The Distribution of Tsetse-Flies in the Province of Mozambique, East Africa : Bull. Entom. Res., London, Vol. II, Part 1, pp. 39-42, and map (May 1911). JACK, RUPERT W. : ,66. Preliminary Notes on the Habits of the Common Tsetse : Rhodesia Agric. Journal, Salisbury, Vol. IX, No. 1, pp. 60-72, 4 plates. (October 1911) ; 67. Observations on the Breeding Haunts of Glossina morsitans : Bull. Entom. Res., London, Vol. II, Part 4, pp. 357-361, Plates XVII- XXI (January 1912) ; 68. Tsetse-Fly and Big Game in Southern Rhodesia : Bull. Entom. Res., London, Vol. V, Part 2, pp. 97-110, maps I and II (September 1914) ; 69. Tsetse-Fly Investigations, Sebungwe, August-September 1916 : Rev. Appl. Ent., London, Ser. B, Vol. V. p. 26 (February 1917) ; 70. Natural Transmission of Trypanosomiasis (T. pecorum Group) in the Absence of Tsetse-Fly : Bull. Entom. Res., London, Vol. VIII. Part 1, pp. 35-41, 2 maps (August 1917) ; 71. Tsetse Fly Investigations : Visit to Melsetter District and Portuguese East Africa : Rev. Appl. Ent., London, Ser. B, Vol. VI, pp. 134-135 (July 1918) ; 72. Tsetse-Fly in Southern Rhodesia, 1918 : Rhodesia Agric. JL, Salisbury, Vol. XV, No. 5, pp. 406-415, 2 plates (October 1918) ; 73. Operations against Tsetse-Fly in Southern Rhodesia : Rhodesia Agric. JL, Salisbury, Vol. XVI, No. 4, August 1919, pp. 292-298, 1 map (August 1919) ; 74. Tsetse Fly in Southern Rhodesia, 1918 : Bull. Entom. Res., London, Vol. X, Part 1, pp. 71-90, 3 plates, 3 maps (November 1919) ; 75. Some Notes and Remarks on the Bionomics of Glossina morsi- ians: Bull. Entom. Res., London, Vol. XI, Part 2, pp. 113-132, 2 sketch-maps (September 1920). JAMOT, E. : 76. Essai de Prophylaxie medicale de la Maladie du Sommeil dans rOubangui-Chari : Bull. Soc. Path. Exot., Paris, T. XIII, No. 5, pp. 343-376 (12th May 1920). JOHNSTON, Dr. J. E. L. : 77. A Summary of Entomological Survey of Kaduna District, Northern Nigeria : Bull. Entom. Res., London, Vol. VII, Part 1, pp. 19-28, 2 sketch-maps (May 1916). KING, H. H. : 78. Observations on the Occurrence of Glossina in the Mongalla Province of the Anglo-Egyptian Sudan : Bull. Entom. Res., London, Vol. Ill, Part 1, pp. 89-93, and map (May 1912). 171 KINGHORN, Dr. ALLAN : 79. Notes on the Preliminary Stages of Glossina morsitans, Westw. : Bull. Entom. Res., London, Vol. II, Part 4, pp. 291-295 (January 1912). KLEINE, F. K. : 80. Die Schlafkrankheit in Kamerun : Arch. f. Schiffs- u. Trop.- Hyg., Leipsic, Bd. XXIII, No. 15, pp. 315-331, 1 map (July 1919). KOCH, H. : 81. Bericht iiber einen Versuch Glossina palpalis durch Fang zu beseitigen : Arch.f. Schiffs- u. Trop.-Hyg., Leipsic, Bd. XVIII, No. 24 .(December 1914). LAMBORN, W. A. : 82. A Preliminary Report on the Problem of Controlling Glossina in .Nyasaland: Bull. Entom. Res.f London, Vol. VI, Part 1, pp. 59-65 .(June 1915) ; 83. Second Report on Glossina Investigations in Nyasaland : Bull. Entom. Res., London, Vol. VI, Part 3, pp. 249-265, 3 plates (December 1915) ; 84. Third Report on Glossina Investigations in Nyasaland : Bull. Entom. Res., London, Vol. VII, Part 1, pp. 29-50 (May 1916) ; 85. Some further Notes on the Tsetse-Flies of Nyasaland : Bull. Entom. Res., London, Vol. XI, Part 2, pp. 101-104 (September 1920). LAVIER, G. : 86. Les parasites des invertebres hematophages. Parasites qui leur .sont propres. Parasites qu'ils transmettent aux vertebres. Travail du laboratoire de Parasitologie de la Faculte de Medecine de Paris. Pp. 218. Paris : Vigot freres (1921). LLOYD, LI. : 87. Notes on Glossina morsitans in Northern Rhodesia : Bull. Entom. Res., London, Vol. Ill, Part 1, pp. 95-96, fig. 1 (May 1912) ; 88. Notes on Glossina morsitans, Westw., in the Luangwa Valley, Northern Rhodesia : Bull. Entom. Res., London, Vol. Ill, Part 3, pp. 233-239 (November 1912) ; 89. Further Notes on the Bionomics of Glossina morsitans in Northern Rhodesia : Bull. Entom. Res., London, Vol. V, Part 1, pp. 49-60, 4 plates, and map (April 1914) ; 90. Note on Scratching Birds and Tsetse-Fly : Ann. Trop. Med. and Parasitol., Ser. TM., Liverpool, Vol. VIII, No. 1, p. 83 (21st April 1914) ; 91. Report on the Investigation into the Bionomics of Glossina morsitans in Northern Rhodesia, 1915 : Bull. Entom. Res., London, Vol. VII, Part 1, pp. 67-79, 2 figs., 1 plate, 3 tables (May 1916). MCCONNELL, Dr. ROBERT E. : 92. Notes on the Occurrence and Habits of Glossina fuscipes in Uganda: Bull. Entom. Res., London, Vol. Ill, Part 1, pp. 55-59 .(May 1912). MACFIE, Dr. J. W. SCOTT: 93. Experiments and Observations upon Glossina palpalis : Bull. Entom. Res., London, Vol. Ill, Part 1, pp. 61-72 (May 1912) ; 94. The Distribution of Glossina in the Ilorin Province of Northern Nigeria : Bull. Entom. Res., Vol. IV, Part 1, pp. 1-28, pi. I-VII, and map (May 1913) ; 172 95. The Results of Dissections of Tsetse-Flies at Accra : Kept. Accra Lab. for the Year 1915, London, pp. 49-54, 4 figs,, 2 tables,, (no date). MARTIN, Dr. G., LEBOEUF, Dr., ROUBAUD, Dr. E. : 96. La Maladie du Sommeil au Congo Frangais, 1906-1908. — Rapport de la Mission d'fitudes de la Maladie du Sommeil au Congo^ Frangais 1906-8. Pp. vii + 721, viii plates, 136 figs., 1 map. Royali 8vo. Paris : Masson & Cie (1909). MANSON, Sir P., G.C.M.G., F.R.S. : 97 Manson's Tropical Diseases, 7th ed., edited by P. H. Manson- Bahr, D.S.O., M.D. Pp. xvi + 960, 21 col. plates, 6 half-tone plates, 404 figs, in the text, and 31 charts. London : Cassell & Co. Price 31s. 6d. net (1921). MASON, C. : 98. Report of the Government Entomologist. Nyasaland Pro- tectorate, Ann. Kept. Dept. Agric. for Year ended 31st March 1917, Zomba, pp. 9-13 (1917). MOISER, Dr. B. : 99. Description of Haunts of Glossina tachinoides in Bornu Province,. Northern Nigeria : Bull. Entom. Res., London, Vol. II, Part 2, pp. 119-126, 6 figs. (July 1911) ; 100 Notes on the Haunts and Habits of Glossina tachinoides near Geidam, Bornu Province, Northern Nigeria : Bull. Entom. Res., London, Vol. Ill, Part 2, pp. 195-202 (August 1912) ; 101. Notes on a few Photographs illustrating the Haunts and Habits, of Glossina tachinoides in Bornu, Northern Nigeria : Bull. Entom. Res., London, Vol. IV, Part 2, p. 145, plates XIII-XVII (September 1913). MURRAY, Dr. W. A. : lOlfl. Note on Relation between Tsetse-Fly (Glossina morsitans) and Game in the Proclaimed Area, Nyasaland : Trans. R. Soc. Trop. Med. and Hyg., London, Vol. XV, No. 4, pp. 118-121, and map (20th October 1921). NEAVE, Dr. S. A. : 102. Report on a Journey to the Luangwa Valley, North-Eastern Rhodesia, from July to September 1910 : Bull. Entom. Res., London, Vol. I, Part 4, pp. 303-316, 7 figs, and map (January 1911) ; 103. Notes on the Blood-Sucking Insects of Eastern Tropical Africa : Bull. Entom. Res., London, Vol. Ill, Part 3, p. 275-323, plates X, XI (November 1912). NEWHAM, Temp Lt.-Col. H. B. G. : 104. Trypanosomiasis in the East African Campaign . //. R.A.M.C., London, Vol. XXXIII, No. 4, pp. 299-311, 1 map (October 1919). NEWSTEAD, ROBERT, F.R.S. : 105. A Revision of the Tsetse-Flies (Glossina), based on a Study of the Male Genital Armature : Bull. Entom. Res., London, Vol. II, Part 1, pp. 9-36, figs. 1-17 (April 1911) ; 106. On the Genital Armature of the Males of Glossina medicorum,. Austen, and Glossina tabaniformis, Westwood : Bull. Entom. Res.,. London, Vol. II, Part 2. pp. 107-110, figs. 1, 2 (July 1911) ; 173 . A New Tsetse-Fly from the Congo Free State ; and the 'Occurrence of Glossina austeni in German East Africa : Ann. Trop. Med. Parasit., Liverpool, Vol. VII, No. 2, pp. 331-334, figs. 1-2 {June 1913) ; 107. [With Dr. J. B. Davey.] Royal Society's Commission on Trypanosomiasis in Nyasaland, British Central Africa. First Pre- liminary Report on the Bionomics of Glossina morsitans : Repts. Sleeping Sickness Comm. Roy. Soc., London, No. XV, pp. 142-157 plates 11-13 (1914); 108. Polypneustic Lobes in the Larvae of Tsetse-Flies (Glossina) and Forest-Flies (Hippoboscidae) : Ann. Trop. Med. Parasit., Liverpool Vol. XII, No. 1, pp. 93-107, 7 figs. (July 1918). XEWSTEAD, R. and EVANS, A. M. : 109. New Tsetse-Flies (Glossina) from the Belgian Congo: Ann. Trop. Med. Parasit., Liverpool, Vol. XV, No. 1, pp. 95-102, 6 figs (27th April 1921); 109«. A New Tsetse-Fly from the South Cameroons : Ann. Trop. Med. Parasit., Liverpool, Vol. XVI, No. 1, pp. 51-54, figs. 1-2 (31st March 1922). PATTON, Capt. W. S. and CRAGG, Capt. F. W. : 110. A Text-Book of Medical Entomology. Pp. 768, 89 plates. London : Christian Lit. Soc. for India. 21s. (1913). PICKELS, A. : 111. Nigeria (Northern Provinces) Annual Medical and Sanitary Report for the Year ending 31st December 1917, Lagos, pp. 107-165 (no date). POLLARD, Dr. J. : 112. Notes on the Tsetse-Flies of Muri Province, Northern Nigeria : Bull. Entom. Res., -London, Vol. Ill, Part 2, pp. 219-221, and map (August 1912). RODHAIN, Dr. J. : 113. La maladie du sommeil dans 1'Ouelle (Congo Beige) a la fin de 1914 : Bull. Soc. Path. Exot., Paris, T. IX, No. 1, pp. 38-72, 1 map, 6 tables (January 1916) ; 114. Note sur les trypanoses et les piroplasmoses des grands animaux de 1'Ouelle : Bull. Soc. Path. Exot., Paris, T. IX, No. 2, pp. 95-109, 1 fig., 1 table (9th February 1916) ; 115. La limite septentrionale de 1'aire d'extension de la Glossina morsitans entre le. Lualaba et le Lac Tanganyika : Rev. Zool. Afric., Bruxelles, Vol. VII, No. 1, pp. 57-64, 1 map (25th May 1919). RODHAIN, Dr. J., PONS, Dr. C., VAN DEN BRANDEN, Dr. F., and BEQUAERT, Dr. J. : 116. Rapport sur les travaux de la mission scientifique du Katanga (octobre 1910 a septernbre 1912), pp. 61 et seq. — Contribution au mecanisme de la transmission des trypanosomes par les glossines. Bruxelles : Ministere des Colonies (1912) ; 117. Essai de capture des glossines au moyen de glu, fabriquee avec de 1'huile de lin. — Ibid., pp. 165, et seq. ; 118. Dispersion et biologic des Dipteres hematophages au Congo Beige.— Ibid., pp. 188, et seq. 174 ROUBAUD, Dr. E. : 119. Sur la reproduction et les variations du developpement dans la Glossina palpalis, Desv. : C.R. A cad. Sci., Paris, T. 146, No. 1, pp. 362-365 "(17th February 1908) ; 1190. fitudes biologiques sur les Glossines du moyen Dahomey: C.R. Acad. Sci., Paris, T. 152, No. 1, pp. 406-409 (13th February 1911); 119&. Sur un essai d'elevage de Glossines dans les laboratoires d'Europe: Bull. Soc. Path. Exot., Paris, T. VIII, No. 1, pp. 34-36 (13th January 1915) ; 120. Les zones a tsetses de la Petite-Cote et du Bas-Saloum (Senegal) : Bull. Soc. Path. Exot., Paris, T. VIII, No. 3 (1915) ; 121. Histoire d'un elevage de Glossina morsitans a ITnstitut Pasteur de Paris : Bull. Soc. Path. Exot., Paris, T. X, No. 1, pp. 629-640, 2 figs. (llth July 1917) ; 122. Observations biologiques sur Nasonia brevicornis, Chalcidide Parasite des Pupes de Muscides. Determinisme physiologique de 1' Instinct de Ponte ; Adaptation a la Lutte contre les Glossines : Bull. Scient. France et Belgique, Paris, T ser., T. L, No. 4, pp. 425-439, 1 fig. (8th June 1917) ; 123. Les Particularites de la Nutrition et la Vie symbiotique chez les Mouches Tsetses : Ann. Inst. Pasteur, Paris, T. XXXIII, No. 8, pp. 489-536, 15 figs. (August 1919) ; 1230. Les mouches tsetses en Afrique occidentale frangaise — Nos connaissances actuelles sur leur histoire et leur role pathogene : Bull, du Comite d'£tudes Hist, et Scient. de I' Afrique Occidentale Francaise, Paris, pp. 257-300, 7 figs, and map (1921). RUPPERT, F. and HUBER, F. : 124. t)ber die tierarztliche Tatigkeit im Feldzuge in Deutsch-Ost- Afrika : Deutsche Tierarztl. Wochenschr., Hannover, Bd. XXVIII, Nos. 38, 39, pp. 441-446, 455-458, 10 figs., 1 map (18th and 25th September 1920). SANDERSON, Dr. M. : 125. Notes on Glossina /us ca, Walk., in North Nyasa : Bull. Entom. Res., London, Vol. I, Part 4, pp. 229-302, 2 figs., 1 map (January 1911). SCHWETZ, Dr. J. : 126. La Limite occidentale de la Glossina morsitans dans le Katanga du Nord : Bull. Entom. Res., London, Vol. VI, Part 3, pp. 283-288, and map (December 1915) ; 127. Quelques Observations preliminaries sur les Moeurs de la Glossina brevipalpis : ibid., pp. 289-292, and map (December 1915) ; 128. Preliminary Note on the General Distribution of Glossina palpalis, Rob. Desv., in the District of Lomami, Belgian-Congo : Ann. Trop. Med. and Parasit., Liverpool, Vol. IX, No. 4, pp. 513-526 (30th December 1915) ; 129. The Western and Northern Limit of Glossina morsitans in Northern Katanga : Bull. Entom. Res., London, Vol. VIII, Part 2, pp. 165-168, and map (December 1917) ; 130. Preliminary Note on the Tsetse-Flies of the Kabalo-Albertville (Lualaba-Tanganyika) Railway : ibid., pp. 169-175, and map (December 1917) ; 175 131. A Comparative Study of the Habits of Glossina brevipalpis Newst., G. fusca Westw. and G. pallidipes, Aust. in the Belgian Congo : Ann. Trop. Med. Parasit., Liverpool, Vol. XI, No. 4, pp. 365-398 1 map, 1 chart (llth May 1918) ; 132. Recherches sur les Glossines (Mouches Tse-Tse), pp. viii +151, 4 maps, 5 figs.. 8vo. Bruxelles : Hayez (1919). (1) Note preliminaire sur la Distribution de la Glossina palpalis dans le District du Lomami (Congo Beige). Pp. 1-16. (2) La Limite occidentale de la Glossina morsitans dans le Katanga du Nord. Pp. 17-20. (3) La Limite occidentale et septentrionale de la Glossina morsitans dans le Katanga du Nord (2e note). Pp. 21-25. (4) Quelques Observations preliminaires sur les Moeurs de la Glossina brevipalpis. Pp. 27-34. (5) Note preliminaire ur les Glossines du Chemin de Fer Kabalo- Albertville (Lualaba-Tanganika), avec un Croquis. Pp. 35-44. (6) Contribution a 1'fitude des Moeurs de la Glossina brevipalpis, de la Glossina fusca et de la Glossina pallidipes (lre Etude). Avec un plan. Pp. 45-83. 7. Contribution a 1'fitude des Moeurs de la Glossina brevipalpis, Glossina fusca, Glossina pallidipes, Glossina palpalis et Glossina morsitans (2e £tude). Avec une carte. Pp 85- 120. 8. Contribution a 1'fitude des Moeurs de la Glossina brevipalpis, de la Glossina fusca, de la Glossina pallidipes et de la Glossina palpalis (3me fitude). Avec cinq figures. Pp. 121- 141. 9. Quelques Remarques concernant les Moeurs de Glossina tabaniformis, Westw. Pp. 143-144. 10. Note explicative sur la Carte de la Repartition des Glossines dans le Nord-Katanga. Avec quelques Considerations generates. Pp. 145-151. 133. Quelques Remarques concernant les Moeurs de la Glossina tabaniformis : Ann. Trop. Med. Parasit., Liverpool, Vol. XII, No. 3-4, pp. 279-280 (28th February 1919) ; 134. L'Identite des Conditions geo-botaniques des Gites a Pupes de la Gl. palpalis, de la Gl. fusca, de la Gl. brevipalpis, de la Gl. pallidipes et de la Gl. morsitans : Bull. Soc. Path. Exot., Paris, T. XII, No. 5, pp. 234-238 (14th May 1919) ; 135. La Maladie du Sommeil dans le Nord-Katanga (Congo Beige) en 1913-18: Bull. Soc. Path. Exot., Paris, T. XII, No. 9, pp. 671-680, and map (12th November 1919) ; 136. La Maladie du Sommeil dans le Moyen-Kwilu (District du Kwango, Congo Beige) en 1918: Bull. Soc. Path. Exot., Paris, T. XII, No. 10, pp. 798-812, and map (10th December 1919); 136a. La Presence de Pupes de Glossina palpalis a 1.500m. de 1'Eau: Bull. Soc. Path. Exot., Pans, T. XV, No. 1, pp. 23-25 (llth January 1922). SHARPE, SIR ALFRED, K.C.M.G., C.B. : 137. Notes on the Habits of Glossina morsitans in Nyasaland and the Adjoining Territories : Bull. Entom. Res., London, Vol. I. Part 3,. pp. 172-175 (October 1910). 176 SHIRCORE, Dr. J. O. : 138. On two Varieties of Glossina morsitans from Nyasaland : Bull. Entom. Res., London, Vol. IV, Part 1, p. 89 (May 1913) ; 139. Suggestions for the Limitation and Destruction of Glossina morsitans : Bull. Entom. Res., London, Vol. V, Part 1, pp. 87-90, and sketch map (April 1914) ; 140. A Method for the Trapping of Glossina morsitans Suggested for Trial : Trans. Soc. Trop. Med. and Hyg., London, Vol. IX, No. 3, pp. 101-102, 3 diagrams (January 1916) ; 141. Traps for Glossina', Correspondence: Trans. R. Soc. Trop. Med. and Hyg., London, Vol. XIV, No. 2, p. 30 (18th June 1920). .SIMPSON, Dr. J. J. : 142. Entomological Research in British West Africa : I. — Gambia : Bull. Entom. Res., London, Vol. II, Part 3, pp. 187-239, map and 14 figs, in text (October 1911) ; II. — Northern Nigeria : Bull. Entom. Res., London, Vol. II, Part 4, pp. 301-356, 3 maps and 8 plates (January 1912) ; III. — Southern Nigeria: Bull. Entom. Res., London, Vol. Ill, Part 2, pp. 137-193, map and 4 plates (August 1912) ; IV. — Sierra-Leone : Bull. Entom. Res., London, Vol. IV, Part 3, pp. 151-190, map and 5 plates (November 1913) ; V.— Gold Coast : Bull. Entom. Res , London, Vol. V, Part 1, pp. 1-36, map and 4 plates (April 1914) ; 143. Colour Conventions to Indicate the Distribution of Blood- .'Sucking Insects and the Diseases they Transmit : Bull. Entom. Res., London, Vol. II, Part 4, pp. 297-299, Plate VI (January 1912) ; 144. Bionomics of Tsetse and other Parasitological Notes in the Gold Coast : Bull. Entom. Res., London, Vol. VIII, Parts 3-4, pp. 193-214 (February 1918). .SWYNNERTON, C. F. M. : 145. An Examination of the Tsetse Problem in North Mossurise, .Portuguese East Africa : Bull. Entom. Res., London, Vol. XI, Part 4, j>p. 315-385, 9 plates, 1 map (March 1921). TEICHMANN, Dr. E. : 146. Zur Biologic der Tsetse-Fliegen., Beobachtungen und Vor- .schlage : Zeitschr.f. angew. Entomologie, Berlin, Bd. I, Heft 1, pp. 147- 159, Taf. I-II (April 1914). THEILER, Sir A. : 147. Veterinary Research : Annual Report of the Director, 1915-16 : Mept. Union of S. Africa Dept. Agric.for Year ended 31st March 1916, Capetown, pp. 45-49 (1917). TURNER, R. E. : 148. A New Species of Mutilla parasitic on Glossina morsitans : .Butt. Entom. Res., London, Vol. V, Part 4, p. 383, 1 fig. (March 1915) ; 149. On Mutillidae parasitic on Glossina morsitans : Bull. Entom. .Res., London, Vol. VII, Part 1, pp. 93-95, figs. 1-2 (May 1916) ; 150. On a Braconid Parasite of Glossina : Bull. Entom. Res., London, Vol. VIII, Part 2, p. 177 (December 1917) ; 151. On a new Mutillid Parasite of Glossina morsitans: Bull. .Entom. Res., London, Vol. X, Part 3, pp. 327-328, 1 fig. (April 1920). 177 TURNER, R. E. and WATERSTON, J. : 152. A new Parasite bred from Glossina morsitans in Nyasaland : Bull Entom. Res., London, Vol. VII, Part 2, pp. 133-135" figs 1-2 (October 1916). UNION OF SOUTH AFRICA, DEPT. OF AGRIC. : 153. Tsetse-Fly Investigation : //. Dept. Agric., Union 5. Africa Pretoria, Vol. I, No. 9, p. 799 (December 1920). VAN SACEGHEM, R. : 154. Contribution a 1'fitude de la Transmission du Trypanosoma cazalboui : Ann. Med. vet., Bruxelles, T. LXV, Nos. 8-9, pp. 369-374 (August and September 1920) ; 155. La Trypanosomiase du Ruanda: C.R. Soc. Biol., Paris, T. LXXXIV, No. 5, pp. 283-286 (5th February 1921) ; 156. La Trypanosomiase du Ruanda — Deuxieme Rapport : Bull agric. du Congo Beige, Bruxelles, T. XII, No. 2, pp. 294-302 (June 1921). VAUGHAN-KIRBY, F. : 157. Game and Game Preservation in Zululand : S. African Jl. Science, Cape Town, Vol. XIII, No. 9, pp. 375-396 (April 1917). WOLFEL. K. : 158. Beitrag zur Kenntnis der Tsetse (Glossina morsitans) und der Trypanosomiasis : Zeitschr. f. Infektionskrankh., parasitdre Krdnkh. u. Hyg. der Haustiere, Berlin, Bd. XVII, No. 1-2, pp. 19-36, and map (17th August 1915). WATERSTON, J. : 159. Notes on African Chalcidoidea — II : Bull. Entom. Res.+ London, Vol. V, Part 4, pp. 343-372, figs. 1-17 (March 1915) ; 160. Chalcidoidea bred from Glossina morsitans in Northern Rhodesia : Bull. Entom. Res., London, Vol. VI, Part I, pp. 69-82, rigs. 1-5 (June 1915) ; 161. Chalcidoidea bred from Glossina morsitans in Nyasaland : Bull. Entom. Res., London, Vol. VI, Part 4, pp. 381-393, figs. 1-9- (February 1916) ; 162. Notes on African Chalcidoidea — V : Bull. Entom. Res., London, Vol. VII, Part 2, pp. 123-132, figs. 1-5 (October 1916) ; 163. Chalcidoidea bred from Glossina in the Northern Territories, Gold Coast : Bull. Entom. Res., London, Vol. VIII, Part 2, pp. 178-179, 1 fig. (December 1917). WOOSNAM, R. B. : 164. The Question of the Relation of Game Animals to Disease in Africa : J/. E. Africa and Uganda Nat. Hist. Soc., Nairobi, Vol. IV, No. 7, p. 21 (December 1913) ; 165. Report on a Search for Glossina on the Amala (Engabei) River. Southern Masai Reserve, East Africa Protectorate : Bull. Entom. Res. London, Vol. IV, Part 4, pp. 271-278, and map (February 1914). ZUPITZA, M : 166. Ein Weg zu erfolgversprechender Bekampfung der Schlafkrank- heitsfliege am Tanganika : Arch. f. Schiffs- u. Trop. Hyg., Leipsicf Bd. XXIV, No. 6, pp. 161-166 (June 1920). (5979) M 178 MISCELLANEOUS : 167. A Method of Destroying Tsetse-Flies : Bull. Entom. Res., London, Vol. I, Part 1, p. 85 (April 1910). 168. An Investigation into the Habits and Distribution of the Cattle Tsetse-Flies (Glossina] : Bull. Entom. Res., London, Vol. I, Part 2, pp. 152-153 (July 1910). 169. Handbook of Instructions for Collectors, 4th ed., pp. viii -f- 222, 75 figs., 1 plate : London, British Museum (Natural History] (1921). 170. Hygiene-Mesures a prendre pour combattre la Maladie du Sommeil au Congo Bels^e : Bull. off. du Congo Beige, Bruxelles, 13me annee, No. 10, pp. 1229-1256 (15th October 1920). 171. Report of the Glossina Sub-Committee of the Imperial Bureau of Entomology : Trans. R. Soc. Trop. Med. and Hy^., London, Vol. XIV, No. 4, pp. 59-62 (19th November 1920). PLATE I. Fig. 1, GLOSSINA PALPALIS, Rob.-Desv. $ (x about 4£). Fig. 2. GLOSSINA TACHINOIDES, Westw. $ (x about 4£). PLATE II. Fig. 1. GLOSSINA MORS1TANS, Westw. ? (x about 4J). Fig. 2. GLOSSINA PALLIDIPES, Austen. $ (x about 4i).. PLATE III. PLATE IV. Fig. 1. A very typical breeding place of GJossina. morsitans under a long-fallen trunk of Burkea afrieana, North Mossurise, Portuguese East Africa. After Swynnerton (145). 2. A much used Glossina brevipalpis breeding place in North Mossurise ; fallen trunk of Piptadenia buehanani, under which (at the spots indicated by the white marks) Mr. C. F. M. Swyn- r.erton took nearly 100 puparia of the species mentioned. After Swynnerton (145). PLATE V. Fig 1. A breeding place of Glossina morsitans at Ngoa, Northern Rhodesia ; recumbent tree, very close to path (in foreground). Beneath this tree Mr. Lloyd found 11 pupag and 43 empty cases. After Lloyd (89). Fig. 2. Glossina taehinoides, photographed in the natural position of rest. After Dr. Bernard Moiser (101). 179 INDEX A. Abyssinia, Glossina longipennis in, 20 ; G. morsitans in, 14. Acacia, in relation to habitat of Glossina, 54, 55. Adansonia digitata (Baobab Tree), in relation to habitat of Glossina, 37, 71 . Aepyceros melampus (Impala), relation of Glossina to, 132 ; infected with Trypanosoma rhodesiense, 94. Africa French West, Glossina fusca Group in, 46 ; G. longipalpis in, 55 ; G. pallicera in, 46 ; G. palpalis in, 9 ; G. morsitans in, 50, 51. Africa, Portuguese East, Glossina austeni in, 12, 35, 48 ; G. brevipalpis in 19, 43, 51, 54, 58, 64, 71, 75, 80, 88, 92, 101, 119; G. morsitans in, 16, 41, 42, 48, 51, 54, 70, 71, 75, 80, 88, 92, 101 ; G. pallidipes in, 16, 42, 53, 71, 75, 80, 88, 92, 101. America, North, an ancient habitat of Glossina, 6 (note). Amomum, Glossina brevipalpis sucking leaf of, 58. Anastatus viridiceps, parasite of Glossina morsitans in Northern Rhodesia, 122, 124, 131. Angola, Glossina brevipalpis in, 19 ; G. palpalis in, 8. Anopheles, measures against, in Central America, 111 (note). Antelopes, relation of Glossina spp. to, 35, 41, 73, 80, 86, 91, 92, 101, 112, 132, 138 ; suggested experiments with Glossina and, 106 ; behaviour of, when attacked by Tabanus, 112; in relation to spread of trypano- somiasis, 94, 95 ; experiments in destruction of, 103 (see Game). Ant-lions, in breeding places of Glossina morsitans, 40. Ants, attacking Glossina, 33, 117; attacking preserved specimens of Glossina, 157 ; relation of, to Glossina larvae and pupae, 28, 117. Arabia, Glossina tachinoides in, 2, 6, 11, 46, 47. Ashanti (see Gold Coast). Austenina brevipalpis (see Glossina}. B. Babal Thorn, relation of, to habitat of Glossina tachinoides, 47. Baboons, relation of Glossina to, 91, 92, 93, 102, 103. Banana, relation of, to habitat of Glossina pallidipes, 42, 53. Bats, relation of Glossina to, 84, 93 ; destroying mosquitos, 113 (& note). Bauhinia, relation of, to breeding places of Glossina, 43. Bechuanaland, Glossina morsitans in, 14, 15. Bembex, 128; predacious on Glossina and Asilids, 117, 119. Birds, relation of Glossina to, 47, 85, 86, 90, 91, 97, 113, 114 ; investigations on food of, 115 (note). Borassus Palm, relation of, to Glossina breeding places, 40. Brachystegia, relation of, to habitat of Glossina, 43, 51, 52, 54, 58. Bubalis lichtensteini (Lichtenstein's Haartebeeste), infected with Trypano- soma rhodesiense, 94. Buffalo, relation of Glossina to, 41, 43, 70 (& note), 96, 97, 99, 100, 101, 132, 138 ; attacked by rinderpest, 48 ; infected with trypanosomes, 95. Bufo regularis, fed upon in captivity by Glossina palpalis, 91. Bushbuck (see Tragelaphus scriptus}. Bush-pig, relation of Glossina to, 92, 93, 102, 112 ; danger of spread of sleeping sickness by, 100. 180 INDEX— continued. C. Cacer gates leucosticta, predacious on Glossina, 117. Cactus, relation of, to habitat of Glossina tachinoides in Southern Arabia, 47. Calliphora, parasitised by Nasonia brevicornis, 134, 135. Calosoma sycophanta, introduced into U.S.A., 132 (note). Camels, repellents for protection of, against biting flies, 149. Cameroon, Glossina caligineain, 10 ; G. fusca in, 17 ; G. haningtoni in, 17 ; G. pallicera in, 10; G. tabaniformis in, 18; G. tachinoides in, 11 ; G. ziemanni in, 13. Cattle, relation of Glossina to trypanosomiasis of, in Tropical Africa, 1-3, 34, 56 (& note), 103, 151 ; Glossina spp. attacking, 54, 57, 70, 75, 76, 78, 101, 103, 104, 119; protection of, against Glossina, 147-150 ; relation of game to trypanosomiasis of, 95 ; trypanosomiasis of. apparently due to biting flies other than Glossina, 106. Cercopithecus albogularis beirensis, 92. Cercopithecus pygerythrus, relation of Glossina to, 92. Chalcis amenocles, parasite of Glossina tachinoides and G. morsitans submorsitans in Gold Coast, 130, 131. Chameleon, relation of Glossina to, 84, 90, 91, 96. Chrysomyia, parasitised by Nasonia brevicornis, 135. Chrysops, attacked by mites, 136. Cissus, relation of, to Glossina breeding places, 43. Civet Cats, destruction of, against Glossina, 142. Cobus ellipsiprymnus (Waterbuck), relation of, to Glossina, 35, 86, 138 ; infected with trypanosomes, 94, 95. Cockroaches, in breeding places of Glossina morsitans, 40. Coelalysia glossinophaga, parasite of Glossina morsitans submorsitans in Gold Coast, 131. Combretum, relation of, to habitat of Glossina morsitans, 51. Congo, Belgian, distribution of Glossina in, 60-64 ; G. brevipalpis in, 19, 42, 43, 53, 56, 58, 63, 64, 76 ; G. fusca in, 17, 42, 53, 56, 58, 63, 76 ; G. fusca var. congolensis, 17 ; G. fuscipleurisin in, 18, 56, 64 ; G. longi- palpis in, 64 ; G. morsitans in, 14 (note), 34, 36, 49, 53, 61, 62, 63, 136 (note) ; G. nigrofusca in, 18, 64 ; G. pallicera in, 11, 64 ; G. pallidipes in, 16, 42, 53, 56, 58, 63, 76 ; G. palpalis in, 8, 10, 34, 35, 53, 60, 61, 93, 121 ; G. schwetzi in, 22, 64 ; G. severini in, 21, 64 ; G. tabaniformis in, 18, 56, 64 ; G. tachinoides not recorded from, 11. Congo, French, Glossina fusca in, 17 ; G. morsitans in, 14 ; G. tabaniformis in, 18; G. tachinoides in, 11. Conostigmus rodhaini, parasite of Glossina palpalis in Belgian Congo, 121, 131. Copaifera mopani (Sanya or Iron-wood Tree), relation of, to Glossina breeding places, 37 (& note), 41. Coracias caudatus, probably predaciou? on Glossina, 114. Crateropus, relation of, to Glossina breeding places, 41 ;• probably predacious on Glossina, 114. Crocidura, relation of Glossina to, 90. Crocodiles, relation of Glossina to, 72, 76, 85, 86, 91 (& note), 142. Crocothemis erythraea, of little use as an enemy of Glossina, 119. Crossarchus (Banded Mongoose), relation of Glossina to, 90. Crow-shrikes (Prionops and Sigmodus), probably predacious on Glossina in Nyasaland, 114. Cuba, mosquitos disseminating malarial and yellow fevers in, 111. D. Dahomey, Glossina longipalpis in, 34, 46, 54, 55, 66, 79 ; G. morsitans in, 14 ; G. palpalis in, 46, 54, 55, 79 ; G. tachinoides in, 23, 46, 55, 65, 79. Dicrurus afer (Black Drongo), predacious on Glossina morsitans in Nyasa- land, 114. 181 1 NDEX — continued. Diplorhynchus mossambicensis, Glossina resting on trunk of, 71. Dirhinus inflexus, parasite of Glossina morsitans submorsitans in Gold Coast, 131. Dog, transmission of trypanosomes by Glossina to, in Africa, 2 ; destruction of, against Glossina, 142. Dormouse, relation of Glossina to, 90. Dragonflies, predacious on Glossina, 116, 117, 118, 119. Ducks, relation of Glossina to, 84. Duiker, attacked by Glossina tachinoides, 47 ; infected with trypanosomes, yo . Duiker, Blue, relation of Glossina to, 35, 92. E. Ebony, relation of, to Glossina breeding places, 37, 46. Eland, relation of Glossina to, 138 ; infected with trypanosomes, 95. Elephants, relation of, to habitats of Glossina, 35, 79, 101 ; not infected with trypanosomes, 96. Erythrina, relation of, to Glossina breeding places, 37. Eugenia, relation of, to breeding places of Glossina morsitans, 41. Eupelminus tarsatus, parasite of Glossina morsitans in Nvasaland, 127, 129 131. Euphorbia, relation of, to habitat of Glossina tachinoides in Southern Arabia, 47. Euponera senaarensis, attacking larva of Glossina morsitans in Nyasaland, 28. F. Fernando P6, Glossina palpalis in, 6 (note), 9. Ficus, relation of, to Glossina breeding places, 35. Fish, Glossina experimentally fed on, 91 ; list of, destroying mosquito larvae, 114 (note). Fowls, relation of Glossina to, 83, 84, 90, 119. Francolinus, relation of Glossina morsitans to, 37, 41, 115. Frogs, relation of Glossina to, 91. Fuirena, relation of, to habitat of Glossina morsitans, 52. G. . Galago demidofft, relation of Glossina to, 91. Gambia, Glossina morsitans in, 14, 50, 51, 67 ; G. palpalis in, 67. Game, in relation to Glossina, 2, 35, 38-41, 43, 47, 50, 52, 56, 70, 73, 79, 80, 86, 91, 94-107, 112, 132, 138-142; effect of destruction of, on Glossina, 99, 100, 101, 102, 151 (see Antelopes). Gecko, relation of Glossina to, 41, 84, 90. Georychus (Mole Rat), relation of, to Glossina, 90. Glossina (Tsetse-flies), methods of distinguishing, from other flies, 4 ; classification of, 7 ; reproduction of, 23-34 ; parthenogenesis in, 32 ; larvae of, 25-27 ; pupae of, 27-32 ; breeding places of, 34-43 ; breeding season of, 43 ; habitats of, 44-60 ; food and feeding habits of, 78-90 ; drinking of water by, 87 ; sucking of vegetable juices by, 58, 88 ; geographical distribution of, 6, 8-20, 60-64 ; seasonal migrations of, 65-67 ; range of flight of, 67-69 ; travelling habits of, 69-70 ; resting surfaces used by, 71 ; proportion of sexes of, 71-78 ; effect of external factors on, 107-110, 146 ; mode of death of, 109 ; influence of colours on, 109-110, 145, 146; influence of odours on, 110, 145; parasitic enemies of, 119-136; predacious enemies of, 110-119 ; indirect enemies of, 111-112 ; relation of, to game, 2, 35, 38-41, 43, 47, 50, 52, 56, 70, 73, 79, 80, 86, 91, 94-106, 112, 132, (5979) M* 182 INDEX — continued. 138-140, 151 ; relation of, to smaller animals, 90-93 ; relation of, to trypanosomiasis of man and animals, 1-3, 8 (& note), 14, 34, 45, 56, 57, 73, 77, 93, 94, 95, 97, 101, 102, 103, 104, 136, 137, 140, 142, 146, 150, 151 ; trypanosomiasis of cattle apparently sometimes due to biting flies other than, 106—107 ; means of limiting and destroying, 110-147; methods of clearing against, 136-141; grass-burning against, 137, 139, 140-141 ; methods of catching and trapping, 141-146 ; artificial breeding places for, 146-147 ; repellents for, 147—150 ; protection of stock against, 147—150 ; protection of man against, 150-151 ; destruction of big game against, 151-152 ; keeping and breeding of, in captivity, 158—159 ; methods of collecting, pre- serving and studying, 152-157 ; dissection of salivary glands of, 159-161 ; symbols to indicate distribution of, 161-162 ; suggestions for future work on, 162-164 ; bibliography of principal publications connected with, 165-178. Glossina austeni, systematic position of, 7 (note) ; characters of, 11 ; distribution of, 12 ; pupa of, 29, 30, 31 ; pupal stage of G. brevipalpis longer than that of, 34 ; breeding places of, 35 ; habitats of, 47, 59 ; breeding of, in captivity, 158 ; symbol to indicate distribution of, 162. Glossina brandoni, synonym of G. austeni, 11, 12. Glossina brevipalpis, systematic position of, 7 ; characters of, 4, 18, 19 ; distribution of, 19 63, 146 (note) ; copulation in, 23 ; reproduction in, 23, 24 ; larva of, 26, 28 ; pupa of, 29, 30 ; pupation of, 28, 29, 34 ; emergence of imago of, 32 ; parthenogenesis in, 32 ; breeding places of, 42 ; bionomics of, 56-59, 60, 63, 70, 71, 79-80, 87 ; entering railway carriages at night, 52, 59 ; habitat of, 47, 51, 53, 54, 56-59, 63 ; seasonal migrations of, 65 ; proportion of sexes of, 75-76, 77 ; resting surfaces used by, 58, 71, 79 ; relation of, to game, 101, 105 ; relation of, to smaller animals, 92 ; relation of, to trypanosomiasis, 56 (& note), 57, 77; dragonflies probably predacious on, 119; Mutilla glossinae bred from, 129 ; protection of cattle against, 150 ; clearing measures against, 140-141 ; breeding of, in captivity, 24, 158 ; symbol to indicate distribution of, 161, 162. Glossina caliginea, systematic position of, 7 ; characters of, 10 ; dis- tribution of, 10; habitat of , 45 ; symbol to indicate distribution of, 162. Glossina fusca, species allied to, 7, 17 ; characters of, 16 ; distribution of, 17, 63 ; pupa of, 29, 30, 31 ; breeding places of, 42 ; habitats of, 53, 55, 58, 59 ; habits of, 60 ; proportion of sexes of, 76 ; symbol to indicate distribution of, 161, 162. Glossina fusca var. congolensis, in Belgian Congo, 17. Glossina fuscipes, considered the eastern race of G. palpalis (q.v.}. Glossina fuscipleuris, systematic position of, 7 ; characters of, 17 ; dis- tribution of, 18, 64 ; habitat of, 56 ; symbol to indicate distribution of, 162. Glossina haningtoni, characters and systematic position of, 17; in Cameroon, 17. Glossina longipalpis, systematic position of, 7, 13 ; characters of, 16 ; distribution of, 16, 64 ; duration of gestation in, 24 ; duration of pupal stage of, 34 ; habitats of, 46, 51, 54-55 ; not habitually asso- ciated with man, 50 ; seasonal migrations of, 66 ; food of, 79 ; relation of, to game, 101 ; Bembex predacious on, 117; symbol to indicate distribution of, 162. Glossina longipennis, systematic position of, 7 ; characters of, 4, 20 ; distribution of, 20 ; entering railway carriages at night, 52, 59 ; habitats of, 59 (& note) ; habits of, 59, 60 ; symbol to indicate distribution of , 162. Glossina maculata, not considered a valid species, 7 (note). Glossina medicorum, systematic position of, 7 ; characters of, 19; dis- tribution of, 20 ; a little-known species, 59 ; symbol to indicate distribution of, 162. 183 INDEX — continued. Glossina morsitans, characters of, 4, 13-14 ; copulation of, 23 ; distribution of, 14, 61-63 ; species allied to, 7, 13 ; races of, 15 ; stages of reproduc- tion in, 23, 24 ; larva of, 26, 27 ; pupa of, 29, 30, 32 ; pupation of, 27 ; breeding places of, 35-41, 42 ; breeding season of, 43 ; habitats of, 44 (& note), 48-52 (& notes), 53, 54, 59; seasonal migrations of, 65, 66 ; range of flight of, 67-68 ; travelling habits of, 70 (& note) ; resting surfaces used by, 71 ; proportion of sexes of, 74, 75, 76 ; food and feeding habits of, 78, 79 (& note), 82, 83, 84, 85, 86, 88; drinking of water by, 84, 87 ; experiments in feeding, with extravasated blood, 88-90 ; relation of, to man, 49 (note), 101 ; relation of, to game, 94 (& note) -106 ; relation of, to smaller animals, 90, 91 (& note), 92, 93 (& note) ; relation of, to trypanosomiasis, 2, 8 (note), 14, 34, 140, 146; influence of colours and odours on, 110; parasitic enemies of 120-133; predacious enemies of, 114, 115, 116, 117, 118-119; methods of clearing against, 137, 138, 139, 140, 141 ; methods of catching and trapping, 143-145 ; artificial breeding places for, 146-147 ; breeding of, in captivity, 158, 159 ; salivary glands of, 160 ; symbol to indicate distribution of, 161, 162 ; suggestions for future work on, 162-164. Glossina morsitans var. pallida, in Nyasaland, 15. Glossina morsitans var. paradoxa, in Nyasaland, 15. Glossina morsitans siibmorsitans, 7 (& note) ; distribution of, 15 ; larviposition by, 24 ; protection of larva of, from ants, 28 ; habitat of, 52 ; food of, 84 ; natural enemies of, 119, 131 ; symbol to indicate distribution of, 162. Glossina nigrofusca, systematic position of, 7 ; characters of, 18 ; dis- tribution of, 18, 64 ; habitat of, 56 ; symbol to indicate distribution of, 162, Glossina pallicera, systematic position of, 7 ; characters of, 10 ; dis- tribution of , 10, 64 ; habitat of, 45 ; symbol to indicate distribution of, 162. Glossina pallidipes, systematic position of, 7, 13 ; characters of, 15 ; distribution of, 15, 63 ; reproduction of, 24 ; larva of, 28 ; pupa of, 29, 30, 31 ; duration of pupal stage of, 33 ; breeding places of, 41-42 ; entering railway carriages at night, 52, 59 ; habitats of, 16, 52, 53 (¬e), 54, 56, 57, 58, 59 ; resting surfaces used by, 71 ; proportion of sexes of, 75 ; feeding habits of, 79, 80 ; drinking of water by; 88 ; relation of, to game, 92, 101, 103, 105; relation of, to man, 101 ; relation of, to trypanosomiasis, 103; clearing measures against, 141 ; breeding of, in captivity, 158 ; symbol to indicate distribution of, 162. Glossina palpalis, species allied to, 7 ; characters of, 8 ; copulation in, 23 ; larva of, 25 ; pupa of, 29, 30, 31 ; breeding places of, 34 (& note) ; - habitats of, 44-45, 46, 47, 50, 54, 55, 59 ; distribution of, and of its varieties, 8, 9, 10, 11, 60-61 ; in Fernando P6 and Principe, 6 (note), 9 ; seasonal migrations of, 66 ; range of flight of, 67 ; travelling habits of, 69-70 ; proportion of sexes in, 71-74, 76 ; food and feeding habits of, 78, 79, 80, 82, 83, 84, 85, 86, 87 ; drinking of water by, 84 ; game in certain localities unnecessary to, 97, 105 ; relation of, to reptiles and smaller animals, 91 (& note) ; relation of, to pigs, 93, 137, 142 ; relation of, to trypanosomiasis, 2, 8, 34, 73, 93, 137 ; effect of external factors on, 107-110 ; influence of colours on, 109 ; influence of odours on, 110; parasitic enemies of, 121, 131, 135; predacious enemies of, 116; indirect enemies of, 112; methods of clearing against, 137; methods of catching and trapping, 142-143 ; artificial breeding places for reducing the numbers of, 147 ; breeding of, in captivity, 158, 159 ; salivary glands of, 160 ; symbol to indicate distribution of, 161, 162. Glossina palpalis var. fuscipes, 1 (note), 9 (note), 10 ; habitats of, 45 ; proportion of sexes of, 76 ; probably not destroyed by birds in Uganda, 114; attacked by Cacer -gates leucosticta, 117; symbol to indicate distribution of, 162. 184 INDEX— cont inued. Glossina palpalis var. pallida, proposed name for variety occurring in Gold Coast, 9. Glossina palpalis var. wellmani, 8, 9. Glossina schwetzi, description of, 21 ; in Belgian Congo, 64. Glossina severini, systematic position of, 7 (note), 22 ; description of, 21 ; in Belgian Congo, 64. Glossina submorsitans, considered a race of G. morsitans (q.v.). Glossina tabaniformis, systematic position of, 7 ; characters of, 18 ; dis- tribution of, 18 ; habitat of, 56 ; in Belgian Congo, 64 ; symbol to indicate distribution of, 162. Glossina tachinoides, systematic position of, 7 ; characters of, 4, 11; distribution of, in West and Central Africa, 1 1 ; in Southern Arabia, 6, 11, 47; variety of G. palpalis associated with, in Nigeria, 9; G. austeni confused with, 31 ; copulation in, 23; stages of reproduc- tion in, 23, 24 ; breeding places of, 34 ; habitats of, 46, 50, 55 ; seasonal migrations of, 65 ; range of flight of, 68 ; food of, 79, 84 ; parasitic enemies of, 130, 131, 135; predacious enemies of, 119; symbol to indicate distribution of, 162. Glossina ziemanni. systematic position of, 7 (note) ; characters of, 12 ; in Cameroon, 13. Goats, relation of Glossina to, 73, 75, 84, 117, 119, 144; suggested experiments with Glossina and, 106 ; relation of game to trypano- somiasis of, 95. Gold Coast, Glossina caliginea in, 10 ; G. fusca in, 17, 55 ; G. longipalpis in, 55 ; G. medicorum in, 20 ; G. morsitans in, 14, 24, 28, 52, 84, 119, 131 ; G. nigrofusca in, 18, 56 ; G. pallicera in, 10, 45 ; G. palpalis in, 11, 34, 46, 55, 84 ; G. palpalis var. pallida in, 9 ; G. tabaniformis in, 18; G. tachinoides in, 11, 34, 46, 68, 84, 119, 130. Grass-burning, against Glossina, 137, 139, 140, 141. Grasshoppers, Glossina tachinoides probably feeding on, 47. Guinea, French, Glossina morsitans in, 14, 50, 51 ; G. tachinoides in, 11. Guinea Fowl, relation of Glossina morsitans to, 37, 41; food of, 115 (& note). H. Haartebeeste, relation of Glossina to, 41, 86, 101, 132, 138 ; infected with trypanosomes, 95. Haematopota, method of distinguishing Glossina from, 5, 6 ; characters of, 6 ; possibly transmitting trypanosomiasis to cattle, 107. Haematopota vittata, 5. Haltichella edax, parasite of Glossina morsitans in Nyasaiand, 131. Hares, relation of Glossina to, 97. Hippopotamus, relation of Glossina to, 56, 79, 91 (note), 101, 117; not infected with trypanosomes, 96. Hornbill, attacked by Glossina morsitans, 91 (note). Horse, transmission of trypanosomiasis to, by Glossina, 2 ; attractive to Glossina when moving, 78. House-fly, fecundity of, compared with that of Glossina, 141. Hyaena, infected with trypanosomes, 95, Hyphaene veniricosa, relation of, to breeding places of Glossina morsitans, 4 1 . Hystrix (Porcupine), relation of Glossina to, 97. I. Iguana (Monitor), relation of Glossina to, 85 (see Varanus}. Impala (see Aepyceros melampus}. Ivory Coast, Glossina fusca in, 17 ; G. longipalpis in, 16 ; G. medicorum in, 20 ; G. nigrofusca in, 18 ; G. pallicera in, 11 ; G. tabaniformis in, 18. 185 INDEX— continued. J. Jackals, relation of Glossina to, 97. E. Kenya Colony, Glossina austeni in. 12, 48 ; G. brevipalpis in, 19, 48, 52 ; G. fuscipleuris in, 18 ; G. longipennis in, 20, 52, 59 (& note) ; G. pallidipes in, 16, 48, 52, 53 (& note) ; G. palpalis in, 8 ; G. tachinoides not recorded from, 1 1 . Khaya nyasica, relation of, to habitat of Glossina pallidipes, 54. Kordofan, Glossina morsitans in, 14. Kudu, infected with trypanosomiasis, 95 ; relation of Glossina to, 132. Landolphia, relation of, to Glossina breeding places, 43. Lemur, relation of Glossina to, 91, 93. Leopard, relation of Glossina to, 97 ; importance of, as a " deprivative " enemy of G. palpalis in Uganda, 112. Liberia, Glossina fusca in, 17 ; G. medicorum in, 20 ; G. pallicera in, 10. Lizards, relation of, to Glossina, 41, 84, 85, 87, 90, 91 (& note), 96. Lucilia, parasitised by Nasonia brevicornis, 135. Lycaon, relation of Glossina to, 97. Lyperosia, probably transmitting trypanosomiasis to cattle, 106 ; repellents for protection of stock against, 149. Lyperosia irritans, parasitised by Spalangia spp.. 135. M. Malaria, and mosquitos in Central America and U.S.A., Ill (& note). Man, attacked by Glossina spp., 39, 46, 47, 48, 49, 51, 52, 54, 56, 57, 75, 76, 78, 79, 81, 100, 117, 118, 142, 144 ; relation of Glossina to, 72, 73, 78, 101 ; relation of Glossina to sleeping sickness in, 1-3, 34, 56, 73, 93, 94, 136, 137, 140, 142, 150 ; relation of game to sleeping sickness in, 95, 100 ; protection of, against Glossina, 150. Markhamia, relation of, to Glossina breeding places, 35. Melittophagus meridionalis, predacious on Glossina morsitans in Nyasaland, 114. Mermis, parasitic in Glossina, 136 (note). Meteorological Conditions, effect of, on Glossina, 33, 47, 50, 55, 59, 79, 108, 139. Mimosa, relation of, to habitat of Glossina longipalpis in Dahomey, 55. Mites, attacking Glossina, 135, 136. Monitor, relation of Glossina to, 85 (see Varanus) . Monkeys, relation of Glossina to, 47, 84, 85, 92 ; not infected with trypanosomes, 95 (note). Mosquito Larvae, destroyed by fish, 114 (note). Mosquitos, destroyed by bats, 113 (& note) ; probably transmitting trypanosomiasis to cattle, 106 ; mites attacking, 135. Mouse, relation of Glossina to, 90. Musca, sp. incert., parasitised by Spalangia, 111 . Musca domestica, 6 ; parasitised by Spalangia spp., 135. Mutilla auxiliaris, parasite of Glossina morsitans in Portuguese East Africa, 131, 132. Mutilla benefactrix, parasite of Glossina morsitans in Nyasaland, 128, 129, 131, 133. Mutilla glossinae, parasite of Glossina morsitans in Northern Rhodesia and Nyasaland, 121, 122, 128, 130, 131 ; bionomics of, in Nyasaland, 122-126, 128 ; parasitised by Syntomosphyrum glossinae, 125, 127 (note), 129, 131. 186 INDEX — continued. N. Nagana (see Trypanosomiasis of Domestic Animals) . Nasonia brevicornis, experiments with Glossina and, 134 ; parasite of blow-flies, 134, 135. Nematodes, parasitic in Glossina, 136 (& note). Nephele pilipes, relation of, to scarcity of G. palpalis on islands in Lake Victoria, 116. Newsteadina fusca (see Glossina). Nigeria, Glossina caliginea in, 10, 45 ; G. fusca in, 17, 55 ; G. longipalpis in, 55 ; G. medicorum in, 20 ; G. morsitans in, 14, 15, 51, 52 ; G. pallicera in, 11 ; G. palpalis in, 9, 46, 47, 71, 80, 108-110, 135; G. tabaniformis in, 18 ; G. tachinoides in, 11, 24, 46, 47, 135. Nyasaland, Glossina brevipalpis in, 19, 23, 42, 56 (note), 57, 58, 77, 79, 80, 87, 129 ; G. morsitans in, 2, 14, 27, 37, 39, 40, 42, 43, 48, 49 (& note), 56 (& note), 67, 68, 71, 74, 77, 78, 86, 87, 91, 103, 104, 114, 117, 122, 123, 128, 132, 133, 140, 144, 146; G. morsitans var. pallida in, 15; G. morsitans var. paradox a in, 15 ; G. pallidipes in, 16. Nygmia phaeorrhoea (Brown-tail Moth), 132 (note). 0. « Oribi, infected with trypanosomes, 95. Orthetrum chrysostigma, predacious on Glossina morsitans in Nyasaland, 1 18. P. Palms, relation of, to Glossina breeding places, 40, 41. Paltothyreus tarsatus, in breeding places of Glossina morsitans, 40. Panama Canal Zone, measures against mosquitos in, 111 (note). Periophthalmus, Glossina palpalis experimentally induced to bite, 91. Phacochoerus ethiopicus (Wart-hog), infected with trypanosomes, 94, 95. Pheidole megacephala, in breeding places of Glossina morsitans, 40. Pigs, relation of, to breeding places of Glossina spp., 35, 38, 43 ; relation of Glossina to, 73, 92, 93, 100, 101 ; destruction of, against Glossina, 142 ; danger of spread of sleeping sickness by, 93, 100 ; trypanosomiasis of, 95, 106. Piptadenia buchanani, relation of, to Glossina breeding places, 43. Plexippus paykulli, predacious on G. palpalis in Gambia, 116. Porthetria dispar (Gipsy Moth), 132 (note). Principe, Glossina palpalis in, 6 (note), 9 ; extermination of G. palpalis in, 142, 143. Prionops, probably predacious on Glossina, 114. Procavia, possible relation of Glossina to, 93. Prolaelius glossinae, parasite of Glossina morsitans in Nyasaland, 131. Pternistes, relation of Glossina morsitans to, 37, 115. R. Rabbits, relation of Glossina to, 93 (note). Rats, relation of, to Glossina, 90, 92, 102. Reed buck, infected with trypanosomes, 95. Repellents, for Glossina, 147-150. Rhodesia, Northern, Glossina brevipalpis in, 19 ; G. morsitans in, 2, 14, 32, 33, 38, 39, 40, 41, 43, 48 (& notes), 49 (& note), 56 (note), 65, 76, 79, 80, 83, 90, 100, 115, 116, 120, 121, 122, 123, 124, 130 ; G. pallidipes in, 16 ; G. palpalis in, 8, 9. Rhodesia, Southern, Glossina morsitans in, 14, 36, 40, 49, 65, 79, 96, 97, 99, 102, 103, 106. Rinderpest, relation of, to game and Glossina, 95, 96, 101, 103 ; attacking buffa'o, 48. Robber Flies, as possible enemies of Glossina, 117. 187 INDEX— continued. S. Sarcophaga, parasitised by Chalcids, 131, 135. Savannah Forest, relation of, to habitat of Glossina, 50, 51, 52, 53, 55, 58. Senegal, Glossina longipalpis in, 16 ; G. morsitans in, 51, 159 ; G. palpalis in, 159 ; G. tachinoides in, 11. Senegambia, Glossina morsitans in, 14. Sheep, relation of game to trypanosomiasis of, 95 ; Glossina morsitans experimentally fed on, 89 ; suggested experiments with Glossina and, 106. Sierra Leone, Glossina fusca in, 17, 55 ; G. pallicera in, 10. Sigmodus, probably predacious on Glossina, 114. Silurus, relation of Glossina to, 91. Situtunga, relation of Glossina to, 73, 91 (note), 112. Sleeping Sickness, relation of Glossina to, 1-3, 8, 14, 34, 45, 56, 73, 93, 94, 104, 136, 137, 140, 142, 150. Somaliland, British, Glossina longipennis in, 20. Somaliland, Italian, Glossina longipennis in, 20 ; G.pallidipes in, 16, 24, 33 41, 53. Spalangia, parasite of Musca, 127 ; suggested experiments with Glossina and. 135. Spalangia muscidarum, hosts of, 135. Spiders, destroying Glossina, 112, 115, 116 (& note), 118, 119. Squirrels relation of Glossina to, 97. Stegomyia, disseminating yellow fever in Central America and U.S.A., 111. Stomatoceras (Centrochalcis) exaratum, parasite of Glossina morsitans in Nyasaland, 131. Stomatoceras micans, parasite of Glossina morsitans in Northern Rhodesia and Nyasaland, 122, 123, 127, 130, 131, 133, Stomoxys, method of distinguishing Glossina from, 5 ; characters of, 5 ; attacked by mites, 136 ; probably transmitting trypanosomiasis to cattle, 107 ; repellents for protection of stock against, 149. Stomoxys calcitrans, 5 ; probably transmitting trypanosomiasis to cattle, 106 ; effect of odours on, 110 ; parasitised by Spalangm muscidarum, 135. Strepsiceros kudu, relation of Glossina to, 132 ; infected with trypanosomes, 95. Sudan, Anglo-Egyptian, Glossina longipennis in, 20 ; G. morsitans in, 14, 15, 94 (note), 100 ; G. palpalis in, 8. Sudan, French, Glossina palpalis in, 9 ; G. tachinoides in, 11. Syntomosphyrum glossinae, parasite of Mutilla glossinae, 122, 125, 127 (note), 129, 131, 133. T. Tabanidae, outbreaks of trypanosomiasis among cattle probably due to, 106. Tabanus, drinking water, 88 ; an indirect enemy of Glossina, 1 12 ; dissection of salivary glands of, 161. Tamarind, relation of, to habitat of Glossina, 46. Tamarisk, relation of, to habitat of Glossina tachinoides in Southern Arabia, 47. Tanganyika Territory, Glossina austeni in, 12, 31, 48, 158 ; G. brevipalpis in, 19, 24, 27, 32, 34, 57, 158 ; G. morsitans in, 14 ; G. pallidipes in, 16, 158 ; G. palpalis in, 8. Termites, pupae of Glossina morsitans in nests of, 38 ; attacking pinned insects, 157. Thrixion, larva of, compared with that of Glossina, 26. Thyridanthrax abruptus, parasite of Glossina morsitans in Rhodesia and Nyasaland, 120, 126, 130, 131. Ticks, Glossina tachinoides probably feeding on, 47. 188 INDEX— continued. Toads, relation of Glossina to, 90, 91. Togoland, Glossina fusca in, 17 ; G. morsitans in, 14 ; G. palpalis in, 11 ; G. tachinoides in, 11. Tragelaphus scriptus (Bushbuck), infected with trypanosomes, 94, 95 ; relation ot Glossina to, 35, 47, 112, 132. Transvaal, Glossina morsitans in, 14, 48, 98, 99. Trombidium, larval forms of, probably attacking Glossina, 136. Trypanosoma angolense, Glossina transmitting, 136. Trypanosoma brucei, Glossina transmitting, 136. Trypanosoma brucei vel rhodesiense, game infected with, 95 (note). Trypanosoma caprae, game infected with, 95 (note). Trypanosoma congolense, Glossina transmitting, 136. Trypanosoma gambiense, transmitted by Glossina palpalis in Central and West Africa, 8 ; causing sleeping sickness, 2, 8, 94, 136. Trypanosoma ingens, game infected with, 95 (note). Trypanosoma pecorum, game infected with, 95 (note) ; apparently trans- mitted to cattle by biting flies other than Glossina, 106, 107. Trypanosoma rhodesiense, transmitted by Glossina morsitans; 8 (note), 14 ; causing form of sleeping sickness in Nyasaland and Northern Rhodesia. 2, 8 (note), 14, 94, 136. Trypanosoma simiae, game infected with, 95 (& note). Trypanosomiasis of Domestic Animals, relation of Glossina to, 1-3, 34, 45, 56, 57, 77, 94, 95, 97, 101, 102, 103, 136, 137, 142, 146, 151 ; due to biting flies other than Glossina, 106. Trypanosomiasis, Human (see Sleeping Sickness). Tsetse-flies (see Glossina}. U. Uapaca, relation of, to habitat of Glossina brevipalpis, 58. Uapaca sansibarica, relation of, to breeding places of Glossina, 43. Uganda, Glossina brevipalpis in, 59 ; G. fusca in, 17, 42, 55, 56 ; G.fuscipleurism, 18, 56 ; G. longipenms in, 20, 59; G. morsitans in, 14 (& note), 16, 100 ; G. pallidipes in, 16, 52, 59 ; G. palpalis in, 9, 34, 44, 45, 55, 69, 72, 73, 76, 85, 87, 114, 116, 117, 135, 137, 147; G. tachinoides not recorded from, 11. V. Varanus (Monitor — sometimes wrongly called Iguana), relation of Glossina to, 72, 85, 87, 91 (note), 96. Villa lloydi, parasite of Glossina morsitans in Northern Rhodesia, 120, 130, 131, 133. Vultures, relation of Glossina to, 91. W. Wart-hog, relation of Glossina to, 47, 80, 92, 140 ; relation of, to breeding places of Glossina morsitans, 38, 41 ; danger of spread of sleeping sickness by, 100. Waterbuck (see Cobus ellipsiprymnus] . Wildebeeste, relation of Glossina to, 132. Y. Yellow Fever, disseminated by Stegomyia in Central America and U.S.A., Z. Zebra, relation of Glossina to, 132. Zululand, Glossina austeni in, 12 ; G. morsitans in, 14 ; G. pallidipes in, 16, 31, 54, 103. NON-CIRCULATING BOOK BIOLOGY ,13 686636 BIOLOGY UNIVERSITY OF CALIFORNIA UBRARY