Mote e 2 FR OT teat orhtiecin ted eSe-gies! Shir ier 8 APE AD I Melt « Ce lh dal tin be MS oh: PO SIN tri es ehcp ia > the A Sindh ded ra°pia lata ttebalerarare-avirgretcate. PUT MOREE eral el oie! viehripte sy ere eR Twit ree NP rig ety iphate by tpt gag re - errr n a FOS wlytacerptetens: 2 : 2 iabamendy pores Vy ewes Pre eree wre ile a i ie alk | : bw PeVY rr t? i A tee wea ee eae oe eee Sy" ee ewe ee FW ce F ole ew wee we” EEE vr ‘ ie ‘ *: a a te! ete sie cl " weet reer ete : ceo we era 6 ‘ 4 rch naw ? : : ‘rr ce Ceres HEE tes gy ; de adhe d a ae we a hey i ae cerry ss ee vn ie ee PEP OO CRETE ENE Pe ee Py ‘vo ere wt + 2 ‘ee ? ve 2 vw & pln 4 Pad . , ri ‘ psy at - ° a . we} . 7 oe ireve eva Teese gee es eA Ae eI ee : we ere ; wna / evepane ‘ trere vy reer ee ee 22 6s ese yh. tote byt ‘ : rr i — w'w's ere , 4 & —— 7 ‘ves pte a hs - ; 5 : 3 — . —_ ¢ "ew ee i> + * . a . “ . ue ov : ‘ errs od % . cin e® Pepe ey wee ote ve’ Gwetwelel ere ers ee ee ee SMITHSONIAN. DEPOSIT Ps a Py Lt A. ‘ r i 4 k Vv ; a) 4 % * wy te : Z af YY ° OF ENTOMOLOGICAL RESEARCH ISSUED BY THE IMPERIAL BUREAU OF ENTOMOLOGY. EDITOR: THE DIRECTOR. VOL. X. LONDON: 1919-1920. IMPERIAL BUREAU OF ENTOMOLOGY. Honorary Committee of Management. VISCOUNT HARCOURT, Chairman. Lieutenant-Colonel A. W. Ancocx, C.LE., F.R.§8., Juondon School of Tropical Medicine. Major E. E. Austen, D.S8.O., Entomological Department, British Museum (Natural History). Dr. A. G. Bagsnawe, C.M.G., Director, Tropical Diseases Bureau. Major-General Sir J. Ross Braprorp, K.C.M.G., F.R.S., Secretary, Royal Society. Major-General Sir Davip Brucz, K.C.B., F.R.S., Chairman, Governing Body, Lister Institute. Mr. J. C. F. Fryer, Entomologist to the Ministry of Agriculture and Fisheries. Sir Sipney F. Harmer, K.B.H., F.R.S., Director, British Museum (Natural History). Professor H. Maxwsiit Lerroy, Imperial College of Science and Tech- nology. Hon. E. Lucas, Agent-General for South Australia. Dr. R. Stewart MacDoveaut, Lecturer on Agricultural Entomology, Hdinbureh University. Sir Joon McFapynan, Principal, Royal Veterinary College, Camden Town. Sir Parrick Manson, G.C.M.G., F.R.S8., Late Medical Adviser to the Colonial Office. Sir Danret Morets, K.C.M.G., Late Adviser to the Colonial Office in Tropical Agriculture. Professor R. Newstreap, F.R.S., Dutton Memorial Professor of Medical Entomology, Liverpool University. Professor G. H. F. Nurratz, F.R.S., Quick Professor of Protozoology, Cambridge. Professor E. B. Pouuton, F'.R.S., Hope Professor of Zoology, Oxford. Lieutenant-Colonel Sir Davip Pratn, C.M.G., C.I.H., F.R.S., Director, Royal Botanic Gardens, Kew. Sir H. J. Reap, K.C.M.G., C.B., Colonial Office. The Honourable N. C. RotuscuHip. Dr. Hueu Scort, Curator in Entomology, Museum of Zoology, Cambridge. Sir ArntHur H. Suretey, G.B.E., F.R.8., Master of Christ’s College, Cambridge. Mr. R. A. C. Speruine, C.M.G., Foreign Office. Sir Stewart Stockman, Chief Veterinary Officer, Ministry of Agriculture. Mr. F. V. TuHeopatp, Vice-Principal, South Hastern Agricultural College, Wye. Mr. C. Warzsurton, Zoologist to the Royal Agricultural Society of England. The Chief Entomologist in each of the Self-governing Dominions is an er officio member of the Committee. General Secretary. Mr. A. C. C. Parxrnson (Colonial Office). Director and Lditor. Dr. Guy A. K. Marsnatu, C.M.G. Assistant Director. Dr. 8. A. NAVE. Head Office—British Museum (Natural History), Cromwell Road, London, 5.W.7. Publication Office. — 88, Queen’s Gate, London, 8.W.7. : BFE " wit evi ‘ j & | . fern *s, e . Wiberg eg ’ ¥ Tae * Ws te Seid Tis 4 | Li yon PRLHASTS | Taye | * iy a yate : AAe ‘ hid ‘y Aeeae ; : if , \ sj | ia ; am , ia cee : : A i : ee j par a ays ‘ ef ¢ (vale 1 ri y : . ns pe 1% ; . : m cay aha] Bes | e : a jt i sedate ; | : rau P jad : #« i" ae hey ey rN ‘| Ab weft Gt i ay ny ed ih an So el Aire im es ae hs Sa fain | re aa nat fe Vii paugdr ae ne ‘7 ; i es , j wis . . ; ; rv aya ae i WELT tt dick Y OW A aad ) oe a io r ¥ 3 - a ae | j ‘eit ii } an a es Pe | ete : bi it Tk ae aWaliite® he he hin TS Yuta ; Aaa the | shld heey ules nih Lig TiN i beans ne iW iuaites, ae ose esl ey 7 l re ik a GA orn itii v eon ‘pb Niglae a er tt AL Rg rt LB iii Aste Bil beh s ats : : ' ; ens, law i eu ‘ ; ¥: CONTENTS. ORIGINAL ARTICLES. ADERS, W. MANSFIELD-. PAGE. Insects Injurious to Economic Plants in the Zanzibar Protectorate.... 145 Notes on the Identification of Anophelinae and their Larvae in the AFA Zany Aled toate GE C UOT UCI) oy 0 so) wc a0 eevee! «8 hao] s+ o> Gow dh> atch opegae = enprstass 329 Austen, Masor E. E. A Contribution to Knowledge of the Tabanidae of Palestine........ 277 Baker, Dr. A. C. Wicaranaprena wiolae, Lheo,, and its Allies 7.520). 4552462525 c4e5 oes 45 BaRRAUD, Lizut. P. J. Notes on some Culicidae collected in Lower Mesopotamia.......... 323 Brzzi, Pror. M. A new Australian species of Reora with a remarkable Life-habit (Dipt. ; TURF JEIOCI CES) ages a 3 eae rr ee eree ABE. Notes on the Ethiopian Fruit-flies of the Family Trypaneidae, other 1H SEI g JOCKS Ul aie tick args re ie. Aa ee tee Beare aera 211 Bratn, Cuas. K. ites Coccidae om South Adricas—ivs SG) See, AA GE 95 Cresrm, i.’ D., Jnr. Acuschubenrcedimor Caress. pow cont Sob uisite kale oll wad al. 43 Distant, W. L. . A new Lygaeid Bug found among Stored Rice in Java.............. 41 Duke, Dr. H. LynDHURST. An Enquiry into the Relations of Glossina morsitans and Ungulate Game withiepectal reterence to imderpest.. 2.02 02.25..-.02-)--- 7 Epwarps, F. W. Pios@ui vo NO@vesh spate PLE COR tote ME a oi wine's cee aie, « 129 Fisker, W. F. Investigations into the Bionomics of Glossina palpalis.............. 347 Incram, Dr. A. The Domestic-breeding Mosquitos of the Northern Territories of the Cre Gln Eee erect tS o's) Se us cucievatt Blelaline eco ea Y si wle-ers aye e ede 47 (693) Wt.P2/154. 1,000. 9.20. B.&F.,Ltd. G11/14. Vi. CONTENTS. InecraM, Dr. A., & Macriz, Dr. J. W.S. The Early Stages of West African Mosquitos—iv.................. 59 JACK, RUPERT W. Tsetse Ply im Southern Rhodesia, AGS 20... ... 22. eee 71 Macrir, Dr. J. W.S. The Chaetotaxy of the Pupa of Stegomyia fasciaia................ 161 Macriz, Dr. J. W. S., & Ineram, Dr. A. The Karly Stages of West African Mosquitos—iv................. 59 MacGrecor, Cart. M. E. On the Occurrence of Stegomyia fasciata in a Hole in a Beech Tree in Epping Borest.\ . sles oa oe ia Ge eee ee ee een eee een eee oe MarsuHat., Dr. Guy A. K. Some Injurious South Atricam Weevil Saye. eyelet eee ee ere 273 Mauvtikx, S. A new Hispid Beetle injurious to the Oil Palm in the Gold Coast...... al Mutr, F. On ‘some, African Delphacidae (Homoptera)--\ 5.22.0 epeee ee 139 NEWSTEAD, ROBERT. Observations on Scale-Insects (Coccidae)—vi...................... 179 Sranton, A. T. The Mosquitos of Far Eastern Ports with special reference to the Prevalence of Steqomyva fasciata, B.S sce See ee 333 TRAGARDH, Ivar. On the Use of Experimental Plots when studying Forest Insects..... 157 TURNER, R. E. On a New Mutillid Parasite of Glossina morsitans.........0+-+-+00- BAN VeItcH, RoBERT. oar olbte of Notes on the more important Insects in Sugar-cane Plantations in Fyji 21 MISCELLANEOUS. Collections received: : 20 ii.0. 2. oe ee eee 93, 209, 345, 465 Vi. PLATES. PAGE. ule Wings of Neotoxopiera violae and its allies.......... facing 46 II.—IV. Views of habitats of Glossina morsitans in 8. Rhodesia ,, 90 V.—XII. SOM na AeaMCat COCCIMACK le. sie ents sd oles oe wks we ts XIIT.—XV Views showing ravages of Oryctes in Coconut Palm.. a LOG 2OWil SOIC NS eL CLS Biv | ko en eee a 208 eV avail Wines of Trypetidae. 32 oii... eee eee ge XIX. Injurious South African Curculionidae............ ap AS: MAPS. PAGE. PiearGot hx periment OW GIOSSUNG. ca... wee eke bee ee eee es ees to face 17 Charts showing density of Glossonain Uganda.......... 372, 388, 411, 423, 458 Map showing increase of G'. morsitans in Sebungwe District......... io face 90 Pepeol Soummern ty Nodesian (2.00 Nese Mie menenel SUL ay ORD cote) cove, we) < ol aceu cue, sche 0/4 cirn pd: Sielnce eda) olen «ma, 6 es a. BO Nout? ae ILLUSTRATIONS IN THE ‘ Details of :— Rioxa termitoxena, Bezzi,sp.n. .. Rhabdocnemis obseura, Boisd. . Pupa of Anopheles pretoriensis, Theo. eae . rufipes, Gough Ochlerotatus hirsutus, Theo. Bey gi Be nigeriensis, Theo. » » Culex ager var. ethiopicus, Edw. Pe hiie's » guasigelidus, Theo. » wunivittatus, Theo. Male cenitalia, Culex aurantapex, Tidw. 39 39 cs 49 Inicranotroms nigeriensis, Muir, Sp. n. a Ps bridwellt, Muir, sp. n. ay ae Ae ibadanensis, Muir, SI. Tsny' li fs Delphacodes bridwelli, Muir, Sp. n. we Meqamelus furcifer, Horv. ve is ue Bs kolophon, Kirk as i x Be nigertensis, Muir, var. n. flavolineatus, Muir, a n. Oly peococe cus hempeli. CKIL, g . Pseudococeus mquilinus, Newst. , Sp. n., on se perniciosus, N. & W., var., 2 Yilamentosus, CkIll., Q s Pseud lophilippia inguilina, Newst., sp. 0., ce Antonina waterstont, } Newst., sp. n., e 183; Pulvinaria brevicornis, Newst., sp. n., - broadwayt var. ae Newst., nov. 4 Lecaniwm subacutum, Newst., sp. n., DLecanium (Eucalymnatus) decemplex, Newst. ae sp. 1, a inguilinum, Newst.,spn., Q . Ke yh deformoswm, Newst., sp. n., on a nigrum var. nitidum, Newst., Q ae Platysaissetia montrichardiae, Newst.,sp.n., Q . Aspidiotus (Chrysomphalus) apicatus, Newst., sp. Dn. cs Be umbontferus, Newst. s Be re articulatus var. magnospinus, Newst., A (Odonaspis) rhizophilus, Newst., sp. t., 9 on florineides, Newst.,sp.n., 2 .. Chionaspis madiunensis, Zehnt., Q nd fenundieutus. Newel: Spe ile 9 5 praelonga, Newst., Sp. n., 2 a auratilis, Newst., sp. n., @ “a dura, Newst., sp. n., © laniger, New st., sp. Tie a H aematopota sewellt, Aust..sp.n.. ¢& o Chart of effect of Rinderpest Blood on Glossina on Trochorrhopalus strangulatus, Gyl. Ar Rhopaea subnitida, Arrow ; R. pee Arrow Adoretus versutus, Har: ne Simodactylus cinnamomeus, Boisd. Lacon eiehcollis, \QeriACe ee Monocrepidius pallipes, Esch. Ampera intrusa, Dist., sp. n. Portable breeding Gace... _ OAR), B40, TEXT. x. ILLUSTRATIONS IN THE TEXT. Map of Northern Territories, Gold Coast : Map of experimental plot in pine forest, Sweden Diagram showing distribution of pine pests Cephalo-thorax of pupa of Stegomyta fasciata Abdomen of pupa of Stegomyia fasciata .. a ns Coelaenomenodera elaeidis, Maulik, sp. n., 172; larva.. Aspidoproctus gowdeyi, Newst., sp. n., g genital armature Aspidiotus longispina, Morgan, 9 puparium Head of :— Chrysops punctifera, Lw., 2 4 Haematopota sewelli, Austen, sp. D., J 3 wing of 0 x minuscularia, Aust., sp. Nn. : Bs minuscula, Aust., sp. 0. innominata, Aust. a Ml. Tabanus alexandri nus, Wied., i insecuior, nist, sp. n., ie i yt > rupinae, Aust., Sp. ., 3 Q, 304 : ‘imago, OTe > arenwagus, Aust. NSD. Dent) a 308 ; imago, © >» accensus, Aust., Sp. n., co 311; imago, ¢ 5% leleant, Aust., Sp. n., $2 O, 315; imago, 2 >» pallidipes, Aust., Sp. 0, Q, and i imago dalei, Aust., Sp. D., 3 ; 31 9; wing Mutilla auxiliaris, Turner, sp. n., 2 ¢ Anopheles funestus, thorax of larva, plumose hair enlarged Sketch-map of the Kome-Damba Group of the Sesse Islands harts re Glossina on islands and shores of Victoria Nyanza 48 158 159 SiG 163 165, 168 173, 174 ss Liregl . 194 279 283 286 286 286 298 298) 303 306 310 312 318 329 32 330 352 "370, 371, 429. 431, 445 Xj. ERRATA. Page 20, line 42, for Sleep read Sleeping i ee Sky San NACL » world » 135, fig. I, ,, ame >) | Same ee AON a 1 ,, Dicranotropis ,, Delphacodes el] ene 34." 5: Ehyncophorus ,, Rhynchophorus Pts Se Cyelas » Cylas * 183, LOO LACES » facies of he POU ea MMe Saag) So ae) > 200, 5, 39, »,, pernicillaris » perspicillaris 29 321, ) a, Dy) Mie Oe M an 465, » 42, 5 Jawaera » Goldberg Page 64, line 15, for “6 mm.” read ‘‘0°6 mm.” 9 3) 29, Transpose ‘‘ 8—9 hairs’ and ‘ 5—7 hairs ”’ 69, fig. 7, The upper enlarged pecten spine should bear no secondary spines. 161, line 7, For “ Circum-ocular’’ read ‘“‘ Post-ocular ”’ 993 Lol, » “Superior” 28, 29 99 £ yr) 99 29, ,, ‘“* Posterior’”’ » ‘Median ”’ ole 4) “oC Amterion=’ », ** Lower anterior ”’ » ‘** Upper anterior” 33, 5, ‘*Antero-inferior”’ ,, ‘‘ Lower posterior ”’ 34, ‘‘ Postero-inferior”’ ,, ‘“‘ Upper posterior ”’ 36, After “these setae” insert ‘<(previously referred = as the inner lateral row.—Bull. Ent. Res. x, p. 59.)” 41, After «these setae” insert “ (previously referred to as the outer lateral row.—Bull. Ent. Res. x) py O94)” 19, For “‘ seta single ”’ read ‘‘seta; single ”’ Si es (iP) 99 Bs 66 (ee) 39 ee ee Er a A SN nat io ex IMPERIAL BUREAU OF ENTOMOLOGY. BULLETIN OF ON TOMOLOGICAL. RESEARCH. Won, X:. 1919. A NEW AUSTRALIAN SPECIES OF RIOXA, WITH A REMARKABLE LIFE- HABIT (DIPT.; TRYPANEIDAK). By Prof. M. Bzzzz, Turin, Ttaly. Dr. Guy A. K. Marshall has recently submitted to me a Trypaneid, discovered in Northern Australia by Mr. G. F. Hill. The fly is said to have been bred from larvae having habits very different from those of the related members of the family ; and being moreover interesting from a morphological and biogeographical standpoint, it forms the object of the present note. The Oriental species of CeRaTITINAE with 6 bristles on the scutellum, with a complete thoracic chaetotaxy, with a bristly third longitudinal vein, with a not wavy second longitudinal vein, and with a wing-pattern consisting of an extended brown patch with hyaline spots and hyaline indentations, have been ascribed by me* to the two genera Diarrhegma and Riova. While the first of these genera is less numerous and is well characterised by the form of the head, the second contains heterogeneous elements, which have been divided by me into three groups. Fol- lowing on this, Prof. Hendel} erected the new genus Hexacinia for the peculiar group of stellata, Macq., and divided the remaining species into the genera Rvoxa (with a pectinate arista) and Rioxoptilona (with a plumose arista). But this last distinction seems to be a premature one, as it is impossible to draw a sharp line of separation between the species in which the arista is plumose on the upper side alone, and those in which it is plumose on both sides. I have before me two species of Australian TRYPANEIDAE which are referable to the genus Rioxa, s.l.; they closely resemble one another both in body coloration and wing markings, but may be distinguished as follows :— 4 1(2). Two upper and two lower pairs of orbital bristles present; arista regularly pilose, with some short hairs only on the under side; mesonotum and scutellum without dark spots, the latter with the middle pair of bristles weaker than the others; discoidal cell with two hyaline spots, which are confluent with the hyaline indentations of the hind border musae, Froggatt * Mem. Ind. Mus., Calcutta, 1913, iii, pp. 53-175, pl. viii-x (vide pp. 108 and 111 t Wien. Ent. Zeit., 1914, xxxiii, pp. 73-98 (vide pp. 78 and 82). (C572) P4§140. 1,000. 10.19. B.&F.Ltd. G11. A PROF, M. BEZZI. 2(1). Only a single upper and a single lower pair of orbitals; arista pilose on both sides and in the male with a tuft-like pilosity at the end; mesonotum in front of the scutellum with two blackish spots, which are extended over the scutellum itself; middle scutellar bristles as strong as the others ; discoidal cell with but one isolated hyaline spot .. termitoxena, sp. n. Rioxa musae, Froggatt, 1899. oe musae, Froggatt, Agric. Gaz. N.S.W., Sydney, 1899, x, p. 501, pl. u, figs. -3; id., “ Austral. Insects,” Sydney, 1908, p. 308; id. ae on Fruit-fly sad ihe Pests in various Countries, 1907-8, Dept. Agric. N.S.W., 1908, p. ai pl. vu, figs. 1-3. Rioxa (Trypeta) musae, Froggatt, Proc. Linn. Soc. N.S.W., 1911, xxxv (1910), p. 872. Rioxa musae, Bezzi, Mem. Ind. Mus. 1913, i, pp. 76 and 114. The present species was originally obtained from bananas brought to Australia from the New Hebrides; but it has been found subsequently breeding in both culti- vated and native fruits in Queensland and New South Wales. 2. Rioxa termitoxena sp. nov. (fig. 1). Differing from all the other known species in having only a pair of upper orbital bristles, and very distinct on account of the peculiar arista of the male. 42. Length of body, 65-7 mm.; of the ovipositor, 155 mm.; of the wing, 7-75 mm. Head (fig. 1, A) entirely reddish-yellow ; occiput quite unspotted and shining, chiefly near the borders ; it is paler on the sides beneath, and the lower swellings are not developed. Frons opaque, with the middle stripe more dark reddish (sometimes brownish on the fore half in the female), with more yellowish sides and with a black ocellar spot; it is clothed with rather dense, short and equal, black hairs; lunula yellowish. Face shallowly but distinctly concave above the rather prominent mouth-border ; the broad keel between the antennal grooves is flat; cheeks very narrow, linear, whitish; jowls broader than the third antennal jot, unspotted. Fig. 1. Rioxa termitoxena, sp. n. a, head of male ; b, wing of fully coloured female. Eyes rather broad, but with the vertical diameter longer than the horizontal one. Antennae inserted near the middle of the eyes, shorter than the face, and entirely reddish ; first joint very short, with black hairs at the sides ; second joint short and globular, with short black hairs; third joint about twice as long as the first two joints together, almost bare, poe at the end, only slightly concave above: arista. with rather long and scattered hairs from the base to the end on the upper side, and on the apical half alone beneath; in the male the terminal hairs of the A NEW AUSTRALIAN SPECIES OF RIOXA. 3 arista are closer together on both sides, forming thus a kind of apical tuft, which is not clearly indicated in the female. Palpi broad, yellowish, reddish at the outer border, with yellowish hairs and some black bristles; proboscis dirty yellowish, with pale hairs; the hairs of the lower part of the head are likewise pale yellowish. All the cephalic bristles are black, those of the occipital border being numerous and pointed ; the postvertical are long; outer vertical half as long as the inner one ; no ocellar; only a single superior orbital,* the basal one; only a single inferior orbital, the apical one, which is placed very near the antennae, and is rather weak and short ; genal bristle well developed, black. Thorax on the back entirely reddish and rather shining, the very short pubescence black and the whitish dust very sparse ; the humeri, the sides above the notopleural line, and a broad postsutural stripe above the root of wings, are of a more yellowish colour, but not sharply defined; along the dorsocentral lines there is sometimes in fully coloured specimens a brownish, ill-defined longitudinal stripe ; the two blackish spots in front of the scutellum are rounded and lie between the praescutellar and inner pair of the posterior supra-alar bristles, not extending in front beyond the praescutellar bristles. The pleurae are entirely reddish and shining, quite unspotted, even the hypopleural spots being not paler than the surrounding parts; the short hairs are mainly black. Scutellum triangular and of rather large size; it is flattened above, with well developed lateral keels; it is entirely yellowish, but on each side there is a triangular blackish or brownish spot in contact with that at hind border of the mesonotum ; it is clothed on the sides with short black hairs, and bears six equally strong, black bristles, even those of the middle pair being as strong as the others, and those of the apical pair being parallel or diverging. Post- scutellum and mesophragma shining black (in fully coloured specimens), with a narrow, reddish, middle longitudinal stripe. Thoracic chaetotaxy complete; all the bristles are black, even the scapular ones, the middle pair of which is as strong as the lateral ones and has its bristles rather distant ; praesutural bristles rather weak and short; dorsocentral bristles placed much behind, about on a level with the praescutellar ones; one strong mesopleural bristle, and 2-3 more weak ones below it; pteropleural bristle as strong as the sternopleural one. Squamulae and halteres whitish, the former with white hairs at the border. Abdomen elongate, slightly constricted at base, and about as broad as the thorax at the end of the second segment; it is shining reddish, with broad, shining black bands at base of segments 3-5, the bands being in the female broadly interrupted in the middle, while in the male they are entire or nearly so, the last segment being entirely black; the hind borders of segments 2-5 are pale yellowish with whitish dust, thus contrasting strikingly with the black basal bands. Male genitalia black, rounded, prominent below; ovipositor shining black, depressed, with the basal segment as long as the last two abdominal segments together. Venter entirely reddish-yellow, unspotted. The short abdominal pubescence is black, like the bristles at the sides and end; the two basal segments are clothed with soft, long whitish hairs at the sides. * In the allied genus Pitlona there is likewise but a single s. or., but it is the apical one. In the recently described genus Ortaloptera, Edwards (Trans. Zool. Soc., 1915, p. 419, pl. xxxvii, fig. 9), from New Guinea, the cephalic chaetotaxy is very like that of the present species, while that of the thorax is much more reduced. (C572) | a2 4. PROF. M. BEZZI. Legs entirely and evenly reddish from the coxae to the end of the tarsi; their short pubescence is pale, but it is black on the outer side of the femora; front femora with long black hairs above, and with 3-5 black bristles below at the end; middle tibiae with two apical black spurs, and with a short row of 2—4 short and stout black bristles at the base on the hind sides ; hind tibiae with a complete row. Wings (fig. 1, B) long and broad, with a well developed costal bristle; the stigma is of usual length; the second longitudinal vein is straight; the third is bristly throughout, moderately bent backwards beyond the middle of its last portion, and parallel with the last portion of the fourth; small cross-vein below the end of the first longitudinal vein and beyond the middle of the discoidal cell; hind cross-vein long, as long as or even a little longer than its distance from the small one, reaching below the fifth vein at an obtuse angle ; point of the anal cell acute and longer than the second basal cell. The base of the wing is hyaline to a little before the end of the second basal cell and to the base of the stigma; but in fully coloured specimens the second costal cell, the basal half of the first basal cell, the base of the second basal and more narrowly that of the anal cell, are pale yellowish. The stigma is entirely dark brown, without any hyaline part. The brown patch extending across the greater part of the wing is darker on the upper half, lighter on the lower, and distinctly yellowish in some parts towards the middle. At the fore border there is a single hyaline indentation of triangular shape just beyond the stigma, reaching with its point the third longitudinal vein, a little beyond the small cross-vein. The three rounded hyaline spots in the middle are as follows: one at the end of the first basal cell, before the small cross-vein; one a little before the middle of the first posterior cell, above the upper end of the hind cross-vein; and one of greater size before the end of the discoidal cell, in contact above with the fourth vein, and distant beneath from the fifth vein. At the hind border there are two hyaline indentations ; a smaller one of more triangular shape in the second posterior cell, pointing with its inner corner to the upper basal angle of the same cell, without reaching it ; a larger and broader one towards the middle of the third posterior cell, with its inner, broad end in contact with the fifth vein, but without entering the discoidal cell. The axillary lobe is mainly hyaline, with a brown spot at the end of the anal cell, not extending beyond the axillary vein; the alula is hyaline. The narrow extreme base of the marginal cell is hyaline, like a small elongate spot into the first basal cell, just before and above the end of the second basal cell; the whitish oblique stripe into the brown base of the third posterior cell is very striking. The indentation of the fore border and the three hyaline spots in the middle are distinctly whitish, like the inner or superior part of the broad indentation of the third posterior cell. Type 3 (British Museum) and an additional specimen of female sex from North Australia, Port Darwin, 14. x. 1914 (G. F. Hill), bred from galleries of a termite, Mastotermes sp., in tree-trunks ; the specimens of this couple are not fully coloured, as is frequently the case with bred flies. Type 9 (British Museum), and an addi- tional specimen of the same sex, from same locality and same collector, 8, vii, 1913 ; these specimens are fully coloured, having been taken on the wing. Macquart (Mém. Soe. Sci. Lille, 1855, p. 144 [124], pl. vu, fig. 7) has described Urophora bicolor from Adelaide (type in Bigot’s Collection), which is evidently a A NEW AUSTRALIAN SPECIES OF RIOXA. 5 Rioxa, but differs from both the preceding species in the wing pattern. Froggatt (Australian Insects, p. 308) reports it as a Trypeta as follows: “A larger native species with reddish brown head and thorax; with black body; with beautifully mottled black wings having the base and sides unclouded. I have taken it on the trunks of wattle trees near Bathurst, N. 8. Wales.” It is probable that this species is different from that of Macquart; the habit of settling on trunks of trees is very suggestive of that of the new species here described. In his report of 1909 Froggatt has briefly compared (p. 114) this 7. bicolor with his own species 7. musae. Macquart in the above-cited work (1847, p. 109 [93], pl. vi, fig. 7) has also described from Australia (coll. Bigot) Tephritis lugubris, which is likewise possibly a Rioza, but has a very different wing pattern and body coloration. The Trypeta por nia of Walker (List, 1849, p. 1039), from Port Stephenson, is also probably a species of Rzoxa. It would be interesting to know if the reduced cephalic oe (by the want of the second upper orbital bristle) is a character of the true Australian species of Riora ; in this case it may be considered of generic value, chiefly when supported _ by such a different kind of life-history as that observed for the new species 7’. term- toxend. Of the numerous species recorded or described from New Guinea, only the following seem to belong really to Rioxa: formosipennis, Walker, 1861; lateralis, Kertész, 1901; sexguttata, De Meijere, 1913; and seriata, De Metjere, 1915; but they all have a wing pattern very different from that of the species here described. The other species, like debeauforii, De Metjere, 1906, nigra, De Meijere, 1906, msigms, De Meijere, 1913, nigripennis, De Meijere, 1913, and flava, Edwards, 1915, seem to belong to other genera ; quadrifera, Walker, 1861, was placed by Prof. Hendel in his new genus Themaroides in 1914. Pa, a seb laa mee ce 4. We y Ay ayy e we Wee ie Wi ba « 5 sy Re: AN ENQUIRY INTO THE RELATIONS OF GLOSSINA MORSITANS AND UNGULATE GAME, WITH SPECIAL REFERENCE TO RINDERPEST. By H. Lynpuurst Duxe, M.D., D.T.M. & H. (Camb.), Government Bacterrologist, Uganda Protectorate. In the course of a conversation in November 1917, Mr. E. Hutchins, the Chief Veterinary Officer, Uganda Protectorate, drew my attention to an outbreak of rinderpest in the Northern Province of the Protectorate, which first manifested itself in the buffalo of the Chopi Country. The disease was then rapidly spreading through the game in the Glossina morsitans area between Masindi Station and the Kafu River. He pointed out the excellence of the opportunity for investigating the alleged relationship between rinderpest and fly. As I had worked from May till September 1914 in the Northern Province fly belt in association with Mr. W. F. Fiske, I was in a particularly advantageous position to estimate any alteration in the fly distribution which might have occurred as a result of the epidemic. The question had a special local significance, as, should the disease prove fatal to the fly, the introduction of rinderpest among the situtunga on the Islands of Lake Victoria would simultaneously remove the reservoir and the carrier of the mammalian trypanosomes. My release from military duties was duly sanctioned in November 1917, but unfortunate circumstances over which I had no control postponed its fulfilment until April 1918. By this time the disease had almost worked itself out in the fly area, and, what is more to be regretted, the long-deferred rains had com- menced throughout the district. The heart of a unique opportunity was thus lost. The Masindi fly belt had been studied in 1914 for several months by Fiske, a highly trained entomologist with considerable experience in the scientific handling of insect pests on a large scale. His careful observations on the bionomics of the fly in this belt are of the greatest value in arriving at conclusions regarding the present problem. There is a strong consensus of opinion among Huropeans of experience in South Africa that the great rinderpest epidemic of the nineties resulted in a marked decrease in the tsetse in the fly areas through which the disease passed, amounting, in some districts at any-rate, to the immediate or eventual disappearance of the fly. The explanation of this phenomenon most generally advanced is that the fly dis- appeared as a direct result of the great destruction by the disease of the wild game upon which it depended for food. The fly were, in other words, starved to death. In this paper it is proposed first to review the somewhat meagre evidence bearing on the interrelation. of game, fly, and rinderpest, and to discuss the same with the aid of personal experience and observation. Then to proceed to the actual experi- ments which were undertaken to decide whether rinderpest blood, per se, is or 1s not destructive to Glossina ; and, finally, to endeavour to arrive at some kind of under- standing regarding the relations of fly and game and the conditions which lead to a disappearance or diminution of Glossina. My apology for presuming to attack . so complicated a problem is that I have recently been privileged to witness the 8 DR. H. LYNDHURST DUKE. occurrence in Uganda of a marked and independent diminution in the numbers of two of the best known species of Glossina, and also an outbreak of rinderpest among the game of a morsitans area. It is to be hoped that the observations made on the spot may be of value in throwing additional light on the bionomics of Glossina. My thanks are due to His Excellency the Governor, Sir F. J. Jackson, who has throughout evinced an interest in these investigations, and has kindly given me the benefit of his wide experience of game and fly in South Africa. Also to Mr. Hutchins, C.V.O., whose valuable advice and assistance I have found of the greatest help. I am indebted also to the Administrative Officers for their assistance in obtaining food in the face of a general scarcity caused by the famine. Historical and General Evidence from the Literature. As regards the disappearance of Glossina morsitans from various districts in Rhodesia and the Transvaal following on the rinderpest epidemic of the early nineties three explanatory theories demand special consideration :— (1). That a fly which absorbs the blood of an animal suffering from rinderpest dies as a direct consequence. Death might be immediate, and due to poisonous by-products circulating in the blood; or it might result subsequently as a result of the development of the rinderpest organism in the tissues of the fly. Again, rinderpest blood might exert an injurious effect on reproduction. (2). That the fly disappeared owing to loss of its normal food supply, the wild game. A special development of this theory supposes that certain species, especially buffalo, are more responsible than others. (3). That the phenomenon had no direct connection with the rinderpest, being due to some obscure coincident cause. With regard to rinderpest blood, per se, killing fly, we have reliable evidence that in certain districts no difference was observed in the numbers of the tsetse, in spite of the passage of the disease through the game of the fly belt. Sir Alfred Sharpe (1) writes that when the rinderpest visited Nyasaland, there was a noticeable decrease in the numbers of big game in some of the fly belts, especially of buffalo, “but I have never noticed that this had any effect on the Glossina.” Val Gielgud (2) writing from N. E. Rhodesia says: “ Since the almost total destruction of buffalo by rinderpest these flies have not disappeared, although, perhaps, not so numerous as formerly. Stevenson- Hamilton (3) remarks, ‘“‘ while there is no doubt that @. morsitans (?) abso- lutely disappeared from considerable areas during or immediately after the epidemic, we have it on reliable authority that elsewhere it was in no way affected.” Sir David Bruce (4), in his evidence before the Sleeping Sickness Committee i October 1913, states that, in response to his request, Montgomery in British East Africa fed “ a lot of flies on rinderpest animalsand it didthem no harm.” I recollect sending the pupae from Mpumu to Montgomery for these experiments, the report of which I have not been able to obtain. The pupae were those of G. palpalis. I have, however, always understood that owing to difficulties of transport of the pupae, etc., it was found to be impossible to pronounce an opinion as to the effects of the rinderpest organism on the fly, except in as much as the blood of the sick animals was apparently not immediately fatal, Further evidence on this important point will be produced in a later section. THE RELATIONS OF GLOSSINA MORSITANS AND GAME, 9 Regarding the second explanation, that fly disappeared because their food supply, the game, was destroyed by the rinderpest, there is considerable difference of opinion. While the majority of the observers hold that the fly are more or less intimately dependent on the game for their existence, others, and among them Stevenson- Hamilton, Sir F. J. Jackson and Neumann, point to the existence of fly areas in which, they state, game is either absent or in numbers inadequate to constitute the sole food supply for the existing tsetse. If it were possible to discover an area in which tsetse exist year after year and from which the presence of ungulate game and their attendant carnivora could be completely excluded, we should then know that the fly could subsist entirely on baboons, monkeys, small mammals, birds and reptiles. But, so far as I can discover, no such game-free fly belt has been found. Even if such did exist, however, it would not necessarily follow that the sudden or even the gradual disappearance of the ungulates from a tsetse area rich in game would not ultimately be detrimental or even fatal to the fly. Apparently rarity of the game is a different proposition ; the human standpoint may well differ from that of the fly in such matters. Thus, little is known at present regarding the length of time wild fly can exist without blood, or the relation between food supply and the production of larvae. I have kept wild captive palpalis alive without any food for 10 days, and it is highly pro- bable that fly under natural conditions can exist much longer without blood. The hardiness of the fly, coupled with the local habits of such game as duiker, dikdik, warthog and bushbuck, must be borne in mind in dealing with an apparent absence of game from a fly belt. One of the best authenticated examples of the occurrence of tsetse in a locality almost totally devoid of ungulate game is the fly belt mentioned by Stevenson- Hamilton (3) in the Nyassa district of Portuguese East Africa. His personal observations as to the rarity of the game at the time of his passage, the dry season of 1908, must be accepted as correct as coming from a careful observer and a keen hunter. It is, however, plain from his reports that big game did exist in the neighbourhood, though in very smal! numbers. He himself mentions a “ few kudu ” in the narrow strip of country between the fly belts; and Howard (5) re- ferring to Hamilton’s trip mentions one waterbuck having been shot, the presence of a few elephant, and the finding on one occasion of buffalo spoor. The party traversed the bush along a main native track, and their opportunities for studying the fauna of the belt were limited to the vicinity of this track. Howard produces evidence showing that 5 or 6 years before Hamilton’s visit this district was full of big game, and it is evident from the account of Barnett’s trip along the Rovuma River (6), that in 1910 both game and fly were present in fair numbers in the country immediately to the north of the belt. Reference to Howard’s map of the fly distribution in Portuguese East Africa shows that in 1910 the first of the two belts was a mere focus, possibly a disappearing remnant of the greater numbers seen by Hamilton. Hamilton describes the fly as in ‘‘ swarms ” and “‘ extremely troublesome, often biting ourselves and our natives after sundown.” It is, indeed, usual to find references to areas where game is rare couched in similar terms. But it is possible, indeed probable, that under these circumstances, their number and persistence testify rather to the desperate vigilance of flies assembled from a large 10 DR. H. LYNDHURST DUKE. radius than to a similar prevalence throughout the belt. Hamilton notices that the fly were concentrated along the main track in the neighbourhood of camps and halting places, localities in which experience had taught the insects to expect their supply of blood. They are even reduced to so far departing from their usual habits as to bite at night time. Fiske, among other reliable observers, has drawn attention to the following propensities of G. morsitans. ‘‘In the Masindi fly belt, the only morsitans area in which I have seriously studied this fly, I have noticed that the passage of a ‘ safari’ of porters or, better still, of a cyclist, will to a very large extent clear a road of its waiting tsetse, with the result that an observer following on a short time afterwards may find few or no fly until arriving near the edge of the belt. The swarm of morsitans which sweeps along in the wake and round the person of a cyclist passing through a fly belt gives him an entirely false impression as to the true prevalence of the fly.” I labour this point because the fallacy of in- ferring from the numbers of tsetse buzzing about the observer a similar infestation of the belt as a whole is by no means rare. A consideration of all these points reveals an alternative or, at any rate, a modified explanation of the facts noted. (Game was very rare in the part of the belt traversed, and the local fly were probably hard-pressed for mammalian blood. Hence their voracity and persistence, and their concentration at spots where, in addition to their natural liking for the edge of a path, they could best satisfy their hunger. Further, it being the dry season, these fly would tend to concentrate in certain favoured localities (cf. infra). Probably game such as duiker, oribi and warthog existed scattered throughout the belt ; also a few large game remained. The exis- tence of fly in the country immediately to the north of this belt in 1910 may mean that, in consequence of a migration of the game, the tsetse were gradually shifting northward to merge with the Rovuma belt, in a richer game country. | Hamilton himself realises the danger of dogmatism on so important a question as the result of a single trip ; but none the less, he has laid considerable stress on the apparent independence of these fly. His facts show that, in certain localities, morsitans can temporarily adapt itself to the almost total absence of ungulate game. Natives and their domestic animals, monkeys, birds, reptiles, and probably hyaenas, were available to the fly in the present instance. Whether the survival under these gameless conditions 1s permanent, or merely pending an alteration of range, is at present an unsolved question. In the majority of the tsetse areas of British East Africa it appears from the writings of Jackson, Neumann, and others (7) that game occurs only in very small quantities. Neumann (8), describing the country south of the Athi River above its junction with the Tsavo, writes : ‘“‘ Here are great stretches of uninhabited bush country with a perennial river running through it, and hardly any animals, though plenty of birds and of ‘fly,’ Tsetse.” But in his description of this scanty game he says “‘a very few waterbuck at intervals, an occasional lesser kudu or two, a little lot of impala, sometimes at one particular point, and rarely an odd bushbuck, are all that are seen, without counting the diminutive paa, and now and again a rhino, giraffe or hippo spoor.” Here we have a fairly comprehensive menu, however barren the district may seem to the hungry hunter; yet it is just “sufficiently ‘restricted to render the fly THE RELATIONS OF GLOSSINA MORSITANS AND GAME. 11 appreciatively attentive to a human visitor. Such conditions are, I imagine, fairly typical of many of the so-called gameless fly belts. A less keen observer than Neumann might very well have overlooked some of the species mentioned. Competent observers such as Woosnam, Yorke, Bevan, Lloyd, Shircore, etc., are unanimous that they know of no instance where morsitans occurs in the total absence of big game (9). It is possible, however, that the tsetse of a belt are adapted to the special con- ditions of that belt as regards the distribution, species and numbers of the game. In a district where buffalo and other big animals are numerous these will be most easily accessible to the fly, especially in the long grass season. The tsetse will there- fore learn to depend on these bigger animals. In another belt the regular food supply will be the smaller antelopes, baboons, warthog and carnivores. Any cataclysmic alteration in the game distribution of a belt, especially if it occurred at the critical time when the grass is up, might well be seriously detrimental to the tsetse, whereas a more gradual alteration would give time for a varying degree of adjustment to the new conditions. G. palpalis on the islands of Lake Victoria often shows a deliberate preference for reptilian over mammalian blood. In other regions, such as the neighbourhood of the crater lakes near Lake Albert Edward, crocodiles are absent and Varanus rare,and the palpalis are plainly dependent on the numerous hippo and other game animals. Evidence of a similar independence in G. morsitans is wanting. The smaller mammals in morsitans country, though doubtless occasionally fed upon by the fly, will, by their habits, be difficult to find, especially when the grass is up. I have found Varanus in the middle of the Masindi morsitans belt, though there were no fly around at the time. The extraordinary partiality of G. palpalis for this saurian suggests the possibility of the more fastidious morsitans feeding upon it on occasion, and this fly has been found to contain non-mammalian blood in nature. In the present state of our knowledge, however, these creatures must be looked upon as accessory sources of food, and it is the general opinion of experts that mammalan blood is necessary at any rate for the propagation of the fly (Lloyd). My own experience of morsitans, apart from the Masindi belt which will be dealt with later on, is founded on visits to the following belts in “ German ” Hast Africa under active service conditions :—(1) The great fly belt on the Tabora and Mwanza road ; (2) the Buzinza area to the west of Mwanza ; (3) the Namirembe-Biharamulo belt, possibly continuous with (2), around the south-west corner of Victoria Nyanza ; (4) between Ikoma and the Mara River and KE. towards Olgoss; (5) the area along the Kagera River, near the south frontier of Uganda. In all these belts game is common, excepting the first, in which I spent about 14 days at Shinyanga, well within the fly zone, at a time when game, excepting dikdik, was rare, having recently migrated in search of water. In none of the above belts was buffalo found within many miles of my camps. The fauna observed in each case respectively comprised :—(1) Zebra, Thomson’s gazelle, giraffe, duiker, dikdik (plentiful), hyaena, roan, impala, carnivores; (2) sable, topi, reedbuck, kudu (few), hyaena, duiker, roan, impala, zebra, dikdik, giraffe, carnivores ; (3) the same as in (2), plus rhino and dikdik (very common); (4) Thomson’s and Grant’s gazelles, 12 DR. H. LYNDHURST DUKE. oribi, rhino, hartebeest, topi, roan, wildebeest, duiker, zebra, impala, dikdik, carnivores ; (5) roan, impala, oribi, duiker,eland, zebra, topi, impala, wild dog and other carnivores. In these belts fly persisted throughout the dry ‘season and the burning of the grass. They appeared much addicted to the neighbourhood of paths, and, in places where motor transport was employed, illustrated to a remarkable degree the “following ” propensity of morsitans. On the whole, the fly were not particularly troublesome, except in connection with motor vehicles, around which they collected literally in swarms. In the North-East Transvaal Stevenson-Hamilton (3) states that the fly com- pletely disappeared immediately after or during the rinderpest of 1896-1897, and has never come back. Apparently it disappeared also from the adjoining Portu- guese territory. As regards the effects of the disease on the game he writes, “after the rinderpest it was found that the last of the eland in the N. EK. Transvaal had disappeared, and that the buffalo were reduced to a herd of about 20, which remained in the densest part of the Sabi Bush in the heart of the former fly country. A small number of kudu and bushbuck survived in the same locality. Impala, wildebeest, and other species, native to the district, appeared much as formerly.” In this district the rinderpest arrived among the game at the season of the long grass. Gibbons (10) writes of North Barotseland: “I may mention that between 1896 and 1897 the tsetse has disappeared in many districts where it was previously rampant, and is so decimated in others as to be scarcely perceptible.” Again Jack (11) speaking of the Lomagundi district of Southern Rhodesia says: “It would seem that the fly in this part survived the rinderpest in very small numbers in scattered localities, failed to increase to any extent, but persisted until recent years, although on the verge of extermination.”” Hamilton contends that “‘ in the absence of any other destructive cause, the fact that even a few bufialo, kudu, and bushbuck survived, supposing these animals to provide the favourite food of the morsitans, must surely have induced a partial survival of the latter, a nucleus which in the course of years would have increased with the increasing herds.”’ From a consideration of the above evidence the following points emerge :— (1). That there is no known instance of the prolonged existence of G. morsitans in an area where game is entirely and permanently absent. (2). That the fly is generally found associated with game in considerable numbers and variety. It may, however, occur in localities where wild ungulates are apparently scarce. Under these circumstances the tsetse are probably pressed for food and their appreciation of the human visitor is apt to give their exasperated victim an exaggerated impression of the degree of infestation of the belt as a whole. This impression will be enhanced by the “ following ” propensities of the fly. (3). That during the rinderpest epidemic of the nineties the tsetse of South Africa were affected differently in different localities. The fly were either (a) apparently unaffected ; or (6) markedly reduced innumbers, temporarily or permanently; or (c) reduced to the point of ultimate extinction. Generally, however, the passage of the rinderpest throughout the fly country was marked by a great diminution in the numbers of the tsetse. (4). That the rinderpest affected some species of game very severely and left others almost untouched, with the result that a large number of animals escaped destruction. THE RELATIONS OF GLOSSINA MORSITANS AND GAME. 13 (5) That the theory regarding buffalo as essential to the subsistence of the fly is untenable. It is possible that the fly of an area where buffalo are very numerous may resort to this species as its chief food supply. This would, however, apply equally to whichever of the larger ungulates predominated, and probably depends on the ease with which these big animals are accessible to the fly, especially at the season when the grass is high. The arrival of the rinderpest at this season would enhance its effect on the fly through its food supply. (6) That much of our information regarding the relations of game and _ tsetse is based on superficial observations by untrained observers on isolated occasions. In such circumstances the persistence of small isolated foci may easily be overlooked. Description of the Masindi Fly Country. The Masindi fly area comprises an extent of bush country in which tsetse has always been fairly numerous and evenly distributed in the wet season. In the dry season, after the annual grass fires, a diminution in number has been noticed both by Viske and by the native inhabitants of the belt. During this dry season Viske says that the fly appear to be concentrate in certain localities, which show special characteristics and resemble Shircore’s ‘‘ primary centres” (12). Game is fairly plentiful throughout the fly belt. The main motor road from Masindi Port to Butiaba traverses the belt, which towards the River Kafu is limited by open plains swarming with kob and reedbuck. A few small villages are scattered about here and there. The fly scrub itself consists of scattered small trees and bushes rising above the coarse grass, which grows in tufts and, in July and August, reaches a height of 6 or 7 feet. Long low ridges are separated by valleys of varying extent, sometimes narrow and thickly covered with bush, sometimes wider with thmner bush or open stretches of shorter grass. In the rainy season these valleys hold swamps. In the dry season there are two types of waterhole found in the belt :— (1) Muddy depressions or holes containing filthy water and pitted with game tracks. These holes are occasionally found on the ridges, but more generally lie along the bottoms of the valleys. This type of waterhole dries up slowly, and the gaping eracks in the sun-baked bottom will absorb many showers without leaving any standing water on the surface. (2) Punched out holes or ponds often of clear water, situated on the ridges. Of these holes Fiske says “ They fill and overflow during the rains and usually dry quickly following their cessation.” Also, and this is probably important, they are replenished by any chance shower of rain. Such a waterhole is a feature of Fiske’s “foci,” i which the fly concentrate during the dry season. Other characters of these foci are freedom from bush, and shorter finer grass; they are surrounded by bush which is often unusually thick. Game is also always present. Fiske assumes, as a result of his work in the belt, that morsitans finds particularly favourable conditions for breeding within these foci, at all events during the least favourable season of the year, and that it disperses from them freely, especially during the wet season. No fly are found in the open plains about the Kafu except such as may follow passengers along the roads. As regards pupae, I have found these in fair numbers in the crevices of ant-hills and under fallen tree-trunks. Fiske is of opinion that 14 DR. H. LYNDHURST DUKE. throughout the belt, when the grass is up, large numbers of pupae are deposited indiscriminately in the grass-tufts, and that the crevices and tree-trunks are more especially a resort for the dry, short-grass season. He finds that the fly breeds less freely during the dry season; an observation also made by Lloyd in Northern Rhodesia. During my stay in the belt in 1914, in May and June, fly were fairly numerous throughout the area, especially along the various roads traversing the belt. It was practically impossible to bring a dog along the Nakasongola-Masindi road where this traverses the fly scrub without 1t becoming mfected with trypanosomes. Cycling and even marching along this and the other roads was unpleasant to a degree on account of morsitans ; whilst trekking through the bush in search of game the fly were in evidence, and often a nuisance. Observations in the Masindi Belt during April and May, 1918. On my arrival in the fly area in April 1918, I was at once struck by the marked decrease in the numbers of the fly met with. I visited all the localities seen in 1914, including some of Fiske’s foci, and am confident, from a comparison with the number of flies caught in May 1914, that there has been a very marked reduction. I had with me fly boys who had previously worked with me in this belt, and they were unanimous that fly were much diminished everywhere and often unobtainable where they were formerly fairly abundant. On only one occasion did I find fly at all numerous, and this was in the immediate wake of a large herd of elephant at a waterhole. The fly had disappeared from this spot the next morning. Cycling along the main motor road was no longer a penance, and on trekking about from point to point in what used to be the most thickly infested parts of the belt the fly were few and far between. I found no evidence of a corresponding local increase in any part of the belt north of the Kafu River. The grass was growing rapidly, though, perhaps, not so long as it was in places during my visitin May 1914. There is reliable evidence that fly were still worrying cyclists along the motor road throughout the earlier part of 1917. In May of that year two independent and reliable observers report that the fly were numerous. In November 1917 Mr. Bain, the District Engineer working in the Masindi district, tells me that he visited the fly belt along the main road and was struck by the apparent absence of tsetse. A considerable number of buffalo were killed by the rinderpest in this fly area. Bushbuck, warthog and duiker also suffered severely, waterbuck and hartebeest being little affected. The kob and reedbuck on the Kafu plains also died in considerable numbers. A certain number of bufialo survived the epidemic. I next visited the Bugungu fly country, through which the disease had spread irom Chopi. These parts I had visited in August and September 1914. Owing to the famine of 1918 it was impossible to “ safari’ through this big area. I therefore went by sailing boat along the north-east shore of Albert and up the Victoria Nile to Fajao, camping at intervals along the bank and striking inland to visit the scenes of my former tour. As a direct result of the epidemic, the bufialo moved south along the Bugungu plain to the point where the escarpment joins the lake near the outfall of the Waki River. They left a train of dead along their course, a striking feature of these bodies being that most of the skins remained intact and the bones were not generally pulled about by carnivores. Bodies of kob and warthog were ~ also found. THE RELATIONS OF GLOSSINA MORSITANS AND GAME. 15 Throughout the parts visited, G. palpalis was still found where it occurred in 1914, and in large numbers. On several occasions I found it abundant, as testified by actual catching in the bush surrounding the dried-up remains of a buffalo, the sick animals almost always making for water. G. morsitans was found in various parts of the area, but here the reduction in numbers was much less marked than in the Masindi Port belt. In one place, however, I found (and caught) considerable numbers, and here the fresh tracks of a herd of some 20-30 buffalo were observed. In the Bugungu region numbers of waterbuck have survived, and elephant are common and frequent visitors. Bufialo have survived in small numbers out of the numerous herds which existed throughout the district before the rinderpest. Kob and reedbuck also occur ; bushbuck and pig used to be common, but I saw none alive during my visit. In order to ascertain whether the fly had migrated southwards across the Kafu River with the buffalo, I sent two reliable fly boys for a three weeks’ trip through the Buruli district. In 1914 VWiske and I found fly exceedingly rare in this area, but its existence was known. The boys obtained G. morsitans from several localities, sparsely distributed. At one spot, some 15 miles south of the Kafu, they described the fly as numerous. A single specimen of G. pallidipes was taken. It appears, therefore, that fly are more numerous in Northern Buruli than they were in 1914. There is no known focus of morsitans in the south from which these fly could have spread. Whether this increase is a direct result of the rinderpest, it is difficult to say. The buffalo are in the habit of visiting and crossing the Kafu in normal times, and it is almost certain that, in their panic of the rinderpest, considerable numbers must have trekked south into Buganda. The disease has at the time of writing reached within 40 miles of Kampala, and the main agent in this spread is beyond doubt the game, especially buffalo. To sum up, there has been a marked diminution in the numbers of G. morsitans in the Masindi fly belt, comciding with the passage of the rinderpest through the game of the district. The fly is not exterminated, but the reduction in numbers is much more marked than that which ordinarily results from the annual hot season. In Bugungu the reduction, if such has occurred, is less apparent. There is no apparent reduction in the numbers of G. palpalis in either district. In both areas considerable quantities of waterbuck have lived through the epidemic. Bufialo have also survived in some numbers, though, on the Bugungu plains especially, the number of dried carcases testifies to a heavy death roll. Experimental Investigation of Effect of Rinderpest Blood on Glossina. It was found impossible to obtain G. morsitans in sufficient numbers to carry out feeding experiments in the actual belt. 1 was therefore obliged to utilise G. palpalis, and, to avoid all danger of introducing the rinderpest into the populous districts around Entebbe and Kampala, it was decided to select a suitable island on Lake Victoria for the scene of the experiments. Nsazi Island some 15 miles south-east of Hntebbe was chosen, and Mr. Hutchins supplied me with rinderpest blood from Singo, where immunity inoculations were in progress. He most kindly performed the inoculations and twice came out to the island to assist me with his advice. regarding the progress and diagnosis of the disease in the calves. The temperature charts will be found below. Calf B. R. was seen by Mr. Hutchins, who confirmed DR. H. LYNDHURST DUKE. Ui) Daily G, 2 3] ' ' 27'28|29'30 16.718. 24'25|26 22123 inoculated 1; OO OU gd car (2 =| + N 24610 burinp s31g eal] i} ! Es I { ai l22 ' ' 20 JULY (op) - = — G ———-— N] pa ia a Fil Sa ala gael 0 | —— © E Reece o : — rates ee = pea at = Re 2 a se - ° : aa : = 17 JULY. dx } cies pee Pf apace Se eared aMbh ake sew 1h bahay vr ow ie Wii siaiienecneorsat Ame adi se nor : Ty bg ee ai “ nee Piety peecnuntnet shi we L S i ‘ a ¢ < i + hee ete go ng PORES As ed eto iF doce tal Bela Reten ar ’ i Se ‘ aren aes 8 BULL. —— Ee < o. 2 = O > 4 ri Z| Wi eS ~ rm « 5 G SO : < be) S G e * by ® << b SS) i omimencement THE RELATIONS OF GLOSSINA MORSITANS AND GAME. 17 the diagnosis. The other two rinderpest animals, calves C. R. and D. R., showed typical symptoms, both before and after death. The following animals were used in these experiments :— Calf A—not immunised against and not infected with rinderpest. Kept isolated from rinderpest animals on the island. Used for control feeding. Calf B. R—inoculated with citrated rinderpest blood on 6.vui.1918, 34 hours after removal from sick animal. Died with typical symptoms of rinderpest on 19.vu.1918. Calf C. R—inoculated on 6.vi.1918 with 20 c.c. of pooled citrated blood from three rinderpest animals, 30 hours after removal from body; this apparently failed to infect. Stalled in closest contact with Calf B. R. and died on 24,vu.1918 with symptoms of rinderpest. Post mortem showed mouth ulcers and acute inflammatory exudation into lower bowel. Calf D. R—inoculated on 16.vu.1918 with pooled citrated blood from three cases of rinderpest, 24 hours after removal from body. Killed on 1.vin.1918 after showing typical rinderpest symptoms for some days, confirmed by post mortem examination. | Sheep E, F, G, H, and K—healthy animals kept for control feedings, and stalled at a distance from the rinderpest animals. None of these sheep showed any signs of illness. Calf A—developed symptoms strongly suggestive of “ amakebe,” 1.e., great en- largement of lymphatic glands, transient diarrhoea, and anaemia. The blood proved negative to Piroplasma a few days before it was killed. The animal also showed extensive skin lesions, which Mr. V. F. Richardson, to whom I described the case, said were very suggestive of a blastomycotic infection. This animal was killed on my returning to the mainland, together with all the other animals then alive. There was no sign of any inflammation of the gut. The animal may, therefore, be considered as clean as regards rinderpest, and was rapidly recovering condition when it was killed. On 17th and 22nd July the dead flies in the experiments were entered on the day © following in each case, except in Experiment 22. It will be seen from this table that rinderpest blood exercises no apparent effect on the fly. It was impossible to pronounce decisively regarding the effect on larva production. 29 A certain number of larvae were produced during the course of the experiments, of which a number were prematurely aborted. No difference in this respect was noted between flies fed on rinderpest blood.and those fed on clean animals. Meteorological Conditions in the Masindi Belt in 1918 and Previous Years. Masindi is 29 miles along the main motor road to Butiabwa from Masindi Port. The fly belt extends up to about 12 miles from Masindi Port. The following table shows the monthly rainfall for the years 1914-17, and January—May 1918. In regard to these figures, which were taken at Masindi Station, it must be recollected that the actual fly scrub receives less rain per annum than Masindi, which is surrounded by high hills. The right-hand figures in each annual (C572) B BULL. ENT. RES. VOL.X. PART |. [Ot H.L. DUKE} 5 3 Number of Flies x DAILY TI AND DEATHS OF MALES we = in the Experiment pee N = wy REATMENT AN. HS OF FLIES, MALES AND FE: . REMARKS. di x § £8 ale ie =a] ill ipa > ie alae] ca] = = & a5 wos > | 3 comms | 8 &2| 8 5 © S$ See = flies Bg & & & 9 |lo| 7 \/2 | 13 | 74115 | 16 \ 717 | re | 19 @5|26|27|26|29|30)a/ |/ |}2 13 |*#|5)6)7 |e | 9 |10 | 17 | 12 | 12 | 14 | 15'| 16 | 17 | 78 | 19 |20 | 27 | 22 | 28 | 24 | 25 | 26 | 27 = iS NaS 39 = Kimmi % ape PEN ew Ps) I) eh NIDA a on A lA “ll Island 8-1 |8-1|0-0]0:0|2-0}2-0}1-0/2:0]0-0]1-0]0-oj0- : : -110-010-0 x AIlAITAITA[ATASA;)ATATATS Sains = es “ 0-0|0-0|0-0|0-0|0-0|0-0|/0-0] of 1-0lo-0 7111-0 Control to Expt, 8. by. A|A[AIAIAIATATA [A TATA AIA eee a es | 8-0 | 1-0}2-1 |5-1|4-0]2-1 |3:0] 2-0} O-1}1 |0:0 | x AJlAI[ATA AIAITATATATS Fr ue 1-0 |0-0]0-0]0-0]1-010-0|0-0]0:0| __|0-1)0-0]0:0|0-0 -0|0:0]0-0 ci R July 16 BSBA RACERS 78S AS BR Es En Sea Si Pee (eS Sait | Fed on calf for the last four days 3-0] 3-0|0-0/0-0|0-0/0:0|0:0/0:0 0 | 0-1 |0-0] 0-0 :0| 1-0 |0:0 of its disense, R July [5 A A A a A [A BR/BR /BR|BRIBR/| S DOR|S DR Fed on two rinderpest a Y 17-2| 3:1 |3:2}0-0|0-0|0:0 |0:0|0:0 |0-0] 0-0}0-0|0-0|0-0|0-0 0-0 | animals. R Meee A|A |A /BRIBRIBR|IBR/IBRIBR|S |[DR|DR]| Ss 2 (DRI ry uly 6: 0-0 0-0 ” | A Ss Ss Ss 1 lpoeee 5 A Has rindergest blood any weakening _|z: 0:0 2-1 |I- : effect on fly? of Ex.2 R A A |S Gicisfelelcf{ec s [kK |[s me fs fis fs] Fed on calP from day after First u, re 7-0|4-2|1- : 0-0}0-0 0-0/0-0/0-0]0-1]0-0|1-0]0-0 o|0-0]4-1 | 1-0} 5-1 |3-1|0-1 |4-0]2-0|3-0 |2-0|1-0 |2-0 CHEESE! R ATA s Als G/Gi{s|{c/Glclca S|His [RH |[H|sS [Hs [a [Ss [H Is pee | ” 4-0/8:1| 1-1 |0-0]2:1 | 2-1|1-0|0-0/0-0]0-0/0-0|0-0]0-0 0-1 |0-0|0:0|4-0|4-0| 1-0] 0-1 -0|1-0}3-1|1-0 | 4-1 |3-2|0-0)0-0]1-0|0:0|1:0 |3:0)4-0 IL ii R A |A/A/A /BR/BR/BR/BB/SR/(BRIBR] S [DR] S DRIDR|S |/DR/DR|DR|DR H|TH|S |H [H|S [H/|S/TH {STH |S Fad on tno Hinder aasE aE a a 1-2 |9-4|1-1 |1-0]0-0/0-:0]2-1]0-0 2:0|0-0 | 0-0] 1-0 0:0 |0-0]0:1 |2-0}1-0 |2-:0}0-0 10/71 |6:1 |o-0]2-1 [1-1 | 12 [oo] 04 | F1 |3-0)2-0|4-0 i A |A |e |CRICRICRICRICRICRIBR| S IDR|S DR|DR|S |DR|DR|DRIDR TK STS te els fers fe tS ik 1s Nsazi ES nS Island R July 18-19 Z | Fed on three rinderaest animals Sem 2:010:0|/2:0/0-0|20|3-0/0:1 2:2|0-0/0-0]1-0 0-1 !0-0/2-0/5-0]1-0|0-0]0-0 -0/0:0]0-0] 1-0 | 6-1 |4-0/4-0|0-0/2-0]0-2|0-0]1-0 1-0 | 1-0 | 1: +0) 0- :0|3- : : ao fETE/ElE/eE/E;SETE!E lpRis lets G | .Gujisn | Gh lnc iicmG KiOofmiys fms ficlS lw lS tx 1s Ip ; 5 2zg|a } Kimmi July 19 Single feed on rinderpest. srimel | Istand | mle i2rol3-0|3-0 [5-0 1-0 |1 2139. 1-0 |0-1 |0-0|0:0}0-0 9:0 |0:0|0-0|0:0|0-0]0:0] 0:0 0:0} 0-0 |4-0/0:0/0-0/0-0|5-0| 1:1) 11 |3:0|0-0|1-0 {1-1 |: : : : on last day of disease Baus | R ies | E E | £ [DRIDR|DR]|S DR/IDR|S /DOR|S |F | F KolSele Kol SmpKal Sulake) Sal Kul Salat iS s i al ees waahs = antes : | __|12:3]3:0} 0-0} 9-0 1:0 |0-0] 0-0] 1-0 |0-0/0-1 0-0} 7-0 |2+1|1-0]2-0|0:0|o-1 12-0 |6:2|1+1 [2:2 |0-0| 1-1 | 3:1 2-0) 0-0]0-0/0-0}1 0 j2'3 on [on el | 4 see Eaateeaees es zie pS ite) oie S/H |s |) [s) fs (| js |) (ss) ffs i | | al 2 ‘ [2: _| 1:0 | 0-0] 0-0} 0-0 0-0/0-:0]0-0|1-0 |o-0 0-0|0-0 0:0|2:0|4:0|0-0|1-0 |2:0|0-0| 0-0} 1-0] 0-1 |0-0] 1-0 |0-0/0-0 | 1-0 i: 19 | 2 | Bulago x Ee ETEeVTE;LETeE|S Py Se eee S|H|s | 1S | 1s |/ls [His [Ss [5 |S /s Island 6-0| 6:0 |0:0]0-0] 1-0 |0-0/2-0|0-0/0:0/0:0|3:0/0-0} 0-0 |0-:0|0-0|0-0|0-0|0-0/0:0 2:0|/0-0|1-0 0-0 |0-0 |2-0|3-0 [2-1 |1-0|4-0|1-1 |0-0]0-0] 1-0 |1-0|0-0/1:0) [ R uly 18 or23 E/E;E;/E;SE,JE Creu or Ey) SojDR DRIDR[S |DR|S |G |G SiepS belo iis fst) oS loa 1S le i Siiw)sS lace ls Fed on two rinderpost animals. : 1 R NS Sasen 9:0}2:0|4-0|1-0|3-0 |0.0|0:0/0.0) 0.0|0-0/0-0/0-0] 0-0 |0-0|0-0|0-0|0-0/0-:0/0-0 0} 1-0/a-0)0-0 | 1-1 |3-1|1-2 |o-0|5-2| 1-1 |o-1|0-0]0-0|3-0|0-0|0:0 Nees} r R ali ion Eye Pe jer CRICR/CR|CR|CR/CR CRICR]|S E |e Sas Ere |e Sees 1S lis PS packs K |S K/[s |K ss ins 0-0|1-0|2-0|0-0|0-0|0-0| 0:0/0:0| 0:0|0-0]0-0|0-0|0-0 Q-0}0-0]0-0]|1-0 |0-0]0-0]0-0 0-00-01: |0-0] 0-1 |o-0/0-1 |o-0|0-0] 0-0 0-0|0:0|0-0/0-0 |o-0|1-0 _|l R E|E |e IcR|CR|CRICR|CR|CRICRICR|CR| 5S El[Fis/El/eE/fE/e Riso (eS (S'S |s 1s jes alt | sf | 2 1-0] 8-1 |0-1 |0-0/0-0|0-0| 0-0/0: 0)0-0j0-0]0:0 |0-0 5-0|2-2|2-0|4-0] 2:1 |0-0|0-0 0|0-0|0-0|0-0| 1-0 |6-0|1-0 |2-0|9:4|3-2|/0-2| 0-1 ale alt | I I) a R efe|/sfelele le Hi{s/H |s /H/s |sS|s ls 7 Island He o-o|2 1 |o-0 2-0}0-0/0-0/0-0 *2|5-4/0-O| 1-3 |3-0 |3:0|}0:0]0-0/1-0|1-0|1'0 at Bia _| ; R EES (ele Te |e jH |S) |s |S |H 1s 1s [Ss [ | aL ny i : e {O20} (020111020: /]G70)(Ox0)}t020)/0x01/020 0-0 | 0-0] 1:0 |O-0|0:0|0:0|0-0|0-0|1-1 | 0:0 ll a + ale = Kimmi awe | ae Fie rye ee | eye SP (Een S| EO SN ton (RS [US PLE ESO || |S eee ESI ES) Fed once on rinderpest wnimal Island ay, | - | 0-0|4-0|13-:2| 6:0 1-0 | 1-0] 0-0 : o|2-2\2-1|4-0|3-4 |2-2) 1-1 ene )| 0-0 |2-0| 0:0/ 0-1 |3-1 | 0-0] 0:0/2-0/2-1 | 3-4 |3-0]1-3 | 9-0] 1-1 |0-0/0-0]/ 0-0 |2-0|0-0|1-0 |3-0|2 se | I A A[A[S|[A|[ATATA|[S [KITK/[S|TK]¢ [ki s]{xK ris }mis feisi=s fois ise Is im ry | | | I J 0-0 | 2-0] 0-0] 0+1 | 1-1 | 0-0!.0:0}0-0] 0-0] 0:0] 0:0] 1-0 | 2-0 | 5:0] 1:0 |0-0| 2+) | 2-0] 0-0] 0-213-0] 0-1 |2:0| 1:0] 70| 42) Ihe] 6-31 75 [21 Bred cut betwee R July 23 (eal fa S [pris |orR|s |pRioR| HIS |S |HIs |H 1S | )S |H|SlH|S/fHlslHIisis|is]s|/sis]s ned cut Between u , SE TE247, 2 0:0|0:0]0:0|0-0|0:0|0:0|0:0]0:0]0:0 |0-0!0:0/0:0|0:0|0-0| 1-0 | 0-1 |o-4 | 0-0|0-0]4-2| 2-0] 0-0] 0-3 | 0-0] 0-0 0-1 | 1:1 | 0-0] 0-2 | 83 8-6] 2:1 1 9°S | 1S { = ——|_82' g LOZIEOSO)) Kimmi xX ‘a ETS)TE Ve | ele lm lS isis fx Ss ll Ssa) 1S fxs th papier TS (Ss (TiS TST “|| I; | Isd. 23-7 : _I| -0|1:0 | 1:0 | 1-0 | 1:0 | 0-0] 0-0] 1-0 | 5-2| 0-0] 0-0] 0-0 |0-0] 2:0] 1-0 | 7-0 | 6:0] 2:1 |0-.0|2:0/3-0| 0:0] 2-0] 3:0| 4-0] 6-0] 4:0] 6:0) 9:1 | $0) 10 SsOi test a = x ie e/elsie@ irFlse |eolixnxls |s lis |nNis iH | SlH/|S/H|/S [HIS [H[ sis , 3 1:0 10-0] 0:0} 9-0] 9:0] 0-0/0: |0:0]0:0] 0:0] 0:0] 1:0 | 0-0} 4:0] 6:1 4:0} 9-1 | 2-0] 1-1 2-0|2:0| 2:0| 3:0} 1-0] 0-1 1-0 S Ss Ss S = in DRIDRIDRIDR|DRIDRIORIS |H|H|S|H|]S |H|/S |H1S /"1S/4/S|H lal Sys isisys Brad out between R July 26 2:0| 4-0| 3-7| -0| 5-1 | 2-1 |o-4 ca toleh J 4 | 0:0 0:0] 0-0]0-0 | 0:0] 0:0|.0:0|0:0|0:0 |0-0] 0.0] 0-0| 0-0/ 0-1 0-1 | 0:0] 0:1 |O-0/ 0-0 | 0-0] 0:1] 0-0 | 0-1} 0-0] 0-0 }2-1 |e + H ST Bred oul be July 30 if DRIDR|ORIS |K |S /S]|K]/siwKiS [Si[K{/STS {sys s {sists iS Tris b red out between sed. 77 + 307 R ees ai} 0:0] 0-0] 0:0] 0:0] 0:0/0:0]0:0] 0:0| 0:0 | 0-0] 0:1 | 0-0] 1-1 |0-0 0.0] 08 | a1 | 41] 2-1] 48 }41{13js0t tt =k iE sie * | Gap | Sei |ns | S| Kilesmlsels (si ]'s) ls: Siisais isis | ie red out on “ele 0: 00:0 |0:0] 0:0] 0-0 | 0-:0/ 0:0] 0-0/0:0}01 10-0) 1-7 |2:0 | 3:0} 4-6 | O12}2-2 | 1- hk 29-7 |. Ke I [aa } | LL io) OOO alt | Lo 5 | seal 1 Malby &Sons Lith 9772.P4.140, 1055 O19 In the “Daily Treatment and Deaths” columns the ocourrénce of capital letters A-G, show that the flies fed that day on the calf or sheep corresponding to that letter (see list. above). ‘fhe addition of R shows that the animal had rinderpést. =fed throughout on clean control animals. The first figure gives the male, the second the female deaths. S—starved that day. j a 18 DR. H. LYNDHURST DUKE. column of the table represent the rainfall taken some 15 miles north-west of Masindi in a still more hilly and much fcrested district with a proportionally heavier rainfall. In consequence of the exceptional drought in November 1917 the grass fires of the 1917 hot season began early in December, somewhat sooner than usual. A severe famine prevailed during the early months of 1918 throughout the whole Province of Bunyoro, | Masind: Area Rainfall. | 1914. | 1915. | 1916. | 1917. | 1918. Pn ANS: oe da ie B. iW Ber ans AC B. AY B January 2-31} 1:00} 1:53} 0°84} 0:00] 0:00 4:82] 3°79 | 0:00 | 0-00 February 1:45 | 2°97 | 2:99) 1:14] 3:96] 4:71] 559! 7281 0-00 | 0-42 March | 4°17 | (4:61-] 697 | 3°98 5:12 | 5:01) 71-23 | 1-71 |) 9-6) ae April | 380] 380] 4371 603| 748] 7:16 | 4:961 6:87 | 10-06 May | 675 | 825] 645 | 810] 993] 576) 865] 7:09 | June 4°36 | 7:09 3°88 | 4°54 | 2:44] 4:56) 7:04 | 4:45 | July | 3-46 | 6°70 | 2:57] 2°00 | 0-56} 0-88! 1:49] 1-33 August 3-68 | 3:82 | 2°05 | 2:26! 520! 552] 3:38 | 5:00 September | 7°40] 6°74] G14] 675] 5:94 | 10°79 | 11°85 | 6:80 October 6:00 | 5:10] 689} 9:39 | 8291 10°58} 7:51 | 7°75 | November | 9-07 9°82] 3°58} 600! 3:80} 4:50/| 0-80] 1:46 | December | 1:33 | 0°78 | 292| 250.) 2:19] 200| 4:17] 1:99 | | i The column A shows the rainfall at Masindi, and B the rainfall at a spot 15 miles _ north-west of Masindi. In normal years the grass fires begin at the end of December and extend through January into February. The big rains begin in March. The grass is burnt some- what irregularly, but the result of the hot season and its fires is that most of the grass in the fly belt is burnt to the ground and the smaller shrubs severely scorched. Growing trees are hardly affected ; fallen trunks are more or less destroyed, often smouldering for days. The grass grows rapidly after the onset of the rains, and in July and August has reached a height of 6-8 feet in many parts. Conclusions. Of the game species available to the fly in the Masindi belt, warthog, bushbuck, duiker and carnivores are undoubtedly the most important, considering the belt as a whole. EKlephant and bufialo come and go; hartebeest and waterbuck are local in their range, especially the latter; the kob and reedbuck of the open plains sometimes encroach upon the edge of the fly scrub, but are probably seldom bitten ; and there are, I believe, no dikdik in this area. In Bugungu the conditions are similar, except that elephant and waterbuck are much more numerous, and hippo from the Nile and Lake Albert roam far inland. Fiske (13), discussing the food supply of the Masindi belt fly, applies the principle, established almost beyond doubt for G. palpalis, that the percentage of females among the active flies caught is a direct index to the abundance of food animals. He writes that ‘ The flies in this belt are well fed during April, May, and a portion of June, but during the latter part of June and in July begin to find difficulty mn obtaining sufficient food. The percentage of females caught rises, and coincidently THE RELATIONS OF GLOSSINA MORSITANS AND GAME. 19 the flies become more persistent in their desire to feed upon man. This change in their behaviour is also coincident with the coming to maturity of the grass, which in early May would not suffice to hide even a duiker, but by the middle of June would completely conceal bushbuck and pig in most localities, and larger animals in some.” : Now the rinderpest commenced in Chopi a few miles to the north of the scene of Fiske’s and my work in 1914. The disease spread rapidly through the buffalo, which, as usual, migrated, some going towards the Bugungu escarpment, some making south for the River Kafu and Buganda. Right through the long-grass season the game was dying in numbers throughout the fly area. The elephant, which descend towards the River Kafu in the rains, were still in the thick elephant-grass country of the Chopi reserve. Bushbuck and warthog were dying in numbers. The food supply of the fly was thus severely reduced just at the time when food is, under ordinary conditions, most difficult to find. The drought, beginning in November, led to an early drying of the grass and commencement of the fires. Similarly the intensity and prolongation of this drought led to an unusually complete burning of the bush. The fly were thus doubly handicapped. Firstly, their food supply was diminished at the most critical season of the year; and secondly, the adverse conditions of deprivation of shelter and destruction of breeding grounds were inten- sified by the prolonged drought. I consider that these phenomena afford a reasonable and probable explanation of the remarkable diminution in the numbers of tsetse in the Masindi fly belt. The Bugungu country, being uninhabited, is less subject to burning in the dry season, as these fires are started and kept up by natives. It remains now to be seen whether the principles elucidated in connection with the Masindi morsitans belt are applicable to the problem of the North-East Transvaal. In this latter district the rmderpest arrived about June in 1896, and the game were still dying fast in October, the hot season. The grass reaches its full growth in June and in some years the fires may not commence until September. I can find no data regarding the particular climatic conditions obtaining in 1896-7 in this part of the Transvaal. It is therefore impossible to say whether or not the dry season that year was unusually prolonged. One of the two factors already discussed in dealing with the Masindi fly is, however, common to both cases, i.e., the arrival of the disease in the district at the season of the year when the grass is up and game most difficult to find. Regarding the complete and permanent disappearance of the tsetse from a large strip of country, described by Hamilton as coincident with the passage of the rinder- pest, it would appear from his paper that he arrived in the district in 1902, some five. years after the epidemic. From the date of his arrival he is satisfied that no recurrence of the tsetse in their former haunts has taken place. That the fly disappeared completely in the season [ think improbable. The truth probably is that, in consequence of the combination of adverse conditions afforded by the arrival of the disease in the cold weather, together with a prolonged dry season with consequent more complete burning of the grass, the fly were very markedly reduced between June 1896 and April 1897. Small isolated foci probably (572) B2 20 DR. H. LYNDHURST DUKE. persisted which, for reasons undetermined at present, instead of increasing, diminished to the point of absolute disappearance before Hamilton’s arrival in 1902. This explanation is compatible also with the apparent irregularity in the effects © of the disease on fly distribution in different areas. To exert its maximum effect the epidemic should arrive during the season when the grass is long. In its passage through the country it may reach certain fly belts when the grass is yet short and the game easy to see. In such case the effect on the fly might be less severely felt. The tsetse would be better able to follow the migratory movements of the susceptible game, and to locate the whereabouts of such relatively immune species as waterbuck, impala, and hartebeest, whose distribution is local. The practical outcome of these observations is that morsitans should be attacked during the dry season of the year, when natural conditions are most unfavourable to it. The two main indications are :—(1) Destruction of the primary centres or foci referred to by Shircore and Fiske by filling in or drainage of the water supply ; destruction of game in the vicinity, and careful burning of the grass at the most suitable time. (2) Careful control of bush burning so as to obtain the maximum effect of the fire ; indiscriminate and patchy burning by natives to be stopped, and the fires started and maintained in a systematic manner so as to ensure universal burning of the grass. An observation made by Dr. Sander (14) in German Kast Africa in a pamphlet on tsetses in 1905 is of interest in this connection. He says that both Europeans and natives attribute the increase and spread of tsetse in the area between Tanga and Kilimanjaro to the cessation of annual grass fires. The game in this area had diminished very greatly. bj References. 1. Bull. Ent. Res., 1, p. 174. 2, Austen, E. E. Monograph of the Tsetse Flies, p. 290. 3. Bull. Ent. Res., ui, p. 113 et seq. 4. Minutes of Evidence of Departmental Committee on Sleep Sickness; Question 321, | 5. Bull. Ent. Res., 1, p. 40. 6. Bull. Ent. Res., u, p. 44. 7. Austen, EK. E. Monograph of the Tsetse Flies. 8. Neumann, O. ‘‘ Klephant Hunting in East Equatorial Africa,” pp. 141-142. 9. Minutes of Evidence of Departmental Committee on Sleep Sickness. 10. Gibbons, Major A. St. H. ‘Africa from South to North, through Marotse Land.” 11. Jack, R. W. Bull. Ent. Res., v, pp. 97-110. 12. Shircore, J. O. Bull. Ent. Res., v, pp. 87-90. 13. Fiske, W. F. Report on Preliminary Investigations into the Bionomics of | Glossina morsitans in Uganda. 14. Minutes of Evidence of Departmental Committee on Sleep Sickness. Mr. G. A. K. Marshall; Quest. 2020. 21 NOTES ON THE MORE IMPORTANT INSECTS IN SUGAR-CANE PLANTATIONS IN FYJT. By Rosert Veitcx, B.Sc., F.E.S. Entomologist to the Colonial Sugar Refining Company, Lid.* The writer of the following notes has been engaged on entomological investigations in Fiji, Australia and Hawaii since 1914 ; the work has been entirely confined to the pests of sugar-cane and has been carried on mainly in the first-mentioned country, where insect depredations are very serious and materially reduce the profits on a considerable number of plantations. The investigation of these Fijian cane pests open up a field that is practically new, and so the work will probably continue for some considerable time. The length of the period that is likely to elapse before the publication of final results is such that it seems desirable to place on record a few of the main points in the life-histories of the more important insects of the sugar-cane plantations of that country, and so the writer has compiled these notes containing summaries of their life-histories in so far as they are at present known to him. The Sugar-cane Industry m Fj. The six sugar-cane mills operating in the Fiji Islands are all modern well-equipped factories handling large quantities of cane. Four of the mills are the property of the Colonial Sugar Refining Co. Ltd. (Fiji & New Zealand), the remaining two being owned by other interests. The writer is not conversant with the conditions at the latter mills, and so his remarks on the status of the various insects are based mainly on observations made in the districts supplying the mills of the Colonial Sugar Refining Company, namely, Lautoka, Rarawai, Nausori and Labasa, the first three being situated on the island of Viti Levu, and the fourth on the island of Vanua Levu. Nausori is the only one situated on a windward coast, and 1s generally spoken of as a “ wet” mill on account of the rainfall there being so much heavier than at the other three, which are usually referred to as “dry.” The use of the term “ dry ” does not mean that any of the mills so designated suffer from an insufficient rainfall; on the contrary it is generally ample at all the mills, and indeed in some seasons it is excessive. The average rainfalls for the eight seasons ended December 1916 are as follows :—Nausori 120 inches, Labasa 87 inches, Rarawai 82 inches and Lautoka 66 inches. Most of the rain falls during the hot season, which lasts from December to April. In Fiji sugar-cane is grown on comparatively small isolated blocks of land, which generally form a narrow interrupted strip along the coast, or along the banks of the more important rivers, large compact areas such as exist in Hawaii and other sugar-cane countries being quite unknown. The smallness and isolation of the * Published by permission of the General Manager, 22, ROBERT VEITCH. cultivation, the interior being one mass of mountains, which not infrequently run down to within a mile of the coast line. The most profitable soil for the growth of sugar-cane is found on the alluvial flats along the banks and near the mouths of the rivers; between these cultivated river flats and the mountains there is a variable quantity of undulating land and foothills on which some cane is grown, but only a small percentage of that land can be profitably put under sugar-cane, and consequently very little cane is grown at an elevation of more than one hundred feet above sea-level. The cultivated soils may be classified under the following headings: (1) alluvial flats, (2) red hill soils, (3) sands, (4) stiff clays and (5) “ tiri” or reclaimed salt- swamps. The alluvial flats are very rich and give heavy crops of cane, while the red hill soils are deficient in humus and give a light but sweet yield; the sands give good sweet crops in wet years, but in dry seasons the results are frequently poor. The stiff clays are difficult to work, and consequently they rarely give good returns. The reclaimed salt-swamps vary greatly in physical properties, but are all alike in giving only a medium or poor yield of cane. The season for replanting fields that have been under fallow is in the early months of the year, while the harvesting season usually commences in June and ends in December. The leading varieties grown are Badila, Malabar and Pompey. Badila is the best all round variety, and in 1917 it constituted 97 per cent. of the crop at Lautoka, 97 per cent. at Labasa, 88 per cent. at Rarawai and 32 per cent. at Nausori. At the latter mill Malabar is the leading variety, as it is believed to be the most suitable for the poorly drained flats surrounding that mill| Pompey is now being exten- sively grown on the second-class lands at Rarawai, and on such soils it seems to give a better financial return than Badila. The general rotation 1s a 25 per cent. one, which means that 25 per cent. of the land is under plant cane, 25 per cent. under first ratoons, 25 per cent. under green fallow and 25 per cent. under preparation for replanting. Mauritius bean (Stzzo- lobium aterromum) is the green fallow crop usually ploughed in, although cowpea is also used to a certain extent. The Cane Beetle Borer (Rhabdocnemis obscura, Boisd.) (fig. 1). The beetle borer has long been known as the most destructive pest of sugar-cane in Fiji, where it now constitutes a very serious entomological problem. The available evidence indicates that its original habitat is New Guinea, whence it has spread to North Queensland, Fiji, Samoa, Tahiti, Hawaii and other islands of the Pacitic. It is believed that the pest reached Fiji in a consignment of seed cane imported from Hawaii. The borer shows a marked preference for soft rank cane, and as this occurs most commonly in the heavy crops on rich alluvial flats, it follows that this pest is more destructive on such soils than elsewhere. The amount of damage does not depend solely on the type of soil on which the cane is grown, for it has been found that the susceptibility of the different varieties varies greatly. Badila is a soft sweet cane liable to fall and rot, while Malabar is a hard cane with a low sugar content and with a less marked tendency to fall to the ground when heavy tonnages INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 23 are grown, and as the borer prefers soft sweet stalks of cane to tunnel in, it almost invariably happens that on similar types of soil Badila suffers much more severely than Malabar. The white elongate oval egg of this pest is about one-sixteenth of an inch in length, and is laid in the rind of the stalk behind a half-loosened leaf-sheath, and a week after oviposition the grub hatches out and bores into the cane, becoming full-grown in eight to eleven weeks. In boring the cane the grub destroys a very large amount of tissue, partly by eating and swallowing it, and partly by chewing it to extract the sap. The tunnels made are not infrequently two feet long, but generally they do not extend to more than six inches; in tunnelling the cane the grub frequently eats too close to the rind, and in doing so it makes small rupture holes about the size of a pin’s head. The borer grubs can be readily located by the presence of these holes, which stand out somewhat conspicuously as a result of the rind im their immediate vicinity being of a rather lighter shade than elsewhere. The full-grown grub measures somewhat more than half an inch in length, and is legless, wrinkled and yellowish grey in body colour, the head being light reddish brown. The fifth and sixth adbominal segments are greatly swollen ventrally. Fig. 1. Rhabdocnemis obscura, Boisd., x 3. The full-grown grub prepares for pupation by eating a hole in the rind of the cane somewhat less than a quarter of an inch in length, and at a short distance above this hole it pupates in a cocoon of fibres separated from the tissue in which it has been tunnelling. The cocoon is extremely strong, and as it is very difficult for anything to penetrate it, the pupa is generally immune from the attacks of the small brown ant (Pheidole megacephala, F.); were the cocoon less strongly made a large proportion of the pupae would undoubtedly perish, for this ant is extremely voracious and a very formidable predator. 24 ROBERT VEITCH. The newly formed pupa is cream-coloured, but as the pupal stage advances the the darker markings of the future beetle appear and eventually show up distinctly. The pupa is half an inch in length and the pupal stage lasts for about two weeks. The newly formed beetle rests in the cocoon for a considerable number of days before venturing forth to feed and mate, and when it does so, it leaves the cane by the hole made by the grub prior to pupation. The period that elapses between the laying of the egg and the emergence of the fully hardened beetle is between three and four months. The beetle is half an inch in length and is a typical Curculionid ; the colour is dark brown, broken by several black patches on the thorax and elytra. The sexes can be distinguished by the fact that in the male the rostum is shorter, straighter, thicker and also rougher on the under surface than in the female (fig. 1). The damage caused by this pest is often very severe, for it 1s no uncommon occur- rence to harvest fields in which it is difficult to find a single stalk unattacked by it. The percentage of stalks damaged by borers at the Fiji Mills during the 1917 crushing season was 14, a figure that is sufficiently high to cause considerable anxiety. The activities of this pest lead to a very considerable reduction in the crop, for the sugar yield is affected in several ways. In the first place, there is the loss of sugar in the cane tissue swallowed or chewed by the borer grubs, and in badly attacked fields this is very considerable, because many of the stalks are so extensively tunnelled that they contain little or no sound cane. The loss of sugar is not confined to the tissue actually eaten, for it has been shown that the boring of the grubs materially reduces the sugar content of the untunnelled portions of infested stalks. Another source of loss in attacked fields is the increased lability of bored stalks of cane to break and fall to the ground, where they become still further infested and rot rapidly. The general effect of the attacks of this pest is to depreciate the value of a large proportion of the crop on the richer lands, and to render much of it absolutely valueless for milling purposes. The following control measures have been found to be of great assistance in Fiji :-—- (1) Seed cane for replanting fallow fields should be cut in areas known to be free from borer ; where this is not possible the seed used should be carefully inspected, and all infested stalks should be rejected and destroyed. (2) The trash on badly infested fields should be burned as soon as possible after cutting, as by so doing numbers of beetles underneath the trash and most of the eggs, stubs, pupae and beetles in the discarded cane will be destroyed. Cane stalks that are badly diseased or badly attacked by the borer are generally left on the field, and unless the trash is burned off it finds an excellent breeding ground in such discarded cane. Trash should not be burned on poor land, because on such soils 16 is necessary to preserve all humus-forming material ; fortunately, however, cane on poor land is seldom badly infested by the weevil, and so there is rarely any necessity for burning trash on such soils. (3) Fields to be replanted should be ploughed out as soon as possible after harvestins, thus depriving the pest of such an excellent breeding ground as neglected volunteer ratoons. (4) The beetles should be collected by means of traps of split canes laid in suitable fields. These traps attract large numbers of weevils which can be collected and destroyed, i INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 25 A prolonged effort was made to increase the efficiency of the control of this pest by the introduction of a Tachinid fly parasite, Ceromasia sphenophori, Vill., which was introduced by Mr. F. Muir from New Guinea to Hawaii, where it soon brought about a reduction in the losses on the plantations. As the conditions in Fiji appeared to be favourable to the introduction of the parasite, it was thought possible to repeat the success, and accordingly Dr. J. F. Ilhingworth, of the College of Hawaii, was commissioned by the Colonial Sugar Refining Company to introduce the parasite to this country. He arrived on 30th May 1913, and succeeded in rearing the parasite in breeding cages, from which he liberated large colonies, all of which failed to become established. He then returned to Hawaii after spending two or three months in Fiji, and from the date of his departure until the end of 1914 the work was carried on by Mr. D. 8. North (Mycologist to the Company) ; from January 1915 to March 1917 the work was superintended by the writer. Mr. North was successful in estab- lishing the parasite in several fields on Natova plantation, and it looked as if success was in sight, but unfortunately this field colony died out shortly after the breeding cage on the plantation was closed. The writer’s experience while in charge of the work was similar ; so long as the breeding cages were maintained, it was possible to find considerable numbers of parasitised borers in the fields adjoining the cages, but when these cages were closed down the field colonies dwindled and eventually died out ; 1t is thus evident that these field colonies depended for their continued existence on a steady stream of new parasites from the breeding cages ; when that steady stream ceased the rate of multiplication in the field seemed unable to counterbalance the very heavy mortality rate, and so the colonies gradually died out. The effort to establish the parasite lasted almost four years, and ended in failure in spite of the fact that very large colonies were liberated, and that at one time breeding work was being carried on at five different centres in order to give the fly a trial under all the different conditions prevailing in the sugar districts. This failure illustrates the uncertainty of all parasite work ; for the success in Hawaii seemed to assure an equally favourable result in Fiji. Climatic conditions cannot be held responsible for the failure, for the parasite was bred in very large numbers in captivity, clearly showing that the climate was favourable. The writer is of opinion that the large jumping spiders and the small brown ants, combined with peculiarities in agricultural methods and in the growth habits of the leading variety of cane, have been responsible for this disappoint- ing and unexpected result. The ease with which this parasite establishes itself when conditions are favourable is shown by the fact that it became established in North Queensland from a few flies that escaped from an intermediate breeding cage used by Muir and Kershaw when they were transporting the parasite from New Guinea to Hawaii. The Small Cane Beetle Borer (Trochorrhopalus strangulatus, Gyl.) (fig 2). This weevil is frequently found in cane. It is much smaller than the more des- tructive species, from which it can be easily separated in the larval, pupal and adult stages. The larva of the larger species has the 5th and 6th abdominal segments distinctly swollen ventrally, a feature which is absent in this insect ; and in the pupal stage the two can be readily separated by the fact that the smaller species does not make a cocoon. The adults can be distinguished by their colour, the smaller species being unicolorous, while the larger one has its brown colouring broken by black 26 ROBERT VEITCH. marks on the elytra, the thorax and elsewhere. Other less easily observed differences exist, but the above are quite sufficient to separate the various stages. So far the writer has never observed this species attacking perfectly sound cane, and his experience is that it generally breeds in rotten fallen cane, and less frequently in sickly standing stalks. Were it to turn its attention to sound cane it would un- doubtedly become a serious pest, as its rate of multiplication is very great ; but at Fig. 2. Tvrochorrhopaius strangulatus, Gyl., + 5. present its importance is very slight. The control measures recommended for the larger cane beetle would also be effective with this species. The White Grub of Sandy Soils (Rhopaea vestita, Arrow) (fig. 4). White grubs frequently occur in large numbers in certain types of soil, and are often the cause of very severe damage, so serious in fact that several tracts of coastal land have been thrown out of cultivation because of bad infestation. Two species have been found attacking cane on an extensive scale,—and being new to science, they have been described by Mr. Gilbert J. Arrow (Ann. Mag. Nat. Hist. (8) xvi, p. 319) under the names of Rhopaea vestita and Rhopaea subnitida ; both species are natives of Fiji. The local distribution of these grubs is very interesting; for on no occasion has the writer found the two species together in the same type of soil. A. vestita has always been found in sandy fields situated close to the coast, while A. submitida occurs in rich alluvial lands and in red hill soils, but not in sandy lands. Although the latter species is found in alluvial soils, it is never present in large numbers and has never been observed doing appreciable damage ; for although it undoubtedly attacks the cane, its numbers in such soils are so small that it has no apparent effect on the crop. On several occasions, however, this same species has caused considerable damage on red hill soils. &. vestiia is responsible for extensive injury year after year on sandy soils, and is by far the more destructive of the two species. The life-history of R. vestita has been more extensively studied than that of R. subnitida, and the main features of it are outlined in the following paragraphs. INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. ii The finely reticulated, elongate oval eggs are on an average about 2°5 mm. in length and 1:25 mm. in breadth when newly laid, but the egg increases in size as the incubation period advances, and just before hatching it is spherical rather than elongate oval, the diameter being approximately 3°5 mm. The eggs are laid singly at a depth of 15 to 20 inches, having an average incubation period of 31 days, the maximum and minimum being 34 and 28 days respectively. Dissection shows that each of the paired ovaries consists of six tubes each producing three eggs, the maximum number a female lays being 36, although in captivity it is usually considerably less. The eggs are laid from May to September inclusive, the chief egg-laying months being June and July. The grubs feed on both living and decaying vegetable matter in the soil, very large quantities of soil being swallowed in order to extract this ; but living vegetable matter, such as cane roots, is also attacked, and in this respect the damage is particularly severe in January, February and March, ie., when the grubs are well grown. When full-grown the white grub averages 27 mm. in length and 6 mm. in breadth; the body colour is creamy white, with the exception of the last abdominal segment, which appears greyish black as a result of the contents of the alimentary tract showing through the skin. The great majority of the grubs are found within six or seven inches of the surface of the soil all the year round, and so they are always within reach of the plough. In April, or, even earlier in some cases, the grubs enter the pre-pupal stage which lasts for a couple of weeks, at the end of which time the pupa is formed in an earthern cell at a depth of 6 to 18 inches below the surface. The pupa averages 19 mm. in length and 8 mm. in breadth; the pupal period lasts for 31 days. Pupae are found from April to September, but are most plentiful in April and May. The beetles do not leave the soil immediately after the transformation from pupa to beetles, but lie resting in their earthern cells for two or three weeks before emerging to feed and mate. Beetles have been found in the soil on 7th February, but this is unusually early, for they do not reach their maximum abundance until June. During the swarming season they are found in thousands on their food- plants, the chief of which are acacias, guavas, coconuts and cane, but for the purpose of mating they will also swarm in large numbers on any prominent object such as windmills, telephone poles or large trees. In the swarming season the beetles emerge about twenty minutes after sunset, and mating proceeds for two hours, after which they settle down on their food-plants, where some of them feed. while others remain quite inactive; shortly before sunrise they return to the soil or to their hiding places under cane trash and fallen leaves. The most prominent feature of this return to the soil is the rapidity with which it is accomplished, as not a single beetle is to be seen a couple of minutes after the first one has taken flight. The beetle is reddish brown in colour, and is clothed with very small yellowish setae, except on the breast, which is covered with long yellow hairs. The average length is 17 mm. and the average maximum breadth 9 mm. The life-history may be summarised as follows :—The female lays twenty to thirty eggs singly in sandy soil in June and July at a depth of 15 to 20 inches, and after an incubation period of 28 to 34 days the grubs emerge and feed on humus and 28 ROBERT VEITCH. living vegetable matter, being most destructive in January, February and March of the next year, by which time they are nearly full-grown. They pupate in April and May in earthern cells at a depth of 6 to 18 inches, and after 31 days or so in the pupal stage the beetles appear. The adults do not leave the soil immediately after their transformation, but rest in the earthern cells for a week or two, and then emerge to feed and mate and repeat the life-cycle. There is only one generation in a year. This pest is very destructive on sandy soils, in which it attacks cane of all ages. Fields replanted in January, February and March almost invariably suffer severely, as the grubs at that time are well grown and immediately attack the young plants, destroying the roots, or eating the young shoots at the base and thus severing them from the set and the root system. Many of the attacked plants die, but others survive, although more or less crippled, the older cane being better able to with- stand the attacks, as its recuperative powers are greater, but even such crops are frequently severely checked. The following natural enemies take a heavy toll of the white grubs of this species. The wireworms of Monocrepidius pallipes, Esch., attack and destroy the grubs in large numbers, and the Scoliid wasp, Discolia ovalauensis, Sauss., parasitises very many ofthem. An unidentified species of mite attacks the grubs in considerable numbers when kept in captivity, and the attacked individuals generally die. In the field conditions do not seem to be so favourable to the mite, and it is only rarely that one there finds grubs as badly infested as in the laboratory. Minah birds destroy large numbers of grubs during cultivation operations, and are undoubtedly of great assistance in minimising the losses from this pest. Fields known to be badly infested by white grubs should not be planted until April, as by that time most of the grubs will be in the pupal or prepupal stage, and so the crop will be free from serious attack until December or January, when the cane will be well established and so better able to resist the grub attacks. The collection and destruction of beetles and grubs is also of assistance in controlling this pest, while the frequent ploughing of infested fields will expose many grubs to destruction by minahs. The White Grub of Alluvial and Red Hill Soils (Rhopaca subnitida, Arrow) (fig. 3). The cane grub of sandy soils has been dealt with in considerable detail in the preceding paragraphs, but a much briefer reference will suffice for the less destructive species, R. subnitida, partly because it is of less importance, partly because less detail is required owing to the similarity between the life-histories of the two species, and partly because it has been less closely studied than RA. vesiata. The seasonal occurrence of the two species is practically the same, and so far the writer has not observed any difference in feeding habits. The beetles eat acacia, coconut, cane and guava leaves, but the amount of food consumed is small. The beetle, pupa, grub and egg of R. submitida are each larger than the corres- ponding stages of R. vestita. The beetle of the latter species can always be readily identified by the presence of a smooth median area on the pygidium, which in the former is setose all over. The writer cannot distinguish between the pupae and INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 29 eggs of the two species, but the well grown grubs are not difficult to separate, as that of R. subnitida is more hairy and has longer legs than the grub of R. vestita. The egg-laying records of females kept in captivity and the dissection of ovaries show that in reproductive powers the species are similar. It is therefore difficult to explain why this grub should be so much less numerous than the other. The beetles are just as hardy, and in the laboratory the grubs are not nearly so easily injured as those of the sandy soil species. The explanation of the relative scarcity of this grub will probably be found in the presence of some enemy which attacks R. subnitida only and has so far escaped observation, or in the existence of some other adverse factor in the soil types frequented by this grub. This is a point worthy of further investigation. The control measures recommended for R. vestita will also be of agsistai:ce in checking the destructive activities of this species. ql er i a A a Fig. 3. Rhopaea subnitida, Arrow, X 4; Fig 4. Rhopaea vestita, Arrow, x 4: a, portion of elytra highly magnified @, portion of elytra highly magnified to show the scaling. to show the scaling. The Rose Beetle (Adoretus versutus, Har.) (fig. 5). The rose beetle is not generally regarded as a pest of cane, but on a recent plantation inspection the writer observed the grubs of this species attacking the germinating eyes of cane sets planted some two or three weeks previously. The grubs of this Rutelid beetle have been found in all types of soil, and almost invariably occur within six inches of the surface. The beetles are found all the year round and are a great nuisance to all rose growers, as they attack and destroy the 20 ROBERT VEITCH. leaves of the rose bushes, and in many districts it is impossible to grow roses suc- cesstully unless they are screened every evening before sundown. Fig. 5. Adoretus versutus, Har, <4. The beetle feeds on the leaves of sugar-cane, guava, acacias, cacao and rose bushes, This species is fortunately of slight importance in the cane-fields, and no control measures are required. The Introduced Parasite of White Grubs (Scolia manilae, Ashm.). While on a recent visit to Hawaii the writer was much impressed by the degree of success obtained in the introduction of Scola manilae, Ashm., to control the white grubs of Anomala orientalis, Waterhouse. This pest was accidently imported from Japan in quite recent years and is now firmly established on two plantations on the island of Oahu, where it is seriously reducing the cane crops. A considerable number of parasites were introduced to control it, and of these the Scolia shows most promise ; at the end of 1917 it had become extremely numerous, and showed such splendid prospects of success that the writer determined to take a colony back to Fiji, in the hope that they might attack the Fijian cane grubs and rose-beetle grubs. The officers of the Hawaiian Sugar Planters Association gave all possible assistance, and supplied laboratory accommodation while the writer was breeding the colony, and also much helpful advice based on their own experience in handling this parasite. The colony was successfully transported to Fiji, and six hundred fertilised females and over two hundred males were liberated in an area that is always badly infested with stubs. It is still too early to say whether or not success has been attained, but the laboratory trials showed promising results, for the parasite attacked the grubs of R. vestita and A. versutus and laid eggs on them ; adult parasites were subsequently reared from some of these eggs. A distinctly unfavourable factor was introduced a few weeks after the parasites were liberated, for a heavy flood occurred, resulting in a large portion of the grub-infested area being under water for a few hours, and if larvae and pupae of the parasite were present in the soil, it is almost certain that a considerable number of them perished. INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. Pl | The Sugar-Cane Wireworm (Simodactylus connamomeus, Boisd.) (fig. 6). The following species of wireworms have been found in the soils of the Fijian cane- fields : Simodactylus cinnamomeus, Boisd., Lacon stricticolis, Fairm., and Monocre- pidius pallipes, Esch. The first two species, but more particularly S. cownamomeus, are very destructive to young cane plants, while the third is distinctly beneficial, as it is a very formidable enemy of the white grubs of Rhopaea vestita. S. cinnamomeus is by far the commonest of the three species, being found in all five types of soil, but reaching its maximum abundance in the rich low-lying flats, where it frequently occurs in enormous numbers. The full-grown wireworm is sub-cylindrical in shape, and measures 25 mm. in length and 2mm. in breadth. The dorsum varies in colour from a very light brown to a blackish brown, an intermediate shade being typical ; the pleura and venter are pale brown. The newly hatched wireworm measures 2 mm. in length, and for a few days is of a semi-opaque white colour, but rapidly assumes a faint brown tint. It grows very slowly, the available evidence pointing to a larval period of not less than three years. The older wireworms under laboratory conditions moult every ten or twelve weeks. Fig 6. Simodactylus cinna- momeus, Boisd., x 3. This wireworm pupates in a small earthen cell with smoothly plastered inner walls, and ten or eleven days later assumes the adult form. The beetle does not leave the soil immediately after its transformation, but generally rests in its cell for a fortnight or longer before emerging to feed and mate. The beetles are inactive during the day, and may be found in large numbers hiding behind the leaf-sheaths of cane, maize and various weeds, and in maize cobs, they also hide under lumps of earth on the surface of the soil. Towards dark they become more active and leave their hiding places. The females lay their eggs in clusters or strings in small cavities formed in lumps of moist friable soil, the number of eggs per cluster or string varying from five to fiitty. The elongate oval eggs are pearly white in colour and when newly laid measure ‘7mm. in length and 3 mm. in breadth ; in the cool months of June and July the mcubation period varies from 16 to 19 days. 32 ROBERT VEITCH. The ravages of this pest are particularly severe in newly planted fields of cane. Stools of cane of all ages are attacked, but it is only in the earlier stages that the plant succumbs to the injury, for once it is properly established it is well able to survive the attacks of the wireworms, being then sufficiently vigorous to send out new eyes and roots to replace those destroyed by them. The wireworms in newly planted fields commence the attack by eating the butts or the ends of the sets or cuttings of cane, and then five or six days later they turn their attention to the eyes and roots, which by that time, provided germination conditions have been favourable, have started to swell and come away. The small roots are frequently destroyed, and in a bad attack only 20 or 30 per cent. of the eyes escape injury. The wireworm attacking the eye eatsa hole at the base and then devours the whole of the softer inner tissues, thus killing it. Eyes that have germinated and sent up shoots nine or twelve inches long are also attacked and frequently so badly damaged that they die off. The greatest damage to the eyes and young roots occurs in the second and third weeks after planting. In a bad attack as many as 75 per cent. of the sets or cuttings are completely destroyed. ' Heavy additional expenditure is incurred when a newly planted field is badly injured, for the numerous misses or failures have to be replaced by new sets or trans- plants, and this is a costly operation, especially if the latter are used. Preventive and remedial measures are urgently required for the control of this beetle, but unfortunate:y it seems to be particularly healthy, being apparently free from parasitic and predaceous enemies and also immune to the most virulent poisons ; this latter fact was demonstrated in a poisoning trial in which the wireworms feeding on the poisoned cane seemed if anything even healthier than those feeding on the control sets. The following measures have, however, been found very beneficial in minimising the losses :—(1) In fields known to suffer from wireworm attacks it is well to make provision for failures by the continuous planting of a certain proportion of the rows ; usually there is a space of at least fifteen inches between the sets or cuttings in the rows, but in continuous planting no such space exists and hence the number of sets per row is much greater than in the ordinary rows. When required the surplus stools in the rows planted continuously should be dug up to fill the blanks or misses in the ordinary rows. This transplanting should be done only in wet weather. (2) Transplants only should be used to fill the gaps, as new sets are at once attacked by the wireworms and many of them are destroyed ; the transplants being older are better able to resist the attack, and if good rains fall shortly after transplanting, very few of them fail, and so a full stand of cane is obtained. (3) The cane on alluvial flats should be second ratooned, as by so doing the annual planting area is reduced, and thus the annual loss from wireworms is minimised. (4) Drainage should be im- proved wherever possible, and the importance of clean cultivation cannot be over- estimated. The Yellow Wireworm (Lacon stricticollis, Fairm.) (fig. 7). This species 1s not nearly so destructive as S. cumnamoneus, being much less numerous and not nearly so voracious; it is occasionally met with im numbers in alluvial soils and it also occurs in red hill and sandy soils, but 1s never so abundant | as the common species. INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 33 The pearly white egg is more nearly spherical than that of S. connamomeus, and measures “6 mm. in length and ‘4 mm. in breadth; the mcubation period is 12 or 13 days. The larva or wireworm is pale clay-yellow, except for the head, the first thoracic segment and the last abdominal segment, which are brown; the full-grown wireworm measures 18 mm. in length and 2°5 mm. in breadth, and the duration of the larval stage is probably three years or longer. The pupa is Fig. 7. Lacon stricticollis, Fairm., X 4. typical of the ELaTeRIDAE, and the blunt posterior angles of the thorax serve to identify the pupal stage of this species, as in the other two the posterior angles are backwardly produced as sharp teeth. The pupal stage lasts for 13 or 14 days, The feeding habits of this species seem to be similar to those of S. cinnamomeus, and the control measures recommended for that species would be just as effective for this one. The Predaceous Wireworm (Monocrepidius pallipes, Esch.) (fig. 8). This species has been found only in sandy soils in association with the white stubs of Rhopaea vestita, and has never been observed attacking cane. The larva Fig. 8. Monocrepidius pallopes, Esch., X 4. is very similar in appearance and size to that of Lacon stricticollis, from which it can be distinguished by the shape of the caudal notch, which in Monocrepidius (C572) C 34 ROBERT VEITCH. pallipes. is smaller and carries smaller teeth on the margin. The larval stage probably lasts for three years or longer and the pupal stage for nine or ten days. The most interesting point in the life-history of this species is the fact that it feeds voraciously on the white grubs of Rhopaea vestita, and it is undoubtedly one of the most important factors holding that pest in check. It has been observed eating grubs in the field, and in a prolonged laboratory experiment it was found to consume on an average one grub a week; in the experiment referred to, the wireworms were placed in tins containing rich soil, a white grub and a piece of fresh cane, thus giving them the choice of animal food, living vegetable matter and decaying vegetable tissue or humus. The animal food was almost invariably preferred by this species. The Sugar-Gane Army-Worm (Cirphis unipuncta, Haw.). The caterpillars of this Noctuid moth frequently appear in considerable numbers, and are responsible for an appreciable amount of damage to the younger crops. The caterpillars can be found all the year round, but generally are most abundant in the cooler months. This pest is a notorious enemy of maize and other cereals in the United States, but fortunately the damage in Fiji has never reached serious dimensions. The most extensive outbreak of this insect in the writer’s experience in Fiji occurred in May and June 1918, in young plant fields six to ten weeks old. The caterpillars were in large numbers on ten plantations in the Lautoka and Rarawai Mill districts ; no damage was observed in cane older than nine months, and with one exception the attacks were confined to plant crops. The army-worms shelter during the day in the folded leaves of the central shoot of the cane plant, and after dusk they move out on to the tips of the leaves, where they feed ; when they are present in large numbers they frequently strip the leaves right down to the midrib, thus severely retarding the development of the young plant. The young caterpillars feed on the leaves of tender grasses, until they are able to attack the harder tissue of the cane leaves. There are usually five moults at intervals of three to five days, the caterpillar being full-grown in three weeks. The full-grown caterpillar is 36 to 39 mm. in length, and is greenish brown in colour. When ready to pupate it enters the soil and forms an earthern cell at a depth of two or three inches, and inside this it pupates a few days later. The pupal stage lasts ten or eleven days. This pest is well held in check by its enemies, and its attacks rapidly die down, for parasites and predators take a heavy toll of it. The minah bird destroys numbers of the caterpillars, and a hornet (Polistes macaensis, F.) kills even more ; the imago of the latter is very fond of the caterpillars, which also form an important item in the food given to its grubs. The Tachinid fly, Sturmia bimaculata, Htg., and a Braconid (Apanteles sp.) have been bred from this species of army-worm, and the larvae and adults of various CaraBipaE have been found to feed on the cater- pillars. Natural enemies are usually sufficient to hold this species in check, but in the event of a serious outbreak resulting from a combination of circumstances unfavourable to the natural enemies it could be controlled by the use of poisoned sprays. INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. gO A Second Species of Sugar-Cane Army-Worm (Cirphis loreyi, Dup.). The army-worms of this species are occasionally destructive in the cane areas although generally their attacks are confined to a few stools along the edges of fields. During a residence of several years in Fiji the writer has seen only one attack in which the damage was at all serious; this occurred in an eleven-month- old crop of plant cane, and was so severe that many of the leaves were stripped down to the midrib. This species occurs in the old word tropics, and in Africa it is a serious pest of maize. Minahs, hornets and the Tachinid, Sturmia bimaculata, Htg., keep this pest in check, and the use of artificial control measures is unnecessary. The Mauritius Bean Army-Worm (Prodenia litura, F.). Mauritius bean is extensively used on all the plantations, it being part of the normal rotation to long-fallow land once every four years with Mauritius bean as a covering crop, which is eventually ploughed in as a green manure. The Mauritius bean army-worm occurs in every fallow field and is responsible for a very appreciable check to the growth of the crop. The eggs of this moth are laid in clusters or nests, each of which contains from 500 to 700 eggs, the usual spots chosen for egg-laying being the leaves of the food- plant. The caterpillars hatch out after an incubation period of three days, and immediately commence eating small irregular patches on the under-surface of the leaves, but as they grow older they eat through the leaf tissue, destroying epidermis, ground tissue and small veins. The caterpillars also attack cotton, tobacco and cacao, and the writer has recently bred them from the fruits of the “ivi” or Tahitian chestnut. They are full grown in three weeks, and then pupate in earthern cells two or three inches below the ground, the moth emerging after a pupal period of eleven or twelve days. The enemies of this insect are the minah, the hornet and the small brown ant (Pheidole megacephala, F.). The ant destroys large numbers of the eggs. Remedial -measures are not at present required for the control of this pest. The Moth Borer of Cane (Zrachycentra chlorogramma, Meyr.). The larvae of this Tineid moth are occasionally found tunnelling cane, and causing a slight amount of damage. Their work has usually been most conspicuous in low-lying reclaimed tiri fields, which frequently contain a considerable proportion of sickly canes. Some of these moth borers have been found in half-rotten cane, but others have been found in soft light-coloured Badila stalks which were standing up well and were far from being rotten, although they were of a somewhat sickly appearance. No stools of perfectly healthy cane have been attacked by this insect. The full-grown larva is about 25 mm. in length, and lives in a case which it drags about from place to place; this case is a spun web to which numerous small particles of cane, earth and vegetable debris are attached. The case affords a considerable amount of protection to the larva, which is able to wriggle into or out of it with astonishing rapidity. The damage caused by this species bears some resemblance to the work of the beetle borer, but it differs from it in several important respects. The tunnels of the moth are wider, shorter and more irregular than those of the beetle borer and they (C572) a2 36 ROBERT VEITCH. contain a much rougher, darker coloured frass. Other points of importance in the mode of attack of this pest are firstly the fact that, unlike the beetle borer, it swallows all the cane it chews, and secondly that it frequently leaves its tunnel and moves up the cane-stalk or even to another stalk, where it commences a new tunnel. Otherwise the larva destroys cane in a manner somewhat similar to the beetle borer by eating tunnels along the stalks, destroying large quantities of tissue and lowering the sugar content of the untunnelled portions of the affected stalks. The full-grown larva pupates in an extremely tough case made of small pieces of cane fibre overlapping each other like roofing tiles ; it is securely closed at both ends and must afford very good protection to the pupa. The pupa is formed sometimes in the old larval tunnel, and at other times behind the leaf-sheaths or under debris, the pupal period lasting for three weeks. The burning of the trash after harvesting and the exposure of tunnelled stalks to the heat of the sun will largely check the increase of this insect should it ever show signs of attacking sound cane. The Cane Leaf-Miner (Cosmopteryz sp. N.). The larva of this very small and beautiful moth mines in the midribs of cane, being particularly abundant in young plant fields, where it does a slight amount of damage by tunnelling along the midrib, interfering with the supply of sap to the leaves and discolouring the tissues surrounding its tunnels. This species is well controlled by parasites. The Hornet (Polistes macaensis, F.). The common hornet of the cane-fields is far from being a pest, but it is so mtimately connected with the sugar plantations that no reference to the insects of the cane- fields would be complete without an outline of its life-history. This hornet is only a recent introduction, but it has already become one of the commonest insects in the islands, and has probably attracted more general interest than any other species in Fiji. It is widely distributed throughout India and the Kast, and appears to have reached these islands some fifteen years ago, having in all probability been introduced on a labour ship from India. It showed up first at Suva and has now spread all over Viti Levu; it has also gained a footing in Vanua Levu, and in a few years time it will doubtless be generally distributed throughout the islands. The life-history of this species may be summarised as follows :—A fertilised female that has hibernated during August and September commences nest-building at the beginning of October. The cells of the nest are made of the usual fibrous material mixed with cement, and in each the female lays a small greyish white egg from which the young larva hatches in afew days. At first the larva is fed on sugary substances, but after it is a few days old the diet is changed to an insect one. The larva pupates 14 to 18 days after hatching, and the imago emerges after a pupal period of 15 or 16 days; after the emergence of the first new generation progress on the nest is very rapid, and by April it may be as large as 12 inches in diameter. Towards the end of April males and females appear in equal numbers whereas previously the new hornets emerging were nearly all females. This last generation leaves the nest and the hornets hide in sheltered spots, emerging on sunny days in June and July to mate, but by the INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 37 end of the latter month mating has ceased and all the males are dead. The females hibernate for seven or eight weeks during August and September, and then they reappear and commence nest-building. This species is a desirable addition to the Fijian fauna, because it acts as a useful check on many pests. The writer has observed the hornets feeding pieces of cater- pillars, flies and grasshoppers to their larvae, while the hornet itself has been seen eating full-grown crickets, young grasshoppers, army-worms, jumping spiders, flies and small beetles. The sting of this species is very painful, and therefore many neaple. losing sight of its value as a predator, have passed a verdict of unqualified condemnation on it. While it is true that labourers and others, including the writer, are sometimes badly stung, yet it is fortunate that during the height of the cane-cutting season the hornets are hibernating, and so the labourers can work without fear of being stung. At the beginning of the season a few nests are met with and towards the end of the season the small new nests begin to show up, but even then the danger is not nearly so great as it would be were the cane cut from January to April, when the nests are large and defended by numerous individuals, During the early months of the year there is little occasion for going into big cane at Lautoka, Rarawai and Labasa Mills, and so the labourers do not generally encounter the hornet at its worst; but at Nausori Mill the position is different, for there the usefulness of the hornet is distinctly lessened by the fact that the cane is trashed during the season when the nests are at their maximum size, and considerable expense is incurred in sending special labourers (in canvas suits, gloves and veils) into the cane to destroy the nests before the ordinary labourers can enter the field to trash the cane. The trashing of cane has now been abandoned at the other three mills, being regarded as an unprofitable measure under the climatic conditions of the dry districts. The Leaf-Hopper (Perkinsiella vitiensis, Kirk.). This pest can be found in any cane field, but fortunately it does little damage, as it is kept in check by numerous natural enemies, the most effective of these being the ege-parasites Ootetrastichus, Paranagrus and Anagrus. The Stylopid parasite, Elenchus tenurcornis, Curt., attacks both young and adults, and the numerous spiders in the cane-fields also take a heavy toll of this pest. The eggs of this leaf-hopper are laid beneath the epidermis of the cane plant, usually along the midribs of the leaves, but also in the internodes of the stalk, or even in the leaf-sheaths ; generally three eggs are laid in each cavity, the position of which is indicated at first by the piece of white wax covering the opening of cavity, and later by the reddish discoloration of the tissues surrounding it. The young leaf-hoppers hatch out and feed on the cane, sucking the sap and ex- creting honey-dew, on which a black fungus grows. In Fiji the writer has never seen the leaf-hopper present in numbers large enough to stunt the growth of the crop appreciably, and control measures are unnecessary. The Mauritius Bean Bug (Brachyplatys pacificus, Dalli.). This little. black Pentatomid bug occurs wherever Mauritius bean is grown, and it can also be found feeding on a number of the leguminous weeds that flourish along the . 38 ROBERT VEITCH. drains and roads on the plantations. If the upper leaves of any bean plant be pushed aside the bean bug can often be found in such numbers that the stem of the plant is literally black with them. : The bug is 5 mm. in length and belongs to the sub-family ScUTELLERINAE. The eggs have been found on the leaves of Mauritius bean, on the leaves of the commoner weeds of the cane-field, on pieces of timber lying in fields and even on the doors and window-sills of bungalows. The eggs are laid in a double row, the number in each set varying from ten to thirty. Hach egg is about one millimetre in length and is an extremely neat white cylinder with a little lid that is opened when the young bug emerges ; in most cases the lid falls off when the bug leaves the egg, but sometimes it remains hanging half open as if it were hinged to the cylinder. The eggs are laid end to end in a double row with the lid ends directed outwards. They hatch out after an incubation period of six or seven days. The newly emerged bugs immediately start sucking the plant sap. They moult frequently, but the exact number of the moults and the intervals between them have not yet been definitely ascertained. The numbers of this pest are often so great that they considerably retard the growth of the green crop, but the damage is not sufh- ciently serious to warrant the adoption of control measures. The multiplication of the bugs is checked by a Chalcidid egg-parasite described by Mr. J. Waterston as Ooencyrtus pacificus, and an important fungous enemy is an unidentified species of Isaria which attacks the nymph and imago. The Gane Aleurodid (Alewrodes comata, Mask.). This Aleurodid is occasionally found in extremely large numbers on the under surface of cane leaves, but as a rule it is not common, and even in its worst attacks it is only a minor pest. The pale yellow eggs are attached to the leaves and are usually found in clusters containing sometimes as many as 150 eggs. The various stages are all typical of this family. The damage to the cane is twofold ; for in the first place, it is stunted by the loss of sap sucked out of the tissues by the Aleurodid, and in the second place, a sooty fungus grows on the excreted honey-dew and interferes with the process of photosynthesis. The most effective check to the increase of this pest is the maggot of a Syrphid which feeds on all its stages. The life-history of this predator has been worked out and is briefly as follows. The white cylindrical egg of the Syrphid is firmly cemented to the underside of the cane leaf, and is almost invariably laid alongside an actively ovipositing Aleurodid. The young maggot hatches out in three days, and imme- diately starts to feed voraciously, attacking all stages of the pest, but seeming to prefer the eggs and the adults to the larvae and pupae. The quantity of food consumed is enormous, as is illustrated by the fact that a single maggot has been observed to destroy as many as one hundred eggs in a day, as well as a considerable number of adults; while another was seen destroying thirty eggs and four adults in twenty minutes. The maggot is typical of the Syrphid family, and is full-grown i seven or eight days ; it forms a greenish pear-shaped pupa, from which the adult emerges six days later. This Syrphid is an unidentified species of Xanthogramma. INSECTS IN SUGAR-CANE PLANTATIONS IN FIJI. 39 The Cane Mealy-Bug (Pseudococcus bromeliac, Bouche). This mealy-bug is often found at the nodes of the cane-stalks behind the leaf-sheaths, where it punctures the rind and sucks the sap, thus lowering the vitality of the plant. It rarely occurs in large numbers and is of only slight economic importance. The older females are quite active, but the young nymphs move about freely until they find a suitable spot at which to pierce the rind of the cane ; when once that is found they become very sluggish, and settle down to feed and reproduce. Locusts (Locusta danica, L., and Cyrtacanthacris guttulosa, Walk.). These two species sometimes become numerous along the edges of fields, stripping the cane leaves down to the midribs, but the damage is rarely serious. The minah bird is a very efficient check on the locusts, which form its principal food; whenever trash or any rubbish is being burned off in a field the minahs gather in large flocks round the edges and reap a rich harvest from the insects that are driven before the fire and smoke ; these consist very largely of grasshoppers. Conclusion. The three most important pests in the cane fields are undoubtedly the beetle borer, the wireworm and the white grub, each of which constitutes an entomological problem of the greatest importance. Their life-histories have been as fully studied as the time at the writer’s disposal permits, but there are still many points of the greatest im- portance that require investigation in the immediate future. Although much research work remains to be done, a number of satisfactory control measures have been evolved, and their enthusiastic adoption will undoubtedly lead to a great reduction in losses ; while there is every reason to believe that with an increasing knowledge of the habits of these pests further and still more effective measures will be devised, The other pests more briefly referred to in these notes are of minor importance only, and although they have to be recognised as injurious on the plantations, the damage they cause is at present so slight that control measures are unnecessary. In conclusion, the writer has to acknowledge the courtesy of the Imperial Bureau of Entomology in identifying the species referred to in these notes. Babloography. (1) Muir, F. “ Notes on some Fijian Insects,” Bulletin No. 2, Hawanan Sugar Planters’ Experiment Station, 1906. (2) Jepson, Frank P. “ Report of Economic Entomologist,’ Annual Report of Department of Agriculture, Fiji, pp. 53-54. (3) Ilhngworth, J. F. “ Further Notes on the Breeding of the Tachimd Fly parasitic on the Cane Beetle Borer,” Jl. Econ. Entom., Concord, vii, no. 5, October 1914, pp. 390-398. (4) Ilingworth, J. F. “A New Pest of Cane in Fiji,” J]. Ecom. Entom., Concord, vu, no. 6, December 1914, pp. 444-445. (5) Veitch, R. “ The Sugar Cane Wireworm in Fiji,” Agricultural Report No.1 of the Colonial Sugar Refining Company, 1916. (6) Veitch, R. “ The Hornet in Fiji,” Agricultural Report No. 2 of the Colonial Sugar Refining Company, 1917. hub rare ie, ‘ wy, ve bt i Py) ee rey i oy ee cieth yy) bagtt 4 ts aa 484 ceatiag rere f Ppa pein, § el inhigs id. alt ye hhaXe MRD pie onal: Vise 41 A NEW LYGAEID BUG FOUND AMONG STORED RICE IN JAVA. By W. L. Distant. Dr. Guy A. K. Marshall, Director of the Imperial Bureau of Entomology, has requested me to indentify four specimens of a Lygaeid sent to him by Dr. C. J. J. Van Hall from Java, where it was found among stored rice. It is a very distinct species and represents a new genus of which I here give figure and description. Fam. LYGAEIDAH. Subfam, APHANINAE. Div. Myopocuaria. Ampera, gen. nov. Head about as long as anterior lobe of pronotum ; first joint of antennae a little shorter than head; pronotum longer than broad, transversely constricted before middle, collar narrow, anterior lobe distinctly globose, posterior lobe strongly centrally longitudinally incised, the basal lateral angles subnodulosely prominent ; scutellum broadly subangulate ; rostrum with the first joint slightly passing base of head— imperfectly seen in carded specimens; anterior femora moderately incrassated ; other characters generally as in Pamera, from which it principally differs by the structure of the pronotum and scutellum. Ampera intrusa, sp. nov. (fig. 1). Head, anterior lobe of pronotum and scutellum, black or blackish, posterior frontal lobe more castaneous ; corium more or less stramineous, the apical area more or less infuscate ; membrane black or piceous, body beneath blackish ; legs pale stramineous ; Fig. 1. Ampera intrusa, Dist., sp-n. second joint of antennae considerably longer than first; membrane not or very slightly passing abdominal apex ; rostrum stramineous, imperfectly seen in carded specimens. Var. Head and pronotum and body beneath, dark ochraceous, Other structural characters as in generic diagnosis. Long, 33-4 millim. Java (Dr. Van Hall—from stored rice). A USEFUL BREEDING CAGE.* By Laurence D, CiEare, Jnr., F.E.S., Biological Division, Dept. of Scvence and Agric., British Gurana. Entomological investigations in the tropics usually entail a certain amount of difficulties not experienced in temperate regions, and it is often necessary to devise various kinds of apparatus to overcome these. The breeding cage described here was devised by myself as the result of a number of difficulties experienced in br2eding experiments. Cages of this type have been in constant use in this laboratory for the past four years and have been found to meet almost every requirement. The ease with which they are constructed, together with the possibility of varying their size to suit one’s needs, as well as their portability, are all highly commendable features. Fig. 1. A portable breeding cage The cages are composed of cylinders of brass mosquito-proof wire netting, 20 meshes to the inch, held together by brass paper-fasteners. To the top and bottom of these cylinders covers are fitted. In this laboratory large glass petri dishes, between 8 and 10 inches diameter, are used for this purpose. Galvanised iron pans may, however, be used instead of these dishes, and if made in nests would be a decided advantage when travelling ; the wire being carried in rolls. It will be seen that the diameter of these cages is limited only by the size of the pans, while the height can be * Published by permission of the Director of Science and Agriculture, British Guiana. 44 LAURENCE D. CLEARE, JNR. varied from 3 feet, the width of the netting, down to a few inches. If it is desirable to use cages taller than 3 feet, and this will only be very rarely, the height may be increased by joining on an additional strip of netting by means of paper-fasteners. Originally, fine muslin, held in place by rubber bands, was used for the tops of these cages ; but after some large larvae were attacked by rats the dishes came into use. Obviously muslin can still be used in certain places with advantage, especially when weight has to be considered. The wire netting is usually fitted inside the dishes both top and bottom, but in cases where earth is necessary for pupation it is advantageous to fit the cylinder outside the bottom ; the advantage of this will be found when lifting off the wire to replace the food-plant. Food-plants, of course, have their stems placed in small bottles or other receptacles containing water. Cages of this type have been used in tick breeding experiments by Mr. G. KE. Bodkin, and when thus employed were placed in larger dishes containing water and kerosene oil to prevent the escape of the ticks. This method could also be used to prevent attacks of ants, which, in this Colony at least, have always to be considered in insect breeding, It may here be mentioned that in this laboratory all tables used for breeding ex- periments are kept standing in bowls containing water and kerosene oil. Were the bare wooden legs of a table placed in this mixture the oil would penetrate the wood and ruin the legs. Professor Harrison has devised a method to avoid this ; small iron rods about 6 inches long and 1 inch in diameter are driven perpendicularly into the legs and it is these that come in contact with the mixture of oil and water.* *[As an alternative to water and kerosene, dusting une saucers occasionally with pyrethrum powder has been found very effective in keeping off ants; this has the fdaeekare that the legs of the table require no special adaptation.—ED.]_ 45 NEOTOXOPTERA VIOLAE, THEO., AND ITS ALLIES. By A. C. Baker, B.S.A., Ph.D., US. Bureau of Entomology, Washington, D.C. (PuateE I.) In 1915 (Bull. Ent. Res. vi, p. 1381) Theobald described an Aphid taken on violets in Africa as Neotoxoptera vielae. He has also recorded a statement by Davis that it probably is Rhopalosiphum violae, Pergande, 1900 (The Entomologist, xlix, p. 149). A study of Pergande’s species as it is to be found in greenhouses in the United States and Canada shows that the form described by Theobald is really an aberration from Pergande’s species. This is shown clearly by the figures accompanying this note. Figure | is the typical forewing of wolae, Perg., and from examples of this type aber. rations have been found as shown in figures 2 to 6. It will be noticed that there are several cases in which the media is once branched, the condition met with in Theo- bald’s form. Figures 4 and 6 show a much greater departure from the type, and yet these forms as well as all the others figured can be obtained from one colony and in one “ family ” of the species. The hindwing as figured by Theobald is quite typical for Pergande’s species (fig. 7), but here also figures 8 and 9 represent in each case an aberrant form. The writer is unable to state that similar conditions would be met with also in Africa and that all of these types would occur. From the fact that they occur in _ America, and bearing in mind that nearly all the allies of this species present similar conditions, one would expect to find them. In order to understand the proper generic designation for this species it is necessary to review briefly the small group to which it belongs. In the subfamily APHIDINAE there is small group, the PENTALONINA, the members of which are mostly tropical or subtropical. This is based on Pentalonia nigronervosa, Coquerel, 1859 (fig. 10), an Aphid with most remarkable venation. This venation, as far as the writer’s expe- rience goes, is constant, although no large rearings have been made. To understand this venation it is only necessary to look at Idiopterus nephrolepidis, Davis, 1909, and some aberrant forms of this species, which forms, by the way, are not at all rare. Figure 11 shows the typical fore-wing of this insect ; here the radial sector is quite distinct. In figure 12 this is seen to be coalesced with the media in such a way as to produce a closed cell comparable to the closed cell in Pentalonia. Figure 13 is one of another aberration in which the media does not exhibit its typical forked character. : ‘In 1915 Del Guercio (Redia, vii, p. 463) erected the genus Fullawayella for Macro- syphum kirkaldyz, Fullaway, 1909. This insect is closely related to violae, Perg., but differs in the dilation of the cornicles and other details. In describing the species Fullaway (Ann. Rep. Haw. Agr. Exp. Stn. 1909, p. 22) stated that the venation is normal, that is for Macrosiphum, but he figured what evidently is an aberration similar to those found in other and related species (fig. 15). It seems evident that violae, Perg., and kirkaldyi, Fullaway, belong to the same genus, and it is equally evident that it is not Rhopalosiphum, the type of which is nymphaeae, Linn. Essig (Pomona College Jl. Ent. iii, p. 541) has stated that he believes kirkaldyi and 46 A. C. BAKER. nephrolepidis to be identical and treats the species as a native of the Hawaiian Islands imported into the United States. While this view may be correct, the figures given by Fullaway do not agree, as far as the cornicles are concerned, with any specimens of nephrolepidis. They do agree, however, in being swollen somewhat like those of Micromyzus nigrum, V. d. Goot, 1916. We have been unable to study the type of the species, and since Essig’s view is a surmise, we consider korkaldyi to be distinct from nephrolepidis as indicated by Fullaway’s figures.. The genus Micromyzus was erected by Van der Goot for his new form, nigrum (fig. 16) (Cont. 4 la Fauna des Indes Neérland. 1, fasc. 3, p.52, 1916). M. nigrum is undoubtedly closely related to kurkaldyz, if Fullaway’s figures are correct, and is congeneric with it. Mzcromyzus must therefore fall for Fullawayella. One other genus which evidently belongs to this group has been described (Patch, Ent. News, xx, p. 338, 1909). This is Microparsus, with variabilis Patch (fig. 14) as type, a form which was first figured and described by Sanborn but not named by him. This species, hke the others in the group, shows aberrations. The different genera and species may be separated as follows :— 1—Forewing with a constant closed cell Pentalonia. a—vVeins heavily bordered .. a'—-Veins not heavily bordered 1'—Forewing typically without such cell 2——Hind wing with one vein only a—One species only . We 2'-_Hind wing typically with more Shar one vein nogronervosa, Cql. caladu, V. d. Goot. 2. Microparsus. varvabilis, Patch. 3. 3—Cornicles cylindrical Idiopterus. a—OQne species only .. nephrolepidis, Davis. 3'—Cornicles more or less swollen Fullawayella. a—Cornicles broadly swollen 4g ie en? Ok a+—Cornicles slightly swollen - Be 6—Cornicles very broadly swollen and distinctly” reticulate near distal extremity b'—Cornicles moderately swollen and not reticu- late near distal extremity vrolae (Perg.) c—Segment iv of antenna without sensoria kirkaldys (Fullaway). eo Semment i iv of antenna with 3-5 sensoria .. nigrum (V. d. Goot). Of the species in the key only two, violae, Perg., and nephrolepidis, Davis, have been studied with a view toward determining the commonly present aberrations. If the opportunity to study other species in this way should arise, it is not improbable that other names would sink in synonomy. ‘This brings up again the question of the advisability of retaining a name, in a lower category than a species, for an aber- ration or of giving names to such forms if undescribed. In some groups of the APHIDIDAE these forms are remarkably abundant and their naming would place in the literature an almost endless series of designations. If, however, certain of these forms were found constantly 1 in certain regions (as for example if the violae of Theobald, as he believes, is a somewhat fixed African form) it would seem advisable for the sake of convenience to use some designation for them. tulipaella (Theo.) Ce me eee ee Ge 7 2 i is ema ee ae Bioysanlh 7 ta Se aie J a 28 oe ae. , i aes aa ‘ as yi ph er a8 um ne ear Gui aveR mIONS cues, Ai 2 e) 2) & Ay oe « — i uF : & < ts na i oy 5 ‘ 1 Ve ‘i ‘. my th Noe : mi y 7 F ’ Tot ° ith. - * . 4 j ~ : , ; ¢ hut « a , bi ‘ 45 . ies r

Rea oe ee Ag ee re er SW, JENA, IRESBARGHE Wl, XX. JPA i, Ieee JIU. Fig 1. Mopani forest in full leaf in the Zambesi Valley, April 1914. Fig. 2. Mopani forest in leafless condition, Wankie’s District, near Gwaai River, October 1918. es > Aden’ ied Weer ore =f ae Whee ake A005) BuLvt. Ent. RESEARCH. VoL. X. Part lI. Peas IW, Fig. 1. Large numbers of pups of Glossina morsitans were found in a hollow in the trunk of this tree; Sipane Vlei, Southern Rhodesia, November 1914. Fig. 2. Enlarged view of the hollow in which the G/ossina morsitans pupee were found. “7 = es ; ; E . Ze "7 e G > = R me > = 4 ¥ : J ae = = [pa 1 Gore a : Pi = a HA ah = Es 5 ee rt 3 ys j 1 ri 3 K = a ’ Ea a i j t r J : te hs ; e \ E J = a t e 5 us i ‘ oo a =} Eee = iy ; pit ah - a 98) ap en SP ae i os | ier i. i es aga es 5 COMPILED & CRAWN IN TH S8Auw ane a a eh Vy ‘ hr t pid | 4" Butt. Ent. Research. Vo. X., Part 1. Map I. Part inhabited by tsetse immedietely after the rinderpest. Limits of tsetse in 1904... Limits of tsetse in 1907._..__ Limits of tsetse in 1910, _ Limits of tsetse in 1913. Limits of tsetse in 1918... NORTHERN RHODESIA Bale & Danielsson, Map showing the increase in recent years of GLOSSINA MORSITANS in the Sebungwe District. Southern Rhodesia. Mae ax Wa ty ta ine ow hinge 2 : Laden hs BULL. ENT. MESEAnKN INVYANGA {INYANGA COMPILED & ORAWN IN TH SAL! Burt. Ent. Researcn. Vor. X.. Parr |, MAP OF SOUTMERN RHODESIA ISIS Showing belts of Glossina morsitans. NORTHERN] RHODESIA CTDRIA FAL UMVU AS, ILI aNzif COMPILED & DRAWN INTHE SURVEYOM GENERALS OFFICE BALIEBURY RHODESIA feted il; ee: Pia Te cy ae ng i ia tg few! ’ ‘ ) e > ? o Pe iar " - 2 Lite — 4 , “ ‘ / . Fi ae ri Le — r ; AD, - > a Sa > y , wx) - > I y Fut f i Vy. 2m A és Yim on J ‘ Y. 1 m = ji oF += ae ie - * 4 ; y r : 3 7 ’ . ‘ > 4 t > ‘ : 5 . me ‘ 4 i f M $ 14 Fi ; ey : 4 sex : ‘ 4 Bs ; : = ? ~ > 7 5 P * : Be ¢ \ a > -s o “a an ay www dp vane a

aS tog iets pepo i Po ‘ Sa teidap hr Eh 5 sire a cl oe eileen r { Silda ih apes on i i | = re se) eT * 2 7 “ cap eae ih ; i we : : kK: feat s he 5 oe eer ens . ie Satie wh ; 4 aac vi ras ns * tk een : fs . iS 4 of ' *« . it, ——_ wy sy Aas etx be he . de Pa Tel ae Ce ets Be el ee Aven aU Butt. Ent. RESEA - ee : ri ‘ i ; ; i i) i ae : ' ) ie } . Ms i 2 } iy % j Vg, : tt 4 Ce aoe a a if } i i } j Sy fl kK AS \ i { i) 1 : fs Ji fall er ‘ J Yi 1 ' , on ue ee me t Ny 7 , i Butt, Ent. RESEARCH. VoL. X., Part I. Map III. Dreadnought Scale of Miles. 2 & G G&G ee ee eee eeeerionn Gatooma ¥ Forest more or less untouched in 19138. ail) Forest cleared by wood contractors between 1913 and 1915. Forest being cut at the end of 1915. THE SURI-SURI FLY BELT IN THE HARTLEY DISTRICT, SOUTHERN RHODESIA. a aie echaptaee eae es ees Mees seein seg oem deren eR ee ands 8 a ¢ at pe Fi ty %, 91 ON THE OCCURRENCE OF STEGOMYIA FASCIATA IN A HOLE IN A BEECH TREE IN EPPING FOREST. By Captain M. E. MacGrecor, R.A.M.C., Officer in charge of the Entomological Laboratory, Sandwich, Kent. For the last few months I have been using a beech tree-hole in Epping Forest as a source of supply of Anopheles plumbeus ; and it has been our custom to collect larvae from the hole and transfer them to the laboratory at Sandwich, where the development is continued under artificial conditions. In this way we have been able to obtain large numbers of Anopheles plumbeus and Ochlerotatus geniculatus, together with the recently discovered Orthopodomyia albionensis * as an associate. Not long after finding the Orthopodomyia, I was surprised to find yet another species from the same tree-hole, two male Stegomyia fasciata emerging from the tank containing the mixed larvae from Epping Forest. The specimens were of normal size, and we now have them preserved in our collection here. As is well known, Stegomyia fasciata has been brought to England for experimental purposes, and has been kept breeding in captivity by several workers for some years. It is therefore probably accounted for in nature in England by stray individuals which have accidentally escaped from the laboratories. It is, however, very surprising that the first recorded specimens to be found under natural conditions in England should have occurred in a forest to a large extent removed from human habitation, since Stegomyza fasciata abroad is essentially a domestic mosquito. The insect may therefore be indigenous. The larvae were collected, and bred out in the laboratory, under my personal supervision, so that I can vouch for the absence of any error in the observation. * Journal of Royal Army Medical Corps, Nov., 1919. if rae ee || i ,' Wise erties v i 93 COLLECTIONS RECEIVED. The following collections were received by the Imperial Bureau of Entomology between Ist January and 3lst March, 1919, and the thanks of the Managing Com- mittee are tendered to the contributors for their kind assistance :— Dr. W. M. Aders, Government Economic Biologist :—-23 Diptera and 2 Lepidoptera ; from Zanzibar. Mr. T. J. Anderson :—9 Diptera, 137 Chalcids, 267 other Hymenoptera, 3 Lepidop- tera, 27 Coleoptera, a number of Coccidae, and 12 other Rhynchota ; from British Kast Africa. Mr. B. N. Bandyopadhyay :—11 Culicidae, 1 Tabanid, and 1 Glossina ; from British Hast Africa. Lieut. P. J. Barraud :—315 Culicidae, 131 Culicid larvae, 17 Culicid preparations, 44 other Diptera, 36 Slides of Siphonaptera, 5 Hymenoptera, 11 Coleoptera, 2 Coleop- terous larvae, 29 Lepidoptera, 10 Orthoptera, 26 Odonata, 5 Odonate nymphs, and 2 Mites ; from Mesopotamia. Mr. G. KE. Bodkin, Government Economic Biologist :—3 Diptera, 49 Hymenoptera, 18 Coleoptera, about 30 Mallophaga, 2 species of Coccidae, 14 other Rhynchota, 8 Orthoptera, 14 Odonata, and 3 Ticks ; from British Guiana. Capt. P. A. Buxton :—1 Simulium, 14 Tabanidae, 29 other Diptera, 9 Coleoptera, and 23 Rhynchota ; from Mesopotamia. Dr. G. D. H. Carpenter :—9 Diptera, 1 Chalcid, 64 other Hymenoptera, a number of Coleoptera, 2 Planipennia, 2 Lepidoptera, 4 Cimicid bugs, 34 other Rhynchota, and 26 Orthoptera ; from Portuguese East Africa. Mr. W. C. Crawley :—81 Coleoptera ; from Hampshire. Mr. J. Sydney Dash :—5 Lepidoptera, 1 Hymenopteron, 31 Coleontees and. 6 Rhynchota ; from Guadeloupe. Mr. D. d’Emmerez de Charmoy, Government Entomologist :—17 Opiliones ; from Mauritius. Mr. P. R. Dupont, Curator of the Botanic Station :—12 species of Coccidae and a number of fungi; from Seychelles. Mr. C. C. Gowdey, Government Entomologist :—2 Tabanus, 4 Hippoboscidae, 37 other Diptera, 10 Chalcids bred from Coccidae, 205 Hymenoptera, 34 Coleoptera, 6 Mallophaga, about 100 Aphids, 18 species of Coccidae, 30 other Rhynchota, 3 Orthoptera, 3 Ticks, and 1 Scorpion ; from Uganda. Mr. H. Hargreaves :—212 Culicidae, 1 Haematopota, 47 other Diptera, 2 Hymenop- tera, 19 Coleoptera, about 150 Anoplura, 27 Rhynchota, and 2 Orthoptera; from Taranto,. Italy. Mr. J. C. Hutson :—86 Coleoptera ; from Ceylon. Imperial Department of Agriculture :—9 Hymenoptera, and 7 Coleoptera ; from the West Indies. Mr. Nigel K. Jardine :—278 Diptera, 38 Lepidoptera, 34 Hymenoptera, 144 Coleop- tera, 29 Rhynchota, 10 Orthoptera, 9 Odonata, and 2 Arachnida ; from Ceylon. 94. COLLECTIONS RECEIVED. ; Capt. Malcolm E. MacGregor :—4 Diptera, 3 Hymenoptera, 16 Mallophaga, about 80 Ticks, and a number of intestinal worms ; from Portuguese Hast Africa. Mr. K. D. Shroff, Assistant Entomologist :—a collection of Crustacea, Mollusca and worms ; from Burma. Mr. R. Veitch :—8 Diptera, 16 Hymenoptera, 3 Thysanoptera, 42 Coleoptera, 14 Lepidoptera, 29 Orthoptera, 7 Rhynchota, and 6 Spiders ; from Fiji and Queens- land. Capt. Jas. Waterston, R.A.M.C.:—6 Phlebotomus, 13 Siphonaptera, and a large number of lice; from Macedonia. Dr. W. G. Watt :—100 Culicidae and about 50 Culicid larvae; from the Gold Coast. i \ & { f y t - { . \ z. : » re , j % Ww % THE COCCIDAE OF SOUTH AFRICA —IV.* By Cuas. K. Bratn, M.8c., M.A., Division of Entomology, Pretoria, South Africa. (Puates V—XII.) CONTENTS. PAGE. Genus Chionaspis (continued) .. He be ap - wn . 95 ,, Lepidosaphes ai ig ae wig ot ih * a 105 oy . Lschnaspis? 5: sah as ou Ae si ae ist 107 Subfamily ASTRROLECANINAE ie st s ‘ei i. A o4 108 Genus Asterolecanvum , 0) iy cs , a He an 108 », Lecaniodiasprs i i! bt u. iis 4 ay, KK 115 » Cerococcus .. J; sf LM a bi ve ng oe 119 Subfamily TACHARDIINAE .. mM ms ¥, ee ue Mi He 122 Genus Tachardia .. Ws at HG a4 8 a2 ail a 122 Subfamily ? .. i ee a if - " a « ‘2: 126 Genus Halimococcus aN e aT un KS ae i a 126 Baccacoccus .. Bi a a ny ne af $f re 127 2) 147. Chionaspis scutiae, sp. n. (Plate v, fig 176). Scale of adult 2 about 2 mm. long, moderately broadened behind, convex, usually curved, white, not glossy, but with =: conspicuous growth-lines; exuviae orange- brown to dark brown. The colour of the second exuviae is only slightly obscured by a faint secretionary covering. Puparium of g comparatively large, white, non-carinate, with yellow or orange exuviae. Adult 2, when mounted, regular in outline, about 1°3 mm. long, elongate, widest behind the middle, hyaline, except the median lobes which are very dense, and the middle part of the pygidium which is yellowish. The segmentation of the body is unusually distinct but is not strongly marked at the margins, nor are the abdominal segments produced. The pygidial margin is simple. There is only one pair of lobes present, which are short and broad, densely chitinised, brown, with the rounded outer margin once notched. Plates are apparently absent, but there are two stout spines on each side. Pygidium as illustrated (fig. 176). Circumgenital glands in 5 groups: 94 GAO 0219 12299 iid 3299 Habitat: “ Wacht-een-beetje” (Scutia indica, Brohn), eee C.P. ; collected by C. P. Lounsbury, September 1907. Collection No.: 182. *For Part II, see Bull. Ent. Res. ix, p. 107; Part III, Bull. Ent. Res’ ix, p. 197. (C605) P4/140. 1,000. 1.20. B.& F.Ltd. G.11. A 96 CHAS, K. BRAIN. 148. Chionaspis (Pinnaspis) chionaspitiformis (Newst.) (Plate v, fig. 175). ’Draspis chionasprformis, Newst., Bull. Ent. Res. i, p. 198, 1910. Hemachionaspis chionaspitiformis, Lindinger, Jahrb. Hamb. Wiss. Anst. xxvii, Beih, 3,191: Scale of adult 2 about 2 mm. long, white, appearing very loosely constructed and soft in texture, moderately broadened behind, with yellowish or brownish exuviae. The second exuviae are covered with a thin layer of secretion. Puparium of ¢ elongate, non-carinate, with yellow exuviae. Some specimens collected by Claude Fuller on twigs of a native tree at Busi, Portuguese East Africa, appear as matted, dark brown to blackish scales, with brown exuviae. This, I think, is due to an accumulation of sooty fungus, as the insects agree in all other respects with those on African mahogany from Rhodesia. Adult 2 mounted, about 14 mm. long and 0°7 mm. broad at the widest point, which is situated at a considerable distance beyond the middle. The insect is narrow, rounded in front, gradually broadening to the free abdominal segments, from which it suddenly narrows to the hind extremity, which is pointed. The whole body is hyaline except the mouth-parts and median portion of the pygidium, which are yellow. The median lobes are densely chitinised and brown. The antennal tubercles are rather large with one very long flagellum and a few short spurs. Parastigmatic glands present, 5 to 8 at each anterior spiracle. The pygidium is characterized by a single pair of lobes, which are broader than long and have their inner margins in close contact or fused ; their distal end is evenly rounded and the outer margin deeply notched. From the outer edge of each lobe a short spine arises. The plates are longer and more curved than in C. cassiae—as illustrated in fig. 175. Circumgenital glands in 5 groups, as in cassie, or with the posterior lateral glands slightly more numerous. The most striking differences between this species and C. cassiae are :—that the scale is white, without the transverse brownish ridges, and the pygidium appears more acute, with the plates longer and more curved. Habitat: On African mahogany, Zomba, Nyasaland ; collected by Ross-Townsend, December 1908 (Cape Collection 2100). On native tree, Busi, Portuguese Hast Africa ; collected by C. Fuller, May 1915. Collection No. : 322, 322a. 149. Ghionaspis simplex, Green, var. (Plate vi, fig. 178). Chionaspis simplex, Green, Cocc. Ceylon, ii, p. 160, 1899. Scale of adult ? large, 3 mm. long, and 1°5 mm. broad, broadest about middle, white, dull, without distinct growth-lines. First exuviae colourless to yellowish ; second exuviae large, roundly convex, glassy, usually -+ buffin colour, with a depressed median area bordered on each side by a prominent beaded ridge. In living material the second exuviae are apparently covered by a thin, dull white layer of secretion, which is easily flaked off. In the majority of dry specimens this layer and the first exuviae are missing. Puparium of ¢ not observed. Adult 2° contain many well developed embryos with 5-jointed antennae. THE COCCIDAE OF SOUTH AFRICA. Sif Adult 9, when mounted, elongate, narrowly rounded in front, broadest immediately behind the middle, and somewhat broadly pointed behind, hyaline, with the pygidium slightly denser ; abdominal segments well marked and flatly rounded at the margin. Pygidium without lobes or plates, coarsely corrugate, with conspicuous spines and dorsal glands as illustrated (fig. 178). Circumgenital glands in 5 groups, which are often almost contiguous in a wide bow : 18—25 24—-32 24—32 48—60 48—60 Remarks.—This species is very much like C. semplex Green, and is in all probability the variety he mentions from Mauritius ; it has but 5 groups of circumgenital glands instead of 7 as in C. simplex. - Habitat: On bamboo, Durban, Natal ; collected by A. Kelly, May 1915. Collection No. : 175. 150. Chionaspis caffra, sp. n. (Plate v, fig. 174). Scale of adult 9 about 2 mm. long, narrow, but gradually widening to near the posterior end, where it is flattened and broadly rounded, straight or slightly curved. Exuviae brown ; second exuviae covered. G puparium small, white, with orange exuviae ; generally non-carinate, sometimes faintly tri-carinate. Adult 9°, mounted, about 1:4 mm. long, narrow in front, then gradually widening to some distance behind the middle ; anterior part of body and median parts of the anterior abdominal segments considerably chitinised ; abdominal segments not rounded, nor produced at the margin. Antennal tubercles small, each with one very stout spine. L, close together, inner margin slanting slightly outward, crenulate. L, small, roundly conical. Circumgenital glands in 5 groups: 5—T 8—14 8—l4 19—25 19—25 Habitat : On Acacia sp., probably A. caffra, The Thorns, Pretoria; collected by C. P. Lounsbury, September 1915. Collection No.: 150. 151. Chionaspis ambiguus, sp. n. (Plate v, fig. 177). Scale of adult 9 elongate, mytilaspiform, narrow in front, broadly rounded behind, usually + curved, arched, 2°3 mm. long, somewhat covered by outer bark tissues. Colour of scale pale to dark brown, with a greyish surface covering ; exuviae yellow. ¢$ puparium not observed. Body of adult 9 elongate, broadest just behind middle, anterior end tapering slightly and broadly rounded in front, posterior end tapering abruptly to the two densely chitinous median lobes. These, together with the chitinous thickenings extending into the pygidium, and the mouth-parts, are yellowish ; remainder of body hyaline. Free abdominal segments (4) slightly rounded, not much produced. Antennal tubercles moderately large, with two stout spines. Parastigmatic glands 0. (C605) a2 98 CHAS. K. BRAIN. Pygidium (fig. 177) with one pair of lobes, which are close together on their inner surface, their outer edges being crenulate. Between the median lobes there are two large spines but no plates. At the outer margins there is also a large spine, then a gland from which arises a small sharply pointed plate, then a very small rudimentary lobe, which is pointed and chitinous and in some cases seems to be composed of two lobules of which the outer is the smaller. These are followed by a long simple dagger- shaped plate, then a large gland opening and another plate. Beyond this the margin appears thickened lke a minute rudimentary lobe, beyond which there are two additional pairs of plates. There are a few plates on the margin of the first abdominal segment from the pygidium in addition to the glands. Extending into the pygidium are three pairs of chitinous thickenings. Circumgenital glands in 5 groups, of which the median and anterior laterals are often almost united into a bow. 7—l1 10—14 10—14 6— 9 6—9 Remarks. A large percentage of the @ scales showed the circular exit holes of some Hymenopterous parasite. Habitat : On twigs of lilac, Fort Beaufort, C.P.; collected by C.P. Lounsbury, June 1913. Collection No. : 263. 152. Chionaspis leucadendri, sp. n. (Plate v, figs. 166, 171). Scale of adult 9 about 2°6 mm. long, white, smooth, slightly glossy, not very convex, long and narrow, somewhat widened and flattened behind. Exuviae brownish ; second exuviae covered by a very thin layer of secretion. 2 scales chiefly on stems ; ¢ puparia most common on leaves. g puparium comparatively large, white, flat, shghtly roughened, not carinated, with pale exuviae, which are sometimes + greenish yellow in colour. Adult 9, mounted, about 1:4 mm. long, narrow in front, but quickly widening to about middle, whence it gradually narrows again to the median lobes. Abdominal segments well indicated but not prominently produced. Body moderately chitinised, the mouth-parts, lobes, and margin of pygidium denser. The four abdominal seg- ments adjoining the pygidium have their margins with a few gland-pores and a number of short, stout, very acute spine-glands. These, seen in optical section, appear like rose-spines, but the points are often long, slender and curved. They occur chiefly on the margin, but extend in scattered formation some distance within the edge. P, small, spine-like. lL, rounded, striate, arising from the thickened margin of the pygidium. P,—6 long, dagger-shaped. Circumgenital glands in 5 groups, which are almost contiguous, in the form of a horse-shoe : 4—7 11—18 11—18 26—32 26—32 Hatitat : On silver-leaf tree (Leucadendron argenteum), National Botanic Gardens, Kirstenbosch, Capetown; collected by the late Professor H. H. W. Pearson, August 1914. Collection No. : 145. THE COCCIDAE OF SOUTH AFRICA. 99 153. Chionaspis (Phenacapsis) lounsburyi, Cooley (Plate v, fig. 169). Chionaspis lounsburyi, Cooley, Can. Ent. xx, p. 87, 1898. Phenacaspis lounsburyi, Fernald, Catalogue, p. 238, 1903. Adult 92 scale about 3 mm. long, white, glossy and pearly, sides almost parallel when at the margin of a leaf, but usually broadened behind the middle when on the blade ; evenly rounded behind, often with faint transverse ridges. Ventral scale absent or represented only by the incurved margins of the dorsal scale and an extremely delicate layer which remains on the leaf. Hxuviae orange-yellow or orange- brown; first exuviae paler, second covered by a pearly white layer of secretion. S puparium about 1 mm. long, similar to that of 2 but smaller, non-carinated. Body of the adult @ elongate, all colourless and hyaline, except the mouth-parts and lobes which are faintly yellow. The free abdominal segments are not produced: at the margin, but are slightly rounded. oe L, low, broad, and very widely divergent, usuallv appearing as though they pro- jected from behind the thickened median area of the pygidium. J, composed of two small lobules which are distinctly separated ; the inner lobule is the larger. Lobes distinctly striate. Plates simple, thin. | The antennae are placed a long way back near the mouth-parts and consist of minute tubercles each with one long thin curved hair. Parastigmatic glands present, usually 3 to 5 at anterior spiracles. Circumgenital glands in 5 groups : 4—T WI—15 11—15 20—26 20—26 Habitat: On Rhus thunbergi, Ceres, C.P.; collected by C. P. Lounsbury, , 1896. On a native Proteaceous plant, Wolseley, C.P.; collected by C. P. Lounsbury. June 1915. Collection Nos. : 166, 170. 153a. Chionaspis (Phenacaspis) lounsburyi ekebergiae, var. n. (Plate v, fig. 168). © scale similar to that of C. lounsburyi, but readily distinguished by the following characters :— Parastigmatic glands present, 5 or 6 glands in a close crescent. Pygidial margin as figured (fig. 168). Circumgenital glands more numerous, ¢.g., 6—9 11—20 11—20 29—46 29—46 Plates more distinct and last row of dorsal glands more numerous than in C’. lounsburyt, usually 9 : 4. Habitat : On Ekebergia sp., Durban ; collected by C. Fuller, July 1915. Collection No. : B. 186. 100 CHAS. K. BRAIN. 154. Ghionaspis (Phenacaspis) natalensis, Ckll. (Plate v, fig. 165). Phenacaspis natalensis, Ckll., Ann. Mag. N.H. (7) ix, p, 25, 1902. Phenacaspis natalensis, Fernald, Catalogue, p. 238, 1903. Professor Cockerell’s description is as follows :— “2 scale white, about 3 mm. long, pyriform ; exuviae pale orange-brown. “Five groups of circumgenital glands; median of 10, anterior laterals 22-26, posterior laterals 19-26. Anal and genital apertures opposite. Median lobes large, widely diverging, broader than long, the long inner margin strongly serrulate ; their bases well apart, the space occupied by the usual pair of short spines. Second lobe represented by three elongated and rounded lobules, the first of which is largest and bears a spine. Third lobe represented by a very long narrow lobule bearing a spine followed by a broad and much shorter lobule, and then a very broad serrulate lobule, having its outer slope much the longest. The fourth lobe is represented by a triangu- lar lobule bearing a spine and two slight swellings of the margin, too slight to be called lobules. “ 2 scale feebly tricarinate or barely keeled at ail.” Habitat: On mango, Durban, Natal ; collected by C. Fuller, 1901. On palm, Durban ; collected by A. Kelly, July 1915. Collection Nos.: 171, 171a. 155. Chionaspis (Dinaspis) imbricata, sp.n. (Plate vi, fig. 179). in material seen by the writer the scales invariably occupy cracks in the bark on the stems of the host-plant. Scale of adult 2 small, elongate, almost parallel-sided, white, with orange to brown eXUV1ae, Puparium of 3 similar but smaller. Adult 9, when cleared and mounted, small, averaging 0°75 mm. long and 0:48 mm. broad. Theanterior endis broadly rounded ; the sides of the body are almost parallel to a little more than half the length, when they gradually taper to the pointed pygidium. The anterior half of the body and the pygidial area are more densely chitinised than are the free abdominal segments and appear yellowish. The chitin on these parts is thickened in minute ridges, giving the impression of a finger-print, a character only found in a few Coccids. It is impossible to describe the pygidial margin as having definite lobes, and plates are absent. The chitin is here folded so as to form numerous rounded more or less imbricated prominences, some of which reach, or extend beyond, the margin, which thus appears festooned. There are a few spines of moderate length at intervals around the margin. The anal opening is set well back from the margin and around it are a number of conspicuous thin spots in the chitin, appearing as perforations. In the second stage 9 there are about three simple plates on each side of the pygidium. Antennae set well forward, tubercles moderately prominent with one or two curved setae. Parastigmatic glands 0. Circumgenital glands O. Habitat : On stems of Huclea natalensis, Point Road, Durban ; collected by C. P. v. d. Merwe, July 1916. Collection No. : 157. THE COCCIDAE OF SOUTH AFRICA. 101 156. Chionaspis (Dinaspis) diosmae, sp.n. (Plate v, fig. 167). Insects -- clustered on leaves of host-plant, especially on the upper surface. Adult 9 scale about 2-2 mm. long, comparatively broad, moderately convex, silky iN appearance, with conspicuous growth lines. First exuviae yellow ; second exuviae brownish, covered by a dense layer of secretion similar to remainder of scale. The layer covering the second exuviae is very easily removed and carries the small first exuviae with it. Owing to this fact many of the specimens appear to have brownish, shining pellicles. Adult © viviparous ; mounted specimens contain many well-developed larvae. Adult 9, mounted, about 1:6 mm. long and 0°8 mm. broad ; widest about middle and narrowing to each end. Body moderately chitinised. The pygidial margin is quite different from that of any other species of Dinaspis known to me. The median notch is angular, with narrow straight thickened margins, below which the median lobes project. Owing to this fact the appearance of the lobes varies greatly according to how the specimens are mounted, L, in some cases appearing quite distinctly, in others being almost invisible. There are apparently no plates. Pygidium as illustrated (fig. 167). Circumgenital glands O. Habitat: On buchu (Diosma crenata), Wellington, C. P.; collected by C. P. v. d. Merwe, November 1904 (Cape No. : 1554). Collection No. : 147. 157. Ghionaspis (Dinaspis) lounsburyi (Leonardi) (Plate v, fig. 172). Dinaspis lounsbury2, Leon., Bol. R. Sc. Agr. Portici, p. 216, 1914. Chonaspis capensis, Newst., Bull. Ent. Res. vii, p. 378, 1917. Scale of adult 9 about 2mm. long, elongate, generally straight, but varying in shape according to the position on the plant, condition of crowding, etc. On straight stems the sides of the scale are + parallel from the end of the second exuviae to beyond the middle, where it becomes slightly broadened with the hind margin broadly rounded. Other specimens situated in the angles of thorns, etc., are much shorter and broader, sometimes almost as broad as long. Colour of scale white ; first exuviae bronze to bronze-brown ; second exuviae brown, slightly pointed behind. Puparium of § about 0°8 mm. long, white, non-carinated, with orange pellicles. Posterior margin of puparium broadly rounded, opening by the upper part splitting from the lower so that the dorsal flap is exactly like the lower one. Adult 92 viviparous ; when mounted, elongate. Anterior part suddenly narrowed and rounded in front, paler yellow ; median portion of body darker yellow, wider, with almost parallel sides ; posterior extremity broadly rounded. The free abdominal segments are broadly rounded at the margins. Antennal tubercle small, with two setae of medium length and thickness. Pygidium as illustrated (fig. 172). Circum- genital glands O. Habitat: On stems and leaves of native spiny plant (Gymnospora buaxrfolia) ; extremely common around Pretoria and also received from Umtali, Rhodesia (sent by R. Lowe Thompson, of Salisbury). Collection No.: 251. 102 CHAS. K. BRAIN. 158. Chionaspis (Dinaspis) distincta (Leonardi) (Plate v, fig. 170). Dinaspis distincta, Leonardi, Bol. Sc. R. Agr. Portici, p. 213, fig. 33, 1914. Chionaspis distorta, Newst., Bull. Ent. Res. vii, p. 377, 1917. Scale of adult 2 about 2-2 mm. long, moderately broad, convex and roughened, dull white or greyish white in colour, with exuviae dark orange-brown; second exuviae covered. , Puparium of ¢ robust, bufi-coloured and non-carinate, with yellowish or orange exuviae. Very few gj specimens are present in the large amount of material before me and in a number of cases the § puparium does not lie close to the stem in the usual manner, but seems to be attached at the anterior end and projects outward between several female scales. Adult 9 viviparous ; elongate, narrow in front, gradually broadening to behind the middle, to which point the body is highly chitinised ; posterior to this it is thin, hyaline. There are three pairs of lobes, which are more heavily chitinised than the remainder of the pygidium : L, broad, inner margins converging at the base, faintly trilobed or broadly crenulate ; median lobe small, L, and L, small, + triangular. Antennae with one long flagellum and four short stout spurs. Circumgenital glands O. Habitat : On stems of Protea hirta, Pretoria ; collected by the writer, October 1914. Professor Newstead’s reference “‘ Windersboom, Transvaal,’ should be Wonderboom, Pretoria. Collection No. : 163. 159. Chionaspis (Poliaspis) carissae (Ckll.) (Plate v, fig. 164). Poliaspis carissae, Ckll., The Entom. xxxv, p. 112, 1902 ; Fernald, Catalogue, p. 243, 1903. Scale of adult 2 about 1:8 to 2 mm. long, usually straight, widest shortly behind the second exuviae and somewhat abruptly narrowed and attenuated posteriorly ; white, glossy, with brown exuviae. Puparium of 3 white, very long, distinctly tri-carinate ; exuviae almost colourless. Specimens on “ umkavoti”’ at Durban had the 9 scales faintly marked by trans- verse ridges and the ¢ puparia with a pronounced median ridge. The body of the adult 9 is long, narrow in front, then almost parallel-sided to the free abdominal segments, with the posterior margin regularly and broadly rounded. The median portion of the body is highly chitinised and appears brownish-yellow in mounted specimens. The anterior end is paler in colour, and the free abdominal segments and the pygidium are almost colourless. Circumgenital glands in 8 groups : 3— 4 3-4 3— 4 1l—17—_s 24s W1—17 17-—27 17—27 Remarks. Professor Cockerell’s original description is as follows :— ‘“ © seale similar to that of P. cycadis, but perhaps narrower. Second skin paler, as in cycadis. “ similar to P. cycadis, but the strongly serrulate reddish-brown median lobes are wide apart, the interval being nearly as great as the breadth of a lobe; the second lobe consists of two lobules, of which the inner is the larger, and its tip projects a little THE COCCIDAE OF SOUTH AFRICA. 103 beyond the level of the tips of the median lobes ; the margin just beyond the second lobe bears two large dorsal glands, like those of the series on the next segments anteriorly ; there are only four dorsal glands in the short rows nearest the anal orifice. The middle of the body is red-brown, and strongly chitinised. Circumgenital glands in eight groups ; the posterior laterals 19, middle laterals (cephalolaterals of other genera) 11, median 5, in a transverse row, and the anterior groups characteristic of Poliaspis form a transverse series broken into three linear groups of three or four, which are widely separated. “ § scale tricarinate. “ On Carissa (? C. grandiflora, D.C.) plant, which belongs to the Apocynaceae, Durban, Natal (Fuller). In these species and P. cycadis the anterior group of glands are in transverse lines; in P. media, and the species described by Fuller from Australia, the groups are circular.” Habitat: On Carissa grandiflora (amatingula), Durban; collected by C. Fuller (part of original material). On “umkavoti” (Chaetachme aristaia), Durban ; collected by C. Fuller, October 1914. Collection Nos. : 181 and 167. 160. Chionaspis (Pinnaspis) cyanogena (CkIl.) (Plate vi, fig. 180). Hemichionaspis cyanogena, Ckll., The Entom. xxxiv, p. 226, 1901 ; Fernald, Catalogue, p. 240, 1903. Scale of adult 2 about 2 mm. long, narrow, moderately convex, white, but appear- ing slightly greyish in massed specimens. Exuviae pale yellow to orange ; second exuviae covered by a thin secretionary layer which is easily flaked off and therefore often missing. Puparium of ¢ white, non-carinate, or with a distinct median ridge. HExuviae pale yellowish. | The body of the Qis elongate, often rather narrow in front and suddenly broadening to about the middle and suddenly narrowing again to the pointed pygidium. The margins of the free abdominal segments are not conspicuously produced, but are supplied with numerous gland-openings and a few short, pointed projections. The pygidium has the median lobes moderately produced, their inner margins straight and in close contact, their outer extremities sloping gently backward and twice notched, forming three sub-equal rounded crenulations. The median lobes and the posterior median part of the pygidial area are yellow, and the yellowish coloured spots on the margin may indicate three other pairs of rudimentary lobes. At these points gland- openings are apparent, and small rounded projections may be seen on the margin ; the plates area httle longer than the median lobes and are distributed as shown in fig. 180. Antennal tubercles + globular, with one curved flagellum. Circumgenital glands in 5 large groups : 12—18 15—19 15—19 13—17 13-17 Habitat: On Alternanthera sessilis, R. Br., Durban; collected by C. Fuller. On Alternanthera sessilis, R. Br., Scottsburg, and on native weed, Natal coast; collected by C. Fuller, July 1915. Collection Nos.: 177 and 177a. 104 CHAS. K. BRAIN. 161. Chionaspis (Pinnaspis) aspidistrae, Sign. (Plate vi, fig. 181). Chionaspis aspidistrae, Sign., Ann. Soc. Ent. Fr. (4) 1x, p. 443, 1869. Choonaspis brasihensis, Sign., Ann. Soc. Ent. Fr. (4) 1x, p. 444, 1869. Choonaspis latus, Ckll., Psyche, vii, Suppl. 1, p. 21, 1896. Chionaspis aspidistrae, Green, Cocc. Ceylon, ii, p. 110, 1899. Hemichionaspis aspidistrae, Cooley, Spec. Bull. Mass. Exp. Sta. p. 45, 1899. Chionaspis aspidistrae, Newst., Mon. Brit. Coce. i, p. 187, 1901. Hemachionaspis aspidistrae, Fernald, Catalogue, p. 239, 1903. Pinnaspis aspidrstrae, Lindinger, Die Schildlause, p. 79, 1912. Scale of adult © variable, largest specimens about 2°5 mm. long, more or less pear- shaped, thin, usually semi-transparent, whitish or yellowish to reddish brown. Kixuviae colourless or faintly yellow. : Adult 9 elongate, broadest across the abdominal segments, which have the margins produced into rounded conical processes in the young adult stage. Pygidium as illustrated (fig. 181). Circumgenital glands in 5 groups: 5—15 15—23 15—23 17—23 17—23 Habitat: On ferns, Aspidistra, etc., Cape, Natal, and Transvaal. Collection No. : 176. 162. Chionaspis (Pinnaspis) proxima (Leonardi). Hemichionaspis proxima, Leon., Bol. R. Sc. Agr. Portici, p. 193, 1914. A translation of Professor Leonardi’s description is as follows :— “Female. Body elongate, with the extremity rounded and having the greatest width about the middle. The abdominal segments narrower than the cephalothorax and projecting laterally in well defined lobes, which have their free margins rounded. The lobes of the last abdominal segments have 2-3 long and robust hairs along the free margin. The buccal apparatus with the maxillary-mandibular bristles radiating and extending beyond the posterior extremity of the body. “Antennae tuberculiform and surmounted with a long flagellum. Posterior stigmata without glands. “ Pygidium with two pairs of lobes, of which the middle pair, which is deeply coloured, has the inner margins approximated and the outer more deeply incised. Those of the second pair are separated from the first by a robust hair, and by the mouth of a large wax-gland. They are very small, slender, somewhat hyaline, with the free margin rounded and entire. On the outer side of each of the second lobes there is first a hair, then the mouths of two large wax-glands, and at a certain distance - from these another hair followed by the mouths of two other wax-glands, and lastly near the pre-anal segment a fourth hair. ** The circumgenital glands are in five groups, e.7., 10 8 12 21 1 17 19 gl 21 17 1g 16 16 22 23 THE COCCIDAE OF SOUTH AFRICA. 105 “The sexual aperture is situated in the middle of four groups of lateral glands. The anal aperture is almost the same height as the sexual opening, but rather more towards the pre-anal segments. “ Colour of the body in specimens treated with acetic acid, yellowish, except the pygidium which is yellow-ochraceous. “Female scale. Oval, flatly convex, with the secreted portions scant, whitish grey, and the exuviae at the apex and narrower than the scale, and of an ochroleucous colour. “* Dimensions of the scale : Length of scale a ‘e Me a e 1,280u Width of scale .. a an a bee ee 800u Length of larval exuviae if ms o 350u Width of larval exuviae ig Me me im 220u Length of nymphal exuviae .. Fi me me 750u Width of nymphal exuviae .. 470u “Male scale. Elongate, sides parallel, terete or sliehtly sarzed, rounded behind, with the dorsum deeply grooved, and the resulting ridges well marked. Larval exuviae vellowish, small and situated at one end. Waxy scale white. Size of scale : length 960u ; width 320u. “ Habitat: Collected at Thies (Senegal) and at Mamu on mango, at Konakry on Anona, at Hann (Dakar) on Calotropis procera and on an undetermined plant ; at Dodowa, at Lagos, at Cotonou (Dahomey), at Quifangando (Angola) and at Pretoria on undetermined plants. “ Notes: This species is near Hem. orlandz, Leon., from which it can readily be distinguished by the following characters. It has glands at the anterior stigmata and the glands around the sexual aperture are always more numerous; there are present, although not very apparent, a second pair of lobes, and further, the last four abdominal segments are supplied with hairs, not only the last two as is the case in Hem. orlandi.” This species is not represented in this collection and although a thorough search has been made around Pretoria it has not been found by the writer. Habitat : On an undetermined plant, Pretoria ; collected by Professor Silvestri. Collection No. : 178. Genus Lepidosaphes, Shimer. Lepidosaphes, Shimer, Tr. Am. Ent. Soc. i, p. 373, 1868. Mytilaspis, Sign., Ann. Soc. Ent. Fr. (4) viii, p. 841, 1868 (no description). Mytulasprs, Sign., Ann. Soc. Ent. Fr. (4) x, p. 91, 1870. Scale of adult @ elongate, usually narrowed in front and gradually broadening behind. The colour is very variable, but often orange-brown to dark brown. The exuviae are terminal, with the second exuviae covered. The female scales are usually + curved. Ventral scale variable, similar to those of Chionaspis spp. The ¢ puparium is similar to that of the 2, but smaller and narrower; the larval exuviae are terminal. The posterior part of the puparium is often separated by a thin transverse band of secretion, which acts as a hinge, allowing the extremity of the scale to be raised to permit the adult J to emerge. 106 CHAS, K. BRAIN. Adult 2 elongate, with the margins of some of the segments often produced. Circumgenital glands in 5 groups, but each group usually consisting of fewer glands than is normal in Chionaspis. 163. Lepidosaphes pinnaeformis (Bouché) Kirk. (Plate vi, fig. 183). Aspidiotus pinnaefornus, Bouché, Stett. Ent. Zeit. xu, p. 111, 1851. Coccus becku, Newm., The Entom. iv, p. 217, 1869. Aspidiotus citricola, Pack., Guide to Study of Insects, p. 527, 1869. Mytilaspis fulva, Targ., Bol. Soc. Ent. Ital. p. 131, 1872. Mytilasprs flavescens, Targ., Ann. R. Min. Agr. p. 84, 1876. Mytilaspis citricola, Comst., Rept. U. 8. Dept. Agr. p. 321, 1881. Mytulaspis citricola, Lounsbury, Rept. Ent. C. G. H. p. 67, 1896. Mytilaspis citricola, Green, Coce. Ceylon. i, p. 78, 1896. Mytilaspis pinnaeformis, Newst., Mon. Brit. Coccidae, i, p. 204, 1901. Mytilaspis beckiz, Ckll., Ent. News, xiti, p. 17, 1902. Lepidosaphes pnnaeformis, Kirkaldy, Fauna Haw. iii, 2, p. 110, 1902. Lepidosaphes pinnaefornus, Fernald, Catalogue, p. 313, 1903. Lepidosaphes beckui, Fernald, Catalogue, p. 305, 1903. Common Name: Mussel Scale. Scale of adult 9 elongate, usually + curved, narrow in front and broadened behind, with a flattened paler marginal area. Colour varying greatly with host-plant, etc., from pale greyish or yellowish brown to reddish- or olivaceous-brown. In size it may reach 3 mm. long and nearly 1 mm. in breadth. The © puparium smaller, narrower, and more delicate in texture, with sides = parallel, and colour usually a little paler. Adult 9 elongate, narrow in front, widest at the first free abdominal segment, 1.¢., behind the middle. The colour is generally whitish, or faintly yellowish with the terminal segments darker. The antennae are represented by a small tubercle with two stout curved hairs. The pygidium as illustrated (fig. 183). Circumgenital glands in 5 groups : , 5—8 9—17 9—I17 7—1l1l 7—Ill1 Habitat : On croton, Capetown, Port Elizabeth, and Uitenhage (C.P.), Johannes- burg (Tr.), and Durban (Natal). On orange, Cape Province, Natal and Transvaal. On Murraya, New Guelderland, Natal. Collection Nos. : 262—262e. 164. Lepidosaphes gloveri (Pack.) Kirk. (Plate vi, fig. 182). Coccus glovert, Packard, Guide to Study of Insects, Ed. 1, p. 527, 1869. Mytilaspis flava, Targ., Catalogue, p. 44, 1869. Mytilaspis glover, Comst., Rept. U. 8. Dept. Agr. p. 323, 1881. Mytilaspis glover, Lounsbury, Rept. Ent. C. G. H. p. 71, 1896. Mytilaspis pallida, Green, Ind. Mus. Notes, iv, p. 5, 1896. Mytilaspis glovervi var. pallida, Green, Cocc. Ceylon, i, p. 85, 1896. Leyidosaphes glovert, Kirkaldy, Fauna Haw. Is, 1, 2, p. 111, 1902. Leyidosaphes glover, Fernald, Catalogue, p. 309, 1903. Lepidosaphes gloverr, Lind., Die Schildlause, p. 106, 1912. Common Name: Long Scale. THE COCCIDAE OF SOUTH AFRICA. 107 The scale of the adult 9 is very long and narrow, 2°5 to 3 mm. long and about 0°5 mm. broad, long comma-shaped, usually reddish or greyish-brown with paler margins ; on some host-plants the colour is paler. Hxuviae terminal, yellowish. Ventral scale well developed, whitish. & puparium much smaller and more delicate than the Q scale, and paler in colour. Adult 2 elongate, sides + parallel, abdominal segments slightly wider, but without prominent lateral projections. The colour when alive is pale yellow. Pygidium as illustrated (fig. 182). Circumgenital glands in 5 groups : 4 6—8 6é—S8 4—5 4—5 Habitat: On citrus, Cape Peninsula; collected by C. P. Lounsbury. Also at Port St. Johns, Stanger, and Warmbaths. Collection No.: 261. Genus ischnaspis, Douglas. Female scale extremely long.and narrow, with the first exuviae extending beyond the anterior margin. Male puparium elongate, similar to that of the second stage female, without the central hinge of the 3 of Leprdosaphes. Adult 2 very long and narrow, posterior margin of pygidium forming a regular shallow concave depression, from which the median lobes project. Dorsal surface of pygidium with a peculiar lattice-work design. Circumrgenital glands in 3 or 5 small groups. | 165. Eschnaspis longirostris (Sign.) (Plate vi, fig. 184), Mytilaspis longirostris, Sign., Bull. Soc. Ent. Fr. (6) ii, p. xxxv, 1882. Ischnaspis filrformis, Douglas, Ent. Mo. Mag. xxiv, p. 21, 1887. Ischnaspis filiformis, Newst., Mon. Brit. Coc. i, p. 210, 1901. Common Name: Black Thread Scale. Scale of adult 2 3 to 35 mm. long, very narrow, with sides parallel ; posterior extremity slightly broader ; colour shining black, with greyish margins. Scale very convex, often showing slight indications of transverse ridges, usually straight, but where a number are crowded together the scales often make a regular bend to avoid other scales and then continue again in the original direction. Larval exuviae terminal, brownish in colour; second exuviae occupying about one-quarter of the length of adult scale, covered with shiny black secretion like the remainder of the scale. Ventral scale complete, robust, white or greyish in colour. Adult 9 very long, sides parallel, broadest at the free abdominal segments. Colour yellow. Pygidium as illustrated (fig. 184). Circumgenital glands in 3 groups: 3—4 4—6 4—6 A few large (to 4 mm. long) specimens in the Cape collection, otherwise typical, have 5 groups of glands. Three of these have: 4 5 4 108 CHAS. K. BRAIN. Habitat: On citrus, Warmbaths (Tvl.). On Chaetachme aristata (umkavoti), Durban, Natal; palms, Kast London, Durban and Pretoria; Aucuba, Hast London. On Dracaena australis, Kast London ; collected by C.P. Lounsbury, November 1906. This species is particularly abundant on citrus at Warmbaths, Transvaal. Collection Nos. : 277-2776 and 278. Subfamily AsTEROLECANIINAE. Adult females fixed to the stem of their host-plants ; sometimes causing pits in the stems and thus becoming slightly or almost entirely embedded in the outer tissues. Adults usually enclosed in a more or less horny or glassy cyst, which is most often yellow in colour and, in one genus, has a marginal fringe of glassy filaments. The legs and antennae are most often rudimentary or absent in the adult stage. Figure-8 © glands are present in one or more stages. This subfamily is represented in South Africa by four genera which may be distinguished as follows :— Test of 9 usually yellowish, + transparent, with marginal fringe .. Asterolecanium. Test of Q dense, almost ligneous, without fringe .. : .. Lecanodiraspis. Test of Q = = waxy and without fringe, 2 without piemaee spines Cerococcus. © naked, causing shallow rounded galls in stem* By et .. Amorphococcus. Genus Asterolecanium, Targioni-Tozzetti. Asterolecanium, Targ., Inter. 2nd Mem. Studi Cocc. Catalogue, p. 41, 1869 ; Sign., Ann. Soc. Ent. Fr. (4) x, p. 276, 1870. Planchoma, Sign., Ann. Soc. Ent. Fr. (4) x, p. 282, 1870. Asterodiaspis, Sign., Bull. Soc. Ent. Fr. (5) vi, p. ecix, 1876. ‘Female insect completely enclosed within a thin but compact glassy or horny test, characterised by a continuous marginal fringe of glassy filaments. The test may be hemispherical, flat, or even somewhat concave above. It varies in outline from circular to linear. The surface is usually smooth, with, in some species, erect or curling filaments on the disc, similar to those of the marginal series. . . . At the posterior extremity there is a small opening through which the larvae escape. This aperture is sometimes placed at the end of a tubular extension, which is often more or less elevated. The colour of the test is invariably of a greenish or yellowish tint, but the-fringe and dorsal filaments are sometimes tinged with red... . “The adult female insect, denuded of its covering, is at first approximately of the same form as its covering. After oviposition it shrivels up and hes at the anterior extremity of the test, the remaining cavity being packed with ova. The antennae are rudimentary, consisting of a single short joint with a few curved hairs at its extremity. The limbs are totally absent... . No stigmatic spines. Anal lobes absent or minute, usually represented by a pair of small tubercles and each bearing a stout seta. In a few species, both tubercles and setae are wanting. Between the setiferous tubercles are usually from two to four smaller tubercles, each bearing a small spine. The anal ring is sunk in a tubular pit which sometimes opens on to the extreme margin, but usually terminates shortly before the margin on the dorsal surface ; it normally carries six stout hairs, but is sometimes hairless. . . . * See Bull. Ent. Res. 1x, p. 112, THE COCCIDAE OF SOUTH AFRICA. 109 “Male puparium of similar structure to that of the female, but smaller and pro- portionately narrower. The fringe is simpler, consisting only of the nymphal and larval elements. The winged adult emerges from beneath the posterior margin without the aid of any hinged operculum such as occurs in the allied genus Lecaniodiaspis. “Adult male with distinct neck. Antennae ten-jonted, with three or four knobbed hairs at apex. Ocelli large, in two pairs—dorso-lateral and ventral respectively. Rudimentary eyes small, colourless and inconspicuous. Scutellum ample. Halteres absent or obscure. (Genital sheath long and sharply pointed. Two long caudal filaments. “Larva of typical form. Antennae distinctly six-jomted. A marginal series of (usually twenty-eight) paired glands. Posterior extremity with a pair of longish caudal setae” . . . (Green). Key to South African Species of Asterolecanium. A. Anal ring hairless. a. Antennae rudimentary. Anal extremity of test turned up, tubular ; caudal setae of adult 9 short, stout brevispinum, sp. D. Anal extremity of test not prominently upturned or tubular ; caudal setae long variolosum, Ratz. aa. Antennae long, 10-jointed, caudal setae long wu .. borboniae, sp. n. AA. Anal ring large, with 6 long spine-like hairs. a. Marginal fringe of adult test reddish. Test small, flattish, usually naked when olds: ie iN pustulans, Ckll. Test small, flat ; on acacia a ” - .. conspicuum, sp.n. Test large, rounded ; on bamboo ss net S, .. bambusae, Boisd. aa. Marginal fringe ae adult test white. Test yellow, globular, enveloped in very long ney ae .. euryopsis, Fuller. Test greenish or creamy white, fringe short . £ stentae, sp. n. 166. Asterolecanium brevispinum, sp. n. (Plate vu, fig. 188 ; Plate vim, fig. 197). Test of adult 2 about 3 mm. long, 2°2 mm. broad, very convex, with the caudal extremity produced into a narow upturned process. The colour is bright yellow and somewhat transparent, except at the anterior end which is bright brown in the few specimens at hand, probably owing to the presence of the shrivelled female body. Of the eight specimens in this collection five of the tests are entirely naked. The other three show the presence of long large curved glassy filaments (fig. 188), which were apparently equally long and equally numerous over the whole dorsal and marginal areas. The integument of the body is very thin ; the antennae are small, -+ cordate, with two short spurs. The caudal extremity is simple, with a small hair- less anal ring ; the caudal setae are represented by short stout spines. The figure—8 glands are comparatively large (fig. 197). Remarks. This species is somewhat like A. coffeae, Newst., but the fringe is white instead of yellow, and there are other distinctive differences to be seen in the mounted specimens, particularly the hairless anal ring and short caudal spines. Habitat : On veld bush, Ceres, C. P.; collected by C. P. Lounsbury, October 1908. Collection No. : 21 110 CHAS. K. BRAIN. 167. Asterolecanium variolosum, Ratz. (Plate vii, fig. 190 ; Plate viii, fig. 193). Asterolecanwum quereicola, Sign. et Auctt. (nec Lecaniwm quercicola, Bouché). Coccus variolosus, Ratz., Tharander Jahrbuch, xx, p. 187, 1870 (not seen; fide Judeich & Nitsche). Lecanvum quercus, Altum (nec Linné), Forstzoologie, iii, Insecten, p. 365, 1881. Coccus variolosus, ‘“‘ Ratzeburg MS.,” Hagen, Can. Ent. xix, p. 60, 1887 (no description). Planchoma fimbriata (Boyer) Maskell, Extract, Trans. N. Z. Inst., 1894, p. 63. Coccus quercicola, Sign., Judeich & Nitsche, Lehrb. Mitth. Forstins., ii, p. 1252, 1895. Planchona quercicola (Bouché) Maskell, Trans. N. Z. Inst. xxviii, p. 396, 1895. Asterodiaspis variolosus, Boas, Dansk. Forstzoologie, p. 395, 1896. Planchonia (Asterolecanium) quercicola (Bouché) Froggatt, Dept. Ag. N.S.W. No. 75; poi eo7, Asterodiaspis quercicola (Bouché) Newstead, Coccidae of Brit. Is. (Ray Soc.) 1, pp. 1, 14, 34, 35, 36 and 39, 1900. Asterolecanwum variolosum, Newst., Mon. Brit. Coce. 11, p. 156, 1902. Ova pale yellow to light brownish yellow according to age ; about 190u long and 115 broad. Larva, just emerged, pale yellow, with legs and antennae hyaline, about 260 long and 145u broad. Antennae obscurely six-jointed. Caudal lobes small, tuberculate, each with one long seta, averaging about the same length as the antennae. LHyes prominent, lemon-yellow. In the half-grown female the colour of the insect is clearly visible through the thin transparent test. It is then 0°7 mm. long and 0°55 mm. broad, caramel-brown, with a distinct white fringe and distinct transverse ridges on the dorsum (fig. 190). At maturity the adult 2 completely fills the test, which has become stouter and yellower but is still transparent enough to allow the greenish brown colour of the adult insect to show through plainly. At this stage the tests vary in size from 1°2 mm. long by 0°9 mm. broad and 0:26 mm. high to 1-5 mm. Jong by 1°3 mm. broad and 0°6 mm. high. As the eggs are matured and laid the body of the female shrinks and the posterior end of the cyst serves as an ovisac. At this stage the test is regularly domed and practically smooth, while the fringe has usually more or less worn away. The line indicating the end of the shrunken body is often very distinct, so that the front half of the test appears deep brown and the hind half pale yellow. After clearing, staining and mounting, the adult 9 is almost circular, with the anal extremity slightly narrowed. The anal orifice is very small and hairless, with two small spines, one on either side. The caudal lobes are obsolete, but the two long setae persist as shown in fig. 193. The figure-8 glands are small, in a single series all round the body. The antennae are rudimentary, with one long and one short spiny hair (fig. 193a). Habitat: On Quercus sp., Cape Peninsula, Elsenburg (C. P.), Johannesburg, Irene, Pretoria and Vereeniging (T'vl.). Collection No. : 18. THE COCCIDAE OF SOUTH AFRICA. lll 168. Asterolecanium borboniae, sp. n. (Plate vii, fig. 198). Test of adult 2 small, about 1 mm. long and 0°75 mm. broad, rounded in front and rather pointed behind, slightly convex, with the margin rather thickened and slightly crenulate. The extreme posterior extremity appears tubular and is upturned or recurved over the back. There is apparently no marginal or dorsal fringe. The colour is pale greenish yellow and the test is almost transparent, showing the dark body of the female at the anterior end. When mounted the body is broad pear-shaped, about 0°9 mm. long and 0°6 mm. at greatest width, with the posterior extremity slightly produced,and broadly rounded. The characters are very indistinct. The anal ring is small ; the caudal setae stout, about 65 to 70u long, and the margin has but a single row of figure-8 glands, which are small, measuring approximately 8u across the pair. They are rather wide apart (about 30u) and are not associated with simple glands (fig. 189). The mouth-parts, with the rostral loop, average about 185, long. A g mounted on the type slide from the Cape collection, measures 0°77 mm. long without the antennae and genital spike. The latter is broad and strong and measures 150u. The antennae are 10-jointed, measuring approximately : (1) 20, (2) 26, (3) 60, (4) 64, (5) 74, (6) 70, (7) 68, (8) 58, (9) 50, (10) 46. Habitat : On leaves of “ gorse” (Borbonia cordata, Linn.), Ceres, C. P. ;- collected by T. F. Dreyer, November 1906. : Collection No. : 302. 169. Asterolecanium pustulans (Ckdll.) (Plate vu, fig. 187 ; Plate vii, fig. 200). Asterodiaspis pustulans, Ckll., Jl. Inst. Jamaica, 1, p. 143, 1892. Planchonia pustulans, Ckll., Science Gossip, xxix, p. 77, 1893. Asterolecanium pustulans, Ckll., Bull. Bot. Dept. Jamaica, p. 8, 1893. Test of adult 2 about 2 mm. long and almost as broad, nearly circular, with the posterior extremity slightly produced. Dorsum convex, with a rather conspicuous rounded median ridge, but without transverse ridges. When young the test is flat and entirely covered with pinkish filaments, which appear to be divided on the marginal area. When a little older the dorsal filaments are missing, but there is a distinct marginal fringe of rather long pink processes. When full-grown the marginal fringe also is usually lacking. The colour of the test is greenish yellow, except when it contains the shrivelled female or eggs, which show through the semi- transparent test. On apple the insects do not make pits in the bark, nor is there a ridge around them. Marginal row of figure-8 glands single, with the addition of a single row of simple glands. The figure-8 glands are of two sizes and are scattered over the body surface ; the smaller are about equal to the marginal series in size, the others almost twice as large (fig. 200 a). Antennal tubercles prominent, with two curved spines and a minute spur (fig. 200 6). Caudal lobes moderately prominent. Caudalsetae 75u. Anal setae 40u. Habitat : On apple, Lourengo Marques ; collected by C. B. Hardenburg. On stems of oleander, Lourengo Marques ; collected by C. P. Lounsbury, October 1914. Collection Nos.: 300 and B.300. (C605) B 112 CHAS. K. BRAIN. 170. Asterolecanium conspicuum, sp. n. (Plate vii, fig. 186 ; Plate vui, fig. 196). Test of adult 9 about 1°5 mm. in diameter, almost circular, with a small, narrowly rounded posterior extremity. On some species of Acacia the insects cause distinct pits in the bark of the twigs (A. robusta ?), but on A. horrida this is apparently not the case. Occasionally the bark has been seen to cause a low rounded mound around the insect, producing an effect on the twig very similar to that caused by Amorphococcus acacwue. The test is at first flat, yellowish, with short reddish hairs over and around it. Later it becomes moderately convex, smooth on top, semi-transparent and greenish yellow, with only the marginal fringe persisting. When cleared, the integument is extremely delicate and quite hyaline. The figure-8 glands in the marginal series are comparatively eae and are accom- panied by small circular glands of two sizes. The caudal extremity has low rounded tubercles which are scarcely produced, each with one stout spine of moderate length and a shorter one on the inside. The anal ring has six spines almost as long as the longer ones of the lobes (fig. 196). Remarks. This species is often observed to be heavily iaestea with a Hymenopterous parasite, and in many cases the majority of old tests show the exit holes of such insects. Habitat : On Acacia spp. (native species only), Pretoria, Marikana, and South-West Protectorate. Collection Nos.: 17 and 303. 171. Asterolocanium bambusae, Boisduval (Plate vii, fig. 185 ; Plate vii, fig. 195). Asterolecanium bambusae, Bdv., Insectologie Agricole, 1869. Asterolecanium bambusae, Sign., Ann. Soc. Ent. Fr. (4) x, p. 280, 1870. Asterolecanium bambusae, Newst., Mon. Brit. Coce. 11, p. 151, 1903. Asterolecanium bambusae, Green, Cocc. of Ceylon, iv, p. 328, 1909. Mr. Green’s description, without reference to figures, is as follows :— “ Test of adult female oval, convex above, the posterior extremity slightly produced into a blunt point where is a small terminal aperture ; smooth, glassy ; colourless, or tinged with pale green or yellow ; transparent, revealing the form of the insect and eggs beneath. Dried examples assume a more definite ochreous colour, a brown patch at anterior extremity representing the dead body of the insect. Marginal fringe pinkish, consisting of a double series of glassy filaments springing from the margin in pairs, each pair contiguous at the base and for the greater part of the length, but diverging laterally at their free extremities which meet the ends of the adjacent . filaments, thus forming a series of narrow loops irregularly crossed by fine web-like threads; the outermost series continuous, except at anal extremity, and longest; the second (nymphal) series less than half the length of the outer, and interrupted at more or less regular intervals ; an innermost (larval) widely spaced series of crook- shaped filaments. Length (without fringe) 2 to 25 mm. Breadth 1°10 to 1°60 mm. Fringe, outer series 0'1 mm. ; inner series 0:04 mm. ‘* Adult female insect at first more or less filling the test. After the deposition of the eggs, the abdomen becomes shrunken and the insect occupies the anterior part of the test only, the remaining cavity being packed with ova. Colour of insect dull THE COCCIDAE OF SOUTH AFRICA. lal green, the dorsum more or less mottled with reddish brown. Dried insect uniform reddish brown. Rostrum conspicuous, pyriform, approximately central. Antennae submarginal, rudimentary, consisting of an irregular tubercle surmounted by two longish stout bristles and one short spiniform hair. Spiracles conspicuous, subglobular, at a considerable distance from margin, with scattered series of minute pores connecting them with the margin. Abdominal extremity slightly cleft; anal tubercles broad and stout, but not very prominent, each with a long seta at its apex and one or two short stout spines near the base. Anal ring stout, with six stout hairs which just project beyond the margin. Margin with a continuous series of large approximately circular thick-rimmed pores, in pairs, and a double ventro-marginal series of minute pores. There are some scattered paired pores of a slightly larger size on the dorsum, and many simple circular pores on the ventral surface of the abdomen. Length of extended insect 1:25 to 175 mm. Breadth 0-80 to 1:25 mm. After oviposition the insect becomes greatly shrunken and does not resume its original proportions during maceration. “ Karly adult female at first pinkish, later dull green, finely maculated with reddish brown. ‘The outermost series of the fringe is at first the shorter, but soon outgrows the nymphal fringe. ‘Traces of the larval fringe can often be distinguished at intervals within the margin. “In the nymphal stage the fringe is in a single series, but within the margin are remains of the divaricating filaments of the later larval stage. ‘““ The newly hatched larva is of an oblong oval form, very pale pinkish or reddish brown. Limbs well developed. Antennae six-jointed, the sixth much the longest, with truncate apex. Margin with fourteen 8-shaped spinnerets on each side, and a median dorsal pair of similar pores immediately above the rostrum. Length 0:2 mm. “ The larva subsequently secretes a fringe of short glassy filaments which are paired and diverging on the cephalo-thoracic area, but single on the abdominal margin (although the spinnerets giving rise to these latter are of the same form as the others). ‘“ Ova very pale pinkish or yellowish. “Male unknown. A single example was observed, in which the dorsum was distinctly tricarinate and bore a marginal and three dorsal series of curling filaments. This may possibly be the male puparium.” The caudal extremity and part of the marginal series of figure—-8 glands of the é adult female are illustrated (fig. 195). Habitat : On stems of large bamboo, Durban ; collected on Aina occasions by Claude Fuller, C. . cnn and A. aes Collection No. 172. Asterolecanium euryopsis, Fuller (Plate vii, fig. 192 ; Plate vin, fig. 201), Asterolecanwum euryopsis, Fuller, Agr. Journ. C. G. H. xiv, pp. 93-95 and 818, 1899. As no fresh material of this species has been seen by the writer and all the numerous specimens of this collection are somewhat rubbed the following particulars are taken from Mr. Fuller’s description (l.c.) :— ‘“ The infested twigs have the appearance of being covered with a fine soft floss of a yellowish or white colour which is easily rubbed off and the insects exposed. The (C605) B2 114 CHAS. K. BRAIN. scales within which the females live very much resemble small glass beads, or the drops of resin which the plants exude and for which they have no doubt often been mistaken. They are thin and transparent with a yellow tinge of colour, smooth and shining, and very convex, being nearly spherical. Generally the yellow colour of the scale is not noticeable as it appears either brown or dark green, according to the different stages of development of the female inside. In size they are about one-fourteenth of an inch in diameter and a little less than that in height. “Tf these female scales are carefully examined with a magnifying glass it will be noticed that the floss is a production of the insect and that it 1s secreted through the scale at all parts, but mostly around the margin, and curls up in various directions, so that in fresh specimens the scale is quite hidden. If one of the scales is open at about this time of the year, mid-January, the eggs, small pinkish brown particles, will be found inside. Scales which contain eggs are generally parti-coloured, one half being yellow and the other brown or black. It isthe yellow half which contains the eggs, and the yellow colour is due to the scale, whilst the darker portion is the dried remains of the female showing through it.”’ Puparium of g about equal to the diameter of the female test in length, shghtly less than half as wide, flat, with a slight median keel, yellow and transparent, likean empty female test. Thereis no definite operculum, such as isfound in many species, but the extreme hind margin is split to allow the exit of the ¢. Adult 9, cleared and mounted, broad pear-shaped, 1°5 mm. Jong and 1:2 mm. broad. Body hyaline, with figure-8 glands very large and uniformly scattered except at the posterior extremity, where they are practically absent. The caudal extremity (fig. 201) is deeply excavate, with the part around the anal opening densely chitinised. The caudal setae are replaced by two stout spines, and the spines of the anal ring are comparatively very long and stout. Remarks. This insect became so numerous on the harpuisbosch in certain parts of Cape Colony about the years 1898 and 1899 and appeared to have such a toxic effect upon the host-plant that it was suggested that the insect be spread as widely as possible in an endeavour to kill off the bush, which rendered large tracts of the country around Tarkastad useless. It was recorded in the Agricultural Journal (l.c.) that the bushes grew to a height of 8 to 10 feet and that no grass would grow under them. Habitat : On harpuisbosch (EHuryops tenuissimus, Less.). Collection No. : 20. 173. Asterolecanium stentae, sp. n. (Plate vu, fig. 191; Plate vii, fig. 199). Test of adult 9 about 3 mm. long, regularly oval and very convex, with the extreme posterior extremity slightly upturned. Colour greenish yellow suffused with brown, which is more intense at the margin. Dorsal surface scantily supplied with short white glassy processes. The young female is similar in form, but is pale translucent green with a white marginal fringe and with numerous glassy filaments on the dorsum. The median line is not pronounced, but bears a large number of long glassy filaments in a longitudinal crest. The marginal fringe is about one-fourth the greatest width of the body in length (fig. 191). puparium not observed. THE COCCIDAE OF SOUTH AFRICA. 115 Cleared and mounted the adult 9 is broad pear-shaped, hyaline, except for the mouth-parts, spiracles and anal portion. The figure-8 glands form a continuous single row around the margin, except at the sides in the vicinity of the spiracles, where a double row is present in association with single glands of two sizes. Figure-8 glands are also present in scattered form over the dorsum. A little removed from the posterior end is a transverse series of large opaque glands reminding one of the grouped glands of the Diasprn Az (fig. 199). Theantennae are small tubercles with two or three small spines (fig. 199, 6). Remarks. This insect is obviously very similar to A. thes, Douglas, both in general appearance and microscopic characters. It is, however, slightly larger and could be distinguished by the slight difference in colour and the slightly upturned posterior end of the test. Habitat: On stems of Caralluma caudata (Asclepiadaceae), sent in by Miss 8S. Stent, July 1916. Also on Huerma transvaalensis, Stent, and Stapelia sp., in Rockery at Division of Botany, Pretoria. Collection No. : 29. Genus Lecanicdiaspis, Tare. Lecaniodiaspis, Targ., Bol. Soc. Ent. Ital. i, p. 261, 1869. Lecaniodiaspis, Ckll., Can. Ent. xxxi, p. 267, 1899. Prosopophora, Douglas, Ent. Mon. Mag. xxviii, p. 207, 1892. Burchippia, Green, Ann. Mag. Nat. Hist. (7) vi, p. 450, 1900. The adult 9 in this genus is entirely enclosed in a compact tough papery test, which is generally broad oval in shape and buff or yellow in colour. Its upper surface may be flat or convex, and may be + smooth or ribbed or carinated. Occasionally there are small waxy processes on the dorsum, and the insects form shallow pits in the bark of the host-plant. In the South African species only are there waxy plates on the dorsum. There is a small, + circular aperture at the posterior extremity of the test to allow the larvae to escape, but this is usually obscured until the eggs have hatched. The test, when first formed, fits closely to the body of the female, which gradually shrinks with oviposition until the test becomes in reality an ovisac. The adult © is sometimes apodous ; in other cases there are rudimentary legs or, more rarely, they are well developed. The antennae are usually well developed, seven to nine segments being most common. The anal extremity is slightly cleft, with the outer angles generally rounded and bearing a stout seta and a few spines. The base and inner angles of the cleft are supplied with a bilobed thickening, which reminds one very strikingly of the anal plates of the Lecanimaz. The dorsal anterior margin of the cleft has a transverse chitinous band, in front of which is situated the anus. The anal ring bears 8 to 12 hairs, usually 10. The derm is plentifully supplied with 8-shaped glands, and on the dorsal surface of the abdomen there are often two longitudinal rows of cribriform plates. Key to South African Species of Lecaniodiaspis. A. Legs absent. Test smooth, flat, not divided into il m, at .. mimosae (Mask.) Test divided into plates i a o .. natalenis, sp. n. AA. Legs rudimentary. 3 Test smooth, rounded above .. + oe re AG magna, sp. N. AAA. Legs well developed. Shorter than antennae K 3 et chy sf - braber, sp. n. Longer than antennae ue BS ci ae it tarsalis, Newst. 116 CHAS. K. BRAIN. 174. Lecaniodiaspis mimosae (Mask.) (Plate ix, fig. 202 ; Plate x, fig. 215). Prosopophora prosopidis var. mimosae, Mask, N. Z. Trans. xxix, p. 316, 1897. Lecaniodiaspis mimosae, Ckll., Check List Suppl. p. 392, 1899. ; Test of adult 9 about 4°5 to 5 mm. long., 3°5 mm. broad and 1°7 mm. thick, with the dorsum almost flat, the upper and lower surfaces almost parallel, with the margins rounded. When not crowded together the specimens are glued flat to the bark, button-like, but when a number are massed together they are often much distorted. The colour of the young is creamy, but later becomes suffused with brown, with a more distinct median line. The dorsal surface is flaky, without keel or transverse ridges. 3 puparium 1:8 mm. long, 1 mm. broad, elongate oval, rather more pointed in front, flat, with a median keel and faint transverse ridges, pale brown, with a distinct semi- circular operculum. | When boiled in KOH and cleared, the derm of the adult 9 is thin and hyaline, and the numerous gland-pores are very conspicuous. There are two dorsal rows of cribriform plates, 5 in each row. These are + circular, small, 24u in diameter and minutely perforate (fig. 215, b). The antennae are long (about 180), with seven or eight joints, sometimes appearing 9-segmented owing to a pseudarticulation in segment 6. The terminal segments bear several long stout processes with blunt extremities (fig. 215). Legs absent; in two cases indications of extreme rudiments of legs were found, in one case represented by leg i. and in the other by leg ii. The two pairs of spiracles are broad, and the stout tracheae generally remain after treatment with potash. The anal lobes are short ; each bears one very strong spine and 'a few short conical ones. The stigmatic spines are long, stout, slightly curved, and clubbed (fig. 215, a). Habitat: On Acacia horrida and other species, Fort Beaufort ; collected by C. P. Lounsbury, September 1900. Port Alfred ; collected by A. Kelly, March 1915. Namaqualand (Cape No. : 1254). Collection No. : 22. 175. Lecaniodiaspis natalensis, sp. n. (Plate ix, fig. 205; Plate x, fig. 213). Test of adult 2 about 2°5 mm. long and 1°6 mm. broad at the widest part, which is about the middle, flat, somewhat elliptical with the two ends narrowed. In some cases the anterior end is broadly rounded and the posterior extremity pointed. The dorsum is flat and covered with a layer of white material, which is distinctly divided into, three series of + rectangular plates, the appearance of which suggests an Orthezia. The median series is not quite so broad as the two lateral ones and consists of nine patches, the number which is apparently constant for each of the two lateral series also (fig. 205). When :placed-in boiling KOH the body becomes pinkish but the liquid is not appreciably coloured. | The body of the adult @ is flat, about 1°8 mm. long, and 1:2 mm. broad at the widest part, which is just in front of the middle. The anterior end is often suddenly narrowed at the level of the antennae and the anterior margin rounded. - The posterior portion tapers to the deeply cleft extremity. The mouth-parts are THE COCCIDAE OF SOUTH AFRICA. a7 comparatively small, with a long rostral loop. The antennae (fig. 213, a) are short, annular, of four or five segments. Legs entirely absent. The whole dorsal surface has numerous scattered, minute 8-shaped glands. Cribriform plates (fig. 213, b) small, inconspicuous, few in number. The anal cleft is as shown in fig. 213. The caudal setae are strong, and shorter than those of the anal ring (60u). In addition to the two caudal setae there are five or six short spines on 1 each caudal lobe. Anal ring with 8 long hairs (100,). Habitat : On stems of Hibiscus, Durban ; collected by C. P. v. d. Merwe, July 1916. Collection No. : 301. 176. Lecaniodiaspis magna, sp. n. (Plate ix, figs. 206,209 ; Plate x, fig. 214). Adult females ae on on crown of the host- eee at just about ground- level. Test of adult 2 about 6 mm. long, 4°5 mm. wide, and 3 mm. high, regularly broad oval, or slightly narrowed in front and with the hind margin very slightly flattened, with a faint median indentation. The dorsum is very convex, ventral surface slightly rounded. The test is entire and homogeneous in texture, smooth or very faintly roughened, without ridges, but occasionally with very faint ribbed corrugations atthe sides. The colour is of a uniform biscuit tint (fig. 206). When received, the largest specimens removed for mounting contained many eggs. The body entirely fills the test, which is thin, tough and papery. In boiling KOH the test breaks down and the liquid is stained a rich brown colour, while the 92 takes on a rose-pink hue. When boiled for some time the body wall is clear, hyaline, and very delicate. Adult 9°, cleared and mounted, about 3:7 mm. long, almost circular, hyaline, with the antennae and anal armature dense. Legs extremely rudimentary, represented by a small (50x) conical tubercle, with a minute process at the tip and several delicate hairs which appear to represent the digitules (fig. 214, a). Antennae comparatively short (170), of nme segments, which are annular and slightly tapering to the apex (fig. 214). Anal ring with 10 stout hairs. Cribriform plates small, with comparatively — large pores. Caudal spines short (27), two on each lobe. The anal plates are broad, each with deep wrinkles and two stout spines, and possibly two more slender, one above and one below (fig. 214,6). These latter are missing in the slides, but the pores indicate their presence. Remarks. Two plants were submitted, the one crown bore seven large females ranging from 5 to 6 mm. in length, the other about 50 specimens identical in appearance but smaller, all the tests being about 3 mm. in length (fig. 209). Habitat : On crowns of native bush with small narrow leaves ;_ collected at Groot Drakenstein and sent'in by C. W. Mally, June 1916. Collection No.: 27. 177. Lecaniodiaspis brabei, sp. n. (Plate x, figs. 218-218, a). Test of adult 9 about 3:2 mm. long, 2 mm. wide, and 1°5 mm. high, oval, convex, ochre-yellow, with a thin covering of greyish secretion which is easily flaked off. The dorsum is not quite smooth, but has faint rounded rib-marks and occasionally a faint 118 CHAS. K. BRAIN. median ridge. With the roughened secretion removed the colour and general appear- ance of this species is very like the figure Green gives of his L. azadwrachtae. § puparium of the usual type, pale bufi-coloured, not yellow as is the Q test. Larva yellow, 0°5 mm. long, and 0:24 mm. wide. Antennae and legs well developed. Antennae 5-segmented ; basal segment cylindrical, hardly as wide as ii. The posterior extremity of the body is narrowed and deeply and roundly indented, so that there appear to be two large caudal lobes. The inner margins of these have chitinous plates similar to those of the adult ; the transverse chitinous bar in front of these is not very conspicuous. The caudal lobes each bear one very long seta, about 0°2 mm. in length and one or two short spines. Second stage 2 about 1'2 mm. long. In this stage the body has become more rounded, especially at the posterior extremity. The caudal lobes are comparatively shorter and their extremities are curved together, and the long caudal setae of the larva are replaced by two stout spines about 35 in length. The antennae and legs are situated further from the margin and appear proportionately smaller. The antennae are 6—jointed. Adult 9, cleared and mounted, about 3 mm. long, almost circular, with the derm colourless, hyaline. The legs are well developed, but small, shorter than the antennae (180u: 250u). The three pairs of legs are about equally developed (fig. 218, c). The antennae are normally 8—jointed, but occasionally appear 9-jointed through the pseudo-articulation of the 5th segment (fig. 218). There are scattered stout spines, about 27u long, on thick bases, at intervals around the margin (fig. 218, 6). The stigmatic spines are usually unequal, one long and one short, curved, slightly clubbed (fig. 218, a). Figure—8 glands small, uniformly scattered over the surface. Remarks. This species is similar in many respects to Lecaniodiaspis africana, Newst., but is readily distinguished by the smaller size and the fact that the antennal segments ill, iv, v, and vi, are longer than wide and the tarsus is considerably longer than the tibia (fig. 218, c). Habitat: On wild almond (Brabeium stellatifolium, Linn.); collected by C. P. ' Lounsbury, Newlands, 1896 (Cape No.: 1274); also collected by C. W. Mally, December 1915. | Collection No. : 298. 178. Lecaniodiaspis tarsalis, Newst. (Plate ix, fig. 203 ; Plate x, fig. 217). Lecanwum tarsalis, Newst., Bull. Ent. Res. viii, p. 16, 1917. Test of young 9, about 1:5 mm. long, flat, with a distinct median keel, purplish grey in colour. Test of adult 9 bufi-brown, about 3 mm. long, 2 mm. broad and 1 mm. high, with the dorsum roughened by a series of low tubercular projections. There is a distinct median series and a sub-dorsal row on each side, the marginal areas being irregularly roughened, S$ puparium creamy white, about 1°3 mm. long, and 0°9 mm. wide; similar in form to the 9 test but more flat and much lighter in colour. Adult 9, mounted, about 1°6 mm. long and 1‘1 mm. broad, regularly oval, with a complete regular marginal row of stout spines 30u in length, about 15 on each side THE COCCIDAE OF SOUTH AFRICA. 119 (fig. 217, 6). Legs and antennae present. All three pairs of legs about equally developed, all longer than the antennae. Antennae 9-jointed (fig. 217). Caudal setae stout, about 80u in length. Cribriform plates small (124 diameter), yellow (fig. 217, a). On the venter, a little anterior to the position of the anal ring, are two long straight slender setae, about 115 in length, pointing backward. Habitat: On native tree (Dombeya rotundifolia), Pretoria; collected by the writer, September 1914. On Hibiscus in a nursery, Pretoria; collected by A. Kelly, June L1G: ! Collection No. : 23. Genus Geroceccus, Comst. Adult 9 enclosed in a dense waxy test without a marginal fringe as in Asterolecanium, and generally without waxy processes. The test 1s usually dense, opaque, not semi-transparent, and simulates the test of a Tachardia more than that of an Asterolecanium. The posterior end of the test is sometimes slightly produced and uppurned, with the anal aperture at the extremity. Adult 92 with the caudal extremity produced into a rounded protuberance which bears the anal armature. There are two stout lobes, each with a long seta and several spines ; the space between them is + chitinised, with a prominent rounded tongue-shaped plate. Antennae and legs rudimentary. Derm with conspicuous S-shaped glands. Stigmatic spines absent. G puparium elongate, tubular, with a large oval or circular operculum above the posterior extremity. Key to South African Species of Cerococcus. A. Test smooth, + globular, brown .. a ig be .. ovordes (Ckll.). AA. Test rough, ornate, brick-red ipl be n ey, .. ornatus, Green. AAA. Test yellowish or brownish. Marginal 8-shaped glands much larger than those on dorsum passerinae, sp. 0D. Marginal 8-shaped glands little if any larger _... we .. royende, sp. Nn. 179. Gerococcus ovoides (Ckll.) (Plate x, fig. 219). Pollina ovoides, Ckll., The Entom. xxxiv, p. 225, 1901. Professor Cockerell’s description is as follows :— “© scale a rounded conical object much like a Lepidopterous egg, about 14 millim. diam., roughened radiately, pale brown with four longitudinal stripes of white secretion converging to the top of the scale, which is usually reddish. “ § scale elongated, about # millim. long, roughened, yellowish or pink, with an oblique terminal cap. “ 2 scales, soaked in liquor potassae, give a deep orange-brown colour, and the insects themselves turn deep crimson; Q adult globose ; skin with many simple round glands and tubular glands, and some figure-of-8 glands; anal ring with numerous hairs ; caudal lobes prominent, conical, about 45u long, with a few small spines, and ending in stout bristles about 90u long ; mouth-parts well developed, but small; labium short and broad, dimerous, the last joint with bristles on its margin; antennae 120 CHAS. K. BRAIN. represented by small rounded tubercles about 15u long, with a little terminal prominence which appears to represent a second joint, and about six stout bristles about 18 long ; spiracles small but distinct ; legs wanting. “ Embryonic larva with a row of figure—of-8 glands down each side, and dorsal and subdorsal rows of small round glands, the latter failing caudad, the last five glands of the dorsal rows being absent in the subdorsal ; labium very short and broad, cup- shaped ; form of insect elongate-pyriform ; antennae thick, 6-jointed, last joint not very greatly longer than the one before, and notched asin P. pollini.” The antennae and caudal extremity are figured (Plate x, figs. 219 and 219, a). Habitat: On stems of undetermined native tree, Back Beach, Durban; collected by C. Fuller. On fig, Pietermaritzburg ; collected by A. Kelly, 1911. Collection No. : 26. 180. Gerococcus ornatus, Green. Cerococcus ornatus, Green, Cocc. Ceylon, iv, p. 306, 1909. Mr. Green’s description, omitting figure references, is as follows :— “ Test of adult female irregularly oval, the posterior extremity (in fully matured examples) produced into a short upturned tube. In earlier examples this caudal extension is in the form of a trough, partially closed above by stout curling filaments. Dorsum strongly convex, the sides sloping inwards to the comparatively narrow ventral area, which is thin and easily ruptured. Dorsum with three prominent transverse waxy ridges produced, at the middle and two sides, into stout pointed processes ; the lateral processes longest, tapering to fine points, and projecting (in fresh examples) considerably beyond the margin, which is itself thickened and tubercular. Posterior margin with two or more tapering processes directed backwards. There are several pairs of short white curved filaments on the median line, behind the third transverse ridge. In older examples the waxy processes are gradually reduced until they practically disappear, and old worn tests are more or less globular in form. Colour of early test purplish brown ; the ridges, processes, and thickened margins bright coral-red or crimson. Older examples become at first uniform pale red, and finally dull reddish brown. There are two white waxy ridges on each of the sloping sides, extending from the first and third lateral processes, respectively, to the stigmatic openings. Length 2 to 2.5 mm. Breadth (exclusive of processes) 1°12 to 1°50 mm. ‘“ Male puparium oblong, narrow, rounded at the extremities. An inconspicuous median and two prominent transverse ridges, the latter terminating laterally in pointed processes. A smaller pointed process, on each side, close to anterior extremity. Posterior third occupied by a large circular operculum with raised margins. Colour pale red tocrimson. Length,1mm. Breadth, 0°45 mm. “ Adult female insect of a delicate greyish tint, in life. Broadly oval: the terminal segments abruptly narrowed. Abdomen terminating in two prominent irregularly conical lobes, with a bluntly conical median plate between them ; each lobe with a triangular chitinous plate on its inner side and a longish stout seta at its apex. Ventral lip of anal aperture with four or more long stout spines. Anal ring with eight hairs. Rudimentary antenna with from eight to twelve stout hairs on its apex. THE COCCIDAE OF SOUTH AFRICA. 121 Limbs entirely wanting. Dorsum with numerous large and conspicuous paired glands, disposed principally in three broad transverse bands across the middle of the body, leaving the anterior and posterior areas comparatively free. The first and third bands divide near the margins, where they enclose a series of small circular glands. ‘There is a straggling group of the large paired glands within the anterior _margin; a short marginal series on each side of the abdomen, just before the constriction ; and another short series, on each side, close to the anal lobes. Amongst the larger paired glands, and scattered over other parts of the dorsum, are many of a much smaller size. Cribriform plates small, each with a broad densely chitinous border and areolate centre; in two groups of four, immediately anterior to the narrowed part of the abdomen. Venter with small circular glands in loose transverse series across the abdominal segments, and some scattered paired glands of the smaller size. Length, 2 mm. * Adult male uniform dark brown. Wings hyaline ; nervures colourless. Genital sheath very broad at base; sharply pointed at extremity. Although the single example examined showed no waxy caudal filaments, there is a distinct glandular pit on each side, giving rise to fine paired setae such as usually support caudal filaments. Length,1 mm. Expanse of wings, 1°75 mm.” - Habitat: On stems of Aberia caffra and Calodendron capense, Thunb.; collected at Pietermaritzburg by A. Kelly, April 1917. Collection No.: 95. 181. Gerococcus passerinae, sp. n. (Plate x, fig. 216). Test of adult 2 about 2°4 mm. long, elongate, convex, with the posterior end slightly produced, thin, brittle, dull straw-yellow. The dorsum is convex, with a wide shallow groove from about the middle to the caudal end. The remainder of the test _ is uniformly arched, but in a few specimens there appears to be a submarginal series of low tubercles. Puparium of ¢ usual, elongate, paler in colour than the @ test. Cleared and mounted the adult 9 is broad pear-shaped, with the prominent anal portion produced. The integument is thin and hyaline, with 8-shaped glands numerous but scattered. Those of the marginal series are large, and mostly in a single row, but at the sides the series becomes double or treble for a short distance. The 8-shaped glands scattered over the dorsum are smaller, being only about half the diameter of those in the marginal series. The rudimentary antennae consist of small conical protuberances each with about 6 spines (fig. 216, a). The caudal lobes are prominent, with their inner faces chitinised and the spines strongly developed (fig. 216). Habitat : On native shrub (Passerina ericoides, L.; Thymelaeaceae); collected by C. P. v. d. Merwe at Montague, C. P., October 1914. Collection No.: 24. 182. Gerococcus royenae, sp. n. (Plate ix, fig. 210). Test of adult 9 varying in size to 3 mm. long and almost as broad and high, very roundly convex, with a short caudal prominence around the aperture. The test is comparatively thick and brittle, dull orange-yellow to orange-brown in colour, without fringe or processes of any kind. 122 CHAS. K. BRAIN. The g puparia are elongate, tubular, open behind, straw-coloured, with a brownish secretion, which often appears in + rectangular patches on each side of the median line (fig. 210). The mounted @ is almost circular, with the caudal extremity slightly produced. The integument is clear and has a marginal series of 8-shaped glands of medium size. Other similar glands are scattered over the dorsum and these are of about the same size as those of the marginal series, not much smaller as in C. passerinae. The antennae are small conical protuberances with several curved spines, and are longer and narrower than in C. passerinae. The caudal lobes are moderately long, but are not thickened on their inner edges, and a short distance in front of the anal ring are two chitinised discs. Habitat: On stems of blauwbosch (Royena pallens, Thunb.; Ebenaceae), Fauresmith, O. F.§.; collected by J. C. Faure, March 1915. Collection No.: 96. Subfamily TACHARDIINAE. “ Insects enclosed in a resinous cell with three orifices. Adult females apodous, with the terminal segments produced into a tail-like organ bearing, at the extremity, the anal orifice, which is surrounded by a broken setiferous ring ”’ (Green). Genus Tachardia, R. Blanchard. Professor Cockerell (‘“ The Entomologist,” xxxiv, p. 249, 1901), before describing Tachardia albida from Natal, gives the following synopsis of this genus :— “ Tachardia, Blanchard (Lac Insects). This genus contains some very diverse elements, which will no doubt eventually be treated as genera. We may for the present recognise three subgenera :— (1) Tachardia proper. Type, T. lacca, the East Indian commercial lac. Female very elongated, vasiform ; the individuals enclosed in masses of lac surrounding the twigs, never separate. I know of only one species of this group. (2) Tachardiella, subg. nov. Type, 7. cornuta, Ckll. Female more or less globular ; individuals often separate. This includes the species of America and Australia. (3) Tachardina, subg. nov. Type, 7. albida, with the characters given below.” It will be noticed from the description of 7. albida given later that the main points of difference between this and the other species mentioned by Professor Cockerell are (a) the tests are collected in large masses ; (b) the caudal process is peculiar ; (c) the spine is absent. Of these characters none is peculiar to the African species, and T. albida is the only species I know of in which the spine is absent. 1 therefore follow Newstead in disregarding the subgenera for the present and including the six South African insects in the genus Tachardia. I might mention that the insects in this sub-family are by far the most difficult to make satisfactory microscopic slides of, as nearly all the old dry specimens seem to be eaten out, and after dissolving the lac test there usually remain but a few fragments which are quite useless for determination. Fresh material is therefore most desirable in this group. THE COCCIDAE OF SOUTH AFRICA. 123 Key to South African Species of Tachardia. - A, Antennae present. a. Test crimson or deep red, with pale radiating lines .. actinella, Ckll. & King. aa. Test deep shellac-brown, b. Rudimentary legs represented by small conical spines .. munor, sp. n. bb. Legs entirely absent .. me a a i -» ». karroo, sp. n. AA. Antennae absent. B. Spine present. a. Test deep purplish brown, with radiating ridges .. .. decorella (Mask.) aa. Test yellowish brown .. +. Me a ie ne affluens, sp. n. BB. Spine absent. a. Test whitish or pale yellow... ae Ee as) re albida, Ckll. 183. Tachardia actinella, Ckll. & King (Plate ix, fig. 208; Plate xi, fig. 221). Tachardia actinella, Ckll. & King, The Entom. xxxiv, p. 342, 1901. The original description is as follows :— “0, Scale about 3 mm. long, 3 broad, and scarcely 2 high, rounded and depressed, dark crimson, with about sixteen strong, but obtuse, radiating yellowish white ridges ; ~ centre of scale formed as in 7. decorella. The scales are mostly separate, but some- times two or more coalesce. “Q. Dark red, 24 mm. long, 14 broad ; bright pink when boiled in taustic potash ; this colour is due to the internal juices, the skin being perfectly colourless. Antennae stout, cylindrical, pale, about 140u long, 52 broad at base, 36 in middle, 28 at end, obscurely four-jomted. Lac glands with over 60 orifices. Mouth-parts about 141 broad. Spine well-developed, 120u long, rapidly enlarging 35u from tip to a very broad (95) base. Anal ring with 10 long bristles; chitinous anal plate roughly semicircular, posteriorly with a deep linear incision 60 long ; on each side of the anal plate is an elongated process terminating in two sharp spines, the structure being apparently the result of a fold in the plate. Anteriorly, the anal plate is tuberculated. “ 3. Scale cylindrical, elongated, of the usual form ; dark red.” A large collection of 3 material was received from Grahamstown in March 1915 when the insects were just emerging. The following particulars are therefore added from notes made at the time : & test deep madder-red, glassy, about 1 mm. long and 0°5 mm. broad ; flat, upper surface slightly arched, with a median flatly-rounded keel with transverse striae. The anterior end is rounder but narrower than the posterior extremity, which is somewhat elevated, so that the opening from which the § emerges appears dorsal. Prior to emergence this orifice is closed by a glassy cap. When the insect is ready to emerge the cap is raised from behind and the two white caudal filaments protrude. Body of ¢ adult (freshly mounted) 1:2 mm. long, and 0°33 mm. broad across the thorax. Genital spike 0°24 mm. long. Wing 0°95 mm. long. The body and scutellum are of a beautiful madder-red colour; the eyes are black; the legs and antennae pale, almost colourless ; and the wings are clear but iridescent. 124 CHAS. K. BRAIN. Caudal filaments (2) long, from 1} to 14 times the length of the body. These are dense white, slender, and in hfe widely divergent. The antennae are 10-jointed, the joints, when freshly mounted, measuring approximately :—(1) 30, (2) 58, (3) 100, (4) 92, (5) 85, (6) 68, (7) 58, (8) 68, (9) 50, (10) 65. Habitat : On orange, Qumba, C. P. On pomegranate, Ladysmith, Natal. Also on native trees at Kast London and Grahamstown, C. P., Natal and Transvaal. Collection Nos. : 71, 75, 325. 184. Tachardia minor, sp. n. (Plate xii, fig. 227). © test 15 to 2 mm. in diameter, rich castaneous brown, with the larval cast conspicuous. Viewed from above the test is broad heart-shaped ; edges of caudal orifice shghtly elevated ; spiracular openings obsolete. In the material examined the © tests always appeared singly on the twigs, never massed as in 7’. actinella. ¢ puparium about | mm. long, varying in colour from pale yellowish brown to deep castaneous ; dorsal area strongly ridged. When cleared and mounted the adult ° is almost circular and very transparent ; the mouth-parts are comparatively large, averaging about 170u long by 82 at their greatest width. On each side, extending backward from the middle of the mouth- parts, is a series of three small conical spines, which appear as though they might indicate the extreme rudiments of legs. The antennae average 60u in length and are obscurely 3 or 4-jointed. Lac plates varying from 57 to 68u in length; about 48u broad ; with few glands, the number varying from 14 to 28 in specimens examined. Spine about 68 long and 54 across the base, distinctly funnel-shaped, narrowing at about half its length. Chitinous anal plate almost semi-circular, with the basal portion rugose and its apex four lobed. Anterior spiracle with 16 simple glands. Habitat: On rhenosterbosch (Elytropappus rhinocerotis, Less.) ; collected by C. P. v. d. Merwe, Montague, C. P., November 1914. Collection No. : 25. 185. Tachardia karroo, sp. n. (Plate xii, fig. 230). © test 2°5 to 3°5 mm. in diameter, sometimes sing!e, but usually massed together on the thicker stems of the plant. Somewhat similar to 7. minor but more spherical, larger and paler in colour. In a number of cases the test is semitransparent, deep amber-yellow, with three distinct pale lines radiating from the median dorsal ridge to the margins in the form of a broad Y. S puparium comparatively short, deep chestnut-brown. When cleared and mounted the following characters may be determined :—The antennae are obscurely 3-jointed, about 105u long and 27 wide at the base. The apical joint has at its tip 3 or 4 short spines. The anterior spiracle has a few (2 or 3) simple glands. Lac plates about 120u long by 110u broad, with 62 to 84 gland openings. Spine about 90u long and 78w across the base; deep funnel-shaped, narrowing about one-third to one-fourth its length from the tip. Anal plate broader than long ; basal portion coarsely wrinkled, apical part with 4 small pointed lobes. Habitat: On rhenosterbosch (Elytropappus rhinocerotis, Less.) in association with 7. minor; collected by C. P. v. d. Merwe, Montague, C. P., November 1914. Collection No. : 324. THE COCCIDAE OF SOUTH AFRICA. 125 186. Tachardia decorella (Mask.) (Plate ix, fig. 212 ; Plate xi, fig. 222). Carteria decorella, Mask., N. Z. Trans. xxv, p. 247, 1892. Tachardia decorella, Newst., Bull. Ent. Res. vin, p. 127, 1917. The. ° tests generally coalesce so as to form a mass completely surrounding the thin twigs of the host-plant, often for a length of 30 to 40 mm. The individual tests, at maturity, are almost globular, but flattened above, 3 to 4 mm. in diameter, deep purplish brown, or almost black, with dull grey speckles arranged in radiating ridges. The intermediate spaces are somewhat glossy.. The lacis very hard and brittle. The larval tests are deep red. Those of the young females are flat and button-like, brown, with the reddish larval cast superimposed and with the grey ridges more pronounced than in the adult. When mounted the body is broad ovate and transparent. The anal plate is densely chitinised, about 190 broad and 170u long, with its basal portion embossed with very small “ grease spot ” design. The mouth-parts are about 185 long and 126 across the middle. The lac-gland plates are approximately 170u long and have numerous gland openings, all the specimens examined having over 100. The spine is shallow funnel-shaped, about 120u long and 112u across the base, becoming abruptly narrowed at about two-thirds its length from the apex. Near the margins of the segments are more or less circular groups of simple glands. This insect differs in some small details, such as the colour of the test, from Maskell’s description of 7’. decorella, but the marginal gland groups and other characters are so close that, for the present at least, 1 ascribe his name to the species. Habitat: On Acacia karroo, Linn., Crocodile River, Transvaal. On Acacia melanoxylon, Zoological Gardens, Pretoria. Collection No. : 323, 326. 187. Tachardia affiuens, sp. n. (Plate xi, fig. 224), Test of adult °° usually found singly on the stems of host-plants ; occasionally in groups of two or three, never in large groups. Test + globular, almost as deep as wide, sometimes slightly tapering to the top, about 3 mm. in diameter, yellowish to dull shellac-brown with a reddish spot in the centre. The test is generally smooth, but may show indications of faint ridges to the margins, which are prominent in young specimens. d puparium exceptionally long and narrow, pale yellowish brown, with reddish larval cast ; posterior aperture closed by a rough buff flap. When mounted the characters are indistinct, but the following measurements have been supplied from the examination of a range of specimens. :—Mouth-parts about 1704 long and 100u wide. Spine about 100u long and 85y at base, rapidly tapering so that the apical half is almost parallel-sided. Lac plates small, about 85 long and 75 broad, with approximately 40 gland openings, all in a compact group. The anal plate is more straight-sided than usual and ends in four almost equal lobes. Its base is slightly wrinkled aud has scattered transparent spots, giving it the grease-spot appearance something like the design on the dorsum of a Pseudaomdia. 126 CHAS. K. BRAIN. Remarks. This species resembles 7’. longisetosa, Newstead, but is easily distinguished by the fact that the hairs of the anal ring are of normal length, and the glands of the lac-plates are fewer in number and not scattered. Habitat: On Huclea sp. and other native plants ; very common around Pretoria and apparently widespread throughout the Union. Collection No.: 76. i88. Tachardia albida, Ckll. (Plate ix, fig. 211). Tachardia albida, Ckll., The Entom. xxxiv, p. 249, 1901. Professor Cockerell’s description 1s as follows :— “ Forming smooth yellowish-white masses on the twigs ; the extremely dense and hard lac of the several individuals running together ; masses up to 10 mm. diam., and 30 in length. The individuals are marked externally by orange patches, each presenting a small corrugated or segmented ridge, and an aperture. Cavities for females globular to subpyriform. Male scales of the usual elongated form, red, with a very short dorsal segmented ridge, about one-third of total lengths of scale. “ ©. After boiling in liquor potassae globular, giving a very fine crimson colour. Skin after boiling transparent, truncate and caudal processes remaining ferruginous. Mounted female on slide about 5 mm. diameter. Truncate processes (or ‘lac tubes’) very short, orifices very small and numerous. Spine apparently absent. Caudal process peculiar ; transversely oblong or subreniform, with a deep posterior notch, on each side of which are two lobules ; lateral hind margins bearing a sharp spine ; surface finely reticulated ; anal ring hidden, only the ends of the numerous bristles projecting. Spiracles large and well-developed. Mouth-parts well- developed, but small, about 135u broad; “lobes oraux ” (as figured by Targioni- Tozzetti in T. lacca) very large. ‘“ Larva in female fusiform, narrow, tapering posteriorly, about 560u long and 240 broad ; caudal bristles very long.” Habitat : On Acacia karroo, Linn., Verulam, Natal (Fuller No. 5), and Pienaars River, Transvaal ; collected by the writer, January 1917. Collection No. : 72. Subtamily ? Genus Halimococcus, Ckll. Halvmococcus, Ckll., The Entom. xxxv, p. 15, 1902. Only one species has yet been described 7.e., H. lampus, Ckil. Protessor Cockerell’s original descriptions of the genus and type species are as follows :— ‘““ A Dactylopiine Coccid enclosed in a horny sac shaped like that of Solenococcus, without legs or antennae in the adult. Larva with no rows of dorsal spines, no hairs on anal ring, and no caudal tubercles, but four long caudal bristles as in Phoenicococcus. Closely related to Phoenscococcus (which lives in Algeria), but distinguished by the form of the sac, which exactly imitates that of Solenococcus.” THE COCCIDAE OF SOUTH AFRICA. 127 189. Halimecoccus lampas, Ckll. (Plate ix, fig. 204 ; Plate xu, fig. 225). Halimococcus lampas, Ckll., The Entom. xxxv, p. 15, 1902. “©. Enclosed in a dark brown horny sac (which is not dissolved by liquor potassae), which is shaped like a Terebratula shell, 2.e., oval, with the end raised and terminating in an orifice. Length of sac 510u, breadth 300, breadth of orifice about 66u. The orifice is closed by a reticulated plate, except basally, where there is a semilunar opening. In immature examples the sac is prominently segmented on the ventral side. “ @. Scale small, cylindrical, horny, ferruginous, of the same texture as that of the female, but usually somewhat paler. Length 350u, breadth 140. The end comes off, leaving a round opening, as in Muscid pupae. “ ©. A mere bag, with well-developed mouth and spiracles. ‘“ Larva. Rather narrow ; legs and antennae present. No caudal tubercles, but two pairs of long caudal bristles ; two small bristles close to these. Antennae about 45 long, six-jointed, last jot much the longest. No dorsal spines. Last antennal joint with two long bristles. Femur remarkably stout, about 15u broad ; length of femur + trochanter about 30u. “ Older specimens have actually shorter (36u) antennae, with joint 6 longer than 4 +5; 5longer than 3, 3 longer than 4, | large. “ A few white curled waxen threads protrude from beneath the sacs of the females.” Habitat : On leaves of palm (Hyphaene crimta), Durban, Natal ; collected by C. Fuller about 1901. Collection No. : 42. Genus Baeeacoceus, nov. Alhed to the Lecanttmnaz. Body of adult 9, at maturity, naked, without test or waxy covering, almost globular, appearing like a berry on the twigs of the host- plant. In the type species the colour is yellow to orange, often with a faint metallic bronzy sheen, the whole insect simulating a Margarodes cyst. Antennae well-develoved, 7-jointed. Legs well-developed, long, upper and lower digitules present, clubbed. Anal ring large, with numerous (16 2) hairs. Stigmatic clefts, two on each side, with 4 or 5 broad conical protuberances supphed with glands. Type, B. elytropappi, sp. n. 190. Baccacoccus elytropappi, sp.n. (Plate ix, fig. 207; Plate xii, fig. 226). Adult female almost globular, to 3 mm. in diameter, yellow to brownish, glossy, hike a Margarodes cyst, with the small brown anal plates occupying the position of the caudal pore of an Asterolecanium test. In the middle of the dorsum there is a prominence, like a minute keel, occasionally with two rounded prominences in close proximity but a little posterior to it. There is no fringe in the material at hand, but four short white bands, two on either side from the stigmatic clefts at the point where the insect is attached to the thin twig or leaf of the host-plant. When cleared and mounted the derm is almost hyaline, appearing yellowish, with the antennae and legs of about the same density. The integument is without gland-pores and hairs except immediately around the anal plates (fig. 226, d). The (C605) C 128 CHAS. K. BRAIN. legs are long and slender, with long slender digitules, which bear spherical knobs. The lower digitules of the claw are stouter and shorter. The antennae are 7-seg- mented with joints 3 and 4 long, and almost equal. The stigmatic clefts are broad and shallow with 4 or 5 low conical glandular protuberances (fig. 226, c). The anal ring has numerous (10 to 12) long hairs and is closed above by two plates as illustrated in the figure. Habitat : On thin twigs of rhenosterbosch (Hlytropappus rhinocerotis, Less.), French Hoek, C. P. ; collected by C. P. Lounsbury, 1904. Collection No. : 297. ee x ita F a 1a ary iy 4 rigs yy eH) i a vy , sheep nih Poa hh “a ie ie, ah uae Sh ‘ El Veit oe = rie y mt i | a vie ie v EXPLANATION OF PLATE V. . Chionasprs (Poliaspis) carissae, Cll. (Phenacaspis) natalensis, Ckll. 39 i. leucadendri, sp. n. (Dinaspis) diosmae, sp. n. a (Phenacaspis) lounsburyt var. ekebergiae, var. Nn. . Chionaspis (Phenacasprs) lounsburyr, Cooley. ig (Dinaspis) distencta, Leon. A leucadendri, sp. n. (see fig. 166). ‘ (Dinaspis) lounsburyi, Leon. ss sp. on shrub (name of host-plant and locality not known). . Chonaspis caffra, sp. n. 35 chionaspitiformis (Newst). ES scutuae, sp. 0. a ambiguus, sp. N. ISWiLjt, JSAP, IRESENAGE, Wel, OX, IBAsGe Ws PLaTE V del. cx.8 South African Coccide. A AO Morr AVALIX? yy soi igang | : ‘ fbnastou) rely Paina va Moke ik stent) Biaevely aa } i EXPLANATION OF PLATE VI. Fig. 178. Chionaspis simplex, Green, var. 179. a (Dinaspis) embricata, sp. n. 180. hs (Pinnaspis) cyanogena (CkIL.). 181. * (Pinnaspis) aspidistrae, Sign. 182. Lepidosaphes glovert (Packard). 183. ss pinnaeformes (Bouché). 184. Ischnasyis longirostris, Sign. years, W/Al. MOE Me ART 2: Bucci. ENT. RESEARCH. South African Coccide. oe Aad 6 ‘ \ ‘ bart ) oh f % i t fe i ( H i € A (ete = { d \ 1 ne 7 , 5 , i = A % y int Y i MS a) » iva = “i ad a is A Vi ivi i i oe Ati i f EXPLANATION OF PLATE VII. Hig. 185. Asterolecanium bambusae, Bdv. 186. et conspicuum, Sp. L. evs ¢ pustulans, Ckll. 188. s brevispinum, sp. N. 189. - pustulans var. brachylenae, nov. 190. i variolosum, Ratz. 191. = stentae, sp. 0. 192. euryopsis, Fuller. ISUILIL, JEW, IRissianyecw, Wot. O<, Ieee YD, Piynieis, WAGE. IS). Size South African Coccide. _aeypoloat ‘ rnin ssbopteh " ‘ bil sini pa! cf ity tel ne Peeters H siti aa a) f i i bil allan: gh sal kay even iat mr Ne ry i Daa Ny uM 198. 0: 200. 201. EXPEANATION OF PEATE VIE . Asterolecanvum variolosum, Ratz.; a, marginal gland series. - pustulans var. brachylenae, nov. ; 5 a, marginal gland series. bambusae, Bdv.; a, marginal gland series. | a conspicuum, Sp. N.; a, marginal gland series. brevispinum, sp.n, caudal extremity of 2; a, antennal tubercle. Fe borboniae, sp. n., caudal extremity of 2; a, 8-shaped glands of derm; b, antennal tubercle. » stentae, sp.n.; a, marginal gland series; 6, antennal tubercle. e pustulans, Ckll., caudal extremity of Q, euryopsis, Fuller, caudal extremity of ©; a, dermal glands. BULL, JEG, IRESENR@s. WoL, IX, [Paina Ws yas WAUEL ° o® © Oy, rok eae a © A) CO ROMNG WON CHONamcntS es) ° f 6 @ G&D gy GG g@ a @ 8 @D ED GD GD GP oer i ® 0 ty OGRE Os a South African Coccidee, PPT) Gi ue feta at Di ie By oy an OY ah } rf i hat vo ay il rage Wie sh oy ais iss Vee Feet! EXPLANATION OF PLATE IX. . Lecaniodiaspis nimosae (Mask.) i tarsalis, Newst. . Halimococcus lampas, Ckll. . Lecanodiaspis natalensis, sp. n. a magna, sp. n. . Baccacoccus elytropappi, sp. n. . Tachardia actinella, Ckll. . Lecaniodiaspis magna, sp. n. (smal! form). . Cerococcus royenae, sp. Nn. . Tachardia albida, Ck. decorella (Mask.). PMA VX ENT. RESEARCH. VoL. X. Parr 2. BONL IC, South African Coccides. yey, : ¥ ad a Pe MR el ah ebm al oy a AP inpay 13 *% 1dis9 o oe he RON, od 8 pligunnatins pod iene ie i. sek nea ln eee ng aie cuait Pie oa aes re ce f ( v yi : { ms re x ov ate py a wires)’ Ole pe i igh ng * Papnin “ | © $i: nich , a ate Oe Ui ee tthe Hees Iobsies (IED)’ sthevo. dyoninbelh | Lar ae Ae abe coal ba ee fab | a ' a j ue } es ; aa iv - Y é apis | AIS yi | i” ‘ ee nt "i n 2 = oh ial “] ‘ < ¢ th = - Wa fod were, Fig. 218. 214. 215. 216. Zils 218. 219. 220. EXPLANATION OF PLATE X. Lecaniodiaspis natalensis, sp. n. ; a, antenna of Q ; b, eribriform plates. magna, sp. MN. antennae of 9; a, rudimentary leg ; b, anal plate, 9. mimosae (Mask.), antenna of 9; a, stigmatic spines; 6, cribriform plate. Cerococcus passerinae, sp.n., caudal extremity ; a, antennal tubercle of 9. Lecaniodiaspis tarsalis, Newst., antenna; a, cribri- form plates; b, tarsus; c, mar- ginal spines; d, stigmatic spines OFS: brabei, sp.n., antenna ; a, stigmatic spines; 6, marginal spine; ¢, leg 1, of 9; Cerococcus ovoides (CkIl.), caudal extremity ; a, antennal tubercle of 9. Amorphococcus acaciae, Brain, caudal extremity ; a, antennal tubercle of 9. 29 JBiGiijL, JSG, JRIESIe VAG, Wolk, X. IPA B, PLATE X. South African Coecide. EXPLANATION OF PLATE XI. Tachardia actinella, Ckll., lac-gland plate ; a, spine of 9. ,, decorella (Mask.), lac-gland plate ; a, anal plate of 9. ss decorella (Mask.), spine of 9. a affluens, sp.u., spine of 9. eyes DX RESEARCH. VoL. X. ParT 2. ENT. SHOVEL. South African Coccide. ee STAIN IO ViOITAVAIIKS Ibe i Re Ric) barbers pide - 16. Nel an ’ Be: e wos 8 aft tye abe viyths Mv ive eh: \ ‘Avirs Ly 7 Aol ‘ontaa nite i Fh A a hie Ki A t Ho yg x (tenlt an sibiulial ar, ae 7, i wl @ os thet ieya lbh Ailoatiton q any: rus! > , pais fl pay never i A % bt pate oy balgy he bo mLALCTR ef GS Bato TALE en a ae hata pense ol if aay Nog) al wet? Lieb . TAD Miia Rios lo Th, 8 on pouch becca etd fe evi thing: at ’ vt oe oe ee sb 143 | ne on , a . mo , : yy) tJ vars | ‘ / Ee ¥ 1 Se t AY ai rl! 4) folk wa itp y' 2 aj veo i , Me Sees 3 . Sas "eat i stk We im = ‘ en EXPLANATION OF PLATE XII. Halimococcus lampas, Ckll., dorsal view ; «a, lateral view; b, stigmatic cleft; c, antenna Ob 2, Baccacoccus elytropappi, sp.n. ; a, antenna ; J, leg i.; ¢, stigmatic cleft; d, anal lobes of 9. Tachardia minor, sp.n., lac-gland plate of 9. . decorella (Mask.), spine; a, anal plate of 9. actinella, Ckll., spine of 9. karroo, sp. n., spine; a, lac-gland plate of 9. affluens, sp. n., spme of 9. decore la (Mask.), lac-gland plate of °. PEATE) XI, Wile, 2X, Jena 2. BuLc. ENT. RESEARCH. South African Coecide. ary ne ae: ry oo ne ——— a_i ~ apy Ww re Ni e 4) ns Pi. ¢ ? \ ’ ‘ 129 MOSQUITO NOTES. By F. W. Epwarps. Anopheles hyrcanus, Pallas. Culex hyrcanus, Pallas, Reise durch Russischen Reichs, }, p. £75 (1771). Anopheles sinensis, Wiedemann, Auss. Zweitl. Ins. i, p. 547 (1828). Anopheles pictus, Loew, Dipt. Beitr. 1, p. 4 (1845). Anopheles pseudopictus, Grassi, Atti R. Ac. Lincei, vii, 1, pp. 102 & 560 (1899). The salient points in Pallas’s brief description are as follows :—“‘Cinereus . Frons hirtella. Antennae triarticulatae ; extremum trinode, basis vero pilis nigris subpennata. Palmi [? palpi] nulli: ensis nudus . . . basi pilosus. Pedes longissimi, grisei, posticis corporis fere tripla longitudine, praesertim tarsi, elongati. Alae lanceolatae, cinerascentes, venis hirsutis ad crassiorem marginem nigro maculatae Comes prioris [C. caspius, ? =O. dorsalis, Mg.], rarior red ferocior.” If we make the highly probable assumption that Pallas described the palpi as the antennae, and that the latter were either broken off or overlooked in his specimen, the above description becomes understandable, since the sentence referring to the “antennae ” might very well be used to express the appearance under a low-power lens of the shaggy palpi of an Anopheles of the Myzorhynchus or Nyssorhynchus groups. Although no collections of mosquitos have been made since Pallas’s time in the North Caspian region, it is unlikely that any species, other than A. sinensis, to which the description could possibly apply, occurs there. The name hyrcanus will therefore have to be substituted for senensis. | Anopheles subpictus, Grassi. Anopheles subpictus, Grassi, Atti R. Acc. Lincei, vin, 1, p. 101 (Feb. 1899). Anopheles rossi, Giles, J. Trop. Med. 11, p. 63 (Oct. 1899). Grassi’s short description clearly indicates A. rossi, and the fact that he states that his description was drawn up from a specimen sent him by Ross from Calcutta removes any possible doubt which might exist. Itis to be regretted that the operation of the rule of priority will prevent the commemoration in zoological nomenclature of Sir Ronald Ross’s work. Stegomyia fasciata* var. atritarsis, nov. Tarsi of fore and mid legs almost entirely black, two or three white scales at the © bases of the first two joints ; hind tarsi black, with very narrow white rings (narrower than the diameter) at the bases of the first three and the last joints. White rings on the male palpi narrower than usual. Markings of thorax, abdomen and femora, and structure of male genitalia, normal. Gotp Coast: Accra, vi. 1919,1g 192 (Dr. J. W. Scott Macfie). Perhaps the most remarkable colour variation yet recorded in this or any mosquito. * Dr. Guy Marshall urges the retention of this name, rather than the adoption of the earlier argenteus, Poiret, owing to its wide use in medical literature ; a course which has been sanctioned in afew cases by the International Commission on Zoological Nomencla- ture. With this suggestion I readily concur. The names Stegomyia and Ochlerotatus though used here as genera are to be understood in a subgeneric sense. (605) D2 130 F. W. EDWARDS. Ochlerotatus caspius, Pallas. Culex caspius, Pallas, Reise d. Russ. Reichs, i, p. 475 (1771). Culeaz dorsalis, Theobald, Mon. Cul. ii, p. 16 (1901) (2 nec Meigen). Grabhamia dorsalis, Theobald, Mon. Cul. 11, p. 251 (1903). Grabhamia subtilis, Kd. & Et. Sergent, Bull. Mus. Paris, xi, p. 240 (1905). Grabhamia willcockst, Theobald, Mon. Cul. iv, p. 294 (1907). Grabhamia longisquamosa, Theobald, Ann. Mus. Nat. Hung., i, p. 102 (1905). Mansonia arabica, Giles, J. Trop. Med. p. 130 (1906). Culex arabicus, Becker, Denkschr. k. Ak. Wiss. Wien, Ixxi, p. 140 (1910). ? Culex maculwentris, Macquart, Dipt. Exot. Supp. 1, p. 7 (1846). Culex pulchripalpis, Theobald, Mon. Cul. 1, p. 13 (1901) (nec Rondani). Culex penicillaris, Rondani, Bull. Soc. Ent. Ital. iv, p. 31' (1872) ; Ficalbi, Bull. Soc. Ent. Ital. xxxi, p. 160 (1869). } Recent discoveries (see below, under O. currier) make it appear doubtful whether Meigen’s C. dorsalis has been correctly determined by Theobald and others, but: as it is an abundant species over a wide area of Hurope and Asia it is desirable to find a name for it which is not likely to be upset, and I believe that C. caspius, Pallas, is such a name. Pallas’ original description is of course short and vague, but as he states that the species was abundant and very vicious in the marshes near Guriev (North Caspian) it is probable that he met with one of the common European salt- marsh breeders. The salient phrases in the description, “ Color subgriseus, thorace cinereo-fasciato . . . pedes subannulati” would seem to indicate the present species rather than O. currer, while definitely excluding O. salinus. Some confirm- ation of this conclusion has recently been obtained by Capt. P. A. Buxton’s discovery of O. dorsalis, Theo., at Resht, South Caspian. Pallas’ statement that the antennae are ‘‘ utrique sexu filiformes ” may be safely disregarded ; and whatever “ vaginae multae ” may mean, Theobald’s translation of “ sheath of proboscis snowy white ”” has obviously no justification whatever. Large series of this species have recently been received at the British Museum from Italy, Macedonia, Palestine, Egypt and Mesopotamia, which show every gradation in the variation in the colour of the abdominal scales. In some (as in most British specimens) the prevailing colour of the abdomen is black ; in the majority the abdomen has pairs of black spots on each segment, or on the anterior segments; while in the palest the black is entirely absent, leaving only the ochreous or whitish ochreous ground-colour. The thoracic coloration varies less ; the two white longitudinal lines are nearly always present, though sometimes very narrow ;. in two specimens from Italy, and in the type of G. longisquamosa, they are absent, the mesonotum then being uniformly fawn-coloured. The amount of dark scaling on the wing varies very considerably, but the dark scales are always fairly evenly distributed. No variation is discernible in the male genitalia. The comb-scales of the larvae may or may not have a differentiated terminal spine, a variation which has also been noted Ly Dyar in O. currier. Ochlerotatus caspius var. hargreavesi, nov. Among a series sent from Taranto, Italy (H. Hargreaves), are six females which have the whole of the central area of the mesonotum covered with whitish scales, MOSQUITO NOTES. 131 though in regard to abdominal markings they agree with moderately dark specimens of the species. Nothing approaching this variation has been seen from elsewhere, and it therefore seems justifiable to distinguish them under a separate name. It will be interesting to find whether any difference exists in the male. Ochlerotatus curriei, Coq. Culex curriei, Coquillett, Can. Ent. xxxiil, p. 259 (1901). Culex onondagensis, Felt, N. Y. State Mus. Bull. 79, p. 278 (1904). Aédes curriet, Dyar, Insec. Inscitiae Menst. v, p. 122 (1917). ? Culex punctatus, Meigen, Klass. 1, p. 6 (1804). ? Culex dorsalis, Meigen, Syst. Beschr. vi, p. 242 (1830). * Grabhamia broquettu, Theobald, Entomologist, xlvi, p. 179 (1913). A small series recently received from Wareham Heath, Dorset (H. Scott), proves on close examination to represent a species distinct from the one we have usually known as dorsalis, though this latter was taken in company with it and O. salinus, Fic. The two forms differ as follows :— O. caspwus, Pallas (dorsalis, auct.). Scales of mesonotum ochreous brown or fawn-coloured, with two narrow white lines running the whole length. Scales of proboscis mostly pale except towards base and tip. Abdominal scales usually of three colours, black, white and ochreous. Wing scales mostly rather broad, dark ones being evenly spread over all the veins, though in varying proportions with the light ones. Dark scales more numerous on the femora. Apical part of basal lobe of side-piece of male genitalia not prominent. O. curriei, Coq. Scales of mesonotum mostly greyish ochreous ; a dark brown band of varying width in the middle, extending only about half way from the front ; patches of the same colour on each side in front. Scales of proboscis mostly black. Abdominal scales of two colours only, black and greyish white. Wing scales mostly quite narrow ; at the base of the wing they are all pale ; dark ones predominate on the apical half of the costa, on the first longitudinal vein, and on the forks of the fourth, while on the third and fifth longitudinal veins the scales are almost all dark ; elsewhere they are mostly pale. Femora, except towards tips, almost. all pale. Apical part of basal lobe of male genitalia prominent. Meigen’s descriptions of C. punctatus and C. dorsalis do not apply very well either to O. caspius or to O. curried, though they almost certainly must refer to one or the other, and perhaps with more probability to the latter. An examination of the types would be essential to settle the point. The British examples of O. curriet only differ from North American specimens in having the dark scales of the wings rather more numerous, especially on the costa. The male genitalia are identical. Ochlerotatus rusticus, Rossi. _ Culex rusticus, Rossi, Fauna Etrusca, 1, p. 333 (1790). Culex pungens, Robineau-Desvoidy, Mém. Soc. Hist. Nat. Paris, 11, p. 407 (1827). Culex quadratumaculatus, Macquart, Suit. 4 Buffon, 1, p. 34 (1834). Culex dwersus, Theobald, Mon. Cul. 1, p. 73 (1901). Culex quadratumaculatus, Villeneuve, Bull. Soc. Ent. France, p. 55 (1919). 132 F. W. EDWARDS. On a careful comparison of fresh specimens with Robineau-Desvoidy’s description I entirely agree with Dr. Villeneuve in his identification. He has however overlooked the fact that the name quadratimaculatus was an unwarranted substitution by Macquart for Desvoidy’s pungens ; moreover I consider it is equally evident from Rossi’s description that his C. rusticus refers to the more strongly marked form of dwersus with a continuous median dorsal yellow line. This being the oldest name must replace all the others. Ochlerotatus antipodeus, sp. nov. 2. Head with narrow golden scales in the middle ; on each side a patch of narrow black ones, and external to these a small patch of flat white ones ; upright scales and bristles blackish, except for a pair of golden bristles directed forwards over the eyes. Proboscis and palpi black-scaled, the latter about one-seventh as long as the former. Thorax dark brown; mesonotum with blackish bristles and dark reddish-brown scales ; narrow golden scales arranged in five rather definite lines : one median, extending from the front margin to just before the scutellum, where it forks ; a subdorsal pair on the posterior half of the mesonotum ; and a lateral pair on the anterior third ; some scattered narrow golden scales about the middle, and a patch of the same above the bases of the wings. Prothoracic lobes with a few golden scales ; area behind them (proepimerum) with small flat blackish-brown ones. Mid lobe of scutellum with narrow golden scales in the middle, narrow dark brown ones laterally ; lateral lobes with a few narrow dark brown scales. Pleurae with patches of flat white scales. Abdomen clothed mainly with blackish brown scales ; small white lateral basal patches on each segment and narrow yellowish white basal bands on segments 2-5. Cerci elongate, blackish. Legs black-scaled ; femora whitish beneath ; narrow white rings on the bases of the first three tarsal joints, broadest on the hind legs, where there are also a few white scales at the bases of the last two jomts. Claw formula 1.1:1.1:0.0. Wangs with blackish scales, those in the lateral series almost linear. First fork-cell fully twice as long as its stem, its base a little nearer the wing-base than that of the second. Cross-veins separated by quite twice the length of the posterior one. Halteres yellowish with brown knob. Length (without proboscis), 5 mm. New ZeauanD: Karikari Bay, 5.1.1916, 149 (incl. type) (Albert E. Brookes) ; Te Horo, near Wellington, 39 (received through R. F. L. Burton). Ochlerotatus lepidonotus, sp. nov. 6S. Head with a small area of yellowish narrow curved scales in the middle, mixed yellow and black upright forked scales on each side of these, and whitish flat scales more outwardly. Proboscis black. Palp: longer than the proboscis by the last joint. Long joint mainly pale scaled, but also with many black scales which are specially numerous towards the base and apex; its apical fourth much swollen, with long dense hairs on its outer side, which are orange basally, black apically. Penultimate joint about the size of the swollen portion of the long joint ; with whitish scales basally, black-scaled apically ; internally and ventrally with long dense orange hairs with black tips ; a few shorter stiff black hairs on the outer side MOSQUITO NOTES. — 133 at the apex. Terminal joint slightly longer and distinctly thicker than the penultimate ; with whitish scales at the base, otherwise black, black-haired. Thorax with black integument ; mesonotum with pale yellowish scales and bristles (denuded in middle). Pleurae (except sternopleura) densely clothed with flat whitish scales. Postnotum with a tuft of pale yellowish narrow curved scales apically. Abdomen clothed mainly with pale greyish ochreous scales, with some blackish brown ones intermixed to a varying extent; in the darkest specimen the pale scales are confined to rather narrow bands across the bases and apices of the segments, the remainder being all black ; venter mainly whitish. Genitalia : side-pieces less than four times as long as broad, with rather dense dark brown hairs on the apical half, but none overhanging the harpagones; external margin. slightly curved; internal practically straight, except for an indentation just posterior to the basal lobes.. Basal lobes very small, with a tuft of hairs, none differentiated ; apical lobes barely distinguishable, with a few short hairs. Claspers with a straight and rather stout terminal spine. Harpagones just over half as long as the side-pieces, the stems pubescent, four times as long as the appendage, which is rather narrow, slightly curved and frayed at the tip. Harpes undivided, a little shorter than the stem of the harpagones. Legs: femora and tibiae mainly with pale ochreous scales except towards the tips, where the scales are black. Tarsi black, except for numerous pale scales towards the base of the first joint. Claw formula 2.1:2.1:1.1. Wangs very scantily scaled; the scales brown, except for those on the subcosta, and the base of the costa and first, second, fourth, fifth and sixth longitudinal veins, which are pale yellowish. Fork cells about as long as their stems, the upper with its base much beyond that of the lower. . Palpi nearly a third as long as the proboscis, black, with scattered whitish scales. Scarcely any black scales on the abdomen. Claw formula 1.1:1.1: 1.1. Wing-scales rather more numerous, and mostly pale yellowish, except those on the costa and first longitudinal vein. Length (without proboscis), 7-8°5 mm. Maceponia: 43 (including type), 10 9, swept in meadow near Galiko R., Salonica, 26. iv. 1918 (Capt. J. Waterston). A very distinct species, with one remarkable characteristic, the possession of scales on the postnotum. In some respects, notably the form of the male palpi, it resembles O. rustocus, which, it is interesting to note, was taken at the same time and place. Ochierotatus (Finlaya) echinus, sp. nov. Ochlerotatus lateralis, Edwards, Bull. Ent. Res. ii, p. 250 (1911) (nec Meigen). Ochlerotatus ornatus, Hdwards, Bull. Ent. Res. iii, p. 21 (1912) (? nec Meigen). Ochlerotatus geniculatus, Edwards, Bull. Ent. Res. iv, p. 49 (1913) (nec Olivier). Difiers as follows from the common. O. geniculatus :—The four dark stripes of the mesonotum are brown instead of black, and are separated by narrow golden, instead of whitish hnes; the sides of the mesonotum are almost pure white, instead of yellowish white ; the scutellar scales are broad, flat and white; the abdominal seg- ments have narrow basal yellowish bands in addition to the white lateral spots ;. and the hind femora have a dark dorsal line on the basal half. 134 F. W. EDWARDS. Maceponi4 : Stavros, near Salonica, 1918, 19 (type) reared from larva (Capt. J. Waterston). Morocco: 19 caught on horse and mule outside Fez, v. 1909 (Major C. E. P. Fowler). Aucreria: 19 (Dr. Sergent; no data). In the absence of a male, the best distinctive character which can be adduced is the presence of flat scales on the scutellum; the larva also is very different from that of O. geniculatus, hence this form must undoubtedly be regarded as specifically distinct. The Moroccan and Algerian specimens have the flat scutellar scales, but in thoracic markings appear to agree with typical O. geniculatus; hence they are somewhat doubtfully conspecific with the type. Culex apicalis, Adams. Culex apicalis, Adams, Kans. Univ. Sci. Bull. ii, p. 26 (1903). Culex territans, Howard, Dyar & Knab, Monogr. iti, p. 293 (1915) (nec Walker). Culex territans, Schneider, Verh. Natf. Ver. Bonn, lxx, p. 45 (1913). Culex hortensis, Edwards, Ent. Mo. Mag. (3) i, p. 167 (1915) (nec Ficalbi). Culex saxatilis, Dyar, Insec. Inscitiae Menstr. vii, p. 36 (1919). Culex pyrenarcus, Brolemann, Ann. Soc. Ent. France, 1918, p. 427 (1919). Specimens of C. pyrenaicus sent me by Dr. Villeneuve from ‘Rambouillet, France, agree in almost every detail with North American specimens in the British Museum, the only distinction discernible being that the harpagones of the male genitalia are more noticeably serrated on their tips. This difference is so minute that I con- sider the specific identity of the European and North American forms to be unquestionable, particularly in view of the fact that the male genitalia vary slightly among American specimens. The species can be readily separated in both sexes from C. hortensis, Fic., by the white spots at the tips of the hind femora and tibiae being very small or absent ; the wing scales also are perceptibly broader. Capt. J. Waterston has recently discovered the larvae of both C. apicalis and C. hortensis in the neighbourhood of Salonica, and from an examination of his material it can be positively stated that the larva figured by Schneider is that of C. apicalis and not of C. hortensis as I suggested in 1915. Dyar, in the paper quoted, discusses the synonymy of the species, and from the evidence he adduces I should certainly conclude that C. apicalis is the correct name for the species, C.. testaceus, v. d. Wulp, being too doubtful to be made use of. Culex aurantapex, Edw. Culex aurantapex, Edwards, Bull. Ent. Res. v, p. 74 (1914). Taenwrhynchus domesticus, Leicester, Cul. of Malaya, p. 169 (1908) (nec Culex domesticus, Germat). I described this species from a single female from Nairobi. Recently a female and three males reared from larvae have been received from Dar-es-Salaam (A. W. J. Pomeroy). Although these specimens differ from the type in having slightly broader wing-scales and more numerous pale scales on the wings and legs, there seems little reason to doubt their specific identity with the Nairobi example. They appear to agree in every respect with specimens of Leicester’s T. domesticus. As Leicester MOSQUITO NOTES. 135 points out, the wing scales are decidedly narrower than in C. bitaeniorhynchus, a difference which, together with a slight distinction in the male genitalia, will probably suffice to distinguish the two forms specifically. The genitalia are rather peculiar (fig. 1). The apical lobe of the side-pieces has the flat plate much reduced, little more than a flattened bristle ; in addition there are four or five undifferentiated bristles and two stout spines. The harpes have only a minute basal appendage. The “ harpagones”’ are divided into two pairs of strong Fig. 1. Male genitalia of Culex auraniapex, Edw. ; a, apex of side pieces ; b, harpes and harpagones, side aes “ the ame, dorsal view (one side sickle-shaped plates directed dorsally, the ventral pair is the longer, and gives off ventrally from its base an appendage which divides into a rounded knob and two strong teeth. In the typical C. bitaeniorhynchus of the Oriental region, as well as in the var. ethiopicus, Kdw., this ventral appendage of the second harpagonal plate is much reduced and differently formed. The var. ethiopicus differs from the typical bitaenrorhynchus chiefly in having the upper (or basal) sickle-shaped plate of the harpagones much broader than the lower (or apical). The other points of distinction given (Bull. Ent. Res. ii, p. 30, 1912) do not hold good. Specimens of this form were also reared by Mr. Pomeroy at Dar-es-Salaam. The larvae seem to be indistinguishable. Culex watti, sp. nov. g. Head scales mostly whitish. Palpr longer than the proboscis by nearly the length of the last two joints, which are densely hairy. Long joint with a narrow whitish ring before the middle, and another broader one in the middle of the apical half. Last two jomts with narrow pale basal rings, creamy above, white below ; hairs at tip of last jomt also pale; penultimate joint with a short white streak on underside just beyond middle. Proboscis brown, with a narrow, distinct though not sharply. margined pale ring beyond the middle. Thorax brown-scaled, without special ornamentation, except that the scales on and in front of the scutellum and in front of the wing-bases are paler. Abdomen dark brown, the segments with dull whitish basal bands of about even width, those on segments 6 and 7 considerably expanded laterally. Genitaiwa (four specimens mounted): side-pieces normal, rather closely resembling those of C. trifilatus, Edw. (Bull. Ent. Res. v, p. 64), but the clasper is a little more narrowed apically, and on the lobe the modified bristles in each set of three (anterior to the plate) are of about equal length; the one adjoining the plate has a more flattened tip. Uncidistinct, pomted. Harpagones 136 F. W. EDWARDS. with only two incompletely separated divisions, the first sickle-shaped, the point directed ventrally and touching the second, which is broad and flattened horizontally. Harpes with the usual crown of spines, but without any trace of a basal thumb-like projection ; their sides straight and parallel. Legs brown ; femora whitish beneath almost to the tips; a distinct pale spot at apex of hind tibiae, which are also obscurely pale along the inner side ; the faintest suggestion of pale rings at the tarsal articu- lations, most distinct at the junction of the first and second hind tarsal joints. Wings with linear-lanceolate brown scales; bases of fork cells practically level. Q. Resembles the male, but the pale band of the proboscis is much more evident on the underside than above, and is much broader and very ill-defined basally. Middle tibiae in one specimen with a very indistinct pale longitudinal stripe in front. Length (without proboscis), 5 mm. GOLD Coast: 4¢ (including type) 29, Kumasi, 1919 (Dr. W. G. Wait). Hast Arrica: 4g 29, Dar-es-Salaam, 30.1.1918 (A. W. J. Pomeroy). The most obvious characters of this species are the banded proboscis and palpi of the male, together with the practically unbanded legs. The specimens agree in most respects with Theobald’s description of C. ataenata ; I have however examined the type in the Liverpool School of Tropical Medicine, and consider it to be only C’. univittatus, with which indeed the present species might easily be confused. The genitalia are most distinct, and clearly show that OC. watti is more nearly related to C. duttont than to any other African species ; O. duttoni differs chiefly in having the harpagones undivided, elongate, and bent at right angles in the middle. In adult coloration C. duttoni differs obviously in the ringed tarsi and striped tibiae. Theobaldia arctica, sp. nov. §. Differs from 7. annulata, Schrank, as follows :—Palpi rather more slender and darker, both hairs and scales on the last two joints being almost all black. Femora without any trace of pre-apical pale rings; hind tibiae without pre-apical pale patch on the outer side, but with a narrow longitudinal pale streak along the inner side ; first tarsal joint of all the legs almost entirely black, only a very few white scales at the base. Genitalia: seventh sternite pomted, with an irregular row of ten stout close-set spines at the tip. Basal lobes of side-pieces small, with two stout bent setae and a number of hairs. Side-pieces also with a distinct pre-apical prominence on the inner side which is densely hairy. Harpes rather slender, more swollen at the tip, where they have four or five small serrations. Harpagones* rather slender, a little shorter than the harpes, their tips sharply pomted and hooked dorsally. Unci small, pointed, half as long as the harpagones. N. Russia: 1g, Archangel, 25.1x.1918 (Capt. A. G. Carment). In coloration and genital structure this insect agrees almost entirely with T. alaskaensis, Ludlow, and may in fact be the same. Judging from Dyar and Knab’s * These are the structures called ‘‘unci’’ by Howard, Dyar & Knab, but as they are undoubtedly homologous with the “‘ harpagones’”’ of Culex I use this term for them. The same names are used in quite different senses in different genera of CULICIDAE, and the terminology of the parts of the male genitalia of these and other Diptera is badly in need of revision and unification. MOSQUITO NOTES. LST figure, however, the harpes are differently toothed and the subapical prominences and hair-tufts of the side-pieces are more strongly developed than in alaskaensis. The genitalia of 7. annulata differ in the entire absence of stout spines on the seventh sternite and of any preapical prominence or hair-tuft on the side-pieces, and in the differently toothed harpes and less distinctly hooked harpagones. Theobaldia indica, sp. nov. Coloration entirely as in T. arctica. Genitaha : seventh sternite with a group of about seven short stout spines on its pointed tip. Side-pieces over three times as long as broad ; basal lobes and apical prominences practically as in 7. arctica. Harpes very slender, very little expanded apically, the tip being slightly cleft into two, but otherwise untoothed. Harpagones slightly longer than the harpes and very broad, nearly halt as broad (vertically) as they are long; a small sharp, backwardly directed hook on the dorsal margin at some distance before the tip, which is rounded. PunsaB, Inpia: 1g (type), Bakloh, 28.11.1900 (Capt. Lindesay) ; 19, Umballa, 9.111.1905, and 19, Dalhousie, 4.v.1906 (Col. H. J. W. Barrow). The specimen recorded by Giles (Gnats, ed. i, p. 206) from Bakloh is probably this species ; the male recorded by Theobald (Mon. Cul. i, p. 335) is the type described above. No specimens of the true 7. annulata from India are in the British Museum collection. Pe A 3 PRN NaS bev Pe ee ie ac ng jenny j | zy | | EA ’ Rain tal ie . ra pm : Rp’ . 4 ue : ie Mi iy a see vin rsatron mr wi migrtall, ‘ aaa mal tps 2 ey, fae hey in ee i | hala a aginey ms be ‘ pre etd ra! ‘tet e a) svi wis a bait penne take Ain Nh ik SN ee re elie, i i yi ie i: Rye Mae: ti aaa Rowe POC manet bonne at Cabhidst @ a aie Bi nh is te iy | ae ae Ap Weg i 2 TREY i TA Wie ire Shale: tin bi ae wis 6 i il asin vi Metal ¢ i Me mie 3 ra | nih aK i) ng ae ° . ‘ DAEs Re aan ve he in ua pea ‘yee Hit, oe ae es Waa Vie 5 i, See Sone. | Galt ee ae ae ‘i, Maley f ; ail ia Pee Leer ot in! MS da apa ire ig sia Lis ts AM " mow pon fetid TaN SR ON Ve AR, ROT Raa Oe aie _ebctiee ar awa ot! eC ah A AR ine cc nc inane ‘iid ita uate aD. ee ey Ch AS tig i i ee rg yt yyy s a I te | “ig a " er me H ye i i a hit a ny diay. co rp 7 139 ON SOME AFRICAN DELPHACIDAE (HOMOPTERA). By F. Murr. Very little is known about the African DELPHACcIDAE, especially the tropical forms. For this reason it is worth recording the few species represented in a small collection made by Mr. J. C. Bridwell at Oloke-Meji, Ibadan, Nigeria, in 1914. All the specimens were swept from grasses and reeds. It is to be hoped that larger collections will soon be available and that those who work at them will examine and give figures or adequate descriptions of the genitalia, especially of the aedeagus. Owing to the wide distribution of some species and the great difference in colour of some of the macropterous and brachypterous forms, and to the variation in colour of other species, and the paucity of good specific distinctions in the chrodtic characters, it is almost impossible to identify many species of the Delphacides group without making use of the phallic characters. The types of the new species are deposited in the collection of the Hawaiian Sugar Planters’ Association, Honolulu, Hawaii. Measurements are from the apex of vertex to anus and from the base to apex of one tegmen. Tropidocephalus brunnipennis, Sign. One male and one female from Oloke-Meji. The male genitalia of this specimen are identical with those of specimens from Australia (7. eximeus, Kirk.) Perkinsiella sp. One female, which is closely allied to P. bakeri, Muir, from the Philippine Islands, but I fully expect the male to be distinct. This genus in the Pacific is always attached to sugar-cane. It would be of interest to know if this is so in Africa, and to ascertain what parasites keep them in check. Delphacodes disonymos (Kirk). Delphax disonymos, Kirkaldy (1907), Haw. Sugar Planters’ Assn. Ent. Bull. i, p. 151, line 10, p. 155, Pl. xviii, fig. 4. Delphacodes disonymos (Kirk.), Muir (1917), Proc. Haw. Ent. Soc. iu, 4, p. 333. Delphax matamtu, Kirkaldy (1907), op. cit. p. 151, line 13, p. 155, Pl. xvi, figs. 4,5, Pl. xvi. fig. 15. Delphacodes matantu (Kirk.), Muir (1917), op. cot. p. 333; Muir (1918), op. cit. ii, p. 427, fig. 3. | I consider these two species of Kirkaldy’s as being the long and short winged forms of the same species. His two figures of the pygofers are drawn from different positions and so appear slightly different ; the aedeagi are the same. A similar case of difference in colour of the tegmina is found, I believe, in D. erectus nigripennis (Crawiord), which I consider to be the brachypterous form of D. erectus (Crawford). 146 F. MUIR. Deiphacodes nigeriensis, sp. nov. (fig. 1). G. Macropterous; head slightly narrower than pronotum, vertex as long as base is wide, which is slightly broader than apex, base slightly behind the middle of eyes ; first joint of antennae more than half the length of second (1 to1‘7); lateral carinae of pronotum slightly divergingly curved, not reaching to the hind margin; first joint of hind tarsus equal to the other two together. zs fee rate EN | 2205 ms > = Fig. 1. Dicranotropis nigeriensis, Muir, sp. n.; a, aedeagus, lateral view; b, right genital style. Face and clypeus between carinae black, carinae light brown; antennae, vertex, thorax and legs ochraceous tawny, abdomen dark brown. Tegmina hyaline, very light yellowish, veins darker, with sparse minute granules. Pygofer similar in outline to that of D. disonymus (Kirk.), and the genital styles somewhat similar, aedeagus distinct, figured from the right side (figure 1,a); the left side with a row of small spines from a dorso-apical point to the middle; anal styles not so large as in D. disonymus, curved and slightly flattened, bases moderately wide apart. Length, 1:8 mm. ; tegmen, 2 mm. NicErta : Oloke-Meji, Ibadan, 1914 (J. C. Bridwell). Described from one male specimen. I consider this to come near to D. disonymos (Kirk.), and it is very probable that there is a short-winged form in which the tegmina are dark shiny black. Delphacodes bridwelli, sp. nov. (fig. 2). 3. Macropterous; head nearly as broad as pronotum, width (including eyes) twice the length ; vertex square, base about middle of eye; length of face slightly more than twice the width, slightly narrowed between eyes and at apex; antennae reaching slightly beyond the base of clypeus, first joint about half the length of Ge 20. = er Fig. 2. elphacodes bridwelli, Muir, sp. n.; a, aedeagus, lateral view; b, right genital style. second ; lateral pronotal carinae divergingly curved, not reaching hind margin ; hind tibiae equal in length to tarsi, first joint of tarsus equal to the other two together. light brown; black or dark brown between carinae of face and clypeus and over thoracic pleura, coxae and abdomen. Tegmina hyaline, very light brown, veins darker ; wings hyaline, veins brown. ON SOME AFRICAN DELPHACIDAE. 141 Opening of pygofer about as broad as long, dorsal emargination large, anal angles well produced, slightly curved ; genital styles straight, outer edge slightly sinuous, inner edge produced into a quadrate process near base, apex truncate ; aedeagus flattened laterally, broad, apex rounded, an irregular row of spines from near apex on dorsal aspect across left side to near the medio-ventral area, a somewhat similar row on right side ; anal spines large, curved, their bases near together. Length, 2 mm. ; tegmen, 2°2 mm. NicErta: Oloke-Meji (J. C. Bradwell). Described from one male. Dicranotropis bridwelli, sp. nov. (fig. 3). g. Macropterous; head as broad as prothorax ; width twice the length, including eyes ; vertex square; length of face nearly twice the width (1 to 1-8), narrowed between eyes and slightly at apex, median carina furcate about middle; length of the first joint of antennae slightly more than half the second, terete. Length of first joint of hind tarsus equal to the other two together; spur as long as first joint of tarsus, thin, concavo-convex, with many small teeth on hind margin. Fig. 3. Dicranotropis bridwelli Muir, sp. n.; a, pygofer, back view; b, do., side view. Dark chestnut-brown or black; antennae, carinae on head and thorax, rostrum and legs lighter brown ; base of dorsum of abdomen and middle of apical tergite hght. Tegmina hyaline, fuscous, a clearer hyaline area over subcostal, radial and median apical cells, veins of the same colour as membrane, granules small, bearing small black hairs ; wings hyaline with brown veins. Pygofer as figured (fig. 3). Length, 2°9 mm. ; tegmen, 3°8 mm. NicErRIA : Oloke-Meji, Ibadan, 1914, on coarse grass (J. C. Bridwell). Described from one male specimen. There are two female specimens in the collection which may be this species ; they are much lighter, being nearly uniformly light brown. Dicranotropis ibadanensis, sp. nov. (fig. 4). 3S. Macropterous; head as wide as pronotum, width 2:2 times the length including eyes ; vertex square or very slightly wider than long, base slightly behind middle of eye; length of face twice the width, sides slightly curved, narrowest 142 F. MUIR. between eyes and at apex, furcation of median carina one-third from base ; antennae reaching to near middle of clypeus, first joint slightly shorter than second ; first joint of hind tarsus as long as the other two together, spur slightly shorter than first joint of tarsus, with many small teeth on hind margin. a / : / oe \ = oe ple aie! a 4 ¥ Fig. 4. Dicranotropis ibadanrensis, Muir, sp. n. a, pygofer, back view; 6b, do., side view. Light brown, nearly black over clypeus, coxae and abdomen ; carinae of thorax lighter, traces of lighter spots on face. Tegmina hyaline, a curved, fuscous mark from cross-veins over cubitus extending along veins to the apex of media and cubitus, veins basad of cross-veins yellowish, distad of cross-veins darker, granules few and small; wings hyaline with brown veins. Pygofer as figured (fig. 4). Length, 1‘°9 mm. ; tegmen, 2.8 mm. NicEria: Oleke-Meji, Ibadan, 1914 (J. C. Bradwell). Described from one male. Genus Megamelus, Hieber. There are a number of closely allied species of this genus having similar facies, with a wide distribution. Only a close comparison of series from various localities will enable one to decide where one species ends and another begins. I sank M. kolophon (Kirk.) into M. furcifer (Horv.), but I believe it will be better to retain the name as a variety until more material has been studied. I now recognise three varieties of this species, viz. :— 1. M. furcifer (Horv.) (fig. 5). The fork at apex of the genital styles small ; clavus with a small black mark at apex. Fig. 5. Megamelus furcifer, Horv., (eee style. 6. ie kolophon, Kirk ae # Tis Ps nigertensis, Muir, var. n.; a, aedeagus ; b, genital style. ON SOME AFRICAN DELPHACIDAE. 143 2. M. kolophon (Kirk.) (fig. 6). Fork at the apex of the genital styles much larger ; no black mark at the apex of clavus ; aedeagus slightly smaller, with smaller spines. 3. M. nigeriensis, var. nov. (figs 7a, 7b). Fork at apex of genital styles wider and the emargination between them shallower ; a dark mark at the apex of clavus. These forms come near to Kelisia kirkaldyx, Muir, and Kelisia fieberi, Muir, and eventually will be placed together in the same genus. Until I can revise these genera I prefer to let them remain as they are for the present. Megamelus furcifer, var. nigeriensis, nov. (figs. 7 a, 7 5). gS. Macropterous; length of head, prothorax and mesothorax about 1°5 times: the width of the head including eyes, vertex shghtly longer than wide ; length of face 2°5 times the width, narrowest between eyes, furcation of median carina at extreme base ; antennae terete, reaching beyond the base of clypeus, second joint 1-4 times the length of first ; first joint of hind tarsus as long as the other two together, spur longer than first joint of tarsus, large, with many small teeth on hind margin. Carinae of face and clypeus, vertex, middle and lateral portions of pronotum, middle of mesonotum, tegulae and base of abdomen yellow; face and clypeus between carinae, medio-lateral portions of pronotum, lateral portions of mesonotum. and greater portion of abdomen black ; front and middle legs dark brown, hind legs lighter brown. Tegmina hyaline, a dark mark at end of clavus and fuscous over apical veins. The pygofer is similar to that of M. kolophon, Kirk.,* but the genital styles are wider at the apex and not so deeply indented, the aedeagus is slightly larger and the spines stronger. Length, 2 mm. ; tegmen, 2°3 mm. Niceria : Oloke-Meji, Ibadan, 1914 (J. C. Bridwell). Described from two males. Megamelus flavolineatus, sp. nov. (fig. 8). 3. Macropterous; width of head including eyes 1°6 times the length, nearly as wide as the pronotum ; length of vertex 1-4 times the width ; antennae reaching beyond the base of clypeus, first Joint more than half the length of the second. ms 8b. oe a \° Fig. 8. Megamelus flavolineatus, Muir, sp. n.; a, aedeagus; b, anal spines ; ¢, genital style. (1 to 1°7); length of face 2°5 times the width, shghtly narrowed between eyes,. sides subparallel, median carina forked at middle of face ; first joint of hind tarsus. as long as the other two together, spur nearly as long as the first joint of tarsus, large, thin, with many small teeth on the hind margin. * Haw. Sugar Planters Assn. Ent. Bull. 1, pl. xv, figs. 9-10. (C605) E 144 F. MUIR. Dark brown or black; vertex, carinae of head, middle and lateral margins of pronotum, middle of mesonotum, tegulae, antennae and legs yellow or light brown. Tegmina hyaline, veins yellow or light brown, a dark mark at apex of clavus, fuscous over apical portion of cubital veins. Aedeagus short, deep at base where it is laterally compressed, a row of three or four spines on the right side and another somewhat similar row on the left; anal spines large, shghtly curved and pointed ; genital style as figured (fig. 8, c). Length, 1-8 mm. ; tegmen, 2°5 mm. . Macropterous ; similar in colour to the male. Length, 2°0 mm. ; tegmen, 3°0 mm. Nigeria: Oloke-Meji, Ibadan, 1914 (J. C. Bridwell). Described from six males and eleven females. Some of the specimens are lighter in colour than the type, especially between the carinae of clypeus and genae. Phyltlodinus badius, sp. nov. ©. Macropterous; head as wide as pronotum, width 2°4 times the length; vertex wider than long; length of face double the width, narrowest between eyes and at apex, median carina furcate slightly distad of middle; antennae terete, reaching shghtly beyond the base of clypeus, first joint slightly shorter than second ; femora and tibiae of first and second legs and femora of hind legs flattened and expanded ; lateral pronotal carinae divergingly curved, not reaching the hind margin. First joint of hind tarsus equal in length to the other two together ; spur as long as first joint of tarsus, many small teeth on hind margin. Dark chestnut or nearly black; carinae of head and thorax, second joint of antennae, middle and lateral margins of pronotum, middle of mesonotum, four or five pairs of small spots on face, middle and hind femora and hind tibiae lighter brown. Tegmina hyaline; before the cross-veins the veins are light narrowly bordered with brown, with small hght granules; beyond the cross-veins the veins are of the same colour as the membrane, which is brown over the cubital area and over the radial apical veins. spreading into the membrane ; wings hyaline. with brown veins. Length. 2°9 mm. ; tegmen, 4°0 mm. . Brachypterous; n colour similar to the macropterous form. As in most brachypterous Delphacids the reduction is mainly in the apical cells. The middle of the membrane between the veins is raised in longitudinal ridges, giving the tegmina the appearance of having numerous veins. Length 2°7 mm.; tegmen. 2°00 mm. NicgERIA: Oloke-Meji, Ibadan, 1914 (J. C. Bridwell). Described from two macropterous and one brachypterous females, the type being one of theformer. The second macropterous specimen fs darker than the type. 145 INSECTS INJURIOUS TO ECONOMIC CROPS IN THE ZANZIBAR PROTECTORATE. By W. MANSFIELD-ADERS, Government Economic Biologist, Zanzibar. (Purates XIII-XV.) The following notes on the various insects injurious to the economic crops of the Zanzibar Protectorate are based on investigations made during the last few years. The chief economic products of the two islands (Zanzibar and Pemba) are cloves and coconuts, the former being of greater importance. INSECTS AFFECTING CLOVES. Fortunately no insect pests of the clove tree (Hugenia caryophyllata) have been discovered, but it is apparent in many plantations that a number of trees are unthrifty and die. The cause of death still remains uncertain, but most of the evidence points to some physiological condition, such as waterlogging of the roots near swamps, denudation of alluvial soil on slopes, or the tap-root reaching an impervious layer of stone. The bark of certain unhealthy trees is attacked by termites (Termes bellicosus) and dead branches are infested with various species of Bostrychid beetles. Some young clove trees from Pemba had been killed by T. bellacosus feeding on the bark of the lateral roots. The adults of a large Tenebrionid beetle, Pycnocerus passerinw, Bertol., have been found on several occasions lurking under dead bark on clove trees, while larvae and pupae occur in deep tree-holes filled with decomposing vegetation ; I am of opinion that they do no actual damage, but only feed on dead bark, etc. Several experiments have been undertaken to test whether various omnivorous Lepidopterous larvae, such as those of Huproctis producta, would feed on clove trees when deprived of other food, but all died. INSECTS AFFECTING COCONUTS. Oryctes monoceros and O. boas. These two rhinoceros beetles are common through- out the two islands and are a serious menace to the coconut industry, the former being the commoner species. The damage done to young trees is often very serious ; in some plantations more than half of the trees from 23 to 3 years old are killed, while others are stunted and their productiveness much reduced. A certain number of trees survive initial attacks, but as their growth is retarded, they take a long time to reach maturity. Mature trees rarely succumb to the attacks of the beetles, although a number of felled trees were found to harbour the insects. The following record of the life-history of Oryctes monoceros has been obtained :— Eggs found in dead coconut trunk .. tne beg lee Vai HOTT, Kegs hatched .. oe si " PY Hee [Os Viti LAs. Larvae started to build pupal cell .. i sal) AGe xan LOM Ge Pupal cell completed and pupa formed * JAY BOyaa.) VOUT: Adults emerged ue ae My Pee. Mule ho tT (C605) E2 146 W. MANSFIELD-ADERS. It will be seen from this that the length of the larval life is 113 days, that of the pupa 18 days. After the mature larvae had built their pupal cell they shrunk considerably in size and became very sluggish before pupating ; this period of inertia lasted 5 days. Pupae are often found in association with larvae in old coconut trunks, generally among the coarse fibres where disintegration has not yet taken place; the larvae prefer the centre, where the fibres have been reduced to the consistency of a warm moist debris. The larvae of Oryctes boas are more usually found in rubbish heaps composed of decomposing vegetable matter and manure. I have found manure heaps riddled with larvae of all ages ; horse and donkey droppings seem the most attractive. The following preventive measures have been tried :— (1). Traps have been set in various localities filled with rotting coconut trunks, decaying vegetation, etc.; the addition of a little horse manure enhances their attractiveness and they have given excellent results. (2). The collection of adults and larvae by natives has been tried in Hast Africa, but the results were not encouraging. (3). In small plantations the daily examination of all young trees (from 24 to. 3 years old) for adult beetles should be quite efficacious. The easiest method is to insert a small piece of wire with harpoon points into the entrance hole, transfix the beetle and withdraw it. After a little experience the palms harbouring beetles are easily recognised by the moist tow-like frass protruding from the entry holes. Sprinkling dry earth or sand in the crowns of the young trees may possibly act as a deterrent to adult beetles. There is no doubt that bare plantations and isolated trees are more severely attacked, probably because the beetle being a heavy sluggish flier is able to alight more easily on its food-plant under these conditions. The planting of some tall bushy catch-crop, such as Cajanus wdicus, around the plantation might be efficacious. Some imported larvae infected with a fungus (Metarrhizwum amsopliae) were received, and it was proved that our local larvae are susceptible to the disease ; further, in some control experiments one case of actual indigenous infection was found. A long series of experiments with a view to introducing this fungus was planned, but owing to unforeseen circumstances all work in this direction was abandoned. I have on several occasions seen young trees suffering from what might be called bud-rot. The whole of the crown is easily pulled out leaving a hollow depression full of dark fluid with a most foetid odour. It may perhaps be proved later that bud-rot is secondary to the attacks of the beetle. To summarise the above :—The rhinoceros beetles (Oryctes monoceros and O. boas) are common throughout Zanzibar and Pemba, and are the most serious pests of the coconut industry. Young trees from 24 to 3 years old are most usually attacked, many are killed, others greatly delayed in reaching maturity. Trees in isolated positions and on bad soil are more often attacked. Fully matured trees harbour adult beetles, but the damage to them is not serious. The most useful preventive measure is the trapping of larvae in pits filled with rotting coconut trunks, vegetable debris and a little manure. INSECTS INJURIOUS TO CROPS IN ZANZIBAR. 147 No insect parasites have been found preying on the eggs, larvae, pupae or adults. Among mammals the most useful in destroying larvae are the MAcroscELIDAE and SoricipaEz. The following list includes all known to me. Rhynchocyon adersi. A large elephant shrew, which is somewhat local in its distribution and chiefly found in scrub on the east coast of the island. It has not been observed feeding on larvae under natural conditions, but partook of them greedily in captivity. Petrodomus sultan. A rare species, concerning the habits of which no observations have been obtained. Pachyura murina. These small shrews are widely distributed throughout the island and have been found in beetle traps on several occasions. Two species of crows (Corvus scapulatus and Corvus splendens) have been noticed feeding on the larvae of Oryctes boas in manure heaps. Many species of birds have been collected and their stomachs carefully examined for larvae or fragments of adults, but none have been found. The following are minor pests of coconut palms :— Rhynchota. Aspidiotus destructor, Sign. This dangerous coconut scale-insect has been found occasionally on husks and leaves, but at present does no appreciable damage. Aspidiotus cyanophyllr, Sign., A. lataniae, Sign., and Hemichionaspis minor, Mask., have all been found on the husks of coconuts in association with Aspodiotus destructor. Cerataphis lataniae, Licht. This cosmopolitan Aphid has been found frequently on the leaves of coconut palms but never in great numbers. Ornamental palms of various varieties are heavily infested. Coleoptera. Rhina amplicollis, Gerst. Aduits, larvae and pupae of this large weevil were found in a dead coconut tree in Pemba. MNocalandra frumenti, F. These small Curculionids were found innumbers in crevices on coconut husks. The infested nuts showed in the majority of cases cracks and fissures from which gum was exuding. Attention had been drawn to the trees on account of the shedding of immature nuts and peculiar deformities in the shape of the nut, but it is by no means certain that the injuries were due to the weevils, which were probably secondary visitors. Rhyncophorus phoenicis, ¥. This large weevil has been recorded from Pemba. Larvae, pupae and adults were found in an old stump of a coconut tree, others in a dead oil palm (laeis guineensis). As far as can be ascertained at present it is not a serious pest of coconuts. Isoptera. Termes bellicosus, Smeath. Termites often attack seed-beds and cause great havoc among the nuts, in some instances 50°% of them having been destroyed. The soaking of nuts in Cooper’s dip before planting gave good results. Mr. T. J. Anderson, Government Entomologist in British East Africa, advised watering the seed-beds with sea-water, but as the beds attacked were situated inland a considerable distance from the sea, the method has not been tested. 148 W. MANSFIELD-ADERS. INSECTS AFFECTING COTTON. A few experimental plants of the following varieties have been grown, viz., Egyptian :—Abassi, Nubari, Yannovitch, Mit Affifi, Asil, Sakellarides ; American :— Allen’s Long Staple and Sunflower. All the plants were heavily infested by a variety of insects of which the following proved the most injurious. ae Lepidoptera. Pectinophora (Gelechia) gossypiella, Saunders. This has proved to be the most serious pest, as every variety of cotton was attacked. All stages of development were taken throughout the year. Other food-plants were the silk-cotton tree (Eriodendron anfractuosum) and Ladies’ Fingers (Hibiscus esculentus), the latter plant being always heavily infested. Pyroderces simplex, Wism. Some adults emerged from stored seed. Earias insulana, Boisd. A few specimens were obtained from all varieties of cotton. A very common pest of Hibiscus esculentus. Sylepta derogata, F. This leaf-roller occurred abundantly on all varieties and was responsible for serious damage. Prodenia litura, F. Remarkably few larvae were found. Other food-plants are Hibiscus esculentus, young tobacco plants, castor plants, and on one occasion a ripe tomato fruit. Acrocercops bifasciata, Wism. Cotton plants of all ages were attacked by this leaf-miner. The attacked leaves have a very characteristic appearance, the whole of the upper epidermis being raised to form one large blister. Bucculatrix lozoptila, Meyr. Larvae were taken feeding on the upper epidermis of Abassi cotton ; they construct a tough white cocoon shaped like an upturned boat on the leaf. Orgyia vetusta, Hmp.* One small swarm of larvae appeared, completely defoliatmg the attacked plants. Euproctis producta, Wlk. Larvae occasionally swarm on cotton; they are omnivorous, moving from one food-plant to another, castor oil plants seeming to be their favorite food. Rhynchota. Dysdercus fasciatus, F. This stainer was prevalent during the bolling season and caused much damage to the lint ; all varieties of cotton were attacked, especially Caravonica. These insects were observed plunging their proboscis into young oily seeds in bolls which had opened prematurely. They were also seen on many occasions feeding with avidity on fresh mammalian carcases, skins and skulls. Dysdercus superstitiosus, F. This has been taken occasionally, but is by no means a common species. Oxycarenus albidipennis, Stal. All instars are common in well matured open bolls. In neglected crops they occur in vast numbers, although the actual damage done seems to be negligible. INSECTS INJURIOUS TO CROPS IN ZANZIBAR. 149 Pseudococcus obtusus, Newst. At times this is a serious pest. It is one of the commonest Coccids of the island and feeds on a variety of hosts. Many experimental plants of cotton were completely covered, the main stem being smothered. Pseudococcus virgatus, Ckll. Found occasionally on leaves and beneath the sepals. This Coccid is very abundant on a number of ornamental plants and shrubs. Pseudococcus citri, Risso. A rare species on cotton. INSECTS AFFECTING CEREALS. The principal cereals cultivated are rice, maize, and millet (Sorghum vulgare), none of them in any great quantity. Lepidoptera. Cirphis loreyi, Dup. The larvae of this Noctuid are common in all maize fields. They bore and tunnel through the main stalk, attack the base of the cob, and occasionally feed on unripe seeds. Pupation takes place either in the stalk orin the cob. The moist white excrement, very typical of the presence of the larva, is generally found at the base of the leaves or cob. The large exit holes of the imago are easily detected. The species also attacks millet. Sesamia calamistis, Hmp. Very similar in its habits to Curphis. In young cobs the tassel is often attacked, the larvae working downwards into the cob ; it likewise feeds on millet. Busseola fusca, Hmp. Larvae and pupae were found in the main stalk of millet, but never in great numbers. Chilo suppressalis, Wik. This species is the commonest borer of maize and the most harmful, as in some instances it will destroy a full half of the crop; it also attacks millet. Parnara mathias, F. Larvae of this butterfly have been found occasionally feeding on the leaves, but as a pest it is insignificant. Rhynchota. Peregrinus maidis, Ashm. A field of imported maize (Hickory King) was found heavily infested with this cosmopolitan pest ; but the common native variety of maize is rarely attacked. INSECTS INJURIOUS TO VEGETABLES. CABBAGES. Crocidolomia binotalis, Z. Cabbages are always heavily infested with the larvae of this moth and several valuable crops have been completely ruined. Pupation takes place on the leaf under a light silken web or in the rolled end of a leaf. The pupal stage averages 10 days. Other food-plants are turnip, watercress, radish and nasturtium. Pluiella maculipennis, Curt. The small brilliant green larvae of the diamond- back moth were taken in conjunction with those of the former species. Pupae are found on the leaf in a silken web. Phytometra signata, F. This moth is a minor pest. 150 W. MANSFIELD-ADERS. Eeag Piant (Solanum meiongena). Acanthocoris fasciculatus, F. This bug is a serious pest. In many market- gardens these useful plants are smothered with it ; the tips of the voung shoots wilt quickly when heavily infested. It occurs as a minor pest of the chilli (Capsicum sp.). CUCURBITACEAE. Dacus vertebratus, Bezzi. The larvae of this fruit-fly are extremely common in all varieties of native gourds and cause a large amount of damage. Other food- plants are Luffa acutangula and various squashes. Market-gardeners have attempted to grow melons and cucumbers, but very few reach maturity owing to the ravages of this fly. Two other species of Dacus also occur in gourds, D. brevistylus, Bezzi, and Dacus punctatifrons, Karsch. PicEON PEA (Cajanus indicus). Rhynchota. Pseudococcus obtusus, Newst. This insect, one of the commonest Coccids of the island, is found in enormous numbers on this useful plant. Coleoptera. Tragocephala variegata, Bert. Adults of this Longicorn are prevalent during the hot season from October to March. Eggs found in the field in April hatched in 10 days ; larvae have been kept in captivity for 3 months without pupating ; pupae taken in the field produced imagos in 12 days. The female gnaws a small hole in the bark of a lateral branch and there deposits an egg. The larvae feed for a short time in this branch, then tunnel down the main stalk, making many vent holes during their transit and eventually pupating in a plugged cell low down in the trunk or even in the root. A large exit hole denotes the escape of the adult. a Lyprops breviusculus, Gerst. Numbers of these beetles have been found feeding “ on the peas. All instars are present in the pods. They are generally found in neglected fields where old pods have been left on the trees. Several species of MrtorpAr feed on the leaves but are only minor pests. The following have been identified ;—Mylabris dicincta, Bert., Mylabris amplectens, Gerst., and Coryna ambigua, Gerst. Lepidoptera. Marasmarcha atomosa, W\sm. These small delicate moths have been taken towards dusk hovering around the pods. The characteristic larva, studded with spines and hairs radiating from tubercles, feeds through a hole on the pea. The pupa closely resembles the larva and is generally found on the outside of the pod. Deudoriz antalus, Hopfi. The large flat conspicuous larvae of this butterfly are an occasional pest of the pigeon pea. They are voracious feeders and attack the seeds in young green pods with avidity, at times causing a considerable amount of damage. Pupation takes place in the pod. INSECTS INJURIOUS TO CROPS IN ZANZIBAR. 151 Diptera. Agromyza sp. The larvae of this fly attack the young green seeds in the pod, making a ring-like track around the seed ; the pupae are formed within the pod. Lapigs’ Fincers (Hibiscus esculentus). These useful plants harbour a number of important insect pests of which the following have been recorded :—Pectinophora gossypiella, Harias insulana, Sylepta derogata and Prodenia litura ; as previously mentioned all of these occur on cotton. Nisotra weiset, Jac. Numbers of these small flea-beetles swarm on the leaves, giving them a typical shot-hole appearance. The same species has been found on roselle (Hibiscus sabdariffa). Sweet Potato (Ipomoea batatas). Cyclas formicarius, F. This cosmopolitan weevil is a common pest throughout the two islands, many tubers being completely riddled and destroyed. All instars are found in the tubers. Asyidomorpha puncticosta, Boh. The leaves are often severely attacked fe the larvae of this Cassidid beetle, which are conspicuous on account of their lone anal process. Cassava (Mambhot utihssema). This useful plant, the chief food supply of the natives, is exceptionally free from ‘pests. Pseudococcus (Dactylopius) virgatus var. madagascarensis, Newst. A few plants have been observed to be infested with this Coccid. Tetranychus sp. This small red mite is abundant in some plots, and causes ‘wilting and curling of the leaves. INSECTS INJURIOUS TO FRUIT TREES. CITRUS. ‘Rhynchota. Lepidosaphes becku, Newman. This scale-insect is the worst enemy of Citrus, and in many plantations the trees are thickly encrusted. The main stem and laterals are the favourite sites, but leaves and fruit are also attacked. I have seen numbers of oranges packed ready for export infested with this scale. Icerya purchast, Maskell. Fortunately very few trees are attacked by this Coccid, which at present is quite a minor pest. Icerya seychellarum, Westw. Has been observed on lemons. Pseudococcus obtusus, Newst. This Coccid has been taken on all varieties of orange, in addition to its other food-plants. Pseudococcus citrt, Risso. A minor pest which has never been found in abundance. Ceroplastes rubens, Maskell. One record from a young orange tree. Coccus viridis, L. This scale is very prevalent on young trees in some plantations and shows a marked partiality for imported species of Citrus. 1.52 W. MANSFIELD-ADERS. Aspidiotus (Pseudaomdia) trilobituformis, Green. Occasionally met with; this Coccid has a variety of hosts. Cerataphis latanae, Licht. This is another of the commonest citrus pests; the leaves of many trees both old and young were found to be heavily infested. This Aphid is also abundant on many varieties of ornamental palms. Aphis tavaresi, del G. The black cotton aphis is extremely seasonal in its occurrence and is generally found on the leaves and young tender shoots. Lepidoptera. Argyroploce leucotrzta, Meyr. (Citrus Codling Moth). Larvae have been found on several occasions in mandarins, but not in oranges. Papilio demodocus, Hsp. A very common pest on all young citrus trees. Many voung plants are completely defoliated, and the larvae are occasionally found feeding on tender shoots of mature trees. Coleoptera. Tragocephala variegata, Bert. The habits of this Longicorn have been described under pests of the pigeon pea. It attacks all varieties of Crtrus. Porphyronota maculatissima, Boh. A few adults of this unimportant Cetonud beetle have been taken from time to time feeding on the leaves. Gyponychus cervinus, Gerst. A number of adults of this weevil have been obtained, they attacked the leaves of young orange trees ; a minor pest. Diptera. Ceratitis capitata, Wd. This serious pest does not seem to be very widespread. Oranges and mandarins of imported varieties were found to be infested, but indigenous trees (oranges, mandarins and lemons) are far less susceptible to the attacks of this fruit-fly. | MANGOs. Sternochetus (Cryptorrhynchus) mangiferac, F. Adults are commonly found in the stone of the fruit, larvae and pupae in fallen immature fruit. The following Coccrpar have been found on the leaves and fruit :—Asprdrotus (Chrysomphalus) dictyospermi, Morg., Aspidiotus destructor, Sign., Pseudococcus obtusus, Newst., Lecanium adersi, Newst., L. (Saissetia) nigrum, Nietn., L. (Sacssetia) punctuliferum, Green. Of these the only one of importance is Pseudococcus obtusus, which heavily infests both the leaves and fruit of many trees. BANANAS. Aspidiotus destructor, Sign. Has been found occasionally both on the fruit and leaves. SOURSOP. Ceratitis rosa, Karsch. One fruit was found to harbour larvae of this fly. INSECTS INJURIOUS TO CROPS IN ZANZIBAR. 153: INSECTS AFFECTING SHADE TREES. Lepidoptera. The African almond (Lerminalhia catappa) is common throughout the two islands and is much appreciated for its ornamental foliage and shade properties. In certain. situations, especially in the town, it is severely attacked by bag-worms (PSYCHIDAE). Unthrifty trees are heavily infested, most of the leaves being almost skeletonised. Miresa melanosticta, Baker. A minor pest of the African almond. The large green slug-like larvae with lateral tubercles are not easily detected in spite of their size. They have the peculiar habit of arranging themselves in circles on the leaves, their heads facing inwards. Pupation takes place in the soil in a hard round cocoon, with a distinct lid. Curia forda, Westw. The larvae feed on young casuarina trees (Casuarina equisetifolia) and are easily detected by their conspicuous yellow markings. Pupation takes place in the soil at a depth of about 3 inches. Another food-plant is the naseberry. Asura saginaria, Wik. The larvae of this moth have have been taken occasionally on cinnamon trees, but itis quite a minor pest. Rhynchota. Pseudococcus obtusus, Newst. When in poor soil casuarina trees are often heavily infested with this scale-insect, the lower branches being principally attacked. Aspidiotus (Pseudaonidia) trilobitiformis, Green. Many rubber trees (Ficus elastica). are heavily infested with this Coccid. Dysdercus swperstitiosus, F. The open bolls of the silk-cotton tree (Hriodendron anfractuosum) are often attacked by this common cotton-stainer, and D. fasciatus is found in conjunction with it. INSECTS AFFECTING TIMBER. Coleoptera. Macrotoma palmata, F. The larvae of this Longicorn are a serious pest in the mangrove wood rafters in native huts. On one occasion several larvae were extracted from large tunnels which they had made in a piano case. Numbers have been found in the field in old dead mango stumps. Dinoderus minutus, F. Extremely common and destructive to dried bamboo and a very serious pest. Adults have been found in Madagascar teak, a shght attack. On one occasion the bark of an avocado pear (Persea gratissima) was found to be heavily infested with it. Isoptera. Termes bellicosus, Smeath. Ubiquitous in its distribution, the white ant is the worst timber pest of the two islands. Imported European soft woods are reduced to ruin in a few years, though Indian teak is rarely attacked. The structural timber of the native huts (mangrove, Ceriops candolliana) and the roofing made of plaited coconut leaf are often attacked. Termites have been reported as injuring the following living trees :—Seed and seedling coconuts, young avocado pears (the roots badly attacked), and cloves. 154 W. MANSFIELD-ADERS. INSECTS INJURIOUS TO MISCELLANEOUS PLAN®SS. Lepidoptera. Thalassoides digressa, Walk. The larvae occasionally feed on the leaves of the castor plant. Duomitus capensis, Wik. The larvae have been found tunnelling in the main stem of the castor plant. Brithys pancratu, Cyr. The caterpillars are voracious feeders on all species of hhes ; when young they are gregarious and feed under the epidermis of the leat, later they separate and devour the whole plant. Pupation takes place in the soil or at the base of the leaves, the pupal stage averaging from 8 to 10 days. Euchromia formosa, Guér. The larva is clothed with tufts of hairs, dull-coloured and inconspicuous, and forms a cocoon of silk and hairs on the branches of the food-plant. 1tis common on all species of creepers belonging to the genus [pomoea. Glyphodes sericea, Drury. The larvae are common on gardenias. Adult larvae are of a transparent green colour with four black spots on the dorsum. They are leaf-rollers and live and pupate in twisted leaves, their pupal stage averaging 9 days. Coleoptera. Entyposis impressa, Kolbe. Larvae of this weevil have been recorded as feeding on the roots of castor plants and caladiums. Brachycerus atrox, Gerst. Great numbers of adult weevils have been taken in the soil around the tubers of lilies (Amaryllis sp.). During the day they hide in burrows near their food-plant and come out to feed on the leaves at night. Larvae feed on the tubers, eating into the centre. Mausoleopis amabilis, Schaum. Adults of this Cetonid are common in the blossoms of various ornamental plants, especially roses. Rhynchota. Aspidiotus (Chrysomphalus) aonidum, L. A common scale-insect on rose stalks. Pseudococcus virgatus, Ckll. Has been recorded from various ornamental creepers, etc. Orthoptera. Chrotogonus hemrpterus, Schaum. A very troublesome grasshopper; one experimental field of castor was ruined by its depredations. INSECTS INJURIOUS TO STORED PRODUCTS. In the tropics, where food is abundant throughout the year, insects attacking stored grain are more in evidence than in Europe. Owing to the custom of storing grain in loose receptacles and exposing it in open bins for sale nearly every sample showed evidence of insect attack. INSECTS INJURIOUS TO CROPS IN ZANZIBAR. 155 Coleoptera. Calandra oryzae, L. This cosmopolitan pest of gram is extremely common in rice. and is also found in maize and sorghum. Laemophloeus pusillus, Schénh. Very abundant in maize. Tribolium castaneum, Hbst. Common in maize and occasionally found in rice. Silvanus surinamensis, L. Abundant in maize. Tenebroides mauritanicus, L. An occasional pest of maize. Bruchus chinensis, L. A serious pest of all beans and pulses. Numbers are always to be found in Phaseolus mungo and Cajanus wmdicus. Bruchus ornatus, Fhs. Not so common as B. chinensis, but has been recorded from beans. Carpophilus humeralis, F. Very common in maize cobs in the field, generally in those attacked by fungus. Cossonus suturalis, Boh. An occasional pest of stored sweet potatoes. Dermestes vulpinus, F. Abundant in dried fish. Latheticus oryzae, Waterh. An occasional pest of maize. Rhizopertha domanica, F. Occasionally in maize. Necrobia rufipes, de G. Both larvae and adults are abundant in dried copra, feeding on the kernels; the former bore long cylindrical tunnels into the dried copra. Lepidoptera. Ephestia cautella, Wik. Larvae very common jn rice and various flours. ye a, atl pees ; iSesqp ly am Le er 4 z ate : ‘ My An th oy ia , ie i an 2h A ih : ae a [ato pee. a ern “ase, | 5 oe 4, it 4 i A e , aad x, i" wie ‘ i ' aw fs yy a Ph i2 a Vicoitanis wis as ‘waening ae phils Rik My yt ak «Hie pune | % vy ~ f= i cy : J J ' ot : ‘ a oe DO J . } f } i { 4 A Geena. ern, ee a ny a ‘ enn . 8-No 2 Aha AL Heaney ON a sire afi Bacar” v bana he vy ae Loar: ie es oe eh laitagls WP: Wh aibtodh e. NOW a vellacaenbtersinh he fay ut fads, cay Pym noc me Hino! gabe txeep al D iy . A ry nae oye a é % Se lei 7 k : ‘tite alle Sivien sivas Bes a = 7 ay PY v vee, deri hy aa dan ee heyy r 7 eR TP Oy. od - iat a ‘ Pree aa if a Lah BULL, IEW, IRESENRGH. WOilbe 2X. JPART 2. Pete XO A young Coconut Palm with stunted leaves badly attacked by Oryetes, showing the entry hole of an adult Beetle at the base of a leaf-stalk. Bui. JBiwae, RESEARCH, WO, XM, IPasar Ze PLATE XIV. Leaf of Coconut Palm broken off just above the Bore-hole of an Oryetes. fant ty =, , WEB: :|] | A ay. ip =} A : * ( if aN |) VAN ANN Rin : Par Nit HRA F\Whit Wk be mh: \ Wipy > Y \i\\ 4 y Wh 7 \ Y \ ‘ Fig. 3. Larva of C. elaeidis, mouth-parts ; md.j, mandibular joint ; mz, maxilla ; ch.s, chitinous supports ; ms, muscles controlling the movement of the mandibles ; l, labium; s, setae; B, roof of the mouth. on each side of the head a little posterior to the mandibular joint ; it is two-jointed, with a simple lobe at the apex of the second joint and another stouter one bearing several papillae (fig. 2, a). The foregoimg observations are taken from preparations made from two larvae, one full fed and the other younger. They were cleared in potash and washed and put in 90 per cent. alcohol shghtly coloured with Congo red for about 12 hours; the chitm was stamed well, showing good differentiation. With more material the structures may be made out with more precision and in greater detail than I have been able to do at present. The figure of the larva has been drawn from a balsam preparation ; m clearing in potash the specimens get stretched to a certain extent. In drawing the figures of the mouth-parts of the larva the artist was aided by three photomicrographs taken by myself. 175 OBSERVATIONS ON SCALE-INSECTS (COCCIDAK).—VI. By Rosert Newsteap, F.R.S., The School of Tropical Medicine, the Unversity, Liverpool. (Prare xX VT") Clypeococcus, gen. nov. Female. Dorsum nude and densely chitinised. Marsupium occupying practically Venter (sternites) remaining attached to the food-plant. the whole of the interior. Cephalic margin with a strongly developed clypeus, Legs and antennae present. lying prone over the mentum. Abdominal stigmata present. Larva of the Icerya type, and with several pairs of abdominal stigmata. Type, C. hempelr, Ckll. Clypeococcus hempeli, Ckll. Icerya (Crypticerya) hempelr, Ckil., Can. Ent. xxxi, p. 43 (1899). Crypticerya hempeli, Hemp., Rev. Mus. Paul., iv, p. 376 (1900). Female adult (PI. xvi, fig. 1). Sphaeroid and hollow at gestation, the marsupium filling practically the whole of the interior. Integument densely chitinised, faintly ribbed and polished ; very dark plum-coloured, with a delicate ‘‘ bloom-like”’ secretion, the segmentation of the dorsum indicated by faint blackish interrupted lines. Venter ° (fe REY NU Ki NCA A: = BEEZ ” ag = F > fi eH —= =, cB are | PY “4 Clypeococcus hempeli, Ckll., 9; a, the relative position of the various Fig. °h. appendages on the anterior portion of the venter: an, antenna; cl, clypeus ; m.sh, membranous sheath enclosing the buccal filaments ; b, antenna of larva ; ce, tarsus of larva; d, gland-pores and hairs on venter of 2; e, dorsal spines of larva; f, gland-pores of larva. membranous, resting on a thin layer of hard resin-like secretion, segmentation distinct. Mentum biarticulate, the “filaments ’’ enclosed in a membranous sheath (m.sh. in G (C605) 176 ROBERT NEWSTEAD. fig 1, a), the latter lying obliquely to one side of the median line in mounted pre- parations. Clypeus (fig. 1, a, cl.) strongly produced, densely hairy and com- pletely overlying the mentum. Antennae, legs and thoracic stigmata lying in very deep depressions. Ventrally the margins and the whole of the thoracic area, together with the terminal abdominal segment, densely hairy ; the hairs (fig. 1, d) distinctly knobbed at the tips. Antennae (fig. 1, a, an) robust, of 8-9 strongly chitinised segments. Legs (fig. 1, a) robust, and strongly chitinised ; claw simple. Abdominal stigmata ‘small and apparently in six pairs. Thoracic stigmata large; the gland-pores mul- tilocular, the cylindrical ducts long and apparently truncate proximally. Dermal ‘gland-pores (fig. 1, f) large and with a deep cylindrical rim; they are irregularly disposed on the venter and for a short distance also beyond the margi1; the dorsum rather thickly studded with minute pores surrounded by dark granular bodies. Anus large and roughly circular in the outline formed by the dense chitinous walls of the body; no other exact details observed, but portions of a ruptured membranous cuticle project into the lumen, suggesting that the true structure is wanting. Diameter, 6°5-9mm.; antennae, 0°6 mm. long. Larva. Form elongate. Antennae (fig. 1, 6) clavate, of six segments ; 2nd, 4th, and 6th longest ; 3rd and 5th very short and subequal ; all the segments with fine slender hairs, the longer ones less than the length of the terminal segment ; terminal segment with several slender spinose hairs at the tip ; in the clear membrane which connects the 2nd and 3rd segments is a relatively very large chitinous ring with a granular centre; as no hair or spine has been seen attached to this organ it is suggested that it may be a sensorium or gland-pore; three similar structures also occur on the coxae. Legs long and slender; claw (fig. 1, c) with a minute, subapical denticle. Dorsal gland-pores with a broad quoit-like rim and a very short cylindrical duct, the whole structure when seen in profile reminding one of a very short-stemmed agaric or broad-flanged stopper to a glass jar; these are arranged more or less in transverse rows. The terminal segments of the abdomen, more especially in the region of the anus, with a group of large spines, packed very closely together in the median area. Terminal abdominal bristles in five pairs ; these are of great length, or slightly longer than the antennae. The chitinised hind gut with its flange-like papillae very distinct. Abdomen with at least six pairs of stigmata ; the external opening minute and membranous; the atrium long, gradually narrowed distally, strongly ribbed transversely or irregularly moniliform, and the proximal portion with a deep wide cleft ; the connective tube slender, chitinised, and bifurcated, each branch having well-formed taenidia; length of the atrium about twice the diameter of the large quoit-like gland-pores. Braziu: San Paulo; on the spiny branches of an unknown tree or shrub (mimosa ?), 1906. (In a small collection of Brazilian Coccidae purchased from Mr. O. EK. Janson ; no other data). The determination of the species is based upon an examination of a specimen (ex coll. T. D. A. Cockerell) kindly supplied by Mr. E. E. Green, to whom I express my thanks, as in the absence of his material I should most certainly have given this insect a new specific name. I have described the morphological characters in some detail as hitherto the true characteristics seem to have been almost entirely overlooked. The presence of a OBSERVATIONS ON SCALE-INSECTS. rire strongly developed clypeus in this Coccid it is unique, and is probably homologous with that of certain members of the FuncoripEa belonging to the genus Cixius and other allied forms. Aspidoproctus gowdeyi, sp. nov. Female adult. Completely covered dorsally with a thick and densely felted layer of dusky white and pale yellow wax ; the wax more or less divided into segments corresponding to the segmentation of the body. Venter pale castaneous, slightly mealy and in parts covered with white woolly filaments. Legs pitchy red. Marsupium well developed, the secretionary operculum wanting, but apparently broken away. Fig. 2. Aspidoproctus gowdeyti, Newst., sp. n., 6 genital armature. Male. In dry specimens the body is uniformly black or piceous. Wings very dark smoky brown or blackish ; costa black, subcostal vein dull crimson ; forked pseudo-vein dull white and very conspicuous. Halteres with 4-5 strongly hooked bristles. Body mealy. On maceration in KOH the softer parts of the integument change to dull crimson ; the sclerites and also the legs and antennae, piceous. The single pair of central processes well developed ; terminal bristles varying in length and thickness, the longest about four times the length of the process. Genital armature (fig. 2) shaped somewhat like a flask or ampulla in minature ; the proximal portion broadest ; near the centre of the dilated proximal portion is a shallow saucer- shaped appendage suspended to the walls of a large, ovate, chitinous ring ; distal portion cylindrical and densely chitinised. Intromittent organ not observed. UcanDA: Kampala, on plumbago and rose, 15.x.1918 (C. C. Gowdey). The males hatched apparently during transit through the post and were rather badly crumpled and otherwise injured. Walkeriana digitifrons, sp. nov. Female, adult (Plate xvi, fig. 3). Stationary. Somewhat pyriform, narrowest in front ; highly convex and sloping gradually towards the cephalic margin. Dorsum more or less covered with an easily deciduous layer of white or pale buff-white, granular wax ; margin with relatively robust, coalescing plates of white wax, which, when perfect, are distinctly laminated transversely ; cephalic margin with a short stout, centrally placed, cylindrical process. Integument very hard, dark castaneous or piceous; margin, when denuded, with a series of large blunt tooth-like projections or tubercles, arranged rather widely apart ; dorsum with 3-4 bilateral rows of gland pits, im which the granular wax generally remains more or less intact; abdominal (C605) a2 178 ROBERT NEWSTEAD. segmentation distinct. Venter hollow, but filled with loose white flocculent material. in which the eggs are laid and the larvae subsequently hatch. Vaginal orifice transversely linear, and without a secretionary operculum or an invaginated mar- supium. The integument of the dorsum, after long maceration in KOH, becomes pale brown in colour; but partly retains its hard and somewhat brittle nature. Antennae with 10-11 segments; 4th segment the shortest, but its articulation with the 5th is often incomplete, though the constriction between these segments is well marked and relatively deep. Eyes in the form of a truncated cone. Legs rather sparsely hirsute, but normal in shape. Compound gland-pores of the dorsal pits relatively large, rather narrowly ovate, and collectively forming a rather coarse and irregular reticulation. Dorsal gland-pores of the usual multilocular type ; they are fairly numerous and more or less evenly distributed ; those on the marginal tubercles are surrounded by a radial and somewhat petaloid pattern, due apparently to the thickening of the integument. Ventral gland-pores similar to those of the dorsum, but slightly larger and with an ovate central pore ; they are thickly packed together at the margins and also on the marginal tubercles. Anus surrounded by a broad concentric band of dark chitin, which is thickly studded with simple, more or less circular gland-pores. Fine slender hairs are scattered over both sides of the body ; marginal hairs also slender, but longer than those on other parts of the body. Length, inclusive of the finger-like process on the cephalic margin, 6-9 mm. Larva. This is of the usualform. Five of the long hairs on the terminal segment of the antennae equalling the length of the body. Sides of the abdomen with numerous long fine hairs ; anal hairs or bristles apparently in three pairs, of which the central pair are much the more slender and shorter ; outer pairs (2nd and 3rd) relatively stout and longer than the body. Uaanpa: Damba Isl., Sesse Islands, Lake Victoria ; on Barkea eminir, 8.x.1912 (C. C. Gowdey). Pseudococcus inquilinus, sp. nov. Female, adult. Uxternal covering and lateral appendages destroyed by the medium in which the specimens were preserved. Form, when denuded, similar to that of P. longispinus, Targ. With 16-17 pairs of cerarii; anal pair with two, the others with from 4 to 7 sharply conical spines. Lateral cerarn (fig. 3, c) with 2-3 auxillary setae and numerous, obtusely triangular pores rather closely grouped together near the cerarn, but scattered and mergmg into the body-pores beyond. Anal lobe cerarii (fig. 3, a) surrounded by a small circular chitinised area; ventral surface of each lobe (fig. 3, 6) with a broad and somewhat rectangular chitinised area. Anal ring well formed; setae (fig. 3, e) a very little shorter than the anal lobe setae. Dorsal body setae (fig. 3, d) numerous, slenderly spinose, with, in many imstances, flagellate tips. Obtusely triangular pores very numerous, imter- mingled with larger simple pores ; and on the last few segments of the abdomen a few multilocular pores ; tubular ducts very short and scanty (?) the rims ,of which are chitinised. Dorsal osteoles large, in two pairs. Legs relatively stout. The young adult female resembles the old adult, but has the gland-pores more closely packed together. OBSERVATIONS ON SCALE-INSECTS. 179 b) BritisH Guiana: “‘ Cattle Trail Survey,’ on an unknown plant; ‘‘ the insects enclosed by ants (Acromyrmez sp.) in small paper nests,” 1919 (A. A. Abraham, per G. #. Bodkin). This species is very near P. comstocki, Kuwana, as defined by Ferris (Leland Stanford Jun. Univ. Pub. “The California Species of Mealy Bugs,” p. 41) but differs in having a larger number of cerarian spmes and in the form of the chitinised area on the ventral surface of the anal lobes. Taken in association with Lecanium inquilinum, Newst., and L. deformosum, Newst. Fig. 3. Pseudococcus inquilinus, Newst., sp. n. 9; a, aorsal and, b, ventral surface of anal lobes ; ¢, lateral abdominal cerarii; d, body setae; e, hair of anal ring. Pseudococcus perniciosus, Newst. & Willcocks, var. Ovisac of female. Arranged m a similar way to those of typical P. perniciosus,* but the more or less globular masses are smaller. The examples are so badly weathered, however, that it is not possible to give details of the structure of the mdividual ovisacs. Female, adult. Under pressure of the covering glass the form is narrowly ovate. Antennae of seven segments. One pair of cerari present on the anal lobes (fig. 4, a), the spines sharp and somewhat slender, with a few obtusely triangular gland-pores scattered around them. No typical lateral abdominal cerari present, but their position is indicated by a single, faintly lanceolate spine (fig. 4, c), with from 1 to 2 simple supplementary setae at some distance away from it: the spines are traceable m some individuals on the last 3-4 segments, in others on the penultimate segment only. Body spies (fig. 4, d) mmute, famtly lanceolate, and very scanty indeed ; hairs small and also very scanty. Anal lobe setae (fig. 4, 6) shghtly longer than the anal rime setae (fig. 4, e). Integument very thickly set with gland-pores (fig. 4, c, c), more especially so along the margin; these are of three kinds: multilocular, tubular and obtusely triangular ; the first-named are arranged in narrow transverse bands on the dorsal surface of some of the abdominal segments, elsewhere they are regularly disposed on both surfaces. Length of 53 adult 99 varying between 2:1 and 2°7 mm. : ——— * Newstead & Willcocks. Bull. Ent. Res., i, p. 138 (1910). 180 ROBERT NEWSTEAD. Larva. Anal lobes with a pair of slightly lanceolate spmes; setae longer than those of the anal ring. Dorsal spies minute, similar m shape to those on the anal lobes. BritisH East Arrica: Kabete, on coffee, November 1918. ‘‘ Coffee bush infected in the lab. has been killed by this scale” (F. W. Dry, for T. J. Anderson). Typical examples of B. perniciosus, N. & W., have from 5-6 pairs of lanceolate cerarian spies on the distal segments of the abdomen, with 2-3 obtusely triangular pores scattered near them; and the body spines, though smaller than the cerarian spmes, are relatively much larger, and also much more frequent than are those in the variety from coffee at Kabete. Moreover P. perniciosus is much larger, measuring from 3-4 mm. - 1 i] s ‘ 8 J ' 1 8 + ‘ ' ' i} ¢ RB Perniciosus, var. FP. filamentosus. Fig. 4. Pseudococcus perniciosus, N. & W., var., 2; a, b, dorsal and ventral aspect of the anal lobes; c, lateral abdominal cerarii; dd, gland-pores ; e, hair of analring. Pseudococcus filamentosus, Ckil., 2; f, gland-pores; gg, lateral abdominal ceraril. Brain * has sunk P. perniciosus as a synonym of P. filamentosus, CkiL, without giving reasons for so domg; this action has led me to re-examime examples of the latter (part of the type lot kindly presented to me by Professor T. D. A. Cockerell in 1892) and I find that it differs to a marked degree from P. perniciosus in having very few gland-pores (fig. 4, f), relatively shorter anal lobe setae and small groups of obtusely triangular pores round the cerarian spines (fig. 4, g,g). Some of the tubular ducts of the dorsum are also much shorter and many of them are accompanied by 2-3 obtusely triangular pores. Clearly therefore P. pernaciosus is specifically distinct from P. filamentosus, and the former name must be retamed. * Brain, C. K. Trans. R. Soc. 8S. Africa, v, pt. 2, p. 99 (1915). OBSERVATIONS ON SCALE-INSECTS. 181 Phenacoccus ballardi, Newst. The original description * of this rather remarkable insect was given without illustrations. I have thought it desirable, therefore, to give a photomicrograph of both old and young adult females (Pl. xvi, fig. 4) m the hope that it will enable students to determine the species with greater ease. Pseudophilippia inquilina, sp. nov. Female, adult. Form short ovate and slightly tumid. Colour in life mauve pink. Dorsum nude; venter protected. by a thin vesicular glassy scale, which is firmly attached to the bark of the food-plant. Antennae (fig. 5, @) and Jegs quite rudimentary; the former, which are much shorter than the stigmata, are composed Fig. 5. Pseudophilippia inquilina, Newst., sp. n. Q; a, antenna ; b, leg; c, ¢,, ¢,, stigmata; d, gland-pores. Second stage @: e, stigmatic cleft; f, anal lobes. apparently of three segments, the apex having several stiff hairs. Legs (fig. 5, 6) slightly smaller than the antennae, bare, composed of a single tubercular-shaped segment, with a relatively large claw; lower digitules stout and bluntly pointed. Stigmatic clefts (fig. 5, c) clearly defined. The stigmata (fig. 5, ¢, ¢,,¢,), which are placed close to the cleft, are protected by a well-defined external arch (fig. 5, gl. a7.) shaped somewhat like a horse-shoe in minature, and closely set internally with thick-rimmed multilocular gland-pores. Anal cleft nearly four times the length of the lobes; the latter surrounded by a distinct chitimous arch. Venter rather * Newstead, R. Bull. Ent. Research, viii, p. 17 (1917). 182 ROBERT NEWSTEAD. thickly studded with circular gland-pores (fig. 5, d). Dorsum with numerous spines, especially towards the margin. Rostrum relatively large; the filaments protected by a very long membranous sheath. Length, 4°2-6°8 mm ; width, 4°3-5°5 mm. Female, second stage. Short ovate; dorsum flat. Colour and the structural characters of the antennae and legs as in the adult female. Stigmatic clefts (fig. 5, e) relatively deep. Stigmatic spmes two, one on either side of the external glandular arch ; the latter much more extended towards the margm, and with fewer gland- pores than in the adult; fulcrum to the atrium of the stigmata strongly produced. Anal lobes (fig. 5, /) with a tongue-shaped sclerite between them at the base. Larva. Ovate. Antennae and legs well developed ; the former of six segments, of which the 3rd is much the longest and equal m length to the 2nd, 4th and 5th together. Anal lobes large; apical hair very long and stout. One large stigmatic spime in each cleft. Marginal hairs m an irregular double row. Abdommal segments with a transverse series of hairs, slightly smaller than those at the margin. Rostral filaments as long, apparently, as the circumference of the body. JAMAICA: on the banks of the Great River, near Montpelier; attached to the bark of an unknown tree beneath a large, blackish coloured “‘ paper” nest of Cremastogaster brevispmnosa, Mayr, var. tumulifera, For. The nest in question was attached to the bole of the tree about six feet from the ground; 10.xu1.1908 (R. Newstead). , I have placed this rather remarkable msect in Cockerell’s genus Pseudophilippia as it agrees best, m its morphological characters, with Cockerell’s diagnosis ; but the absence of an ovisac may be thought by other students to preclude its admission here. It seems to me, however, that the presence of a glassy ventral scale beneath the body of the female and the curiously protected stigmata do not m themselves call for the erection of a new genus. Antonina waterstoni, sp. nov. Female, adult. Colour, in life, pale buff to dusky buff. Form flat, narrowly ovate to very elongate, broadly rounded in front, widest generally in the region of the proximal segments of the abdomen, narrowing rather suddenly behind; last two segments of abdomen (fig. 6, a) strongly constricted. Antennae quite rudimentary and apparently unsegmented ; they are placed quite close to the margin. Legs absent. Mentum very small and unisegmented; just below it there are several minute tubercles, each with a short stiff hair. Stigmata large and widely separated ; first pair in a line with the rostrum ; the small group of parastigmatic glands merging into those at the margin. On the dorsal surface just behind the 2nd pair of stigmata is a large group of minute spines (fig. 6, b) occupying the whole width of the 1st proximal segment of the abdomen. Margin, all round, with an almost contmuous band of relatively large pores (fig. 6, c) and a few minute pointed spines. Anal segment (fig. 6, a) markedly distinct ; dorsal surface with a pair of forwardly directed bristles ; ventrally it is almost covered with pores, which are almost as numerous as those on the preceding segment. Anal lobes quite rudimentary, each bearing a few stiff hairs. Anal ring (fig. 6, a,) placed in a slight depression; hairs six in number, and rather stout. Length of young adult, 3-42; width, 2-32 mm. Old adults measure : length, 5°7-6 mm ; width, 3-3-2 mm. OBSERVATIONS ON SCALE-INSECTS. 183 The dorsum and venter are very thinly protected with white powdery wax, this secretion adheres to the food-plant, but readily comes away from the insect ; beyond the body, especially in the posterior region, the secretion is dense and completely fills the narrow space between the stem and the leaf-sheath. The general of facies the female, together with the secretionary matter (ovisac), bears a striking resemblance to Aclerda berleser, Bufta. RW. pL aim Fig. 6. Antonina waterston, Newst., sp. n., 9; a, terminal seg- ments of abdomen; a,, terminal segment of abdomen with anal ring, etc.; b, gland-pores and body spines; c, marginal gland-pores ; d, dorsal aspect of pupa; e, leg i. of pupa; f, antenna of larva. Male (fig. 7, a). Rather robust. Head almost as broad as the thorax, articula- tion faint. Eyes well within the margin, and enclosed by two longitudinal curved sclerites ; ocelli slightly smaller, placed between the antennae, just within the margin of the frons, ventrally. Antennae (fig. 7, b) of nine segments; Ist and 2nd very robust; 184 ROBERT NEWSTEAD. 3rd and 4th very slender and markedly narrower than the rest ; 4th much the smallest and less than half the length of the 3rd ; all the segments with the exception of the 4th and 9th with short stout, bluntly pointed spines ; terminal one with several long hairs. Leg i. (fig. 7, c) with the tibia and tarsus more robust and much shorter than the corresponding segments in legs i. and il. (fig. 7, d); all the tibiae with distal spines, ventrally. Genital sheath short. Caudal bristles long. Wings long, but rather narrow. Length to end of genital sheath, 0°8. Pupa (fig. 6, d). Robust. Antennal sheaths short and composed, as far as one can ascertain, of nine segments. -Leg i. (fig. 6, e) more robust than legs ii. and i. Claws to tarsi very slender. Length, 0°8 mm. Fig. 7. Antonina waterstoni, Newst., sp.n., ¢; a, ventral view ; b, antenna ; c,legi; d,legili; e, vestigial buccal cavity (?) Puparvum of Male. Three of these were found closely packed together under a leaf-sheath in association with the females. Collectively they formed an irregular mass of white, loosely felted and brittle strands of wax, completely enclosing the pupae. Larva, adult (¢ Male). Very elongated, parallel-sided. Antennae (fig. 6, f) of six segments ; 3rd shortest ; 6th longer than the first three together ; all the segments. with fine hairs, the 5th and 6th with long slender spinose ones. Anal segment of abdomen with a long stiff bristle indicating the position of the lobes. Anal ring with six long hairs. Rostrum very broad. Eyes small but prominent. Length, 055 mm. MacrepontA: beneath the leaf-sheaths of Arundo phragmites, 1917 (Capt. James Waterston). OBSERVATIONS ON SCALE-INSECTS. 185 It affords me infinite pleasure to dedicate this newly discovered species to our esteemed colleague. The males were all dead, and more or less imperfect ; and although one failed to trace the long caudal filaments, the presence of these structures is indicated by the setae which supports them in life. All the examples were lying beneath the leaf- sheaths and were flattened out as if by pressure of the sheaths. The male of Antonina australis, Green, has been seen by Froggatt (Agricultural Gazette, N.S.W. No. 742, p. 3, 1904), but so far as I can ascertain it has not been described. The discovery therefore of the male of A. waterstona is of interest, as the members of this sex in all the other species hitherto described are unknown. The female of A. waterstoni is nearly related to A. socialis, Newstead, but differs in having much smaller antennae, in the presence of a large isolated group of minute spines, and in the character of the anal segment, including also the relative position of the anal ring. Pseudokermes marginatus, sp. nov. Female Test. Roughly hemispherical ; glassy white, with a median longitudinal suture, which renders the two halves easily separable ; dorsal surface with faint traces of small and somewhat rectangular patches of secretion ; sides with wavy conchoidal striae ; stigmatic ridges more or less distinct ; margin wavy. Female, adult. Shape somewhat like that of a soldier’s steel helmet in miniature, with a narrow mediodorsal ridge, a relatively very broad flat margin (rim) and prominent anal lobes. Surface faintly uneven but shining. Colour pale castaneous. Boiled in KOH, the integument of the dorsum changes to pale straw-colour ; the broad flange becomes quite transparent, and the extreme margin brownish—the three grades of chitin showing in marked contrast to one another. Antennae represented by exceedingly minute tubercles bearing 4-5 short stout setae. Legs entirely absent. Margin wavy and irregular. Stigmata robust, somewhat cylindrical, and externally obtusely conical. Stigmatic clefts and spines absent. Marginal spines relatively stout, acutely pointed, and very widely separated. Anal cleft deep ; lobes somewhat triangular, the proximal and inner margins longer than the outer ; apices with several fine hairs. Dorsal gland-pores in the median longitudinal ridge, circular, surrounded by a small pale area, and often divided into two linear groups. The broad flat marginal flange, in very old and heavily stained examples, with numerous cell-like clear areas, the inner series forming an irregular dactyliform pattern; in younger forms these structures are wanting and in their place are seen a large number of narrow tubular ducts. The extreme margin presents, on its inner surface, an irregular crenulated appearance, the depressions occupied by a rather ill-defined duct. Length, 2°1-1°8 ; width, 2°1-2 mm. British Guiana; Ituni Savannah, on Nectandra sp.," 28.11.1919 (A. A. Abraham per G. EH. Bodkin). The test of the female resembles that of the young forms of P. nitens, Ckll., but the fine vertical striae are wanting and the surface is much more uneven. In the female of P. nitens the broad margin is wanting; the integument is uniformly membranous after maceration in KOH, and minute vestigial legs are present. Male puparia not observed. 186 ROBERT NEWSTEAD. Pulvinaria brevicornis, sp. nov. Female, adult. More or less oval in outline and highly convex, or sub-hemispherical ; generally with two longitudinal rows of rather deep pits, one on each side of the median line. Colour, in alcohol, varying from pale buff to pale castaneous ; some are unicolorous, others with two interrupted longitudinal black limes following the course of the pits, the outer line, in some examples, giving off lateral lmes on the abdommal segments. Integument thin. Antennae (fig. 8, a) relatively very short and robust, equal in length to the anterior tibio-tarsal segments together; of 6 segments (the articulations somewhat ill-defined in some examples); 5th and 6th each with a rather long slender spine. Legs (fig. 8, b) short and very robust. Stigmatic clefts obsolete ; spines three (fig. 8, c), stout, the central one generally slightly Fig. 8. Pulvinaria brevicornis, Newst., sp. n., 9; a, antenna; b, leg ; ¢, ¢,, stigmatic and marginal spines ; d, anal lobes ; e, preanal gland-pores; f, ventral tubular ducts; g, stigmatic gland-pores. Male, pronymph : h, antenna; 1, leg. longer than the laterals, but in some instances it is of the same length as the others. Marginal spines (fig. 8, c,) simple, pointed and rather widely separated. Anal cleft short, or two to three times longer than the lobes. Anal ring of 10 hairs. Anal lobes (fig. 8, d) with the proximal margin much longer than the distal margin ; apex with several hairs. Dorsum, in heavily stamed preparations, with widely separated, broadly oval or subcircular cells. Venter with innumerable circular gland-pores, the tubular connections of which (fig. 8, f), are suddenly truncate on one side near the proximal end and furnished with a rosette-like extension. Length, 3-3°75 mm ; width, 1°75-3 mm. British Guiana: Turkeyn, East Coast, on Avicennia mitida, 22.vi.1917 (G. H. Bodkin). The mtegumental characters of this msect are unusual, and should serve, together with the form of the antennae, the anal lobes and stigmatic spmes, to distinguish it from its allies. OBSERVATIONS ON SCALE-INSECTS. 187 Pulvinaria broadwayi var. echinopsidis, nov. Female, adult. Ovate, usually very slightly narrowed in front. Antennae of eight segments ; 3rd a little longer than the 2nd ; formula 8, 2, 8, 1 (4, 5, 6, 2) or 3, 2, 8, 4 (5, 6, 7). Legs robust and relatively long. Lower digitules strongly merassate proximally and broadly dilated distally. Stigmatic clefts (fig. 9, a) very shallow ; spines three, all of them stout, the middle one usually a little more than twice the length of. the laterals. Marginal spines (fig. 9, a) set rather closely together and of two types—one relatively short and simple, the other longer and slightly divided Fig. 9. Pulvinaria broadwayi var. echinopsidis, Newst., nov., 9; a, stigmatic cleft with marginal and stigmatic spines ; b, anal lobes. at the tip. Dorsum without glands or cellular structures; venter crowded with glandular tubes, more especially so m the abdommal region. Anal lobes (fig. 9, 6) rather narrow and furnished distally with several long hairs. Anal ring of eight hairs, of which one pair is much smaller than the rest. Length, 2-15; width, 1-1°3 mm. Ovisac. More or less rounded and formed of loose and somewhat brittle material, at the side of which the shrivelled body of the female rests. Greatest width, 15-3 mm. British GUIANA: Botanic Gardens, Georgetown, on Echinopsis latiflora, 1x.1918 (G. EH. Bodkin and H. Morrison). Lecanium subacutum, sp. nov. Female, adult (fig. 10, a). Colour of dead examples pale dusky yellow. Flat and very thin; dorsum wrinkled, the wrmkles at the margin radial. Form long and narrow ; extremities subacute ; the length three times as great as the greatest width ; one side of the body is usually more or less straight, the other strongly arched. Antennae (fig. 10, 6) of six segments, the third very long, and almost equalling the length of the 2nd, 4th, 5th, and 6th together. Legs long and slender. Stigmatic clefts (fig. 10, c) shallow ; spines three, all of them very stout and bluntly pointed ; the central one rather flattened and more than twice the length of the laterals. Marginal spines (fig. 10, c) simple, rather stout and strongly curved backwards ; they are placed very closely together, so that the tip of each spme almost reaches the strongly curved portion of the spine below it. Anal lobes (fig. 10, d) long and narrow ; the length equallmg that of the 3rd segment of the antennae. Dorsum with numerous large circular gland-pores (fig. 10, e, e) having strongly chitinised rims and fine granular centres; they are irregularly scattered over a relatively broad area between the anal lobes and the antennae. Dorsal spines (fig. 10,e, e) short, stout, and bluntly pointed. Dermal cells absent. Submarginal tubercles in two 188 ROBERT NEWSTEAD. pairs: one anterior, the other posterior. Anal cleft relatively short, and from one-seventh to one-ninth the length of the body. Length, 2-4-3:1 mm. Young females are relatively narrower than the adults; but do not otherwise differ from them. Male puparwm. Glassy white. Very elongate ; outline similar to that of a young female ; median plate or “ coronet”’ long and very narrow ; stigmatic ridges well defined ; margin with one division along the line of the lower pair of stigmata. Length, 1:8-1°9; width, 0°5-0°6 mm. = Pore P f a - \ mY | | | \ | / / f 7 Sade” \ mt Fig. 10. Lecaniwm subacutum, Newst., sp.n., 2; a, adult 2; b, antenna; c, marginal and stigmatic spines; d, anal lobes; e, e, dorsal gland-pores and spines, Ucanpa: Jana Isl., Sesse Islands, Lake Victoria, on Coffea robusta, 9.x.1918 ; Butumira Isl., Sesse Islands, on the leaves of an unknown plant, 12.x.18 (C. C. Gowdey). In both imstances this species was living in association with Aspidiotus articulatus var. magnospinus, Newst. Lecanium (Eucalymnatus) decemplex, sp. nov. Female, adult. Circular, or more or less so; fiat and very thin. Colour trans- lucent amber-yellow, often with a tinge of red or pale castaneous. The whole of the dorsum covered with a thin hard glassy test, the presence of which is extremely difficult to detect and which is equally hard to detach. Boiled in KOH the female OBSERVATIONS ON SCALE-INSECTS. 189 (fig. 11, a) presents the following morphogical details: Dorsum divided into five bilateral plates—two cephalic, three thoracic, and two abdominal; the sutures separating the cephalic from the first thoracic plates terminating at the stigmata ; the other sutures are connected with the mesal one (these sutures in the dried examples appear as well defined narrow ridges). Mesal suture between the anal lobes and the rostrum with large circular gland-pores (fig. 11, 6). Antennae of six segments; the third almost as long as the 1st, 4th, 5th and 6th together. Legs well developed ; lower digitules large and strongly incrassate. Anal lobes (fig. 11, c) somewhat triangular ; inner margin longest; distal margin shortest; the sclerites beneath (fig. 11, d,d) stout and somewhat spine-like. Anal cleft deep,and apparently partly fused but separable. Stigmatic clefts (fig. 11, e, e) small, but deeply invaginated; spines three, very robust, and blunt at the tips. Marginal spines (fig. 11 e. e) simple, and set rather widely apart. Submarginal gland pores (fig. 11, f) very small, number- ing from six to eight on either side. Oval cells can be seen, near the margin, in old and well stained examples. Length, 3°4-3°7 ; width, 3-2-3-°7- mm. Fig. 11. Lecaniwm (Hucalymnatus) decemplex, Newst., sp. n., 2; a,adult 9; b, mesal gland-pores ; c, anal lobe; d, d, sclerites of anal lobes ; e, e, stigmatic clefts and spines. Male Puparium. Broadly ovate; divided into eleven plates: two median, one cephalic, and four bilateral; the lateral plates with partial subdivisions. Length, 1°8; width, 1:4 mm. British Guiana: Ayaria, Thuraka Lake, Ituribisci Creek, Essequebo, on leaves of Lecythis sp., 6.x.1918 (G@. H. Bodkin). This somewhat remarkable species evidently belongs to the subgenus Hucalymnatus, its distinguishing features being the small number of plates into which the dorsum is divided. Lecanium inquilinum, sp., nov. Female, adult. Ovate or elongate and highiy convex ; more or less circular when mounted under pressure. Integument pale brown, but thin and transparent after maceration in KOH. Antennae (fig. 12, a) of eight segments. Legs relatively robust 190 ROBERT NEWSTEAD. and rather long ; digitules normal. Anal cleft free and a little more than twice the length of the lobes ; the latter (fig. 12, b) rather elongate and obtusely rounded distally. Stigmatic clefts very small ; spines three, all robust and blunt, the laterals almost equal in size to the central one. Marginal spines (fig. 12, c) long and very acute, the tips in many cases appearing almost flagellate; they are placed very closely together and are continuous along the stigmatic clefts. Stigmata with a large trumpet. shaped peritreme, and placed unusually near the margin of the body. There is a large closely packed group of multilocular gland-pores on the inner walls of the anal cleft, close up to the lobes. A few minute gland-pores occur in the densely chitinous patch surrounding the anal lobes. Length, 1°7-2°3 ; width, 1:4-1°8. Female, young adult. Similar to the old adult, but the marginal spines are much more bluntly pointed and the chitinous patch surrounding the anal lobes is wanting. Britisa Guiana: “Cattle Trail Survey,”’ the insects enclosed by ants (Acromyrmex sp.) in small paper nests 1919 (A. A. Abraham per G. E. Bodkin). ALA AU Fig. 12. Lecanium inquilinum, Newst., sp. 1., Q; a, antenna; b, anal lobes; ¢, marginal spines. The form and arrangement of the marginal spines recall those seen in certain species of Pulvinaria; but the species clearly belongs to Lecanium and somewhat resembles the hollow hemispherical species. Taken in association with Lecanium deformosum, sp.n., and Pseudococcus inquilinus, sp. n. Lecanium (Eulecanium) deformosum, sp. nov. bi Female, adult (fig. 13, a-d). Dorsum rather flat; sides relatively thick ; general form very irregular and distorted, some examples being broader than long and others more or less elongate, but the margins in all cases (12 examples) are asymmetrical and often distorted to a marked degree. Antennae (fig. 13, e) of six segments ; the 3rd and 6th longest. Legs with the tibio-tarsal segments either distinctly articulated or partly so, rarely completely fused; leg i. (fig. 13, f) with the tibia generally strongly curved. Anal lobes (fig. 13, 9) very broadly dilated distally. Anal cleft faintly fused, but easily separated after maceration in KOH. Stigmatic clefts small, or seated in faint depressions; spines (fig. 13, h) three, the laterals normally very short, stout, OBSERVATIONS ON SCALE-INSECTS. 191 and obtusely pointed : in one example (fig. 13, 4,) the lower group on one side has the lateral spines as long and as stout as the central one. Marginal spines (fig. 13, 2) simple and hair-like. Dorsal gland-pores minute and widely separated. In well stained specimens the integument of the dorsum is faintly divided into broad plate- like radial areas enclosing numerous irregular pigmented markings. Length, 1:2- 1°8; width, 1:1-1°5, mm. British Guiana: ‘Cattle Trail Survey,” on an unknown plant, the insects enclosed by ants (Acromyrmex sp.) in small paper nests, 1919 (A. A. Abraham per G. E. Bodkin). Fig. 13. Lecanwwm deformosum, Newst., sp. n., 2; a-d, outline of four adults; e, antenna; f, leg; g, anal lobes; h, h,, stigmatic spines ; 2, marginal spines. A small oviparous species, remarkable for its markedly deformed shape, the presence or absence of the tibio-tarsal articulation, and the apparent variability of the stigmatic spines. Found in association with Pseudococcus inqulinus, sp. n., and Lecaniwmn mquilinum, sp.n. Lecanium (Saissetia) nigrum var. nitidum, nov. Female, adult. Usually more or less hemispherical, but some examples are slightly ovate and narrowed in front; margin markedly flattened, and often with regular rectangular patches of silvery secretion ; dorsum smooth and shining; anal lobes usually porrect. Colour varying according to the age of the individual: young forms pale red-brown, old adults rich dark castaneous. Eyes relatively large, black and prominent. Antennae (fig. 14, a) of eight segments; the 3rd scarcely as long as the 4th and 5th together. Legs rather slender. Derm cells forming a closely reticulated pattern as in Lecamwum nigrum, Niet. Stigmatic clefts very shallow ; spines three, the central one about five times the length of the laterals, (fig. 14, b). Marginal (C605) rr 192 ROBERT NEWSTEAD. spines (fig. 14, c) short and sub-palmate, the distal portion being broadly flattened and deeply divided. Anal lobes (fig. 14, d) short and obtusely rounded ; inner margin slightly longer than the proximal; external margin strongly arched. Length, 1-8-2 mm. Female, young adult. Flat ; straw-coloured when dry. Form when mounted under pressure somewhat ovate. Antennae (fig. 14, a) with eight segments and similar in form to those of the adult. Legs i. and ii. much longer than leg i. Stigmatic clefts (fig. 14, b) mdicated by a very slight indentation of the margin ; spines three, the laterals minute and acutely pointed. Marginal spines (fig. 14, 6, c) as in the adult. Anal cleft free, its length slightly less than twice the length of the lobes ; the latter (fig. 14, d) as in the old adults. Ucanpa: Bukeke Isl., Sesse Islands, Lake Victoria, on Luzebarziba, 9.x.1918 (C. C. Gowdey). | Fig. 14. Lecanium nigrum var. nitidum, Newst., 92; a, antenna; b, stigmatic and marginal spines; c¢, marginal spines; d, anal lobes. ‘ : Its small size and generally hemispherical form, together with the highly polished integument and the sub-palmate marginal hairs, are the distinctive features of this well-marked variety. Platysaissetia montrichardiae, sp. nov. Female, old adult. Blackish or sooty brown. (Generally more or less ovate and very slightly narrowed in front ; but occasionally the outline is irregular, indented, or markedly asymmetrical; dorsum very low convex or almost flat, scabrous, the minute elevations often carrying small particles of the test of the young female. Vertical sides markedly shallow. Pseudo-margin relatively faintly produced. Tubular glands at the extreme margin. Stigmatic clefts rarely traceable. Venter hollow, but more or less covered with a rather thick pellicle of white wax. Dermal - cells (fig. 15, a, from an unstained preparation) very irregular in outline, each pale area with a distimct pore at the extreme edge ; the walls of each cell area thick, dark, and very regular. Other details as in the young adult. Length 4-5:2; width 2°9-3°9. mm. OBSERVATIONS ON SCALE-INSECTS. 193 Female, young adult. Pale brown to dull castaneous. Dorsum flat and covered with a dirty white, glassy test, consisting of mmute plates, which collectively present a faintly imbricated appearance. The test is easily deciduous, and when removed the integument presents a polished appearance. Antennae (fig. 15, 6,5) of eight seg- ments, of which the 4th, 5th and 8th are the longest ; formula: (4, 5, 8) 3, 2, (6, 7) 1 or 5 (4, 8, ) 3 (2, 6, 7) 1; terminal hair of great length. Legs relatively slender ; tarsus of anterior pair (fig. 15, c) with a well defmed constriction ; coxa and trochanter each with a very long hair. Stigmatic clefts obsolete ; spines (fig. 15, d,d) three, the central one equal in length to the 4th and 8th segments of the antennae ; lateral spines poimted and about one-fourth the length of the central one. Marginal spines (fig. 15, d) slightly shorter and stouter than the lateral stigmatic spines ; they are Fig. 15. Platysaissetia montrichardiae, Newst., sp. n., 2; a, dermal pores and “‘cells.””> Youngadult 9; b, b, antennae ; c, leg; d, d, stigmatic spines ; e, post anal gland-pores ; f, leg of larva. separated by a distance equal to three or four times their length. Dermal gland- pores irregularly ovate, large, placed close together, and most conspicuous towards the margin. Anal lobes with the outer margin strongly arched ; and surrounded by a narrow wall of dense chitin. Anal cleft fused ; from one-fifth to one-sixth the length of the body. Postanal glandular pores (fig. 15, e) relatively large and circular, forming a broad scattered group extending as far forward as the mentum. Length, 2°6-3°3 mm. Female, second stage. Differs from the adult chiefly in the following details : Antennae of six segments, of which the third is much the longest. Postanal gland- pores absent ; postanal bristles in three pairs. Length, 0°7-0°8 mm. (C605) H2 194. ROBERT NEWSTEAD. Male Puparium. Oblong, opaque, glassy, white; anal cleft distinct; surface composed of minute rough polygonal plates ; the marginal series forming a roughly serrated fringe. Length, 1°5 mm. Larva. Differs from the second stage female in the followmg details: Antennae with the 3rd and 6th segment equal and longest. Legs (fig. 15, f) with the distal femoral bristle of great length ; tarsal digitules markedly unequal in thickness and length respectively, the longer one arisimg from the tarsus some distance behind the smaller one; digitules to the claw normal. Anal rmg with six hairs. Length, exclusive of the caudal bristle, 0°5 mm. Britiso Guiana: Ikruaka Lake, Essequibo, on Montrichardia aculeata, 11. m.1917 (G. EB. Bodkin). A very heavy infestation, so much so that the insects covered a very large propor- tion of the branches. A somewhat remarkable species, distinguishable, in the adult female, by the curious character of the derm and the relatively short anal cleft (0°7-0°8 mm.); in the second stage female by the absence of marginal cylindrical ducts to the glands ; and in the larva by the unusual character and relative position of the tarsal digitules. Aspidiotus longispina, Morgan. Aspidiotus longispina, Morgan, Ent. Mo. Mag. xxv, p. 352, pl. v, fig. 1 (1889). Morganella maskellt, Cockerell, Bull. 6, T.S., U.S.A. Dept. Agric., p. 22 (1897). Hemaberlesia longispina (Morg.), Leonardi, Riv. Pat. Veg. vi, p. 120, fig. 4 (1897). Asprdiotus longispina var. ornata, Maskell, Trans. N.Z. Inst., xxx, p. 225 (1897). Aspridiotus (Morganella) maskelli (Ckll.) Brain, Bull. Ent. Research, ix, p. 136, pl. vii, fig. 109 (1918). Female Puparwm (fig. 16).—This, when perfect, is narrowed, strongly produced, and slightly involute posteriorly, resembling the long curved toe of an Oriental slipper in miniature. This very remarkable appendage is composed of both dorsal and ventral portions of the capsulate puparium ; but it is rarely found intact in examples which have been submitted to even light pressure, as it 1s very brittle and readily breaks away. Fig. 16. Aspidiotus longispina, Morgan, 2, puparium. Male Puparwum. Elongate, exuviae terminal or sub-termimal. Colour as in the female puparia. Male. Not differmg in its morphological characters from typical members of the genus. British Guiana: Botanic Gardens, Georgetown, on papaw, 1919 (G. EH. Bodkin). OBSERVATIONS ON SCALE-INSECTS. 195 The synonymy given above is, I believe, correct, the determmation bemg based upon an examination of material received from both Morgan and Maskell. A co-type ° from Morgan in my collection was origmally mounted in Canada balsam without staiming, so that the true form of the strikmgly characteristic squamae on the pygidium of the 2 could not be detected and was thus overlooked by Morgan in the first instance, and subsequently also by Leonardi, to whom I sent my example for examination, and for the purpose also of figuring it m his memoir (I.c.). In 1906 I had occasion to stain the female given to me by Morgan for comparison with examples submitted to me by Kotinsky, on Ficus sp., from Honolulu. I then found that the squamae were strikingly different from what they appeared to be when unstained. Subsequently I stained and mounted an example of Maskell’s A. longispina var. ornata and found it to be specifically identical with Morgan’s co-type female. Clearly therefore the specific names maskella of Cockerell and ornata of Maskell must sink. Aspidiotus (Chrysomphalus) apicatus, sp. nov. Female Puparium. More or less circular, convex and very thick ; covered with a relatively thick epidermal layer of the bark ; colour, when denuded, opaque black, larval exuviae nude, shining black, forming a well defined nipple ; second exuviae black ; ventral surface shining black. Ventral pellicle rather stout ; white or dusky white, with a dark brown or blackish periphery. Diameter, 1°6-1°9 mm. Fig. 17. Aspidiotus (Chrysomphalus) apicatus, Newst., sp. nz, 2; a,adult; 6, thoracic spine ; ¢, pygidium ; d, paraphyses of do. ; e, fringe of do.; f, anal plates, etc., of larva. Female, adult (fig. 17, a). Broadly ovate, with a well marked and rather highly chitmised, cephalic projection, and immediately below it a minute spiny process (fig. 17, 6). Margin of body more or less strongly and finely crenulated. No parastig- matic glands. Pygidium (fig. 17, c) broadly rounded, with a well marked callosity extending along it proximally, from which there arises on either side a somewhat 196 ROBERT NEWSTEAD. lanceolate sclerite (fig. 17, d). Margin (fig. 17, e) with nine pairs of well defined para- physes and a few rather indistinct ones beyond them ; those arising from the distal lateral portions of the lobes with the sides almost parallel and the ends more or less suddenly truncate. Lobes in four pairs; the 2nd, 3rd and 4th pairs dentate. Length, 1-1:1 mm. Male Puparium. Broadly ovate, with the ends equally rounded ; colour dark brown or blackish, but covered with the epidermal layer of the bark ; larval exuviae black and more or less exposed; under surface dark brown or piceous, the central area partly covered with a white mealy secretion. Uength, 9'5-0°S mm. Larva. Broadly ovate. Antennae of five segments, of which the 5th is much the longest and distinctly ringed. Pygidium (fig. 17, f) with two pairs of finely dentate lobes ; the median pair much the larger and converging towards the middle line. British GUIANA: Enmore Forest, East Coast, Demerara, on Avicenmia nitida, 1x.1918 (G. H. Bodkin and H. Harrison). Aspidiotus (Chrysomphalus) umboniferus, sp. nov. Female Puparium. More or less circular to broadly pyriform; flat, thin, and somewhat transparent; pale reddish-brown to pale chocolate-brown. Exuviae Fig. 18. Aspidiotus (Chrysomphalus) wmboniferus, Newst., sp. n., a, larval and nymphal pellicles; b, adult 9° after treatment in KOH; ¢, unrestored 9; d, antenna of ¢ ; e, pygidium of @. central or subcentral ; the first nipple-like and black when denuded ; the second, sooty brown. Ventral pellicle very delicate and composed generally of concentric rings. Diameter, 1°6-2 mm. Male Puparium. Elongate ovate, slightly narrowed posteriorly. Colour similar to that ofthe female. Length, 1 mm. Femaie (fig. 18, 6, c). Very broadly pyriform; width of cephalothorax slightly greater than the entire length of the body ; metathorax with a relatively large, blunt, marginal tubercle, which, together with the margin below it, is strongly OBSERVATIONS ON SCALE-INSECTS. Us Vie chitinised. Pygidium (fig. 18, e) unusually narrow and bluntly pointed. Circum- genital glands in four groups, the posterior lateral groups placed in a line with the vaginal orifice ; formulae of four individuals :— 8-6 66 7-8 6-6 66 66 56 6-6 Lobes in three pairs ; median pair somewhat quadrate, broader than long; 2nd and 3rd pairs slightly shorter than the first and more than twice as broad as long. Squamae between the»median and 2nd lobes, very short; the second of the proximal pair, immediately beyond the third lobe, trifid. Paraphyses in six pairs, all rather narrow but clearly defined. Dorsal gland-pores small and few in number, with the tubular ducts filiform. British Guiana: Ayaria Creek, Fssequibo, on Lecythis sp., 6.x.1918 (G. H. Bodkin). In the structurat details of the margin of the pygidium this species very closely resembles A. perseae, Comst., but it has a much more strongly pointed pygidium, fewer circumgenital glands, and an extra pair of paraphyses ; these characters, taken in conjunction with the large metathoracic tubercles and the strongly chitinised margin below them, form the salient features of this Coccid. Aspidiotus (Chrysom- phalus) paulistus, Hempel, also possesses very distinct cephalo-thoracic tubercles ; but in this species there are only two pairs of paraphyses and the pygidium is relatively shorter and much more rounded distally. Aspidiotus (Selenaspidus) articulatus var. magnospinus, nov. This variety (fig. 19, a) differs from typical Aspidiotus articulatus, Morgan, in the following details : Cephalothoracic margin finely but distinctly serrated. Thoracic Fig. 19. Aspidiotus articulatus var. magnospinus, Newst., 9; a, adult; b, thoracic spine; e¢, fringe of pygidium; d, thoracic tubercle of typical © articulatus for comparison. 198 ROBERT NEWSTEAD. spine (fig. 19, 6) relatively very large, curved and rather acutely pointed, the contour gradually merging into the cephalo-thoracic margin, and considerably longer than its greatest width. In arteculatus, the thoracic spine (fig. 19, d), drawn to the same seale as in fig. b, is suddenly produced, short, and bluntly pointed, and its length about equal to its greatest width, Ucanpa: Bufumira Isl., Sesse Islands, Lake Victoria, on the leaves of an unknown plant, 12.1x.1918 (C. C. Gowdey). Aspidiotus (Selenaspidus) kamerunicus, Lind., has a similar thoracic spine and a serrated margin, but in this species the broad palmate squamae, between the second pair of lobes and the spiny process, are replaced by squamae of a much narrower type. Aspidiotus (Odonaspis) rhizophilus, sp. nov. Female Puparium. Dense, hard and capsulate, but the two halves slightly separated posteriorly. Form irregular, but old examples are slightly narrowed and produced posteriorly ; convex above and flat beneath. Larval exuviae generally Fig. 20. ) ® j J . , wt t di ‘i 4 iv a j 2 y Fi J rin E } } if ] % j s t r ee i Ph “| - : f a) es ; yal i "2 . a ie ' i m i } ? 4 } Ny / = i. 2 8 : y \ } ‘ Ne Lat & tiled 2h wi” agpegebchenge Baty ‘ ch oid eee wut awit neath . ay ie were a a ee eT @ a tn hs Raed ay We . ie eh | ay irom Lalasd aiid al es + é ; F ‘ nine 4 et od He Wl - el SS Bea Pe: Eae V a Y nat me we nod wy ei ay fy M IBULi. JEN, JIRESEWRGEH., Wok, 2X. IJPARr 2. Pini SOWIE Fig. 1. Clypeococeus nena gel, Ckll., Fig. 2. Chionaspis laniger, Newst., adult females, x 234. puparia of females, x 3. Fig. 3. Walkeriana digitifrons, Newst., Fig.4. Phenacoceus ballardi, Newst.,. adult females, x 2. young and adult females, x 3. — : ~ ts. 209 COLLECTIONS RECEIVED. The following collections were received by the Imperial Bureau of Entomology between 1st April and 30th June, 1919, and the thanks of the Managing Committee are tendered to the contributors for their kind assistance :— Dr. W. M. Aders, Government Economic Biologist :—13 Culicidae and 38 other Diptera ; from Zanzibar. Mr. T. J. Anderson, Government Entomologist :—15 Culicidae, 10 Haematopota, 1 Tabanus, 490 other Diptera, 596 Coleoptera, 10 Planipennia, 4 Isoptera, 189 Rhynchota, 37 Orthoptera, 45 Odonata ; from British Hast Africa. Mr. B. N. Bandyopadhyay :—23 Culicidae, 5 other Diptera, and 4 Lepidoptera ; from British East Africa. Lieut. P. J. Barraud :—2 Chrysops, 4 Tabanus, 1 Asilid, 6 Hymenoptera, and 19 Coleoptera; from Macedonia: 842 Culicidae, 36 slides of Culicid larvae, 3 Chrysops, 1 Stomoxys, 1 Hippoboscid, 116 other Diptera, 24 Hymenoptera, 124 Coleoptera, 5 Planipennia, 50 Lepidoptera, 5 Trichoptera, 6 Isoptera, 50 Anoplura, 9 Rhyn- chota, 24 Orthoptera, 13 Odonata, 1 Spider, 3 Centipedes, 3 Wood-lice, and 13 Karthworms ; from Mesopotamia. Mr. G. E. Bodkin, Government Economic Biologist :—8 Diptera, a number of Ants, 13 Chalcids bred from Coccidae, 12 other Hymenoptera, 27 Coleoptera, 4 species of Coccidae, 22 other Rhynchota, 4 Orthoptera, 4 Ticks, 4 Spiders, 2 Centipedes, 6 Millipedes ; from British Guiana. Mr. J. H. Burkill, Director of the Botanic Gardens :—4 Agromyzid Diptera and 16 larvae and pupae ; from Singapore. Mr. J. E. A. den Doop :—20 Weevils, with 17 larvae and 21 pupae; from Java. Durban Museum :—90 Coleoptera ; from Natal. The Government Entomologist, Madras :—69 Coleoptera and 6 Rhynchota ; from South India. Mr. C. C. Gowdey, Government Entomologist :—24 Ticks ; from Uganda. Mr. EK. Hargreaves :—146 Culicidae, 1 Pangona, 3 Chrysops, 1 Haematopota, 11 Tabanus, 25 other Diptera, 39 Hymenoptera, 110 Coleoptera, and 3 Orthoptera ; from Taranto, Italy. The Imperial Entomologist, Pusa :—3 Stephanidae ; from India. Imperial Institute :—9 silk cocoons damaged by Anthrenus ; from Morocco. Mr. 8. Leefmans, Instituut voor Plantenziekten, Buitenzorg :—35 Coleoptera and 58 Rhynchota; from Java. Mr. F. Muir :—5 Weevils and 2 larvae, and 29 Delphacidae. Prof. G. H. F. Nuttall, F.R.S. :—59 Coleoptera, from Burma: and 44 Culicidae, 7 Ceratopogoninae, and 38 other Diptera ; from British East Africa. Mr. W. H. Patterson :—10 Coleoptera and 2 larvae; from the Gold Coast. Mr. R. Swainson-Hall :—2 Coleoptera ; from Portuguese Congo. Mr. R. Veitch, Entomologist, Colonial Sugar Refinmg Company :—9 Diptera, 25. Hymenoptera, 9 Coleoptera, 1 Lepidopteron, and 14 Rhynchota; from Fiji. 210 COLLECTIONS RECEIVED. Dr. W. G. Watt, Medical Officer of Health :—195 Culicidae and 107 larvae; from the Gold Coast. Mr. C. B. Williams :—12 Diptera, 616 Ants, 3 Moths, 306 Rhynchota, 3 Thrips, 1 Centipede, and 1 Spider; from Trinidad. Mr. Rodney C. Wood :—9 Tabanus, 45 other Diptera, 101 Coleoptera, and 2) 3 Rhynchota; from Nyasaland. IMPERIAL BUREAU or ENTOMO Loc. a a ; HON ORARY. COMMITTEE OF MAN AGEMENT a THE VISCOUNT HARCOURT, ‘Chairman, | Limvr. -CotoneL A. ALCOCK, C.LE., F.R. g. Masor E. E. AUSTEN, D.S.0. Dre. A. G. BAGSHAWE, CMG. Masor-GEnzRAL Sm JOHN R. BRADFORD, K.0.M. G3 F. RS. “ee th “ Masor-Genzrat Sr DAVID BRUCE, pe BRB? 65 ae ene Me. J. CO. F. FRYER. Da S. F. HARMER, F.RS._ | eh ea Pror, H, MAXWELL LEFROY. — | 8 ie Hon. E. LUCAS. ’ Da R. STEWART iieptenii Sm JOHN MoFADYEAN. _Sm PATRICK MANSON, G.C.M.G., F-.R.S. Si DANIEL MORRIS. K.C.M.G. | Prov. R. NEWSTEAD, F.RS. OR. 8 Coa Paor, G.H. F. NUTTALL, WR O ree Pror. E. B. POULTON, F.R.S. SER Lrzot. -CotonEL Str DAVID PRAIN, C. MG, C.I. B. F.RS. Sm H. J. READ, K.O.M.G:,C.B. i Aa aeaas ay Hon. N.C. ROTHSCHILD, = Hs PEN ce Dr. HUGH SCOTT. Ae See Oa Og Hk a en nt AOE, SHIPLEY, RR Se oF me 5 are ee, Me. Ry A. C.-BPERBING “OMEGs gibt tc hoe" Bs Aas Sfe STEWART STOCKMAN, 2:2, 0008 (2 Vas eae ‘Ma Bi Vi THROBALD Se oS gl ee es C. WARBURTON. ve am mea mR thane tae, | | ‘Director. Pg pind hy ‘ gee ee) Dr. GUY AK. MARSHALL Ae AE EN | _ Besistant Director. NC Ree ane kas peer CS ae ies | Secretary. i. a . . Dat, A. ee ©. FARKINSON. 211 NOTES ON THE ETHIOPIAN FRUIT-FLIES OF THE FAMILY TRYPANEIDAE, OTHER THAN DACUS.—III.* By Prof. M. Brzz1, Turin, Ltaly. (PLates XVII-XVIIT.) The new material received from the Imperial Bureau of Entomology through the kindness of the Director, Dr..G. A. K. Marshall, enables me to prepare this third paper on the Ethiopian fruit-flies. It is a proof of the great richness of the African Trypaneid fauna that of over 41 species received only 10 were referable to previously described species, and that no less than 8 new genera have been erected, most of them for very strikingly characterised forms. * * * x x a The TRYPANEIDAE at present known from the Ethiopian Region, with the addition of the new forms here described for the first time, are enumerated in the following systematic catalogue, which numbers 216 species. I. Subfamily DACINAE. 1. TrRipacus, Bezzi. lounsburyi, Coq. sphaeristicus, Speis. fuscovittatus, Grah. armatus, F. fuscatus, Wied. nebulosus, Walk. d’emmerezi, Bezzi. bivittatus, Big. (pectoralis, Walk. ; bipartitus, Grah.; cucumarius, Sack ; armatus, Bezzi). momordicae, Bezzi (bipartitus, Bezzi). eburneus, Bezzi. xanthopterus, Bezzi. humeralis, Bezzi. disjunctus, Bezzi. flavicrus, Grah. scaber, Loew. stylifer, Bezzi. punctatifrons, Karsch. 2.;Dacus, Ee. hamatus, Bezzi. trigonus, Bezzi. inornatus, Bezzi (modestus, Bezzi). oleae, Gmel. mesomelas, Bezzi. semisphaereus, Beck. blepharogaster, Bezzi. rufus, Bezzi. annulatus, Beck. erythraeus, Bezzi. mochii, Bezzi. woodi, Bezzi. macer, Bezzi. immaculatus, Coq. brevis, Coq. fieicola, Bezzi. bistrigulatus, Bezzi. brevistriga, Walk. brevistylus, Bezzi. vertebratus, Bezzi. binotatus, Loew. bistrigatus, Loew. ciliatus, Loew. sigmoides, Coq. africanus, Adams. sexmaculatus, Walk. longistylus, Wied. kingt, Frogg.). (testaceus, Macq. ; Il. Subfamly ADRAMINAE. 3. MERACANTHOMYIA, Hendel. antennata, Hend. * For Part I, see Bull. Ent. Res. vili, pp. 215-251; for Part Il, ibidem ix, pp. 13-46. 1,000. 3.20. B.&F.Ltd. G.11/4. A (637) W+.P4/140. 4. SOSIOPSILA, Bezzi. trisetosa, Bezzi. Py PROF. M. BEZZi, Ill. Subfamily CERATITINAE. 5. COELOPACIDIA, End. strigata, Bezzi. melanostigma, Bezzi. madagascariensis, End. 6. CONRADTINA, End. longicornis, End. acrodiauges, Speis. acroleuca, Wied. suspensa, Bezzi. 7. CELIDODACUS, Hendel. obnubilus, Karsch (apicalis, {Hend.). coloniarum, Speis. 8. CARPOPHTHOROMYIA, Austen. vittata, F. sceutellata, Walk. tritea, Walk. pseudotritea, Bezzi. superba, Bezzi. 9. LEUCOTAENIELLA, Bezzi. guttipennis, Bezzi. trispila, Bezzi. pentaspila, Bezzi. 10. CHELYOPHORA, Rondani. magniceps, Bezzi. 11. BISTRISPINARIA, Speiser. fortis, Speis. 12. CLINOTAENIA, Bezzi. grata, Wied. anastrephina, Bezzi. 13. CERATITIS, McLeay. capitata, Wied. catoiri, Guér. 14. PTERANDRUS, Bezzi. anonae, Grah. (pennipes, Bezzi). colae, Silv. rosa, Karsch. fasciventris, Bezzi. rubivorus, Coq. volucris, Bezzi. ? penicillatus, Big. - 15. PARDALASPIS, Bezzi. morstatti, Bezzi. punctata, Wied. brémei, Guér. melanaspis, Bezzi. cosyra, Walk. (giffardi,) Bezzi). silvestril, Bezzi. senegalensis, Macq. antistictica, Bezzi. flexuosa, Walk. quinaria, Bezzi. stictica, Bezzi. aliena, Bezzi. 16. PERILAMPSIS, Bezzi. pulchella, Austen. formosula, Austen. * The genus Cladoderris, Bezzi, is better placed in the ORTALIDAE,,and will be dealt with in a forthcoming paper on this family. 17. HopLoLopHa,"Bezzi.. cristata, Bezzi. 18. TRIRHITHRUM, Bezzi:. lycu, Coq. nitidum, v. Rod. albonigrum, End. validum, Bezzi. nigrum, Grah. (obscurum, End.).. gagatinum, Bezzi. occipitale, Bezzi. bimaculatum, v. Rod. nigerrimum, Bezzi. leucopsis, Bezzi. inscriptum, Grah. (coffeae, Bezzi).. ? albomaculatum, v. Rod. 19. XANTHORRHACHISTA, Hendel... alata, Beck. (cephalia Hend.). 20. THEMARICTERA, Hendel, flaveolata, F. rufipennis, Hend. laticeps, Loew. 21. THEmMARA, Walker. fallacivena, End. 22. BARYGLOSSA, Bezzi.*- histrio, Bezzi. 23. PTILONIOLA, Hendel.. neavel, Bezzi. preussi, Hend. tripunctulata, Karsch. (Nov. GENUS.) guttatolimbata, End. 24. RHACOCHLAENA, Loew... fasciolata, Loew. pulchella, Bezzi. hammersteini, End. 25. PHORELLIA, R. Desv:. brunithorax, R. D. tristriata, Karsch. 26. TaomytA, Bezzi.. marshalli, Bezzi. ocellata, Lamb. 27. NotomMa, Bezzi,, bioculatum, Bezzi. (Nov. GENUS.) jucunda, Loew. 28. AcrpiA, R. Desv. seychellensis, Lamb. fossataeformis, Bezzi. homogenea, Bezzi. 29. ** OcnEROS,”’ O. G. Costa.. mundus, Loew. excellens, Loew. sinuatus, Loew. undatus, Bezzi. bigemmatus, Bezzi. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 213 III. Subfamily CERATITINAE.—cont. 30. OCNERIOXA, Speiser. pennata, Speis. woodi, Bezzi. disereta, Bezzi. gracilis, Loew. 31. TErrparRitis, Latr. indecora, Loew. © vernoniicola, Bezzi. 32. CarpomytA, A. Costa. incompleta, Beck. 33. CRASPEDOXANTHA, Bezzi. marginalis, Wied. manengubae, Speis. 34. TeRELLIA, R. Desvoidy. ? hysia, Walk. 35. ALLOTRYPES, Bezzi. brevicornis, Bezzi. 36. AcrIuRA, R. Desv. perspicillaris, Bezzi. angusta, Loew. semiangusta, Bezzi. caeca, Bezzi. oborinia, Walk. tetrachaeta, Bezzi. capensis, Rond. ternaria, Loew. binaria, Loew. 37. SPHENISCOMYIA, Bezzi. sexmaculata, Macq. neavel, Bezzi. 38. TEPHRELLA, Bezzi. bezziana, End. nigricosta, Bezzi. cyclopica, Bezzi (w-fuscum, End.). tephronota, Bezzi. rufiventris, Bezzi. sexfissata, Beck. hessii, Wied. IV. Subfamily TRYPANEINAE. 39. PLATENSINA, Enderlein. diaphasis, Big. (Nov. GENUS). lunifera, Loew. 40. EuTRETOSOMA, Hendel. oculatum, Hend. frauenfeldi, Schin. bipunctatum, Loew. millepunctatum, Bezzi. polygramma, Walk. 41. ELAPHROMYIA, Bigot. adatha, Walk. (melas, Big. ; wlula, Loew). 42. SpaTHULINA, Rondani. semiatra, Loew. bioculata, Bezzi. parceguttata, Beck. acrosticta, Bezzi. aldabrensis, Lamb. margaritifera, Bezzi. 43. PLIOMELAENA, Bezzi. brevifrons, Bezzi. strictifrons, Bezzi. 44. EUARESTA, Loew. planifrons, Loew. amplifrons, Bezzi. 45. Ensina, R. Desv. sororcula, Wied. gladiatrix, Bezzi. myilopitoides, Bezzi. anceps, Loew. magnipalpis, Bezzi. dubia, Walk. ignobilis, Loew. siphonina, Bezzi. 46. Eurtsia, Meigen. perpallida, Bezzi. discipulchra, Bezzi. practexta, Loew. caffra, Loew. dissoluta, Loew. tristrigata, Bezzi. cyana, Walk. 47. CAMPIGLOSSA, Rondani, perspicillata, Bezzi. 48. CAMAROMYIA, Hendel. acrophthalma, Bezzi. ? helva, Loew. 49, ACANTHIOPHILUS, Becker, helianthi, Rossi. ? ochraceus, Loew. 50. TRYPANEA, Schrank. subcompleta, Bezzi. hemimelaena, Bezzi. peregrina, Adams (wrophora, Bezzi). augur, F'rauenf. tristicula, Hend. * auguralis, Bezzi. hexapoda, Bezzi. confluens, Wied. aucta, Bezzi. repleta, Bezzi. aira, Walk. amoena, Frauenf. decora, Loew. diversa, Wied. * This recently described species was erroneously included by Prof. Hendel in his great work on the South American Trypaneids, 1914, p. 82; but it is from Mozambique, and was accidently omitted by me in my previous paper ; it is very near augur, has about the same wing pattern, and has likewise only two scutellar bristles. (637) »2 214 ' PROF. M. BEZZI. V. Subfamily SCHISTOPTERINAE. 51. PERIRHITHRUM, Bezzi. 53. RHOCHMOPTERUM, Speiser. marshalli, Bezzi. neuropteripenne, Speis. 52. RHABDOCHAETA, de Meijere. 54. SCHISTOPTERUM, Becker. spinosa, Lamb. moebiusi, Beck. neavel, Bezzi. * * * * * * The addition of the very interesting new genus which follows makes necessary a | modification of the beginning of the table of genera in my first paper, p. 216, nos. 1-5, which may be modified as follows :— 1(6). Thoracic chaetotaxy incomplete, hm., prst., dc. and st. being always wanting (and also the prsc. in African forms). a(d). Arista bare; abdomen short and broad, usually broader than the thorax ; Wings with the first three longitudinal veins closely approximated, the anterior cross-vein long and oblique, the second basal cell dilated and the anal cell drawn out into a very long point oe Subfam. Dactnar. b(c). Only two sa. present, the anterior one being entirely wanting Dacus, ¥. (8. sir-): c(b). Three sa., the anterior one being well developed .. .. Tridacus, Bezzi. d(a). Arista plumose; abdomen long and linear, more narrow than the thorax ; wings with the first three veins not approximated, the anterior cross-vein short and placed perpendicularly, the second basal cell not dilated and the anal cell with a short point .. .. Subfam. ADRAMINAE. e(f). Femora not spinose beneath; thoracic suture interrupted in the middle ; antennae shorter than the face ; point of the anal cell very short Sosiopsila, gen. nov. f(e). Femora spinose beneath; thoracic suture complete; antennae much longer than the face; point of the anal cell longer .. Meracanthomyra, Hend. Il. Subfam. ADRAMINAE. The present subfamily 1s here accepted in a wider sense than that attributed to it by Hendel in 1914, and corresponding to that adopted by me in 1916.* The subfamily was believed to be almost exclusively Oriental; but the discovery of the following new form shows that it is probably well represented in the Ethiopian Region also. Sosiopsila, gen. nov. The present new genus seems to be allied to the Oriental Neosophira, Hendel (Gen. Insect., 76, pl. i, figs. 57-59 and 64) and to Colobostroter, Enderlein (Zool. Jahrb., xxxi, p. 445, fig. T.), but is distinct in having well developed outer vt., and also in having only a pair of sct. (the basal pair) ; on the wings the third longitudinal vein is not sinuous. Head in front view more broad than high, a little broader than the thorax ; occiput convex and prominent, with developed lower swellings; the head is attached to the thorax about in the samé manner as described for the genus Xanthorrhachista. * Bull. Ent. Res., vii, October 1916, p. 120. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 215 Eyes oval, less compressed, their vertical diameter being only a little longer than the horizontal one. Frons flat, not at all prominent in profile, with parallel sides, only a littie longer than broad ; lunula narrow, but much extended laterally ; face flat in the middle, with shallow antennal grooves and with distinctly prominent mouth-border. Cheeks lnear; jowls not broader than the breadth of the third antennal joint. Antennae inserted at the middle of the eyes, only a little shorter than the face ; third joint pubescent, rounded at tip; arista plumose, its feathering being about as broad as the length of the third antennal joint. Proboscis short and thick, with very broad flaps, which cover the palpi. Cephalic chaetotaxy rather developed, only the oc., put. and genal bristles being wanting ; no ocp.; outer vt. half as long as the strong inner pair; one strong s. or., and two pairs of long and distant 7. or. Thorax elongate, with incomplete suture and reduced chaetotaxy ; there are only two npl., one a., and two p. sa., one mpl., one pt.; the scp. are rudimentary, only the middle pair being distinct. Scutellum triangular, flat above, a little broader than long, distinctly carinate on the sides, with only the 0. sct., which are very long, strong and diverging. Abdomen narrow and long, of cylindrical shape, narrow at base but not properly pedunculate, in the male with 4, in the female with 5 segments; the last two segments bear some strong bristles at the sides behind. Male genitalia small and rounded; ovipositor short, depressed, with the basal segment as long and nearly as broad as the terminal segment of the abdomen, and with the apical segment much narrower and acute. Legs long and simple, with non-spinose femora, even the front pair being without bristles below ; middle femora elongate, as long as their tibiae, and a little thickened ; middle tibiae with a single long spur at end; hind tibiae shorter than their femora, distinctly curved, with scattered bristles on the outer side; claws very short. Wings short and proportionally narrow, hyaline, with a broad black spot at apex. Costa densely ciliate, with no distinct bristle; auxiliary vein closely approximate to the first longitudinal vein, which ends at middle of the fore border, before the small cross-vein, and is pilose above. Second longitudinal vein straight; third bristly throughout its whole length and slightly curved but not sinuous ; last segment of the fourth vein slightly diverging at end; small cross-vein a little beyond the middle ; hind cross-vei straight and perpendicular, 24 times as long as the small one; anal cell a little broader than the second basal cell, its inferior angle acute but not produced and shorter than the upper one; sixth vein reaching the hind border. Axillary lobe equal in the two sexes ; alula longer than broad. Type: the following new species :— Sosiopsila trisetosa, sp. nov. (Pl. xvu, fig. 1). A narrow and elongate, entirely yellow species not unlike a Psila, with a broad rounded black spot on the mesophragma, and with a broad black spot near the apex of the wing. ¢ @. Length of body, 65-7 mm. ; of wing, 4°5-5 mm. ; of ovipositor, 0°7 mm. Head and its appendages completely yellow. Occiput shining, unspotted ; frons more yellow, dull and finely pubescent in the middle, but with shining ocellar and vertical plates, the ocellar not being reddish brown, like the upper border of the lunula. Face shining in the centre; cheeks and jowls with a faint whitish dust, 216 PROF. M. BEZZI. like the lower occipital swellings. All the cephalic bristles are yellowish, only the et. and the s. or. being a little darker. Thorax entirely yellow; darker, less shining and punctulate on the dorsum, lighter, more shining and smooth on the pleurae ; the very short and scattered pubescence of the dorsum is yellowish, like all the macrochaetae, only the p. sa. being darkened, chiefly the interior one; the pleurae have only sparse and very thin pale yellowish hairs on hind border of meso- and on upper border of sternopleura. The scutellum is coloured like the mesonotum and shining ; it bears some pale yellowish hairs, while its macrochaetae are black. The mesophragma is shining yellow, but it has in the middle a broad, rounded, shining black spot, which is prolonged above in the middle of the postscutellum. Halteres and squamulae pale yellowish, the latter with a pale fringe. Abdomen shining yellow like the mesonotum and similarly punctate, with longer and denser pale yellowish pubescence; its terminal macrochaetae are black; venter yellow. Male genitalia shining yellow and yellowish pubescent; the ovipositor likewise. Legs entirely yellow, with pale pubescence and pale hairs, only the spurs of the middle tibiae being black ; the middle femora have below in the middle three rigid long hairs, directed downwards, stronger and more bristly in the male than in the female. Wings quite hyaline and strongly iridescent; the veins are yellow, with blackened ends, but the whole costa (with its ciliation) is black; the costal cell is yellowish, and the short and narrow stigma dark yellow. The black apical spot is attached to the costa by its base, which extends from the end of the second to the end of the third vein ; it is continued inwardly, crossing the third vein and ending roundly on the fourth vein ; the end of the first posterior cell is therefore completely hyaline, while the end of the submarginal cell is broadly black. In the female the apical spot extends only to the middle of the first posterior cell, without reaching the fourth vein ; but even in the male the part of the spot placed beyond the cubital fold is lighter. Type ¢ and type 9, a single pair of specimens from Portuguese East Africa, KH. of Mt. Mlanje, 21-23, xi., 1913 (S. A. Neave); an additional female specimen from Nyasaland, Mt. Mlanje, 8.x.1913 (S. A. Neave). IiI. Subfam. CERATITINAE. In the new material are present some very important species, which belong to the two genera Coelopacidia and Celidodacus, previously unknown to me; they may be comprised in my table of the genera on p. 216, just after no. 8 (67), as follows :— A(B). Prst., dc. and st. wanting; face very concave in the middle; arista bare ; femora not spinose beneath ; third ne vein bristly throughout its whole length .. ; ; he Coelopacidia, End. B(A). The above-named eeieles wer or at elite not all wanting at the same time ; face flat or even convex in the middle. 9(10). Femora spinose beneath; arista shortly plumose; third longitudinal vem with only a few bristles near its extreme base. a(b). Wings very broad, with the second and third longitudinal veins sinuous, the last portion of the third being moreover bent downwards at the end. Conradiina, End. b(a). Wings narrower, with the above-named veins not sinuous and with the last portion of the third straight hg. Bi a .. Celdodacus, Hend. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 217 Coelopacidia, Enderlein, 1911. This genus was placed by its author near Acidia, while in Prof. Hendel’s table it comes near Platyparea. I have before me two species, which agree in the form of the body and head, and in the reduced wing pattern, with the description of Dr. Enderlein ; but the author has not described the chaetotaxy, which is very reduced ‘in the species before me. The systematic position of the genus is dubious. In the want of prst., dc., prscl. and st. it agrees with the DactnaE and ADRAMINAE, but differs from them in the presence of the hm. Owing to this last character, the presence of distinct ocp. and to the presence of bristles on the underside of the front femora, I prefer to locate the genus at the beginning of the CERATITINAE, as a connecting link between this subfamily and the ADRAMINAE. I have not seen the genotype, C. madagascariensis, End., but I will give here a characterisation of the genus drawn from the two continental species at hand. Head in front view about as broad as high, not broader than the thorax ; occiput convex and prominent, with developed but not sharply differentiated lower swellings ; neck inserted just below the middle of the occiput, which bears a tubercle above it. Frons distinctly concave in the middle, parallel-sided, about twice as long as broad ; lunula very narrow and not extended at sides. In profile the eyes are much rounded, their vertical diameter being only a little longer than the horizontal one; they are higher above than the middle of the hollowed frons, while in front the frons is promin- ent as a distinct conical protuberance, which bears the antennae. Face deeply hollowed in the middle, without median keel and with undifferentiated antennal grooves, but distinctly prominent below at the mouth-border; cheeks linear ; jowls as broad as the breadth of the third antennal jot. Antennae inserted at middle of the eyes ; in consequence of the protuberance of the fore part of the frons and of the excavation of the face, they are rather porrect; third jomt rounded at tip, bare, not reaching the mouth; arista long, thin, with very short scattered pubescence. Proboscis short and thick; palpi less clavate and shortly setulose. Cephalic chaetotaxy reduced ; ocp. reduced to some thin acute yellowish bristles ; put. very short, thin, almost indistinct ; no oc. ; four vt., the inner pair not much longer than the outer one; one s. or., erect and directed backwards, placed near the middle of the frons, at the front end of the long narrow curved ill-differentiated vertical plates ; one 7. or. just above the lunula, directed inwards and decussate ; genal bristle thin and short. Thorax elongate, with the suture narrowly interrupted in the middle; scp. long and strong, those of the middle pair approximate ; two npl.; one a. sa. and two p. sa.; one mpl. ; no distinct pt. and no st. Scutellum flat, triangular, as long as broad, with lateral keels and with two pairs of strong bristles, the apical ones parallel and not much weaker. Abdomen elongate, narrower than the thorax, of linear shape, being as broad at end as at base; there are 4 segments in the male and 5 in the female, but the sutures are not very distinct ; last segment with some macrochaetae at hind border in both sexes ; male genitalia rounded and concealed ; ovipositor with the basal segment depressed, triangular, obtuse, as long as the last two abdominal segments together ; apical segment narrow, acute. Legs slender ; front femora with only one bristle below near the end, middle femora distinctly thickened, shorter than their tibiae, which bear two equally strong spurs 218 PROF. M. BEZZI. at the end; hind tibiae straight, shorter than the femora, with some short bristles on outer side; claws short. Wings long, hyaline, with a very small apical black spot. Costa not ciliated, shortly pilose, without bristle; auxiliary vein closely approximated to the first vein, which is pilose and ends before or opposite to the small cross-vein; second and third vein straight and diverging at their ends, the third bristly throughout ; small cross-vein beyond the middle of the discoidal cell; hind cross-vein straight and perpendicular; last segments of the third and fourth veins parallel; second basal cell not dilated, but very long; sixth vein produced to the hind border ; anal cell not broader than the second basal cell, its lower angle drawn out into a point which is rather long, but is always shorter than the second basal cell; axillary lobe equal in the two sexes; alula longer than broad. The known species may be distinguished as follows :— 1(2). Front femora with some bristles below near the end; wings without black apical spot, but with a dark border to the hind cross-vein madagascariensis, End. 2(1). Front femora with only one bristle below; wings with a black apical spot, but the hind cross-vein not or less infuscated. 3(4). Thorax on the dorsum with a whitish pollinose, longitudinal, middle stripe ; wings with yellow stigma and with a slightly infuscated hind cross-vein strigata, sp. nov. 4(3). Thorax not striped; wings with black stigma and the hind cross-vein not infuscated .. - ve wi sie ae melanostigma, sp. nov. Coelopacidia strigaia, sp. nov. An elongate, entirely yellow species, distinguished by the whitish thoracic stripe and the yellow pterostigma. 3 2. Length of body 9-95 mm. ; of ovipositor 1 mm. ; of wing 8-8°5 mm. Head entirely yellow, unspotted, only the very small ocellar dot being black. Occiput very shining and bare, except for a few short hairs near the neck and on the lower part ; frons wholly glistening, quite bare, with the front part slightly reddish above the lunula, the upper border of whichis narrowly brown. Face very glistening ; cheeks and jowls not pruinose. Antennae entirely yellow, even on the apical part of the last joint ; second joint short and nearly bare ; third joint about twice as long as broad; arista yellowish. Palpi pale yellowish; proboscis dirty reddish, with pale hairs. Of the cephalic bristles the two pairs of vt. and the s. or. are black, the thin pvt. and the ocp. are yellowish, like the 2. or. and the genal bristle ; the sparse and short hairs of the lower part of the occiput are pale yellowish. Thorax entirely yellow and shining; but the dorsum, being punctulate, is less shining than the very glistening pleurae, which are moreover of a paler colour ; the whitish pollinose middle stripe is broad, but only distinct when the thorax is viewed from in front ; this stripe begins just beyond the the middle scp. and ends before reaching the scutellum broadening gradually behind. The very short and sparse pubescence of the dorsum is pale yellow, while the pleurae are quite bare, with only a few, hardly visible, thin and whitish hairs on the mesopleura. Scutellum entirely glistening yellow, bare, with some scattered pale hairs at the hind border. All the macrochaetae of the thorax and scutellum are black. Postscutellum and NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 219 mesophragma shining yellow, the latter more or less reddish in middle or on upper border. Squamulae very small, with sparse and short pale hairs at border ; halteres pale yellowish, with more reddish knob. Abdomen entirely shining yellow; but there is a narrow dark longitudinal stripe along the middle of the two basal segments, rather faintly marked; the pubescence is more dense than that of the thorax, but is always short and yellowish; the apical macrochaetae are black. The venter is shining yellow in the middle, whitish dusted on the sides. Male genitalia shining yellow, with pale pubescence; ovipositor shining yellow, with very short and sparse pubescence. Legs entirely shining yellow, with more reddish tarsi; their sparse pubescence is pale, but the bristles of the femora and tibiae are black, like the apical spurs of the middle tibiae ; the middle tibiae have a longitudinal row of three rather strong black bristles towards the middle of the posterior side ; the hind tarsi are ciliated below, chiefly on the praetarsi ; the claws are black, the pulvili whitish. Wings yellowish-hyaline, very shining and iridescent ; costa and veins yellowish, but the thickened basal part of the first longi- tudinal and the common stem of the second and third, are black ; the cross-veins and the ends of the longitudinal veins are blackish ; the stigma is entirely yellow. The apical dark spot is small, extending symmetrically on both sides of the end. of the third longitudinal vein, along the wing border; it is broader in the female than in the male. The infuscation of the hind cross-vein is very small and faintly indicated, sometimes even indistinct. The first longitudinal vein ends in the costa a little, but distinctly, before the small cross-vein. Type ¢ and type 9, a couple of specimens from Nyasaland, Limbe, Chiromo,. Ruo R., 22.1x.1916 (R. C. Wood). ee Coelopacidia melanostigma, sp. nov. Closely allied to the preceding species, but differing in some particulars of the coloration and in the longer third antennal joint. 6. Length of body 85 mm. ; of wing 75mm. Head and its appendages exactly as in C. strigata, but the put. more developed and blackish ; the third antennal joint is distinctly longer and reaches the mouth-border, being about 2°5 times as long as broad. Thorax, scutellum and halteres as in the preceding species, but there is no trace of the whitish middle stripe; there is moreover a small but distinct pi., which is not to be seen in C. strigata. Abdomen the same, with the basal stripe blacker and more sharply defined. The wings likewise, but the stigma is entirely of a deep black colour ; the first longitudinal vein ends opposite to the small cross- vein, not before it; the apical dark spot is similarly shaped, but there is no trace of the infuscation of the hind cross-vein. The legs are identical. Type J, a single specimen caught in the same locality and on the same day as the specimens of C’. stregata, from which it is certainly distinct (R. C. Wood). Conradtina acrodiauges, Speiser, 1913 (Pl. xvii, fig. 2.) A couple of specimens of this characteristic fly from S.{Nigeria, Ibadan, 29.x!.1913—20.1.1914 (Dr. W. A. Lamborn). 220 PROF. M. BEZZI. Celidodacus, Hendel, 1914. Prof. Hendel has differentiated the present genus from Conradtina only on account ot the wing pattern; the chaetotaxy of the head and thorax seems to be the same in both genera ; but I think that the form of the wings and the shape of the second and third longitudinal veins can be regarded as sufficient for generic separation. I have before me what I believe to be the genotype, named apicalis, but not described, by Prof. Hendel ; I think that it is the same as Acidia obnubila of Karsch, even though the latter author does not mention the spines on the underside of the femora. A close ally of this species seems to be Acidia coloniarum, Speiser, which has a different wing pattern, having an isolated dark band across thesmall cross-vein. I will give here a characterisation of the genus Celzdodacus, with some additional notes on the type-species. Head in front view distinctly broader than high. Occiput rather flat above and less prominent below, the lower swellings being not developed. Eyes rounded, with the vertical diameter longer than the horizontal one. Frons flat, parallel- sided, 1°5 times longer than broad; in profile it is only a little prominent above the root of the antennae; lunula very small. Face convex in the middle, with broad antennal grooves ; mouth-border prominent ; cheeks linear; jowls narrow. Antennae inserted a little above the middle of the eyes, with the third jomt acute but rounded at tip, long, extending a little beyond the mouth-border; arista plumose, the breadth of the feathering being equal to the breadth of the third antennal joint towards the middle. Proboscis short and thick; palpi broad and bristly. Cephalic bristles strongly developed ; ocp. acute and black; two vé., the mner pair longer ; pvt. well developed ; no oc. ; two s. or. and three 7. or.; genal bristle well developed ; there is also a bristle towards the middle of the lower part of the occiput. Thorax broad and stout, with complete and strongly developed chaeto- taxy; scp. long, those of the middle pair not approximated ; the dc. pair is placed much behind; two mpl.; pt. strong. Scutellum triangular, as long as broad, flat above, with lateral keels and four long strong bristles. Abdomen with 4 segments in the male and 5 in the female, narrowed at base, about as broad as the thorax in the middle, with terminal bristles ; male genitalia small; ovipositor short and broad, depressed. Legs short and stout, all the femora spinose beneath, but those of the hind pair less so than the others; middle femora thickened ; middle tibiae with a single spur at the end; hind tibiae with less developed posterior row of bristles. Wings proportionately narrow, with well developed pattern. Costa not ciliated and without bristle; auxiliary vein stout and well separated ; first longitudinal vein ending before the middle of the costa and before the small cross- vein ; second vein a little sinuous; third bristly throughout, straight on its last portion and diverging from the fourth; small cross-vein much beyond the middle, placed on the last third of the discoidal cell; hind cross-vein straight ; sixth vein reaching the hind border; point of the anal cell acute, as long as the second basal cell; axillary lobe broad; alula rounded. Celidodacus obnubilus, Karsch, 1887 ; (apicalis, Hend., 1914). (P!. xvu, fig. 4). A robust species, at once distinguished an account of the peculiar coloration of the thorax and wings. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 221 6 9. Length of body 75-8 mm. ; of ovipositor 1 mm. ; of wing 65-7 mm. Head entirely reddish yellow, only the small ocellar dot being black ; frons opaque and with scattered hairs in the middle, shining at vertex and on vertical plates ; antennae entirely yellowish, with the third joint three times as long as broad ; palpi _and proboscis yellowish. All the cephalic bristles are black. The rather opaque thorax has in the middle of the dorsum two broad longitudinal stripes of whitish dust, which are visible only in certain lights; on their outer border there is on each side a small black stripe, which is sometimes wanting ; the middle brown stripe, _as described by Karsch, is formed only by the space between the two whitish stripes. Very characteristic is the white, black-margined stripe on the notopleural line. All the bristles are black, and the short pubescence of the dorsum is likewise black. Scutellum like the mesonotum, with short and sparse pubescence on the sides. Pleurae shining, with soft thin pubescence. Squamulae and_halteres whitish. Mesophragma and postscutellum shining black, but sometimes reddish or yellowish. Abdomen above either entirely shining black, or with the first two segments more or less broadly yellowish ; the pubescence is blackish, the bristles black. Male genitalia shining black, ovipositor shining black, sometimes reddish brown towards the middle. Legs entirely yellow, with black bristles, spines and spurs ; front femora with two rows of 5-6 spines each near the end, middle femora with 7-8, hind femora with 4-5 much smaller ones. Wing pattern,as figured and _ described by Karsch, the apical hyaline spot being whitish, as stated by Hendel. The stigma is entirely black. Sometimes there is a small subhyaline dot in the - dark base of the first posterior cell, near the fourth vein. The veins are yellowish _ on the hyaline parts, and blackish on the dark parts. Originally described from Pungo Andongo, and recorded subsequently from S. Nigeria, the species seems to be widely distributed in Central Africa, from Hast to West. There are some specimens of both sexes from S. Nigeria, Ibadan, 29.x1.1913 (Dr. W. A. Lamborn); from Nyasaland, Mt. Mlanje, 15.iii.13—vili.1913 (S. A. Neave) ; from Chintechi, Lake Nyasa, ii. 1913 (J. G. Morgan). The other species, C. coloniarum, Speiser, has likewise a wide range, extending from the Hast to the West Coast. * *k * ES * * As important additions have to be made to the forms grouped around the old - genus “ Ceratitis,” a modification of my table on pp. 217-218 in my previous paper is rendered necessary, beginning with no. 17 (28), as follows :— 17(28). Middle scp. well developed ; legs short and stout; wings of broader shape ; head not balanced. - 18(19). Antennae short, with the third joint distinctly pointed at end, and with the second joint prominent and spinulose ; arista long plumose; usually three 7. or. present ; genal bristle very strong. A (B). Basal segment of the costa destitute of bristles before the costal one; basal dark band of the wings perpendicular, beginning at stigma Chelyophora, Round. B (A). Basal segment of costa provided with two groups of strong bristles before the costal one ; basal band oblique, beginning beyond the stigma Bistrispmaria, Speis. 22(23). 93(22). PROF. M. BEZZI. . Third antennal joint rounded at end. . Antennae very short, with the second joint prominent above and densely spinulose; basal dark band of the wings oblique, beginning at fore border beyond the stigma; costal bristle long and strong, as in the preceding genera, like the genal bristle of Clinotaenra, gen. nov. . Antennae less shortened, with the second jomt not prominent, never spinulose above ; basal band perpendicular, beginning at stigma ; costal bristle short and less developed; genal bristle much less developed. . Head, in side view, less shortened, the eyes being therefore more rounded ; thorax and scutellum prevalently yellowish or reddish, or densely grey-dusted, with black spots (when the thorax is shining black, the scutellum is yellow and not spotted) ; wings with yellowish bands, which are only exceptionally broadly infuscated, and with blackish basal streaks ; the band passing over the hind cross-vein is not, or only exceptionally, united with the basal one. Arista with short pubescence, which is more distinct on the upper side ; frons of the male with conspicuous spatulate appendages Ceratitis, MacLeay (s. str.). Anzista with longer pubescence or even with short plumosity, Be developed on both sides ; frons of the male without such appendages. . Frons flat, normally shaped, with rather thin s. or., which are inserted on Jess distinct and less converging plates. . Middle legs of the male broadly feathered, at least on the tibiae; wings with the dark marginal band separated from the basal one beyond the stigma .. a Pterandrus, Bezzi. . Middle legs of nals eptreh eho band typically united with the basal one. . Scutellum rounded, swollen, without distinct lateral keels, more or less distinctly trilobate, black- spotted. . A pair of well developed oc. present; first posterior cell not narrowed at end ; fourth longitudinal vein ending much beyond the tip of the wing Pardalaspis, Bezzi. . Oc. quite wanting ; first posterior cell distinctly narrowed at end; fourth vein ending at tip of wing - us Capparimyia, gen. nov. . Scutellum flat and bluntly as ras with distinct lateral keels, entirely yellow, not spotted .. wi : Re Perilampsis, gen. nov. . Frons with very stout s. or., which are inserted on well developed and converging plates, which are besides in the male very prominent and cristate .. ae : f aS Hoplolopha, gen. nov. B (A). Head distinctly en in ne view, the eyes being more narrow ; thorax and scutellum shining black, sometimes with whitish markings, or even the scutellum white with black spots; wings with blackish bands. radiating from the base, which is typically destitute of streaks; arista with long plumosity .. Me ai af .. Triuhithrum, Bezz. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 223 Leucotaeniella, Bezzi, 1918. Leucotaeniella guttipennis, sp. nov. (Pl. xvu, fig. 3). Very like trispila in the scutellar pattern, but distinct from it and from the other species onaccount of its peculiar wing pattern, which approaches that of the reticulate type. ®. Length of body 6-6°5 mm. ; of ovipositor 1 mm. ; of wing 5-6°5 mm. Head exactly as in trispila, but without the black ocellar dot, and the semi- circular spot in the middle of the frontal stripe is paler and less distinct. Thorax, scutellum and halteres as in érispila ; the black apical scutellar spot is less developed, being much smaller than the space between the apical pair of bristles, and not at all visible from above. Abdomen similar, but the ovipositor not darkened at end; legs entirely similar. The wings in shape, coloration and venational characters are exactly as in trispila, but they show conspicuous differences in the pattern. This last in its general features is of the same type, but the middle and cubital bands are broadly fused together and even with the marginal and anal bands, forming thus a single broad patch filling up the apical two-thirds of the wing, interrupted by some rounded hyaline spots, which form a kind of broad reticu- lation. The basal band is broadly united with the anal one in the submarginal and first basal cells, thus forming a rounded hyaline spot in the marginal cell just below the stigma, and a hyaline indentation which ends above in the middle of the first basal cell. The stigma is blackish on the basal half, and almost hyaline in the apical one; the second costal cell is hyaline, with the two ends narrowly infuscated. The anal band is broader, and infuscated on its lower part; the marginal band has a broad hyaline spot before the end of the marginal cell (which shows therefore two broad hyaline spots), and another at wing border before the end of the submarginal cell, which has another hyaline spot in the middle just below but much smaller than the apical one of the marginal cell; the first posterior cell is entirely infuscated, yellowish on the basal part and fuscous on the apical, with three rounded hyaline spots, two of which are in the middle (one behind the other), and one at the end in the lower apical corner just above the end of the fourth longitudinal vein and passing a little below it. The second posterior cell has two hyaline spots at the border with a less defined light (but not hyaline) space above them; the discoidal cell has a rounded hyaline spot before its apex, united with the indentation of the third posterior cell. All these hyaline spots and indentations are broadly margined with fuscous; but the hind cross-vein is entirely on the yellowish part, being nearer to the fuscous border of the spots of the second posterior cell than to that of the indentation of the discoidal cell. Type 9, a single specimen from N. Nigeria, Zungeru,18.xi.1910 (Dr. J. W. Scott Macfie). An additional female specimen from the same locality and collector differs only in having three (instead of two) hyaline spots in the submarginal cell along the costa, a third (half as small) hyaline spot being present between the two spots of the type. Bistrispinaria, Speiser, 1913. Owing to the fact that the present very distinct genus was shortly characterised by Dr. Speiser as a section of Ceratitis, I will give here a more complete description of it. It is more closely allied to the genera Chelyophora and Acroceratitis, than to those that have been separated by me from Ceratitis, s.]. 294- PROF. M. BEZZI. Head in front view a little broader than high, and distinctly broader than the- thorax; eyes rather narrow; lower occipital swellings much developed; jowls rather broad. rons broader than the eye, as broad as long, a little concave before: the ocelli. Antennae broadly separated from each other at base, inserted near the middle of the eyes, with the second joint very prominent and beset with strong spinules, with the third joint pointed at the end and reaching to the middle of the face ; arista long plumose on both sides. Cephalic bristles long and strong, black ; the two s. or. are placed on converging plates and the second is very stout, curved backwards and inserted over a small black tubercle ; three strong 7. or.; oc. very long and strong; ocp. black and well developed ; genal bristle very long. Frontal stripe with only a few hairs towards the middle. _ Thoracic chaetotaxy complete ; dc. placed a little before the line of the a. sa.; two mpl.; pt. long and strong. Scutellum rounded, convex, swollen, less distinctly trilobate, shining black, with four bristies and some stout whitish hairs above. Abdomen ovate, as broad as the thorax, acute at the end, bristly at hind border of the segments ; male genitalia small. Legs stout, simple; front femora with a complete row of bristles beneath; middle tibiae with two spurs at end ; hind tibiae without distinct: posterior row. Wings rather long, with fuscous basal streaks and fuscous bands ; basal band oblique, with its upper end beyond the stigma, united with the marginal band ; middle band complete ; cubital band isolated. Basal segment of costa with two groups of 2-3 erect spines each, the first placed just above the humeral cross-vein, the second just before the costal bristle, which is well developed. First longitudinal vein ending before the small cross-vein ; second vein nearly straight ; third vein a little curved about the middle of its last section, and then parallel with the fourth or slightly diverging, beset with scattered bristles to beyond the small cross-vein, which is placed near the middle of the discoidal cell ; hind cross-vein long and rather sinuous, inwardly oblique, the lower angle of the discoidal cell being acute ; discoidal cell about three times as long as the hind cross-vein; anal cross-vein deeply sinuous; the anal cell with a broad point, which is not longer than the second basal cell. 1. Bistrispinaria fortis, Speiser, 1913. (Pl. xvu, fig. 5). A stout species, distinguished by its frons, antennae and legs. ¢ Length of body 5 mm. ; of wing 6 mm. Head entirely pale yellowish, opaque, with a dark brown subocular spot; face whitish, with a dark yellowish stripe at the mouth-border; frons with a darkish rounded middle spot. Antennae entirely yellowish, with the third joint distinctly darker; the spinules of the second joint are black and thick. Palpi whitish, with black bristles; proboscis yellowish. All the bristles are black, lke the hairs on middle of the frons. The lower occipital swellings are whitish and clothed with white hairs. The thorax is blackish on the dorsum but densely grey-dusted and clothed with pale yellowish hairs; on the disc there are three narrow longitudinal darker stripes, the lateral ones being broadly interrupted at the suture; on the sides there are two broad, subquadrate, shining black, ill-defined spots; the humeri are yellowish; on the postalar calli there is a rounded, deep black, opaque spot. The pleurae are reddish brown, rather shining, with a whitish yellow patch on the upper border of the mesopleura, continued below the root of wings, and NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 225, there united with the double whitish hypopleural spot; the hairs of the pleurae are white on the whitish parts, and black on the dark parts. Postscutellum shining black ; mesophragma reddish brown, but densely grey-dusted. Scutellum shining black, with a narrow middle longitudinal yellow stripe, and with a yellow stripe on each side; the bristles are black and the hairs pale yellowish. Halteres pale yellowish. Abdomen with black bristles and yellowish hairs ; it is blackish, but grey-dusted, with a narrow whitish hind border to the segments. Legs pale yellowish, with the femora broadly infuscated and the four posterior tibiae darker on the basal half; bristles black. Wings whitish hyaline, with blackish basal streaks and with fuscous bands ; veins black, but the costa and the first longi- tudinal vein whitish at the stigma. At the base there is a short blackish band, beginning at the humeral cross-vein; there is a blackish streak in each cell, broader and more sharply defined near the fore border; but the anal cell has no middle streak. Stigma whitish, with the base narrowly darkened. The first band is oblique, beginning at fore border just beyond the stigma, passing over the small cross-vein, and ending evenly and symmetrically at the end of the sixth vein. The costal band is united with the above-named band, extending below to the third. vein and on the apical part passing even a little beyond it; it includes 6-7 darker dots along the fore border, being moreover narrowly hyaline near the costa. About. at middle of the last section of the third longitudinal vein there arises from the costal band the middle band, which is narrow and oblique, irregularly interrupted. at base, and crosses just beyond the middle the last section of the fourth vein, ending at the border a little after the end of the fourth vein. The cubital band is broader. than the middle one, extends from the hind border across the hind cross-vein, and. ends roundly above it in the first posterior cell, being distinctly separated from. the basal band. Axillary lobe hyaline, only with a aiffused greyish shade at the apex. Originally described in the male sex from Kamerun, Soppo, xu. 1912; there is: a male from Uganda, Kampala, 10.xi.1915 (C. C. Gowdey). Clinotaenia, gen. nov. Notwithstanding the obtuse third antennal joint, the present new genus seems to be more closely allied to the preceding one than to the following genera. Its. wing pattern is very like that of Bastrispunaria, and is not very different from that of Trypeta grata, Wied., a species which might therefore be better placed here than in Leucotaenella. Loew (Berl. Ent. Zeits., v, 1861, p. 268) has pomted out that the allies of 7. grata are to be found only in the neotropical fauna; the American forms with a similar wing pattern are those of the genus Anastrepha, which are however very different in their structural characters; but it is interesting to note that the end of the fourth longitudinal vein in the species here described shows a slight tendency to be curved upwards. The new genus Clinotaenia is closely allied to the Oriental genus Gastrozona, but has only two z. or., and different body and wing patterns. It must be recorded that Prof. De Meijere has recently described * from Java a species of Anastrepha (A. extranea), which however cannot be placed in the American genus, nor in the present one. * Tijdschr. v. Entom., lvii, 1914, p. 193, pl. v, fig. F. 226 PROF. M. BEZZI. Head in front view distinctly higher than broad, as broad as the thorax; frons quite flat, proportionally narrow, as broad as the eye, longer than broad ; occipital lower swellings distinct, but less developed than in Bistrispinaria; eyes narrow, the head being very narrowed in profile; jowls rather broad. Antennae approxi- mate at base, inserted distinctly above the middle of the eyes ; they are very short, with the second joint rather prominent above and beset with short and thick spinules ; third joint rounded at tip, not reaching the middle of face; arista long plumose. Cephalic bristles well developed and black; the s. or. are inserted over less distinct and not converging plates, and the anterior one is not specially dilated; only two 2. or.; oc. well developed, but rather short ; ocp. black and acute ; genal bristle of medium size. Thoracic chaetotaxy complete, with black and strong bristles ; middle scp. long and approximated; dc. on a line with the a. sa. or only a little beyond it; two mpl.; pt. as strong as the st. Scutellum swollen, rounded, not trilobate, with four bristles, the apical ones decussate, and with sparse thin hairs ; it is whitish yellow, with three black spots. Abdomen oval, as broad as the thorax, bristly on sides and at end, with whitish transverse bands; ovipositor as long as the last three abdominal segments together, triangular, rather swollen at base, broadly obtuse at end. Legs short and stout; front femora with a complete row of strong bristles beneath; middle tibiae with a single spur at end; hind tibiae with a row of strong bristles in the middle of the outer side. Wings proportionally long, with blackish basal streaks and with fuscous bands, the first of which is placed obliquely (whence the generic name); costal bristle strong and long, but there are no other strong bristles before it. First longitudinal vein ending before the small cross-vein ; second vein about straight ; third vein beset with dense bristles through- out its whole length, its last portion parallel with the last portion of the fourth ; fourth vein slightly but distinctly curved upwards at end; small cross-veins long and oblique, placed beyond the middle of the discoidal cell ; hind cross-vein straight, inwardly oblique, the lower angle of the discoidal cell being acute; discoidal cell three times as long as the hind cross-vein ; anal cross-vein deeply angulate in the middle ; lower point of the anal cell of medium length, as long as the second basal cell. Type: the following new species. Clinotaenia anastrephina, sp. nov. (Pl. xvi, fig. 6). A pretty species readily distinguished by its antennae and wing pattern. Length of body 5:5 mm.; of ovipositor 1 mm. (when completely exserted 2:2 mm.); of wing 5°5 mm. Head entirely pale yellowish, with a broad blackish patch on the occiput from the neck to the upper corner of the eyes, with a small brownish subocular spot on the jowls ; towards the middle of the frons there is a reddish yellow spot. The frons 1s opaque, only the narrow plates bearing the s. or. being shining; ocellar dot blackish. Antennae reddish yellow, the second joint appearing infuscated owing to the numerous black spinules. Proboscis and palpi yellowish, the latter with rather thick black bristles. Occipital swellings whitish and clothed with whitish hairs, with a few short and thicker black ones at lower border. Frontal stripe in the middle with a few short black hairs. The thorax is blackish on the NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. wealth dorsum, but densely dark grey tomentose and with short golden-yellow pubescence ; the sides above the notopleural line and above the root of the wings are shining blackish brown, while on the postalar calli there is a rounded, velvety black spot; on each side, behind the suture and above the blackish supra-alar patches, there is a narrow and ill-defined, yellowish stripe. Pleurae blackish, partly reddish brown, shining, with a broad oblique whitish patch across the meso- and pteropleura, uniting in front with the whitish immaculate humeral callosities; there is also a broad double whitish hypopleural spot; sternopleura shining black; mesopleura hairs whitish. Postscutellum and mesophragma shining black. Scutellum shinin whitish, with whitish hairs ; there is a narrow black stripe at base, and three broad black spots, which are hardly visible from above, the apical one occupying the whole space between the two apical bristles, and the lateral ones having the lateral bristles inserted at their outer borders. Squamulae and halteres whitish. Abdomen blackish, but clothed with dense grey and white dust; first segment shining black on the basal half, white-dusted on the apical half, thus forming a complete and broad cross band; second segment entirely brownish grey; third and fourth entirely white, with a narrow blackish basal band. The abdominal hairs are white on the white, black on the dark parts; all the bristles are black. Venter reddish brown. Ovipositor entirely reddish, darker at base, with short black or dark brownish hairs. Legs with dark reddish coxae and with blackish femora, which are dark reddish at base; tibiae and tarsi pale yellowish, but the latter darkened at end; bristles black ; hairs dark on femora and whitish on tibiae. Wings whitish hyaline, with black veins, but the costa and the first longitudinal are whitish on the hyaline parts, like the fourth and fifth veins. At the extreme base there is a short fuscous band, extending from the humeral cross-vein to the base of the second basal cell; after this band there are the blackish streaks, which are broad and placed one in each cell, except the anal cell, and are present even in the base of the submarginal cell and in the upper corner of the base of the discoidal cell. The second costal cell is infuscated at the two ends; the stigma is hyaline, with a blackish base. The bands are deep fuscous, the darker spots of the marginal band along the costa being thus less distinct. Basal band oblique, beginning at fore border beyond the stigma, passing across the small cross-vein and ending narrowly at the tip of the sixth vein; it is almost straight on its outer border, with fuscous teeth on the inner edge, two of which extend into the third posterior cell; at base of the discoidal cell it is united with the fuscous basal streak of the same cell. The marginal band extends below to the third vein, passing even below it near the apex and ending at wing border near the middle of the first posterior cell. The oblique middle band is rather broad ; it begins at middle of the last portion of the third vein and ends at wing border just below the end of the fourth ven; the hyaline indentation between the marginal and the middle band is narrower than the terminal part of the marginal band itself. Cubital band less oblique, beginning at hind wing border, passing across the hind cross-vein and ending rather narrowly above it in the first posterior cell, quite separated from the basal band; its external border is convex, while the internal one is concave, the band being thus distinctly arcuate. Axillary lobe entirely hyaline, like the alula. Type 9, a single specimen from Nyasaland, Mt. Mlanje, 3.xn.1913 (S. A. Neave). (637) B 28: PROF. M. BEZZI. Pterandrus, Bezzi, 1918. Pterandrus rosa, Karsch, var. fasciventris, nov. Differs from the type in having well developed blackish bands on the abdomen. 3. Length of body 5-5.5 mm. ; of ovipositor about 1 mm.; of wing 55-6 mm. Head, its appendages and bristles exactly as in rosa. Thorax and scutellum like- wise, but the dorsum of the mesonotum in fully coloured specimens seems to be darker and more densely grey-dusted. The chaetotaxy is the same, the single mol. being present. The abdomen in both sexes shows a distinct, complete, blackish. rather shining cross band on the apical half of the second segment; the third and fourth segments are whitish, but in the female they have a distinct dark cross band at base. The ovipositor is shining reddish, with a distinct black base and with a less intense apical black border, but these dark parts are not sharply separated. The legs are as in rosa, the male having the middle femora quite simple, and the middle tibiae feathered on the apical half alone; the front femora of the male are not densely ciliated. The wing pattern is the same as in rosa, but the stigma is blackish, with the apical third hyaline. Type ¢, type 9, and an additional female specimen from Uganda, Entebbe, 17.vii.1911 (C. C. Gowdey); another male specimen (more lightly coloured owing to immaturity) from Uganda, Kampala, 5.xu.1916 (C. C. Gowdey). The species Pterandrus anonae, P. colae, P. rosa and Pardalaspis morstattv are very closely allied and have a similar wing pattern; the feathering of the middle legs in the male decreases from the two former (which have femora and tibiae feathered) to the third, in which only the tibiae are feathered, and to the fourth, in which they are not feathered at all. The two former species agree also in the ciliation of the front femora, which is lacking in the two latter; morstattc in the length of the ovipositor agrees with colae, while anonae and rosa have a short ovipositor. Ina more natural classification P. morstatt: must be placed in the genus Pterandrus, notwithstanding the lack of feathering on the middle legs in the male. Pardalaspis, Bezzi, 1918. I will give here another table for determining the rather numerous species of the present genus, excluding znscripta, Graham (belonging to Trirhithrum), and with the addition of two new species. 1(18). Wings with the band across the hind cross-vein quite isolated, and without a complete middle band, or rarely with a small isolated streak on the last portion of the fourth longitudinal vein. 2(15). The black spots of the scutellum very broad, quadrate or rectangular in shape, closely approximated, and nearly covering the whole surface. 3 (4). Wings with the aE band isolated (as in Pterandrus); arista shortly plumose .. ae morstaitt, Bezz. 4 (3). Wings usually with ie high eta, bevel united at stigma with the basal one ; when they are separated, the arista is pubescent only. 5(14). Last portion of the fourth longitudinal vein destitute of an oblique dark streak across its middle. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 229 6 (7). Two strong mpl. present; wings broad, with strikingly developed basal streaks and dots, and with the bands much infuscated ; head of the male with a peculiar, striking coloration .. a .. punctata, Wied. 7 (6). Only one mpl. present as a rule; wings with a less developed basal pattern and with pale yellowish bands, which at most are infuscated only towards the hind border ; head of the male not peculiarly coloured. 8 (9). Thorax destitute of black spots ; species of greater size (8 mm.) brémer, Guér. 9 (8). Thorax with distinct black spots, at least on the sides beyond the suture ; size smaller (not over 7 mm.). 10(11). Occiput above with broad, shining black spots; the shining black parts on back of mesonotum more developed than the yellow ones; scutellum black, with a narrow yellow base a e:: melanaspis, sp. nov. 11(10). Occiput yellow above or only a little infuscated; yellow parts of back more developed than the black ones; scutellum yellow with black spots. 12(13). Humeri with a distinct black spot; sides of thorax with black spots before the suture; scutellum with two basal black spots besides the three apical ones 2: ; t .. cosyra, Walk. (giffards, Bezzi). 13(12). Humeri and sides of thorax eked the suture destitute of black spots ; scutellum with no distinct basal spots.. de fe silvestrir, Bezzi. 14 (5). Last portion of the fourth vein crossed over the middle by a well developed, oblique, dark streak ; wings short and broad .. .. antestictica, Bezzi. 15 (2). The black spots of the scutellum much smaller and broadly separated, the outer ones being more rounded; the scutellum therefore mainly yellow. 16(17). Scutellum with three black spots ; wings with the Set and basal bands wholly separated beyond the stigma .. flexuosa, Walk. 17(16). Scutellum with five spots; the two Lge tania Bands united and fused © with the stigma .. Ne : Bye quinaria, Bezzi. 18 (1). The band over the hind cross-vein not ceomiart or when its union with the other band is less distinct, there is a complete middle band. 19(20). The marginal band united with the basal one and with the stigma; the band over the hind cross-vein united with the basal one; no complete middle band Ht oP fi stectuca, Bezzi. 20(19). The marginal band separated gi Te ale one beyond the stigma; the — band on the hind cross-vein united with the marginal one; a “complete middle band across the last section of the fourth vein. . aliena, Sp. Nov. Pardalaspis melanaspis, sp. nov. Allied to P. cosyra (giffardz), but smaller and at once distinguished by the broad black occipital patches, and by the more extended shining black pattern of the thorax and scutellum. 39. Length of body 3:2-3°5 mm.; of ovipositor 0°5 mm. (when completely exserted 1°5 mm.); of wing 3°2-3°5 mm. Head pale yellowish, with reddish-yellow frontal stripe, with a black, rather broad, rounded ocellar spot, and a broad, shining black patch on upper part of the occiput, extending from the neck to the eyes, but leaving a narrow yellow border near the (637) BQ 230 PROF. M. BEZZI. eyes. There is no subocular spot on the jowls. The antennae are entirely yellow, about as long as the face; second joint not prominent, never spinulose; third joint broad and long, attenuated but rounded at end ; arista dark, shortly pubescent on both sides. Upper mouth-border rather prominent and shining. Palpi broad, flat, yellow, with short black bristles ; proboscis yellowish. Cephalic bristles black, but the pvt. and the very thin and indistinct genal bristles are yellowish, the former sometimes infuscated ; the hairs of the jowls and of the lower part of the occiput are yellowish ; frontal stripe with sparse indistinct darkish hairs. Thorax entirely pale yellowish and rather shining, with broad black spots on the dorsum ; it is clothed with short, pale yellowish hairs. Humeri with a broad rounded black spot; from the humeri to the transverse suture extends a broad rectangular blackish-brown shining patch, reaching inwardly to the dorsocentral lines, and including a deep black, opaque spot above the notopleural line with the anterior npl. Between these two black patches there is a rather broad, shining black, middle stripe, extending from the black scapular region to the suture, and ending there truncately. Beyond the suture the back is entirely shining black, with a concave front edge; along the middle line there is a pale yellow, longitudinal stripe, which is a little narrower than the presutural black stripe and ends with a transverse, pale yellowish streak in front of the scutellum, thus forming a § shaped mark. Above and in front of the root of the wings, and on the postalar calli there are rounded, deep black, opaque spots. The pleurae are of a shining pale yellowish colour, whitish on the upper part of mesopleura, and there with a narrow longitudinal black stripe, extending from below the humeri to the mpl. bristle. The upper border of the sternopleura is pale ; there is a broad double whitish hypopleural spot. All the bristles are black and strong, but the well developed middle scp., and the thin pt. and st., are pale yellowish ; one mpl. only ; dc. placed before the line of the a. sa. Scutellum rounded, convex, slightly trilobate ; it is entirely shining black, with a narrow, pale yellowish, basal border ; it has rather long and numerous whitish hairs on the disc, and four long black bristles, the apical ones not decussate, or only shghtly so at the tips. Postscu- tellum shining black; mesophragma yellow, clothed with dense whitish dust on the upper half. Squamulae whitish ; halteres yellowish, Abdomen entirely yellow, rather shining, with numerous blackish hairs, and with black bristles at end and on sides. The narrow hind border of the first segment in both sexes and almost the whole of the third segment in the female, are densely clothed with whitish dust, forming thus two white cross bands. Ovipositor shining yellow, with yellowish hairs, with the basal segment bluntly triangular, about as broad as long, swollen. Legs entirely pale yellowish, with yellowish pubescence ; the long bristles of the underside of the front femora, and those of the hind tibiae are likewise yellowish ; the single spur of middle tibiae is black. Wings proportionally long, with short and less developed costal bristle ; the veins are entirely pale yellowish, and typically disposed ; the stigma is about twice as long as broad, quite yellow, or rarely with the apical half subhyaline. The pattern is yellowish and shaped as in cosyra, but the basal dark streaks are less numerous, the second basal cell having a single blackish dot, and the anal cell having no dots at all. Type dg, type 9, and an additional female specimen from Cape Colony, Grahams- town, 9.0.1905 (C. W. Mally). NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 231 Pardalaspis aliena, sp. nov. Allied to cosyra and silvestrw, but distinct from them as well from all the other species in having the costal band of the wings separated from the basal one, and in having a complete middle band across the last section of the fourth longitudinal vein. 9. Length of body 4°2 mm. ; of ovipositor 0°6 mm. ; of wing 4:2 mm. | Head pale yellowish, without any dark spot, only with a small black ocellar dot. Face whitish, not prominent at the mouth-border. Antennae wanting in the type, only the shortened and pale yellowish basal joints being present. Palpi whitish, not very broad, without bristles, with whitish hairs ; proboscis pale yellow. Cephalic bristles black ; oc. well developed. Thorax shining yellow, with short whitish pubescence. Humeri whitish, immaculate ; above them there is on each side a shining black, longitudinal stripe, which ends before the suture, but is con- tinued behind it along the dorsocentral lines by a black subquadrate spot on the dc. bristles, and by a broad stripe in front of the scutellum, including a broad whitish patch, in the shape of a rounded spot. Above the notopleural line and above the root of the wings there are the usual velvety black, opaque spots, which are separated by the whitish notopleural callus; on the postalar calli there is also a rounded, deep black spot ; above the shining supra-alar patch there is on each side a short whitish stripe. The pleurae are pale yellowish and shining, with a white meso- pleural spot and a double white hypopleural spot. Postscutellum deep black; mesophragma pale yellowish, with faint whitish dust. Scutellum shining whitish, with three subquadrate, equally broad spots at hind border ; it is convex and less distinctly trilobate. All the thoracic bristles are black, even the middle scp. and the pt., only the st. being yellowish ; one mpl. Squamulae white ; halteres whitish. Abdomen shining yellow, with whitish pubescence and with black bristles at sides and at end; the narrow hind border of the first and third segments is covered with whitish dust, forming thus a very narrow and ill-defined white cross band ; ovipositor shining yellow; venter pale yellowish. Legs entirely pale yellowish and with concolorous bristles and hairs, only the spur of middle tibiae being black. Wings with entirely pale yellowish veins, and with a very short and black costal bristle ; the bands are pale yellow, infuscated towards the hind border ; the stigma is yellow, with the extreme upper corner subhyaline. At base there are some darkish dots and streaks, three in the second costal cell, one streak in the other cells and none in the anal cell. The basal band is perpendicular, beginning at stigma, including the small cross-vein, which is situated just at its outer border, and ending with a sheht curve at apex of the sixth longitudinal vein. After this band there is a complete hyaline band, which extends to the fore border just beyond the stigma ; the marginal band is therefore separated from the basal one. The marginal band extends below to the third vein, passing beyond it with its infuscated end, and ending at apex about in the middle of the first posterior cell ; 1t includes two darker marginal dots in the marginal] cell, and a rounded hyaline spot at end of the second vein. A little before the middle of the marginal band arises the narrow and complete middle band, which, passing across and beyond the middle of the last section of the fourth longi- tudinal vein, ends infuscated at hind border in the upper third of the second posterior cell. The cubital band extends across the hind cross-vein and is only slightly oblique ; above it becomes gradually paler, but is distinctly united with the costal band just 232 PROF. M. BEZZI. near its base. The small cross-vein is placed noticeably before the middle of the discoidal cell ; the last portions of the third and fourth longitudinal veins are slightly divergent ; the hind cross-vein is straight and placed a little obliquely ; the produced lower angle of the anal cell is rather broad, and a little shorter than the second basal cell. Type 9, a single damaged specimen from Cape Colony, Grahamstown, 30. xi. 1908 (C. W. Mally). Capparimyia, gen. nov. I will erect here this new genus (which, so far as at present known, is not Ethiopian but only Mediterranean) with Ceratitis savastani, Martelli,* as type. This species shows all the characters of Pardalaspis, as well as the same coloration of body and the same wing pattern, but it is at once distinguished by the complete want of the oc. bristles, a character which is present in none of the allied forms. The ending of the fourth longitudinal vein at the tip of the wing is also unique among the allies of Ceratitis, and it is rendered even more evident by the distinct narrowing of the first posterior cell, due to the curving downwards of the third longitudinal vein in its last portion. The erection of a new genus for the present species 1s supported by the fact that it is a gall-making insect ; the young buds of the flowers of Capparis spinosa, L. (whence the generic name), in which the larvae live, become hypertrophied, deformed and arrested in their development, the effects being similar to those caused by the Itonid midge, Asphondylia capparis, Riibsaamen. The gall of Capparimyia savastani is included in the great work of Prof. Houard}; and it must be remembered that all the larvae of the allied forms live in fruits and are not galligenous, and that the faculty of making galls is very rare in the subfamily CERATITINAE, while it is not rare in the subfamily TRYPANEINAE. The main characters of the new genus are as follows :—Pvt. yellow; ocp. black and acute; two 2. or.; the s. or. are inserted on indistinct and not converging plates; genal bristle thin and yellow; dc. placed only a little behind the line of the a. sa.; one mpl.; pt. as developed as the st., and both yellowish in colour, like the middle scp., while all the other bristles are black. Wings with a short but strong costal bristle; first longitudinal vein ending noticeably beyond the small cross-vein, which is placed before the middle of the discoidal cell; third vein bristly to the middle of its last portion ; first posterior cell narrower at end than in the middle, the last portion of the third vein being curved downwards before the end; fourth vein ending very near the tip of the wing; lower angle of the discoidal cell acute, the hind cross-vein being however only a little oblique ; lower point of the anal cell acute and produced, but a little shorter than the second basal cell. The single species at present known has been found only in Sicily and in Southern Italy. * Boll. Arboric. Ital., vii, 1911, p. 19, figs. 1-4, and Mem. R. Accad. Zelanti, (3) vii, 1912, p. 50, figs. 1-4. In the Zoological Record, vol. L, p. 423, the species appears as published in 1913. t Les Zoocécid. des plantes d’Europe, etc., ili, 1913, p. 1355, n. 6730. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 233 Perilampsis, gen. nov. This new genus is erected here with Carpophthoromyia pulchella, Austen, as type, and includes Carpophthoromyia formosula, Austen, but not Trypeta grata, Wied., which probably belongs to Clinotaena, as stated above. It comprises very beautiful (whence the generic name) Ethiopian flies, which agree in coloration of the body and in wing pattern, as well in the following characters :— Head in front view broader than high, the face being a little shorter than the frons and the jowls very narrow; eyes rounded; occipital lower swellings less developed ; frons about as broad as long, flat, with the shining vertical plates short, not prominent, slightly converging in front ; face distinctly concave in the middle and prominent at mouth-border. Antennae inserted at middle of eyes, rather long but always shorter than the face; second joint not prominent, never spinulose above; third joint twice as long as the second and obtuse at end; arista with long pubescence, or even shortly plumose, the feathering being much narrower than the breadth of the third joint. Palpi broad and shortly bristled ; proboscis short. Cephalic bristles black; ocp. numerous and long, black, acute ; oc. strong and long; two 7. or.; genal bristle black and rather long. Thorax shining black on the back of mesonotum and adorned with characteristic transverse bands of minute, pale-coloured hairs; pleurae with whitish mesopleural stripe and with double hypopleural spot ; scutellum quite yellow, destitute of black spots. Thoracic chaetotaxy complete; middle scp. long and approximated; dc. placed a little beyond the line of the a. sa.; one mpl.; pt. as strong as the st.; all the bristles are black. Scutellum broadly triangular, not trilobate, rather flat above, with distinct lateral keels, pubescent above and with four bristles. Abdomen shortly oval, distinctly broader than the thorax at base, and with black bristles at end and on sides. Legs stout, with complete rows of bristles on front femora and on hind tibiae ; middle tibiae with a single spur. Wings broad and long, with a blackish and characteristic pattern, the middle band of which is more or less complete, and the basal band perpendicular and beginning at the stigma. Costal bristle short but strong; first longitudinal vein ending opposite to the small cross-vein ; second vein straight; third vein bristly to the middle of its last section ; third and fourth veins slightly divergent towards the end, and thus the first posterior cell not narrowed outwardly ; fourth vein ending much behind the tip of wing; small cross-vein before the middle of the discoidal cell; hind cross- vein oblique inwardly ; point of the anal cell a little shorter than or as long as the second basal cell. Perilampsis puichella, Austen, 1910. Of the present species, known from Uganda and Kamerun, there is a specimen from 8. Abyssinia (Rk. J. Stordy). The species seems therefore to have a very wide distribution over the whole centre of the Ethiopian Region. Hoplolopha, gen. nov. This new genus is a very interesting one, because, although it is without doubt closely allied to Ceratztis (s. 1.) and has a wing pattern very like that of the preceding genus, it shows in the male sex on the frons the cristiform protuberances with 234 PROF. M. BEZZI. _Incrassated bristles (whence the generic name), which are characteristic of other Trypaneid genera, like Straussia, Vidalia, Stemonocera, ete. Head in front view a little higher than broad. LHyes rather narrow; lower occipital swellings well developed; jowls as broad as the breadth of the third antennal jot. Frons longer than broad, narrower than the eye; in the female it is flat, while in the male it is distinctly concave in the middle ; the shining orbital plates are produced beyond the middle of the frons, and converge in front, and moreover in the male they are prominent and cristiform. Face shorter than the frons, rather flat, and with a less prominent mouth-border. Antennae inserted at the middle of the eyes, proportionally long, but always shorter than the face ; second joint not specially prominent, nor spinulose; third joint elongate linear, rather narrow, about three times as long as the second, rounded at tip; arista pubescent, the breadth of its feathering being much narrower than the breadth of the third antennal jomt. Palpi broad; proboscis short. Cephalic bristles black and strong; ocp. thin and pointed; oc. strong and long; the three s. or. are in both sexes very long and incrassated, chiefly the two former, curved back- wards, but in the male they are distinctly longer; two 7. or., rather short and much shorter than the superior ones ; genal bristle rather short. Thorax yellowish grey, pale pubescent, with ill-defined black pattern, and with less striking, whitish, pleural markings ; scutellum swollen, rounded, yellowish, with three broad, shining black spots. Middle scp. long, approximated, black lke all the other bristles ; de. a little beyond the line of the a. sa.; one mpl.; pt. as strong as the st. ; four sct., the a. sct. diverging. Abdomen short and oval, with bristles on sides and at end ; male genitalia developed ; ovipositor narrow and cylindro-conical, swollen, as long as the last three abdominal segments together. Legs short and stout ; front femora with a complete row of bristles below ; middle tibiae with one spur; middle and hind tibiae with an external row. Wings long but proportionally not broad, with a short but strong costal bristle, with basal dark streaks and dots, and yellowish bands ; basal band beginning at stigma and perpendicular ; middle band complete. First longitudinal vein produced beyond the small cross-vein ; second vein straight ; third vein densely setigerous to the middle of its last portion; small cross-vein oblique and placed considerably before the middle of the discoidal cell ; third and fourth slightly divergent at end, the fourth ending much after the tip of wing and the first posterior cell not narrowed outwardly ; hind cross-vein before the middle of the first posterior cell, very oblique inwardly, the lower angle of the discoidal cell therefore acute ; discoidal cell three times as long as the hind cross- vein; anal cross-vein deeply sinuous, the lower angle of the anal cell very long. longer than the second basal cell; axillary lobe broad and semicircular, in the male separated by a deeper incision of the hind border at the end of the sixth longitudinal vein. Type: the following new species. Hoplolopha cristata, sp. nov. (Pl. xvii, fig. 7). A robust fly of comparatively large size, at once distinguished from all the allied forms with a typical Ceratitis-like wing pattern on account of the cristate frons of the male. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 255 $2. Length of body 6°5-7 mm.; of ovipositor 1 mm.; of wing 75-8 mm. ; breadth of wing about 3 mm. Head yellowish, but more whitish on the face, on the linear cheeks and on the lower occipital swellings ; the frons is opaque and immaculate, but the ocellar and vertical regions and orbital plates are shining and of a more reddish colour; below the lower corner of the eyes there is an ill-defined brownish genal spot, more developed an the female than in the male. Antennae entirely yellowish, the third joint infus- cated along the upper border ; palpi pale yellowish, with some black bristles at the end ; proboscis reddish. All the cephalic bristles are black, even the pvt. and the genal ones ; the lower occipital hairs are white, but along the lateral borders of the oral cavity there is a rather dense row of short bristly black hairs. Thorax on the -dorsum reddish brown, but covered with a dense grey dust and with pale pubescence ; ‘it is opaque, with a more or less distinct trace of a middle longitudinal blackish -stripe ; on the dorsocentral lines there is on each side a blackish stripe, which is ‘interrupted to form a rounded postsutural spot, and a broader and shining elongate -spot in front of the scutellum ; above the wings there is also on each side a blackish -stripe, and two broad blackish spots, the anterior shining, the posterior deep black. Humeral calli reddish or pale yellowish, not spotted. Pleurae reddish yellow, rather ‘shining, with white dust and with whitish mesopleural stripe, more distinct in the ‘male, and with rather long, pale yellowish hairs, which are however dark on the lower part of the mesopleura, while on the breast they are white. The double whitish hypopleural spot is more distinct in the male. Postscutellum shining black, more -or less reddish below and on sides ; mesophragma reddish with grey dust. Scutellum with three very extended, shining black, rectangular spots, and between them very narrow, yellowish or reddish stripes; in the narrowly yellow base there are two dark spots, in contact with the hind border of mesonotum. MHalteres whitish, -with reddish knob ; squamulae whitish, with narrow dark border. All the bristles -are black, even the middle scp. Abdomen reddish yellow, with grey dust ; the second, third and fourth segments are blackish at base, more broadly in the second, which is almost entirely black, while the third is more densely clothed with whitish dust, thus forming a transverse band, more distinct in the female. The hairs are whitish on the pale, and blackish on the dark parts; the bristles are black. Male genitalia black and black-haired ; ovipositor shining reddish, darkened at tip, with -darkish hairs. Legs entirely pale yellowish, with blackish pubescence, black bristles and black claws. Wings hyaline, iridescent, with yellowish veins, only those of the lower part of the base being blackened. The extreme base, before the humeral cross-vein, is infuscated as a short yellowish band. The basal blackish dots -and streaks are numerous, and are more developed in the female than in the male; they invade the anal cell also, and even the base of the discoidal and third posterior cells. The stigma is yellowish, darker on the basal half. The bands are yellowish, infuscated towards the apex and the hind border of wing. Basal band perpendicular, going from the base of the stigma to the end of the sixth longi- tudinal vein, but not extending into the axillary cell, or only slightly so; the small cross-vein is placed just at the anterior border of this band. The costal band is broadly connected with the basal one, its lower border going exactly along the third “vein, surpassing it only near the tip of the wing, and there ending (infuscated) 236 PROF. M. BEZZI. before the middle of the first posterior cell; it contains two rather large fuscous. spots on the fore border in the marginal cell. From the costal band, about at middle of the last portion of the third vein or a little beyond it, departs the fuscous narrow middle band, which, crossing obliquely beyond the middle of the last portion of the: fourth vein, ends at the hind border just below the tip of the fourth vein itself. The fuscous cubital band is about twice as broad as the middle one; it begins at the hind border of the wing symmetrically with the end of the fifth vei, and passing over the hind cross-vein fills up broadly the end of the discoidal cell, and ends, more pale-coloured, in the first posterior cell, but without reaching either the basal or costal band; the base of the first posterior cell is thus completely hyaline. The broad axillary lobe is hyaline, with an ill-defined fuscous middle spot; the alula is. quite hyaline. The oblique folds in the discoidal and third posterior cells are about: as dark as the axillary vein. Type dj and type 9, a single couple of specimens from British East Africa, Kabete,. 28.vi.1914 (7. J. Anderson). Trirhithrum, Bezzi, 1918. The peculiar character of the wing pattern in the present genus is not (as wrongly indicated on p. 233 of my previous paper) that the middle band is united to the: marginal one, but that the cubital band (or the band on the hind cross-vein) is united to the basal or anal band. Another error is to be found in p. 239 in the description of the new species 7'r. occijntale, which has the thorax shining black, and not shining” yellowish. To the present genus is to be ascribed also Ceratitis inscripta, Graham, which is the same as 7’. coffeae, Bezzi, as stated by Dr. G. A. K. Marshall ; and the following’ very distinct and robust new species, which seems to be allied to T. nitidum, v. Roeder, in having a complete middle band on the wings. Trirhithrum validum sp. nov. A robust fly of comparatively large size, entirely shining black, with whitish mesopleural stripe and whitish scutellum, and with a complete middle band on the: wings, which have also rather distinct black streaks at the base. ©. Length of body 7°5 mm. ; of ovipositor 1°5 mm. ; of wing 8 mm. Head with the occiput shining reddish in the middle and above, black on the: sides; frons dull yellowish, face whitish on its lower part, jowls reddish like the lower occipital swellings, which are shining black above. The frons is about as broad as long, unspotted, only the ocellar and the narrow and short orbital plates being” shining reddish. The cheeks are whitish, but their upper half, by the roots of the antennae, is purplish. Antennae wanting in the case of the type. Upper half of the face, just below the antennal root, blackish and grey-dusted. Jowls of a purple: colour, with a dark brown spot below the eye. Palpi and proboscis reddish brown. All the cephalic bristles are black, even the pvt. and the genal ones; two 2. or. ; oc. very strong and long ; occipital hairs black, even below ; mouth-border on the. sides with a row of black bristly hairs. Thorax shining black on the back, with short dense black pubescence ; humeri black, narrowly reddish brown behind ;. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 237 pleurae shining brown, with a broad oblique whitish mesopleural stripe, on which there are long and pale yellowish hairs, while the hairs are black on the rest; no distinct lighter hypopleural spots. Postscutellum and mesophragma shining black. Scutellum of a semicircular shape, flattened above, shining whitish, with three broad shining black spots at hind border which are not visible from above ; it is clothed above with rather long whitish hairs and bears four bristles, the middle pair being not decussate. Thoracic bristles black, even the long and approximated middle scp.; dc. placed on the line of the a. sa.; one mpl.; pt. stronger than the st. Halteres blackish, with yellowish base ; squamulae dark yellowish, with blackish border and with dark fringe. Abdomen with parallel sides, not broader than the thorax and not narrowed at base; it is shining black, clothed with black hairs, and with black bristles on sides and at end; the hind border of the first, and the whole of the third segment are densely clothed with a whitish-grey dust, thus forming two transverse bands, which are clothed with whitish hairs; the fourth segment is reddish brown on the middle ; ovipositor elongate triangular, as long as the last three abdominal segments together, depressed, shining black, with black hairs and with the apical segment reddish ; venter shining black. Legs with reddish brown coxae and femora, the latter more blackened outwardly, and with black pubescence ; tibiae and tarsi pale yellowish and with whitish pubescence ; bristles of the femora black, those of the tibiae tawny. Wings long and narrow, hyaline, with a blackish pattern ; costal bristle strong but short; veins dark brown or blackish, but the costa and the first longitudinal pale yellowish on the basal hyaline indentation. The extreme base is blackened, being separated from the basal band by a triangular hyaline indentation, the base of which is at costa in the two basal cells, and the vertex is below the end of the second basal cell; in this hyaline part there are some blackish dots, likewise in the basal blackish part there are some narrow hyaline streaks, forming thus a Ceratatis-like pattern, though closer and less distinct. The stigma is entirely black. The basal band begins at the stigma and extends to the hind border below the end of the sixth vein, filling up almost the whole of the axillary lobe, and being thus broadly united to the basal blackish part; the whole of the first basal cell, and the base of the discoidal and third posterior cells are broadly united with this band. The marginal band is broadly united at base with the basal one, and does not extend beyond the third vein, passing only a little over it in its terminal part, and ending at costa near the first third of the first posterior cell ; it encloses two darker dots in the marginal cell, and two hyaline streaks at costa before and behind the end of the second longitudinal vein. About at middle of the last section of the third vein arises the narrow, but complete and even, middle band, which, crossing beyond the middle of the last portion of the fourth vein, ends at hind border just below the end of this same vein. The oblique cubital band is broader than the middle one, and is broadly united with the basal one at the small cross-vein ; it runs along the hind cross-vein, leaving hyaline the upper external corner of the discoidal cell, and ends at hind border symmetrically with the end of the fifth ven. Small cross-vein long and oblique, the first posterior cell being as broad as the discoidal cell, which is very acute below, the hind cross-vein being very oblique. Alula entirely infuscated. Type 9, a single specimen from Uganda, Entebbe, 12—13.xii.1912 (C. C. Gowdey). 238 PROF. M. BEZZI. Xanthorrhachista, Hendel, 1914. Xanthorrhachista alata, Becker, 1910. In my previous paper I overlooked the description of Acidia alata, Beck. (Ann. Soc. Ent. France, 1910, p. 28), with which Xanthorrhachista cephalia, Hend. (1914), is undoubtedly synonymous. Themarictera, Hendel, 1914. To the characters of the genus on p. 243 of the first paper must be added : No oc. ; two pairs of vt.; two s. or. and two 2. or.; put. well developed ; genal bristle thin ; ocp. long and acute, black. Scp. long, the middle ones approximated ; no pt.; st. very short and thin, not properly wanting, as stated by Hendel; three p. sa., the inner one thinner. Themariciera laticeps, Loew, 1861. Closely allied to T. flaveolata, F., but differing in the presence of two hyaline spots, one in the submarginal and one in the first basal cell. The first of these spots is placed on the limit between the brown and the yellow pattern, beginning at the third longitudinal vein just beyond the small cross-vein, and extending above nearly to the middle curve of the second vein; in this way is formed the black indentation, which is exaggerated in Loew’s original figure. The second hyaline spot is before the small cross-vein; it is ovate in shape, and extends from the third to the fourth vein. These two spots are distinctly whitish, like the basal streaks, as described by Loew. In the rest the unknown male is very like the female ; the genitalia are rounded, not prominent, entirely shining yellow; the last abdominal segment has numerous black bristles. The eyes are green, with two purplish cross bands, converging anteriorly, on the upper half. A male specimen from Natal, Durban, xi.1916 “reared from native fruit” (C. P. van der Merwe, Agric. Dept. 8. Africa). This specimen is an aberrant one, or is not fully coloured in the wing pattern, hyaline streaks being present in the ends of the marginal, submarginal and discoidal cells and the whole middle of the first posterior cell; the fuscous pattern is thus restricted to some broad borders on the longitudinal and cross-veins. Puparium. To the reared specimen is attached a puparium, which seems to be rather small for such a fly ; it measures 6 mm. only in length, while the adult fly is about 9 mm. long. It is of a dark reddish-brown colour and has a rather hard and thick skin; the segmentation lines are not impressed ; the broadened areas of the ventral side are not prominent ; the posterior spiracles are very approximated, the surrounding parts being of a blackish colour. The surface is smooth, but quite opaque. Rhacochlaena, Loew, 1862. The Ethiopian species of the present genus, with the addition of the new one, may be distinguished as follows : 1(2). Wings destitute of a broad apical fuscous band, the first and second posterior cells being almost completely hyaline wh .. » +. fascrolata, Lw. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 239 2(1). Wings with a broad fuscous apical band, the first and second posterior cells , being for the greater part infuscated. 3(4). Apical hyaline spot of the wings broad, uae on both sides beyond the ends of the 3rd and 4th veins He x pulchella, sp. n. 4(3). Apical hyaline spot small and not extending ee beyond the ends of the above-named veins ae a i .. hammersten, End. Rhacochlaena fasciolata, Loew, 1863. The apex of the wing is occupied by a white rounded spot, which in certain lights is very conspicuous. The ovipositor is 2 mm. long; it is cylindro-conical, shining red, but broadly black at base above and more narrowly at tip ; it is clothed with very short and dense, darkish pubescence. In the present species there is but a single s. or. A couple of specimens from Natal, Estcourt, ix. and x. 1896 (G. 4. K. Marshall). Rhacochlaena puichella, sp. nov. (Pl. xvu, fig. 8). Very near hammersteim, but at once distinguished by the much broader apical hyaline spot of the wings. g. Length of body 45 mm.; of wing 4 mm. Head reddish, face jowls and lower part of occiput pale yellowish or whitish ; frons opaque, with shining vertical plates and slightly whitish sides; the small ocellar dot is black. In the middle of the face there is an ill-defined fuscous cross band, not reaching the sides. Antennae entirely reddish, but the third joint a little infuscated at the end; the third jomt is rather narrow, with the upper border distinctly concave ; it is 2} times as long as broad, with rounded tip; arista shortly pubescent. Jowls very narrow, narrower than the breadth of the third antennal jomt. Hyes greenish, with two parallel cross bands of a purplish colour, one a little beneath the middle, the other on the upper third. Palpi and proboscis yellowish. Cephalic bristles black ; or. 1+ 3; no oc.; put. less developed; the longer bristles are rather ~nrtel at the sit. ilnoree: black, opaque, with greyish dust on the back; a middle longitudinal stripe and the sides are reddish ; pleurae shining yellowish ; along the notopleural line, from the humeri to the root of the wings, there is a rather broad whitish stripe. Scutellum whitish in the middle and infuscated on the sides. Mesophragma black and a little shining ; the hypopleurae are also black, but with a narrow whitish stripe in continuation with the notopleural one. Squamulae and halteres whitish. Chaetotaxy typical; all the bristles are black, but the longer ones have yellowish ends; four scutellar; the short pubescence is pale yellowish. Abdomen black, with the last segment reddish brown behind, rather shining, with dark pubescence and black terminal bristles. Genitalia reddish brown, shining, with short yellowish appendages beneath. Legs entirely yellowish ; bristles of front femora and spurs of middle tibiae black. Wings nearly the same as in Enderlein’s fig. R, p. 441, with the following differences : (a) the first fuscous band is less developed and much shorter, ending near the upper corner of the second basal cell; (6) the second fuscous band is longer, reaching the hind border of the wing or nearly so ; (c) the third abbreviated band is narrower ; (d) the apical hyaline spot is twice as broad, extending symmetrically above in 240 PROF. M. BEZZI. the submarginal and below in the second posterior cell; it is distinctly whitish ; (e) the hyaline indentation into the second posterior cell is narrower; (f) the hyaline dot at the end of the second longitudinal vein and the hyaline streak in the middle of the first posterior cell are faintly developed, or even indistinct. Type 3, a single specimen from the Gold Coast, Accra, i. 1916 (Dr. J. W. S. Macfie). * x x * sk *k The Ethiopian species referred by authors to Acidia, were provisionally placed by me under the head of Philophylla; but after the removal of obnubila and coloniarum to Celidodacus, there remain only ocellata (which belongs to a new genus) and seychellarum, which on account of its eyes certainly cannot be ascribed to Philophylla. This last genus seems therefore to be absent from the Ethiopian fauna. The genera of this group must be disposed as follows, with a modification of my table on pp. 218-219 of the first paper. 43(42). Prst. always present; two s. or. 44(45). Small cross-vein near or before the middle of the discoidal cell. a(b). Wings of usual shape, with straight costa and long second posterior cell, adorned with narrow vertical fuscous bands, which are often partly broken into spots... : a Phorellia, R. D. b(a). Wings towards the middle very Boas er the costa rounded outwardly and with very short second posterior cell; they are adorned with a broad rounded fuscous patch and with some curved dark bands Taomyia, gen. nov. © 45(44). Small cross-vein beyond the middle of the discoidal cell, and often very near the hind cross-vein. c(f). Eyes in profile narrow, much higher than broad. d(e). Occipital lower swellings not developed; wings with yellow longitudinal stripes, the lower one bifurcate at end; thorax with black spots Notomma, gen. nov. e(d). Occipital lower swellings rather developed ; wings with yellowish or fuscous, partly confluent, cross bands; thorax not black-spotted Philophylla, Rond. f(c). Eyes in profile broad and rounded; if they are rather narrow, the occipital lower swellings are not developed. g(h). Wings of usual shape, with short stigma and distinct costal bristle ; oc. well developed ; wings banded, at least on the apical half § Acrdra, R. D. h(g). Wings rather narrow, with parallel sides, with elongate stigma and without costal bristle ; oc. very thin or even wanting ; wings not banded, with the fore half usually entirely infuscated .. Ocneros, O. G. Costa. Taomyia, gen. nov. The present new genus is nearly allied to Acedia and Philophylla, difiering from both in the very different shape and venation of the wings, which recall that of the Neotropical genus Parastenopa, Hendel, which however has no oc. and a differently shaped pterostigma. Head in front view a little higher than broad, in side view NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 241 ‘with rather narrow eyes and distinct lower occipital swellings. Frons as broad as the eye, distinctly longer than broad, flat, with very short orbital plates ; lunula broad. Face shorter than the frons, convex in the middle, with no prominent upper mouth-border and with diverging antennal grooves; cheeks linear; jowls narrow. Antennae inserted a little below the middle of the eyes, rather separated at base, shorter than the face ; second joint less prominent above and not spinulose ; third joint linear, twice as long as the basal ones, rounded at tip; arista quite bare. Palpi not very broad, bristly ; proppscs short. Ocp. black and acute, less numerous ; oc. very long and strong; one s. or. ; 2 or 37. or., only the middle pair being ise the other very short; genal bretle well developed. Thorax elongated, about twice as long as broad, with complete chaetotaxy, but the middle scp. are not developed ; de. placed much behind the line of the a. sa.; one mpl. ; pt. rather thin. Thoracic suture broadly interrupted in the middle. Scutellum bluntly triangular, convex above, bare, with four bristles, those of the apical pair decussate at the end. Abdomen oval, elongate, not broader than the thorax, with terminal bristles; male genitalia prominent and appendiculate. Legs rather slender ; front femora with 5 bristles below ; middle tibiae with one spur ; hind tibiae with only 4-5 longer bristles on the middle of the outer side. Wings of usual shape in the basal half, very broad and nearly semicircular in the apical half, broadly infuscated, with hyaline stripes, surrounding a large rounded middle patch. Costa very curved on its last portion ; costal bristle hardly distinguishable ; first vein not specially curved upwards at the end, the stigma being longer than broad ; second vein straight ; third vein bristly to the middle of its last portion, which is strongly curved downwards towards the end, and thus very divergent from the second and parallel with the last portion of the fourth ; submarginal cell very widened at end ; discoidal cell very long and much broadened outwardly, the hind cross-vein being perpendicular, arched outwardly, about three times as long as the basal cross- vein of the discoidal cell, and about as long as the términal portion of the fourth vein ; small cross-vein on or just beyond the middle of the discoidal cell ; second posterior cell unusually short, higher than long ; lower angle of the anal cell drawn out into an acute point, which is shorter than the second basal cell; axillary lobe broad ; alula well developed and rounded. —_ : the following new species. It is very probable that Acidia (*%) ocellata, Lamb, 1914, from the Seychelles, belongs to the present genus, having a similarly formed head and a very similar wing pattern; but it has the small cross-vein placed beyond the middle of the discoidal cell, and two s. or. on the frons. Taomyia marshalli, sp. nov. (Pl. xvii, fig. 9). A very distinct fly of comparatively large size, at once distinguishable on account of its very characteristic wings. ¢. Length of body 7°5 mm. ; of wing 7 mm.; breadth of a wing 3°2 mm. Head of a dark reddish-brown colour ; occiput shining, broadly blackened on the sides, light yellowish near the eye-borders and below ; frons rather shining, with a eas ocellar dot and an ill-defined blackish nes band in front above the antennae ; lunula shining yellowish ; face shining, dark brown in the middle, more 242 PROF. M. BEZZI. yellowish towards the antennal grooves and on the sides; cheeks whitish ; jowls unspotted. Antennae entirely pale yellowish, the third joint a little infuscated along the upper border ; arista pale yellowish at base. Palpi yellowish ; proboscis dirty brownish. All the cephalic bristles are black and rather thin; the frontal stripe seems to be bare on the middle. Thorax black on the back, but clothed with greyish dust and with three narrow parallel longitudinal black stripes, those on the dorsocentral lines extending about to the scutellum, while the middle one is abbreviated behind the transverse suture. The short pubescence of the back is black. The shoulders, the notopleural suture and the postalar calli are reddish brown, like the pleurae, which are blackened on the middle of the meso- and sterno- pleura, but are pale yellowish on the pteropleura, forming thus a broad perpendicular stripe; hypopleura blackish and shining, like the postscutellum and the meso- phragma. Scutellum entirely yellowish with a narrow black basal band. All the thoracic bristles are black and rather thin. Squamulae dirty brownish ; halteres pale yellowish. Abdomen black, shining, but with dark dust; pubescence and bristles black ; genitalia yellow, black above and behind ; the unpaired and strong appendage of the underside is similarly yellow. Legs pale yellowish, but the four posterior femora above on the apical half, and all the tibiae except the tip, are more or less blackened ; bristles black, those of the front femora very thin. Wings whitish hyaline, with a very broad, dark brown pattern; an abbreviated stripe, extending from the middle of the stigma to the third longitudinal vein is yellow. Veins yellow, but darkened on the dark parts. Costal bristle very small. Extreme base of the wing hyaline; an arcuate brown band begins along the costa, filling up the whole of the first and the base of the second costal cell, and becoming gradually broader passes over the middle of the second basal cell, fills up the whole of the anal cell and is prolonged as an acute stripe along the sixth longitudinal vein, ending with a point at the hind border of the wing. The axillary lobe and the alula are quite hyaline. After this dark band there is a narrow, complete hyaline band, going from the middle of the second costal cell, over the apex of the second basal cell and over the base of the third posterior cell, to the hind border of the wing. The extensive brown pattern of the broadened apical half of the wing is thus sharply separated from the basal one ; it consists of a rounded, very broad patch, which occupies the whole of the discoidal cell, and the greater part of the submarginal and first basal cells, and the entire third posterior cell. At the fore border there is the yellow stigmatic indentation, and after this a broader triangular oblique hyaline indentation, which reaches with its obtuse poimt the fourth vein, entering the end of the first basal cell ; from this indentation departs a rather narrow hyaline arcuate band, which, running along the fore border, from which it is separated by a narrower brown costal band, ends in the upper angle of the second posterior «ell. The yellow stigma is blackish brown at base and on the lower spical half. Type g, a single specimen from Natal, Verulam, vil. 1897, collected by Dr. G. A. K. Marshall, in whose honour this fine insect is named. Notomma, gen. nov. The present genus is erected here for a species which has a wing pattern recalling that of Xanthorrhachista, from which it differs in general appearance and in the NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 243 position of the small cross-vein, and Craspedozantha, from which it differs in the bristly third longitudinal vein and in the position of the dc. bristles. In Prof. Hendel’s table it runs down to Neanomoea, an undescribed Oriental genus from Formosa. | Trypeta jucunda, Loew, seems to be an allied form, but differs in the venation and in the shape of head, belonging probably to some peculiar, as yet undescribed genus. | Head a little narrower than the thorax, in front view about as broad as high, in profile rather flat, the eyes being narrow, with the vertical diameter about twice as long as the horizontal. Occiput flat, even a little convex on the upper half, destitute of distinct lower swellings. Frons gradually narrowed from the vertex to the antennae, flat, narrower than one eye in front, about one and a half times as long as broad, in profile a little prominent above the base of the antennae. Lunula broad. Eyes greenish, with a purplish horizontal band just above the middle. Face flat, with a distinct, broad and rounded middle keel and with shallow antennal grooves ; mouth-border prominent; cheeks linear; jowls narrower than the breadth of the third antennal jot. Antennae inserted a little above the middle of the eyes ; second joint short, small, bristly above ; third joint broad, pubescent, parallel-sided, rounded at tip, not reaching the mouth, less than twice as long as: broad ; arista bare. Palpi broad, bristiy ; proboscis short and thick. Cephalic: chaetotaxy complete and well developed ; ocp. acute, black ; put. thin and parallel ;: two pairs of vt., the inner ones not much longer ; oc. robust but not long; two s. or. and three 7. or. ; genal bristle strong, like the bristle on the lower part of the occiput. Thorax robust, flat, subquadrate, with complete chaetotaxy ; the middle scp. are approximated and as long as the outer ones; a. sa. strong; dc. placed much behind the line of the a. sa; prsct. approximated ; two mpl. ; pt. about as strong as the st. Scutellum convex above somewhat Ceratitis-like in shape but not trilobate, with four strong bristles, the apical ones diverging. Abdomen in the middle broader than the thorax, narrowed at base and at the end; it has five segments, the last with bristles at the sides and behind; ovipositor elongate triangular, rather swollen. Legs short and stout; front femora with a complete row of strong bristles beneath ; middle tibiae with one spur; hind tibiae with a complete row at the outer side; claws curved; middle tibiae without bristles on the hind side. Wings elongated and proportionally narrow ; costa not ciliated: and without costal bristle. They are adorned with two longitudinal yellow bands, one along the fore border and the other along the middle; this last is bifurcate. into a recurrent band across the hind cross-vein and along the hind border of the. wing. Auxiliary vein strong; first longitudinal vein long, ending beyond the. middle of the wing but always before the small cross-vein; stigma long, in the middle about three times as long as broad; second and third veins rather: approximated, slightly sinuous, the third bristly throughout its whole length ; the last portion of the fourth vein is slightly sinuous and nearly parallel with the last portion of the third, which ends nearer the tip of the wing, but the first posterior cell is much broader at end than at base; small cross-vein placed exceedingly near the end of the discoidal cell, its distance from the hind one being about equal to its own length; hind cross-vein curved and oblique outwardly, its lower end: (C637) Cc 244 PROF. M. BEZZI. being very near the hind border of the wing; discoidal cell very broad at end ; second posterior cell very short, above only a little longer than broad at base; second basal cell rather broad and short; anal cell with the lower angle acute and prolonged, but not longer than the second basal cell; sixth vein reaching the hind border of the wing. Axillary lobe broad and short ; alula rounded. Type: the following new species. Notomma bioculatum, sp. nov. (Pl. xvu, fig. 10). A reddish brown species, with two deep black spots on the middle of the back of the mesonotum. ©. Length of body 6°5 mm. ; of wing 6°5 mm. ; of ovipositor 1°5 mm. Head entirely yellowish, with only a small black dot on the ocelli. Occiput shining, on the lower part with faint pale dust and with some rather long and soft pale hairs. Frons opaque, with shining ocellar and vertical plates and some scattered dark hairs on the middle. Face with whitish dust. Antennae entirely yellowish, even the arista; palpi and proboscis yellowish, the latter with dirty blackish flaps. All the cephalic bristles are black. Thorax reddish brown, shining on the back; it is clothed with faint greyish dust and short black pubescence, disposed in regular approximate longitudinal rows; the two very striking, deep black, opaque spots are of oval shape and placed on the dorsocentral lines, just behind the suture ; the pleurae are paler and more shining, and clothed with rather long, scattered black hairs. Scutellum coloured and pubescent like the back of the mesonotum ; at hind border it has three ill-defined, shining brown spots, a smaller one between the apical pair of bristles, and a larger one on each side between the basal and apical pairs of bristles. All the bristles are black. Squamulae reddish, with a dark fringe; halteres reddish yellow. Postscutellum reddish, with two rounded black spots ; mesophragma entirely of a shining reddish colour. Abdomen shining reddish, with black pubescence and black bristles ; the hind border of the segments is more lightly coloured ; the venter likewise ; ovipositor shining reddish. Legs entirely reddish, with paler tibiae and tarsi ; bristles and spurs black. Wings shining and iridescent with yellowish veins. The first yellowish band is complete from the base to the tip of the wing, going throughout along the costa, but leaving two hyaline streaks in the marginal and submarginal cells, just before and behind the end of the second longitudinal vein ; m the middle this band does not extend below the second vein, while at base it reaches the fourth vein just above the second basal cell, and at the end it extends below the third vein to the middle of the end of the second posterior cell; at the end there is a blackish infuscation, on both sides of the apex of the third longitudinal vein; in the middle of the marginal cell there is a black rounded spot, just beyond the entirely yellow stigma. The second longitudinal band is of a paler yellowish colour above, but infuscated below and at the end ; it begins on the anal cell and on the lower part of the second basal cell, and goes across the base of the discoidal and the apex of the first basal cell to the small cross-vein ; it there becomes narrower and is bent downwards, going along the last portion of the fourth vein to the costa, which is reached just below the tip of the fourth vein itself. From this band arises another, which forms a border along the hind cross-vein and extends beyond it as a border along ~ NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS, 245 the hind border of the wing to the end of the sixth vein; this recurrent band is yellowish interiorly and fuscous exteriorly. The hyaline band between these two yellow longitudinal bands is nearly equal in breadth throughout its whole length and is bent just above the small cross-vein, entering there as a small streak into the marginal cell, below and before the rounded black spot of the fore border. The hyaline band between the central yellow band and the recurrent posterior band ends with a rounded tip in the base of the first posterior cell just behind the small cross-vein ; at base this band is united with the hyaline axillary lobe. The second basal cell is almost entirely hyaline, while the discoidal and third posterior cells are almost entirely infuscated except the broad hyaline band mentioned above. Type 9, a single specimen from the Gold Coast, Aburi, 1912-13 (W. H. Patterson). Acidia, Rob.-Desv., 1830. Three species are provisionally placed in the present genus, those here described as new being very like the European A. caesio; they may be distinguished as follows :— 1(4). Discoidal and third posterior cells with hyaline streaks or spots. 2(3). Thorax and femora black ; the apical dark bands of the wing broadly united with the basal patch, the second hyaline indentation of the fore border being quite wanting; discoidal and third posterior cells each with a hyaline rounded spot *: P _ seychellensis, Lamb. 3(2). Thorax and femora entirely yellow ; iis apical dark bands quite separated from the basal pattern by the second indentation, which forms a complete hyaline cross-band ; discoidal and third Laos cells, each with a hyaline streak .. = fossatacformis, sp. n. 4(1). The above-named cells at Aivaley fastens ana deseo of hyaline parts; body and femora entirely black; wings with two hyaline indentations at fore border .. as My. £4 af A homogenea, sp. n. Acidia fossataeformis, sp. nov. (Pl. xvii, fig. 1). An elegant species very like the Oriental A. fossata, F., but at once distinguished by its entirely vellow thorax and legs. g 2. Length of body 45-5 mm.; of wing 4-4'5 mm.; of ovipositor 0°8 mm. Head entirely yellow and unspotted, except the very small black ocellar dot. Occiput flat on the upper half, even a little concave above the median tubercle, convex but not very prominent on the lower half, without distinct swellings. Frons reddish yellow, opaque, with scattered dark hairs and moderately shining vertical plates ; it is parallel-sided, narrower than the eye, 24 times as long as broad ; lunula broad and short, semicircular, whitish, with brown upper border. In profile the frons is only a little prominent above the antennae; the eyes are rather rounded, their vertical diameter being only a little longer than the horizontal. Face flat and long, not prominent at mouth-border, with a broad middle keel and not very deep antennal grooves; cheeks linear, whitish ; jowls whitish, unspotted, as broad as the breadth of the third antennal joint. Antennae entirely yellow, inserted at the middle of the eyes, short, extending only to the middle of the face ; second joint swollen and prominent, with a short bristle above ; third joint rather (637) C2 246 PROF. M. BEZZI. small, short, attenuated apically but with rounded tip; arista with rather long pubescence, but the breadth of feathering narrower than the breadth of the third antennal joimt. Palpi and proboscis short, yellowish, pale-haired. All the cephalic bristles are black and rather thin; ocp. well developed and acute; pvt. parallel ; outer vt. half as long as the inner ones; oc. short and thin; two s. or. and three 2. or. ; genal bristle well developed. Thorax reddish yellow on the back, very little shining, with faint whitish dust and short dark pubescence ; the pleurae are paler and more shining, with scattered longer dark hairs. All the bristles. are black; chaetotaxy complete ; scp. well developed, the middle ones approxi- mated ; dc. placed much behind the line of the a. sa. and near to the prsct.; only one strong mpl.; pt. as strong as the st. Scutellum triangular, as long as broad, ‘flat above, with lateral keels; it is coloured like the back of the mesonotum, but is more bare and shining; it has four black bristles, the apical ones being shorter and decussate. Squamulae and halteres pale yellowish ; the former with a short thin white fringe. Postscutellum and mesophragma entirely reddish yellow and shining. Abdomen of usual shape, as broad as the thorax towards the middle; it is shining reddish yellow, but sometimes very infuscated in the male and even brownish or blackish; the pubescence and terminal bristles are black; venter more yellowish, with faint pale dust. Male genitalia shaped as in fossata, with a pair of long and stout appendages, the basal one reddish, while the apical one is black, and longer and stouter. Ovipositor with the basal segment broadly triangular, shining black, the apical segment narrow and dark reddish; the contrast between the reddish abdomen and the black ovipositor is very striking. Legs rather slender, entirely yellowish ; front femora with 3-4 long black bristles beneath at the end; middle tibiae without bristles, with one black spur at the end ; hind tibiae with no distinct row. Wings shining and iridescent, with non-ciliated costa and with a short but distinct costal bristle. The veins are yellowish at base and on the hyaline parts, black on the rest; the second longitudinal vein is a little sinuous ; the third vein is bristly throughout its whole length; the second, third and fourth veins are equally divergent towards the apex, ending at equal distances ; the last portion of the fourth vein is straight ; the small cross-vein is placed beyond the middle of the discoidal cell; the hind cross-vein is straight, perpendicular, a little longer than its distance from the small one; inferior angle of the anal cell produced, long and narrow, but not longer than the second basal cell; sixth longitudinal vein complete. The base of the wing is yellowish in the basal and costal cells, even the brown stigma being yellow at base. The pattern is blackish brown and about the same as in fossata (vide Bezzi, Mem. Ind. Mus., ui, 1913, pl. ix, fig. 48), which however has the wings distinctly broader. Type 3, type 2, and some additional specimens of both sexes from Nyasaland, Chiromo, Ruo R., 2.1.1917 (R. C. Wood). Acidia homogenea, sp. nov. A black species, with a wing pattern very like that of the European A. caesio, but at once distinguished from it and from the other species by the discoidal and third posterior cells being completely infuscated, without hyaline spots or streaks. Q. Length of body 4 mm. ; of wing 4 mm.; of ovipositor 0°6 mm. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 247 Entirely shining black, the head, antennae, tibiae and tarsi in part yellowish. or whitish. Head as broad as the thorax, rounded in front view and about as high as broad ; it is rather narrow in lateral view, the eyes being considerably higher than broad. Occiput entirely black, shining above, grey-dusted below, without lower swellings ; it is narrowly reddish on its lower portion. Frons narrow and long, parallel-sided, narrower than the eye, twice as long as broad ; it is entirely reddish, opaque, with shining white sides, and with a small blackish ocellar dot, the narrow and short vertical plates are shining; lunula broad, semicircular, whitish. Antennae inserted near the middle of the eyes ; the two basal joints are yellowish ; third joint wanting in the type. Face only a little shorter than the frons, as broad as it above, but twice as broad below; it is flat, with shallow antennal grooves, and no prominent mouth-border; it is blackish, with a shining white dust, yellowish only on the flat middle keel; cheeks linear, hardly distinguishable ; jowls very narrow, reddish, unspotted. Palpi and proboscis dirty yellowish, the former dilated at the end, the latter short and thick. Cephalic bristles missing in the type, but judging by the insertion points there are two s. or. and three equally distant 7. or. Thorax entirely shining black, not dusted on the back; scutellum similar, triangular, flat above ; pubescence and bristles abraded in the type, but the mpl. and st. are present on one side and are black; the scutellum has certainly four bristles; the dc. are placed very near the suture. Squamulae brownish ; halteres black. Abdomen entirely shining black, with short black pubescence and black bristles at the end; ovipositor shining black, triangular, flat, pro- portionally short; apical segment reddish; venter black. Legs with shining black femora ; front tibiae entirely, four posterior ones on the apical half, whitish, like the whole of the tarsi. Wings rather broad, with the membrane shining and iridescent, chiefly on the hyaline parts; costa and veins black, but whitish on the hyaline parts. First longitudinal vein short, the stigma being about as long as broad at base; second vein straight; third vein bristly to the small cross-vein, bent downwards beyond middle of its last portion ; last portion of the fourth vein curved and slightly diverging from the third; small cross-vein beyond the middle of the discoidal cell; hind cross-vein straight and perpendicular, a little longer than its distance from the small one; lower angle of the anal cell narrow and shortly produced. The blackish pattern is very like that of A. caesio, or of the Formosan “ Trypeta” superflucta, End. (Zool. Jahrb. xxxi, 1911, p. 428, fig. J) ; but it differs in having more than the basal half of the wing entirely blackish, without any hyaline indentation of the fore border before the stigma, and without any hyaline spot or streak in the discoidal or third posterior cells ; only the axillary lobe is hyaline. The stigma is deeper black than the surrounding parts. The first hyaline indenta- tion of the fore border beyond the stigma is as in caesio ; the second likewise, but ending with a narrow point on the fourth longitudinal vein, without entering the discoidal cell. The two hyaline apical indentations of the hind border are as in caesio, the black streak between them being complete and not interrupted at base (difierimg thus from that of fossata and fossataeformis, in which it is isolated). Type Q, a single specimen from Nyasaland, Port Herald, iv.-vi. 1913 (Dr. J. E. S. Old). 248 PROF. M. BEZZI. Oeneros, O. G. Costa, 1844. In the typical species of the present genus the dc. are placed near the suture, while in the known Ethiopian species they are placed much behind. It seems thus probable that the latter species are not congeneric with the palaearctic ones ; but I have not erected a special genus for them, because they may perhaps be better placed in the following genus Ocnerioxa, which differs only in the non- bristly third longitudinal vein. The genus seems to be abundantly represented in the Ethiopian fauna; I have before me two additional new species, which belong to the group of O. sinuatus. This group is not only distinguished by its colour characters, but also by the slender body, the longer antennae (which are nearly extended to the mouth- border), the distinctly pubescent arista, the thin and prevalently yellow macrochaetae of the head, thorax and scutellum, the longer pterostigma, and by the last portion of fourth longitudinal vein being quite straight. To the characters of the genus must be added that the 2. or. are accompanied by additional bristly hairs. The known species may be distinguished, by a modification of the table on p. 250 of my first paper, as follows :— 1(2). Scutellum entirely yellow ; discoidal cell wholly infuscated, without fe spots ; stigma as long as or longer than the second costal cell; antennae about as long as the face, with a shortly pubescent arista; macrochaetae | mostly yellow. a(d). Base of stigma narrowly hyaline, at least near the costa; there is a small hyaline dot in the base of the first posterior cell, just above the hind cross- vein. b(c). Hyaline spot in the marginal cell beyond the stigma triangular and extended to the second longitudinal vein; the fuscous portion of the wing has below only a median broad projection, which extends about to the hind border in the hyaline part of the wing ee . J. sinuatus, Loew. c(b). Hyaline spot of the marginal cell in the shape of an elongate stripe near the costa, which does not extend below beyond the middle of the cell; the fuscous portion is wavy below, with 4-5 pro} ections, which are all far from the hind border of wing of : He undatus, sp. n. d(a). Base of stigma with a broad hyaline ale no palit dot in the base of the first posterior cell; fuscous part of wing ending below with one acute median projeeinon: which reaches the hind border of wing along the fifth vein. 7 se .. bigemmatus, sp. n. 2(1). Scutellum Bevedie plaice in ate anianiloe discoidal cell with hyaline spots; stigma shorter than the second costal cell; antennae much shorter than the face, with bare arista ; macrochatae black mundus and excellens. Ocneros undatus, sp. nov. A slender and pale-coloured species of smaller size, very distinct on account of the lower border of the fuscous part of the wing being wavy and remote from the hind border of the wing. Q. Length of body 4mm. ; of ovipositor 1 mm. ; of wing 4°5 mm. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 249 Head entirely yellow and opaque; occiput with a narrow black stripe on each side, extending from the neck to the lower third of the eyes; frons narrower than _the eye and longer than broad, unicolorous, with a very small black ocellar dot. The face and unspotted jowls are yellowish, while the occipital lower swellings are more lightly coloured and surrounded above with the black stripe; upper mouth-border prominent. Antennae entirely pale yellowish, only a little shorter than the face, with the third jomt rather broad, and with a shortly pubescent arista. Palpi and proboscis pale yellowish. All the cephalic bristles are pale yellowish, only the short outer vt., the acute and numerous ocp. and the 2. or. being black ; there are no distinct ocellar bristles ; there are only two pairs of longer 7. or., but on the same line are some black, much shorter, bristly hairs. The lower occipital hairs are whitish; the bristly hairs near the borders of the mouth are yellowish. Thorax entirely yellowish, less shining, with pale short golden-yellowish pubescence ; on each side there are two black longitudinal stripes, one extending from the shoulders along the notopleural line to the root of the wing, and continuous with the black postalar calli and the very glistening black mesophragma; the other is parallel with the former but ends at the suture, being continued beyond it only by a short ill-defined darkish stripe; the occipital black stripe is contmuous with the notopleural one. The pleurae and the breast are entirely whitish yellow, unspotted, the contrast between the black and the whitish parts being very striking, when the thorax is viewed from the side. Hypopleural spot whitish. Scutellum of a light yellowish colour, like the posterior part of the back of the mesonotum, but a little more shining. Postscutellum shining black. All the macrochaetae are yellowish, those of the pleurae being lighter; one mpl.; pt. much weaker than the st. ; even the scp. are black. Squamulae blackish, with a long dark fringe ; halteres pale yellowish. Abdomen shining yellow on the basal half, blackened on the apical half, the last three segments having black lateral borders; venter whitish ; hairs and bristles yellowish. Ovipositor flattened, broadly obtuse at the end, shining reddish with pale pubescence. Legs entirely whitish yellow, those of the front pair wanting in the type; apical spur of the middle tibiae black ; hairs of the posterior row of the hind tibiae brownish. The wings are almost wholly and evenly infuscated ; the lower border of the fuscous part ends about on a line extending from the point of the anal cell to the lower anterior corner of the discoidal and the upper third of the second posterior cell ; but this line is wavy, having some rounded projections, three of which are in the second and three in the third posterior cell. There is besides a very small subhyaline rounded dot in the basal part of the first posterior cell, a little before the hind cross-vein and closer to the fourth than to the third longitudinal vein. The stigma is as long as the second costal cell and is black, with a narrow short subhyaline streak near the base at costa, being as broad as a third only of the stigma itself. The hyaline streak at fore border beyond the stigma is shorter than the stigma itself, and does not extend below beyond the middle of the marginal cell. The small alula is entirely brown, while the axillary lobe is entirely hyaline. The third longitudinal vein has 2-3 bristles beyond the small cross-vein. The last portions of the third and fourth longitudinal veins are quite straight and parallel, or only a little divergent. Type 9, a single specimen from Nyasaland, Mt. Mlanje, 2.11.1913 (S. A. Neave). 250 PROF. M. BEZZI. Ocneros bigemmatus, sp. nov. (Pl. xviu, fig. 2). Very like the preceding species, but distinguished by its larger size, and the different pattern of the thorax and wings. 7 ©. Length of body 5°5 mm.; of ovipositor 1mm. ; of wings 6 mm. Head, its appendages and bristles exactly as in the preceding species; the two occipital black spots are broader, shaped as a triangular patch, the outer corner of which does not reach the border of the eyes ; there are besides two approximated black stripes above the neck, directed towards the vertex, wanting in O. undatus. Thorax and scutellum, and their pubescence and bristles, exactly as in the preceding species ; there is also an equally narrow, black, notopleural stripe in continuation of the occipital one, and therefore there is the same striking contrast with the whitish pleurae ; the interior stripe is however much broader and ends at the suture, but is continued beyond it with a broad stripe on the dorsocentral lines, extending nearly to the lateral edges of the scutellum. The mesophragma is shining reddish, brownish fuscous on the sides, like the postscutellum. Halteres and squamulae as in O. undatus. Abdomen more elongate, shining reddish; the three terminal segments more broadly black on the sides and middle; bristles black; venter darker; ovipositor shining black, with blackish pubescence. Legs as in the preceding ; front femora with only three rather thin and yellowish bristles on the apical half beneath. Wings as in O. undatus, but with the following differences : (a) at base of the stigma there is a broad whitish-hyaline spot, extending below to the first longitudinal vein; (b) the whitish hyaline spot beyond the stigma has the shape of an oblique band, which goes from the end of the first vein across the middle of the marginal cell to the second vein ; (c) there is no trace of the hyaline basal dot of the first posterior cell ; (d) the lower border of the fuscous part is more irregular, showing a blackish tooth in the middle of the upper border of the third posterior cell, and another more elongate tooth on the outer inferior angle of the discoidal cell, reaching the hind border of the wing along the fifth longitudinal vein ; in the second posterior cell the border of the fuscous part forms a straight line, which has no indentations or projections (or only a little so beyond the fourth vein), the cell being thus mainly hyaline; along the middle of the anal cell there is also a subhyaline longitudinal stripe, which is wanting in the preceding species. Type 9, a single specimen from British Kast Africa, Embu, 12.xu1.1913 (G. H. J. Orde-Browne). Ocnerioxa, Speiser, 1915. The species of the present genus show a great resemblance to those of the first group (sinuatus, undatus and bigemmatus) of Ocneros; they have the same appearance and the same colour pattern of body and wings, being probably con- generic. They can be distinguished only by the third longitudinal vein being quite bare, or scarcely setigerous before the small cross-vein only ; and also by the bare, non-pubescent arista and by the less developed orbital hairs between . the fronto-orbital bristles. In the new material there is the following undescribed form. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 251 Ocnerioxa discreta, sp. nov. Very like O. wood: from Nyasaland, but more slender, smaller and distinguished by some minor details in the coloration of the head, abdomen and wings. ~. Length of body 4 mm.; of wing 45 mm. The body is narrower than in woodi, the thorax measuring about 1 mm. only in breadth, while in woodi it is 15 mm. broad. Besides there are the following differences : (a) the occiput has only the two lateral black stripes, the two middle ones being quite absent; (6) the frons is entirely whitish, even above the roots of antennae, the black band of woods being quite wanting, and thus the white lunula is not conspicuous ; (c) the third antennal joint is not at all infuscated ; (d) the black transverse band at the upper mouth-border is reduced to two small spots, one on each side of the mouth-edges, and these spots are moreover dull ; (e) the jowls are entirely whitish, quite destitute of the fuscous subocular spot ; (f) the black notopleural stripe is narrower, and there is a faint indication of an anterior median fuscous stripe, and the mesophragma is dark reddish in the middle ; (g) the abdomen is not at all greenish, and has the fourth segment yellowish, with narrow black sides and hind border; (4) the wing pattern is identical, but the fuscous tooth below the outer lower angle of the discoidal cell along the fifth longitudinal vein is much shortened, not being extended to the hind border of the wing. These small differences, together with the very different habitat, may be con- sidered of specific value. Type 3, a single specimen from N. Nigeria, Zungeru, 5.xi.1910 (Dr. J. W. Scott Macfie). Allotrypes, gen. nov. This is a rather aberrant genus, which on account of its bare third longitudinal vein, black and strong ocp. and its general appearance, comes in the vicinity of Acura, being eee very distinct from it in the chaetotactic caer as well as in the wing pattern. Head a little broader than the thorax, but in front view about as broad as high. Occiput distinctly concave above the neck, and less prominent below, with the lateral lower swellings but little developed. Frons flat or even a little concave before the lunula, with about parallel sides, as broad as one eye, only a little longer than broad. In profile the frons is not at all prominent, not even above the antennae; the eyes are rounded, with the vertical diameter only a little longer. Lunula small. Face long, only a little narrower but distinctly a little longer than the frons ; it is about parallel-sided, flat, with no prominent mouth-border, and with shallow but long antennal grooves. Cheeks linear; jowls very narrow, narrower than the third antennal joint. Antennae inserted distinctly above the middle of the eyes; they are very short, not extending beyond the middle of the face ; second joint not prominent, shortly setulose above, destitute of long bristles ; third joint pubescent, attenuate at end but obtuse, 14 times as long as broad ; arista shortly pubescent. Palpi and proboscis short, retracted into the oral cavity. Cephalic bristles very strong and peculiarly curved; no distinct put.; ocp. numerous, 252, PROF. M. BEZZI. long, unusually strong, black, acute at end ; inner vt. very long, more than twice as long as the outer ones, strong, curved backwards beyond their middle; one s. or. and one 2. or., equally strong and long and both curved backwards ; the single 2. or. is placed much in front, near the lunula ; oc. strong, but not very long ; on the sides of frons there are thus only two pairs of bristles, one near the base and one near the end; genal bristle less strong. Thorax proportionally narrow, less convex above, shining on the back, with the suture broadly interrupted. The: bristles are strong and the chaetotaxy is complete, except the scp., which are not distinct ; de. placed only a little behind the line of the a. sa.; prsct. only a little closer together than the dc. ; one mpl. only ; pt. and st. very strong, curved upwards. Scutellum triangular, as long as broad, flat above, with distinct lateral keels ; it has four bristles, the basal ones very strong and long, the apical shorter, thinner and decussate. Abdomen narrow and elongate, not narrowed at base, with about. parallel sides throughout its whole length; it has five segments, the last one with less developed bristles ; ovipositor flattened, triangular, about as long as the last three abdominal segments together. Legs proportionally slender and bare ; front femora with two long bristles below on the apical fourth ; middle tibiae with a single spur; hind tibiae destitute of a distinct row of bristly hairs on the outer side. Wings of usual shape, but rather narrow at base, the axillary lobe being little developed and the alula very narrow. They have a characteristic and rather aberrant pattern (for the group), consisting in a broad infuscation extending almost: over the whole wing, with the fore border broadly yellowish-hyaline on a little more than the basal half; there are no hyaline spots, and no hyaline indentations. Costa not ciliated ; with a short but strong costal bristle. Auxiliary vein typical, distinct throughout, with diffuse end; stigma short, about as long as_ broad ; first vein hairy and rather short, ending much before the small cross-vein, its distance from the end of the second vein being about twice as long as that of the second from the third vein. Second vein in its last half strongly diverging from the third, which is quite straight throughout and quite bare. Fourth vein straight and parallel with the third ; small cross-vein placed beyond the middle of the long discoidal cell; hind cross-vein straight, perpendicular and rather short, being shorter than, or nearly as long as its distance from the small one; last portion of the fifth vein proportionally long and strong; sixth vein complete, but colour- less on more than its apical half. Second basal cell but little developed ; anal cell broader but shorter than the second basal cell, with an acute but not or very little produced lower angle. Type: the following new species. Allotrypes brevicornis, sp. nov. (Pl. xvii, fig. 3). A narrow elongate species not unlike some forms of Aciwra, but very distinct on account of its wholly testaceous thorax, and its pecular wing pattern. Q. Length of body 4°5 mm.; of ovipositor 15 mm. ; of wing 4°5 mm. Head reddish above, pale yellowish below; occiput with two broad shining black spots of triangular shape, with the base on the eye-border and the narrow apex reaching the insertion of the neck ; these two spots are separated by a yellowish triangle which has the vertex at the neck and the base at the vertical edge. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 253: Ocellar dot smal! and black. Frons reddish yellow, shining, with sparse short hairs in the middle. Antennae entirely reddish yellow, with the third joint paler. Face pale yellowish, shining in the middle below; the cheeks and jowls are clothed with a faint whitish dust. Palpi and proboscis pale yellowish or whitish. All the cephalic bristles are of a deep black colour. Thorax and scutellum entirely shining testaceous ; the back has very short scattered black hairs and no distinct dust; humeri and pleurae paler than the back; all the bristles are black. Mesophragma shining, more or less infuscated in the middle, or even brownish. Halteres and squamulae reddish yellow. Abdomen shining testaceous, distinctly darker than the back of the mesonotum, even blackish brown in the middle of the segments and behind; pubescence and bristles black; ovipositor shining black on the basal half, dark reddish-brown on the apical half, but sometimes entirely black. Legs entirely reddish yellow, with the front tibiae and the front tarsi infuscated ; bristles and spurs black. Wings shining, but slightly iridescent ; veins and costa yellowish on the light parts, and brownish on the dark ones. Stigma yellowish. The yellowish-hyaline basal part of the pattern extends over the costal, marginal and submarginal cells to the third vein; the marginal cell however has its end brown, while the submarginal cell is brown on its apical half ; this clear basomarginal band is yellowish in the middle, being hyaline at the two ends only ; sometimes there is a distinct, but not sharply defined, rounded dark spot in the middle of the hyaline part of the marginal cell near the costa (this spot is wanting in the type). The rest of the wing is uniformly brown, only the posterior border of the third posterior cell, of the axillary lobe and of the alula being hyaline; the second posterior cell is somewhat more clear on the middle of its hind border. Type 9, a single specimen from Durban, Umbilo, 14.vi.1914 (ZL. Bevis); I have seen an additional specimen from Zululand in the collection of the South African Museum, Capetown. Aciura, Rob. Desv., 1830. Of the present “ genus ” there are in the new material before me three species, one of which is not yet described and differs from all the others in having all the occipital bristles black, like binaria, Loew, which is however a very different species, Aciura perpicillaris, sp. nov. (Pl. xviu, fig. 4). A small elongate slender species, which is allied to angusta in having the axillary cell rudimentary, but differs from it and from all the others of the same group in having all the occipital bristles black, and the discoidal cell entirely black with a single hyaline rounded spot near its extreme base. § 2. Length of body 3-3°2 mm.; of ovipositor 15 mm.; of wing 3‘6-4 mm. Head entirely shining black and very glistening on the occiput, vertex and orbital plates; the very narrow orbits and the linear cheeks are clothed with greyish or whitish dust; the very narrow jowls are dark reddish brown, like the anterior part of the frontal stripe. Lunula very small. Antennae inserted below the middle of the eyes, entirely black ; third joint rounded at tip and grey-dusted ; 254 PROF. M. BEZZI. arista long and with rather long pubescence, its feathering being about as broad as the breadth of the third antennal joint. Palpi whitish; proboscis dirty brown. All the cephalic bristles are black, even those of the occipital row; two s. or., the basal one much smaller; three 7. or., the apical ones much smaller; oc. as developed as the basal pair of the s. or. ; occipital hairs sparse and black. Thorax narrow and elongate, shining black, more glistening on the sides and on the pleurae ; the short and very sparse pubescence of the back is black; all the bristles are black; dc. placed before the line of the a. sa.; one mpl.; pt. weaker than the st. Scutellum shining black, with only the b. scf., which are very long and divergent. Postscutellum and mesophragma shining black like the hypopleural callosities ; halteres black; squamulae dark and with a dark fringe. Abdomen narrow and elongate, narrower than but about as long as the thorax ; it is shining black, with sparse black pubescence and short black bristles; in the male it is obtuse at the end, the genitalia being shining black and placed inferiorly at the side; in the female the ovipositor is as long as the abdomen, glistening black and with black pubescence, broad at base and acute at end, depressed ; the apical segment is dark reddish-brown at base. Venter shining black, with a faint dark grey dust. Legs rather long and slender; they have dark brown coxae and reddish trochanters ; the femora are shining black, but narrowly brownish on the sides ; the four anterior tibiae are yellow like their tarsi; the hind tibiae are broadly darkened at base, more broadly in the female, and the tarsi are dark yellowish; front femora with 2-3 thin black bristles beneath on the apical half; hind tibiae destitute of a distinct row of bristles on the outer side. Wings very narrow and elongate, with rudimentary axillary lobe and alula, but appearing not so cuneate as in angusta, because they are not dilated distally and have a much narrower third posterior cell. The veins are black, but the costa and the first longitudinal are yellow in the two hyaline identations of the fore border; costal bristle long and strong. The wing pattern is like that of angusta, differing chiefly in the presence of two hyaline central spots, and in the indentations of the hind border being more spot-like. The base is entirely hyaline, without the obtuse black band of angusta, or only with a faint indication of it below the humeral cross-vein. The stigma is entirely black and about as long as the black apical part of the second costalcell. The two hyaline indentations of the fore border are broader, more regularly triangular and of equal size; the black band between them is broadly triangular and disposed perpendicu- larly to the costa; the tips of the hyaline indentations are placed exactly on the third vein and are obtuse, but at the upper end of the small cross-vein there is a very narrow short hyaline streak, which is separated from the vertex of the second hyaline indentation. A little before the middle of the first posterior cell there is a rather broad, rounded or shortly oval, hyaline spot as in the following species. The discoidal cell is entirely black, but near its base there is a rounded hyaline dot, which is smaller than the above-named one. The three hyaline indentations of the hind border are very different from those of angusta, being much shorter and not produced into the discoidal cell ; the two indentations of the third posterior cell have the shape of two rounded spots; the indentation of the second posterior cell is twice as long as the other, but is broad and not stripe-like. The axillary lobe and the alula are hyaline. The third posterior cell is much narrower than NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 255 in angusta, while the axillary cell is equally narrow; the hind cross-vein is short, nearly straight and perpendicular, the lower outer angle of the discoidal cell being thus not acute and not produced ; lower angle of the anal cell not acute. Type 3g, a single specimen from British East Africa, Embu, 11.11.1913 (G. H. J. Orde-Browne); type 9, a single specimen from Gold Coast, Aburi, 27.x1.1914 (Dr. J. W. Scott Macfie). Aciura oborinia, Walker (1849). (Pl. xvi, fig. 5). A comparatively large species, like the following one, but at once distinguished by the hyaline basal spot of the third posterior cell. I assume the present species to be that of Walker, because it corresponds in the size and answers very well to the description of the wing pattern; but Walker says decidedly that the bristles at hind border of head are black, while in the present specimens they are whitish yellow, like in the following and in the allied species. ‘In this character the species agrees better with the preceding one (persprcillaris), which has also the hyaline spot of the first posterior cell; but I think that persprcillaris cannot be Walker’s species, beg much smaller and having a very different wing pattern. I will give here a short redescription of what I believe to be Walker’s species.* ¢@. Length of body 45-5 mm. ; of ovipositor 2mm.; of wing 5°5-6 mm. Head reddish brown, with black occiput, black ocellar plate and black vertical plates ; antennae with reddish basal joints and infuscated third jomt; lunula not specially developed. Thorax shining black, more glistening on the sides and pleurae ; the short dorsal pubescence is yellowish; the bristles are black. The scutellum is not well preserved, and seems to have no trace of a. sct. Halteres yellowish. Abdomen narrow and elongate, shining black, with black pubescence and black bristles ; ovipositor as long as the abdomen, shining black. Legs long and rather slender, black, with the tibiae at end and the tarsi yellowish; front femora with three black bristles beneath. Wings with a rather broad and quite hyaline axillary lobe. The pattern is as described by Walker; the black basal stripe extends from the first costal cell to the middle of the second, ending there obliquely and separated from the black stigma. The second hyaline indentation of the fore border extends with its acute point into the base of the first posterior cell, along the outer side of the small cross-vein; the oval whitish spot at the base of the first posterior cell is placed much before the middle of the cell itself, only a little beyond the upper end of the posterior cross-vein. The three indentations of the hind border are narrow and oblique, the basal one entering the base of the discoidal cell and reaching the fourth vein, while the much shorter middle one enters only with a very short point and the third stops just at the upper basal angle of the second posterior cell. The last portions of the third and fourth longitudinal veins are slightly but distinctly convergent towards the end; the two cross-veins are very close together, their distance on the fourth vein being shorter than the length of the small cross-vein ; the hind cross-vein is curved and oblique, the lower exterior angle of the discoidal cell being acute and produced ; third posterior cell as broad as the axillary lobe ; lower angle of the anal cell rather acute but not produced. Costal bristle long. *[The specimens described by the author agree well with Walker’s type.—ED.] °* 256 PROF. M. BEZZI, A couple of specimens from Uganda Protectorate, Entebbe (Forest), 3,800 feet, 5-11.vu.1911 (S. A. Neave). Aciura tetrachaeta, Bezzi (1918). Of this characteristic species, described from N. W. Rhodesia, there is a couple of specimens from Natal, Estcourt, 1x.-x. 1896, and Malvern, 1897 (G. A. K. Marshall). The as yet undescribed male is very like the female; its abdomen is obtuse at the end, and the genitalia are black. The pubescence of the back of the thorax is yellowish, and the bristles are likewise yellowish but more or less infuscated, those of the pleurae being black. The front femora are yellow on the inner side and bear four long bristles beneath. The greyish shade in the middle of the first or basal indentation of the hind border is not distinct in the present specimens, Spheniscomyia Bezzi, 1913. I place in this genus a new species, which is rather aberrant in lacking the a. sct. ; but in size, in the chaetotactic characters of the head, the black ocp., the form of the scutellum and in the faint dust of the thorax, it agrees with the typical species. On account of these characters, it is very different from the Ethiopian species which form the very homogeneous group of Tephrella dealt with by me in the preceding paper. Spheniscomyia sexmaculata, Macquart (1843). Of this common species there is a male specimen from Nyasaland, Mt. Mlanje, 4.vii.1913 (S. A. Neave), and another from Natal, Port Shepstone, v. 1897 (G. A. K. Marshall). Spheniscomyia neavei, sp. nov. (Pl. xviu, fig. 6). A pretty species very distinct from any other in having no scé. and in having two hyaline indentations at the fore border of the wing. It is possible that A. capensis, Rond., which is described with a similar wing pattern, belongs to the same natural group as the present species. § 2. Length of body 2°3-2°5mm.; of ovipositor about 1 mm.; of wing 3-3:1 mm. . Occiput entirely black, but clothed with rather dense, dark grey dust; frons reddish, with whitish orbits and black, but densely grey-dusted ocellar and orbital plates ; lunula whitish, like the linear cheeks; face yellowish, with whitish dust ; jowls very narrow, reddish, not spotted. Antennae entirely reddish yellow, with a shortly pubescent arista. Palpi and proboscis pale yellowish. All the cephalic bristles are black, even those of the occipital row ; oc. well developed ; two strong 2.o0r.; genal bristle well developed. Thorax and scutellum entirely black and rather shining, the greyish dust of the back being very faint and even wanting on the scutellum ; the bristles are black; dc. placed on the line of the a. sa.; one mpl. ; pt. above as developed as the st.; no a. sct., the b. sct. being very long but not diverging. The short pubescence of the back of the mesonotum is black, like the rather sparse hairs of the scutellum. Mesophragma shining black; scutellum broader than long and rather convex. Squamulae white; halteres pale yellowish. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 257 Abdomen shining black, with black pubescence and short black bristles; venter black, but less shining; ovipositor glistening black and with black pubescence. Legs black, with yellow tibiae and tarsi; the femora are narrowly yellowish at tip, and the hind tibiae are infuscated at base; front femora with four long bristles beneath. Wings with long and strong, but simple costal bristle; the veins are black, but they are yellowish at base and on the hyaline indentations. Small cross-vein placed on the apical third of the discoidal cell, which is much broadened outwardly ; hind cross-vein convex outwardly ; last portions of the third and fourth veins about parallel, the former being only a little curved downwards near the end ; lower angle of the anal cell acute, but not produced. The base of the wing is quite hyaline, the dividing line being straight and running obliquely from the distal end of the second costal, second basal and anal cells to the hind border; the stigma is entirely black. At the fore border there are two triangular, hyaline indentations of equal size, just beyond the stigma and reaching with their obtuse point the third longitudinal vein. The wholly black discoidal cell has near its base a rounded hyaline spot, which is above in contact with the fourth vein, while below it is distant from the fifth. At the hind border there are four hyaline identa- tions, disposed in pairs; two shorter and broader into the third posterior cell, reaching the fifth vein, the basal being twice as broad as the apical one; the other two into the second posterior cell, the basal one extending along the hind cross-vein (but without being in contact with it) to the fourth longitudinal vein; the apical one is curved inwardly and ends broadly obtuse at the third vein, crossing the first posterior cell in its last third. The hyaline parts are distinctly whitish ; the dark parts are blackish, only the basal band below the sixth vein in the axillary cell being more greyish. Type ¢ and type 9, a couple of specimens from Nyasaland, Mt. Mlanje, 16.vi. 1913 ; an additional female specimen from the same locality, 31.v.1913, all collected by S. A. Neave, in whose honour this interesting species is named. In the Entomological Museum of the University of Kansas, at Lawrence, Kas., U.S.A., there is a female specimen collected by F. L. Snow at Salisbury, Rhodesia, 5050 ft., v. 1901; it differs in having the basal hyaline spot of the discoidal cell of greater size, extending from the fourth to the fifth longitudinal vein. Tephrella, Bezzi, (1913). Tephrelia sexfissata, Becker (1910). Acura sexfissata, Becker, Ann. Soc. Ent. France, 1910, p. 28. In my previous paper (Bull. Ent. Res. ix, p. 22) I overlooked this species, which is nearly allied to but distinct from 7. rufiventris, Bezzi ; it is recorded from British Kast Airica, Nairobi (de Rothschild). IV. Subfam. TRYPANEINAE. Spathulina, Rondani, 1856. This genus, as defined by me in the preceding paper, seems to have a near ally in the South American genus Lamproxyna, Hendel (1914, p. 64), which shows also a shining black, not dusted abdomen, but seems to be different in the much more elongate and bicubitate proboscis, in the antennae being inserted below the middle 258 PROF. M. BEZZI, of eyes, etc.; moreover the type-species, L. nitedula, Hendel, from Peru, has a different wing pattern, with more numerous hyaline spots, which are disposed in rows along the longitudinal veins. In the new material there is the following new species, which is the first to be known from West Africa. Spathulina bioculata, sp. nov. Closely allied to sematra and running down to it in my table (1918, p. 28), but at once distinguished by the pale yellowish macrochaetae of the head and thorax, and by the presence of four (not three) hyaline spots in the second posterior cell, and of two hyaline dots in the first posterior cell.* ©. Length of body 2°5 mm. ; of ovipositor about 1 mm.; of wing 3°5 mm. Head not depressed, in front view about as high as broad, in side view narrower than high; it is entirely pale yellowish, more whitish on the face, and blackish but grey-dusted in the middle of the occiput ; middle frontal stripe more yellow in the apical half, and there with some scattered yellowish hairs; all the bristles are pale yellowish, those of the hind border quite whitish; there are three 2. or., but the apical one is only half as long as the others; oc. long and strong; ocp. rather acute. Antennae pale yellowish, shorter than the face, with the pubescent third joint as long as the first two joints together and a little pointed at its upper terminal angle ; arista microscopically pubescent. The hairs of the lower occipital border and of the chin are whitish. Palpi whitish ; proboscis yellow, rather thick, as long as the lower border of the head, with the terminal portion thick, bent back- wards and a little shorter than the basal one. Thorax black on the back, but clothed with a dense opaque, grey dust ; a narrow stripe just above the notopleural line is reddish yellow; the pleurae are black, with a darker grey dust, but they are yellowish below the prothoracic stigma; the scutellum is grey on the middle, broadly yellowish at the borders and below; the mesophragma is black, clothed with a dark grey dust, like the pleurae. The short pubescence of the back is pale yellowish like that of the pleurae, which is longer and more sparse; all the bristles are yellowish ; the a. sct. are as long as the 0. sct. and are decussate ; the dc. are placed before the line of the a. sa. Halteres whitish. Abdomen shining black, with black hairs and black bristles ; the first two segments are dark reddish-brown at the sides, and the first three segments have a narrow reddish hind border ; ovipositor depressed, shining black. Legs, including coxae and tarsi, entirely pale yellowish; their hairs are whitish, the spur of the middle tibiae reddish; hind tibiae with a well developed posterior row of bristles. Wings proportionally long and broad ; veins pale yellowish at base and on the hyaline dots, blackish on the dark parts. Third and fourth veins distinctly diverging at the end; hind cross- vein perpendicular, a little curved outwardly and shghtly longer than its distance from the small one. The pattern is as in semiatra, with the following differences : the subapical hyaline spot of the submarginal cell is much broader; there are two small hyaline dots in the first posterior cell, the apical one being twice as broad as that placed over the upper end of the hind cross-vein ; the second posterior cell * These two dots are not shown in Loew’s figure, pl. ii, fig. 12 ; but in the description they are recorded as being sometimes present, it seems therefore that it is a somewhat variable character. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 259 is distinctly more elongate and shows a fourth hyaline spot before its upper apical angle, a little smaller than the three others, which are disposed in a triangle; the basal hyaline spot of the third posterior cell is broadly separated from that of the axillary lobe, which shows two small hyaline dots in the dark space below the sixth vein. Alula whitish hyaline and unspotted. The costal bristle is long and black. Type 9, a single specimen from N. Nigeria, Zungeru, 15.1x.1910 (Dr. J. W. Scott Macfie). The character of the long and decussate a. sct. is common to the present species and semiatra, Loew, and is in contrast with the other species in which these bristles are rudimentary or even wanting ; in Lamproxyna there are likewise only two sct., the apical ones being wanting. This character must also be taken with caution in my table of genera ; even the character of the 2. or. can be misleading, as in the present species there are three pairs of these bristles, while the other species have only two pairs. Pliomelaena, Bezzi, 1918. Pliomelaena brevifrons, Bezzi (1918). A female specimen from Natal, Port Shepstone, v. 1897 (G. A. K. Marshall). Euaresta, Loew, 1873. Euaresta amplifrons, sp. nov. (Pl. xvii, fig. 7). A robust species of comparatively large size, very near planifrons, Loew, but at once distinguishable by the presence of a pair only of scutellar bristles (agreeing in this with the nearly allied megacephala, Loew), by the striking abdominal coloration and by the somewhat different wing pattern. g. Length of body 7 mm. ; breadth of abdomen 2 mm. ; length of wing 7 mm. Head very broad, even a little broader than the thorax, in front view broader than high, in side view about as broad as high ; it is entirely yellowish, but clothed with a dense cinereous dust. Occiput concave, with a double blackish spot in the middle, above and on sides of the neck ; the rather stout hairs of its lower portion are whitish, like those of the chin. Frons very broad and quite bare, about three times as broad as the eye, and even a little broader than long ; it is very flat, being not produced above the eyes, but conspicuously prominent in front above the antennae ; it is entirely of a reddish grey colour, the middle band being not distinct from the orbits ; the small ocellar dot is blackish ; lunula very well developed, broad, semicircular, rather prominent. Cheeks rather broad; face a little shorter than the frons, concave in the middle, rather prominent at the mouth-border, clothed with whitish dust; jowls twice as broad as the cheeks, a little broader than the third antennal joint. Eyes rounded. Antennae inserted a little above the middle of the eyes, very short and much shorter than the face; they are entirely yellowish ; first jomt with pale yellowish hairs; second joint rather prominent above; third joint shorter than the first two joints together, with the upper terminal angle rather acute; arista yellowish, thickened at base, microscopically pubescent. Proboscis thick, short, yellowish ; palpi as long as the proboscis, broad, yellowish, with some bristly, yellowish hairs. All the cephalic bristles (637) D 260 PROF. M. BEZZI. are yellowish, those of the hind border of the head being paler ; two 2. or., which are more developed than the s. or. ; oc. about as long as the vt. Thorax, scutellum, postscutellum and mesophragma entirely black, but clothed with dense cinerous tomentum, which on the back is more clear and shows a little bluish nuance ; the short hairs are pale yellowish, like all the macrochaetae, but those of back are darkened at the base and inserted on small blackish dots; dc. placed well in front, only a little beneath the line of the prst.; pt. as strong as the mpl. ; st. even stronger. Scutellum bare, with the b. sct. only, which are long, divergent, inserted on rather large black spots and placed rather distant from the borders ; there is no trace of the a. sct. Squamulae whitish, with a rather long whitish fringe ; halteres pale yellowish, with a little infuscated knob. Abdomen broad and very obtuse at the end ; it is reddish yellow above on the first three segments, and quite black on the fourth, which is twice as long as the preceding segment ; all the hairs are pale yellowish, like the lateral and apical bristles ; venter entirely reddish, even on the last segment; genitalia rounded, blackish brown. Legs rather short and stout; they are entirely reddish and rather bare, with whitish dust ; front femora with a row of four stout yellowish bristles beneath on the apical half; hind tibiae without a distinct row. Wings with a distinct costal bristle and with yellowish veins; the distal portions of the 2nd, 3rd and 4th veins are straight and slightly divergent ; cross-veins rather close together, their distance being shorter than the length of the straight posterior cross-vein, and the anterior cross-vein being placed only a little before the last fourth of the discoidal cell; lower angle of the anal cell rather acute, but less produced. The pattern is very like that of planifrons, but the dark patch is more reduced, not extending basally beyond the stigma, the basal portion of the submarginal cell being thus hyaline, and the point of the bifurcation of the 2nd and 3rd veins being in the shape of a conspicuous black callosity, which is very strikingly developed, while in planifrons it is comprised in the dark patch. The hyaline spots at the apex and hind border of the wing are of greater size, the radiating pattern being thus more developed ; the discoidal cell is almost completely hyaline on its basal half ; the second basal cell is quite hyaline ; the dark spot at the lower end of the anal cell is less developed, and there is no dark spot at all below the end of the 6th vein. Type 3, a single, rather damaged specimen from Natal, Malvern, v. 1897 (G. A. K. Marshall). The wing pattern of the present species is not unlike that of some species of the genus T'rypanea, but is also very like that of Huaresta conjuncta, Loew. It is very probable that the present species may be galligenous, like H. megacephala, Loew, which in Sicily makes very conspicuous galls on the twigs of the Composite plant, Inula crithmoides;* it is therefore possible that EH. megacephala, E. planifrons and E. amplifrons will belong to a natural genus, which is biologically characterized by the faculty of making so-called pleurocecidia on Composite plants, a character rather uncommon in the present group of flower-head flies. * T, De Stefani Perez, Marcellia, Avellino, iii, 1904, pp. 122-125; Trotter e Cecconi, Cecidotheca italica, Fasc. xiv, no. 326 (1906); C. Houard, Les Zoocécidies des Plantes ad’ Europe, etc. ti, 1909, p. 972, no. 5627. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 261 Ensina, Rob.-Desv., 1830. Ensina sororcula, Wiedemann (1830). Of this common species there is a male specimen from Natal, Ulundi, 5,000-6,500 ft., ix. 1896 (G. A. K. Marshall). Ensina giadiatrix, sp. nov. Closely allied to the preceding species, but distinguished by the infuscated antennae, the black femora which have only a very narrow yellowish tip, and by the exceedingly long ovipositor of the female. ©. Length of body 3 mm.; of the basal segment of the ovipositor 1°3 mm., and of the completely exserted ovipositor 3 mm. ; of the wing 3 mm. Head as in sororcula, but less depressed and with the frons distinctly shorter ; ocellar spot blackish ; third antennal joint distinctly infuscated. The thorax seems to be darker and devoid of distinct pattern on the back; the pleurae are more decidedly blackish ; there is no trace of a. sct. The abdomen seems likewise to be darker above and devoid of distinct pattern ; the ovipositor is shining black, depressed, with the basal segment a little longer than the whole abdomen (in sororcula itis about a half as long as the abdomen), the second segment shining black, the third segment reddish ; if completely exserted, as in the case of the type, the ovipositor is longer than the entire body. In the legs all the femora are entirely black, having only the extreme tip narrowly yellowish, while in sororcula they have the apical half yellowish, and the four anterior ones are mainly or even entirely yellow. The wings are proportionally broader and have the same venation ; the pattern is very similar, but the stigma is quite black, even at the base ; the fuscous reticulation is more developed on the posterior half of the wing, chiefly in the discoidal and the second posterior cells ; in the first posterior cell the hyaline spots are disposed in two distinct longitudinal rows, a condition which is never to be observed in sororcula, in which they are not very distinct, never ordinate, but are in one row only at least towards the middle of the cell. Type 9, a single damaged specimen from Natal, Ulundi, 5,000-6,500 ft., ix. 1896 (G. A. K. Marshall). Ensina magnipalpis, sp. nov. Closely allied to dubia, Walker and Loew, but distinguished by the hyaline spot of the stigma, and by the much less numerous hyaline spots of the first basal and discoidal cells. ¢. Length of body 38-4 mm. ; of wing 4-4:2 mm. Occiput black, grey-dusted above, yellowish below. Frons flat, not prominent in profile, broad, parallel-sided, about 1°5 times as long as broad ; it is of a dark orange-yellow colour, but a narrow longitudinal middle stripe and the orbits are whitish ; lunula broad and rounded, yellowish ; cheeks narrow, with white shining dust; jowls yellowish, unspotted, as broad as the third antennal joint. Face narrower and shorter than the frons, with broad and very prominent mouth-border. Eyes in profile rounded; antennae inserted a little below their middle; they are entirely yellowish and a little shorter than the face; second joint with yellow (637) p2 262 PROF. M. BEZZI. hairs ; third joint broad, rounded at tip, about twice the second in length ; arista microscopically pubescent. Palpi pale yellowish, with pale hairs; they are very long and broad, flattened, as long as the basal part of the proboscis; the latter is very long, about as long as the head and thorax together ; its basal part is pro- duced much beyond the mouth-border, and the apical part is acute and much produced behind the chin. Bristles abraded in the type, but they seem to be whitish on the occipital border and black on the frons; pubescence pale yellowish. Thorax clothed on the back with a dense opaque whitish-grey dust, but with a very broad even longitudinal blackish stripe, which extends laterally to the dc.; a smaller lateral blackish stripe on each side above the notopleural line; the back can therefore be described as black with two lateral whitish stripes; the pleurae are entirely greyish and unspotted. The bristles of the back are black; the pubescence is pale yellowish ; on the pleurae the st. and the mpl. are black, while the others are pale yellowish or whitish, like the hairs. Scutellum blackish, greyish at the end and below, with four equally long and strong black bristles, the apical ones diverging. Postscutellum and mesophragma coloured like the pleurae. Squamulae and halteres dirty whitish, the former with a white fringe. Abdomen broad and flat, grey, with two longitudinal rows of blackish spots, which are partly confluent to form two longitudinal stripes; last segment with a broad yellowish hind border ; pubescence pale yellowish ; apical bristles black ; genitalia yellowish ; venter yellowish grey. Legs entirely yellowish, the femora more or less greyish on the outer side; hairs and bristles pale yellowish or whitish, only the apical spur of the middle tibiae being black. Wings with a well-developed black costal bristle ; costa black on the dark, and yellowish on the hyaline parts ; veins black, in part yellowish at base and on the hyaline spots; stigma longer than broad, deep black, with a small circular, sharply defined, whitish hyaline spot before the end. First vein ending before the small cross-vein; 2nd vein straight; 3rd vein bare, straight throughout and parallel with the last portion of the 4th; smalt cross-vein placed considerably beyond the middle of the discoidal cell; hind cross-vein perpendicular, convex outwardly, distinctly longer than its distance from the anterior one; discoidal cell long; anal cell broader than the second basal cell, with an acute but not produced lower apical angle; 6th vein complete but colourless at the end. The pattern is blackish, more intense on the fore half, paler behind ; in general shape it is like that figured by Loew for dulia (pl. u, fig. 20) with the following differences, besides the hyaline spot of the stigma; the hyaline spots in the costal cell are broader ; in the base of the submarginal, 1st and 2nd basal and discoidal cells there are only very few, much broader hyaline spots; the 3rd posterior cell has the reticulation closer, being fuscous with about 8 hyaline spots. The axillary lobe is likewise hyaline, with two fuscous rounded spots at the hind border ; the anal cell is also hyaline. Type 3; and an additional specimen of the same sex from Durban, Umbilo, 28.v1.1914 (L. Bevis). These two species, dubia and magnipalyis, which agree in palpal and other characters, and have a similar wing pattern with much extended black parts, require the formation of a new genus, if they cannot be placed in the genera Spathulina or Pliomelaena. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 263 Ensina siphonina, Bezzi (1918). There is an additional specimen of this very distinct species, likewise from British East Africa, Nakuru, i. 1913 (Dr. B. L. van Someren). Trypanea, Schrank, 1795. Of this genus, numerous species of which are tabulated in my previous paper, there are in the new material before me two forms which are both very different in wing pattern from all the other Hthiopian species before known, and are here described as new. Trypanea sudcompleta, sp. nov. (PI. xvii, fig. 8.). Distinguishable by the four strong scutellar bristles and by the wings, which have the typical star-shaped subapical pattern, but are marked to the base with a faded but well developed reticulation, like that of the American species abstersa, Loew, and reticulata, Hendel. Q. Length of body 3°7 mm. ; of ovipositor 0°5 mm.; of wing 4mm. Head of proportionally smaller size and rather narrow in profile, with the purplish eyes higher than broad ; it is greyish, with a dark spot on the middle of the occiput ; frons about as broad as long, not at all prominent in front, with whitish dust in the middle; lunula rather broad; face short, not prominent below; cheeks linear and jowls only a little broader. Antennae inserted at middle of the eyes, only a little shorter than the face, entirely yellow, with the third joint rounded at end ; arista bare. Palpi and proboscis yellowish, the latter short and retracted into the oral cavity. Cephalic bristles dark yellowish, but those of the hind border whitish ; oc. long and strong; three 7. or., but the anterior one less than half as long as the others. Thorax blackish on the back, yellowish on the sides, but densely clothed with grey dust, opaque and unicolorous; the short hairs are pale yellowish, the bristles dark yellowish, those of the back being inserted over small black dots ; pt. as strong as the st.; hairs of the hind border of mesopleura rather long. Scutellum like the thorax, but broadly yellowish at the hind border; Db. sct. long, diverging, inserted over rather broad black spots ; a. sct. a little shorter, decussate, on smaller black spots. Postscutellum and mesophragma entirely grey. Squamulae whitish, halteres pale yellowish. Abdomen like the thorax, with short pale yellowish hairs and yellowish terminal bristles ; the first three segments have a rather broad yellowish hind border; the 4th and 5th are blacker basally and more yellow apically ; ovipositor depressed, shining black, short and obtuse, a little shorter than the last two abdominal segments together. Venter pale yellowish, with whitish hairs. Legs short and stout, entirely pale yellowish ; front femora above with a complete row of short yellowish bristles, below with an apical row of four very long yellowish bristles ; spur of the middle tibiae blackish ; hind tibiae with a well developed row towards the middle. Wings long and narrow, with a distinct costal bristle and pale yellowish veins, which are infuscated on the darker parts of the pattern ; first vein ending much before the small cross-vein ; 2nd, 3rd and 4th veins straight in their terminal portions and equally and gradually divergent ; cross-veins perpendicular and parallel, the hind one being longer than its distance from the anterior one; lower angle of the anal cell acute but less produced ; 6th 264 PROF. M. BEZZI. vein prolonged to the hind border. The star-shaped apical patch is blackish brown; it encloses a triangular indentation into the marginal cell, two hyaline spots near the border at end of the submarginal cell, one just below the end of the 2nd vein and the other broader and subquadrate before the end of the 3rd vein. The first posterior cell has two hyaline spots at base, the first as long as the breadth of the cell itself and in contact with the small cross-vein, the second rounded, half as long, and in contact with the 4th longitudinal vein ; the cell has at the end the three usual spots, which form the two usual radiating branches ; two other fuscous parallel rays are in the second posterior cell, and another less distinct ray on the hind cross-vein. The fuscous rays at the apex and hind border are therefore five in number. The basal portion of the wing shows a pale but complete reticulation, formed by rounded hyaline spots of rather large size; of these spots there is a row only in the base of the submarginal and first basal cells; the discoidal cell has a row in the basal, and two rows in the apical half; the third posterior cell has ten broader spots disposed in two rows, and along the hind border three more spots ; axillary cell with six spots in two rows. The base of the second posterior cell has three spots, two larger and one small, between the ray of the hind cross-vein and the middle ray. The short stigma is quite colourless, with only a black dot near the costa at base of the costal bristle ; it is not united with the subapical patch ; the oblique ray departing from it and extending to the small cross-vein is very faint and belongs to the pale reticulation. Type 9, a single specimen from British East Africa, Nairobi, 27.1v.1911 (7. J. Anderson). From 7. abstersa (which I have in my collection from California, New Mexico and Colorado) the present species differs in the closer reticulation of the wings, in the wholly colourless stigma, and in the want of the hyaline spot in the marginal cell, just above the end of the second longitudinal vein (not to be confounded with the hyaline spot in the submarginal cell just below the end of the same vein). From 7’. reticulata it differs in the subapical brown patch being much more extended and more typically shaped. Trypanea peregrina, Adams (1905). Having had occasion to see the type of this species in the Entomological Museum of the University of Kansas, Lawrence, I must recognize that it is the same as my T. urophora of the preceding paper, as already suspected. The figure 11 on pl. 1, is too dark, the ray extending from the stigma to the small cross-vein being really much paler and interrupted. The form of the ovipositor is the same, even though it is depressed in the type of peregrina. Trypanea hemimelaena, sp. nov. (PI. xviu, fig. 9). A pretty species recalling 7. guimari, Becker, from the Canary Islands, and hkewise with the apical half of the wing only occupied by a not typically star- shaped pattern, which shows however more numerous hyaline spots. S. Length of body 3°2 mm.; of wing 3°2 mm. Head entirely pale yellowish, clothed with whitish dust, and only narrowly ‘blackened towards the middle of the occiput, above the neck ; it is rather large, NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 265 being in front view broader than high, and in side view higher than broad ; eyes greenish, unicolorous, higher than broad. Frons flat, a little prominent in front above the base of the antennae, as long as the face, with a broad bare yellow middle stripe, and with a whitish-grey ocellar triangle and lateral orbits; lunula broad, semicircular, whitish grey. Face flat, with shallow antennal grooves and the upper mouth-border not prominent, cheeks linear ; jowls only about twice as broad as the cheeks, unspotted. Antennae inserted above the middle of the eyes, much shorter than the face ; they are pale yellowish ; third joint pubescent, with the upper terminal angle a little infuscated and not acute ; arista microscopically pubescent, yellowish, thickened at base. Palpi and proboscis yellowish, the latter thick and short, retracted into the broad oral cavity. All the cephalic bristles are pale yellowish, those of the hind border whitish; oc. long and strong; two v. or.; genal bristle distinct, but short and whitish ; the rather short hairs of the lower part of the occiput and of the chin are whitish. Thorax black in ground- colour but clothed with a dense grey dust, which on the back is more clear and a little yellowish, and darker on the pleurae; there is no distinct pattern on the back, except an ill-defined dark stripe below the notopleural line; all the hairs and bristles are whitish, the latter being inserted over very small but striking black dots; dc. placed near the suture; pt., mpl. and st. well developed, but . intermingled with rather long and stout hairs. Scutellum coloured like the back of mesonotum and likewise pubescent ; the two basal bristles are inserted over proportionally broad dark spots; the apical bristles are medium-sized and decussate at the end. Postscutellum and mesophragma clothed with a dense grey (not yellowish) dust. Squamulae and halteres pale yellowish. Abdomen elongate and rather attenuate at the end; it is densely tomentose like the back © of the mesonotum (in guimarz it is less tomentose, the black ground-colour being visible) and clothed with rather long pale yellowish hairs; hind border of the first two segments distinctly yellowish; venter blackish, with short scattered whitish hairs; genitalia black. Legs rather short and stout, entirely reddish, the front coxae grey-dusted ; they are clothed with a short whitish pubescence ; front femora with five stout yellowish bristles beneath; hind tibiae with no distinct row. Wings proportionally short and broad, with a distinct costal bristle and pale yellowish veins, which are blackened on the apical half; 3rd and 4th veins straight and a little divergent at the end, while the 2nd is more strongly divergent from the 3rd; cross-veins straight and parallel, the small one placed near the last fourth of the discoidal cell, the posterior one longer than its distance from the small one ; lower corner of the anal cell forming a wide angle; 6th vein prolonged to the hind border, even if very faint. The stigma is very short and pale yellowish, with a small black dot near the base of the costal bristle ; the first vein ends much before the small cross-vein (in guemari it ends below or only a little before it). The basal half of the wing is entirely whitish hyaline and quite unspotted. The apical part is occupied by a dark brown patch, which is less dark towards the hind border, and has a yellowish patch on the small cross-vein into. the end of the first basal cell, prolonged above into the submarginal cell towards the stigma. In the brown patch there are the following hyaline spots ; two spots close together in the marginal cell just beyond the stigma, the first very broad * 266 PROF. M. BEZZI. and subquadrate, the second much narrower and subtriangular; in the sub- marginal cell there are three rounded spots, one in the middle just above the upper end of the small cross-vein and at the inner end of the upper part of the yellowish patch, the other at the fore border just below the tip of the 2nd vein, the third of greater size and subtriangular just before the lower apical angle. The first posterior cell has a rounded spot near the base, half as broad as the distance between the 3rd and 4th veins and in contact with this last vein before the upper end of the hind cross-vein ; two other spots before the end, one over the other, that in contact with the 3rd vein being smaller than that in contact with the 4th; and one apical, extending from the 3rd to the 4th vein, and thus forming the normal apical dark fork of the Trypanea-pattern, though a little shorter than usual. Discoidal cell with four about equally small, rounded spots, disposed as a square. Second posterior cell with three larger rounded spots along the hind border, of an indentation-like shape, and one of equal size near its upper inner angle. Third posterior cell with a single rounded spot of greater size before its lower apical angle and in contact with the hind border of the wing. Type 3, a single specimen from the Gold Coast, Aburi, 1912-13 (W. H. Patterson). Owing to the great affinity of the present species with 7. guimari, which shows a very remarkable sexual dimorphism, it is possible that the unknown female has a somewhat different wing pattern. These two species, like 7. peregrina, belong evidently to the gnaphalu-mamulae group. V. Subfam. ScHISTOPTERINAE. This peculiar subfamily is represented in the new material before me by two remarkable forms, one of which requires the formation of a new genus, while the other, being closely allied to Rhabdochaeta spinosa, Lamb (1914), from the Seychelles, may be provisionally placed in the Oriental genus Rhabdochaeta, de Mei}. An important character of the subfamily is that the s. or. are placed on a line converging towards the centre of the frons, being thus not in the same line with the 7. or. Perirhithrum, gen. nov. This new genus is more closely allied to the Oriental Rhabdochaeta than to the Ethiopian genera Schistopterum or Rhochmopterum, but differs from all the known genera of the subfamily in having the first and third longitudinal veins bristly above, in having a produced lower angle in the anal cell and in having a very oblique hind cross-vein. From Rhabdochaeta the new genus differs in the want of the middle pair of bristles on the frontal stripe. These bristles are likewise wanting in Schistopterum and are not mentioned for Rhochmopterum ; they are recorded for the type species of the genus Khabdochaeta, as well as for the species spinosa, Lamb, and are present also in my species baker: from the Philippines. This pair of bristles seems to correspond to the crossed frontal pair of the ANTHoMyIIDAE: but in Rhabdochaeta they are not crossed. It is interesting to note the presence of a similar pair of bristles in the male of Euribia perpallida, Bezzi (as pointed out by me in vol. ix, p. 35, note), the wing-pattern of which rather recalls that of the ScHISTOPTERINAE. NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 267 Body robust, of comparatively large size, with strong bristles of different coloration, but not of different kind. Head in front view broader than high, in side view rather narrow. Eyes bare, of large size, distinctly higher than broad. Occiput rather flat and not swollen below. Frons about as long as broad, with nearly parallel sides, not at all prominent in front; middle band devoid of central bristles; lunula rather small and deeply excavated. Face shorter than the frons, concave in the middle and with a very prominent border at the upper mouth-edge and at the sides of the mouth; cheeks linear; jowls very narrow. Antennae inserted at middle of the eyes and as long as the face; first joint very short ; second joint prominent above and there clothed with short obtuse stout bristles ; 3rd joint about as long as the first two joints together, of a rectangular shape, with the upper terminal angle acute but not very prominent ; arista longer than the antennae, with rather long pubescence. Mouth-opening broadened anteriorly ; palpi long, flattened, bristly, prominent and of the usual shape for the subfamily ; proboscis short, with short and broad terminal flaps. Occipital row formed by stout obtuse whitish bristles; inner vf. very long; oc. long and strong, widely distant from each other at the base; two s. or., not in the same line with the three 2. or., but converging in front and thus the anterior pair about in front of the oc.; middle frontal band bare; genal bristle strong; lower portion of the occiput with scattered stout hairs. Thorax about as long as broad, moderately convex above, with sparse scattered pubescence, even on the pleurae. Chaetotaxy complete, but with no distinct scp.; dc. placed near the suture and a little before the line of the a. sa.; one mpl. ; pt. a little weaker than the st. Scutellum broadly triangular, gently convex above, clothed with sparse hairs ; 0. sct. long and diverging ; a. sct. well developed, but broken in the type. Abdomen elongate, narrower than the thorax, nearly bare above and on the sides, with some rather stout bristles at the end; male genitalia globose and rather prominent downwards. Legs rather slender, the femora not incrassated ; front femora with a complete row of strong bristles below ; middle tibiae with a single spur; hind tibiae with no distinct row on the outer side. Wings rounded and very broad ; costal bristle well developed, with a second much smaller one before it; there is also a group of 4-5 short bristly hairs at the costa near the humeral cross-vein and its incision. The incision at the end of the auxiliary vein is present, but much shorter and less prominent than typically. Auxiliary vein well developed, but ending at a straight angle and there evanescent ; first vein ending about at middle of the fore border and distant from the costa (the costal and subcostal cells being thus very broad), and suddenly bent forwards at tip; it is clothed with bristles, which on its terminal part are very long. Stigma distinct, about as broad as long. Second vein long, straight, a little bent forwards at the apex, ending nearer to the end of the 3rd than to that of the first vein. Third vein quite straight, clothed with short bristles to the middle of its last portion; the submarginal cell is greatly widened at the end, because the 2nd vein is strongly divergent from the 3rd, which is parallel with the longitudinal axis of the wing. Fourth vein likewise straight and in its last portion quite parallel with the last portion of the third. Small cross-vein normally developed, straight, perpendicular, about as long as its distance from the upper end of the hind cross-vein ; it is placed near the terminal third of the upper border 268 PROF. M. BEZZI. of the discoidal cell, but near the middle of the lower border of the same cell. Cubital fold well developed and forked outwardly, evenly chitinised. Hind cross- vein broadly S-shaped, but placed very obliquely so that its upper end is near the middle of the wing, while its lower end is very near the end of the fifth longitudinal vein ; the lower external angle of the discoidal cell is therefore very acute and much produced outwardly. Anal cell broad, with its lower outer angle acute and produced longer than the 2nd basal cell ; sixth vein prolonged to the hind border of the wing. Axillary lobe broad and rounded, with a distinct long axillary vein; alula well developed, rounded. The wing pattern is a very characteristic one, consisting of a broad, light brown patch, occupying the middle and extending over the whole hind portion of the wing, except the apex; this patch shows black spots, shining “bullae ’’ and subhyaline spots near the hind border; from this patch depart numerous long rays, which extend to the costa along the entire fore border and the entire apex. Type: the following new species. Perirhithrum marshalli, sp. nov. (PI. xvii, fig. 10). A strange and beautiful insect of comparatively large size, strikingly distinguished by the very peculiar wings. | g. Length of body 5°5 mm. ; of wing 55 mm. ; breadth of wing 3 mm. Head yellowish, darkened towards the middle of the occiput. Frons with a dark ocellar spot and with a transverse dark band in front before the lunula. Antennae reddish yellow, with the third joint blackened along its upper border; arista whitish, cheeks white, with a small deep black, rounded spot above on each side, near the roots of the antennae ; jowls yellowish, with a reddish spot on each side in front; face yellowish, darkened towards the middle. Eyes reddish brown and apparently unicolorous. Palpi whitish at base, reddish yellow on the apical half, and clothed there with short black bristles; proboscis reddish yellow. The ocp., the outer vt. and the pvt. are whitish; the long inner vf. are black, while the oc. are pale yellowish; the long and strong s. or. are black with a brownish end and are inserted over distinct, black, slightly prominent tubercles, like the first (or inner) of the 2. or., which is likewise black with a brownish tip; the two apical pairs of the 7. or. are on the contrary-shorter and whitish. The genal bristle is black, while the rather long, bristly hairs near it on the chin, and those on the lower part of the occiput, are whitish. Thorax on the back black in ground-colour, but covered with dark grey dust; the suture is broadly interrupted in the middle, with whitish dust on the sides and has inwardly a brownish transverse band, which is interrupted towards the middle; the points of insertion of the bristles are darker and spot-like, and there are in the centre two dark brownish rounded spots, one on each side, interiorly and posteriorly of the anterior dc.; the back is denuded in the type, but in front of the scutellum it shows rather long and scattered whitish hairs. The pleurae are yellowish near the sutures and towards the breast, but are darkened towards the centre of the sclerites; they are clothed with a whitish-grey dust. All the macrochaetae are pale yellowish, but those of the back are distinctly black at base, though less so in the case of the mpl. and st., while the pt. is entirely whitish. . Scutellum coloured like the back of the mesonotum, but less obviously — NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 269 yellowish on the sides and below; the b. sct. are very long, pale yellowish, with a black base, and are inserted on a small black dot; the a. sct. are wanting in the type. Postscutellum and mesophragma blackish, grey-dusted. MHalteres with yellowish stalk and blackish knob. Abdomen reddish, with pale grey dust; each segment bears on each side a very broad, brown spot, which is rectangular in shape on the three basal segments, while it is subquadrate on the fourth; it could be said that each segment is brown, with a narrow whitish hind border and a whitish middle line; each segment bears moreover on the sides, in the whitish portion, a smaller dark spot, and the 4th segment is broadly yellow at the hind border. The entire abdomen seems to be bare, with only a few scattered short black hairs on the sides of the segments, chiefly of the last one, which bears moreover 8-10 not long but strong, black bristles at the hind border and on the sides. Genitalia reddish yellow; venter entirely reddish, with whitish dust and sparse whitish hairs. Legs entirely pale yellowish, even to the end of the tarsi, but each femur is adorned below with two rounded dark spots, one near the base, the other towards the middle; the hairs and bristles of the front femora are pale yellowish, while the posterior femora have blackish pubescence and some dark bristles at the end above; middle tibiae with a dark spur. The wing veins are yellow but darkened on the darker parts of the pattern, and more strikingly on the costa, which is alternatively black and yellow. The short costal bristles near the humeral cross-vein are yellowish, while the two costal bristles at the stigma are black, like the bristles on the Ist and 3rd longitudinal veins. The pattern is brownish, more blackish at the end of the rays of the fore border and on several of the central spots ; the membrane is iridescent, quite hyaline on the fore border and at the apex, while the hyaline spots of the hind border are more whitish. The radiating part shows the following dark rays, which become progressively longer from the base to the apex of the wing; one at the extreme base of the first costal cell; one like a narrow border of the humeral cross-vein ; two in the second costal cell, one broader near the base, and one narrower but ' darker beyond the middle; one narrow and greyish along the diffuse terminal part of the auxiliary vein; three blackish, close together and very marked on the stigma, the first and broadest of all at the base, the second in the middle, the third and narrowest along the terminal part of the first vein. In contact with the last of these rays there is a broader triangular greyish ray departing from the fuscous base of the marginal cell and showing some small, subhyaline, whitish streaks on its base. This ray is followed by two other parallel rays at equal distances in the marginal cell itself; these two rays are narrower and bent a little outwardly, and have their central stripes whitish hyaline. There are subsequently two dark rays in the submarginal cell, narrow and full, one ending at the end of the second vein, the other ending at middle of the distance between the tips of the 2nd and 3rd veins; the Ist posterior cell has likewise two narrow rays which follow exactly the two external branches of the cubital fork, ending at the tips of the 3rd and 4th veins ; the 2nd posterior cell has the last three rays, which are less sharply defined, the last one being even hardly distinguishable from the dark patch; they are more irregular and sinuous and enclose two small, hyaline spots and a hyaline streak at the base. On the disc of 270 PROF. M. BEZZI. the wing there are: (1) the blackish patches, two of which more striking and more defined are of subquadrate shape and placed one in the submarginal cell before the small cross-vein, and the other near the base of the first posterior cell, just above the upper end of the hind cross-vein ; (2) the shining “ bullae,” the more distinct of them being placed around the small cross-vein at unequal distances to form a kind of circle: one in the first basal, one in the submarginal just above the small cross-vein, two in the first posterior above and in front of the black spot, and one in the discoidal cell, at some distance below the above-named black spot. There are moreover some other “ bullae,” one at the extreme base of the submarginal cell, and one at the extreme base of the discoidal cell extending into the end of the 2nd basal cell. The 3rd posterior cell and the axillary lobe are entirely and evenly infuscated and show some hyaline spots forming a kind of reticulation ; 9 of these spots are placed into the axillary cell, 4 above and 5 below the axillary vein, and 7-8 are placed in the 3rd posterior cell, those near the apex being more stripe-like. The hyaline parts of the stigma are colourless, like the corresponding ones of the 2nd costal or marginal cells. The basal and anal cells are infuscated to the base, only the extreme base of wing and the alula being greyish hyaline. Type J, a single partly damaged specimen from Natal, Port Shepstone, v. 1897 : another of the interesting discoveries of Dr. G. A. K. Marshall, in whose honour the species is named. Rhabdochaeta neavei, sp. nov. About the same as R. spinosa, Lamb, from the Seychelles, but seeming to differ in the wing pattern, which is yellowish brown, not black, and shows therefore a very remarkable deep black spot at the base of the first posterior cell, just beyond the small cross-vein. ®. Length of body 2:2 mm. ; of ovipositor 0°8 mm. ; of wing 3 mm. Head yellowish, with the occiput darkened towards the middle; frons yellow on the apical half, with whitish orbits and a blackish ocellar dot ; cheeks whitish, with a small, rounded, rather prominent, deep black dot above, between the root of the antenna and the eye; face whitish, with a brown transverse band above the mouth-border; jowls very narrow and unspotted. Antennae pale yellowish at base, with the 3rd joint more yellow, very attenuated and very acute at the end ; arista whitish, shortly pubescent. Palpi and proboscis dirty yellowish. Middle frontal stripe bare, but with a pair of white bristles in the middle. Cephalic bristles pale yellowish, the or. and the inner vt. darkish at base ; those of the hind border are whitish ; two s. or. and three 7. or., counting that placed far in front, just above the black dot of the cheeks ; the stout short hairs of the lower part of the occiput are whitish. Thorax black in ground-colour, but densely clothed with grey dust, which is darker on the pleurae; humeri and notopleural region yellowish ; on the back there are distinct black spots on the points of insertion of the macrochaetae, those of the prsct. being the most developed ; the short hairs are pale yellowish ; the dorsal macrochaetae are distinctly blackened at the extreme base, but are elsewhere yellowish; the pleural macrochaetae are NOTES ON THE ETHIOPIAN FRUIT FLIES, OTHER THAN DACUS. 271 entirely pale. Scutellum like the back of the mesonotum, but broadly yellowish at the sides and behind ; the 0. sct. are blackish at base and inserted over black spots; the a. sct. are much smaller, decussate at the end, not darkened at the base and not inserted on black spots. Postscutellum and mesophragma black but clothed with grey dust, which is darker on the former. Squamulae and halteres whitish. The abdomen seems to be differently coloured from that of spinosa, being entirely reddish, with the base of the first two segments narrowly blackish ; hairs and bristles pale yellowish; ovipositor broad, flattened, with the basal segment shining reddish with black end. Legs pale yellowish, with dark pubescence and whitish bristles ; front femora beneath before the end with a blackish spot, forming an incomplete ring, and there with blackish bristles ; the four posterior femora have this spot more developed, and have moreover a similar but more elongate spot near the base ; all the tibiae have a narrow black ring near the base, less developed on the front pair; hind tibiae with a short but distinct posterior row of bristles. The wings are shaped as in Lamb’s figure 13, p.321; they have the same venation, with a deeply sinuous fourth longitudinal vein and a nearly straight hind cross-vein ; in the Oriental species pulchella and venusta the fourth vein is a almost straight, but the hind cross-vein is likewise about straight ; in bakert the fourth vein is straight, while the hind cross-vein is broadly S-shaped and oblique. The veins are bare, as stated by Lamb; but on the third vein I can perceive some scattered, almost microscopic, black bristles. The general shape of the pattern is the same; it is however not black, but yellowish brown ; the shining “ bullae’ are of the same number and have the same disposition ; but the wing pattern being more clear, the black spot at the base of the first posterior cell is very striking ; it is placed below the basal “ bulla” of the same cell, and bears at base a small rounded whitish-hyaline dot, corresponding to that figured by Lamb at the inner basal angle of the first posterior cell. Other differences are : (1) the marginal cell has no hyaline spots in the fuscous part at the base of the radiating streaks ; (2) the hyaline dots in the submarginal, first basal and first posterior cells are less numerous and much smaller. It must be remarked that in the present species, as well as in spunosa, the costal nick, even if distinct, is much less developed than in pulchella and bakeri ; de Meiere’s figure of venusta shows it however not developed. Type 9, a single specimen from Nyasaland, 8.W. of Lake Chilwa, 13.1.1914, collected by S. A. Neave, in whose honour the species is named. Two species of Huribia, perpallida and discipulchra,* both likewise from Nyasaland, show a very remarkable affinity with the above-described species of Rhabdochaeta, and are perhaps allied; but they have different wing pattern, unicolorous legs, ete. ~ Bull, Ent. Res. ix, 19138, pp. 35, 36. ij ie Oe aa ean "eee hes i sobre Ne tors, j Whe aol ey. (a elieuea : ian vt Or eave aed y Peek" Ra %,: srg wn bit a jal re tins ‘a dws a y oa oh } ae biwrigy Ae we Re | gate ts naphi nN f say aeties ou ae i es ule ants mea heey wae «ys oa ee Tf! MY : ; a Panam, | tah rte far a So ae wis! “Dae: aa ina sale diy “s = t TAP IO VOW ” uhs Ohh hee Bi Re aed Y So! pave \i \ ‘oh, wh { P et TS ie 0 yi Ke raat Lat Par (ae ar, Fk, a avon ste eens pera INE NPN ALY Yow ho} A * 4 > c ast : ¥ = 7 mn bag bi , MOM 72 ISSOs oN rTM Dy AR voy r unt Wa . rit ; yon ay Cae eit i Pt aN LN y h ee ae silk f al ; 7 Meret , RS re a i oh q = se Meas 8) eR tMIS RONG eh Sp» 48 : Ae had . FID a + i rege . a fi . Se oe < . 5 i tore ‘ t . . i —s, j + 40) ‘y i Vy J i : \ | ower ay ra ic ; -_* — ’ . ¢ ‘ n ‘ %,, ie ‘ f S ht Fig. coc on DMD Oo FP OCF DD FE EXPLANATION OF PLATE XVIL. Wings of Trypetidae. . Sosiopsila trisetosa, Bezzi, sp. nov. . Conradtina acrodiauges, Speiser. . Leucotaenella guttupennis, Bezzi, sp. nov. . Celidodacus obnubilus, Karsch. . Bistrispinaria fortis, Speiser. . Clinotaenia anastrephina, Bezzi, sp. nov. . Hoplolopha cristata, Bezz, sp. nov. . Rhacochlaena pulchella, Bezzi, sp. nov. . Taomyia marshalli, Bezzi,.sp. nov. 10. Notomma bioculatum, Bezzi, sp. nov. PLatTeE XVII. WO, Ihe IPsares BuLtit. ENT. RESEARCH. ArPRICAN BRUIT FEIES. al i re t } ‘ hy aa X hee | i inae Ibe y - ‘ j aul ‘ “ \ ‘ » v ie ; ~ vt she er % £5) 2 ‘Ff ! . vi a x 4 a iy if i « ¢ ba A wiv aT 4 Mort VA. OK . 7 apbseye? Jo: agri You pe an pent: Rey ati Coa ae, oar \ ie 9 a od + 7 insell eaatyosts gage ACOA) 8 Bel eiih is wis’ wt * aed 48 ys writ 1\ “eR 7 cit i WJ “von qa Ai Eh ee ne a Vetere 8 wot We diosall: edi vou hea geno SDN: A. Jobin seh ee SR aon a e tas Neat say ey eo iat tdiha ¢ Nod je 7 - t «oi Palle’ 4 1 oo eet i Fi i EXPLANATION OF PLATE XVIII. Wings of Trypetidae. Fig. 1. Acidia fossataeformis, Bezzi, sp. nov. . Ocneros bigemmatus, Bezzi, sp. nov. . Allotrypes brevicornis, Bezzi, sp. nov. . Aciura perpicillaris, Bezzi, sp. nov. . Aciura oborinia, Walker. . Spheniscomyia neavet, Bezzi, sp. nov. . Euaresta amplifrons, Bezzi, sp. nov. . Lrypanea subcompleta, Bezzi, sp. nov. oO ao NT DO FS O&O DH & . Trypanea hemimelaena, Bezzi, sp. nov. 10. Perirhithrum marshalli, Bezzi, sp. nov. SUL, JBN, IRiIsSisnecs Wot, 24 JP Bi une DW NERUCAIN IEIROIOE JRL Se a I ‘ ‘ AL ak te i i B shen 273 SOME INJURIOUS SOUTH AFRICAN WEEVILS. By Guy A. K. Marswattz, D.Sc. (Pirate XIX.) The following undescribed species of CURCULIONIDAE were all forwarded to the Imperial Bureau of Entomology from the Division of Entomology, Pretoria, with the information that they were attacking various cultivated plants. Subfamily BRACHYDERINAE. Prototsrophus* planatus, sp. nov. (Plate xix, fig. 4). g. Unusually flattened dorso-ventrally. Colour black, with uniform and fairly dense brownish-grey scaling above and grey scaling beneath. Head separated from the rostrum by a slightly curved furrow which does not nearly meet the lateral margins; the forehead flat, with a deep and complete central furrow, and the sculpturing hidden by the scaling and the broad subre- cumbent brown setae; the eyes strongly convex, moderately produced backwards, deepest behind the middle, and the hind edge of the orbit not projecting. Rostrum a little shorter than its basal width, strongly narrowed from base to apex, the sides almost straight ; the upper surface with a rather deep median impression containing a complete narrow central carina, and a very shallow longitudinal impression near each lateral margin; the mandibles densely squamose. Antennae with joint 1 of the funicle much longer and thicker than 2, which is nearly twice as long as 3, and all the jomts much longer than broad. Prothorax twice as broad as long, the sides strongly rounded, broadest well behind the middle, the basal margin arcuate, the apex much narrower than the base and broadly sinuate; the upper surface moderately convex transversely, but flat longitudinally, set with large shallow separated setigerous punctures, which are normally hidden by the scaling, and with a narrow smooth central line running from the base for three-fourths the length ; the setae broad, flattened and recumbent. lytra oblong-ovate, obtusely pointed behind, the basal margin jointly sinuate, with the angles not produced ; the striae very shallow, the first three strongly curved outwards at the base, the very shallow punctures normally quite hidden by the scaling; the intervals broad, flat and minutely punctulate, set with irregular recumbent scale-like setae ; the scales very short ovate or almost circular, smaller than those on the prothorax and closely contiguous. Legs stout, densely clothed with grey scales and numerous oblique scale-like setae ; the front tibiae strongly incurved at the apex, somewhat flattened internally and there with a row of 4 or 5 spine-like bristles, the apical comb with 6 or 7 very short teeth; the hind tibiae slightly convex externally, internally flattened and with 3 stout teeth near the apex, the corbels squamose; joint 2 of the front tarsi a little longer than its apical width. Sternwm with the mesepisterna broadly separated from the base of the elytra. * Ann. Mag. Nat. Hist. (9) i, 1919, p. 14. 274 GUY A. K. MARSHALL. Length, 55-7 mm. ; breadth, 3-4 mm. TRANSVAAL: East Rand, 30.1.1919. Described from three specimens. In its colourmg and general sculpture this species most closely resembles P. (Strophosomus) gravis, Gyl., but is readily distinguished from it by its remarkably flattened form and almost oblong elytra. Recorded as injuring the foliage of young orange trees. Protostrophus noxius, sp. nov. (Plate xix, fig. 1). ©. Colour black, with dense sandy brown scaling above, variegated with some paler and darker markings, and uniform sandy grey beneath; the prothorax with usually a broad lanceolate darker median stripe and a broader pale lateral stripe ; the scaling on the elytra brown, with a broad whitish stripe on interval 6 and usually with a macular stripe of alternating dark brown and pale spots along striae 1, 2 and 5. Head separated from the rostrum by an abbreviated and deeply curved stria, the forehead smooth, with a single deep furrow, and densely clothed with over- lapping scales and raised scale-like setae ; the eyes moderately convex, deepest about the middle, projecting strongly backwards, but the hind edge of the orbit not produced. Rostrwm a little more than half as long as the basal width (5: 9), very strongly narrowed from base to apex, with the sides slightly smuate; the upper surface quite flat, with only a very shallow indistinct impression in the middle of the apical half; the mandibles squamose, the mandibular appendages gently curved, simple and tusk-like, flattened ventrally and with the surface partly striate longitudinally. Antennae with joint 1 of the funicle about as long as the next three, joint 2 about as long as broad, the remainder broader than long. Prothorar twice as broad as long, a little broader than the base of the elytra, the basal margin strongly arcuate, the apical margin somewhat narrower and trisinuate, the lateral margins subdenticulate and gently rounded, broadest at the middle; the upper surface strongly convex transversely and flat longitudinally, quite smooth and evenly covered with dense scaling, without any trace of a furrow or carina. Slytra broadly ovate, widest behind the middle, very broadly rounded behind, the basal margin strongly sinuate, fitting closely to the prothorax, with the external angles slightly projecting ; the extremely shallow striae bear rows of punctures that are visible through the scaling, each containing a minute scale; the intervals broad, almost flat, densely covered with almost circular overlapping scales equal to those on the prothorax and each with a row of short recumbent scale-like setae. Legs with the front tibiae moderately incurved at the apex, the inner surface with two or three spine-like setae, the apical margin with a row of five or six stout teeth and an additional one on the outer edge; the hind tibiae with two internal spines near the apex, and the corbels densely scaly ; joint 2 of the front tarsi much broader than long. Sternum with the mesepisterna separated from the elytra. Length, 3°25-3°6 ; breadth, 2-2°2 mm. ORANGE FREE State: Clocolan, 20.vu.1915. Described from three specimens. SOME INJURIOUS SOUTH AFRICAN WEEVILS. 275 Somewhat like a small P. funestus, Pér., but in the latter species the prothorax is distinctly broader in relation to the elytra, its posterior angles are sharp right angles and its anterior ones acute ; and the elytra fit less closely to the prothorax. Found feeding on young wheat. Protostrophus instabilis, sp. nov. (Plate xix, fig. 2). § 2. Colouring variable: piceous, with dense grey scaling; the prothorax with a more brownish median lyre-shaped patch and two indistinct sinuous stripes on each side of the same colour ; the elytra with interval 1 always fawn coloured, elsewhere irregularly mottled with dark brown; usually pale spots stand out rather conspicuously on intervals 3 and 5 behind the middle, and sometimes these are linked up to form a continuous oblique pale band ; occasionally the elytra are distinctly striped, intervals 3 and 5 being almost entirely dark brown and the rest (except 1) light grey. Head separated from the rostrum by a slightly sinuate stria that reaches the sides, the forehead almost flat, with a single deep furrow and the sculpturing quite hidden by the scaling, and set with obliquely raised flattened setae ; the eyes not very prominent, deepest far behind the middle, the posterior edge of the orbit not projecting. Rostrum distinctly broader than long, not very strongly narrowed in front, the sides almost straight and with a slight rounded projection above the apex of the scrobe; the upper surface nearly flat, with a shallow median depression in the apical half, and without any distinct carina ; the mandibles not squamose. Antennae with joint 1 of, the funicle much longer and broader than 2, which is as long as 3 and 4 together, 3-7 transverse, 7 as broad as 1. Prothorax about twice as broad as long, broadest at the middle, the sides very strongly (subangulately) rounded, the base slightly arcuate, with the basal angles rounded off, no broader than the apex, which is truncate ; the upper surface flat longitudinally and smooth, the sculpturing completely hidden by scaling, except for a shallow median stria, which is abbreviated at both ends. LHlytra oblong-ovate, the sides obtusely angulated near the base and from there subparallel to beyond the middle, broadly rounded behind, with the base very shallowly sinuate and the basal angles not projecting; the shallow striae contain separated punctures that are clearly visible through the scaling ; the intervals broad, almost flat, finely coriaceous, and densely clothed with almost circular overlapping fluted scales (of the same size as those on the prothorax) and each with a single row of raised brownish scaie-like setae, which become longer and more erect behind, a few across the top of the declivity beg white. Legs densely clothed with scales and sub- recumbent scale-like setae; the front tibiae only slightly incurved at the apex, with two spines on the inner face, and an apical row of three small and three stout teeth and two stout ones on the outer edge; the hind tibiae with three spines internally on the apical half and the corbels densely squamose. Sternum with the mesepisterna narrowly separated from the elytra. Length, 3-3°5 mm.; breadth, 1°75-2 mm. TransvaaL: Nelspruit, 25.1x.1918. Described from ten specimens. (637) E 276 GUY A. K. MARSHALL. This species differs from typical forms of the genus in the following points :— the obliteration of the longitudinal impressions on the lower surface of the rostrum ; the extension of the transverse cephalic stria to the lateral margin ; the narrowness of the basal margin of the prothorax ; and the humeral angulation on the elytra. in all these characters 1t agrees with Bradybamon, Mshl. (Ann. Mag. Nat. Hist. (9) 1, 1919, p. 9), from which however it differs abundantly in the structure of the epistome, mentum, eyes, venter, etc. Recorded as injuring the foliage of young orange trees. Eremnus horticola, sp. nov. (Plate xix, fig. 3). g§ 9. Colour black, densely clothed with sandy grey scaling, irregularly mottled above with brown markings. Head with the rugose sculpturing entirely concealed by the scaling; the forehead flattened, as broad as the base of the rostrum and with a central fovea. Rostrum parallel-sided at the base, then gradually widening to the apex, very deeply com- pressed behind the scrobes, leaving a raised dorsal area with perpendicular sides, which rapidly narrows from the apex almost to a point at the base; the dorsal area broadly impressed in the apical half and rugosely punctate, the punctures hidden by almost circular large overlapping fluted scales. Antennae with the scape gently curved just below the middle and slightly exceeding the hind margin of the eye; the funicle with joint 2 half as long again as 1, joints 3-6 gradually diminishing, and 7 as long as but broader than 4; the club as long as the three preceding jomts. Prothorax half as broad again as the median length, the sides rounded, broadest before the middle, deeply constricted and broadly sulcate transversely near the apex; the basal margin slightly arcuate, the apex broadly sinuate dorsally, the postocular lobes broad but rather short ; the upper surface uneven, the central area granulato-punctate, with four foveae forming the corners of a transverse parallelogram and with a shallow central furrow towards the base ; on each side of the central area a very broad and deep furrow, bounded externally by a sharp carina, beyond which the surface is rugosely granulate ; all the granules covered with scaling. Hlytra rather broadly ovate in 9, narrower in g, broadest about the middle and obtusely pointed behind ; the dorsal outline only slightly curved, the posterior declivity very steep and curving inwards near the apex, which is hidden from above (especially in the 9); the basal margin gently sinuate and vertically truncate, almost marginate for a short distance on each side of the suture ; the striae shallow and with separated shallow punctures, which are partly hidden by the scaling ; the intervals moderately convex, the alternate ones higher at the base, and the suture obtusely elevated on the declivity, less so in 3 than in ©; the scales much smaller than those on the prothorax, almost circular, fluted and slightly overlapping, the short recumbent scale-like setae irregularly disposed. Legs with dense brown scaling and pale recumbent scale-like setae ; the femora with a pale band on the thickest portion and each with a small tooth. Length, 7-8 mm.; breadth, 3:25-5 mm. ORANGE FREE StaTE: Bloemfontein, 1.xu.1916. Described from eight specimens. Found feeding on dahlias and chrysanthemums. BIE, laws, IyaSiascisi, Wom, ><, IDAinar 8). PLATE XIX. INJURIOUS SOUTH AFRICAN CURCULIONIDE. 1. Protostrophus noxius, Mshl., sp.n. 2. Protostrophus instabilis, Mshl., sp.n. 38. Eremnus horticola, Mshl., sp.n. 4. Protostrophus planatus, Mshl., sp.n. 277 A CONTRIBUTION TO KNOWLEDGE OF THE 'TABANIDAE OF PALESTINE. By Masor E. E. Austen, D.S8.0O. At the present moment our knowledge of the Dipterous fauna, including the Tabanidae, of Palestine is little better than complete tgnorance, consisting as it does solely of a few isolated records of species—chiefly mosquitos. The following paper, which it is intended shall be succeeded by others dealing with other families of Diptera, based upon material collected by the author in Palestine during the recent campaign against the Turks, is a modest attempt to lift one small corner of the veil of obscurity. While, as in Syria according to previous writers,* the dominant Dipterous family in Palestine is certainly the Bombyliudae,} the Tabanidae, of which twenty-one species are recorded or described in the ensuing pages, are by no means poorly represented. The extent to which the Dipterous fauna of Palestine is identical with that of Syria has yet to be determined, and the present writer unfortunately had no opportunity of pursuing his investigations north of Acre. So far, however, as the Tabanidae—the only family yet worked out—are concerned, Loew’s remarks (Verh. z.-b. Ges. Wien, Bd. vu, 1857, p. 79) with reference to the Diptera of Syria would certainly apply to those of Palestine. “The Syrian Dipterous fauna,’ wrote the author in question, “is on the one hand so closely connected with that of Europe, and on the other exhibits so many peculiar species, that it merits a high degree of interest.” | With the exception of three or four examples previously contained in the British Museum (Natural History), the material, amounting to one hundred and forty-four specimens, upon which the following paper is based was collected by the author, with occasional assistance from friends, during the Palestine Campaign of 1917-18. So far as possible, special attention was devoted to the collection ot Tabanidae in 1918, owing to the prevalence early in that year of a rumour that, during 1917, the Turks in the Jordan Valley had lost a very large number of camels from surra, a form of trypanosomiasis of which the causative agent, Trypanosoma evansi, 18 at present believed to be capable of dissemination by more than one species of Tabanus. In view of the imperative necessity of maintaining a large mounted force in the Lower Jordan Valley until the date of the final general advance, the report in question at first caused some apprehension at General Head- quarters, Egyptian Expeditionary Force. Fortunately, however, any fears that may have been entertained were not justified by the result, while the few cases * Bezzi (Brotéria, Ser. Zool., Vol. viii, fasc. 2, 1909, p. 37), writing on Diptera from Syria, remarks :—‘“‘ The peculiar character of the fauna is to be deduced from the multitude of Bombyliidae: this family is preeminent among all others owing to the beauty and wealth of its species, and constitutes 21:6 per cent of the entire collection.” In a foot-note on the same page Bezzi points out that, among thirty-seven species of Syrian Diptera recorded by Loew (‘‘ Nachricht tiber syrische Dipteren,” Verh.z.-b. Ges. Wien, Bd. vii, 1857, pp. 79-80), the Bombyliidae amount to 56°7 per cent. : + The material belonging to this family collected by the author in Palestine during 1917-18 consists of 289 specimens—nearly twice as many as those obtained belonging to any other family ; the precise number of Tabanidae brought back was 136. (637) 2 278 ERNEST E. AUSTEN. of surra diagnosed among animals belonging to the Camel Transport Corps failed to spread, and were by no means certainly contracted in Palestine. While, somewhat curiously, no species of Pangonia* was met with, and the large numbers of specimens of Chrysops encountered all proved to belong to the same species, the genus T'abanus, having regard to the relative smallness of the country concerned, was found to be decidedly rich im species. Of this genus, examples of no fewer than sixteen species were collected, and of these seven, or the high proportion of 43°7 per cent., are apparently new. With one exception, the remainder are more or less well known European forms, some of which have already been shown to occur further afield, as in North Africa, Asia Minor, or Persian Baluchistan. The applicability of Loew’s generalisation, already quoted, con- cerning the Diptera of Syria, when transferred to those of Palestine, at least as represented by Tabanidae of the genus Tabanus, is therefore obvious. Attention is directed to the pursuit of a fast travelling motor car by specimens of two species of Tabanus (cf. pp. 298, 300), one of which proved to be new. Some three months later, a second instance of the same phenomenon was seen under precisely similar circumstances, within a few miles of the same spot, the species concerned in this case being probably Tabanus eggert, Schin., though, since no specimen was caught on this occasion, the identification is uncertain. In Nyasaland, tsetse-flies of the species known as Glossina morsitans, Westw., have been observed to pursue and keep pace with a motor cycle travelling at high speed, and the attraction exerted upon tsetse-flies by moving animals has been mentioned by more than one observer. So far, however, as the writer is aware, the pursuit of a motor car by Tabanids has not previously been recorded. It should be noted that the chase was confined to short distances, so that the chance establishment of a motor route passing through a local Tabanid area is unlikely to have much effect in extending the distribution of species. It is only necessary to add that the whole of the material referred to in the following pages, including the types of all new species, is in the British Museum (Natural History). PANGONIINAE. Genus Chrysops, Meigen. In so far as a definite conclusion is warranted by the author’s necessarily limited observations, the form mentioned below would appear to be the only representative of its genus in Palestine. This paucity of species, however, is more than made up for by the relative ubiquity and absolute aggressiveness of the exemplar con- cerned. Chrysops punctifera, Lw. (fig. 1). Two 99, Wadi Ghuzze, nr. Tel el Fara, 26.vi.1917, biting horse; 1194, 6292, same locality, 23.vii.1917, the $g numerous, resting on rocky bank of Wadi and on plants above pool, many 929 ovipositing on reeds overhanging water; 3g, 499, same locality, 8.vili.1917 ; 14, Deir el Belah, 8 miles 8.-W. of Gaza, 31.vil. 1917; 1g, 19, Wadi Ghuzze, near El Gamli, 12.ix.1917; 299, Wadi Ghuzze, * Bezzi (loc. cit., p. 41, Tab. ix, fig. 32) describes and figures, under the name P. sulcata, a new species of Pangonia from Syria. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 279 nr. Tel el Fara, 14.1x.1917, biting author ; 1g, Wadi Ghuzze, nr. Hl Shellal, 19.ix. 1917; 19, Wadi Ghuzze, nr. Tel el Fara, 28.1x.1917, biting author’s arm; 14, Wadi Sukkereir, 10 miles 8. of Jaffa, 13.1v.1918, gd alone seen; 19, Jericho Plain, nr. Kasr el Hajlah, 28.v.1918 ; 19, Latron, 18.vi.1918, biting author’s hand ; 19, left bank of R. Auja, nr. Sheikh Muannis, 5 miles N. of Jaffa, 29.vii.1918, biting author’s arm; 19, Wadi Ishkar, R. Auja, nr. Ferekkiyeh, 8.viii.1918 ; 19, Wadi Barideh, nr. Sarona, 21.vii.1918, biting author’s hand; 19, same locality, 2.ix.1918, biting author’s arm; 29° (one biting author’s arm), marshes bordering R. Kishon, nr. Haifa, 2.x.1918. In addition to the foregoing, the Museum already possesses a 2 of this species from the “ papyrus marsh above Baheiret el Hileh (Waters of Merom), Upper Jordan Valley, 20.vii.1901 ” (the late Dr. J. Cropper), with the collector’s field-note “ only amongst papyrus.” In Chrysops punctifera the sexual colour-dimorphism, which is so striking a feature in the case of the majority of the palaearctic representatives of its genus, is especially pronounced, and at first sight it is difficult to believe that the predominantly ochreous or olive-ochreous* female, with its brown-banded wings, can possibly be conspecific with the mainly black and largely black-winged male. This exceedingly aggressive and bloodthirsty fly would appear to be generally distributed throughout the entire country, and to occur wherever the necessary conditions for breeding exist. In July 1917, females were seen busily ovipositing Fig. 1. Head of Chrysops punctifera, Lw., °9, in profile, showing eye-markings. \ about 14. in the Wadi Ghuzze. The elongate eggs, which are white when newly laid and subsequently become opalescent brownish, are deposited in countless myriads, closely packed in sheets, on leaves of reeds overhanging pools of slowly flowing or stagnant water. The female of Ch. punctifera, wherever and whenever encountered, seems almost always ready to bite both man and domestic animals, and, so far as the author’s personal experience goes, the species is more aggressive than any other Tabanid met with in Palestine. * For names and illustrations of colours used for descriptive purposes in the present paper, see Ridgway, “Color Standards and Color Nomenclature’ (Washington, D. C. Published by the Author. 1912). 280 ERNEST FE. AUSTEN. The range of the species extends at least into Syria. The type of Chrysops punctifera, Lw., which was originally described (Neue Beitrdge zur Kenntnss der Dipteren, iv, 1856, p. 24) from the male alone, was obtained at Beirut. Sub- sequently (Verh. k.-k. zool.-bot. Ges. Wien, vii, 1858, p. 633) Loew, without giving precise indication of habitat, stated that he possessed nine males and three females of the species from “ Syria,’’ and added some notes on the female sex. Quite recently, during the preparation of this paper, the British Museum (Natural History) has received three males of Ch. punctifera, taken in Beirut Marshes, 11.iv.1919, by the donor, Lt.-Col. HE. P. Sewell, C.M.G., D.S.O., R.AMOC. TABANINAE. Genus Haematopota, Meigen. The twenty-eight specimens of this genus obtained by the author in Palestine appear to belong to four species, none of which can be identified with any of those already known. They are accordingly described as new in the following pages. and an attempt has been made to construct a diagnostic table for their distinction. Since, however, all but three of the specimens are apparently conspecific, while the remainder, representing as many species, are exclusively females, it has been impossible to draw up a synoptic table for both sexes, and the subjoined key should be regarded as purely tentative. Specimens of Haematopota were met with for the first time on 22nd April 1918, on horses near Jericho. None were encountered during the sprmg and summer of 1917, and their apparent non-occurrence in the Wadi Ghuzze is somewhat remark- able. A noteworthy instance of Syrphid-like hovering in the air by males of Haematopota sewelli, sp. n., is recorded on page 285. Key to Species described below. 1(2). First jomt of antenna, viewed from side, with a deep preapical groove* on upper surface ; femora largely greyish fawn-coloured or greyish cinnamon sewelli, sp. n. [1]. 2(1). First jomt of antenna without such a groove; femora entirely or mainly AG 3(4). First jomt of antenna, viewed from side, not at all swollen. Small or very small species... Pe monuscularia, sp. Nn. [2]. 4(3). First jomt of antenna, ed few als distinctly or strongly swollen. 5(6). First jomt of antenna, viewed from side, strongly swollen, its greatest vertical diameter greater than or fully equal to half the length of the joint ; dorsum of abdomen with a broad, grey, border on each side minuscula, sp. n. [3]. 6(5). First jomt of antenna, viewed from side, moderately swollen but elongate, its greatest vertical diameter considerably less than half the length of the jomt; dorsum of abdomen with a double row of grey spots, but without grey lateral borders .. at i ennominata, sp. n. [4]. * Sometimes obscured by hair, in which case it may be difficult to distinguish. If, however, the head be examined from the left side, with a hand-lens magnifying about 5 diameters, the groove, if present though indistinguishable in the left antenna, can generally be seen distinctly on examining the inner side of the first joint of the right antenna. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 281 1. Haematopota sewelli, sp n. (figs. 2, 3). 3 9.—Length, $ (8 specimens) 11 to 12°2 mm., 9 (17 specimens) 9 to 12°5 mm. ; width of head, 94 to 4:4 mm., 9 3 to 4 mm. ; width of front of 2 at vertex 1:2 to 15 mm. ; length of wing, 3 8°5 to 9:4 mm., 9 74 to 10 mm. Blackish (3) or blackish brown, greyish olive, or deep greyish olive () species, allied to and superficially closely resembling H. pluvialis, Linn.: distinguished in the 3 by the har on the upper and under sides of the first antennal joint being shorter and less fine, while the proximal half of the jount 1s enturely olive-grey pollinose, instead of the pollinose covering being confined chiefly to the upper and inner side of the proximal half or two-thirds ; by the femora of all three pairs of legs in the 3, or at least those of the middle pair, having a complete or incomplete tawny or ochraceous-tawny band, instead of all the femora beng unrcolorous dark neutral grey, vron grey, or dark olwve- grey ; by the first antennal joint in the 2 beng somewhat longer, usually slightly less swollen in proportion to ats length, duller (mouse grey, dark neutral grey, or connamon- drab pollinose instead of partly shining black), and, if anything, more distinctly constricted before the tup, but with rts upper surface not humped up before the con- striction ; and by the femora in the 2 beng largely greyish fawn-coloured or greyish cimnamon, instead of uniformly grey (deep olive-grey or neutral-grey). Head: front in 9 smoke grey pollinose, clothed partly with short, erect, blackish hair, partly with pale yellowish hair, vertex with a pair of admedian, mouse grey or deep mouse grey blotches ; frontal triangle in $ greyish olive, both sexes between and above bases of antennae with a large, dead black or blackish brown, hexagonal or quadrate spot, which in 9 is bounded above by the frontal callus; face in 3 pallid neutral grey pollinose, in 2 whitish pollinose, in both sexes speckled with blackish brown on each side, in 3 clothed on each side with fairly long blackish brown hair, in @ with a pair of small but fairly conspicuous, admedian, blackish brown spots below the antennae, and clothed with whitish hair; jowls whitish pollinose, clothed with whitish hair, part of which in 3 sometimes has a yellowish tinge ; occiput olive-grey, light olive-grey, or deep olive-grey, its upper margin fringed in ¢ with a mixture of fairly long, erect, blackish and yellowish hair, and in @ on each side of front with short yellowish hair; frontal spots in 9 blackish brown or black, median spot usually small but distinct, lateral spots either roughly triangular or more or less rounded, not in contact with eyes; frontal callus in 9 shining black, of moderate depth, its lateral extremities somewhat deeper and narrowly connected below with each eye, its upper margin produced in middle le into a small, upwardly directed angle; eyes in ¢ densely clothed with fine, erect, drab-grey hair, hairy covering of eyes in 2 very short, but distinguishable under a hand-lens magnifying about 5 diameters, larger facets in § contrasting conspicuously with those forming remainder of surface, but not coarse ; proximal segment of ¢ palpy mouse grey or deep mouse grey, clothed with long whitish hair, terminal segment fairly large, ovate pyriform, pointed at distal extremity, greyish pollinose pinkish buff, usually with an ill-defined mouse grey blotch on outer side near base, clothed with whitish hair, which on upper margin beyond base is mixed with fine, erect, black or blackish hair ; palpi in Q greyish pollinose pinkish buff, proximal segment sometimes with a mouse grey spot on outer side near 282 ERNEST E. AUSTEN. distal extremity, terminal segment moderately (in some specimens considerably) thickened towards base, proximal segment and base of terminal segment below clothed with fairly long whitish hair, terminal segment otherwise clothed on outer side with short, appressed, glistening pale yellowish hair, usually mixed on upper surface except at base, or at any rate towards distal extremity, with short, appressed, black hairs; first joint of antennae in ¢ considerably swollen, when viewed from above truncate ovate or elliptical ovate in outline, its distal half shining black ; first joint of antennae in ¢ clothed below with yellowish white hair and above with black or blackish hair, except proximal portion of pollinose area at base, where hair is yellowish; second joint of antennae in ¢ blackish mouse grey, clothed with short and rather coarse black hair; expanded portion of third joint of antennae in ¢ of moderate depth as viewed from side, more or less greyish at base, cinnamon in middle, and dark brownish towards distal extremity, on inner side with a vertical band of minute, appressed, black hairs between blunt angle on upper border and lower margin ; annulate portion of third jomt of antennae in ¢ blackish brown, ending bluntly ; first joint of antennae in @ thick, cylindrical, clothed with semi-appressed, short black hairs, lower surface, except distal extremity, clothed with longer whitish hair; second and third joints of antennae in 9 asin g. Thorax: dorsum blackish brown (¢) or dark greyish olive ( ), in each case with pale smoke grey markings of the usual type; in 9 these markings are but little conspicuous, and consist of a narrow median line, commencing on fore border but disappearing before reaching transverse suture, and a pair of admedian stripes, broader on anterior border but rapidly tapering off, interrupted before reaching transverse suture and reappearing behind latter in shape of a pair of more or less triangular spots resting on suture ; markings in Q similar to foregoing, but in some specimens median line is continuous and reaches prescutellar groove, usual crescentic marks in front of latter are distinguishable, there is a broader pale smoke grey stripe on each side extending from inner extremity of depression at end of transverse suture to basal angle of scutellum, and lateral border on each side, from humeral callus to depression at end of transverse suture, and from latter to basal angle of scutellum is pale smoke grey ; humeral calli in ©, and swelling occupying depression at each end of transverse suture in same sex smoke grey ; scutellum in each sex agreeing in coloration with ground colour of remainder of dorsum ; pleurae and pectus in both sexes pale smoke grey pollinose and clothed with whitish hair (latter on mesopleurae in g sometimes appears to have a yellowish tinge); dorsum, including scutellum, clothed in 3 with fine, erect, yellowish hair, and in 9 with short, appressed, glistening yellowish hair, which sometimes, especially towards lateral borders, appears to have a pinkish tinge. Abdomen: dorsum in ¢ dull black, posterior angles of first (visible) tergite and lateral extremities of two following tergites ochraceous-buff, ochraceous-tawny. or cinnamon, posterior margins of second and following tergites (in case of second, third, and fourth segments, and sometimes of fifth segment also, more or less distinctly expanded in middle line into a wide flattened triangle) dark olive-buff : second or third and following tergites to sixth inclusive in ¢ each with a pair of more or less distinctly marked greyish olive or deep greyish olive, rounded admedian spots (most distinct on fourth and two following segments), resting on CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 283 front margin in each case but usually not reaching or not quite reaching hind border, a certain number of the tergites indicated, at least the fourth to the sixth, also with a narrow median stripe of same colour as the spots; seventh tergite in d, except posterior and lateral margins, deep olive or dark olive, without spots or median stripe, lateral margins and posterior angles of third and following tergites of same colour as posterior margins; dorsum in ¢ clothed mainly with blackish hair, which in central region of first three and sometimes the following tergites also may be largely mixed with glistening yellowish or cream-buff hair, posterior angles of first and lateral extremities of two following tergites, posterior angles of fourth to seventh tergites inclusive, hind margins of sixth and seventh tergites and junction of median stripe with hind margin on the four preceding tergites Fig. 2. Haematopota sewelli, Austen: a, head of g from above; a’, antenna of ¢ in profile; b, head of Qin profile. x 14. aoe Fig. 3. Haematopota sewelli, Austen, wine Of O27 XxX G. clothed with bright, glistening, cream-buff hair, downwardly curved lateral extremities of first tergite clothed with whitish hair; venter in 3 mouse grey or deep mouse grey (seventh segment dark mouse grey), a larger or smaller area at lateral extremities of second and third segments light greyish pollinose ochraceous- salmon, hind margins of second and following segments light ochraceous buff ; venter in ¢ clothed with fine, whitish or yellowish hair, seventh segment, except lateral extremities, with usual coarse, erect, black hair, central area of sixth segment’ usually with finer black hair ; dorsum in @ blackish brown or dark blackish brown, each segment with a pair of relatively large, greyish olive or deep greyish olive, rounded, admedian spots, resting on front margin and not reaching hind border ; Jatter, as well as posterior angles of each segment, a median, sub-triangular fleck 284 ERNEST E. AUSTEN. on first (visible) tergite, and a median longitudinal stripe on each of the following tergites, often indistinct or scarcely developed on seventh tergite but on the others expanded laterally at its junction with hind border (on second tergite so much expanded posteriorly that it has the shape of a median triangle), of same colour as the paired spots; lateral extremities of first two or three tergites pale smoke grey or deep gull grey, those of first (visible) tergite sometimes with a more or less distinct pinkish cinnamon fleck on each side; ground colour of posterior angles and hind margins of second or third and following tergites sometimes more or less distinctly cinnamon-buff or cream-buff; dorsum clothed on darker areas, as also on spots, with minute, appressed, black hairs, elsewhere clothed as a rule with appressed, glistening, cream-bufi or cream-coloured hair; venter in Q light greyish olive, smoke grey or pale smoke grey pollinose, second and following segments each usually with a darker blotch in midde line, and with cream-buff or cream- coloured hind borders; venter in 2 clothed with minute, appressed, glistening, yellowish or cream coloured hairs, last two segments mainly with erect black hairs, a few black hairs in centre of posterior half of antepenultimate segment also. Wangs (fig. 3.) drab or light sepia-coloured ; usual three rosettes, pale mark in marginal cell at distal extremity of stigma, and sinuous mark or transverse streak just before apex of wing (starting from costa at anterior distal angle of first submarginal cell) well developed and sharply defined; inner margin of proximal rosette as a rule perfectly distinct from (not fused with, as is often the case in the wing of H. pluvialis, Linn.) other pale marks crossing basal and anal cells ; distal marginal angles of all posterior cells, or at least of second to fifth inclusive, often con- spicuously pale; zig-zag pale mark running across proximal extremity of distal third of anal and axillary cells, as also loop in base of latter, well developed; discal cell with two pale transverse marks (portions of proximal and median rosettes) at each extremity ; veins or portions of veins surrounded by centres of the three rosettes more or less distinctly infuscated ; no conspicuous dark blotch beneath stegma ; latter mummy brown or sepia-coloured, well developed, its proximal extremity pale. Squamae light drab (3), with darker (drab-coloured) borders, or pale smoke grey (2), with ivory yellow borders. Halteres cream-coloured, knobs clove brown at base above and below. Legs: coxae in § neutral grey, clothed with long whitish hair ; femora in § deep mouse grey, or dark mouse grey, with a more or less conspicuous, complete or incomplete, tawny or ochraceous- tawny band round middle, present at least on middle pair, but sometimes absent in case of front femora and scarcely distinguishable on hind pair; femora in clothed with fine whitish hair, which is long behind and below; distal half of extensor surface of front femora in 3, and extreme tips of same surface in hind pair clothed for-most part with minute, appressed, black hairs; front tibiae in g blackish brown, with a conspicuous cream-buff band just beyond base, and with distal two-thirds conspicuously or distinctly swollen; middle and hind tibiae in ¢ dark brown, each with two cream-buff bands, hind tibiae thickened ; tibiae in ¢ clothed with black or blackish hair, which on middle and hind pairs is fairly long and fine, pale bands on all tibiae, as well as anterior surface of hind pair except distal extremities clothed with glistening cream-coloured or Naples yellow hair ; front tarsi in ¢ black, middle and hind tarsi in same sex dark brown, first joints CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 285 of middle and hind tarsi, except distal extremities, cream buff, hairy covering of all tarsi black ; front coxae in Q greyish pollinose cream-coloured or light drab, posterior surface and distal extremities usually neutral grey or light neutral grey, middle and hind coxae in @ pale neutral grey, all coxae clothed with whitish hair, longest in case of front pair, though front coxae sometimes show shorter, glistening, appressed, cream-coloured hairs on outer surface; femora in Q greyish fawn- coloured or greyish cinnamon, their extremities, at least distal extremities, more or less mouse grey or deep mouse grey ; hairy covering of femora in Q as in J, but shorter ; tibiae and tarsi in 9 as in 4, but hairy covering shorter. Jafia district ; Lower Jordan Valley ; near Tul Keram; and Acre. Type of ¢ and five para-types belonging to the same sex, near Jerisheh, 5 miles N.-E. of Jaffa, 29.iv.1918 ; hovering at 9.0. a.m., poised in the air like Syrphids, in front of a tent on the outside of which two 99 of the same species were resting. Type of 9, same spot as foregoing, 5.v.1918. Additional para-types as follows :— 244, same spot as before, 26.1v. and 5.v.1918 respectively ; 999, Ain es-Sultan, near Jericho, 22.iv.1918, on horses ; 499, same spot as that at which the typical specimens were taken, 29.iv. and—.v.1918; one 9, W. bank of R. Jordan, near Jericho, between El Ghoraniye and the Dead Sea; one Q, Kerkar (between Tul Keram and Zimmarin), 28.1x.1918, biting author’s arm; 19, marsh near Acre, close to mouth of Nahr Naaman (R. Belus), 5.x.1918—one of two specimens, one of which bit author’s arm. As a small acknowledgement of much kindness shown to the author, the species just described has been named in honour of his friend Lieut.-Col. (temp. Col.) HE. P. Sewell, C.M.G., D.S.0., R.A.M.C. (late D.D.M.S., XXIst Corps, HE. E. F.). So far as the experience of 1918 warrants such a conclusion, Haematopota sewellt would appear to be the commonest representative of its genus in Palestine. The remark- able habit of the males, referred to above, of remaining poised in the air like Hover- flies, although new to the present writer, would seem to be a generic character, since Schiner (Fauna Austriaca. Die Fliegen (Diptera), I. Theil, 1862, p. 38), at the end of his notes on the genus Haematopota, remarks :—“ The males too hover in the air in the morning and towards evening.” 2. Haematopota minuscularia, sp. n. (fig. 4, a). O.—Length (1 specimen) 6°6 mm.; width of head 2°6 mm.; width of front at vertex 1 mm.; length of wing 5°8 mm. Small, greyish black species, with somewhat elongate antennae, of which first joint as cylindrical, not swollen, and, when viewed from side, without a noticeable, groove-lke constriction before its distal extremity; wings light drab, with all three rosettes well marked, adjacent margins of muddle and proximal rosettes fused together where they cross discal cell, and with a pale border running along hind margin ; femora, except extreme tups, which are narrowly warm buff, entirely grey, those of second paar of legs, at proxumal ends of distal halves, each with a faint indication of a greyish fawn- coloured band. Head: front pale neutral grey, lateral margins, a narrow border encircling each frontal spot, and a narrow median line from vertex to median frontal spot paler 284 ERNEST E. AUSTEN. (pallid neutral grey); face and jowls pallid neutral grey pollinose, sides of face punctate in usual manner, but pair of dark spots on central region of face, seen in H. sewell1, Austen, 2, absent or but faintly indicated ; occiput pale neutral grey ; with exception of a few minute black hairs in region of vertex, and on or near lateral frontal spots, hair on head entirely white ; frontal spots dead black, median spot small but distinct, lateral spots large, circular, not in contact with eyes ; frontal callus shining black, of moderate and practically uniform depth, narrowly in contact with eye on each side below, its upper margin, at least in case of type, not notice- ably produced into an angle in middle line ; usual spot in middle line below callus dull black, wider than deep, not conspicuous ; eyes clothed with short whitish hair ; proximal joint of palpz neutral grey, clothed with long white hair, terminal joint pale pinkish buff, not much swollen towards base, clothed with longer white hairs on proximal portion of under surface, and on outer side with minute, appressed, glistening creamy-white hairs, mixed towards distal extremity with minute black hairs ; first jomt of antennae olive-grey pollinose (shining black on outer and inner sides at distal extremity, which, though somewhat narrower than remainder of jomt, is not marked off by a conspicuous constriction), clothed above and on b Fig. 4. a, Haematopota minuscularia, Austen; 6b, H. miéinuscula, Austen ; c, H. innominata, Austen. Heads of 99 in profile. x 14. outer side with very short, pale yellowish hairs, and below with longer white hair, second joint of antennae greyish brown, clothed with minute black hairs, third joint clove-brown, paler (cinnamon-brown) at extreme base, expanded portion as viewed from side not very deep, roughly elliptical in outline, longer than annulate portion. Thorax: dorsum blackish brown, clothed (including scutellum) with short, fine, yellowish or yellowish white hair, and marked with six, more or less distinct and continuous, pale olive-grey, longitudinal stripes, the two outer stripes on each side fused together in front of transverse suture, and behind latter much broader than remainder ; scutellum with extreme base of outer border on each side, and a more or less indistinct median longitudinal stripe on upper surface pale olive-grey ; pleurae and pectus pale olive-grey, clothed with long white hair. Abdomen: dorsum blackish brown, seventh tergite mouse grey, lateral borders and hind margins of first six tergites, as well as, on each segment from second to sixth inclusive, a median, forwardly directed triangle, based on hind margin and reaching or almost reaching front margin, and a pair of rounded, admedian spots, resting on or close to anterior margin, pale olive-grey ; dorsum clothed with short, CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 287 appressed, whitish hair, except on blackish brown areas adjacent to median triangles on second to fourth segments inclusive, where hairy covering is black or blackish, and more erect ; venter pale neutral grey, clothed with minute, appressed, creamy white hair, terminal segment with usual longer, coarse, erect, black hairs, extreme hind margins of second to sixth segments inclusive paler (pallid neutral grey or ivory yellow). Wangs: light markings in distal half, at least in case of type, somewhat moniliform ; pale border running along hind margin fairly deep, on distal side of each posterior marginal cell deeper than on proximal, commencing in second submarginal and terminating at base of axillary cell, in proximal half of which it becomes very narrow ; owing to fusion of adjacent margins of proximal and middle rosettes, discal cell enclosing only three pale marks; imner margin of proximal rosette almost or quite in contact with other pale marks crossing basal, anal and axillary cells ; sinuous mark at tip of wing not reaching pale hind border ; stigma light sepia-coloured, paler at proximal extremity ; veins or portions of veins surrounded by the three rosettes more or less distinctly infuscated. Squamae semi-hyaline, borders ivory vellow or cream-coloured. Halteres ivory yellow, knobs clove brown above and below. Legs: coxae and femora light neutral grey, clothed with whitish hair; front tibiae brownish black, with a fairly broad, creamy white band just beyond base, distal two-thirds conspicuously swollen, pale band clothed with glistening, appressed, cream-coloured hairs, a few of which pass over on to the brownish black portion, which is clothed otherwise with black hairs ; middle and hind tibiae clove brown, clothed, except at distal extremities, chiefly with glistening, appressed, yellowish hairs, and each encircled with two cream-buft bands, distal extremities clothed with minute black hairs, hind tibiae somewhat thickened ; front tarsi brownish black, clothed with black hair ; middle and hind tarsi clove brown, first joints, except distal extremities, cream- buff. Ain es-Sultan, near Jericho, 8.v1.1918, biting author’s arm. Apart from its small size, this tiny species can be distinguished, at least in the female sex, from all its congeners described or mentioned in the present paper by the shape of the first jomt of the antenna. 3. Haematopota minuscula, sp. n. (fig. 4, b). @.—Length (1 specimen) 7°6 mm.; width of head 3 mm.; width of front at vertex 1:2 mm. ; length of wing 6:2 mm. Dorsum of thorax olive, with lateral borders and three narrow, incomplete, longitudinal strupes pale smoke grey; dorsum of abdomen with black, median, longitudinal strvpe (tapering and becoming olivaceous black towards distal extremity), and with lateral borders broadly olwve-grey ; first joint of antennae strongly swollen ; wngs light drab, with usual pale markings well developed, and those in discal cell as an H. pluvialis, Zann. Head: front deep olive-grey, lateral margins and a ring encircling each lateral frontal spot pale smoke grey, a median stripe extending from margin of occiput at vertex to median frontal spot, and a small ring encircling latter, light olive-grey ; face and jowls pale gull grey, sides of face punctate in usual manner, but pair of dark spots on central region of face, seen in H. sewelli, Austen, 9, and H. plumalis, 288 ERNEST E. AUSTEN. Linn., 2, absent; occiput light olive grey, upper border darker (greyish olive) ; with exception of a few minute black or blackish hairs on front above median frontal spot, and between upper extremities of inner sides of lateral frontal spots, hair on head entirely pale, that on front and fringe of occiput above with a distinct yellowish tinge, hair on face and jowls white; frontal spots dead black, median spot distinct, lateral spots large, roughly oval, not in contact with eyes ; frontal callus shining black, of moderate and practically uniform depth, narrowly separated from eye on each side, and with upper margin produced in middle line into an upwardly directed angle; spot in middle line below callus dull black, triangular or wedge-shaped in outline (narrower below) ; eyes clothed with minute, pale hairs ; palpi light buff, clothed with white hair, proximal joint light neutral grey on outer side towards base, terminal joint slender, elongate and acuminate, as viewed from side only slightly swollen towards base; first jomt of antennae shining black, neutral grey pollinose on upper surface of proximal half, elliptical oval in outline as seen from above, as viewed from side thickest a short distance before distal extremity, but wrthout groove-like constriction before latter exhibited by both H. pluvralis, Linn., 2 and H. sewells, Austen, 2 (vide swpra) ; first joint of antennae, except distal extremity, clothed above and on upper part of outer surface with glistening yellowish hair, and below with longer white hair, distal extremity clothed with black hair; second joint of antennae pale (light drab on outer side, inner surface greyish pollinose light ochraceous buff), distal angles clothed above and below with black hairs; expanded portion of third joint of antennae clove brown (base paler—ochraceous tawny), elongate and tapering, moderately expanded a little beyond base, annulate portion brownish black, shorter than expanded portion. Thorax: dorsum, including scutellum, clothed with partly appressed, partly erect, glistening pale yellowish or whitish hair of medium length, hair on front border whitish, somewhat longer and more erect ; fore border of dorsum pallid neutral grey ; median pale smoke grey longitudinal stripe extend- ing from fore border to a poimt two-thirds of distance between transverse suture _and prescutellar groove ; admedian pale smoke grey longitudinal stripes not reach- ing transverse suture, but each reappearing behind latter in usual shape of a triangular spot resting on suture; pale smoke grey lateral border on each side partially divided behind transverse suture into two longitudinal stripes ; swelling occupying depression at each end of transverse suture clothed below with fairly long and fine blackish hair; pleurae and pectus pale smoke grey, and clothed with fine white hair. Abdomen: black median longitudinal stripe on dorsum not sharpely defined on first (visible) tergite, attaining its greatest breadth (one- third of transverse diameter of segment) on second tergite and thereafter becoming progressively narrower, scarcely distinguishable on seventh tergite, ground-colour of which is mouse grey ; hind margins of second and following tergites, as well as, on second to sixth segments inclusive, a more or less distinct narrow median line light greyish olive ; inner edge of olive-grey lateral border on each side of each segment formed by a large rounded grey spot, in contact with anterior border of each segment and on first three segments more or less completely fused with lateral border, while on last four segments it may be partly or completely separated therefrom, in which case the spot is partly surrounded with blackish brown ; CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 289 dorsum clothed on basal angles and on extreme lateral margins of first four segments with fine white hair, clothed otherwise on grey lateral borders with minute, appressed, glistening yellowish hairs (longer on hind margins of last three segments, and on spots on fourth to sixth segments inclusive partly or entirely replaced by minute black hairs), black or olivaceous black areas clothed for most part with minute black hairs; venter smoke grey, hind margins of second and following segments ivory yellow, third to sixth segments inclusive and posterior portion of second segment clothed with minute, appressed, yellowish hairs, central portion of seventh segment clothed as usual with longer, coarse, erect, black hair. Wangs: outer borders of rosettes somewhat broken up, apical smuous mark (pale transverse streak just before apex of wing) well developed and (at least in the type) reaching hind margin ; distal extremity of each posterior marginal cell occupied by a large, somewhat wedge-shaped pale blotch, posterior margin of axillary cell, from distal extremity to fused ends of zig-zag transverse pale marks, also pale, so that hind margin of wing has a deep pale border, interrupted at proximal marginal angles of posterior marginal cells (broken up into a series of blotches); stigma light sepia-coloured, not very noticeable, paler at proximal extremity. Squamae milk white, borders ivory yellow. Halteres ivory yellow, knobs with a seal brown spot at base above and below. Legs: coxae and femora neutral grey, clothed with white hair, distal extremities of front and hind femora above and below and of middle femora above clothed with black hairs, which also cover rather less than distal half of antero-superior surface of front femora; extreme tips of all femora cream-buff, middle femora with a more or less distinct, complete or incomplete, light greyish cinnamon band, occupying approximately middle third ; front tibiae brownish black, with a fairly broad cream-coloured band just beyond base, distal two-thirds conspicuously swollen, pale band clothed with glistening, appressed, yellowish hair, front tibiae otherwise clothed with black hair; middle and hind tibiae clove brown, middle tibiae with extreme base and two bands cream- coloured, hind tibiae with extreme base and two bands cream-buff, hair on middle and hind tibiae mainly black on dark areas, glistening cream-coloured or yellowish on pale bands, inner surface of hind tibiae clothed mainly with minute, appressed, glistening, yellowish hairs, extensor surface of middle tibiae sparsely fringed with long, pale hairs ; front tarsi black, clothed with black hair, third and fourth joints somewhat expanded; middle and hind tarsi clove brown or blackish brown, clothed with minute black hairs, first joints, except distal extremities, cream-buff or cream-coloured, second, third, and fourth joints also narrowly pale at extreme base. Jericho Plain, near Makhadet Hajlah, R. Jordan, 9.vi.1918, biting mule. The predominance of pale hair on the front in the 9, the pale second antennal . joint, and the markings of the dorsum of the abdomen and of the discal cell in the wing—the latter agreeing with those seen in the wing of Haematopota pluvialis, Linn.—will serve to distinguish the species described above, at least in the female sex, from H. crassicornis, Whlbg., with which it agrees in the absence of a groove- like constriction before the distal extremity of the first joint of the antennae. So far as it is possible to judge from the original description of Haematopota pallens, Lw., the typical specimens of which were obtained in Turkestan, 290 ERNEST E. AUSTEN. H. minuscula resembles that species, especially in the markings of the abdomen and wings, but is distinguishable, inter alia, by its smaller size, by the less elongate antennae, and by the more swollen first antennal joint. 4. Haematopota innominata, sp. n. (fig. 4, c). °.—Length (1 specimen) 9 mm.; width of head 3°25 mm.; width of front at vertex 1°2 mm.; length of wing 7°5 mm. Dorsum of thorax dark olive, with lateral borders and three narrow, complete, longitudinal stripes pale smoke grey or smoke grey ; dorsum of abdomen black, each segment from second to seventh inclusive with hind border, a more or less distinct median strvpe (practically obsolete on third segment in the type), and a pair of rounded spots (relatively small as compared with corresponding spots in H. sewelli, Austen, ©—vide supra, p. 283) light olive-grey ; first joint of antennae as seen from side distinctly incrassate, roughly elliptical, somewhat narrower in middle, narrowed at each extrenuty, but without a sharply marked preayical groove. Head: front deep mouse grey, posterior angles, border of occiput, a narrow border surrounding median frontal spot, and a median stripe connecting latter with occipital margin pale neutral grey, extreme lateral margins and a narrow ring surrounding each lateral frontal spot pale smoke grey ; face and jowls pale gull grey or whitish, sides of face punctate in usual manner, pair of dark spots on central region of face, seen in H. sewelli, Austen, 9, and H. pluvialis, Linn., 9, faintly marked ; occiput pale neutral grey ; hair on front mainly dusky (blackish ~ brown or blackish), that on margin of occiput pale yellowish and short, upper margin of frontal callus partly concealed by a downwardly directed fringe of fairly long, appressed, glistening whitish hair, face and jowls clothed with white hair ; frontal spots dead black, median spot small but distinct, elongate (fusiform or elliptical), lateral spots large, bluntly triangular, at least in type, not in contact with eyes ; frontal callus shining black, of relatively considerable and practically uniform depth, in contact with eye on each side below, but with its upper margin scarcely produced into an angle in middle line ; spot in middle line below callus dull black, truncate triangular, narrower below ; eyes sparsely clothed with minute, pale hairs ; proximal joint of palpr neutral grey on outer surface and clothed with whitish hair, terminal joint light buff, clothed below at base with fairly long and, on outer side with short, appressed, glistening whitish hair, mixed on outer surface beyond proximal fourth with minute black hairs, terminal joint acuminate, proximal half, at least in type, considerably swollen; first joint of antennaz neutral grey pollinose (narrower distal extremity shining black), clothed above and on inner side with short black hair, on upper part of outer side with short whitish hair, and below, especially on proximal half, with longer whitish hair; second joint of antennae blackish brown, clothed with black or blackish hair (third joint missing In case of type). Thorax: dorsum clothed with short, partially appressed, glistening yellowish hair, dorsal surface of scutellum clothed with longer and more erect hair of same kind; median longitudinal smoke grey stripe very narrow, admedian stripes broader, but tapering to transverse suture and becoming partly obsolete midway between latter and prescutellar groove, pale smoke grey lateral border, behind depression at each end of transverse suture, indistinctly divided CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 291 into two fairly broad longitudinal stripes by means of an olive streak ; scutellum with an ill-defined olive-grey median stripe, and base of each lateral border light olive-grey ; swelling occupying depression at each end of transverse suture clothed below with fine blackish hair; pleurae and pectus light neutral grey, clothed with whitish hair. Abdomen: first (visible) tergite without either median stripe or rounded spots, otherwise agreeing in coloration with following tergites, rounded spots on fourth and subsequent tergites resting on base of segment, and in each case partly concealed by hind margin of preceding tergite; lateral extremities of first and second tergites pale neutral grey, those of first tergite clothed with fine white hair, lateral extremities and posterior angles of third and following tergites and hind borders of all tergites clothed with fine, glistening pale yellowish or yellowish white hair, longer than elsewhere on posterior angles of fifth and sixth tergites, and on hind borders of last two segments ; dorsum otherwise clothed with minute, appressed, black hairs, second tergite with an area of minute, appressed yellowish hairs extending from in front of each rounded spot towards posterior angle on each side, spots on third tergite and area in front of them partly clothed with similar hairs ; venter olive-grey, median third of last three scutes darker (mouse grey), hind margins of second and following segments pale olive- buff; ventral surface, except last segment, clothed with minute, appressed, yellowish hairs, terminal segment clothed as usual with coarse, erect, black hairs, a few black hairs also on preceding segment. Wings as in H. sewelli, Austen ; distal margins of rosettes moniliform ; posterior margin of wing with a rather narrow, interrupted, pale border, extending from distal extremity of first posterior cell almost to middle of axillary cell, somewhat faint in distal extremity of first posterior cell, and in each subsequent posterior cell broader in distal marginal angle and tapering towards proximal marginal angle, where it becomes indistinct ; stigma mummy brown, well marked, its proximal extremity pale. Squamae pale smoke grey, borders ivory yellow. Halteres ivory yellow, knobs seal brown above and below. Legs: coxae and femora light neutral grey or neutral grey, clothed with glistening whitish or silvery white hair, extreme tips of femora as usual cream- buff; front tibiae brownish black, with a fairly broad cream-coloured band just beyond base, distal two thirds conspicuously swollen, pale band clothed with glistening, appressed, silvery white hair, front tibiae otherwise clothed with black hair ; middle and hind tibiae clove brown, each with two cream-buft or cream- coloured bands and largely clothed with minute, glistening, yellowish hairs, fringes on antero-extensor margins of hind tibiae consisting of rows of black and yellowish hair alternately ; front tarsi black, clothed with black hair ; middle and hind tarsi blackish brown, clothed with minute black hairs, proximal two-thirds of first joint in each case cream-buff or cream-coloured. Near Jerisheh, 5 miles N.-E. of Jaffa, first week of May 1918. The species just described is allied to Haematopota plumalis, Linn., and H. sewelli, Austen, but is distinguished from both in the female sex by the first joint of the antenna being less swollen and more elongate, with the groove-like constriction before the distal extremity less sharply marked, although the extremity itself is narrowed. Haematopota innominata, Q, is further distinguished from H. sewellz, 2, by the dorsum of the abdomen being blacker, owing to the much less extensive {637) F 292 ERNEST E. AUSTEN. development of the grey markings; and by the femora, except the extreme tips, which as usual are cream-buff, being entirely neutral grey, without any trace of greyish fawn-coloured or greyish cinnamon markings. From Haematopota minuscularia, Austen, 2°, H. nnominata, 9, is distinguishable owing to its larger size, the hair on the front being mainly blackish or blackish brown, the first joint of the antenna, as viewed from the side, being very distinctly swollen and less conspicuously greyish pollinose, and by the grey markings on the dorsum of the abdomen being much less extensive. Lastly Haematopota innominata, 9, may be distinguished from H. minuscula, Austen, 9, by the first joint of the antenna being less swollen and more elongate, and by the absence of broad, grey, lateral borders to the dorsum of the abdomen. Genus Tabanus, Linn. The seventy specimens of this genus obtained by the author prove, as already stated, to belong to sixteen species, of which seven, or 43°7 per cent—a fairly high proportion—are apparently new. The list of species is as follows : OF 1. Tabanus decorus, Lw. ay a alexandrinus, Wied. De 44 unsecutor, Sp. N. 4 a gigas, Herbst. 5, At mendicus, Villen. 6. of lunatus, Fabr. 7 ‘4 nemoralis, Mg. 8 A eggert, Schin. 9, “3 autumnalis, Linn. 10. Ms regularis, Jaenn. 1% i rupinae, sp. N. 12. 3 arenwwvagus, Sp. 0. Te a accensus, Sp. Ni. 14, ~ leleani, sp. n. 15. = ,, ~— pallidipes, sp. n. 16. c dalet, sp. n. The representatives of Tabanus gigas and T. autumnalis show variations in coloration already noticed more than sixty years ago by Loew (Verh. z.-b. Ges. Wien, Bd. vin, 1858, pp. 582-583, 605), in the case of examples from Sicily, Cyprus, or Asia Minor. In particular, the russet tint of the ground-colour of the dorsum of the abdomen in females of 7. autumnalis, Linn., makes the recognition of such individuals a matter of some difficulty for those who are only familiar with the form of the species usually seen in Northern and Central Europe, in which the abdomen in the female is black or blackish mouse grey. While all of the previously described species of Tabanus met with in Palestine are purely palaearctic, and the entire Tabanid fauna appears free from any admixture of Oriental or Ethiopian forms, certain of the more or less grey or drab- coloured species described below as new, such as 7. leleani, T. arenivagus, and T. rupinae, appear to some extent to show the influence of a desert environment. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 293 Key to the Sixteen Species of Tabanus recorded below.* 1 (1£). Eyes (under an ordinary hand lens magnifying about 5 diameters) distinctly or conspicuously hairy. 2 (3). Body uniformly black, unrelieved by conspicuous lighter markings formed by hair or ground-colour, or both; wings blackish brown, or with veins strongly suffused with that colour .. oe alecandrinus, Wied. [2]. 3 (2). Body not uniformly black, if mainly black or slate-black invariably relieved by lighter markings formed by hair or ground-colour, or both ; wings not blackish brown. 4 (5). Dorsum of abdomen with greyish olive median longitudinal stripe lunatus, Fabr. [6]. 5 (4). Dorsum of abdomen not so marked. 6 (7). Dorsum of abdomen with three longitudinal series of clearly defined, pale neutral grey, white haired or whitish haired spots, those of median row triangular Je A be nemoralis, Mg. [7]. 7 (6). Dorsum of abdomen rath so faabltadh 8 (9). Large or very large species, at least 18 mm. in length; dorsum of thorax, including scutellum, score clothed with bright ochreous or yellowish harelvi¢. aie é ‘ gigas, Herbst (4). 9 (8). Small or medium- ee species, not Meocdine 14 mm. in length. 10(11). Second (visible) abdominal tergite in both sexes with a double, quadrate, black or blackish mouse grey mark, occupying median third ‘and partly divided by an indistinct greyish triangle, based on hind margin decorus, Lw. [1]. 11(10). Second (visible) abdominal tergite not so marked. 12(13). Legs in 9 (only sex at present known), except extreme tips of femora, entirely black or slate-black ; no sharply marked contrast between first two abdominal tergites and the two following ones . . insecutor, sp. n. [3]. 13(12). Legs in 2 (only sex at present known) not entirely black, extreme base of extensor surface of front tibiae, and middle and hind tibiae and first jomt of middle and hind tarsi except tips, cream-buff; first two abdominal tergites in 9 greyish olive pollinose, clothed with fairly long ochreous hair, and, especially when viewed from the side, contrasting sharply with next two tergites, which are shining black mendicus, Villen. [5]. 14 (1). Eyes (under an ordinary hand-lens magnifying about 5 diameters) bare or apparently so. *In this table, as in that for the genus Haematopota above, contrasted and mutually exclusive statements and summaries of characters are indicated by the paired numbers on the left, the contrasted conditions in each case being shown, either subsequently or previously, under the second number of the pair (i.e., the one in round brackets). The number in square brackets [ ] after the name of a species indicates the serial position of the species in the ensuing pages. It should be noted that the characters of a given species shown in the table are only sufficient to distinguish it from the other species included in the present synopsis ; they are not necessarily distinctive as regards known species not yet met with in Palestine, but which may ultimately prove to form part of its fauna. Tabanus insecutor, sp. n., is shown twice in the table, since it is included in both the hairy-eyed and bare-eyed categories the very minute hairs on the eyes in the @ being liable to be overlooked on a first examination. (637) F2 294 15(16). 16(15). 17(18). 18(17). 19(20). 20(19). 21(22). 22(21). 23(24). 24(23). 25(26). 96(25). 27(28). ERNEST E. AUSTEN. Blackish species. Dorsum of abdomen slate-black, transversely banded with silvery-white hair; wings with extreme base and proximal two- thirds of costal border ochraceous-tawny, contrasting sharply with remainder of surface .. - a ss és dale, sp. n. [16]. Not blackish species. Dorsum of abdomen not slate-black, but with markings arranged in longitudinal series ; no part of costal border pre- senting a sharp colour contrast to remainder of wing. Conspicuously narrow-bodied, elongate species. a) Frontal triangle in g and subcallus in 2 with shining transverse band, extending from eye to eye .. ». arenwagus, sp. n.[12]. b) Frontal triangle in g and subcallus in 9 without shining transverse band, entirely dull, pollinose .. = accensus, sp.n. [13]. Species of ordinary shape, not conspicuously narrow-bodied or elongate. Dorsum of abdomen with median longitudinal stripe. a) Ground-colour of dorsum of abdomen ochraceous tawny rupinae, sp. n. [11]. b) Ground-colour of dorsum of abdomen blackish brown ; much smaller species than foregoing (length of 9 13-5 to 14 mm.) | regularis, Jaenn. [10]. Dorsum of abdomen without median longitudinal stripe. Legs in Q (only sex at present known), except extreme tips of femora, entirely black or slate-black .. . msecutor, sp.n. [3]. Legs not entirely black or slate-black, at least proximal half of front tibiae, and middle and hind tibiae except tips, pale (cream-buff or ochraceous buff). an Femora cinnamon-buff or pinkish cinnamon .. palladipes, sp. n. [15]. Femora, at least front pair, black or blackish, neutral grey or pale neutral grey pollinose. Mouse grey, dark mouse grey or blackish mouse grey species ; front in 2 broad (its length equal to rather less than four times its breadth at lower margin of lower callus), with two large and completely separate frontal calli, the lower callus large, quadrate, broader than high leleana, sp. n. [14]. Not mouse grey species ; front in Q narrow or of only moderate breadth (its length equal to at least five times its breadth at lower extremity), upper callus represented only, if at all, by a narrow median line running upwards from lower callus, latter narrow, elongate, usually tapering to its upper extremity, two and a half to three times as high as its breadth below. Sides of pale triangles in middle line of abdomen not concave; first posterior cell in wing greatly narrowed (sometimes even closed) at or before its distal extremity ; front in 9 at least seven and a half or eight times as long as its breadth at lower extremity ; venter, except extreme tip, and a broad, median, longitudinal, shining, slightly or not at all infuscated stripe, uniformly light ochraceous-salmon or light ochraceous- buff: large species, at least 19 or 20 mm. in length .. eggeri, Schin. [8]. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE, 295 28(27). Sides of pale triangles in middle line of abdomen usually more or less concave ; first posterior cell in wing, though often narrowed at distal extremity, seldom exceptionally so; front in Q five or five and a half times as long as its breadth at lower extremity ; ground-colour of dorsum of abdomen in Q either russet, black, or blackish grey—if russet, then venter pale smoke grey pollinose orange-cinnamon or pinkish cinnamon, without or with but faintly indicated, broad, dark, median, longitudinal stripe: fairly large or large species, 15 to nearly 19 mm. in length autumnalis, Linn. [9]. 1. Tabanus decorus, Lw. One ¢, three 99, near Jerisheh, 5 miles N.-H. of Jaffa, between 26.iv. and 8.v.1918; one 9, Valley of Jehoshaphat, Jerusalem, 20.1v.1918 (Captain (acting Lt.-Col.) W. J. Dale, O.B.H., R.A.V.C.). | Notwithstanding that the third joint of the antenna in three of the 99 mentioned above (it is wanting in the other two specimens) is somewhat broader than it should be, according to Loew's own statement and the subsequent description by Brauer (Denkschr. k. Akad. Wiss., xlu, 1880, pp. 152-153), there can be little doubt that the determination here given is correct. The eye- markings in this species appear to be in course of disappearance, and to be some- what variable in consequence. Loew and Brauer describe the eyes as unbanded in the g, but as having in the 9 a single, narrow, greenish-yellow transverse band, or an indistinct trace of such a band. The ¢ mentioned above, and one of the © 9 taken at Jerisheh, had in life green eyes, each marked with a single, faint, narrow, median, golden horizontal band ; while another 9 captured at the same spot had bronze-green eyes, each with three faint, golden, transverse bands, the middle one being the more distinct. It may be added that the type of Tabanus decorus was taken in Syria, while Brauer (loc. cot.) records the species as occurring in the Lebanon. 2. Tabanus alexandrinus, Wied. (fig. 5.). Five $d, three 99: near Jerisheh, 5 miles N.-E. of Jaffa, 26.iv.1918 (one J beginning of May). Although the brevity and lack of precision of the original description (Aussereurop. zwerft. Ins., ii, 1830, p. 624) of this entirely black species would render identifi- cation difficult, there can be no doubt, in spite of a slight discrepancy as regards the coloration of the eyes, that the specimens referred to above really belong to T. alecandrinus, as characterised and redescribed by Brauer (Denkschr. k. A kad. Wiss., Bd. xlii,1880, p. 161, Taf. ii and v, figs. 23). This author records the species as occurring in Syria and Italy, and states that Wiedemann’s collection contains examples of it from Alexandria. A distinctive character of the typical form of T. alexandrinus, especially in the Q, is the shape of the expanded portion of the third joint of the antenna, the upper margin of which is convex, or at least shows no trace of the usual excavation (fig. 5). The expanded portion of the third antennal joint varies in width in different individuals, and is sometimes, especially in the 3, a good deal narrower than it appears in the figure. Macquart (Hist. Nat. Ins., Dipt., i, 1834, p. 199), who redescribed the species under the name Tabanus 296 ERNEST E. AUSTEN. carbonatus, from a specimen from Sicily, notes the shape of the upper margin of the third antennal joint. According to Brauer (loc. cit.), the eyes in both sexes are dark green; in the specimens collected near Jerisheh, however, the eyes of § and Q in life were unicolorous black. The Museum collection contains several examples (2 Jg, 3 99), from Algeria and Tunis, of what must, at any rate provisionally, be regarded as a variety or local race of the present species.* This western form is distinguished from the typical one, as represented by Palestine specimens, by the absence of the greyish- white hairs on the lateral margins of the dorsum of the thorax in the 9, on and in front of the postalar calli, and by the distal extremity of the expanded portion of the third joint of the antenna being distinctly emarginate above. This is at any rate the case in the 9; in the two §¢ from Algeria and Tunis the third joints of the antennae are unfortunately missing. Brauer’s record of Tabanus alexandrinus from Syria, referred to above, gives no more precise indication of locality, but Bezzi (Brotéria, Ser. Zool.,vii,1909, p. 42) records the species as having been taken at Ghazir (about 15 miles N.-E. of Beirut), on 14th June, 1904. It is interesting to note that clinging to the hairclothing the front coxae of one of the 3 specimens taken near Jérisheh on 26th April 1918, were six little, hexapod, Campodea-like larvae, which Dr. Gahan states are the “triungulin” stage of an oil-beetle (Meloe sp.). 3. Tabanus insecutor, sp. n. (fig. 6). Q.—Length (1 specimen) 12°25 mm.; width of head 4 mm.; width of front at vertex 0°6 mm. ; length of wing 10°25 mm. Eyes unbanded, at first sight appearing bare, when more closely examined seen to be sparsely clothed with very short, very minute hairs ; dorsum of thorax shining blackish mouse-grey pollinose; dorsum of abdomen shining black, clothed partly with minute appressed black hairs, and on hind borders of second and following segments with appressed glistening whitish or pale ochreous hair, first and second tergites and hind borders of two following segments greyish pollinose ; frontal callus of 9 large, elongate, shining black ; palpi dark neutral grey, terminal segment in Q slender ; first two joints of antennae dark neutral grey, third joint black, expanded portion broad, its upper margin angulate; wings lightly tinged with brownish drab, extreme base and costal border as far as stigma ochre-coloured, a small, ill-defined, oblique, faint cinnamon- brown blotch extending from narrow, elongate, ochraceous-tawny stigma to discal cell ; legs, except extreme tips of femora, which are cinnamon-buff, entirely black or slate- black. *The ¢ genitalia of this North African form have been examined microscopically and compared with those of the typical race, as found near Jerisheh; a similar comparison has been made in the case of the infra-anal plate of the 9. The differences noted in each instance appear to be merély trivial, and, so far as the limited material available for study permits a definite conclusion to be drawn (a pair of specimens of each form is all that can at present be devoted to dissection), cannot be regarded as anything more than individual. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 297 Head light neutral grey pollinose, face and jowls somewhat thinly clothed with whitish hair, longer in basi-occipital region, hind margin of occiput fringed above and at sides with very short, pale yellowish hair ; front in 2 of medium and entirely uniform width, about three and a half times as long as broad, clothed with minute, appressed, glistening ochreous or pale yellowish hair, ground-colour slightly darker in region of vertex but no trace of an ocellar tubercle; lower and upper frontal calli fused together to form a single, elongate, shining black callus, extending from level of inner angles of eyes to a point slightly more than one-third of length of front from hind margin of vertex; callus exhibiting a trace of a fine, elongate, median groove, and separated from eye on each side by an interspace equal to rather less than half its width, lower margin of callus straight, upper extremity ogival, the middle region somewhat constricted, so that outline of entire callus is suggestive of that of an Ancient Greek, leaf-shaped sword ; eyes of 2 in life plain dark green, unbanded ; proximal segment of palp: clothed with whitish hair, distal segment in 2 acuminate, sparsely clothed on outer surface with minute, appressed, yellowish ‘hairs ; first two joints of antennae clothed above and below with short black hairs, first jomt not swollen and not embracing second joint, upper distal angle of latter moderately produced, annulate portion of third joint about one-fourth shorter than expanded portion, depth of latter, from centre of lower margin to apex of angle on upper border, about equal to its length, angle, which bears a few exceedingly minute black hairs, situate in centre of upper border, which is slightly emarginate in front of it. Thorax: dorsum (including scutellum) and pleurae clothed with pale ochreous or pale yellowish hair, the hairy covering of the pleurae almost whitish at the tips, lateral borders of dorsum and commencement of two longitudinal stripes extending a short distance from front margin cinereous pollinose. Abdomen: first and second (visible) tergites clothed with glistening appressed, pale ochreous hair, the hair forming a tuft in the centre of the hind border of the first tergite paler, and that on the hind border of the second tergite glistening whitish ; glistening, appressed, pale hair on hind borders of fifth and sixth segments extending forwards in middle line in each case in form of a broad triangle, apex of which is almost or quite in contact with hind margin of preceding seoment; sides of sixth tergite and entire surface of seventh clothed with glistening, appressed, pale ochreous or pale yellowish hair ; ground colour of seventh tergite cinereous pollinose ; venter blackish slate-coloured, neutral grey pollinose but moderately shining, clothed with short, glistening, appressed, pale ochreous or pale yellowish hair, extreme hind margins of second and following scutes cream- buff. Wangs: proximal portions of first and fifth longitudinal veins ochraceous tawny, bases of second and sixth veins ochreous, veins otherwise dark mummy- brown or dark sepia-coloured, anterior branch of third longitudinal vein without an appendix. Sguamae light buff, frmged with pale hair. Halteres: knobs dark seal-brown above and below, stalks and tips of knobs cream-buff or cinnamon- buff. Legs: coxae neutral grey pollinose, both they and femora clothed with fine, yellowish hair ; tarsi and tips of tibiae, as also inner sides of front pair of latter, clothed with minute, appressed, black hairs, tibiae elsewhere clothed with short, glistening, appressed, yellowish-white, pale yellowish, or pale ochreous hair ; front tibiae slightly thickened. 298 ERNEST E. AUSTEN. Near Mulebbis (Jafia district), 14.v.1918 : caught by Lt.-Col. H. P. Sewell, C.M.G., D.S.0., R.A.M.C., from a light Ford motor-car (which the insect was pursuing) travelling at 20 miles an hour across a grassy plain ; more than one specimen seen. Apart from its smaller size, the species just described presents a superficial resemblance to Tabanus aterrimus var. auripilus, Mg., but is distinguishable, anter alia, by the elongate frontal callus of the 2, and by the conspicuously ochreous coloration of the base of the wing. Fig. 5. Tabanus alexandrinus, Wied., head of @ (typical form) in profile. X about 10. ‘ Vi Fig. € Tabanus msecutor, Austen, 9: a, head from in front; b, head in profile, showing shape of antenna. x 10. Labanus insecutor is allied to T. umbrinus, Mg., with which it agrees in the character of the frontal callus of the 9, the coloration of the eyes, and the shape of the terminal segment of the @ palpi; the new species may however be distinguished, inter alia, by the much greater development of the pale yellow hair CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 299 on the abdomen, particularly on the venter, and by the absence of the dark brown costal border on the wing, extending from the base of the wing to the end of the stigma, and of the broad dark brown blotch behind the latter. 4. Tabanus gigas, Herbst. Two $3: one near Jerisheh, 5 miles N.-E. of Jaffa, 16.v.1918, hovering above low hill, 9°15 a.m. (one other specimen seen) ; the other in marsh 1 mile S.-H. of Tel Abu Zeitun (within 14 mile of spot at which foregoing specimen was taken), 18.v.1918, on flower (one other example seen). This strikingly handsome species, in which, as shown by the first of the two individuals recorded above, the eyes in life are purplish black, without bands, is already represented in the British Museum collection by, among other specimens, a ¢ from Galilee (B. T. Lowne, 1863-64). Statements by Brauer (Denkschr. k. Akad. Wiss., Bd. xlii, 1880, p. 165) show that Tabanus gigas is found in France, Austria, Greece, Sicily, Russia, Syria and Asia Minor, while the series in the Museum includes specimens from Italy and Bulgaria. In length, the two individuals taken by the writer measure respectively 19 and 18°5 mm.; their dimensions are therefore somewhat less than those given by Brauer (loc. cit.), according to whom the length of the ¢ is 20 mm., while that of the 2 may be as much as 23 mm. The variability of Tabanus gigas as regards the coloration of the hairy covering of the head, body, and legs, according to the country of origin, is dealt with at some length by Loew (Verh. z.-b. Ges. Wren, Bd. viii, 1858, pp. 582-583), who refers to the species under its synonym T. albipes, Fabr. Although the two specimens recorded above as having been caught respectively on 16.v. and 18.v.1918 were captured within less than a mile and a half of each other, they nevertheless show considerable colour-differences of the kind indicated. While in the case of the ¢ taken in Tel Abu Zeitun marsh the pale hair on the head, thorax, tibiae and first two segments of the abdomen is of a rich ochreous tint, the corresponding hair in the case of the other specimen is much lighter in hue, and on the tibiae and the hind border of the second abdominal segment is even almost white. The individual with the paler hair also shows a median triangular patch of whitish hair on the hind border of the third abdominal tergite, and a smaller number of yellowish or ochreous hairs in the corresponding position on the following three tergites. These patches of pale hairs are entirely wanting in the other specimen, in which the paler fringe to the anal region is represented by but a very small number of ochreous hairs. 5. Tabanus mendicus, Villen. One ©: Ain Sinia, 10 miles N. of Jerusalem, 18.iv.1918, on bank of stream, only specimen seen (Captain K. B. Williamson, R.A.M.C.). In the living insect the eyes were bronze, with three horizontal, dark purple bands, the lowest the broadest, the middle one narrow. The type of this somewhat unusual-looking species (a 9, taken in the Oasis of Damascus, in the second fortnight in April) is not inaptly compared by Villeneuve (Bull. Soc. Amis Sc. Nat. Rouen (5) xlvii, 1912, pp. 41-42) to a tiny example of Tabanus (Atylotus) gigas, Herbst. The specimen from Ain 300 ERNEST E. AUSTEN. Sinia, recorded above, would seem to be only the second individual of Tabanus mendicus met with up to the present time. It is perhaps worth mentioning that, when viewed from the side, the insect appears to have a broad, shining, black band across the middle of the abdomen, owing to the third and fourth segments being clothed mainly with black hair, and thus contrasting with the base and distal extremity, on which the hairy covering is honey-yellow or cinnamon-buff, while the contrast is heightened by the ground-colour of the first two segments being neutral grey or olive grey pollinose. 6. Tabanus lunatus, Fabr. Jaffa district and Jerusalem: three 99, near Mulebbis, 14.v.1918, caught by Li.-Col. FE. P. Sewell, C.M.G., D.S.0., R.A.M.C., from a light Ford motor-car (which the insects were following), at the same time and under the same circumstances as the type of Tabanus insecutor, Austen, see p. 298 (car proceeding ~ at 20 miles an hour across a grassy plain; several specimens seen); a fourth 9, in the British Museum collection, bearing the label “Jerusalem. A. Béttcher, Berlin,’ received in 1912 from the Zoologisches Museum, Berlin, and presented by Miss G. Ricardo. The four specimens enumerated above, which are the only representatives of this small or medium-sized, greyish-olive-coloured, hairy-eyed species at present contained in the National Collection, show that, in the female sex at any rate, Tabanus lunatus, Fabr., is subject to considerable variation in size, the three examples brought home by the writer measuring from 10°75 to 11°75 mm. in length, as compared with 13:2 mm. in the case of the Q labelled “‘ Jerusalem.” The eyes of the 9 of 7. lwnatus, which are thickly clothed with short, fine, whitish hair, are bronze-green in life, with three narrow, purplish brown, transverse bands. The description and figures of the head and palpus of the 2 given by Brauer (Denkschr. k. Akad. Wiss., Bd. xlii, 1880, p. 172, Taf. ui, figs. 34)—who records the species as occurring in Dalmatia, Italy, Sicily, Spain, Asia Minor and the Caucasus—are somewhat misleading, and since in the preceding synoptic table (J. cit., p. 134) the upper frontal callus is characterised as “ linear, connected with the lower callus . . .”’ it would be difficult if not impossible to determine female specimens of 7’. lunatus by means of the Austrian author’s well-known work on “The European Species of the Genus Tabanus.” A careful examination and comparison of the specimens brought back by the present writer shows that the upper frontal callus in the Q of 7. lunatus, instead of being “ elongate, more or less linear or fusiform ” (Brauer), is roughly oval or elliptical oval in outline, and sometimes much larger than the lower callus, with which it is not connected ; also that the terminal segment of the palpus in the same sex is acuminate, and only moderately swollen at the base—much less so than would be imagined from Brauer’s description and figure. The sharp colour-contrast in the third joint of the antenna, owing to the expanded portion being cinnamon-rufous and the annulate portion dark seal-brown or black, is very noticeable. 7. Tabanus nemoralis, Me. Two 929, Mount of Olives, 20.iv.1918 (#. #. A.), and 4.v.1918 (Captain (acting Li.-Cal.) W. J. Dale, O.B.E., R.A.V.C.), both specimens taken “ on ma.” CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 301 In life, the eyes of the Q of this species are purplish bronze, with three dark purple horizontal bands, the middle one of which is the narrowest, across the centre. Tabanus nemoralis is recorded by Brauer (Denkschr. k. Akad. Wiss., Wien., Bd. xlu, pp. 175, 208-210 (1880)) as occurring in France, Switzerland, Italy, Sicily and North Africa ; the British Museum collection contains specimens from Switzerland (Upper Engadine), and Algeria (Rouiba, May-June, 1911, Baron J. R. M. Surcouf). 8. Tabanus eggeri, Schin. One 9, Wilhelma (Jaffa district), 30.vu.1918, resting inside small, open shed, at back of house in village ; one Q, near Jerisheh, 5 miles N.H. of Jaffa, 26. viii. 1918, in tent, 7.45 a.m. (Zt.-Col. (temp. Col.) E. P. Sewell, C.M.G., D.S.O., R.A.M.C.) ; one 9, near Yahudiyeh (Jaffa district), 1.1x.1918 (Captain (acting Major) W. S. Corfield, R.A.M.C. (T.)). In this species, as was shown by the specimen taken near Jerisheh by Colonel Sewell, the eyes in life are purplish brown, with a bronze sheen, and are unbanded. Tabanus eggert, which is already represented in the British Museum collection, by, among others, specimens from Gibraltar, Italy and Algeria, is recorded by Brauer (Denkschr. k. Akad. Wiss., Bd., xlii, 1880, p. 184—under the pre-occupied name T. intermedius, Egg.) as occurring in Southern France, Spain, Switzerland, Sicily, ‘Corsica, Egypt and Asia Minor. As an aid to recognition it may be mentioned that, in the species under consideration, the diameter of the first posterior cell is greatly contracted on the hind margin of the wing. According to Brauer (é. cit., p. 183), the hind tibiae are fringed on the outer side with black hair ; the specimens from Palestine, however, show a slight variation from the typical form, in that the external fringe on the hind tibiae consists mainly of ochreous or ochraceous- tawny hair. 9. Tabanus autumnalis, Linn. One ¢, Baharet Katurieh, near El Jelil (about 10 miles N.E. of Jaffa), 28.1v.1918, only specimen seen; two 99, Ain es-Sultan, near Jericho, 22.iv.1918—one taken on a horse, the other, as also a third specimen, which was not captured, resting on face of excavations. This is a very widely distributed species, which, according to Brauer (Denkschr. k. Akad. Wiss., Bd. xlu, 1880, p. 193), occurs throughout Central and Southern Kurope, and is also found in Holland, Sweden, England, S. Russia, Poland, Corsica, Corfu, Syria and Asia Minor. Loew (Verh. z.-b. Ges. Wien, Bd. viii, 1858, p. 605), who draws attention to the variability of T. autwmnalis, like that of so many other species, as regards both size and coloration, according to locality, records its range as including “ the whole of Europe and a great part of the Near East.”” In addition to specimens from Hungary, Bulgaria, England and elsewhere, the species is represented in the British Museum (Natural History) by a Q from Algeria (Biskra, 18.v.1893, Rev. A. H. Eaton) and two 9° from Cyprus (one collected by P. Gennadwus, the second specimen taken by Miss D. M. A. Bate, near Ktima, between 3.v. and 15.vu.1901). It may be added that the three last-mentioned examples belong to the form of the species met with in Palestine, and also recorded by Loew (loc. cit.) as occurring in Cyprus, in which the ground-colour of the dorsum of the 302 ERNEST E. AUSTEN. abdomen in the @ is russet instead of black or blackish mouse grey, while the venter in the same sex 1s pale smoke grey pollinose orange-cinnamon or pinkish cinnamon, and the usual broad, dark, median longitudinal ventral stripe is absent or but faintly indicated in both sexes. Two other Algerian 29Q of this species (taken respectively at Algiers and Hammam R’Irha, in May 1908, by the Hon. W. (now Lord) Rothschild), 11 the National Collection, belong to the typical form. Loew (loc. cit.) remarks that South European examples of Tabanus autumnalis are on the whole decidedly smaller than specimens from Germany and Northern Europe. The respective lengths of the three individuals from Palestine, recorded above, are 15°75, 15, and 19°5 mm.; thus it will be seen that, as compared for instance with English representatives of the species under discussion, the two former are markedly subnormal in size. 10. Tabanus reguiaris, Jaenn. One 2, in Wadi el Kelt, Jericho Plain, 1.vi.1918. The eyes of this specimen in life were dark purplish brown, without bands. Tabanus regularis, which was described from a specimen from 8. France (Marseilles district), and is included by Brauer among the Tabanidae of Italy and Greece, has recently been recorded by Becker (Ann. Mus. Zool. Acad. Imp. St. Pétersbourg, xvii, 1912, p. 596) as occurring in Persian Baluchistan. 11. Tabanus rupinae, sp. n. (figs. 7, 8). ¢ 9.—Length, 3 (2 specimens) 14°5 to 15-4 mm., 2 (3 specimens) 15 to 16-2 mm. ; width of head, ¢ 5°4 to 5°6 mm., 9 5:2 to 5°6 mm.; width of front of 9 at vertex 0°75 to 0°83 mm. ; length of wing, § 11:25 to 12 mm., 2 12°6 to 13°25 mm. Dorsum of thorax light greyish olive (deep greyish olive when rubbed) pollinose, clothed with fine, silky, appressed as well as erect, ochreous har, mingled with fine, erect, blackish hairs, the erect ochreous hair whitish at the tups; dorsum of abdomen ochraceous-tawny, with a pale, median longitudinal stripe running from base of second. to hind margin of penultimate tergite ; anterior branch of third longitudinal vein abruptly angulate at the base, and provided with a long appendix; front femora dark mouse- grey (smoke-grey pollinose on outer side), front tarsi and distal halves of front tibiae black. Head: face and jowls pale olive-grey in both sexes, clothed with whitish hair ; occiput pale olive-grey in g, smoke-grey or pale smoke grey in 9, hind margin of occiput in region of vertex clothed with very short whitish or pale yellowish hair; frontal triangle in § smoke grey or pale smoke-grey pollinose; front in Q olive-buff pollinose, moderately broad and of practically uniform width, its length equal to about four and a half times its breadth between inner angles of eyes, clothed with short, appressed, ochreous hairs, mingled in region of vertex with minute blackish hairs; lower frontal callus in 2 blackish brown or sepia-coloured, transversely oblong, its depth equal to about three-fourths of its breadth, narrowly separated from eye on each side, its lower margin situate just above level of inner angles of eyes; upper frontal callus in Q represented by a somewhat ill-defined, dark brown, crescentic mark, with long horns extending upwards, but its precise CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 303 shape evidently dependent upon degree of rubbing to which the specimen has been subjected ; ocellar tubercle in § very small and inconspicuous, situate at bottom of cleft between eyes at vertex, no trace of ocellar tubercle in 2; subcallus in Q pinkish buff pollinose, with a more or less distinct indication of a tawny-olive or light mummy-brown transverse band extending from eye to eye at level of antennae ; eyes under an ordinary hand-lens appearing bare in both sexes (closer scrutiny sometimes reveals scattered minute hairs on lower portions), eyes in $ each with a bluntly triangular area of greatly and abruptly enlarged facets, partly encircled with a sharply contrasted zone of small facets occupying approximately lower fourth and outer and upper border, latter being continued to vertex in practically its full width ; eyes of 9 in life dark green, with purplish-brown sheen and without bands; palpi of ¢ clothed with whitish hair, terminal segment cream-coloured, greatly swollen, ovate pyriform and ending in a small curved point, proximal segment of 9 palpi clothed with fairly long, whitish hair, terminal segment of 2 palpi long SSSA Se SY Fig. 7. Tabanus rwpinae, Austen, 9. xX 4. acuminate, considerably swollen at base, cream-coloured or ivory-yellow, clothed on outer side with short, appressed, glistening hair of similar colour (occasionally intermixed with a few minute black hairs towards distal extremity), and with longer whitish hair at base below; first joint of antennae cinnamon-buff, cream- buff pollinose, clothed with minute, appressed, glistening yellowish or whitish hair, with longer pale hair below, and with minute black hairs on upper distal angle, considerably swollen and expanded distally and partly embracing second joint, latter agreeing with first joint in coloration and in its hairy covering and with its upper distal angle considerably produced, third joint cinnamon, annulate portion infuscated at tip, at least in dried specimens, expanded portion narrow and elongate in g, of considerable depth in 9, with a prominent but blunt angle near base of upper margin in both cases. . Thorax: anterior border of dorsum in certain specimens showing more or less distinct vestiges of three pallid neutral grey 304 ERNEST E. AUSTEN, longitudinal stripes ; swelling in depression at each end of transverse suture drab- grey or light drab ; pleurae and pectus pale smoke-grey, clothed with whitish hair. Abdomen: pale, median, longitudinal stripe on dorsum sometimes, especially in §, formed by a linear series of anteriorly directed, truncated triangles ; median stripe, at least on third (visible) and following tergites, bordered on each side with a dusky, oblong blotch, extending for the full length of the segment except extreme hind border, and clothed with minute, appressed, black hairs (like rest of hairy covering of dorsum very easily removed by rubbing, in which case the blotches are difficult to distinguish) ; on each side of third and two following tergites, outside dusky blotch just described, a more or less distinct trace of another pale stripe or oblique pale mark, sometimes bordered on outer side on fourth and fifth tergites, at least in case of g, with another dusky, black-haired blotch, not reaching hind margin ; sixth and seventh tergites, except lateral and hind borders of former and its median stripe, dark brown and clothed with minute, appressed, black hairs (posterior angles of sixth tergite in g, and hind border of seventh in both sexes Fig. 8. Tabanus rupinae, Austen : a, head of gin profile ; b, head of 9 from in front, x 10; b’, antenna of 9° from the side, greatly enlarged. clothed with longer black hair); dorsum, except as already stated, clothed with minute, appressed, glistening, ochreous hairs, longer and paler on hind borders of segments, extreme lateral borders of tergites pale drab-grey pollinose and clothed with longer whitish hair; ventral scute of first (visible) segment pale drab-grey pollinose, remainder of venter except terminal segment pale cinnamon or pinkish cinnamon, covered with cream-buff pollen and clothed with whitish or appressed, glistening, pale yellowish or pale ochreous hair, hind borders of segments cream- buff or pale drab-grey pollinose, clothed with longer pale hair; ventral surface of terminal segment mouse-grey or deep mouse-grey, clothed as usual with coarse, erect, black hair, in case of ¢ sometimes a few similar hairs on ventral surface of preceding segment also. Wungs suffused with light drab; stigma inconspicuous, elongate, pale tawny-olive; veins lighter or darker sepia-coloured. Squamae cream-buff, borders cinnamon-buff. Halteres cream-bufi or cream-coloured, knobs CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 305 sometimes darker (seal-brown or light seal-brown) at base. Legs: coxae pale drab-grey or smoke-grey pollinose, clothed with whitish hair; femora and tibiae of middle and hind legs, and proximal halves or rather less than proximal halves of front tibiae, cinnamon-buff or pinkish cinnamon-coloured, middle and hind femora drab-grey pollinose ; middle and hind tarsi dark brown above, first joint of middle tarsi sometimes paler towards base; third and fourth joints of middle and hind tarsi distinctly expanded, rather broad; femora clothed with whitish hair ; middle and hind tibiae and proximal halves of front tibiae clothed with short, glistening, appressed yellowish-white hair, which on anterior margin of outer side of hind tibiae is elongated so as to form a fringe ; upper surfaces of tarsi and distal halves of front tibiae clothed with minute black hairs, extensor surfaces of middle and hind tibiae, especially towards tips, also largely clothed with similar hairs. Wadi el Kelt, Jordan Valley, near Jericho: type of ¢ and one para-type of the same sex, 5.vi.1918, on cliff wall, 5.30 p.m.; type of 9 and one Q para-type L.vi. 1918; an additional 2 para-type 25.v.1918, resting on the author’s tent (pitched above the Wadi, close to the point at which the old Jerusalem-Jericho road enters the Valley), 6.30 a.m. On several evenings at the beginning of June 1918, towards sunset, a number of specimens of this species, which was not met with elsewhere, were found resting on the precipitous earth cliffs forming the sides of the Wadi el Kelt, just where the wild gorge in question leaves the hills. Tabanus rupinae, which is not closely allied to any of its congeners known to the writer, presents a very slight superficial resemblance to the Mesopotamian T. polygonus, Walk., but is distinguishable at once, inter alia, by the somewhat wider front in the 9; by the lower frontal callus in the same sex being transversely quad- rate, instead of vertically ogival and prolonged above into a narrow line; _ by the second abdominal tergite in both sexes being without any trace of a blackish median spot at the base; by the anterior branch of the third longitudinal vein being abruptly angulate at the base and provided with a long appendix; and by the very different coloration of the front legs. 12. Tabanus arenivagus, sp. n. (figs. 9, 10). ¢ 9—Length, 3 (3 specimens) 13°5 to 14 mm.; © (10 specimens) 12 to just over 14 mm.; width of head, ¢ 4°5 to 4°75 mm., 9 3°6 to 4:4 mm. ; width of front of 9, at vertex 0°5 to 0°75 mm., between inner angles of eyes 0-4 to 0°5 mm. ; leneth of wing, ¢ 105tollmm.; 910 to 11:2 mm. Small, light drab-coloured, narrow-bodied species, with sides of abdomen straight, tapering from base of fifth segment in 3, and from that of second segment in 2; eyes under an ordinary hand-lens appearing bare (merely showing sparse microscopic hairs below), in 3 with a sharply contrasted area of greatly enlarged facets, and two purplish-brown transverse bands, in 2 with three sinularly coloured bands : frontal treangle in 3 and subcallus in 2 with a shining transverse band, extending from eye to eye; & with well-developed though sunken ocellar tubercle, of which there 1s no trace m 2; latier with two frontal calla ; third joint of antennae ochraceous-buff or ochraceous- orange, narrow, elongate and tapering, its upper margin not excavated but with a blunt 206 ; ERNEST FE, AUSTEN. angle near base ; dorsum of abdomen with four longitudinal rows of oblique, elongate, mummy-brown markings, more distinct in 2; wings hyaline, stigma brownish olive, elongate and distinct, anterior branch of third vein without an appendia. Head conspicuously enlarged in ¢; face and jowls in both sexes whitish pollinose and clothed with whitish hair, a faint, not always distinguishable infuscation between base of each antenna and corresponding eye; occiput pallid neutral grey pollinose, its upper margin fringed behind with erect, pale yellowish hair, very short in 9, longer and somewhat curved forward in 3, frontal triangle in ¢ and front and sub-callus in 9 pale smoke grey or smoke grey pollinose, front in 2 clothed with minute, appressed, pale ochreous hairs, and its ground-colour somewhat darker (mouse grey) in vertical region; shining transverse band on frontal triangle in ¢ nearly twice as deep as corresponding band on subcallus in © and much darker (clove brown), its upper margin level with upper border of area of small facets forming rather less than lower third of eyes, and its lower NY Fig. 9. Tabanus arenivagus, Austen, 9. X 4. margin separated from base of antennae by a space equal to about half the depth of the band ; [corresponding band in @ brownish olive or isabella-coloured, and occupying rather more than upper half of subcallus; front in 2 of medium width, distinctly narrower below, neither callus in contact with eyes, lower callus quadrate, shining, light sepia-coloured and situate well above inner angles of eyes, upper callus much larger and darker (seal brown or dark brown) than lower, not shining, quadrate, situate just below midway between vertex and lower callus, with which it is not connected, and generally with a more or less distinct longitudinal groove or depression in middle lme; area of enlarged facets in eye of 3 sharply defined, bluntly triangular in outline, and occupying rather more than upper two-thirds, with exception of a narrow hind border of uniform width; eyes of 3 in life with area of enlarged facets pinkish isabella-coloured above, light grey on lower border: area of small facets with hind border dark brown, with greenish reflections, lower border greenish-bronze, with two narrow, purplish-brown, CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 307 transverse bands, extending to hind margin, and, below these, a very fine, and shorter darkish, horizontal, transverse streak ; eyes of 9 in lafe greenish-bronze, with three horizontal, purplish-brown bands, and a very narrow dark streak above and below ; palm, pale cream-buff, those of g and proximal joint of those of 9 clothed with long whitish hair, distal jomt of Q palpi clothed on outer surface with glistening, appressed, pale yellowish hairs mixed with minute, appressed, black hairs, and bearing longer pale yellowish or whitish hairs on under surface ; distal joint of ¢ palpi elongate ovate or ovate pyriform, corresponding joint in Q curved, tapering, pointed at tip, not greatly swollen at base; first and second joints of antennae cinnamon-buff, clothed on outer side with pale yellowish hair, and on distal margin of second joint and upper distal angle of first joint with minute black hairs, first joint somewhat greyish pollinose, moderately swollen in g, upper distal angle of second joint not produced, annulate portion of third joint straight. Thorax: dorsum with four dark mouse-grey or fuscous longitudinal stripes, separated by narrower drab-grey or pale drab-grey interspaces, lateral borders pale drab-grey or pale smoke-grey—swelling occupying depression at each end of transverse suture pale brownish drab ; dorsum of $ clothed with fine, erect, pale yellowish or whitish hair, mixed with short, glistening ochreous, appressed hairs, and sometimes also interspersed with darker erect hairs, post-alar calli and swelling immediately above base of each wing in ¢ clothed with whitish hair ; dorsum of @ clothed with short, appressed, glistening pale yellowish or ochreous hair, mixed with a certain number of short and more erect blackish hairs, hair on postalar calli and on swelling above base of wing on each side as in 3; pleurae and pectus in both sexes pale smoke- grey, clothed with whitish hair. Abdomen: pattern of markings on dorsum as shown in fig. 9, mummy-brown markings on second (visible) and following segments starting from base of segment but not reaching hind border, so that on each segment ‘ the drab-coloured markings, consisting of a median triangle and an oblique elliptical ovate spot between latter and each lateral margin, are as a rule connected together posteriorly ; on fifth and sixth segments, however, mummy-brown markings on each side are often themselves connected together in each case, in such a way as to enclose the drab-coloured ovate spot; lateral margins of dorsum pale smoke-grey ; dorsum clothed with minute, appressed, shining pale yellowish hair (in case of 9, with minute, appressed black hairs on the dark markings), lateral margins in ¢ clothed with longer whitish hair, seventh segment m © and sides of fifth, sixth and seventh segments in ¢ with longer blackish hairs ; venter drab-grey (hind borders of segments cream-buff), clothed with short, appressed, pale yellowish hair, seventh segment also with the usual relatively coarse, erect, black hair. Wings: veins pale sepia-coloured, here and there, especially anterior transverse vein and base of anterior branch of third vein, distinctly darker. Squamae smoke-grey, borders paler. Halteres : knobs seal-brown (cream-bufi at extreme tips), stalks cream-buff. Legs: light buff, femora greyish pollinose, front pair sometimes streaked with mouse- grey above or on outer side, tips of tibiae and upper surface of middle and hind tarsi (at least distal extremities of joints of latter) sepia-coloured, front tarsi clove brown (first jomt in 9 sometimes paler except at tip), slightly expanded in both sexes, third and fourth joints of middle tarsi in g§ also somewhat expanded ; femora clothed with whitish hair, long and fine in case of 4; tibiae clothed with (637) val 308 ERNEST E. AUSTEN. glistening pale yellowish hair, which on outer side of hind pair forms an out- standing fringe (longer in J), which is usually mixed with black hairs; tips of tibiae and upper surface of tarsi clothed with minute black hairs, similar hairs also present to a varying extent on outer surface of tibiae, above tips. S. Palestine : El Fukhari (8 miles W. of El Shellal, Wadi Ghuzze) and vicinity, September-October, 1917. Type of g, Hl Fukhari, 21.ix.1917 (Li.-Col. P. 8S. Lelean, C.B., C.M.G., R.A.M.C.); type of 2, same locality, 27.ix.1917 (entered author's tent at dusk, attracted by lamp); eleven para-types as follows:—1 3, near Wadi Ghuzze, 25.1x.1917 (Zt.-Col. E. P. Sewell, C.M.G., D.S.O., R.A.M Cc); Pee Fukhari, 4.x.1917 (Col. J. A. Stewart, R.A.O.C.); 8 29, El Fukhari, 16.ix—7.x. 1917; 1 9, Weli Sheikh Nuran, 22.ix.1917 (Lt.-Col. Arnold, A.A.M.C.). Tabanus arenwvagus, which was not met with outside the Fukhari area, was fairly common during the latter half of September and early part of October, 1917, in XXth Corps Headquarters Camp at E] Fukhari, where it was the only Tabanus seen. The spot, though sparsely dotted with low scrub, was a perfectly dry, sandy Fig. 10. Tabanus arenivagus, Austen: a, head of gin profile ; b, head of 9 from in front, x 10; b’, antenna of 9 from the side, greatly enlarged. one, entirely waterless, on the edge of a desert region, and there was no obvious breeding-place whence the flies could have come, nearer than an occasional damp hollow among the sand-dunes near the coast, several miles further west. The author twice took QQ inside his tent at dusk, and at the same time on 22.1x.1917 was bitten on the hand by a third specimen in the open; on another occasion (26.1x.1917) an officer was attacked by one of these flies, while sitting inside a mess- shelter constructed of reed matting. In size, general appearance, general coloration, and in the pattern of its abdominal markings, Tabanus arenivagus somewhat resembles T. pulverifer, Walk., the typical series of which was obtained at Baghdad. The species described above, however, is distinguished from 7. pulverifer, inter alia, by the presence of a shining band on the frontal triangle of the gj and subcallus of the 9; by the lower frontal callus in the @ being paler, and usually smaller and Jess sharply defined ; by the presence in the 9 of a relatively large, well-marked, subquadrate CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 309 upper (median) frontal callus, instead of the ill-defined transverse band, which represents it in the case of 7’. pulverifer ; by the distal segment of the palpi in both sexes being narrower (less sharply pointed in the 9) and deeper in tint, and in the ease of the 2 bearing many black hairs on the outer surface; by the first joint of the antennae in both sexes being smaller, and the third joint paler, narrower, and much more elongate; by the wing-stigma being more sharply defined, and by there being no trace of an appendix to the anterior branch of the third vein. Tabanus arenivagus likewise resembles an undetermined species represented in the British Museum collection by a solitary 9 from Arabia (Muscat), but is distinguished by the front in the @ being more distinctly narrower below (the inner margins of the eyes being convergent) ; by the upper portion of the sub- callus in the @ being shining, and the lower frontal callus much smaller, and not reaching the inner angles of the eyes; by the smaller size (less prominent upper angle) of the first joint of the antennae ; by the dorsum of the abdomen having a longitudinal series of pale triangles in the median line, instead of a broad, pale, longitudinal stripe of more or less uniform width ; and by the stigma of the wing being more distinctly marked. 13. Tabanus accensus, sp. n. (figs. 11, 12). °.—Length (3 specimens) 12 to 13°4 mm.; width of head 3°8 to 4:25 mm. ; width of front at vertex just under 0°6 to 0°6 mm. ; length of wing 9 to 9°75 mm. Small, greyish species, with two frontal calli in Q; bare eyes ; dorsum of thorax neutral grey or deep neutral grey, longitudinally strvped with pale olive-grey ; some- what narrow straight-sided abdomen, smoke-grey or pale pinkish buff above, with four longitudinal series of elongate blackish brown marks, lateral borders of segments pale smoke-grey ‘pollinose, hind borders of second and following segments cream-buff ; and parti-coloured legs. Head: front in 9 smoke-grey pollinose, subcallus pale pinkish-buff pollinose, front clothed with short, pale yellowish or bright ochreous appressed hair, in region of vertex sometimes also with short blackish or dusky hair, no trace of an ocellar tubercle but a tiny, dusky, mouse grey or deep mouse grey spot in middle line in region of vertex; face and jowls greyish white pollinose, clothed with whitish hair ; occiput pale smoke-grey pollinose, hind margin fringed above with very short and inconspicuous whitish or yellowish white hair; front in 2 of moderate breadth or relatively somewhat broad, slightly but distinctly narrower below, about four times as long as its breadth at the lower end; lower frontal callus shining blackish brown, large, quadrate and nearly square, extending from eye to eye or separated from eye on each side by narrowest possible pollinose interval, lower margin convex and descending slightly below level of inner angles of eyes, upper edge with two small emarginations and consequently produced into three points, upper frontal callus black or blackish brown, oval or elliptical oval, not connected with lower; eyes of 2 wm life with upper angles and lower borders dark purplish brown, and central portion metallic bronze or bronze-green, crossed by two narrow, deep blue, horizontal bands; palpi ivory-yellow or cream-coloured, proximal segment clothed with long whitish hair, distal segment in 2 acuminate, moderately or considerably swollen at base, clothed on outer side with minute, appressed, (637) G2 310 ERNEST E. AUSTEN. glistening whitish hairs, mixed with minute black hairs; first joint of antennae light drab or light pinkish cinnamon pollinose, moderately swollen towards distal extremity, with upper distal angle considerably produced and partly embracing second joint, clothed on outer side with glistening yellowish hairs and with minute black hairs on upper border and towards upper distal angle, second joint of antennae light drab pollinose, with distal margin fringed on outer side below with minute glistening yellowish hairs and upper distal angle moderately or considerably produced, third joint cinnamon- or orange-cinnamon-coloured (annulate portion sometimes darker, dark brown or blackish brown), expanded portion of moderate depth or rather narrow, with blunt but prominent angle on upper margin near base. Thorax: dorsum clothed with minute, appressed, yellowish or whitish, silky hairs, yellowish tinge in the hair being more apparent in interspace between posterior ends of the two admedian pale olive-grey pollinose stripes (median pale olive-grey stripe becoming obsolete at about level of transverse suture); lateral borders of dorsum pallid neutral grey or pale smoke grey pollinose, swelling in Fig. 11. Tabanus accensus, Austen, 9. 4. depression at each end of transverse suture drab-grey, clothed below with long, dusky or blackish hair, and above with shorter yellowish hair ; pleurae and pectus pale smoke grey pollinose, clothed with whitish hair. Abdomen: middle line of dorsum, from first (visible) to sixth segments inclusive, occupied by a longitudinal series of anteriorly directed, truncated, smoke grey or pale pinkish buff triangles ; blackish brown marks, except in some specimens admedian ones on first tergite, not reaching hind borders of segments; in some cases, on each segment from third or fourth to sixth inclusive, the ends of each pair of blackish brown marks are fused together posteriorly in such a way as to enclose intervening smoke grey or pale pinkish buff area, which then assumes appearance of an elliptical oval, more or less oblique spot ; pale areas of dorsum, including hind borders of second to sixth tergites inclusive, clothed with minute, appressed, glistening, pale yellowish or pale ochreous hairs, blackish brown areas clothed with minute, appressed, black or blackish hairs; seventh tergite, except lateral and hind borders, dark brown, smoke grey pollinose at base, clothed with fine, fairly long, curving, black or blackish CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. a lp | hair; venter pale smoke grey or pale drab-grey pollinose, clothed with minute, appressed, glistening, ivory yellow hairs, hind borders of second to sixth scutes inclusive ivory yellow, seventh segment deep greyish olive or mouse grey pollinose, clothed with usual coarse, erect, blackish hair. Wangs: clear, hyaline, veins lighter or darker olive-brown, anterior branch of third longitudinal vein without an appendix, at least in typical form ; stigma pale isabella-coloured or almost colourless, inconspicuous. Sguamae ivory yellow, borders cream-buff. Halteres: knobs ivory yellow, stalks pale cream-buff. Legs: coxae smoke grey or pale smoke grey pollinose, clothed with whitish hair; femora, except tips which are cream-bufi, light greyish olive or smoke grey pollinose, clothed with whitish hair (extreme tips, at least of front and middle femora, with minute black hairs above), inner and under sides of front femora blackish brown or warm sepia-coloured, inner sides more or less smoke grey pollinose ; tibiae cream-buff, tips (distal sixth in case of middle and hind pairs, distal fourth or rather less in case of front pair) blackish brown or dove brown, pale area of tibiae clothed with minute, appressed, glistening ivory yellow hairs, mingled on extensor surfaces with minute black hairs; front Fig. 12. Tabanus accensus, Austen, head of 9 from in front, x 10; a, antenna from the side, greatly enlarged. tarsi blackish brown, first joint more or less clothed above, except at distal extremity, with minute appressed, glistening ivory yellow hairs; middle and hind tarsi clove brown, proximal two-thirds or three-fourths of first joints cream-buff or cream- coloured, clothed with minute, appressed, glistening ivory yellow hairs, mingled with minute black hairs. Lower Jordan Valley and Jerusalem: type from Wadi el Kelt, Jericho Plain 11.v1.1918, biting author’s arm; a para-type from Wadi Mellaha (about 63 miles N.E. of Jericho, near R. Jordan), 26.v.1918, on horse; a third specimen labelled * Jerusalem,” received in 1912 from the Zoologisches Museum, Berlin. In size as well as in thoracic and abdominal markings Tabanus accensus, at least in the female sex, presents a distinct resemblance to the Ethiopian 7. pertinens, Austen; the female of the new species is, however, at once distinguishable, unter alia, by the much narrower front, by the presence of two black or blackish brown frontal calli instead of the single callus being either absent or scarcely notice- able, and by the very different coloration and markings of the legs, especially the tibiae and tarsi. 312 ERNEST E. AUSTEN. 14, Tabanus leleani, sp. n. (figs. 13, 14.) $9.—Length, J (8 specimens) 12 to 14 mm., 9 (17 specimens) 11 to 14°6 mm. ; width of head, j 4°4 to 5:4 mm., 9 4:25 to 5-4 mm.; width of front of 9, at vertex 08 to 1 mm., across lower edge of frontal callus 0°5 to 0°75 mm.; length of wing, 6 8°75 to 10°38 mm., 2 8°75 to 11°6 mm. In general appearance looking like a greyish form of Tabanus cordiger, Mg., with which it closely agrees as regards pattern of abdominal markings, and dimensions and other details of 9 front. Eyes under an ordinary hand-lens appearing bare in both sexes, in § with a sharply contrasted area of greatly enlarged facets, and in both sexes with a single; horizontal, transverse band ; dorsum of thorax (in undamaged specimens) mouse grey, clothed with fine erect hawr, interspersed with a coating of fine, appressed, short, silky hair, of an ochreous or somewhat paler colour; dorsum of abdomen dark mouse grey or blackish mouse grey (when viewed at a low angle from behind suffused, like thorax, with pale neutral grey blcom), with, in middle line, a longitudinal series of pallid neutral grey triangles, and, on each side, midway between middle line and lateral margin, a series of oblique and roughly elliptical spots of same colour, sides of two or more of the proximal segments, from posterior angles of first (visible) to fourth inclusive, often more or less cinnamon ; wings hyaline, stugma scarcely distinguishable, anterior branch of third vein typically without appendia. Fig. 13. Tabanus leleani, Austen,2. 4. Head: face and jowls whitish pollinose and clothed with white hair, occiput in both sexes, frontal triangle in g, and front and subcallus in 9 pale olive-grey pollinose ; frontal triangle in J, immediately below level of enlarged facets of eyes, crossed by a somewhat deep, dusky, shimmering pollinose, horizontal band ; in both sexes a dark clove-brown horizontal band between base of antenna and eye on each side; hind margin of upper border of occiput fringed with short, erect or forward-curving, whitish or pale yellowish hair, among which are occasionally intermixed on each side a few short black hairs, though, at least in the case of the g, the fringe is much shorter than the corresponding one in T. cordiger, Mg. ; front CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 313 in 2 broad, slightly narrower below, its length equal to rather less than four times its breadth at lower margin of lower callus, clothed above with minute black hairs and below with appressed yellowish hairs, vertex darker (dark olive-grey), but without trace of ocellar tubercle ; two frontal calli in 9, lower frontal callus shining black or blackish-brown, large, broad, quadrate, about four-fifths as high as broad, only separated from eye on each side by an exceedingly narrow, pollinose interval ; upper frontal callus in 9 represented by a large black mark, rather less than half- way between upper margin of lower callus and vertex, separated from eye on each side, bare of pollen but in undamaged specimens clothed with minute, appressed, yellowish hairs, and usually either roughly V-shaped, or transversely oval or elliptical, with a deep, narrow indentation in middle line above; eyes in 3 with sharply defined hind and upper border of small facets, not tapering off but extending in full depth to vertex, while greatest depth of lower border, which also consists of minute facets, is about 24 times that of upper border; eyes of 3 in life with area of large facets greyish brown or a beautiful steely grey, lower border consisting of small facets purplish bronze or reddish brown, a dark horizontal band, coloured like the band in 9, close to upper margin of lower border, hind and upper border dark purplish brown ; eyes of 9 in life bronze-green with a purplish sheen, or purplish bronze, with a single, horizontal, dark purple, dark purplish brown, or dark brown band, starting from inner margin of each eye adjacent to lower frontal callus, extending outwards and backwards, and ending bluntly at about one-fifth of the diameter from hind margin; palpo light buff, or ivory white (proximal segment mouse grey at base), clothed with whitish hair, in case of g usually mingled towards distal extremity of greatly swollen, pyriform or ovate pyriform, terminal segment with a few minute black hairs; terminal segment of 2 palpi acuminate (proximal half greatly swollen), its outer surface clothed with minute, appressed, glistening white or yellowish white hairs, frequently mixed with minute, scattered, appressed, black hairs ; first and second joints of antennae blackish (first joint somewhat greyish pollinose above), clothed below with glistening whitish hair and above with minute black hairs, first joint greatly swollen, its bluntly produced upper distal angle overlapping the sharp corresponding angle of the small second joint, third joint warm sepia-coloured to fuscous-black, its expanded portion moderately broad, with a blunt but fairly prominent angle close to base of upper margin. Thorax: dorsum with usual light grey longitudinal stripes more or less distinctly marked, erect hair partly blackish, partly pale (light buff), except on anterior margin, where it is entirely whitish ; swelling in depression at each end of transverse suture greyish pollinose vinaceous-fawn, clothed mainly with black hair; humeral and postalar calli, and lateral border of dorsum above base of wing on each side, clothed with whitish hair; pleurae and pectus pallid neutral grey pollinose, clothed with fairly long, fine hair of same tint. Abdomen: lateral margins of tergites, from second or third (visible) to distal extremity, and posterior borders of dorsal and ventral scutes of all segments cream-buff; basal angles of first (visible) tergite more or less greyish pollinose ; pallid neutral grey markings of dorsum, described in diagnosis above, clothed with minute, appressed, cream-coloured hairs, and usually most distinct (sharply defined in 9) on second and three following segments, but sometimes distinguishable on first and sixth segments 314 ERNEST E. AUSTEN. also; in the case of the second to fifth segments inclusive the median triangles, which are based on the hind margins and taper sharply forwards, reach the anterior margins, while in the 9 the oblique spots sometimes similarly extend over the full length of the segments ; dark areas of dorsum clothed with minute black hairs— towards lateral margins of fifth and following segments in g, and on sixth and seventh segments in Q, the hair is longer; lateral margins of all segments from second to sixth inclusive clothed with fairly long whitish hair ; venter pallid neutral grey pollinose, clothed, except in case of last (visible) segment, which is bedecked with the usual coarse, outstanding black hairs, with minute, appressed, glistening, creamy-white hairs, a certain number of black hairs on penultimate segment also, on median area in front of hind margin. Wangs: veins dark clove-brown, except basal portions of second and fifth longitudinal veins, which are paler (sepia- coloured) ; stagma when discernible, pale cinnamon, narrow. Squamae cream- buff, borders darker (isabella-coloured). Halteres ivory yellow, stalks somewhat infuscated, and knobs usually seal-brown at base above and below. Legs: coxae and femora (except extreme tips of latter, which are cinnamon-buff, and inner and under sides of front femora, which are black—sometimes greyish pollinose in case of inner side) pale neutral grey pollinose, clothed with whitish hair ; front tibiae black, their proximal halves, or rather more or rather less, cinnamon-buff or pale pinkish buff, and clothed with minute, appressed, whitish hairs, which may extend on to the black area, outer surface of front tibiae also with longer hairs, some whitish others black, or whitish on proximal half and black towards distal extremity ; middle and hind tibiae, except extreme tips, which are tinged with seal brown, cinnamon-buff, clothed with whitish hair, which is long (relatively dense also in the case of the 3) on outer side of hind pair ; it is also long on outer side of middle tibiae, where however it is sparse and intermixed with scattered black hairs (likewise sometimes present on outer side of hind tibiae) ; front tarsi black, not expanded, hind tarsi and upper surface of middle tarsi seal brown or dark brown, under surface of middle tarsi, as also upper surface of fifth joint and of proximal two-thirds of first joint, cinnamon. Type of ¢ in Wadi el Kelt, Jordan Valley, near Jericho, 1.vi.1918; type of 9, and three para-types of the same sex in Wadi el Aujah (64 miles north of Jericho), Jordan Valley, 19.1v.1918; additional para-types as follows :—3 gd, (1 resting on rocky bank above pool) 3 29, in Wadi Ghuzze, near Tel el Fara, 5.vi., 4-23.vi., and 8.vii.l1917; 3 99, Ain es-Sultan, near Jericho, 22.1v.1918; 3 39 (resting on cliff wall, 5.30 p.m.), 1 & (tried to bite author, 10.30 a.m.), in Wadi el Kelt, near Jericho, 3-5.vi.1918; 1 9, Wadi ez Zerka, 5.viii.1918 (Captain (acting Lt.-Col.) W.J. Dale, O.B.E., R.A.V.C.), on horse. As throwing light on the range of this species, it may be noted that in addition to the foregoing, the Museum also possesses the following specimens of 7’. leleant :— 1 g (ex coll. Saunders), 1 9 (1909, Dr. G.A. Williamson), from Cyprus ; 1 9, Sbeitla, Tunis, 6.v.1913 (G.C.Champron) ; 2 29, Biskra, Algeria, 14.1v.1894, 31.v.1893 (Rev. A. EH. Baton), and 1 2 from same locality, 1908 (Hon. W—now Lord— Rothschild); 19, Nasiryah, R. Euphrates, Mesopotamia, April-June 1916 (Major W.S. Patton, I.M.S.); and 1 2 from Kangra Valley, Punjab, 4,500 ft., July 1899 (G. C. Dudgeon). CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 315 The author has much pleasure in naming this species in honour of his friend Lt.-Col. P.S. Lelean, C.B.,C.M.G., R.A.M.C. (his cheery tent-companion during many months of the Sinai-Palestine campaign), whose ingenuity, resource, and untiring energy in the cause of field sanitation contributed in no small degree to the healthy efficiency of the H.E.F. Tabanus leleani, which was the characteristic, and indeed the only representative of its genus met with in the Wadi Ghuzze in 1917, was again common a year later im wadis running down to the Lower Jordan, near Jericho. In the Wadi el Aujah, 19.iv.1918, both sexes were seen resting on stones near the water’s edge, while at the beginning of the following June, towards sunset, males were noticed in some numbers sluggishly resting on the precipitous, cliff-like walls of the Wadi el Kelt, a short distance above the spot at which the latter leaves the hills and enters the plain ; under such conditions the insects could generally be boxed or captured in tubes with little difficulty. The solitary occasion on which T. leleani was observed to attack man has already been recorded. Fig. 14. Tabanus leleani, Austen: a, head of gin profile; b, head of 2 from in front, x 10; b, antenna of 9° from the side, greatly enlarged. Apart from its much paler (greyer) appearance, T'abanus leleanc is distinguished ifrom J’. cordiger, Mg., by the presence of a band on the eyes, by the row of erect chair on the upper margin of the occiput in the ¢ being much shorter, finer and less conspicuous, by the erect hair on the dorsum of the thorax being shorter and finer, and the covering of pale yellowish, silky hair more appressed. In general appearance and the majority of the external characters, except as regards size, which is usually distinctly larger, the species described above agrees with Tabanus albifacies, Lw., found in Egypt and Persia. 7. albsfacies, however, is distinguishable at once by the presence on the eyes (at least in the Q) of three bands instead of only a single band. As described by Loew (Neue Bewtrdge, iv, p. 28 (1856)), the wing of 7’. albsfacies shows a well-marked appendix to the anterior branch of the third longitudinal vein. In the specimens of 7. leleant examined ‘such an appendix is absent, except in the case of two 9 9 from Nasiryah, Mesopotamia (Major W. S. Patton, I.M.S.). So far as can be judged from the original description of Tabanus wnicinctus, Lw., the type of which, like that of the foregoing, was obtained in Egypt, 7’. leleani also 316 ERNEST E. AUSTEN. resembles that species in general appearance, width of front in 9, and size. T. unicinctus is however distinguishable, inter alia, by the veins in the centre of the wing being suffused with brown, by the presence of a dark brown stigma, and by the ground colour of the legs, with exception of a pale ring at the base of the front. tibiae, being entirely black. Finally it may be mentioned that Tabanus leleanz is nearly allied to T. unifasciatus Lw., but may be distinguished by the sharply marked contrast in size between the es and small facets in the eyes of the g, and by the greater = and blunter termination of the purple eye band in the 9. 15. Tabanus pallidipes, sp. n. (figs. 15, 16.) °.—Length (2 specimens) 11 to 13°6 mm. ; width of head 3°8 to 4°75 mm. ; width of front at vertex 0°4 to 0°5 mm. ; length of wing 8°6 to 10°2 mm. Small, greyish species, with bare eyes, narrow front in 9, two frontal callt on same sex, the upper one narrow and elongate or almost linear in shape ; dorsum of abdomen greyish cunnamon-coloured, darker at distal extrennty and with lighter and darker markings ; and pale femora and tibiae. Head: front and subcallus in 9 dark olive-buff pollinose, front lathe for most part with minute, appressed, ochreous hairs, and with short dusky hairs on vertex, no trace of an ocellar tubercle, but a tiny, dark mouse grey spot in ocellar region ; face and jowls pallid neutral grey pollinose, clothed with whitish hair; occiput light olive-grey pollinose, upper hind margin with a scarcely noticeable fringe of short, whitish hair ; front in 9 very narrow, tapering somewhat to lower extremity, about six-and-a-half times as long as its breadth at lower end ; lower frontal callus shining dark brown (mummy brown) or cinnamon-brown, oval or quadrate with its angles somewhat produced upwards and downwards, narrowly separate from eye on each side, sometimes with a fine, median, longitudinal groove, upper frontal callus dark brown, linear or narrowly elliptical, indistinctly connected with lower callus by a fine median line; eyes of type an lufe dark brown, unbanded ; palm creamy white or cream-buff, proximal half of distal segment moderately or considerably swollen, distal half acuminate, proximal segment clothed below with whitish hair, distal segment clothed on outer side with minute, appressed, glistening, cream-coloured or yellowish hairs, mixed with minute, appressed, black hairs, which. sometimes appear to be in greater number than the pale hairs ; antennae cinnamon- coloured, first joint (sometimes second joint also) drab-grey pollinose on outer side, moderately swollen distally, with its upper distal angle considerably produced. and partly embracing second joint, first joint clothed on outer side above and distally with minute black hairs, mixed with a certain number of glistening ochreous. hairs, and below with longer yellowish hairs, second joint clothed on distal margin. with minute black hairs and with its upper distal angle sometimes considerably produced, third joint with expanded portion of moderate depth and having a blunt angle on upper margin a little before middle, expanded portion slightly shorter than annulate portion, which may be infuscated. Thorax: dorsum deep neutral, grey, anterior border pale neutral grey, lateral borders and five more or less distinct longitudinal stripes pale smoke grey or smoke grey, outer stripes visible only behind CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. SLT transverse suture above bases of wings, all stripes sometimes difficult to distinguish ; dorsum clothed with short, appressed, glistening, pale ochreous or greyish hair, on darker areas mixed with short, erect, blackish hair, ground colour of rather less than distal half of scutellum when denuded sometimes greyish cinnamon-coloured ; swelling occupying depression at each end of transverse suture: clothed with fairly long, fine black hair, mixed with yellowish hair, which predominates above; pleurae and pectus pale smoke-grey pollinose, clothed with whitish hair. Abdomen: dorsum from second to sixth segments inclusive with four longitudinal rows of elongate, more or less oblique, blackish brown spots, four: spots on each segment, namely an admedian pair (extending from anterior margin outwards, nearly or quite reaching hind margin and enclosing an anteriorly directed, drab-grey, truncated, median triangle), and a lateral oblique spot on each side: extending backwards and outwards from anterior margin, partly cutting off basal angle but not reaching hind margin, area between lateral and admedian dark. brown spots on each side of each segment thus taking shape of a more or less oval or elliptical oval, backwardly and outwardly directed, drab-grey spot; seventh tergite drab, without darker markings; median drab-grey triangles resting on hind margins and extending to front margins of segments, with their posterior angles somewhat produced outwards; drab-grey or greyish cinnamon areas of dorsum clothed with minute, appressed, glistening, yellowish hairs (hair on extreme. _ lateral extremities of anterior tergites more whitish), blackish brown areas of dorsum clothed with minute, appressed, black hairs, longer black or blackish hairs also present on lateral extremities and posterior angles of last three tergites ; venter light pinkish cinnamon (hind margins of second to sixth segments inclusive: paler—light buff), suffused with pale smoke-grey or pale drab-grey pollen, and clothed. with minute, appressed, glistening, cream-coloured or pale yellowish hair; ventral scute of seventh segment bearing the usual coarse, erect, black hairs. Wangs hyaline, clear or tinged with light drab, veins lighter or darker olive-brown, auxiliary and first longitudinal veins, and proximal portion of second and fifth: veins, paler (tawny-olive or cinnamon); stigma faintly tinged with yellowish or almost colourless, inconspicuous. Squamae whitish or ivory yellow, borders cream- coloured. Halteres ivory yellow or cream-coloured, distal extremity of stalks slightly darker, at least in dried specimens. Legs: coxae drab-grey or smoke- grey pollinose (ground-colour of front pair tawny-olive), clothed with whitish hair ; femora cinnamon-buff or pinkish cinnamon, pale smoke-grey pollinose on outer side, clothed with whitish or pale yellowish hair, front femora also with black hairs above ; tibiae warm buff or cinnamon-buff (distal fourth or rather more than distal fourth of front pair brownish), clothed with minute, appressed, glistening, cream-coloured or pale yellowish hair, infuscated distal extremities of front pair and extensor surfaces of all tibiae also largely clothed with minute black hairs, hind tibiae on outer side with a fringe of longer hair, in which in some specimens. cream-coloured in others black hairs predominate ; front tarsi dark brown or dark sepia coloured, third and fourth joints somewhat expanded, middle and hind tarsi mummy brown (first joints paler at base), all tarsi clothed above with minute black hairs, front tarsi sometimes also with yellowish hairs on proximal two-thirds of anterior surface. 318 ERNEST E, AUSTEN, Lower Jordan Valley and Jerusalem: type from Wadi el Kelt, Jericho Plain, 11.vi.1918 ; a second specimen labelled “‘ Jerusalem,” received in 1912 from the Zoologisches Museum, Berlin. The species just described, individuals of which, as will have been seen, sometimes ‘show considerable variation in size, appears to be allied to the Austrian Tabanus miki, Brauer, from which however it is distinguishable, inter alia, by the much paler legs ; in T. mikii the femora, front tarsi and tips of the front tibiae are black. ™ ———s "2 REDE LAAN Dai Fig. 15. Tabanus pallidipes, Austen, 2. x 4. Fig. 16. Tabanus pallidipes, Austen, head of ° from in front, x 10; a, antenna from the side, greatly enlarged. Tabanus pallidipes is possibly identical with the Arabian T. arabicus, Macq., in which the femora were stated by Macquart (Mém. Soc. Roy. Inlle, 1838, p. 299; Dipt. Exot., i, 2, 1838, p. 183) to be tawny or fawn-coloured (fauves). The original -description of this species, however, which makes no mention of the dimensions of the front im the © and leaves the abdominal markings to the imagination of the reader, is so brief and faulty that, in the absence of the typical specimens, precise identification is impossible. CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 319: The species characterised above is also allied to 7’. laetetunctus, Becker, the typical series of which was obtained in Persian Baluchistan ; so far as can be judged from. Becker’s description, however, 7. pallidipes is distinguished, inter alia, by the much. narrower front in the 9, and by the infuscated tarsi. 16. Tabanus dalei, sp. n. (figs. 17, 18.) g.—Length (1 specimen) 125 mm.; width of head 475 mm.; length of wing 1] mm. Dorsum of thorax dusky neutral grey, shining; dorsum of abdomen slate-black second and four following segments with postervor transverse bands of appressed, glistening, silvery-white ha; eyes bare; wings hyaline, extreme base, costal and subcostal cells, and proximal half (narrow portion) of marginal cell ochraceous tawny, contrasting sharply with remainder of surface ; legs black, extensor surfaces of all tibiae, except at distal extremities, clothed with glistening whitish or yellowish harr. Fig. 17. Tabanus dale, Austen, head of ¢ in pro- file. > about 10. . Head of ¢ small, frontal triangle, face and jowls smoke-grey pollinose, face and jowls clothed with whitish hair; occiput neutral grey pollinose, posterior margin: fringed at sides and above with erect, yellowish hair, longer in region of vertex, and mingled on each side of vertical region with a certain number of black or blackish hairs ; ocellar tubercle inconspicuous, sunk in vertical triangle; eyes in 3 each with an area of moderately enlarged facets abutting on inner margin, but such facets progressively diminishing in size towards exterior and merging gradually into remainder, to which they nowhere present a sharp contrast ; palpi in ¢ dark mouse grey pollinose, clothed with whitish hair, terminal segment small, bluntly elliptical ovate, not pointed at distal extremity ; first joint of antennae in ¢ blackish mouse grey, scarcely swollen distally, sparsely clothed with short yellowish hairs, second joint blackish brown, with a few yellowish hairs on inner side and minute black hairs on upper distal angle, which is considerably produced, third joint dark clove brown or blackishjbrown (slightly paler at extreme base), expanded portion of moderate depth or fairly deep, at least in g, with a blunt but prominent 320 ERNEST E. AUSTEN. angle about middle of upper margin, which is moderately excavated in front of the angle. Thorax: dorsum clothed with fine, erect, silky, pale yellowish hair, scattered among which is a certain number of fine black or blackish hairs, which -on scutellum are much more numerous and even appear to predominate, anterior border of dorsum with commencements of two light greyish olive, pollinose, longitudinal stripes; swelling occupying depression at each end of transverse suture agreeing in coloration with remainder of dorsum and similarly clothed with hair, but without black hair; pleurae and pectus clothed with yellowish white or whitish hair, the tuft of long hair on hind margin of mesopleura conspicuously ochreous at base; pleurae neutral grey pollinose above, pectus and lower part of pleurae deep neutral grey. Abdomen: dorsum of first (visible) segment neutral grey pollinose (greyish olive pollinose on each side), clothed in centre and on each side with glistening yellowish white hair ; posterior transverse bands of appressed, glistening, silvery-white hair, on second to sixth segments inclusive, expanded in middle line so as to form anteriorly directed triangles ; silvery-white hair also extending Saeco Zz tii Fig. 18. Tabanus dalei, Austen, wing of g. x 10. from front to hind margins on lateral borders of segments referred to; on third, fourth and fifth segments commencement of expansion of hind border on each side includes a certain amount of appressed, glistening, ochreous hair; ground- colour of hind margins of first six tergites cimnamon-drab; dorsum, except as already stated, clothed with short, appressed, black hair; venter deep neutral orey pollinose (hind margins of segments as on dorsum), clothed with silvery white or appressed silvery white hair, which is more conspicuous on hind borders of segments, and in central area of anterior part of second, third and fourth segments is largely if not entirely replaced by minute black hairs. Wungs: veins dark brown or mummy brown, proximal portion of fifth longitudinal vein, forming lower boundary of second basal cell, ochraceous-tawny and, like corresponding portion of fourth longitudinal vein, narrowly suffused with similar colour ; anterior branch of third longitudinal vein without appendix; stigma ochraceous tawny, indis- tinguishable from coloured costal border ; proximal portion of marginal cell with a narrow hyaline streak on its anterior border. Sqguamae ivory yellow, borders ‘cream-buff or cream-coloured. Halteres: knobs clove brown, stalks mummy brown. Legs: coxae neutral grey pollinose, clothed with yellowish white hair ; middle and hind femora and posterior surface of front femora clothed with CONTRIBUTION TO KNOWLEDGE OF TABANIDAE OF PALESTINE. 321 glistening, yellowish white or ochreous hair; pale hair on hind tibiae in ¢ fairly long and bushy; distal extremities of tibiae and upper surfaces of tarsi clothed with minute black hairs. Near Jerusalem, May, 1918 (Captain (acting Lt.-Col.) W. J. Dale, O.B.E., R.A.V.C.). The species characterised above, with which the author is glad to have the oppor- tunity of associating the name of its discoverer (A.D.V.S., XXth Corps, 1917-19), is not closely related to any of its congeners known to the writer, and, owing to the ochraceous-tawny coloration of the base and proximal two-thirds of. the costal border of the wing, is readily distinguishable from all other species described or mentioned in the present paper. ) OR) SES aaa NORTON ERROR ¥ Ss SPO Sites is SO, 5) ‘ : yok ney AL ‘ ji 1% 1 f } r . ‘ i Ay on ie ie fi eh hid». AL: few, ’ ee 323 NOTES ON SOME CULICIDAE COLLECTED IN LOWER MESOPOTAMIA. By Lieut. P. J. Barraup, R.A.M.C. From October 1918 to February 1919 inclusive I was stationed at Basrah, Lower Mesopotamia, and during that time I was able to devote some time to the collecting of CuLicrpDaE and their larvae. My investigations were necessarily confined to a somewhat restricted area, possibly five miles from north to south, and one mile wide, embracing the districts known as Ashar, Makina, and Magil, along the right bank of the Shatt-el-Arab river. Some collecting was also done on the left bank from Tonooma, opposite Ashar, northwards. Practically the whole of this ground is within the cultivated belt lying between the river banks and the desert, and is intersected by numerous creeks, and irrigation canals. There were also in some places many pools and swampy areas forming good collecting ground. The wet season commenced towards the end of November and continued to the end of my stay, but the rainfall was intermittent and not considerable in amount. In December there were several violent storms with heavy rain for a time. From official temperature records I made the following notes :— October 1918.—Maximum shade temperature between 102°F. (5th and 7th) and 88°F. (23rd and 30th). Minimum shade between 75 (5th) and 57 (24th). November 1918.—Maximum shade between 89 (3rd, 6th and 7th) and 69 (9th). Minimum shade between 63 (7th) and 49 (several days in the last week). December (up to 22nd, beyond which date I have no records).—Maximum shade between 76 (Ist) and 60 (10th, 11th and 12th). Minimum shade between 60 (17th) and 38 (13th). There was no frost or snow throughout the winter. The “ main river ” referred to in the following notes is the Shatt-el-Arab, formed by the junction of one of the mouths of the Euphrates and the Tigris, about five miles north of Ashar. I am much indebted to Dr. Guy A. K. Marshall, Imperial Bureau of Entomology, and Mr. F. W. Edwards for kind assistance in the working out of my material ; to Major 8. R Christophers, I.M.S., for giving me facilities and help in my work at Basrah ; and to Capt. H. E. Shortt, I.M.S., for his valued companionship on many a delightful excursion. | Anopheles stephensi, Liston. This appears to be the chief malaria carrier of the district and bred continuously during the time I was at Basrah. Numbers of adults and larvae were found in October, November and December. The larvae occur chiefly in the clearer pools in the cultivated area. On 14th January 1919 [ found about one hundred larvae in a small pool near Ashar, many of them full-grown. Adults were bred out in the laboratory during the following ten days. In February only a few larvae were found. (637) H 324 P. J. BARRAUD. Anopheles pulcherrimus, Theo. Anti-malarial work in the cultivated belt seems to have ia more effect on this species than upon the preceding one, and the numbers have been greatly reduced. Large numbers of larvae were, however, found in pools near the edge of the desert in October. Sometimes the larvae of this species and those of A. stephensi occur together in the same pool, but those of A. pulcherrimus usually prefer weed-grown stagnant and more brackish water. Camps near the desert are at times infested with large numbers of A. pulcherrimus adults. Anopheles sinensis, Wied. Major Christophers has described a new form of this species occurring in Mesopotamia, viz. var. mesopotamiae (Indian Jl. Med. Res. i, p. 196). The species is not very common in the Basrah area. On 18th December I saw it for the first time and caught three females and some small larvae possibly of this species, in the marshes on the left bank of the main river. Subsequently a few more were taken in the same place. On 18th January one male emerged in the laboratory from a nymph collected in Makina area. Later on more were found on the banks of a creek between Magil and the Euphrates. Culex fatigans, Wied. Abundant throughout the cultivated area during my stay. The larvae were found in extraordinary numbers in some pools, appearing as a dark mass beneath the surface. Culex pipiens, L. Far less common in my experience than C. fatigans. I did not identify any larvae of this-species until January. In mounting numbers of larval skins I did not find much difficulty in separating the two species. Culex modestus, Fc. Some larvae of this species were collected in October and November, but they were not very numerous. In one case they were found in a pool containing numbers of A. stephensi and A. pulcherrimus. Gulex tritaeniorhynchus, Giles. Larvae were found in October and November in Ashar and Makina areas in small numbers. Culex tipuliformis, Theo. Larvae of this species were not found until January, first in isolated pools on the left bank of the main river, and later in larger numbers in Ashar district. Stegomyia fasciata, I. Very few specimens met with in Basrah area during my stay, and no larvae found. On 23rd November, when visiting Mohommerah, lower down the main river on the Persian side, adults were found in small numbers in a building on the river bank. NOTES ON SOME CULICIDAE COLLECTED IN LOWER MESOPOTAMIA. a2) Ochierotatus dorsalis, Me. This mosquito was found quite commonly over the whole area, both adults and larvae, and is the only species of those mentioned which occurred in about the same numbers in the cultivated belt and in the outlying places visited on the edge of the desert. In one case larvae were found in a collection of rain-water in the bottom of a boat in Ashar town, and some were found in pools in the desert at some distance from cultivation. Adults and larvae occurred from December to February. Theobaldia longiareolata, Macq. The first specimen emerged on 27th December in the laboratory, the larva having been found in Ashar area. Towards the end of January and in February the larvae became more numerous, being found chiefly in the deeper pools. The pupa may often be seen about four inches below the surface of the water and appears to retain that position by gentle movement, remaining at that depth for quite a long time. Theobaldia annulata, Schranck. Larvae were not found until 24th February and were then not uncommon in Ashar district. The adults have the thorax and abdomen almost uniformly light reddish brown, with very little trace of the normally conspicuous abdominal banding. This inter- esting colour modification, which is unaccompanied by any structural change, is no doubt adaptive to desert conditions. A similar variation has been noticed in several other mosquitos. Port Said, June 1919. > (637 a ; la | AAW RS oH: reg, a eh Rio scan | | | te ae pach iia ; dea oe pat ag a abe | : . he hia Spee, ne ' i a ie i ae 7 aces 74 onl eR AD DG nf ; se boieh cis a, Hest nese Saas BP a f ing adie i FS Osh too, ea aed 6 , en ba hee eee ; em Bh phy sth Vea) a, Dag Diya: nieici ae ; ‘ ¥ ie j is " i idle . 4 thay: pie ib add ah iene . “bawet! Deri si ¥ 4 # gt Vial ihe. thw tsa bis ‘salad fee his OR, os | . ibaa fag hate LALLA GON ce ues "ees Anne or AN 4 ie ha & = ’ “is be Pe eae NRE iv aye & 1 se: ’ - y 3 4 ie a bane oe ¥ i trek. yaa? A i vpeeailn ; es sia ‘siete pe Bic ” oii i 83 ie siearenas) dod ‘his wali “fbn ne tig beat, fone. nies es is > etter: Seah “ ; : Bete i's isl dalbha: 7 ian, ingle wane ae aint ean ji Spo | ae Ie ata Spek foeicaaungeth re "et 1H) wk anid ical e AAlinde he | “meh 9 ae es ; \ ‘ ee ee fee AK ; Se ae 9 ay 1S eat ae ENG ad Chee ee Fi H seid Au wis dbs ’ . | ae hc yt es ae lc ke TCLS. wuigtboias i werete Ney F pi iy eo r é \ V2 Te gk a t vii this i = 4 f * s } pe all . iy — »~ rm so Bes! a he atl t eet) eeu sid th gine vont Hse iat ta vie beh 2 amb hl eh i. F “ 7 be dj a : oot) : io rt Ty | pr ’ | a re ; 1 : id 5 o = ai 4 J i" si by ay a +4 4 \ ’ i] ee ae. s sane xipuctn, | i a af ‘ ibe hon pi ero ‘ ; roe Shad Ney, ey An pas 5 Shes Porat: et 3 ea Pao PY ‘, 2% ie Beret nurs: > j fi hy yeey Nas j 1%. a , a es 1 Pi oe cre oat ON A NEW MOUTILLID PARASITE OF GLOSSINA MORSITANS. By Row ann EH. TuRNER. Mutilla auxiliaris, sp. nov. Q. Nigra; thorace brevissimo, fusco-ferrugineo ; tergitis tribus basalibus fascia apicali angusta pallide aureo-sericea ; area pygidiali nulla. §. Niger; prothorace, mesonoto, scutelloque ferrugineis; tergitis quatuor basalibus albido-fimbriatis ; alis fuscis, basi dilutioribus, posticis basi hyalinis, venis nigris, calcaribus intermediis posticisque pallidis. Long. 9,5 mm.; ¢, 7 mm. Q. Mandibles simple ; antennae rather stout ; second joint of the flagellum short, scarcely longer than the third; the apical joint rather slender, longer than the penultimate. Head rugosely punctured ; eyes oval, situated nearer to the base of Fig. 1. Mutilla auxiliaris, Turner, sp.n., 2d. the mandibles than to the hind margin of the head. The whole insect rather sparsely clothed with long blackish hairs. Thorax as broad as the head, punctured-rugose, almost as broad in the middle as long ; the sides distinctly, but not strongly, convex ; the anterior margin not quite straight, widely and very shallowly emarginate, the posterior slope rather abrupt, no scutellar tubercle, the sides of the apical slope with several minute teeth. Abdomen strongly punctured, the punctures more or or less confluent longitudinally ; first tergite broad, transverse, almost as broad as the second, abruptly truncate anteriorly ; second tergite as broad in the middle as long, the sides distinctly, but not very strongly convex ; no pygidial area. Hind tibiae with a row of three spines ; tarsal ungues simple. §. Clypeus strongly concave, shallowly emarginate at the apex ; bordered laterally by carinae which converge towards the base, the apical margin with several pale setae. Mandibles bidentate on the right side, tridentate on the left ; labrum trans- verse. Head closely punctured, the clypeus smooth and shining; antennal tubercles strongly developed, the front with an obscure longitudinal sulcus which does a) a ROWLAND E. TURNER. not extend to the anterior ocellus. Second joint of the flagellum distinctly shorter than the third, the fourth slightly longer then the third. Pronotum and mesonotum evenly and rather strongly punctured, the mesonotum short and broad ; scutellum more coarsely punctured ; median segment short and broad, very coarsely reticulate. Abdominal tergites rather less strongly punctured than the thorax; first tergite short, transverse, narrower than the second which is very broad, being nearly three times as broad at the apex as long, with the sides strongly convex ; seventh tergite broadly subtruncate at the apex. Second abscissa of the radius longer than the third, first transverse cubital nervure curved inwards, second curved outwards, first recurrent nervure received close to the middle of the second cubital cell, second beyond two-thirds from the base of the third cubital cell. PortuGuEsE East Arrica: Kanezi, Mossurise district, vu. 1918 (C. F. M. Swynnerton). Bred from puparia of Glossina morsitans, 5 99, 1 ¢. Allied to M. glossinae, Turn., but easily distinguished by the shorter thorax and second tergite of the female ; and in the male by the much more distinct transverse carina at the base of the dorsal surface of the first tergite and the distinctly shorter second tergite. The two species are very closely related, the neuration of the males being very similar, both differmg in this and many other respects from M. benefactriz, Turn., the only other species which has yet been bred from Glossina. O29 NOTES ON THE IDENTIFICATION OF ANOPHELINAE AND THEIR LARVAE IN THE ZANZIBAR PROTECTORATE. By W. MANSFIELD-ADERS, Economic Biologist, Zanzibar Government. Medical Officers and others interested in tropical sanitation often find great difficulty in rapidly identifying adult mosquitos. Four Anophelines have been found in the Zanzibar Protectorate, two of these being well-known malaria carriers. The identification of the two latter with the aid of a pocket lens is quite simple. Adult Anophelines. Anopheles costalis, Lw. This is the commonest Anopheline in the Protectorate and the species most often brought in by the mosquito brigade for identification. It is a large insect and can at once be recognised by the golden yellow stippling on the first pair of legs, which is specially marked on the tarsal joints. Anopheles funestus, Guiles. Adults of this species are very uncommon in the town, but are constantly captured in the outlying districts. This Anopheline is easily recognised by its small size and generally black colour. There are no signs of stippling on the legs, therefore it cannot be confused with A. costalis. Anopheles mauritianus, Grp. Adults of this form are very rarely captured. It is one of the easiest species to identify, its chief characteristics being its large size, the very distinct long black spots on the costal vein, and especially the pure white tarsal joints of the hind pair of legs. Anopheles squamosus, Theo. This mosquito somewhat resembles Stegomyza fasciata in having the legs spotted with black and white markings, but the outstanding characteristic peculiar to it is the presence of lateral abdominal hair tufts. Anopheline Larvae. The identification of the Anopheline larvae is a more difficult problem, though three of them, A. costalis, A. funestus and A. mauritianus, have marked characteristics. 330 W. MANSFIELD-ADERS. Anopheles costalis, Lw. This is the commonest Anopheline larva. It is generally found in small collections of rain-water, such as road-side puddles, borrow-pits, water-holes (dug by the natives for storing water), shallow swamps surrounded by grass, occasionally in cement tanks, oftener in those with algid growth attached to their sides, at times in rain- water collected in dug-out canoes and boats. The larvae of this species vary greatly in colour, some being quite light, some dark, others almost green, the variations being caused by the type of food ingested. The larva has a small head, the hairs on the first three segments of the abdomen are not markedly feathered and the palmate hairs from the second to the seventh segment are not very conspicuous. Anopheles funestus, Giles. These larvae are generally found in backwaters of streams and at the edge of the sluggish rivers where the water is overhung by grass or other vegetation. Occasionally they are found in association with A. costals in the Zanzibar Fig 1. Thorax of larvae of Anopheles funestus, Giles (4th instar); a, one of the plumose hairs, greatly enlarged. Protectorate. The larvae are small and black in colour and much more heavily plumed than those of A. costalis. With a pocket lens a pair of thick feathered hairs IDENTIFICATION OF ANOPHELINAE AND THEIR LARVAE. ool springing from a chitinous pocket near the middle of the thorax are easily recog- nisable (fig. 1, a). This characteristic alone suffices to distinguish the larvae of A. funestus from those of the other island species. It must be remembered that the larvae of A. costalis carry the same hairs, but these are not nearly so conspicuous and are difficult to detect even with a lens. The palmate hairs are better defined than those of A. costalis. Anopheles mauritianus, Grp. The larvae are easily identified ; the antennae carry a conspicuous hair-tuft in the median region and there are two very characteristic fan-shaped tufts of short black hairs above the mouth-brushes. These two characters are sufficient for the identification of these larvae. The abdomen in both old and young larvae is marked with whitish bands. The living larvae have the peculiar habit of twisting themselves into an S-shape, a habit which I have not noticed in any other Anopheline larva. Anopheles squamosus, Theo. The larvae of this Anopheline are unknown to me. Technique and Methods of Control. A simple method of examining mosquito larvae is to remove one from the glass collecting jar by means of a pipette, place it in a drop of clean water on a microscope slide, and anaesthetise it by placing near it a pledget of cotton-wool soaked in chloro- form covering all with an inverted petri dish for a few minutes. The larva can then be moved about and turned by means of a fine dissecting needle with a minimum of injury to its finer hairy structures. The Public Health Officer will be greatly aided in his work by a mosquito brigade of native boys trained to recognise the larvae of the three genera Anopheles, Culex and Stegomyia. We supply our mosquito brigade with glass jars for collecting larvae and with long-handled white enamelled iron ladles, which make most efficient ‘dippers and in which the most minute larvae are easily seen against the white back- ground. A further valuable means of indicating the presence of mosquitos and of obtaining specimens of larvae is the use of mosquito traps. The traps used are ordinary wooden tubs about 24 feet in diameter cut down to. a height of 6 inches. These are sunk till their rims are flush with the ground, soil to the depth of some 2-inches is spread in them, water is poured on to the requisite depth and finally some aquatic weeds are floated on the surface. These traps, under the control of the mosquito brigade, are placed at intervals round the periphery of the town. The traps are examined every six days for the presence of larvae, all species being recorded. The method of examining the traps is as follows. Each trap is visited by an inspector on the sixth day, and its whole contents after agitation, are emptied into a bucket bearing the same number as the trap. The buckets are brought to the Health Office for critical examination. The contents of the buckets are ‘strained through fine muslin, the residue and debris left adhering to the muslin ‘being washed down into a large white enamelled dish, in which the larvae can easily 332 W. MANSFIELD-ADERS. be identified. The muslin is soaked and washed in clean water after each examination. Generally mixed infections of larvae are found, and these have proved of great use in training the staff of the mosquito brigade. All members of the mosquito brigade are trained to capture and identify adults, a most valuable adjunct to the larval index of a town, and the senior members are taught to recognise males and females of Anopheles, Culex and Stegomyia. For obtaining adults they are supplied with short glass tubes about six inches long and one inch in diameter; a plug of cotton-wool soaked with chloroform is placed in the bottom of each and the tube is corked. With this simple contrivance: resting adults can be caught with ease. The adults so collected are pinned out by their respective captors and on their return to the Health Office are submitted for examination. By these means we secure an ample supply of specimens for exchange: or for teaching purposes. The training of natives requires a great amount of patience, but if one is able to interest them in their work, the trouble spent on explaining the life-history and habits of the insects they are expected to control is well repaid. 330 THE MOSQUITOS OF FAR EASTERN PORTS WITH SPECIAL REFERENCE TO THE PREVALENCE OF STEGOMYIA FASCIATA, F. By A. T. STANTON, Bacteriologist, Institute for Medical Research, Federated Malay States. I. INTRODUCTION. The danger of the introduction of yellow fever into Oriental countries has been. the subject of much discussion since Sir Patrick Manson first drew attention to it in 1903. In 1911 Major (now Lt.-Col.) 8. P. James, I.M.S., was deputed by the Government of India to study the conditions in the endemic areas of Central America and in the principal sea-ports between that country and India. His report, published in 1913, was a most valuable contribution to our knowledge of the problems surrounding this subject. Among other recommendations, Major James suggested that further inquiry should be made with reference to the occurrence of Stegomyra fasciata in Far Eastern ports. Any measures that may be devised for the protection of India and the Far Eastern colonies from yellow fever must take account of the distribution and prevalence of the known carrier of the disease, perhaps also of its near allies, and of the sanitary and other circumstances affecting the prevalence of mosquitos generally. It was considered desirable to supplement the data hitherto obtained, and in September 1915 at the suggestion of the Imperial Bureau of Entomology I was instructed by the Government of the Federated Malay States to pursue these inquiries. The following ports were visited between October 1915 and March 1916 ; Bangkok (Siam), Saigon (Cochin-China), Haiphong (Tonkin), Canton (South China), Batavia, Samarang and Soerabaia (Java), Makasser (Celebes) and Tjilatjap (Java). Through the courtesy of several correspondents I have been able also to examine specimens. from other places in this region. It was not found possible to complete the observa- tions by a survey of the ports of China and Japan as had been planned. Reports of the observations (with specimens of mosquitos from the different ports) were sent to the Imperial Bureau of Entomology as the work progressed and these reports are now reproduced in a form convenient for reference. There is added a summary of the observations and of the conclusions that can be drawn from a study of them. - I am under obligation to many officers of the Governments of Siam, French Indo-China, and the Netherlands East Indies for generous assistance in connection with the enquiry, especially to Drs. M. Carthew, M.O.H., and Malcolm Smith (Bangkok), Dr. A. Denier (Saigon) and Drs. W. T. de Vogel, N.F. Lim, H. Werkman, L. 8S. von Romer and M. L. van Breemen (Java). To Dr. G. A. K. Marshall, Director of the Imperial Bureau of Entomology, and Mr. F. W. Edwards, of the British Museum, I am indebted for confirmation or correction of my identifications of mosquitos. 334 of A. T. STANTON. If. Tur Ports VisitEp. Bangkok. Bangkok, the modern capital of Siam, is situated on both banks of the River Menam, in latitude 13° 45’ North and longitude 100° 30’ East, about twenty-five miles up-stream from the bar at the river’s mouth. The city proper, the government offices, banks and business houses are on the left bank of the river ; on the right bank are rice mills and other industries and the dwellings of a large part of the native population. The city is intersected in all directions by a network of canals. The estimated population is 625,000. T'rade.—The trade of the port is mainly with Singapore, Hong Kong, Swatow and Hoihow. There is also direct steamer connection with Saigon and Batavia. Temperature and Rainfall—The monthly mean temperature of Bangkok is above 80°F. for all months of the year except November, December and January. The annual mean rainfall is about 57 inches. The months May to October are wet months ; December, January and February are dry. Water Supply.—A public water supply for Bangkok was installed in November 1914. The water is taken from the Menam about twenty-five miles north of the city, carried by a canal to the filtering plant at Bangkok and thence distributed in pipes over the greater part of the city on the left bank of the river. In addition to this public supply, which is as yet available for only a small part of the population, there are artesian wells, shallow wells and the river and canals, all of which are drawn upon. The storage of water in jars for household purposes is the rule in Bangkok among the poorer classes. Samitation and Quarantine.—The sanitary administration of the city is controlled by the Ministry of Local Government. Provision is made for three European medical officers, a Bacteriologist, and a number of Siamese assistants. No special measures are in operation for the reduction of mosquitos. This department also deals with quarantine matters. The quarantine station, is situated at Koh Phra, an island about 60 miles from Bangkok. It is proposed to construct a new station on the river about 15 miles from the city. Mosquitos—An investigation was made in late October and early November of the distribution of mosquitos in all parts of the city by examining specimens taken in houses and by the identification of larvae taken in artificial and natural collections of water in the neighbourhood of houses. The following list of species is in order of comparative frequency of their occurrence. | (A). Adult mosquitos taken in houses :—Culex fatigans, Wied. ; Stegomyia fas- ciata, F.; Stegomyia albopicta, Skuse (scutellaris Theo.) ; Mansomoides uniforms, Theo. ; M. annulipes, Walk. ; Culex gelidus, Theo. ; C. concolor, R. D.; C. sitrens, Wied. ; Mansonioides annuliferus, Theo. ; Anopheles sinensis, Wied.; A. senensis var. peditaeniatus, Leic. ; A. rossi var. indefimtus, Ludl.; A. barbirostris, Wulp ; A. fuliginosus, Giles; A. fuliginosus var. nivipes, Theo. ; Culex halifaxi, Theo. ; C. fuscocephalus, Theo. ; Uranotaenia cancer, Leic. ; Anopheles tessellatus,.Theo. MOSQUITOS OF FAR EASTERN PORTS. 335 (B). Larvae from artificial collections of water: jars, tubs, barrels, tins, etc. :— Stegomyia fasciata ; Armigeres obturbans, Walk. ; Stegomyia albopicta (scutellaris) ; Cyathomyra brevipalpis, Giles ; Toxorhynchites immasericors, Walk. ; Culex halifax: ; C. fategans ; Anopheles rossi var. indefinitus. (C). Larvae from natural collections of water: pools, canals, ditches, etc. :— Culex fatigans, C. geludus, Armigeres obturbans, Culex concolor, Anopheles rossi var. ndefinitus, A. sinensis. The cannibal larvae of Toxorhynchites immisericors, Walk., were here frequently encountered in water-jars, and their astonishing voracity for Stegomyia larvae, as many as twenty-five of the latter being devoured in a single night, made one think of the possibility of their utilisation in Stegomyia reduction. In Bangkok through- out the city mosquitos are numerous and Stegomyra fasciaia is abundant. The habits of the people in the storage of water and the restricted distribution of the public water supply combine to make the task of its reduction a very difficult one. Saigon. Saigon, the capital of Cochin-China, is situated on the banks of the River Saigon, a triobutary of the Donnai, in latitude 10° 46’ North and longitude 106° 30’ East. The town lies about 40 miles up-stream from Cape St. James at the mouth of the river Donnai and is accessible to the largest steamers. The population of Saigon is about 70,000. The city of Cholon, which for practical purposes is a part of the port, is situated about four miles further up-stream and is connected with Saigon by a steam tramway. Cholon is the centre of much commercial activity in con- nexion with the milling and export of rice, the staple product of Cochin-China. The population of Cholon is 165,000. Trade.—The trade of the port is with Hong Kong, Singapore, Manila, Japan and the Netherlands Hast Indies by direct services, and with EHurope by way of intermediate ports. Temperature and Rainfall—The monthly mean temperature is above 80°F. for all months of the year except December, when it falls shghtly. The annual mean rainfall is about 65 inches. The raimy season begins about the middle of May and ends about the middle of November, January to March being the driest months. Water Supply.—The public water supply of Saigon is collected from a number of surface wells within the city and from a small catchment area about two miles outside it. The supply is now intermittent, necessitating storage in tanks or jars. A plan for a continuous high pressure supply is under consideration. Samtation and Quarantine.—The sanitary administration is controlled by the Civic Corporation through their medical officer of health, who is assisted by sanitary inspectors. Though no special measures are directed against mosquitos, the efficient civil sanitary service ensures that breeding places are not neglected. The quarantine service is directed by the Chief of the Health Service of Cochin China, an officer of the Sanitary and Medical Services of Indo-China. The quarantine station is situated at Nhabeé, fifteen miles down-stream from the city ; the buildings, accommodation and equipment are excellent. 336 A. T. STANTON. These services throughout Indo-China are well maintained. All the principal ports and many of the smaller ones are provided with rooms for disinfection and mobile apparatus for the disinfection of ships. Mosquitos.—The following species were met with, given in order of frequency. (A). Adult mosquitos taken in houses :—Culex fatigans, Stegomyia fasciata, S. albopicta (scutellaris), Armigeres obturbans, Mansoniordes uniformis, Culex gelidus, C. sitiens, C. concolor, Anopheles rossi var. indefinitus, A. sinensis. (B). Larvae taken in artificial collections of water :—Stegomyia fasciata, S. albopicta (scutellaris), Ochlerotatus gubernatorius, Giles, Cyathomyia brevipalpis, Culex concolor. (C). Larvae taken in natural collections of water :—Culex fatigans, C. gelidus, C. concolor, A. rossi var. wndefimtus. At the time of my visit, in November towards the end of the rainy season, mosquitos were not numerous in Saigon, but Stegomyra fascrata was not uncommonly met with. In the native city of Cholon larvae of this species were common. In the annual report for 1914 of the “ Ambulance du Cap St. Jacques ” Major J. H. Rencurel describes an outbreak of fever of obscure causation among officers and their families. A further small outbreak at Cape St. James occurred in 1915 and similar cases were observed at Saigon. Major Rencurel identified this fever with the “six day fever of ports’ described by Leonard Rogers and attributed its spread to a biting insect, mosquito or fly. Certain officers who had experience of the fever and of the conditions at Cape St. James entertained the suspicion that the disease might be related in some way to yellow fever. Haiphong. Haiphong is the shipping port for Hanoi and other commercial centres in Tonkin and for the Chinese province of Yunnan. It is situated in latitude 20° 50’ North and longitude 106° 42’ East, on the banks of the rivers Cua Cam and Song Tam Bac. These rivers are joined by several channels with the river Song Koi, a great water- way connecting Yunnan with the Gulf of Tonkin. The banks of the rivers are low and consist of alluvial mud, from which the present site of the town of Haiphong has been reclaimed. Scattered throughout the town are extensive pools and areas of marsh land. The population of Haiphong is about 45,000. Trade.—Haiphong is accessible to large ocean-going steamers. The sea-borne trade of the port is mainly with Hong Kong, Saigon, Pakhoi in the Chinese province of Kwang-si, and Hoihow on the island of Hainan. Temperature.and Ranfall.—The monthly mean temperature rises above 80°F. only during the months June to September. From December to February it is about 62°F. The annual mean rainfall is about 69 inches, May to October being the rainy months. Water Supply.—The public water supply is a continuous high pressure supply drawn from storage reservoirs in the hills and distributed over the whole area of the city. Samtation and Quarantine-—These services are controlled by the director of Health Services of Tonkin, whose headquarters are at Hanoi and who is represented at Haiphong by an official of the department. MOSQUITOS OF FAR EASTERN PORTS. 337 The quarantine station is at Binh-Dong, about ten miles down-stream. There is a complete equipment of buildings, materials, and staff. Mosquitos. (A). Adult mosquitos taken in houses :—Culex fatigans, Stegomyia fasciata, S. albopicta (scutellaris), Armigeres obturbans, Mansomordes annulrferus, M.annulupes, Culex sitiens, Anopheles rossi var. indefimtus, A. sinensis, A. minimus, Theo. (B). Larvae taken in artificial collections of water :—Stegomya fasciaia, S. albopicta (scutellaris), Armageres obturbans. (C). Larvae taken in natural collections of water :—Culex fairgans, C. satiens. At the time of my visit to Haiphong in December the temperature was low, 64°— 72°F., and mosquitos were few in number. A few specimens of Stegomyia fasciata were taken. Dr. P. L. Simond, Inspector General of the Sanitary and Medical Services of Indo-China, reports (Bull. Soc. Med. Chir. Indochine, vii. 2,1916) that he found a specimen of Stegomyra fascrata in Hanoi, Tonkin, in November 1915, but considers that the species is rare in this region. Dr. Simond has also made observations in Cochin-China, where he found specimens of Stegomyia fasciata, but concludes that it is a rare species. This conclusion was not borne out by my examination of breeding places in Saigon. Canton. Canton, the capital of the Chinese Province Kwang Tung, is situated on the Chu Kiang or Pearl River in latitude 23° 7’ North and longitude 113° 14’ Hast. The city proper extends to a breadth of about two miles and is about six miles in circumference, being enclosed by a wall. The suburbs extend along the river for about five miles. The population has been estimated by the Customs authorities at 2,500,000. Trade-——Ample means of communication exist between Canton and Hong Kong, a distance of 95 miles by water and 112 miles by railway. Foreign steamers and a large number of native craft ply daily. There is daily steamer communication with Macao, and regular communication with Wuchow and other West River towns and with Shanghai and other Chinese ports. Ocean-going vessels of considerable draught can proceed to Canton. Temperature and Ramfall—The monthly mean temperature exceeds 80°F. only during the four months June to September. During the months December to February it falls to about 60°F. The annual mean rainfall is about 65 inches, April to August being the rainy months. Water Supply.—tThere is a public water supply of limited extent brought from the hills eight miles from the city, but both within and without the city walls there are large numbers of wells. The practice of storing water in jars is very general. Sanitation and Quarantine.—In the Chinese city little is attempted in the way of sanitation as understood in Europe. The Municipality of Shameen (British Concession) employs a medical officer of health. 338 A. T. STANTON. There is a quarantine anchorage down river from Canton, but there are no buildings or staff to deal with infected ships. The Medical Officer of Health, Shameen, has assigned to him by the Chinese Customs Department some duties in connexion with quarantine matters, but these duties appear to be ill-defined. Mosquitos. (A). Adult mosquitos taken in houses :—Culex fatigans, C. sitiens, Armigeres obturbans, Stegomyia albopicta (scutellaris), Anopheles sinensis. (B). Larvae from artificial collections of water :—Culex fatigans, Stegomyia albopicta (scutellaris), Armageres obturbans. (C). Larvae from natural collections of water :—C. fatogans. The weather conditions were unfavourable for the development of mosquitos during the time of my visit. No Stegomyia fasciata, either larvae or adults, were found. S. albopicta larvae were present in fair numbers in jars both within and without the city, but their development was slow. Larvae of this species taken at Canton on 13th December 1915, which were apparently then half-grown, transformed to pupae on 6th January 1916 in Kuala Lumpur and a few adults hatched out 8th-10th January 1916. . Since my visit I have had sent to me collections of mosquitos taken in Canton during the month of March. Among these were no specimens of Stegomyia. Batavia. Batavia, the principal city in West Java, is situated on the Bay of Batavia in latitude 6° 7’ South and longitude 106° 48’ East. It is a long narrow city extending along the banks of the river Tjiliwong and comprises the port (Tandjong Priok), the old town (Batavia), and the new town (Weltevreden). The population is about 138,000. Trade——Batavia has extensive trade relations with all the neighbouring ports and is the point of departure for lines of steamers to Australia, EKurope, America (by way of the Philippine Islands) and Japan. Temperature and Rainfall—The annual mean temperature is about 79°F. The temperature is nearly the same throughout the year, the greatest difference between the highest and lowest monthly mean being only about 2°. The annual mean rainfall is 71 inches, October to April being the wet months. Water Supply—The public water supply is drawn from artesian wells and is distributed throughout the city by means of pipes to hydrants and private houses. From the hydrants water is carried in tins to the houses in the native quarters and 13 there stored in earthenware jars. There are also large numbers of open wells, : A scheme for a constant high pressure supply of water from the hills has been under consideration for some years. Sanitation and Quarantine.—Special officers devote their whole time to sanitary work and important progress has been made during recent years in improving the condition of the native quarters of the city. MOSQUITOS OF FAR EASTERN PORTS. 009 There is a special staff which devotes attention to anti-mosquito measures. Native assistants are taught to recognise the different kinds of larvae and to deal with breeding places. By means of posters in the vernacular languages, cinema- tograph films and lectures, the people are taught the importance of mosquito- borne diseases and the means of preventing them. The quarantine station is situated on two islands in the Bay of Batavia about 6 miles from the town. It is supplied with ample accommodation to meet any emergency. There is a complete equipment of apparatus for disinfection. Mosquitos. (A). Adult mosquitos taken in houses :—Culex gelidus, C. tritaenorhynchus, Giles, Stegomyia fasciata, Mansonioides annulipes, M. annuliferus, S. fasciata var. luciensis, Theo., Mansonioides umformis, Culex fatigans, Armigeres obturbans, Stegomyra albopicta (scutellaris), Culex whitmorer, Giles, C. vishnui, Theo., Aédomyta catasticta, Knab, Anopheles rossi var. indefimtus, A. sinensis, A. ludlowi, Theo., A. fuliginosus, Banksinella lineatopennis, Ludl., Uranotaenia campestris, Leic., Culex bitaenorhynchus, Theo., Rachionotomyia aranoides, Theo., Lophoceratomyva barkert, Theo., Culex fuscocephalus, Cyathomyia brevipalpis. (B). Larvae from artificial collections of water :—Stegomyia fasciata, S. fascrata var. luciensis, Culex fatigans, Armigeres obturbans, S. albopicta (scutellaris). (C). Larvae from natural collections of water :—Culex fatigans, C. vishnu, C. tritaenorhynchus, Anopheles tessellatus, A. koch, A. rossi var. indefinitus, A. sinensis, A. ludlowi, A. barbirostris. At the time of my visit to Batavia mosquitos were very numerous, as may be inferred from the large number of species recorded. Stegomyia fasciata was pre- valent in all parts of the city. Special measures are in operation to reduce the numbers of mosquitos; the task, which is under highly competent direction, is a’ formidable one. Samarang. Samarang, the principal port of Mid-Java, is situated on the north coast about midway between Batavia and Soerabaia, in latitude 7° 15’ South and longitude 110° 38’ East. The population is about 96,000. Trade-—Samarang is the commercial centre of a very large agricultural district. There are rail connections to all points in Java and many steamships touch at this port on their way to Singapore, Australia, China, India, Hurope and America. Temperature and Rainfall—tThe climate of Samarang resembles that of Batavia. The annual mean rainfall is 86 inches. Water Supply—The public water supply is drawn from the hills. It is a con- stant high pressure supply and is distributed over the whole area of the city. Samtation and Quarantine.—These are supervised by officers of the Civil Medical Service. Samarang is a second-class harbour and the quarantine station is not provided with its own apparatus for fumigation. A new quarantine station is projected. There are no special measures in force directed against mosquitos. 340 A. T. STANTON, Mosquitos. (A). Adults taken in houses :—Culex fatigans, Stegomyia fasciata, S. fasciata var. luciensis, Uranotaenia cancer, Anopheles ludlowi, C. tritaeniorhynchus. (B). Larvae from artificial collections of water :—Stegomyia fasciata, S. fasciata var. luciensis, Culex fatigans. (C). Larvae from natural collections of water :—Culex fatigans, C. bitaennorhynchus, Anopheles ludlowi, A. barbirostris. At the time of my visit mosquitos were not numerous at Samarang and Stegomyia fasciata was rarely encountered in houses. Larvae of this species were found in jars used for storing water in houses. The admirable public water supply is, I think, an important factor in keeping down the numbers of mosquitos, which were less numerous than might have been expected. Dr. W. T. De Vogel published in 1909 (Geneesk. Tijd. v. Ned. Ind. xlix, no. 5) an account of malaria infection experiments with Anopheline mosquitos bred from larvae taken in salt-water pools near the sea-coast at Samarang. These mosquitos were then identified as Anopheles (Myzomyia) rossi, Giles, and Dr. De Vogel’s results gave rise to some controversy, as it had been thought that A. rossa was not a malaria-carrying species. During my visit to Samarang I re-examined these pools, but found in them only Anopheles ludlowi, Theo., and A. barbirostris, Wulp. At Soerabaia, however, in similar situations I took large numbers of larvae of Anopheles rossi, and I think it quite probable that Dr. De Vogel experimented with this latter species. Soerabaia. Soerabaia is situated on the Kali Mas and its tributary the Kali Pegirian in latitude 7° 14’ South and longitude 112° 44’ East. It is a long narrow city stretch- ing along the banks of these rivers. Population 160,000. Trade.—Soerabaia is the commercial metropolis of Java, and the greater part of the foreign trade is executed from this port. Regular steamship services are maintained with Makasser and other ports in Netherlands East Indies and with Australia, India and Hurope. Temperature and Rainfall—The climate of Soerabaia resembles that of Batavia and other coast towns. The annual mean rainfall is 68 inches. Water Supply.—The public water supply of Soerabaia comes from the hills. It is a constant high pressure supply and is distributed over the whole city. Dr. J. T. Terburgh, Chief of the Public Health Service of Kast Java, has during several years studied the important question of supplying the native population with pure water at small cost. This problem, of vital importance from the point of view of water-borne diseases such as cholera, is also intimately connected with the question of Stegomyia reduction in sea-port towns. Dr. Terburgh was good enough to demonstrate to me the methods which he advocates and to show me their practical working in a native kampong of Soerabaia. Water is brought to each house in pipes and by means of a permanent water- meter, devised by Dr. Terburgh, a quantity of water, estimated to be sufficient for drinking and domestic purposes, is delivered at a uniform rate, The water is stored MOSQUITOS OF FAR EASTERN PORTS. 341 in cement reservoirs of appropriate capacity, whence it can be taken as required. Dr. Terburgh claims for his method that an adequate supply of water is delivered, that wastage is prevented, and that after making provision for interest on capital expenditure and for repayment, the cost to the consumer may be less than he now pays for the transport of water from hydrants. Owing to the more or less constant disturbance of the water surface in these reservoirs there is less tendency for the female Stegomyra fasciata to lay eggs in them than in the common water-jar. Nevertheless larvae of this species were not uncommonly found in the reservoirs. It was observed also that despite the existence of an adequate supply of water at all times in the reservoir, the natives removed water from them and stored it in jars in accordance with a custom that obtains everywhere throughout Eastern countries. In his experiments Dr. Terburgh had not in mind the question of the reduction of breeding places of Stegomyra, but I understand that he proposes to carry out further experiments in an endeavour to meet this indication also. I was much impressed by the possibilities revealed by Dr. Terburgh’s proposals for the solution of the problem of mosquito control in the densely populated towns of the Hast. Samtary Service and Quarantine.—Special officers are in charge of these services and they are provided with subordinate stafis who are now being instructed in the details of anti-mosquito work. A new quarantine station has been approved and is to be built on the island of Madura opposite Soerabaia. Mosquitos. (A). Adults taken in houses :—Culex fatigans, Stegomyia fasciata, S. fasciata var. luciensis, C. sittens, C. vishnur, C. minimus, Anopheles rossi, A. rossi var. indefinitus, Culex bitaeniorhynchus, Anopheles barbirostris. (B). Larvae from artificial collections of water :—S. fascoata, C. fategans. (C.) Larvae from the water collected in cut bamboos :—S. albopicta (scutellaris) ; Rachonotomyia aranoides. (D). Larvae from natural collections of water :—C. fatigans, C. vishnus, C. sitrens, C. bitaemorhynchus, Anopheles rossi var. indefimtus, A. rosse. Mosquitos were numerous in Soerabaia and S. fasciata was a common species. The problem of their reduction is receiving special attention. Makasser. Makasser, the capital of Celebes Island and its dependencies, is situated on the west coast of Southern Celebes in latitude 5° 20’ South and longitude 119° 40’ Hast. Population 26,000. Trade.—Makasser is a trade centre for Eastern Netherlands East Indies and a port of call for steamships proceeding to Australia, Manila and Hong Kong. Temperature and Rainfall—The mean temperature is about 80°F. The climate is less equable than that of Java and the daily variation may be as much as 18°F. The annual mean rainfall is 115 inches. 342 A. T. STANTON. Water Supply.—The water supply is drawn from surface wells in compounds. The question of a piped water supply has been mooted, but the expense involved is great. Sanitation and Quarantine —These services are supervised by the Military Medical Authorities. There are no special measures in force directed against mosquito breeding. Mosquitos. (A). Adults taken in houses :—Culex fatigans, Stegomyia fasciata, C. tritaeniorhyn- chus, C. vishnu, C. fuscocephalus, Anopheles rossi var. indefinitus, A. sinensis, A. barbirostris. (B). Larvae from artificial collections of water :—Stegomyia fasciata. (C). Larvae from natural collections of water :—C. fatigans, C. vishnui, A. rossi var. undefimtus. . -(D). Larvae in cut bamboos:—S. albopicta (scutellaris), Rachionotomyra aranotdes. Mosquitos are exceedingly prevalent at Makasser and Stegomyia fasciata very numerous, but no special measures were in operation for their destruction. There were large numbers of mosquito larvae in the cement reservoirs used for storing water in bath-rooms. III. OrHER RECORDS OF STEGOMYIA. During the course of this inquiry I have received a number of collections of mos- quitos from different parts of the Malay Pensinsula and neighbouring countries. Among these the new records of the occurrence of species of Stegomyia are as follows :— 1. Stegomyia fasciata, F. Malay Peninsula : Kelantan (Tumpat) ; Kedah (Alor Star) ; Pahang (Kuantan). Borneo : British North Borneo (Jesselton) ; Pulau Laut (Stagen). Java: Tegal; Djokjakarta ; Tjilatjap ; Garoet ; Billiton I. Sumatra : Djambi; Padang ; Indrapura ; Pulau Weh. Celebes : Boetan I.; Soembawa I. New Guinea: Port Moresby. 2. Stegomyia albopicta, Skuse. Malay Peninsula: Kelantan ; Kedah ; Ginting Simpah, Selangor (2,000 {t.). Siam : Krabin ; Koh Chang I. Borneo : British North Borneo. Java: Djokjakarta ; Tjilatjap ; Biliton I.; Lombok I. Celebes: Boetan I. Sumatra : Djambi ; Indrapura ; Padang ; Palambang. 3. Stegomyra variegata, Dol. New Guinea : Port Moresby. 4. Stegomyia fusca, Leic. Siam : Krabin ; Koh Chang I.; Koh Mohsi I.; Koh Kra I.; Koh Klun I.; Koh Rang I.; Koh Kut I. Malay Peninsula : Kuala Lumpur. MOSQUITOS OF FAR EASTERN PORTS. 343 5. Stegomyia W-alba, Theo. Malay Peninsula : Kuala Lumpur. Siam : Bangkok. Sumatra : Pulau Weh. 6. Stegomyia trilineata, Leic. Malay Peninsula: Ulu Gombak, Selangor. 7. Stegomyia desmotes, Giles. Malay Peninsula : Ulu Gombak, Selangor. 8. Stegomyra albolmeata, Theo. Malay Peninsula: Kuala Lumpur. TV. GENERAL REMARKS. Mosquitos on Shvps.—It was pertinent to the subject of my enquiry to endeavour to learn to what extent mosquitos were in fact conveyed on ship-board from port to port. In the course of this enquiry, I travelled on a number of ships, some of them of recent construction, some of them old. Careful examination was made in port and at sea. Once only was Stegomyia fasciata observed, and on that occasion the ship lay in a river on both banks of which this species was breeding in very large numbers. Occurrence of S. fasciata inland.—It is often stated that in the East S. fascrata is found on land only near the sea coast. It is therefore of some importance to note its occurrence at inland points in Java and the Malay Peninsula. This species was taken at Garoet, in West Java, 40 miles from the sea and 2,300 feet above sea- level. Within the past ten years S. fasciata has replaced S. albopicta (scutellaris) as the common Stegomyia species in Kuala Lumpur, an inland town of the Malay Peninsula 27 miles from the sea. - Trade Routes —The old trade routes between America and the Orient have been little affected by the opening of the Panama Canal to traffic. These routes still pass northward by way of Honolulu, Japan and China. There is little direct traffic within the tropical zone. : V. SUMMARY AND CONCLUSIONS. 1. In Far Eastern ports of the equatorial region the yellow fever carrier, Stegomyia jasciata, is prevalent at all seasons of the year. Even where efforts have been made to reduce its numbers it is common; where no such efforts have been made, it is a veritable plague. In this region the conditions are highly favourable for the spread of yellow fever if it should once be introduced. 2. The old trade routes between America and the Orient have not so far been altered with the opening of the Panama Canal to traffic, these routes still pass northward by way of Honolulu, China and Japan. It was found in the course of this enquiry that in ports situated towards the northerly limit of the Eastern tropical belt S. fasciata occurred in relatively small numbers in the colder months or was not met with at all. 344 A. T. STANTON. It is thought probable that a survey of the ports of China and apan would show that the conditions there are unfavourable for the propagation of S. fasciata even in the warmer months. If this supposition should prove correct it would be shown that the path for the conveyance of yellow fever infection to the Orient is cut at this point. It is therefore important that a survey of these ports should be undertaken in completion of the observations here recorded. 3. Systematic observations on a number of ships within the tropical belt failed to disclose the presence of S. fascoata on board ships at sea, even when coming from ports heavily infested with this mosquito. It issuggested that too much importance has hitherto been attached to the possibility of the conveyance of infected mosquitos over long distances in ships, and that with modern steel ships the danger is minimal. 4. With the prospective increase in sea-borne traflic it is more than ever imperative that measures for Stegomyia reduction in Far Eastern ports should be vigorously pursued. Examples of effective action are not wanting,and the work of James and his colleagues in Colombo has shown what may be accomplished in this direction. 5. Trained entomologists should be engaged to give their whole time to the study of the highly specialised problems connected with mosquito reduction and to the supervision of measures. These officers should work in close co-operation with the executive public health authorities. 345 COLLECTIONS RECEIVED. The following collections were received by the Imperial Bureau of Entomology between Ist July and 30th September, 1919, and the thanks of the Managing Committee are tendered to the contributors for their kind assistance :— Dr. W. Mansfield-Aders, Government Economic Biologist :— 6 Culicidae and 53 other Diptera ; from Zanzibar. Capt. P. J. Barraud :—150 Culicidae, 4 Phlebotomus, 1 Hippoboscid, 20 other Diptera, 19 Hymenoptera, 32 Coleoptera, 7 Planipennia, 15 Lepidoptera, 11 Rhynchota, 7 Orthoptera, 1 Mayfly, and 17 Odonata ; from Palestine. Mr. G. E. Bodkin, Government Economic Biologist :—7 Diptera, 1 Flea, 26 Termites, 160 Ants, 8 other Hymenoptera, 39 Coleoptera 4 Lepidoptera, 22 Rhyn- chota, 1 Grasshopper, 23 Ticks, 4 Millipedes, 1 Scorpion, and 3 Worms ; from British Guiana. Dr. C. K. Brain :—A number of Chalcids ; from Surrey. Capt. P. A. Buxton :—15 Culicidae, 1 Tabanus, 1 Hippoboscid, 12 Lepidoptera, and 10 Rhynchota ; from N. W. Persia. Division of Entomology, Pretoria :—4 Hippoboscidae, 38 other Diptera, 12 Hymenoptera, 162 Coleoptera, 71 Rhynchota, and 20 Orthoptera; from South Africa. Mr. E. Melville Du Porte :—5 Coleoptera ; from Canada. Durban Museum :—41 Coleoptera ; from Natal. Dr. Eustace W. Ferguson :—17 Tabanidae : from New South Wales. Dr. Lewis H. Gough, Government Entomologist :—3 Trypetid Diptera ; from Egypt. The Government Entomologist, Madras :—113 Diptera, 12 Hymenoptera, 14 Coleoptera, 1 species of Coccidae, and 6 other Rhynchota ; from Southern India. Mr. C. C. Gowdey, Government Entomologist :—56 Diptera, 1 tube of Braconid parasites, 110 other Hymenoptera, 70 Coleoptera, 11 Lepidoptera, 20 Rhynchota, 5 Orthoptera, and 2 Odonata ; from Uganda. Mr. KE. Hargreaves :—52 Culicidae, 5 Tabanidae, 1 Hippoboscid, 1 tube of Chalcids, 5 Coleoptera, and 4 Rhynchota ; from Italy. Mr. William Harris, Government Botanist :—A species of Coccidae, and 41 Cole- optera associated with it ; from Jamaica. Dr. C. Gordon Hewitt, Dominion Entomologist :—20 Psyllidae ; from Nova Scotia. Mr. C. W. Hobley :—5 Haematopota, 9 Tabanus, 1 Glossina, and 1 Tachinid ; from British Hast Africa. The Imperial Departmant of Agriculture, West Indies :—24 Melolonthid beetles, 11 larvae, and 11 pupae ; from Antigua, etc. Mr. H. H. King, Government Entomologist :— 3 Culicidae, 1 Chrysops, 3 Tabanus, and 30 other Diptera ; from the Anglo-Egyptian Sudan, 346 COLLECTIONS RECEIVED. Mr. A Loveridge :— 30 Glossina, 449 other Diptera, 56 Dipterous larvae, 49 Dipterous pupae, 93 Fleas, 109 Thysanoptera, 1,471 Ants, 71 other Hymenoptera, 2,252 Coleoptera, 332 Coleopterous larvae, 93 Coleopterous pupae, 40 Lepidoptera, 179 Lepidopterous larvae, 12 Lepidopterous pupae, 2 Caddis-flies, 282 Termites, 9 species of Coccidae, 95 Cimicidae, 500 Aphididae, 852 other Rhynchota, 20 Odonate nymphs, 5 May-flies, 21 Mallophaga, 4 Anoplura, 29 Ticks, 400 Mites, 99 Spiders, 17 Scorpions, 3 Centipedes, 41 Millipedes, and 27 Worms; from Portuguese and “ German ”’ Hast Africa. Dr. R. E. McConnell :— 1 Tabanus, 30 other Diptera, and 4 Rhynchota; from Uganda. Dr. J. W. Scott Macfie :— 438 Culicidae, 104 Phlebotomus, 5 other Psychodidae, 17 Ceratopogoninae, 8 Cecidomyiidae, 2 Tabanus, a number of other Diptera, 7 Hymenoptera, 5 Coleoptera, 21 Lepidoptera, 20 Termites, 3 Rhynchota, 3 Orthoptera, and samples of grain attacked by insects ; from the Gold Coast. Capt. Malcolm E. MacGregor :—-54 Anopheles maculipennis, with a number of eggs ; from Kent. Mr. Arthur W. J. Pomeroy, Government Entomologist, Nigeria :—17 Tabanidae, and 354 other Diptera; from the United States of America: 24 Hippoboscidae, 35 other Diptera, 16 Chalcids, 57 other Hymenoptera, 287 Coleoptera, 72 Lepidoptera, 75 Rhynchota, 11 Orthoptera, and 15 Odonata ; from Southern Ni geria. Dr. J. O. Shircore :—5 Diptera and 1 Bug ; from British East Africa. Mrs. W. Smith :—19 Culicidae, 2 Glossina, 49 other Diptera, 11 Hymenoptera, 12 Coleoptera, 16 Rhynchota, and 9 Orthoptera ; from Kumasi, Gold Coast. Dr. F. H. Storey :—181 Culicidae, 10 Tabanidae, 163 Glossina, and 30 other Diptera ; from Koforidua, Gold Coast. Mr. C. F. M. Swynnerton :— 19 Glossina and their pupa cases, 3 other Diptera, and 6 Hymenoptera ; from Portuguese Hast Africa. Mr. F. V. Theobald :— 1 Orthopteron ; from Mesopotamia. Mr. R. Lowe Thompson :—113 Coleoptera ; from Southern*Rhodesia. Mr. F. W. Urich, Government Entomologist :—33 Culicidae, and 24 Mutillid Hymenoptera; from Trinidad. Mr. F. C. Willcocks, Entomologist to the Sultanic Agricultural Society :—24 slides of Mallophaga, Anoplura, and Acarina ; from Kgypt. Mr. C. B. Williams :—54 Chrysops and 45 Tabanus ; from Trinidad. é _% IMPERIAL BUREAU. OF ‘ENTOMOLOGY, “HON NORARY OOMMITTEE Or MANAGEMENT THE VISCOUNT Bue a ne _ Limvr.-Coronen A. ALCOCK, C.LE., ERS. peepee Maton E.R AUBTEN, DS.0/% 0 yo se cc aa Dr. A. G. BAGSHAWE, C.M.G. : : | | Re Mayor-Genzrat Siz JOHN R. BRADFORD, K.0.M.G., ERS. val Ee Masor-Generat Sir DAVID BRUCH, K. C. B., BRR es Met, CROP M RR) oy ea Ree $m §. BF. HARMER, KBE. PRS. ee | fie | Prov. H. MAXWELL LEFROY. Hon: E. LUCAS. | De. R. STEWART MACDOUGALE. Sm JOHN McFADYEHAN. i ‘Sm PATRICK MANSON, OMG. ERS. ) Sm DANIEL MORRIS, K.C.M.G. - Pror, R. NEWSTEAD, F.RS. Pror. G. H. F. NUTTALL, F.RS. Prov. E. B. POULTON, F.R.S. | at ns Lizut.-Cotone, Sir DAVID PRAIN, OMG., OLE, BRS his aie a Sm H. J. READ, K.0.M.G., CB. Pee fue Hon. N. C, ROTHSCHILD. ; Dr. HUGH,SCOTT. a He Sin A. E. PLEY, G.B.E., PRS. he Mn R. A. C. SPERLING, CMG. '.Sm STEWART STOCKMAN. Me. F. V, THEOBALD. | | | ? We Mr. C.\ WARBURTON. | | fees enna a ee sit 3 yet a Director. 4 Dr, GUY A. K. MARSHALL ah Assistant . Director. har ar. 8): A. NEAVE. ‘Secretary. oy : Mr. ae Gi: Cc PAREINSON. 347 INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. By W F. Fiske. INTRODUCTION AND GENERAL DISCUSSION. These studies were made in Uganda on the islands and shores of Victoria Nyanza, In territory which had been depopulated some years before on account of tsetse-fly and sleeping sickness. They were begun in October 1913, and extended over two full years, of which eighteen months were spent in the fly belt on Victoria Nyanza, and three months on a tour into Bunyoro, where for the time being (the spring of 1914) Glossina morsitans was the centre of greater attraction as a possible vector of human trypanosomiasis. Seven tours were made from headquarters at Kntebbe— the shortest, of three days only, being interrupted by the War, and the longest, of eight full months, being much prolonged by the War. The working basis unquestioningly accepted in the beginning was that the measures adopted in 1906-07 for the suppression of sleeping sickness in Uganda were both wise and necessary. These measures called for complete severance of contact between the fly and the native populations so long as infection was present. They had contemplated originally the elimination of infection through complete severance of contact between fly and population, but in this particular had failed of their object, for the trypanosome persisted both within and without the depopulated zone; as a parasite of game and fly within it, and as a human parasite in the tiparian populations of adjoming sectors of the lake region without it. They therefore required the extermination of fly in populated districts, or as a prerequisite to reclamation and repopulation of territory from which the inhabitants had been removed. They were exactly such measures as are absolutely required for suppression and prevention of the cattle trypanosomiasis transmitted by Glossina morsitans in other parts of Uganda; and going on the perfectly natural assumption that a close parallel existed between human and bovine trypanosomiasis, as transmitted by Glossina palpalis and Glossina morsiians, respectively, the wisdom and necessity of the measures was not questioned. The economic or practical objective of the studies was the sanitation and reclamation of the fly-infested territory ; the more immediate and technical objective, therefore (proceeding on the above assumption), was the extermination of fly as a sanitary prerequisite to reclamation. To exterminate a species is to restrict its range or distribution. Therefore, the most immediate objective was made that of identifymg and studying those factors in the bionomics or “ control’’ of the species which operate in nature, or which might be operated by artifice, to delimit -and determine or to “control” its range. (659) Wt.P4/140. 1,000 5.20 B.&F.,Ltd. Gp. 11. A 348 W. F. FISKE. It was an integral and most important feature of the original plan that, after the factors which operate in the “ natural control” of the insect should have been identified and studied, an experiment should be conducted on a “ practical” scale to demonstrate at the same time the accuracy and the economic applicability of the knowledge acquired. By the spring of 1915 sufficiently definite conclusions. had been reached on the particular points at issue to justify this experiment. Until then the investigations had been conducted on the islands. The experiment could not be made on the islands, but must be made on the mainland. Consultation. with the Principal Medical Officer as to the best site for it led to the selection of the Buddu district, which borders the lake for some ninety miles just north of the old German frontier. Therefore a tour was made with the object of surveying this. reach of shore carefully, to ascertain the degree of infestation by the tsetse; the probable cost of exterminating it; the precise location of old village sites, canoe landings, fishing grounds, etc., with respect to fly; and such other points as required consideration preliminary to any experiment of the character proposed. The results of this survey were unexpected, and led to a complete revolution of ideas previously entertained. The reach of shore is naturally divided into a considerable number of semi-isolated districts, having from two and a half to. seventeen miles of frontage on the lake. Some of these districts were found to. be infested by tsetse to a degree of density never before encountered. Others were: very lightly infested. In one district a few days labour with a good gang of men would have sufficed to exterminate all the fly on a reach of some five miles. In this case the natives would very willingly have given the labour in return for the lands and the fishing grounds off shore, for the lands were good and extensive and the fishing grounds productive. In several other districts it was thought probable: that the value of the land and water rights was sufficient to induce the natives. to undertake all the labour requisite to exterminate fly without other expense to. the Government than that of supervision and imspection. Any of these would. have been an excellent site for the proposed experiment. There was one district, however, Bukakata, which had a frontage of some six and a half miles on the lake and included the steamer landing for Masaka station,. that presented a strange problem. The lands adjoining the lake in the depopulated: or forbidden zone were valueless for agriculture. The fishing rights were valuable,. and would have been a considerable inducement to the natives to clear the shore of tsetse, except for the fact that the natives were already occupying them, openly, for the entire reach of six miles. Infestation by tsetse was much heavier than in certain other districts, but still moderate, exceeding the average for the lake shore and islands as a whole at only a few points, and averaging for the district about half the average for the region generally. Careful inquiry failed to elicit a particle of evidence that any of the native fishermen had suffered in the slightest degree. from long-continued exposure to tsetse under these conditions. At certain other points along the shore surveyed poaching was free and flagrant ; at one point in particular, where density of fly was more than seven times greater’ than the average for the region generally, and where, in addition to this, food. (of fly) was very scarce and the flies literally ravenous, twenty-three fish traps, INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 349 some new and some old and discarded, were found along a two-mile reach of shore, indicating regular and long-continued contact between trespassers and fly at its very worst. Less aggravated conditions were frequently encountered along many reaches of mainland shore. On returning to Entebbe a special survey was made for fly within the inhabited precincts and for trespassing by natives beyond them. It was found that a considerable portion of the township was very lightly, but constantly or regularly, infested by fly (to an easily measurable degree), and that at certain points in the environs natives went freely into contact with fly at a moderate—but nowhere excessive— degree of density. Similar conditions prevailed at Jinja and at Kampala, the two other lake ports. With these conditions in mind, inquiry was made concerning the number of cases of sleeping sickness recorded in the official death returns or coming to the attention of Medical Officers; for it was thought certain that some must occur. To my very great surprise I learned that, so far as known, not a single case had been contracted within the Province of Buganda—for which the returns are most reliable—since 1912, and that with the exception of two cases (one of them not surely trypanosomiasis, and the other possibly contracted in Busoga) among the men who accompanied Dr. G. D. H. Carpenter on his tour to the islands in 1911-12, no cases were known or suspected to have been contracted since the islands were depopulated. in 1909. A curious situation was thus created, which led to the abandonment of the proposed experiment and to complete readjustment of ideas and preconceptions. In theory—the theory upon which the suppressive and preventive measures had. been based—complete severance of contact between fly and population was necessary in order to control sleeping sickness, but in actual practice complete severance. of contact was found to be unnecessary. In theory it was necessary either to exter- minate fly from populated districts or to make removal of inhabitants ‘ from the vicinity of tsetse complete and without exception” ; but in practice it was proved. sufficient to reduce the density of fly to within moderate limits in populated districts: or to reduce density of population to within moderate limits in fly-infested territory. If this is really sufficient, knowledge concerning factors which control range of the insect is more or less superfluous, whereas knowledge of factors which operate to control breadth of contact between fly and population—equivalent to frequency of contact between hungry flies and men—is specifically required. Two very different lines of study are thus outlined. Which of them ought to be followed ? As a matter of fact, both were followed: the first up to 1915, and the second during the last months of field work. One of them was impractical, and the specific results of it are practically valueless, except as disjointed fragments of information gained can be salvaged for use in other connexions. Which of these two lines of investigation ought to be reported upon ? In the accompanying manuscript neither has been reported upon, because this. | question has uot been answered, nor any specific object for the preparation of any (659) 300 W. F. FISKE. report defined. It contains a mere mass of unapplied, and for the present inapplicable, information of no practical value until a decision has been reached on the following points :— (1) Shall the original plan for these investigations be followed, and an experiment be planned and conducted to test the accuracy and applicability of the conclusions reached ? (2) If so, shall the object be— (a) to exterminate fly and prevent all contact between fly and population, in accordance with the theories held in 1906-07 ; or (b) to reduce excessive density of fly and to prevent excessive breadth of contact between fly and population, in accordance with present practice —as in the Bukakata District in 1915-16 ? Unfortunately the decision cannot rest on the results of these studies in tsetse bionomics, which have only the most indirect bearing upon the point at issue. Every bit of information which seemed, even remotely, to bear upon them, and which could be gleaned from any available source, was included in a report presented in 1916, upon which no action has been taken at this date of writing. The one phase of tsetse bionomics having any bearing on these points is the fly’s choice of hosts and host preferences. This is enlarged upon in the following pages. The methods used in conducting these studies consisted in the main of a “ fly survey ” of the lake shore and islands. This survey was designed to measure as accurately as possible all peculiarities and variations in the range of the insect, and all variations in its density as they occurred from time to time in the same localities, or from one locality to another at the same time. Coincidently observations were made and notes kept concerning every factor known or suspected to operate in the “natural control” of range and density, with the object of identifying those of major importance, 7.e., such as are responsible for easily measurable peculiarities in range or variations in density. No attempt was made to study specific factors—such as natural enemies—unless they were indicated to be of sufficient importance to account for measurable variations in density. By this procedure much useless work was avoided. These methods were especially applicable to studies having as their ultimate or practical objective the extermination of the insect. But after the readjustment of old ideas and preconceptions of the economic problem, made necessary by the discovery that it is practically unnecessary, and even, for economic reasons, unde- sirable to undertake extermination of fly, entirely different methods of study were demanded. ‘The questions to be answered involved less the range and density of tsetse and the factors controlling them than the injuriousness of the tsetse and the factors in its control. The injuriousness of the insect, or its injurious status in relation to a population occupying the same or adjoining territory, is, im part, but only in relatively small part, determined by density of infestation. Factors of equal or even greater importance are those which control the frequency of contact between flies and persons, and these include the relative abundance of host animals . (such as crocodiles), the principal occupations of the population, the precise location of points of occupation or concourse of the population with respect to colony centres INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS, 351 of fly, and various others to which the original fly surveys had accorded only in- cidental attention or had left out of consideration entirely. Reduction in density of fly is one object ; reduction in the injuriousness or injurious status of fly is, in actuality, a totally different object, and methods of study must vary accordingly. Finally, it was disclosed that the question of paramount importance was one that required yet different methods of study. It is to define the extent to which frequency of contact between flies and persons inhabiting a given region must be reduced in order to bring human trypanosomiasis under effective and satisfactory control. It is a question which can only be answered by measuring the frequency of contact and the coincidence of trypanosome infection in the population. It is, | regret to say, impossible for me to carry on investigations along this most necessary line unaided. It is for the entomologist to measure density of fly and frequency of contact between flies and men, and to identify and study the factors in control, but unless he is specifically trained to diagnose and detect trypanosome infection, his studies are incomplete and of relatively little value. — Every effort was made in 1915 to secure the co-operation of a Medical Officer for a tour along the coast and islands of Nyanza Province (British East Africa), where the riparian populations had not been removed; but, on account of the War, nothing could be done. Some little information on this pomt was gleaned from old (manuscript) reports of Medical Officers employed on “ Sleeping Sickness Extended Investigations ” m Uganda, which, as far as it went, strongly confirmed the presumption that the disease is incapable of spreading unless there is excessive breadth of contact between fly and population—in excess, for example, of that encountered in the survey of the Buddu shore in the Bukakata District—and a few other scraps of information have been found in literature—for example, in Todd and Wolbach’s survey for human trypanosomiasis in the Gambia, which contaims a few brief references to prevailing density of fly and frequency of contact—but it is a lme of investigation which has yet to be systematically followed. I. DispaRiry BETWEEN THE SEXES OF GLOSSINA PALPALIS. 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 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. Various hypothetical explanations for this phenomenon have been proposed by Medical Officers and Entomologists who have observed it, and prior to the inception of these investigations all these explanations presupposed that “ caught flies are hungry flies ”’ (attracted to their captor through desire to feed on his person), therefore *<< Thus of 1,400 flies bred from pupae obtained on Damba Island the proportions were 6:9:: 48:52. Whereas of 5,000 flies caughs during the period in which the pupae were co lected the proportions were ¢: 9:: 78°6: 21:4.” Carpenter: Repts. of the 8.8.Comm. of the Royal Soc., xii, p. 105. W. F. FISKE. 352 2 ‘SpuvIST Ossog oT} JO dnory equIvg-oMoy oy} jo dvur-qoyoyg “[ “O1T The enjwiny Sh AO = nuny ak ebuemg = Vv oO | ebues OF ebvema as oD ambuesemq ose =) (ah a nae.) Pa uss SP/el PROS WY S27/S dwey oesny 0 IMHO] : S v € @ | O NWoOlq that the ratio between the sexes in the catch would be an index to the ratio actually existing between all flies in the locality or district where the catch was made. deduction is logical from every point of view “SAIL gO 2]€9S INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 300 ' But what becomes of the surplus males in localities where females predominate, or of surplus females, when, as usually is the case, males predominate, were questions that had not been answered when these investigations were begun. Whatever the explanation might be, it was certain to be intimately involved with phases in the bionomics of tsetse the nature of which could only be guessed, but the importance of which could not be doubted. A minor problem was thus defined which served as a starting point for further investigations into the broad and complicated subject of the bionomics of Glossina palpalis, and the first field-work done was the series of experiments an account of which follows, I(a). Experiment to determine if Variations in Sex Ratio of Glossina palpalis are due to differences in the purely local, or in the climatic or seasonal Conditions of Life. The Islands of Bulago and Tavu (fig. 1) are separated by only a few thousand yards. Tavu is much the smaller and differs also in topography and vegetation. But the two islands are so near together that any difference in sex ratio, if constant and if occurring at the same seasons of the year, would necessarily be due to purely local conditions. Large collections of fly were made on the two islands as shown in Table I. There was considerable variation in sex ratio observed on Bulago on different days (when collections were made at different points on the island), but in the total catch of 4,405 flies there were proportionately twice as many females as in the total catch of 3,126 flies from Tavu. This difference could only be attributed to purely local differences in conditions of life between these two islands. TABLE J. Catch of Fly from Bulago and Tavu Islands, showing relatively constant difference in Sex Ratio. Daily Catch. Total Catch. 1913. Island. Total. Females. Total. Females. Ist Nov. ..| Bulago hss 170 20% 170 35 % Onde) >.". e e 18 V7 £9, 188 33% Sed is Le af te 60 24 % 248 32 % AU iige Oe o s BS oe 630 25: % 878 28. % Sth >, fe yi 564 27 OF 1442 Dy eR S15 as oe ae As LP 656 1S) 95 2098 25 G 10th?) ~,, --.. * A; 651 207 EG 2749 24 % MG 7 2. 35 Ay 1057 2h ie 3803 24°5 % on Ki 167 ey 3970 24 9 Pst . y 435 29 % 4405 25 % Bon (Wa ys] Taye 43 136 AS of 136 Airy bhi yy ge oe. ¥ Lf 1179 8-3 % 1315 7°83 %, TA LOORY Sa a aa e 945 | 13 2260 yA Bile iiieg bc sli 2 $6601), ty auld 3126 12-4 % 354 W. F. FISKE. Much wider variations in sex ratio between different islands were subsequently noted, and the extremes are given in Table IJ. Im all such cases—or at least in most of them—explanations for variations in sex ratio must be found in the purely local conditions of life, and not in climatic or seasonal differences. The absolute extremes encountered in the course of the investigations, together with the records for all colonies of fly or infected districts in which female percentage exceeded 45°0 are presented in Tables III and IV. TABLE II. Extreme Range nm Variation of Sex Ratio (observed) between different Islands an Victoria Nyanza. Island. Date. ..| January 1914 ..| July 1915 ..| February 1914 ..| March 1915 ..| January 1914 ..| August 1914 ..| Septemberl1914 .| February 1915 TABLE III. Catch of Fly. Total. Females. 7065 63°O) a 557 44°3 % 1413 39° 9% fle buat 39°0 % i yee 79% 704 fie as 73 4°1 % 101 3-0 % Absolute Extremes in Sex Ratio as encountered in course of these Investigations on Islands and Mainland. | Largest percentage of females in catch ‘6 | Largest percentage of females in catch Per Locality. Date. Catch.) cent. Remarks, Fem. | | Manene Island..| March 1915..| 46 0-6 | Largest catch with no females. Bugaba Beach ..| August 1914..| 106 1:9 |\ Lowest percentage of females in catch Manene Island. .| March 1915..| 107 1-9 of over 100 flies. Bukakata a OUly . 19D6..) 230) W900 of over 10 flies. Bale Beach ..| June 1915..| 241 | 84 of over 100 flies TABLE IV. Catch in those Districts on Mainland where Female Percentage averaged 45:0 or over for entire District. District. Date. Mujuzi Creek Colony, Buddu Bukakata South Colony, Buddu Kaziru District, Buddu bes Kitebo District Mawakotu .. Gwamba District, South Buddu ..| June - | July ..| duly ..| July MONS i 1915.. 1915.. ..| August 1915.. HONS Catch. Percentage of Females. 1611 67°6 298 51:0 991 50°6 1199 47°5 3422 45-0 INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 355. I(b). Relative Inactivity of Females of Glossina palpalis as the Explanation of Low Female Percentage in Catch. The figures of the catches of fly from Bulago and Tavu Islands presented in Table I show a discrepancy in that the percentage of females in the total catch from Bulago Island tended to fall as the total grew, whereas on Tavu it rose sharply. This might be accidental, but the circumstances under which the experiment was conducted indicated otherwise. Tavu Island is much smaller than Bulago, and the catching of some 1,300 flies on the 5th and 6th of November seemed to have reduced the local density of the species appreciably, as measured by the average number of flies which could be caught per boy per hour. A corps of 9 or 10 boys was employed. During these first days of the investigations the exact time spent in catching was not recorded. It was approximately known, however, and when the catch of males and females per boy per hour (approximately) is tabulated it stands as in Table V. TABLE V. Reduction in Density of Glossma palpalis on Tavu Island due to catching of Fly. Catch of Fly. «“ Boy Density of Fly Indicated. 1913. | ye) | » Males. Females. | Hours. | Males. | Females. | Total | 6th November .. 1081 98 36 30°0 OI 3207 7th a mn 734. 169 45 16°3 3°7 20°0 8th he a. 698 111 45 15°1 2°5 17°6 TABLE VI. Catch of Glossina palpalis on Lula Islet, showing the relative Inactivity of Females. ; Catch for Period. Catch to end of Period. Period (November 1913). Fe- Female Males. maloe) oa Total. | Females. 18th Afternoon .. a are ..| 280 18 60% 298 60% 19th Mor: ing .. me .-| 308 ee 19:0 % 678 1333 % 19th Afternoon .. Bb Bs yo ess 68 42:2 % 839 18°8 % 20th Morning .. oe ve see So. 4G 64°3 % 1068 28°5 % 20th Afternoon .. a we Clu 89 G2, oF 1193 3279 9 21st All Day... at ee aly oOo 224 10-2, oF 1512 40°38 % 22nd... sa on aig peilia aged 79 60°38 % 1642 42°4 % PATH... ss - Da 2O% ee Teme mgles ©,” | 74d 43°5 9%, 25th 55 ste aie au au Ore 93 TORE 1865 46°2 % 26th sk a8 ath aes eee 65 69°1% 1959 473 % 27th Morning \ BA » » >» camp - - vs eh es Oe Cem Or 12 Number of times bitten per hour spent on island .. af a 0-28 | 0-03 *The catches made during this trip indicate a somewhat heavier infestation than this, but it is known that they were too high; they were made at points where flies were concentrated, and allowance is made for lower average density. The really striking comparison, however, is between the figures as given for Damba, and the note made during field work on the beach at Bale (see p. 362) to the effect that “ nearly every fly tried to bite,” and that 15 actually bit in less than one hour. Density, as indicated by the male catch at Damba, is three or six times greater than at Bale, but at Damba 22 hours were spent in the bush without being bitten once, as compared with 15 bites in one hour at Bale, a proportion of less than 1 to more than 15 x 22 = 330. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 365 Making allowances for error at every point, 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. II (e). Female Percentage as an Index to the Chances favouring Transmission of Human Trypanosomiasis. 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 the same fly will attack man a second time aré stupendous in comparison. The female percentage may thus be a very valuable index to the chances favouring transmission of human disease. Ill. Tot Lonc-sHoRE MOVEMENTS OF GLOSSINA PALPALIS. The explanation of the variable ratio between the sexes of Glossina palpalis provided by the early experiments proved inapplicable in many cases, and numerous vagaries ‘in the sex ratio were observed which were for long inexplicable. Eventually investigations into the long-shore movements of flies and the routes and courses followed by them im their food-hunting and otherwise stimulated flights, led back into the old problem of sex disparity and provided logical explanation for many variations in it which had been observed but which could not be correlated with abundance or scarcity of food. The same study also required consideration of “shelter” (arborescent vegetation sought or required by the flies for their protection) and of the relative attractiveness of different types of it. These three, quite different topics—sex disparity, shelter, and movements of flies from place to place—had to be considered coincidently in the field and cannot be entirely separated in reporting upon field work. In the following sub-sections they are discussed as they were studied, inter- -dependently. IIT (a). Experiment to test the Movements of Flies along the Lake Shore. The larger catches of fly made on Bulago Island and recorded in Table I were ‘from near the extremity of a long, narrow spit of sandy land designated in the notes as “Crocodile Point” (see Fig. I). The area of this spit was considerably less than of the islet of Lula, from which the flies were caught as recorded in Tables VI and VII. But though these catching experiments on Lula quickly brought about reduction in density of active flies, 1t was observed that no such effect was produced on the density of the fly at Crocodile Pomt. On the contrary, although more than 2,000 flies were caught there during the period 10th to 13th November, neither density nor female percentage underwent notable change (see Table XII). (659) B2 366 W. F. FISKE, TABLE XII, Catch of Fly at Crocodile Point, Bulago Island, showing Existence of Rapid Movement along Shore. | Catch of Fly. Density of Active Fiy.* Date. re Total. | Females. Males. Females. Total. 5th November 1913.. 564 72 Tha 14°2 5°8 20°0 10th 4 patie: 630 20 2G 16-0 4-0 20-0 lith ” aca 944 28. 6 12°5 5°0 17°5 12th 53 cree 167 1S: oe 10°3 2°4 12°57 13th > pone 435 29° '% 14°7 6:0 20°7 Totals and averages Ist experiment) 2740 24°4 % 13°5 4-6 18°1 17th January 1914 912 40-3 % 24-4 17-1 42°5 20th 09 5. fee se 1102 48°2 % 27 26°4 53°9 21st ” anh pete a oe 303 44°2 % LB 9°0 20°3T 22nd’ ;, 55%, eee 2a: se 606 48:0 % 26°5 28°9 55°4 Totals and averages 2nd experiment 2923 45°2 % 22°7 20°3 42°9 * “ Density ” is the number of flies caught per fly-boy per hour. + These days were somewhat dull and overcast, rendering flies inactive. The fact that density of both sexes was perfectly maintained on the 13th November after four consecutive days of catching, when viewed in the light of the results of less extensive catchings on Tavu (‘Table V) and Lula (Tables VI and VII), can only be explained by the movements of fly along shore, and these movements must be quite free and rapid in order to account for it. Notwithstanding the semi-isolated position of Crocodile Point, 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 (Table XII) 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 subsequently ascertained, along the borders of woodland, game trails or human pathways, ete. ITI (6). Movement of the Sexes along the Lake Shore. Although it was impossible to reduce the local density of fly under such conditions. as exist at Crocodile Point if an interval of several hours is permitted to elapse between periods of catching, it was easily possible to reduce the density of male flies—not of active females—by even a single hour’s catching, provided no interval elapsed INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 367 before the experiment was repeated. In other words, if catching is continued over several consecutive hours, the density of male flies, but not of active females, will be reduced during the first hour, and will not rise again to normal until several hours have elapsed. TABLE XIII. Catches of Fly for consecutive hours at points on Lake Shore, showing Reduction im Density of Males but not of Active Females, and demonstrating more rapid movement of Females. Catch of Fly. Density of Active Fly. Locality. Date. Hour. Total. | Females. | Males. |Females.| Total. Crocodile Point.) (17th Jan. | Ist 343 30°9 do 35°0 14-1 49°1 HulapoIsland|| gia ..j| 22¢ | 319 | 414% | 267 | 185 | 45:2 : 3rd 250 5652 9 14°5 18°7 33°2 Ist 419 IO Oe 623 15 69'°8 | ond | 250 | 10:4% | 37-4 4:3 41-7 ed isle | Sra) isa) 213% | 24-0 6-5 | 305 | He athe 16s eh 25-294 o088 6:8 27-1 sth 167 26-3 % | 203 75 27°8 | : | ——— These experiments were many times repeated with results similar to those presented in Table XIII. It was found that the local density of male flies could be very quickly and easily reduced, whereas that of active females was not affected in this manner. It was conclusively proved by experiments cited on p. 365, and illustrated by Table XII, that reduction in density of males would be made good by incoming flies if some hours were allowed to elapse before a second catch was made, but that no period of waiting was required to make good any reduction in density of active females caused by catching. 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 cited in Sect. I, 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 (see Sect. IV) 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. This conclusion found ample confirmation as the investigations progressed ; as, for example, by the fact that when food is so scanty on an island (as on Kimmi Island in January 1914) that the female percentage is very high, it is impossible to reduce the local density of either males or females by catching experiments at 368 W. F. FISKE. points where they are passing. Under such conditions nearly all flies are seeking food, and all are equally active, as clearly shown by a comparison between the figures presented in Table XIV and those presented in Table XIII. TABLE XIV. Catch of Fly at a point on Shore of an Island where Food was deficient, and where nearly all Flies were seeking it, showing equal degree of Actwity on the part of both Sexes. a2 Catch of Fly. | Density of Fly. Locality. Date. Hour. | Sg Total. | Females. | Males. | Females.| Total. ; Ist | 205 | 66:3°%!| 93-0. | 45:2) || (ieee ond| 249 | 66:3 ° 28 0 55:0 83-0 Fee aces al pond) 4ae). ost, 1) oi) | AO eay mae L “Hath! A684) Goa, | | S7e0 50:0 77°0 \| sth | 289 | 640% | 17:3 30°8 48-1* * Falling off in density during the 5th hour was due to approach of evening. The sexes. remained proportionately as active as before, as shown by the female percentage for this. period. lil (c). Effect of Shelter, or of Type of Vegetation, upon Long-shore Move- ments of Fly and upon Percentage of Females in the Catch. On 4th September 1914, a catch of fly was made under unusual circumstances on the island of Bukassa. The shore at this point was lined with a very dense fringe of reeds only two or three yards in width. Inside the fringe was open grass land, cropped very short by hippo, with scattered clumps of thick bush and trees o° a sort affording attractive shelter to fly. At the observation point itself was a landing place of hippo with a trail forming a tunnel through the reeds. The prow of the canoe was thrust into this tunnel, with the stern projecting beyond the reeds into the open lake, and collections were made simultaneously by one boy stationed in the stern of the canoe, outside the thick fringe of reeds, and by two boys at and near the point where the hippo trail entered the opening, imside the fringe of reeds, These reeds, it should be noted, were at least 10 feet in height,. or higher than tsetse is at all likely to rise from the ground. The three boys worked for two hours with the following results: Inside Reeds. Outside Reeds. Catch per boy hour— Males a) ie Be fig 510 3D Females .. ans as es 45) ES Female Percentage .. M a 143% 68:29, The extraordinary feature of this catch was the low percentage of females (14°3): wmside and the high percentage (68°2) outside the barrier formed by the reeds; the two points being separated by not more than 10 yards. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 369 This same phenomenon was observed a second time under different conditions in a catch made on the 23rd September 1914 at a point on the island of Bussi, where two boys worked ankle-deep in water outside a thick mass of reedy vegetation that prohibited landing, and two other boys on a point of shelving rock, backed by an open space and bushy forest of a type much favoured by fly as shelter. The poorly sheltered, reed-fringed point on the shore was only about 100 yards distant from the attractively sheltered point. The catch was as follows, for the four boys for half an hour: Well sheltered Poorly sheltered Point. Point. Catch per boy hour— Males sf ot AG a 49-0 8°0 Females .. ie * oe 3°0 13°0 Female percentage .. i we 58 60°8 As before, there is an extraordinary discrepancy in sex ratio between points separated, in this case, by only about 100 yards. And 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. A hypothetical explanation for this phenomenon, which has withstood all tests applied to it, is as follows :— (a) The body or mass of active flies is contmually in movement, and streams of flies are continually passing points along shore (see Sect. III (a) ). (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 (see Sect. III (0)). (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. Itfollows 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,* consist principally of females), When confirmation of this hypothesis was sought it was found on every hand, not only in new catching experiments and fly survey work, but in old records, made *'These reasons include the following :— Ist. Becaus> the females in any district or region appear to outnumber the males (see Sect. I (d) ). 2nd. Becaus, 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. 370 W. F. FISKE. ong before. The records made on the island of Kimmi in January 1914 are an example, As presented in tabular form (Table XV) they are not particularly TABLE XV. Catch of Fly on Island of Kimmi showing effect of variable character of Shelter on Density, Sex Ratio and Movements of Glossina palpalis. Catch of Fly. Date. Locality on Island. (January 1914.) (Observation Point.) Male | Female | Total in Density.| Ratio. | Catch. 27th ..|South-easternmost point. ee a Bee ee 3 J) 53°2 506 29th ..|North-eastern point ne sah “if “fell cael coal 66:9 14z 29th ..|North-western point .. ay ne an) whee 522 272 24th ..|South-western point .. a Ka Aa -Baeow 629 384 27th ../Southern shore .. We nm fs soe luted 53°0 276 * Shelier was appreciably less attractive at these two points. striking, but the correlation between shelter, local density of males, and female percentage is quite strikingly shown in figure 2. In this case the “ shore line ” in the graph represents the entire circumference of the island, a distance of between 2 and 3 miles, with five observation points located as stated. The difference in character of shelter was not particularly notable, but it was enough to produce the effect shown. Observation | South- —— —-North-.—North- South- : | Point eastern eastern ——_—s western western Southern | Percentage lot Females Density of Malés Shore Line | Shelter Superior Inferior Superior Inferior Superior —— ~~ Fig, ae Graphic representation of dala given in Table XV. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. Bel Records made in the course of the fly survey of Bugaba Island as presented in ‘fig. 3 afford a second excellent example of the correlation between character of sheltering vegetation, density of male flies and percentage of females in the catch. In this particular case there was a short reach of sandy shore at observation point A3 which served as a breeding ground for the flies, but which was poorly (too slightly) sheltered. At either end of this bit of sandy shore was forest, with marshy, reed-srown fore-shore of a distinctly unattractive type. Only where the forest ‘growth came down to the more open sandy shore at points 42 and 44 was shelter really attractive [Observation Points a4 Pees [pas ub — as Female Percentage 3050 0 80 a O 2530 Female | | er a Percentage = ; \ \ \ | \ \ tf | \ vals \ “a \ | \ 7 \ { \ oe \ | ee | yy, NUE \ ' vg \ ee asi oe \ Density | so vace-sseerbee = Dom LUE 2. Al out 3000 yards - - Agi ae hea Poor Poor Good Poor Good Poor Poor | Shelter ; Breeding _Ground None None None Poor None None NOs Vig. 3. Correlation between character of shelter, density of male flies ana percentage of females. Figure 3 shows an excerpt from the data secured during the “ fly survey ” of the island of Bugaba (vide Sect. VI. (g) ) and illustrates conditions of infestation by fly along a reach of shore about 12 milesin length. In all, more than 150 miles of mainland and island shore were systematically surveyed in this manner, and the data secured have served in scores of cases to confirm the conclusions herein stated concerning the correlation between character of shelter, density of males and female percentage. A single other example may be given in Sect. III (d) which follows. III (d). Notes on Survey of Lutoboka Bay, Bugalla (Sesse) Island. Figure 4 presents the results of a survey of some 5 miles of shore, principally included in the large circular bay that lies within the peninsula of Lutoboka on Bugalla (Sesse) Island. This survey was undertaken with the object of collecting -data relative to the effect of environment upon range and density of Glossina palpalis. SnD W. F. FISKE. It is impracticable to include in the graph data concerning environmental features. responsible for the striking variations in male density and sex ratio shown by the curves. In general they are as follows : Points 1 to 4 inclusive are typical of shore infested to slightly above the average extent for this general region. This reach of 14 miles lies on the western shore of the: bay, and is densely wooded, with bits of open shore alternating with reed-grown and jungly reaches. There were some sand and gravel deposits affording good breeding grounds. Points 5 to 8 inclusive :—The forest continues, but there is no more open shore, nor any breeding places. Instead there is a fringe of floating sudd (papyrus and saw-grass) lying off-shore and no open space between it and the massive shelter of the forest behind. Catches were made from a canoe outside the sudd, which, with the exception of crass, 1s the most repulsive type of vegetation. Only the hungry food-hunting flies (principally females) pass beyond the limits of the open shore at point 4. The relatively idle males turn back. The minimum density of males and maximum percentage of females is recorded from point 6. At pomt 7 males are more numerous and female percentage has fallen, indicating that a new fly colony is being approached. The catch at point 8 confirms this oa makes it certain that males are coming into the repulsively sheltered reach from the other direction. At point 9 there is a break in the sudd, and an open grassy bit of fore-shore, scattered with bushes and backed by massive forest. Shelter is attractive, but as yet no breeding grounds occur. The excellent shelter makes for a sharp increase. in male density and a corresponding decrease in female percentage. At point 10, the forest, which has continued unbroken until now, ceases and open grassland comes down to near the water’s edge. The foreshore is open, with scattered bushes, and deposits of beach sand afford excellent breeding grounds. For a short space between points 9 and 10 this series of open sandy belts is backed by the massive shelter of the forest. This combination affords complete protection to the flies, both as pupae and adults, and forms the centre of the colony, or centre of infestation for the shore on either side. Beyond point 10 the open fore-shore, with sand deposits and the slight shelter of scattered bush, continues (as at point 10), but is no longer backed by massive shelter. The effect on male density and female percentage is precisely as caused by the fringe of sudd. The idle males turn back when massive shelter ends, and only the food-hunting flies, in part, continue. 7 At point 14 is a relatively thick mass of bush, which tempts a few males to linger, but not for long, and density does not rise again until, some distance further on, forest shelter 1s again encountered. There is not the slightest doubt that in these observation points high percentage of females coupled with low density of males is due to insufficient or unattractive sheltering vegetation, leading to a partial segregation of the food-hunting flies of both sexes from the relatively idle and satiated males. Fema Percen? 0 1g effect of environment Bale & Danielsson, LtY Observation Point Density Female Percentage Density Female Percentage Shore Line Scale of Miles Fig.4. Fly survey of a reach of shore in the district of Lutoboka, Island of Sesse (Bugalla), illustrating effect of environment on range, density and movements of Glossina palpalis. Bale & Danielsson, 67 < ‘ - a : ssanealad Tita iebeeenesl hh mrecdemeettied ieee Ae eee : 7 , . ‘ aye 8am tu, Pers ; ._- ae ¢ ' 4%, ale NOTEVISEOU F . el ( : $ Mio. mee © 1 ht eed nie nae mse rye pip nwa en ne ts tail 5 ¢ " . he ‘, } * i? Ai. iPS a ; 1 , “iene fe Ma oT ry = F + fm. ss hy ' 7" SE —— 8 gy enn a ' a ” o ™ 7% “- o ~ ae - ¥, r \ - | Mb we _ — o") :. ry ; censunhtiiemidiamecddiecca te eect ea ae es pa ieee : at Sry. Ma eT te, aay: 1 SH A SUAS. mG INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. o1e. III (e). Maximum Percentage of Females brought about by Segregation through Movements. The maximum percentage of females amongst active flies encountered at any point on island or mainland was in the district of Buddu between the old landing places known as Kalkosa (or Sekwe) and Bale. The figures of the catches made during a fly survey of this coast are presented in Table XVI. The distance from TABLE XVI. Catch of Fly wn the Mujuzi Colony. Catch of Fly. Date. Observa- | Density | Percentage (June 1915.) tion Point. of Males jof Females. Males. | Females.) Total. 24th by: me aM 40 26 39 65 | 26:0 60°0 % 24th “oH 4s os 4] 12 90 102) 40 88°3 % 22nd a a We 42 33 65 Gar 165 66°0 % 22nd aye die Be 43 23 68 91 115 OEE Ss 22nd i a a 44, 58 Wipe ils 19°3 Tar 0 Of 24th e ne és 45 196 162 358 98°0 AD 5 22nd is “ a 46 106 103 209) 45320 49°2 % 26th 2 t oe 47 4] 73 114 |. 205 64:0 % 26th SS, ie ores 48 45 120 165 19°3 TT 26th fi nae ae 49 20 84 104 10°0 80:8 % 26th Bec ahs oe! 50 10 80 90 5:0 88°8 % Totals and averages A. 570 104] 1611 26°0 67°6°%% Average by points se a oe 26°6 69°5 % l point 40 to point 50 is about 34 miles. Point 45 is at the mouth of Mujuzi Creek, which affords excellent harbour for crocodiles, as well as breeding places for the fly on its banks. It is the colony centre, and centre of infestation for a considerable reach of shore. Points 47 to 50 are-in the Bale fly beach mentioned in Sect. II (a), Table VIII. The average female percentage for this district is either 64°6 or 69°5 according to the method of computation, and compared with the maximum encountered on the islands (57°5 or 64°6 for the colony on Kimmi Island; vide pp. 363 and 370) is about 6°0 points higher. This is undoubtedly due to scarcity of food, which only occurred sufficiently or at all at the very mouth of Mujuzi Creek (point 45). The behaviour of flies bore this out completely. Nowhere else on island or mainland were they so troublesome; it was impossible to avoid being bitten, even with both hands employed wielding fly switches. The extraordinarily high proportion of females at points 41, 49 and 50 break all records made elsewhere (for equally large catches), and are certainly explained by 374 W. F. FISKE. the unattractive character of the environment* at these points retarding the movement of the relatively idle and inactive males along shore. These males were concentrated along the short reach of shore including points 45 and 46 ;_ and elsewhere it was almost entirely the actively moving, food-hunting flies that were caught. This particular colony of Mujuzi is in several other respects one of the most interest- ing and deserving of study of any encountered on either island or mainland. It is mentioned elsewhere, and a graph of it is presented facing p. 388. I: was non- existent in 1906 but came into being subsequently as a result of environmental changes described on p. 458. III (f). The Full Significance of the Phenomenon of Sex Disparity in Glossina palpalis with Relation to Food Supply, Shelter and Movements of Fly. The phenomenon of a variable ratio between the sexes of active flies of Glossina palpalis proved on investigation, as I have attempted to make clear, to be replete with interest and significance, Taken over any considerable district or region it is a fair index to the abundance of preferred hosts, and thereby of the relative immunity of man to attack (see Sects. I and II). Taken locally, pomt by pomt along a reach of shore, and coupled with the local density of active males at these same points, it becomes a subtle index to the attractiveness or repulsiveness of the local environment to fly. This fact having been conclusively proved, thereafter data in such local variations become of really great value in reaching conclusions concerning the relative attraction of various types of vegetation, etc., to the flies. Finally, through study of this sex disparity and its causes, a very clear conception is obtained of the ordinary movements of the flies along the shores of the lake, banks of stream, borders of woodland, game trails, foot-paths, or other favourite courses. A great many little experiments and informal observations were made to ascertain the routes most freely followed by the active food-hunting flies—too many to mention in detail. The lake shore is the most favoured of them all, but, in general, they follow quite closely the line separating sunlight from shadow. The flies are averse to penetrating shadow, unless sunlight is perceptible beyond, and even more averse to crossing sunlit spaces unless shadow or massive shelter is perceptible beyond. *It was not necessarily unattractive shelter that prevented the males from congregating at these particular points. They were distant from all sources of food, and however attractive shelter may have been (it was in fact very attractive at 49 and 50, but not at 41) no permanent concentration of males could occur. The males tended to congregate at point 46, which was well sheltered; but not at point 44, just the other side of the creek, where shelter was not attractive. But by the time the relatively idle males had dispersed much beyond point 46, towards points 49 and 50, they began to require food, and there was no food unless they passed back to point 45, or passed far beyond point 50. In other words, a reach of foodless shore, although supplied with both shelter and breeding ground, produces the same falling off in male density and elevation of female ‘percentage as the shelterless shore described in the pages preceding this. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. od It was found by experiment, details of which need not be given, that flies will cross open channels up to at least 300 yards in width, but that they do not cross them very freely. Itis certain that the width of the channel likely to be crossed is dependent on the character of the farther shore. Certainly they do not cross much wider channels than 300 to 500 yards at all freely, for from the beginning evidence of their doing so was carefully sought, and no shred of it secured. It is the movements of fly along shore, or along the border line between sunlight and shadow, with which we are principally concerned, and these studies have shown how the flies tend to congregate in attractively sheltered localities, where, in conse- quence, density of males is high and percentage of females low, and how, pressed by hunge~, both sexes, but more especially the females, tend to disperse from these centres of greater density and range along the lake shore or other favoured route in search of the sluggish amphibian animals on which they principally feed, thus accounting for the extraordinary preponderance of females amongst caught flies in repulsively sheltered localities. IV. Tue Assempiine Hapits oF GLOSSINA PALPALIS. The stimuli which keep in motion the streams of tsetse described in the preceding section are known to include, and believed in all ordinary circumstances to be limited, to the following :— (A) Hunger. Hvery second or third day the flies of both sexes undertake food-hunting flights. These are relatively rapid and easy to follow. (B) Self-protection. At the close of each period of activity the flies of both sexes are moved to seek massive shelter, where they remain in seclusion during periods of inactivity. They appear very unwilling to move far away from such shelter; hence flights in search of it are usually of short duration. (C) Maternal Instanct. At intervals of eight days, or longer, the females are stimulated to seek breeding grounds wherein to deposit their young. The great concentration of puparia in very attractive breeding places is proof that flights in search of them were of considerable duration. The location. of deposits of puparia near the well-marked courses followed by food- hunting flies is proof that these are followed. But the flights themselves. are very hard to follow. (D) Masculine Instinct. Hvery day, if the weather is good, the males appear all to leave shelter to undertake assembling flights. These are readily followed, and as described in the preceding section, are relatively slow. The insects tend to congregate or to loiter at points or along courses where the females are most likely to congregate or to pass. It is these special activities which are now to be considered. Several points in connection with the assembling habits of Glossina palpalis in addition to those touched upon in the preceding sections are discussed in the sub- sections following. The topic is inter-dependent with several others, including, especially, host preferences of the flies, LO We P. PISKE: IV (a). The “Following Swarm’’ of Glossina palpalis. Many observations, but few detailed experiments were made upon the curious habit of Glossina palpahs of approaching and scrutinising strange objects. It was noted that there is ordinarily little discernible difference in this behaviour toward inanimate and animate objects other than their favoured hosts. When the percentage of females amongst active flies is low, man is almost immune to attack, but his person is none the less, and equally with any other similarly conspicuous object, strangely attractive both to the food-hunting flies of both sexes, and to the relatively inactive males. It is on this account that the males are so easily caught, whether or not they are seeking food. Of all objects, apart from their favoured hosts, the most curiously attractive to flies is, perhaps, an open dark-coloured umbrella, and in the catching experiments and fly survey work the fly boys were each provided with one. When displayed at any point in or near the course followed by food-hunting flies, or the relatively inactive males, it becomes so conspicuous an object as to attract to it virtually all flies which perceive it. They crawl over it, especially its underside, inspecting it curiously until curiosity is satiated—or for, perhaps, one to five minutes—when they pass on. And this is equally true of both idle males and food-hunting individuals of both sexes. Despite the extraordinary attraction of the object there is no tendency on the part of any flies to linger more than the few minutes required for its inspection, so that a swarm will never collect. If the density of the stream of passing flies is 20 or 30 per hour, there will be from one to four or five flies resting on the attractive object at all times, but the same flies do not remain for long. As above stated, the behaviour of the flies toward man 1s, ordinarily—z.e., when the man is alert, food is plentiful and the female percentage low—aindistinguishable from their behaviour toward his coat thrown over a bush or his umbrella openly displayed. It is not necessary to think about the flies or to ward them off; occasionally one will bite, but the number doing so is so small that it is not worth thinking about. Man is attractive to flies, but does not hold them after their curiosity is satisfied. The same is true of goats, sheep, and many other animals. But with certain animals under certain conditions the behaviour of the flies is conspicuously different. When a crocodile or Varanus is basking in half open sunlight, the flies attracted to them are apt not to pass on, after cursory examination, but to linger indefinitely. The result is the gradual accumulation of an accompanying or “ following ” swarm, which may grow to number several times more flies than a boy stationed at the same spot would be able to catch in an hour’s time, and which may certainly represent (sometimes) the accumulation of the majority of flies passing the spot during several hours’ time. Similar swarms may collect upon or near a foraging Varanus, as it wanders slowly through the woodland, or a grazing situtunga, and may then follow the animal out of woodland into open grassy spaces, where the flies would not go of themselves, Mention of these conspicuous swarms of flies in connection with Varanus, crocodile, tortoise, hippopotamus and situtunga is made in Sect. VI (e). A few additional notes excerpted from my field books follow. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPATIS. Old On Varanus. Kitobo Island, 3rd December 1913. Varanus observed excavating burrow in sandy soil some little distance from shelter. On approaching it made off rapidly, and on reaching the spot I was assailed by a great number of tsetse, which swept back and forth and around me like angry bees, “‘ buzzing ” in their flight in a manner never before noted. After a few minutes they all dispersed, without any of them alighting upon me. ‘On Varanus. Nsadzi Island, 3rd February 1914. A Varanus was suddenly flushed in grass land (grass about knee-high) not much less than a hundred yards from any shelter. It made off rapidly and a swarm of flies, numbering several dozen at least, rose and filled the air with their angry, bee-like “buzzing.” As on Kitobo they swept back and forth and around about me, but none alighted or offered to attack, and in a minute or two all had dispersed. On Situtunga. Bugalla Island, 23rd November 1914. A large male situtunga was approached as it was feeding with its head concealed in a dense thicket. With glasses (Zeiss prismatic x12) it was possible to make out that a peculiar dark colour of fore leg, lower shoulder and thorax, which were plainly seen through an opening in the bushes, was due to an unprecedented number of tsetse, which literally blackened its coat. It seemed entirely unmoved and phlegmatic under attack. On being shot, the animal plunged directly through the thicket ; ran a few yards at great speed and fell. On proceeding to the spot where it was feeding, I found a “following swarm ”’ of fly of unprecedented size (probably not less than 200 flies) buzzing like a great swarm of angry bees. They surrounded me, but hardly any alighted on me or followed me to where the antelope lay. On Situtunga. Damba Island, 13th September 1915. A large male situtunga was shot in an opening in the forest in the dusk of evening. It ran into the thickly shaded forest and fell. On reaching it I was amazed to find a considerable swarm of flies, partly outside, but judging by the noise they made, more inside than outside the forest (it was so dark inside they could not be seen). Is it possible that a swarm will follow an animal into the night, and perhaps remain on its body all night ? On Varanus. Manene Island, 16th March 1915. While passing through forest a movement in the vegetation (the ground was completely covered with broad-leaved herbage rather less than knee-high) indicated the approach of some small animal. It proved to be a Varanus, which came very slowly, evidently hunting insects and molluscs, to very near me. At no time could I see its body, but keeping pace with it and evidently following the movements of the herbage was a swarm of perhaps two dozen tsetse. The flies were not anxious to feed or to reach the body of the animal, but merely followed its movements, alighting on the herbage or hovering about. A movement on my part sent the animal scuttling rapidly away, whereupon the disappointed flies rose and swept back and forth, buzzing angrily as usual in such cases, 378 W. F. FISKE. On Situtunga. Bugalla Island, December 1914. (Original note lost.) On entering an open space in the jungle where formerly were plantations, a small herd of two female and two half-grown male situtunga was. seen, with other animals feeding in the edge of the jungle out of sight. Those in sight did not immediately see me, who stood motionless watching them, nor upon seeing me did they betray alarm or more than mild curiosity. The whole herd moved in my direction and one female approached within three yards. Hach animal was followed by a small swarm of tsetse—perhaps 15 or 20 flies—few of them on the animal itself, but principally on the vegetation close at hand, or hovering about. Not one of the flies was seen to feed, nor did the animals show annoyance at their presence. On becoming alarmed the antelopes made off without undue haste, the flies following. The above may serve in some manner to make clear the difference between the behaviour of fly toward a favoured and complacent host, and such another as man, who is neither favoured nor complacent. Anyone passing through infested territory and aware of two or three, or perhaps half a dozen, tsetse constantly hovering about him is certain to receive the impression that this number of flies is persistently following him. In the case of Glossina morsitans the impression would, probably,. be correct, for man will collect a following swarm of this species; but with Glossina palpalis 1t is incorrect. The same individual flies will not follow a man for more: than a few minutes unless they are very hungry, but will quickly drop behind and be replaced, perhaps by others. This is easily proved by catching them just as they appear; one will quickly catch many times more than appear to be following at any one time. The explanation of the “following swarm” with Glossina palpalis is, without doubt, identical with that put forward by Lloyd for the “following swarm” of Glossina morsitans. It is the assembling habit of the males; first, to linger along the routes followed by the food-hunting females; secondly, to examine any strange object coming in range of vision; and finally, on encountering a complacent host of a favoured species, to remain by it ; for in these ways they are most likely to come into contact with the seclusive females. Nothing else in the habits of the fly affords more convincing evidence of their unwillin: ness to feed upon man, unless they are forcedby hunger. Far from attracting a following swarm, man cannot even hold one that has already been attracted to some other host ; if by his approach he startles the favoured host into flight, there is not. the slightest tendency on the part of the following flies to turn their attention to him, but invariably, as in the cases noted above, the swarm has dispersed without the flies paying as much attention to the intruder as is usual when other hosts are about. IV (6). Conditions under which a Following Swarm will Collect. More than 25 separate experiments were made with various wild and dumestic animals tethered at poimts where they were exposed to columns of fvoa-nunting flies, mainly to ascertain the host preferences and feeding habits of the insect, and finally in a specific effort to induce a “‘ following swarm ”’ to collect under observation. In the very first experiment in the series such a swarm did collect upon a tethered Varanus, but all other experiments failed in this respect. ? INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 379 Under entirely natural conditions swarms had been seen on Varanus, situtunga, crocodile, hippopotamus, pig (of domestic stock, but run wild) and tortoise, and most freely of all on Varanus. The fact that repeated attempts to induce one to form on Varanus which were held in constraint all failed, excepting only the first (see Sect. VI (c), experiment 4), convinced one that something more than the mere presence of a host animal of a favoured species is necessary, and failure is believed to be due to the impatience of these animals under restraint. In other words, a host must be complacent under attack as well as of a favoured species or it will not be favoured by tsetse to the extent of attracting a “ following swarm.” The final experiments with Varanus, sheep and oxen are described in Sect. VI (c), experiment 3. They led to the conclusion that a host to be “ favoured,” must always be complacent, and suggested that perhaps almost any of the larger animals, including man, might be favoured, even to the extent of collecting a “ following swarm,” if it were in them to suffer the attacks of Glossina without sign of protest. ‘\V. Tae Errect of OPpEN WATER AND Humip CoNDITIONS UPON THE RANGE OF GLOSSINA PALPALIS. No factor in the bionomics of Glossina palpalis is more obviously correlated with the range or distribution of this species than open water. It is a riparian species, and its occurrence far from shores of lakes or banks of streams is a subject for remark whenever observed. Exceptional dispersion inland, however, is not at all rare, and many cases of it were encountered and studied. In all cases the explanation was the same, and involved, as virtually every topic in Glossina bionomics seems bound to involve, the subject of hosts and host preferences. The two topics are inseparable in field work and cannot well be separated in discussing and recounting field work. These studies were more convincing to the observer than it is feared they may prove to the reader, that the correlation in question is purely coincidental; that water or humid conditions are not of direct benefit to the insect, nor required by it; but that a proper combination of food, shelter and breeding places which is requisite to its existence occurs so infrequently away from the shores of lakes or banks of larger streams that it is perforce riparian in habit. V (a). Range of Fly inland from Lake Shore. Ordinarily the range of fly inland from the lake shore is extremely limited; at 50 yards there will usually be a sharp reduction in density ; beyond 100 to 200 yards flies will appear only as stragglers; and at 300 to 500 yards they will disappear, or, at least, density will drop below any figure that can be conveniently measured. If range inland is more extensive than this, some special cause for it must exist. Attempts to measure the ordinary inland range of fly were limited to a single experiment on the island of Kitobo in December 1913, which resulted as follows— the catch being made at the base of a slope that rose rather steeply to a plateau with abandoned plantations. (659) Cc 389 W. F. FISKE. Catch at base of slope, 54 “boy hours,” showed density (both sexes) of KA se ae BS 15:50 Catch ha.f-way up slope, 16‘ ‘boy hours,” showed density of 1:05 Catch at crest of slope, 28 ‘“‘ boy hours,” showed density of ‘26 (This point was well under 300 yards from the water’s edge.) Catch in old plantations beyond the crest of slope, 14 ‘ en hours,” was Nil. This is probably a sharper deloe, in denear ee cee bee not ea eaustale SO. V (0). Effect of Marshes and Rivulets on Range of the Fly. The effect of lakes or streams on the range of Glossina pa palis is so conspicuous that whenever. any unusual extension of its range inland from lake shore or bank of large stream has been observed it has almost always been accounted for as being due to presence of small streams, marshes or humid conditions of life generally. Therefore if any island were to show inland range of fly too much in'excess of that noted as “ normal” in the preceding sub-section, the question of interior conditions with respect to water and elevation (dryness) is the first to be considered. A survey of the islands was begun in January 1914 to ascertain and measure variations in degree of local infestations, as they might be correlated with differences in local environment. Mostly small islands, of less than 1,000 yards short diameter, were visited ; but by the Ist September out of a total of some 40, 8 of larger size had been included in the survey, and conditions with respect to infestation by fly, humidity and topography of the interior of each of these islands were found to be as stated in the accompanying table (XVII). TaBLE XVII. Fly Survey of Islands in Victoria Nyanza showwmq Lack of Correlation between Humidity of Interior and Infestation by Glossina palpalis. | Conditions in Interior. Infestation by Glossina. Island. Topography. Humidity. Shore Line. Interior. Wema Hilly Diya e, aca Dense None. Damba .. Flat Sinem and Dense Medium. marshes Kome .: Hilly Streams. . Light None. Kitobo .. Plateau Dry* Medium. . None. Bukassa . Hilly Dry* Light Light. Bugaba.. Hilly Streams. . Light None. Bubembe Hilly Dryt Light None. Bugovu. ..| Hilly Dryt Light None. * Careful survey showed no streams near points of interior infestation. + Streams may exist on paris of islands that were not explored. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 381 It will be seen that of the eight islands listed, two differed from the rest in being infested in the interior. One of these, Damba, is flat. with streams and marshes in its interior, and with a dense infestation of fly along shore. Two others—Bugaba and Kome—also had streams flowing down from the hills, but they were hilly and the infestation along shore was light. Hither elevation or lightness of infestation along shore might be a possible explanation for absence of fly mland, and its absence is merely negative evidence that it does not always occur in humid localities near lake shore colonies. But in the case of Bukassa Island we have a dry, hilly interior, with only a light infestation along shore, and for all this a general infestation of the interior. This is not negative, but is absolutely positive and very strong evidence that, whatever it may be which restricts the range of this tsetse to a narrow belt along the shore of the islands generally, it is not the presence of water or soil humidity, for the interior of Bukassa is, perhaps, the driest and hilliest of any of those listed except Wema. And whereas it is perfectly possible that the presence of open water might account for infestation by tsetse of a belt several miles in width along the borders of it. it is impossible ta consider water alone as the determining factor when, on some islands, the inland range of fly is virtually limited to 500 yards or less, whereas on others it is extended to 1,500 yards and more, and with no surface water in sight. Slight though this evidence is, it is of such a positive character that it 1s practically conclusive. Some other factor than open water must be held accountable for limitation to inland range of Glossina palpalis from shores of lakes or banks of streams. V (c). Effect of Food Supply on Range of the Fly. A comparison between the two islands, Bukassa and Damba, in search of points of similarity, which should at the same time be points of difference between these and the other islands named in Table XVII, led to the conclusion that the most probable explanation for their infestation interiorly by tsetse was the presence of unusually large numbers of game animals: of situtunga on Damba, and of domestic pigs run wild on Bukassa. This conclusion was reached reluctantly, because at that time (September 1918) it was believed that the only favoured hosts of this species of tsetse were reptilian. The course of the fly survey of the islands was thereupon set to include careful study of conditions on all such as were known or suspected to be more than usually well stocked with game. These included specifically the group of semi detached islands of Bukone, Serina and Lulamba, and the island of Buvu, upon which pigs were known to occur, and the (relatively) very large island of Sesse (Bugalla- Buninga), which had been reported to be overrun with situtunga. Examination of the pig islands disclosed the fact that though great numbers of the animals had been present (enough to cover them with networks of trails and to uproot the soil nearly everywhere in the woodland) some great catastrophe had reduced their numbers to a few stragglers. More skeletons were found than traces of living animals. But one result of the visit was to discover a “ following swarm” of fly about a pig that was shot, and to prove conclusively enough that this animal may be a favoured host. None of the islands was found to be infested by fly except very narrowly along the lake shore. (659) C2 382 W. F. FISKE. On Sesse the outcome was much more decisive. Situtunga were found everywhere in numbers which are extraordinary for an antelope accounted “ rare’ and a denizen only of marshes. They were even more numerous in the overgrown plantations and village sites on the plateaus, and elsewhere on the higher levels, than in the woodland and marshes along shore. The island is nearly forty miles in length, and eventually was explored from end to end. Practically every bit of woodland or jungle upon it was found to be infested by Glossina, including points in the interior of the western peninsula known as Buninga that are at least 24 miles from the lake. The only exceptions were certain areas of woodland in the central portion of the eastern half of the island (Bugalla) which, though hardly a thousand yards from the lake, were completely surrounded by open grassland. Elsewhere a great diversity of inland environment was found _ high hills and low; marshes and rocky summits ; original forest and the frequently impenetrable jungles which have sprung up in the abandoned plantations; but everywhere that shelter at all attractive to the flies was found, there would be both situtunga and fly. Except at the very beginning formal records of “‘ catches ”’ in the interior were not kept, because the work of exploration was done almost entirely on dull days or at hours (after 3 p.m.) when the flies are not as active as earlier. The records made at the beginning are presented in Table XVIII, in comparison with those from Bukassa, where inland extension of range is due to great numbers of pigs. TaBLE XVIII. Catches of Fly on Islands of Bukassa and Sesse (Bugalla) showing Inland Extension of Range due to Presence of Game, made in September 1914. | Bukassa Island. _ Bugalla Island. Collection Points. _ Female Male Female Rite. ratio. | density. ratio. Lake shore... w ne v adh; O78 Ay) LOMES,. etga-p 22°5 % Sandy plain, 400 yards inland me Ce lene — 25°0 lob Hills, 1,000 to 2,000 yards inland .. a RE Eh EOF 10°0 Tore 57. OUD 59) BLOOD res ‘aila, Se elder rch) hawt 47 94 % » 90,000 ,, 4,000 oe af % ee ee — 1°3 =~|\ None S400) OF more. |”. iy ey ..| — ‘4 |J Caught, Of significance is the low percentage of females in the interior points; indicating attractive shelter or ample food supply or both. The only other host animal occurring on the islands, that is found at all commonly more than a few yards from the water front during hours when flies are active, is the monitor lizard (Varanus).* On certain islands—notably Manene and Dziru, *Execept on the sudd-surrounded islands of Baujako and Binga, where bush-buck, bush-pig and buffalo occur, and which are faunistically a part of the mainland. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 383 both of very small size—it is extremely common, and on the island of Mbugwe it is far more numerous inland than is usual. On these islands, also, unusual extension of inland range of fly was noted, as described more in detail in Sect. V (d) following. At the end of the island survey the conclusion was definitely reached that the range of fly inland up to a distance of approximately 24 miles is controlled in the islands by the abundance of food. A summary of observations on this point in tsetse bionomics is presented in Table XIX. TABLE XIX, Fly Survey of Islands in Victoria Nyanza showing Correlation between Abundance of Host Animals and Infestation by Glossina palpalis in the Interior. Infestation by Fly. Host Animals in Interior. Island. Shore, Interior Pig. Situtunga.| Varanus. Wema ..\Heavy ..\None ..\None None .. Pew. Kome Oy eden” ally a’ ee ..|Very few. .|Very few. Bugabu . .|Laght a as | i Alpe » --iKew, Bubembe . |Light Oe | Rs Saintes . .|Rew Eat Bugovu . Light NW i a Wer le Welles: Buvumira .. (Light i SALA Ess a iets pan ial ne Fumbe . (Light a 4 ane .. (Few Seats Bunyama . (Light sis sialy ess ..|Very few. ./Very few. Lulamba ..|Light NTA ..|Very few. .|None .|Few. Buvu. ..(Medium ..| ,, re S ae ‘¢ Bukone . {Light 5 Bia si - a Serinya ..|(Medium ..| ,, BS ; " is Kitobo ..|Medium ..1 ,, None i = Bukana ..(Light ..(/Light Many 55 5 Damba .|\Heavy ..|Medium ..jNone Many r Zinga a sag .|Very heavy|Light * i a Sesse (Bugalla) :— Western Portion . .|Heavy sHhaghiby ls) .aliag Very many ,, Northern ,,. .. . .|Heavy Medium . a % ae North Central Portion ..|Medium . .|Light ~ s am i South ,, vA ..|Medium . .\Light fA. cH ys 3 Southern Portion Weary.) oc.a) Medi. Al... ipalieas ‘5 me Mbugwe ..(Medium ..|Medium ..' ., .|\Very few. .|Many. Manene* . (Medirim © .(Mediunmy 21° ,, A age rg’ s SI EY emiaiany Dziru* . Light .|Light bie (dha .|None ..|Many. ; *Manene and Dziru are small islands less than 1,000 yards in short diameter, but are remarkable in that infestation of the interior is equally as heavy in the case of Dziru and heavier in the case of Manene than along the shore. V (d). Unland Range of Fly as affected by Varanus on the ‘Islands. Manene Island. Manene Island is, perhaps, 1,000 yards in length by 500 in breadth. Its northern half is elevated and rocky, its southern half low and fertile. Mostly it is covered with thick bush or forest, except for the rockiest places, and also excepting the site of old plantations in the very middle of the southern half, which are kept open and closely grazed by hippopotamus. These spaces are entirely surrounded by thick bush or forest. 384. AW, aE SU UBISIKGR, Varanus was more common than on any island of similar size visited. Perhaps as many as six per hour were flushed during the survey of it. They were everywhere, but were especially common in the clear space above-mentioned, where many had their burrows. There were several bits of sand or gravel beach which offered good breeding grounds and excellent shelter for fly, but the shelter in these was less attractive than that in the close-cropped clearings, surrounded by forest, bush or jungle and with bits of massed vegetation breaking the openings everywhere The catch of fly was as follows : Male Female Density. Percentage. Shore points .. a bt mo ¥ 12°3 ie 217 Inland clearing ay sis si8 21:0 a 18 This is quite an exceptional case of concentration of fly away from the water (150 to 200 yards), due to more attractive shelter and constant abundance of food. No breeding places could be found except on the water front. Dzru Island. Conditions generally similar to those on Manene, except that shelter in the interior was not so attractive and concentration of fly less notable. Mbugwe Island. Mbuegwe Island is about two miles in length by one in greatest breadth. It is densely forested, and appears never to have been cleared for cultivation except at a few points, now overgrown with jungle. The one open space discovered on the island is on the crest of a steep hill that occupies the centre of the broadest portion. The very summit of the hill is flat rock, partly covered with thin soil, rank grass and scattered bush. The shore is generally rocky with here and there bits of beach. Varanus is common. Ordinarily this reptile occupies burrows in rocky places or excavated in light soil in grassy openings, near which are located its habitual basking spots. On Mbugwe there are no such openings as are frequented by i1t— at least none were seen—except on the crest of the hill above-mentioned, the slopes of which to the very edge of the water were covered with dense, old forest. Hence, although this opening was fully 800 to 1.200 yards from the water, it was much frequented by the animals, which had fairly covered the soil in places with their excreta (containing shells of crabs and molluscs from the lake). The catches of fly made on this island (4th and 5th March 1915) were as follows :— Male Female Catch. Density. Percentage. Points on shore at Py Se ks oe 16°6 ie 33°5 Crest of hill .. eS $e 83 Be 13:7, vi 50°6 Breeding grounds were sought for and found in and near the basking spots of Varanus, in dry vegetable debris that partly covered the rocks. Six boys, searching for one hour found : Total. Per Boy Hour. Empty shells :% Ae beh yi ene o she 9°2 Healthy puparia .. ey “4 4 agg" ie 3°5 INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 385 The shells had all hatched, and none had been eaten or destroyed by small predators. This particular colony is the one and only inland infestation discovered at any pomt which appeared to be quite independent of colonies on shore, and this is the only occasion on which breeding grounds adequate to sustain an independent colony have been found except in old or recent deposits of beach sand or gravel. The conditions are unique in many respects. The very dense forest which clothed the slopes of the hill was a strong deterrent to dispersions of fly. The local environ- ment was far from being attractive, and it is exceedingly doubtful if the flies would have remained in it if the thick forest had not been so relatively less attractive. But they were walled in as on an island, or as in a great breeding cage—it chanced that there was a regular supply of food—it chanced that there was protection for the pupae—and in consequence the flies remained and bred. There is no reason why such colonies should not be found at any distance from the water front, except that such conditions are exceedingly rare in this region. V (e). Dispersion of Fly inland from the Mainland Shore. At numerous points on the mainland it was ascertained that inland dispersion of fly was no more extensive than is usual on theislands. The two notable exceptions follow. | Dumo Point, Buddu, June 1915. Dumo Point is a peninsula. with a hill and old plantations, separated from the mainland by an open plain. It is in large part densejungle. Near the very centre, about 500 yards from the lake shore on two or three sides was a semi-open space, kept grazed by hippo, in which grew great quantities of ga Bush-buck was rather common. A herd of bush-pig had a retreat in the jungle and clearly marked trails showed where it made regular rounds of the guava thickets for the fallen fruit. In the immediate vicinity of ae thicket, but not elsewhere except in the usual narrow belt along shore, Glossina was common. The catches stand as follows: Male Female Density. Percentage. Points along shore of peninsula .. #': SONGS ARO ez At guava thicket, 500 yardsinland ia Sorc awa The shelter at this point was quite attractive, but no more so than at many other points ; the inland dispersion of fly was distinctly unusual and the presence of both bush-buck and bush-pig, the latter particularly, led to the conclusion that the presence of fly was thus explained. Bujaju Pernnsula, July-August 1915. Bujaju Peninsula is an extensive tract of land, almost an island, but with the deep bays behind it choked permanently with sudd. _Its shore is very marshy, and principally fringed with papyrus; but inside the papyrus are open spaces, with occasional bits of sandy soil, or traces of old beach line, and these are sparsely infested by tsetse. The interior is much of it “ impenetrable ” jungle, cut with great numbers of hippo trails, and with clearings, closely cropped by hippo, where once were villages 386 W. F. FISKE. and plantations. Not far from the northern extremity are natural clearings, with very short grass and flat outcropping ledges of rock, and where the soil is sufficient, exceedingly dense clumps and thickets of bushes and vines. Everywhere on the peninsula the inland range of fly appeared to be normal, except in these natural clearings, which extended a mile or more and were distant 800 to 1,200 yards from the water. Catches were attempted in the clearings kept open by hippo, but without result, or only straggling flies were caught ; but in the natural clearings the greatest density of fly was found of any point on the peninsula. The records are Male Female Density. Percentage. Points along shore of peninsula opposite clearmgs 59 .. 42°3 Point nearest clearing on foreshore... My. OO ase Eile In clear space kept open by hippo 100 yards inland opposite natural clearings .. oe Gey A sD In hippo trail leading to clearing 400 yardsinland 100 .. 12:0 In clearing, 800 to 1,000 yards inland .. ac RUM vee slid In open space 300 to 1,500 yards beyond clearing nil Wo ks Breeding grounds were sought but not found ; but this is no proof that they may not have existed. The shelter was of a most attractive type, but no more so than in other clearings, kept open by hippo, where the soil was much deeper and the vegetation more luxuriant. The only explanation for the presence of fly in such unusual numbers—for, in fact, the major concentration of fly on the peninsula so far as the survey of 1t extended—was the excellent shelter (probably coupled with breeding grounds) and rather unusual numbers of bush-buck, which appeared to be virtually the only source of food. It was less the numbers, however, than the easy accessibility of these animals which was conceived to be mainly responsible. They found refuge in the dense thickets and clumps of bush, which were so small in extent that the flies could find them at all times without penetrating far into the shadow. Kitebo Peninsula, August 1915. Kitebo Peninsula in the district of Mawakota, is bounded by permanent sudd fields, except for a reach of about 3 miles along its south-eastern and southern shore. Here it is narrowly-fringed with sudd of more recent growth and less permanently lodged, which in 1915 was broken at three points. Hach break was the centre of a colony, or centre of infestation, for the space between the breaks and on either side of the semi-open reach. To the northward, behind the permanent sudd fields that fill the channel between the peninsula and the island of Bujako, infestation falls away abruptly. To the westward, behind the permanent sudd fields that bind the western shore and south- western extremity of the peninsula a curious situation was discovered. The south-western extremity of the peninsula is a marshy plain with belts and patches of drier land. The wetter parts of it are overgrown with a terrible tangle of tall grass, shrubbery and briery vines that defies penetration. The drier parts are in places open game pasture, and in part covered with patches of dense jungle. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 387 The whole area forms a triangle, with its base against the higher ground (densely forested) and its apex lost in the tangled mass of vegetation that stretches outwards into the sudd. Crossing this area, some distance from and very roughly parallel to the forested base of it, is a sandy ridge, densely overgrown with shrubbery and jungle. The tangled marsh is beyond this ridge, and inside it are the game pastures, very closely cropped by hippo and bush-buck. (Hippo crop almost as closely as sheep, using their horny-edged lips.) The ridge is an ancient beach line and affords fair breeding grounds and adequate shelter, which is supplemented by the patches of jungle scattered over the game pasture. In addition to bush-buck, which were unusually common, were plenty of more or less recent tracks of water-buck, bush-pig and buffalo, which came out from the forest, and from other game pastures 2,000 to 3.000 yards through the forest. The infestation along the ridge was interesting :— Male Female Catch. Density. Ratio. At break in sudd, and hippo landing 195 44:5 a eee 200 yards from open water, on ridge 114 29-0 49°1%, 700 ., : . 64 20:0 BY abs 1,100 ip e : BD 20:0 27-29/, 1,500 f 4 17 6-5 93°59, There could be little doubt, in this case, that the bush-buck were mainly responsible for inland dispersion, for they greatly outnumbered the other game animals ; of Varanus there was none, of crocodile but few, and the hippo herd slept far away from its nightly pasture. There was also infestation at open points in the forest at distances believed to be 2,500 yards from the lake, but it was slight and not accurately measured. At the game pastures from 3,000 to 5,000 yards away from the open water—not far from the western sudd field—no fly could be found; water-buck was the principal species grazing in them. V (f). Dispersion of Fly inland from Lake Shore along small Rivers. At various points on the western shore of the Lake, in the district of Buddu, larger streams or creeks than exist on the islands flow into it. One of these, Mujuzi, enters the lake about two miles north of the old canoe landing at Kalkosa, and three miles south of that at Bale. South of the creek all the way to Kalkosa the shore is marshy, with an old beach line overgrown with dense vegetation. At the mouth of the creek an old sand beach begins, at a level of some two to three feet above the present beach line and continues nearly to Bale. The foreshore also is sandy, but except for a reach of 600 to 800 yards just north of the creek, it is overgrown with dense vegetation of an impenetrable character to the water’s edge. At the mouth of the creek the breeding grounds and shelter are of the best, and food is provided by a large number of crocodiles which find harbour there, and which also rest in the sand. To the north there is some breeding ground, but 388 Wi Re RIGIKOR: shelter is repulsive and no indication of any sort of food was seen. To the south there are situtunga and perhaps other host animals, but shelter is repulsive and no breeding grounds were seen. Conditions of life at the very mouth of the creek, over a radius of hardly more than 50 yards, are highly favourable for tsetse, but favourable conditions are extremely localised, making for a sharply defined colony that fuses with a small colony at Kalkosa landing (point 40 in Fig. 5), to the southward, but tapers away to a minimum of ‘5 at Bale (about 1,200 yards beyond point 53), where a colony yet further north fuses narrowly with it. Mujuzi Creek is almost a river, being at points some 10 yards in width and of considerable depth. Its shores are thickly wooded, and cannot well be followed because of the tangled vegetation. There are occasional pools with low sand bars, as at points 45C and 45D. The stream was quite open as far as point 45H, where it was choked with floating vegetation which extended for a considerable distance. At points beyond 45F it was open again, and at one point ran through a tract of woodland which seemed the headquarters of a herd of buffalo, but beyond point 45F, which was approximately 2 miles from its mouth, no fly was seen. At no point from the mouth inland was shelter especially attractive; no breeding grounds were seen, and host animals were certainly not abundant. If allowance is made for the general attractiveness of the shelter from point 45 to beyond point 46 along the lake shore, the falling off in density of fly appears to be approximately the same on the lake shore northward to Bale, and along the creek inland. It is more pronounced along the creek, but not very much more. The conditions of life are unfavourable to tsetse in all three directions from the colony centre, and infestations of the creek banks and lake shore alike for a distance of from two to three miles is plainly due to the existence of exceptionally favourable conditions in the extremely limited space around the mouth of the creek. Hixtension of range inland along other creeks wide enough to create something of an alley-way through the bush or forest for the moving flies to follow has been found to be much as in this instance.* The flies follow the waterways much as they follow the lake shcre. But as yet no colony centre, or semblance of such comparable to the hill colony on Mbugwe Island (Sect. V (d) ) or the inland colony in Bujaju (Sect. V (e) ) has been discovered on the banks of any stream. Every bit of evidence that has been gathered supports this final conclusion con- cerning 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 tt. 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. : It may be added that every instance in which fly was encountered other than as mere stragglers at a distance greater than 500 yards from the lake shore has been covered in this section. *If the stream is so small that no break is caused in the continuity of the forest, fly has not been observed to follow it. Observa Point Observa e Meee lhe {a ’ Density Male Flies Solid li densi along sh Dotted dens! on Mujuzi Creek. 1e er fa Be iy a ; ce 1 ; 4 i re - ee) ; Ate y F > {i ‘ 4 BE j i , : — i a ss Bit ut we rer, oh Ls if « Observation Point Density Observation Point Density Density of Male Flies Sekwe or Solid line Kalkosa Landing density along shore. Dotted line density \ Mujuzi Créek| \ along creek \ \ \ \ \ 1 i) Shore Line Scale of Miles Fig. S. Fly survey of mainland shore from Kalkosa to Bale, Buddu, showing inland spread of fly on Mujuzi Creek. we ¢ ; ) , t ‘ , re ‘ ‘ ‘ pain cl ety a é 4 i A r ae) 1%, Cen Hanis ‘ ( | cee wineemteter te ep LETTE AP INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. 389 VY (g). The Sex Ratio at Points where Range is extended inland. In the records of catches at inland points which have been given in the preceding paragraphs a very significant feature is the falling off of female percentage coinci- ‘dently with the falling off of male density, as distance from the water point increases. This is notable in the catches on Bugalla Island (Table XVIII), at Dumo Point, on Bujaju Peninsula, and on Kitebo Peninsula (Sect. V (e) ). On Bujaju and also on Manene Island the male density was greater inland, but the female percentage ‘strikingly lower. On Bukassa Island male density fell and female percentage remained the same, with no serious exception. The Mbugwe hill-top colony is the one striking exception, and this is also unique in being, apparently, quite independent ‘of colonies along the shore and entirely different in this respect from the infestations at Dumo, Bujaju, Kitebo and on Bugalla Island, which are extensions inland of shore colonies. This same phenomenon has been noted repeatedly at other points on Bugalla Island, where collections were made at distances of 500 to 2,000 yards inland, and the whole series of observations falling into this category stand in the sharpest contrast to the rule that applies along shore (so well illustrated by the curves in figure 4, p. 372) that female percentage rises as the distance from the centre of infestation ‘imcreases and male density decreases. And it stands, perhaps as the best bit of other than purely empirical evidence, in proof of the effect of food supply or ‘distribution of host animals upon inland range of fly. The shelter on Bugalla Island, for example, or at Kitebo, is in every respect as good and attractive to the flies as on the foreshore. Food is better, and more abundant. It is only the protection afforded by the breeding grounds that is lacking to make the inland conditions of life entirely favourable. TABLE XX. Mujuzi Colony. Kitebo Colony. Dispersion forced, by hunger, into an Dispersion induced (inland) by superior intolerable environment. attraction of environment, Distance of observa- Female | Distance of observa- Female A Male : “| Male tion point from colony density per- | tion point from colony) ee per- centre. "| centage. eentre. | CenST'Y- | centage. 0 yards 98°0 45°3 0 yards 44°5 54:3 700." ;; 53°0 49°2 200K Fe 29-0 49:1 300): 555 20°5 64°0 700 ,, 20 0 SSeS) E00 ~., Ons 2h TG ae 20°0 Of ZIaOO . 55 10°0 80°8 T500t 6°5 Pe mS) 2,900 ,, a0) Ses The fly cannot breed inland, and hold its own numerically, on account of lack of good breeding places; but there is continual inland dispersion of flies from the colony centres at the breeding places on the foreshore, induced by relatively better or more food and equally attractive shelter. The flies are drawn away from the 390 W. F. FISKE. colony centre, or centres of infestation, by more attractive conditions elsewhere. In the colony at Lutoboka Bay (fig. 4) or at Mujuzi Creek (Sect. III (e) ) dispersion is, on the other hand, forced, and the flies, instead of being drawn outward by a superior attraction, are forced outwards into a relatively repulsive or intolerable environment by the stimulus of hunger. In this latter case female percentage tends to soar as male density falls; im the former male density and female percentage fall together, as shown in Table XX. And it is certainly superior food in combination with equally attractive shelter which explains inland dispersion in all the cases that have come under observation, except inland from the lake along the banks of the Mujuzi and a few other creeks, which is comparable to dispersion along the shore of the lake. VI. Hosts anp Host PREFERENCES OF GLOSSINA PALPALIS. VI (a). Methods of Studying Hosts and Host Preferences. These methods were used for the study of hosts and host preferences of Glossina palpalis :— (1) Experiment. The animals were tethered along fly-infested shore (see Sect. VI (c)). (2) Approach and observation. Wild and domestic animals were approached and the behaviour of fly toward them was observed under entirely natural conditions (see Sect. VI (d) ). (3) Shooting. When possible, wild animals were shot in the fly belt at hours when flies were active. If the animal falls without running the flies which have been following may linger by it for a time, and some of them will usually feed (see Sect. VI (e) ). In addition to the above the study of sex ratio, as affected by the density of favoured hosts, affords an opportunity for securing supplementary or confirmatory data (see Sect. V (g) ). A list of the principal animals that occur within the fly belt, together with methods. by which they were studied as hosts of fly, and a summary of conclusions reached is presented in Table XXI. The list excludes many small mammals, such as rats, bats, etc., which though common are certainly of no importance as hosts of fiy, and it also excludes others such as Hyraz, the “ edible rat,’”’ and others which are rare or nocturnal and of no importance, in this region at least. All animals which are fed upon by Glossina, and thus technically included amongst its hosts, are not equally favoured or preferred by it, and a broad line of distinction may be drawn between the two following categories :— (A). Hosts which are positively attractive and which are favoured or preferred to such an extent as to cause concentration of fly in localities that they frequent. (B). Hosts which are rather repulsive than attractive and which are avoided to such an extent that the flies tend to scatter and disperse from localities. in which no more favoured hosts occur. INVESTIGATIONS INTO THE BIONOMICS OF GLOSSINA PALPALIS. aol: In numerous cases there is no doubt into which category an animal falls, and such are designated by an exclamation point (!) in the table. In other cases there is con- siderable doubt. These are indicated by an interrogation point (?). In one or two, notably that of the domestic ox (see Sect. VI (c), Experiment 3), some individual animals appear to be attractive and others repulsive. There are two tests for distinguishing between attractive and repulsive hosts :— (1) An attractive host is apt to collect a “followmg swarm” of fly. This curious phenomenon is associated with the assembling of the sexes and is described in Sect. IV. (2) The range of fly—especially its range inland from the shore—is likely to be notably extended through the presence of favoured hosts. This phenomenon is discussed in Sect. V. TABLE XXI. Summary of Observations on Hosts and Host-Preference of Glossina palpalis on Victoria Nyanza in Uganda. Methods of Study. Conclusions. Animals observed in fly Approach Is animal |Is range of belts. Experi- and Sout Is animal | a favoured |fly affected ment. observa- cotms: || fed upon ? host? |by animal ? tion. | | | Crocodile ..| Yes wilh Hees ..| Yes We eet aia Westic ic. i) Mes Varanus ..| Yes wl MUER sale Les blll NAC Lat ctl MOR cons la Mega Tortoise eo) ee eo ees ot NO ..l| Yes aa es Hole NO Serpents :.| Ne oo Mes ..| No | No? .| No | No Lizards . ae. No Sale Gs ..| No | No? .| No! No Situtunga | NG ae fo wal nes sill ee EN Gwe ce alee Ses: Hippo .. ..| No eo ae es Pl) Yes 2. Mes ..) Les? Bush-buck ..| No ..| No .«| Yes LI NGEh Ala New ..| Yes?? Bush-pig =|) DO .2| Noe 2c ees pli INO? \ chcNo? Ves ae Water-buck ..| No eee NO ..| Yes alt NO 2 NO? ..| No? Buffalo . ..| Ne ..| No ..| No ..|| O Bur ae : ! No? Duiker.. ..| No s shaNe ae. lea © spsille