“Ye PS CPS EPCS i a4 Oh FOPOW porte y awenty, o diner I AEs baseaietite ite 4 wily tr ete trad ciggibinde: : * * < oo. » * q : ; . + ied a gil - . v~ : : $e - : . a : ‘ weer Ce lh Life eri Chath ea ee care it ee tag dot Petatabatnintesmbatoad . — ss aaaen “ i ; vee dette te ved eves caemetnngue cs cmmeemeagepepyeocen yee oie eee. wate rele reravery'etgte’ a at vr new ee . a. ee ewe sewn Ee we ee TERY oe wer — elev ole ee ee we Orrre wees . P29 ATE ees We one re wRlelen® ‘were wen oe owes i “woe ae ay ‘walw es Oe wee Sree W Serene O aCe Tiere goer ener: Severn es Setgletieeie eae cbt dbibtdieiekbuniua aid "| 7 oon wrete " iethdith Ad bhok de ath Sok le th ak ae eee oor eh eree Pp ipraieren af eee to Ce hee ee bit ict Sill Meth i tere Sdliedidalirds-tedindh iedind-4-dl ‘ewe a ee = ‘. www tes OVP eee ree cdone We we riewwig indediediin ds vs See tore ews reset payor Pe Pe er ewes Peewee ‘wee * Pee re ree Ve owe oor “eee see eee repr Pe wri we Bhs owe e's bets 'eX9 Peeve rte rey CUOCT CM et eee teres . ore ae PVE STD LWW we Ri Glee er wale Owens t rer ete ee eorrtirrrery wh heuvwe es . ey ee Peer ev bee eee erate Rad € + eed 2 ehagee se se i Pea . res serene “w eratep < fey © 5 SF ewe wert cre ee oP eee e Teer ere e wes ryVeresy re . vrerre serv : e* ree ees oe D4 "ae e ‘ rere eter eres t r yee.» wv? . . erere es vey eee Cow Be . . . 7 : : : : Me : <veeewveerr rer eee. . e 2h ere : — , seer ivewess : 7 : ‘ Perereve ‘ -+> » ’ . 4 Peet seers eve J * dee ’ teve : " + ' » ¢ Md ‘ . ‘ete ; - - : . 5 ° ® ) 3 ~ >! A . te i. . . . : . etetene : : ite lee . vous ost e ; : : oreee y . ‘* : e e v : : : : : " . - 7 oe . 4 > ve . . te re - . + . . ° e» | te erepeewreee 80 eevee 2. . ’ : re * t SMITHSONIAN. DEPOSIT wr. a . a PD BBA ae ae Lt Si a: , | j rh) OR eee Teer ne ~ — >. -| OY \ \p be Ww’ A oy Cy 72 Wn ANS wf oo Se A ar Baa >. BULLETIN 234 \V ? OF ENTOMOLOGICAL RESEARCH ISSUED BY THE IMPERIAL BUREAU OF ENTOMOLOGY. EDITOR: THE DIRECTOR, VOi -VilT. LONDON: SOLD BY DULAU & Co., Lrp., 37, SOHO SQUARE, W. 1. 1917-1918, IMPERIAL BUREAU OF ENTOMOLOGY. VISCOUNT HARCOURT, Chairman. Tropical Medicine. Mr. E. E. Austen, Entomological Department, British Museum (Natural History). Dr. A. G. BagsHawe, C.M.G., Director, Tropical Diseases Bureau. Sir J. Rose Braprorp, K.C.M.G., F.R.S., Secretary, Royal Society. Surgeon-General Sir Davip Bruce, C.B., F.R.S., AMS. Mr. J. C. F. Fryer, Entomologist to the Board of Agriculture and Tisheries. Dr. S. F. Harmer, F.R.S., Keeper of Zoology, British Museum (Natural History). Professor H. Maxwett Lerroy, Imperial College of Science and Technology. The Hon. Sir Jonn McCatt, M.D., Agent-General for Tasmania. Dr. R. Stewart MacDovuea tt, Lecturer on Agricultural Entomology, Edinburgh University. Sir Jonn McFapyeay, Principal, Royal Veterinary College, Camden Town. Sir Parrick Manson, G.C.M.G., F.R.S., Late Medical Adviser to the Colonial Office. Sir DanreEL Morris, K.C.M.G., Late Adviser to the Colonial Office in Tropical Agriculture. Professor R. Nerwsreap, F.R.S., Dutton Memorial Professor of Medical Entomology, Liverpool University. Professor G. H. F. Nurratt, F'.R.S., Quick Professor of Protozoology, Cambridge. Professor EK. B. Poutron, F.R.S., Hope Professor of Zoology, Oxford Lieutenant-Colonel Sir Davip Pratn, C.1.E., C.M.G., F.R.S., Director, Royal Botanic Gardens, Kew. Sir H. J. Reap, K.C.M.G., C.B., Colonial Office. The Honourable N. C. RoTuscHiLp. Mr. Hueu Scorr, Curator in Entomology, Museum of Zoology, Cambridge. Dr. A. K. Sarerey, F.R.S., Vice-Chancellor, Cambridge University. Mr. R. Speriine, Foreign Office. Sir Srewart SrockMan, Chief Veterinary Officer, Board of Agri- culture. Mr. F. V. THeosatp, Vice-Principal, South Eastern Agricultural College, Wye. Mr. C. WarBurTON, Zoologist to the Royal Agricultural Society of England. The Chief Entomologist in each of the Self-governing Dominions is an ex officio member of the Committee. General Secretary, Mr. A. C. C. Parxinson (Colonial Office). Director and Editor. Dr. Guy A. K. MarsHatu. Assistant Director. Dr. 8. A. NEAvVE. Head Office —British Museum (Natural History), Cromwell Road, ~ London, 8.W.7. Publication Office —89, Queen’s Gate, London, S.W.7. cA ie go Se ee Alo \ Honorary Committee of Management. Cn ego Lieutenant-Colonel A. W. Atcock, C.I.E., F.R.S., London School of ° oe 2 ee CONTENTS. ORIGINAL ARTICLES. Arrow, GILBERT J. PAGE Some Melolonthid Beetles injurious to Cacao plants in the Belgian UMMA ore thier ee ares Sin sclre AAS sae HEE ae hoe URL eon cee 111 Bezzi, Pror. M. New Ethiopian Fruit-flies of the genus Dacus ..............00006. 63 Notes on the Ethiopian Fruit-flies of the Family Trypaneidae, other than Dacus (s.l.), with descriptions of new genera and species (Dipt.) 215 Bopkxin, G. E. Mates on tne Coccidge of British Guiana .......0.ce ccc ccc cc ccc ee 103 Davipson, Capt. J. Some practical methods adopted for the control of flies in the Egyptian RAPIQN OT | fo oe ok oa Ae Uc gta be Neh eae A a a. wo! 2 297 d’ EMMEREZ DE CHARMoy, D. Notes relative to the importation of Tiphia parallela, Smith, from Barbados to Mauritius for the control of Phytalus smithi, Arrow.. 93 Distant, W. L. Some Coreid Bugs injurious to Cacao plants in the Belgian Congo .. 118 GAHAN, C. J. Some Longicorn Beetles injurious to Cacao plants in the Belgian EERO 0 EET ESPLES SE eA A Sg 8 OO 117 Gowney, C. C. A list of Uganda Coccidae, their food-plants and natural enemies .. 187 (C481) Wt. P.3/130. 1,000. 12.18. B.&F.,Ltd. Gp. 11/1. A2 iV CONTENTS. IncRaM, Dr. A., and Macriz, Dr. J. W. Scort. PAGE. Notes on some distinctive points in the Pupae of West African (22151 2,8 1UNLEC |< ee Ree aE ME EY iS 0 lek NMR 73 The early stages of certain West African Mosquitos................ 135 JACK, RupERT W. Natural transmission of Trypanosomiasis (7. pecorum group) in the absence-of Esetse-Hy ‘sais vice saree oa Sees ue ee eee be Slee 30 Keiitn, Dr. D. On the supposed first stage larva of Leptohylemyia coarctata, Fall... 121 MacGrecor, Matcorm Evan. A summary of our knowledge of Insect Vectors of Disease....... . 2 haS MarsHatt, Dr. Guy A. K. Some Weevils injurious to Cacao plants in the Belgian Congo ...... 112 A new Weevil pest of Sweet Potatoes in Jamaica ...............-. 269 NEWSTEAD, Pror. R. Observations on Scale-insects (Coccidae).—IV, V. .............-. 1, 395 Sampson, Lt.-Cou. F. WINN. A new Scolytid injurious to dried Sweet Potatoes in Jamaica....... 295 ScHWETZ, Dr. The western and northern limits of Glossina morsitans in Northern BGAN BA isos pda s sels eS ae eee > og Oe ee 165 Preliminary note on the Tsetse-flies of the Kabalo-Albertville (Lualaba Tanganyika) TRailweiy..¢i:. 200s. . Poa ate eee 169 Stupson, Dr. James J. Bionomics ot Tsetse and other parasitological notes in the Gold Coast 193 erty. H. The structure of the mouth-parts of Pangonia longirostris in relation to the probable feeding-habits of the species ...............5.5: 253 THEOBALD, FRED, V. Mivican Aphididae:—Part U1], 0050.45 de bagans shoes owe esses 273 TURNER, RowLanp E. On a: Braconid -parasito ‘of Glossina). ich. sane es igem eee eee Weg is 4 c ' ‘i : . 1 oP ; r ; v ae ! we a oe an a ; » iy | | CONTENTS. Vv ev crpasvaw: wine PAGE. | ’ A new species of Paraphelinus, Psy from British Guiana, with a 3 discussion of the genus and the allied Aphelinus, Dalm. ......... 43 Chalcidoidea bred from Glossina in the Northern Territories, Gold | Winrams, ee B. | ae A new Thrips damaging Orchids in the West Indies.............. : ae a eae ek ae v 7 bai pa Pp | a) MISCELLANEOUS. See 2c cs anche Saas aus al ye apr cts diw'e 18 és o gujaic eee dw. EO 59 Me recedes) GUM os iain arg due 119, 191, 311 Vi. PLATES. PAGE, I. to IV. Views in West Africa showing early stages of certain Mosquitos and their haunts: 3): 0:0 bees oes = - facing 154 i's Wanee Ol Aivican Trypaneidae : ccs sea se eee ses 5 oon V1. Mouth-parts of Pangonia longirostris, Hardw. ......... 33 208 MAPS. PAGE, I. Map showing the northern and western limits of Glossina morsitans in the southern Belgian Congo .............. to face 168 Il. Map showing the distribution of Glossina along the Lualaba- Tameamyilen Tarlywaiy «0 ihc 00s. ees ee i aes Vil. ILLUSTRATIONS IN THE TEXT. Details of scale-insects : — Monophlebus ? hirtus, Brain, 2 . Aspidoproctus neavei, Newst., sp. n., 2 Palaeococcus cajani, Newst., sp. n., 2 Icerya nigroareolata, Newst., sp. n., 2 ah » sulfurea var. pattersoni, Newst., nov., 2 Ditetien buxtom, Newst., sp.n., ¢, 11, ¢ Stictococcus intermedius, Newst., sp. n., @ and larva 9 multispinosus, Newst., sp. n., 5 and larva Lecaniodiaspis tarsalis, Newst., sp. n., 2 Pulvinaria africana, Newst., sp. n., ° eS aristolochiae, Newst., sp. n., 92 - elongata, Newst., sp. n., 2 9 ? flavicans, Mask., O, subterranea, Newst., ep. Ry, o Usroplanes bipartitus, Newst., sp, n., 2. ye destructor, Newst., 3 antenna and euceeusinn S egbarum, Ckll., g pupa es vuilleti, Marcial, 2 antennae and eens wend mi zonatus, Newst., sp. n., & Inglisia theobromae, Newst., sp. n., Q : Sketch-map showing the position of the Gatooma teatne area ‘ Sketch-map showing the position of the fly areas in the Sebungwe Distxiet Paraphelinus perkinsi, Waterston, sp. n., details of 9, 47, 48,55; 3 antenna and forewing, 54; left mandible of ¢g - tomaspidis, How., 9 antenna, 48; © wings re s xiphidii, Perk., details of 9, 47, 48, 50,55; gd antenna Physothrips xanthiws, Williams, sp. n., adult 9 ie a Wings of :— Tridacus @emmerezi, Bezzi, sp n. Dacus annulatus, Becker » blepharogaster, Bezzi, sp. n. » erythraeus, Bezzi, sp. n. » hamatus, Bezzi, sp. n. » mochi, Bezzi, sp. n. Pupae of :— Aédomyia africana, N.L. Anopheles mauritianus, Grp. i pharoensis, Theo. ; Chaoborus ceratopogones, Theo., trumpet ———S oT Vlil ILLUSTRATIONS IN THE TEXT. Pupae of: Culex fatiqans, Wied. » guarti, Blanch. » “%imvidiosus, Theo. » thalassius, Theo. » tritaeniorhynchus, Giles Cyathomyia fusea, Theo. Humelanomyia inconspicuosa, Theo. Mimomyia mimomyiaformis, Newst. ss splendens, Theo. Ochlerotatus albocephalus, Theo. Stegomyia metallica, Edw. = unilineata, Theo. Sketch-map showing five years gradual increase in othe Aiseivedion bf Phytalus smithi che Systates mayne?, Mshl., sp. n., ¢.. a ramosus, Mshl., sp. n., 3 Alcides theobromae, Mshl. 9 Sp. aise : Leptohylemyia coarctata, Fall., cae parts of egeenaes S aes ee 122; mouth-parts of a full- -grown larva .. ae M Pe nists africanus, Theo., details oflarvasiphon .. Head, end of abdomen, and mentum of :— Culex guiarti, Blanch. » thalassius, Theo. Cyathomyia fusca, Theo. Mimomyia splendens, Theo. Ochlerotatus albocephalus, Theo. avicoannulatus, Edw. a minutus, Theo. Lateral view of the heads of :— Dirhinus ehrhorni, Silv. (left) nS inflerus, Waterst. (right) Conradiina suspensa, Bezzi, sp. n., wing Rhacochlaena fasciolata, Loew, wing Ocneros mundus, Loew, wing ; mn ai) 8 st $e Pangonia longirostris, Hardw., 3, O54 : ©, details of head, 258-261; 4, details of maxilla, 262, 263; 4, labrum-epipharyax, 263; 6; hypo- pharynx i< $< my a . ‘ Palaeopus costicollis, Marshall, sp.n., 3 Details of :— Aphis acetosella, Theo, nom. n. » vbauhiniae, Theo., sp. n. » obuddleiae, Theo., sp. n. » durantae, Theo., sp. n. » jeeus, Theo., sp. n. » mathiolae, Theo., sp. n. » mathiolellae, Theo., sp. n. » pomonella, Theo. » pruniella, Theo., sp. n. » tamaricis, Theo., sp. n. zizyphi, Theo., sp. n. 275 alopionse i insignis, Theo., sp. n. ny ae 7 ms "290, 291 yt Details of Macrosiphum dahliafolii, Theo., sp.n. .. Le Myzus pterisoides, Theo., sp. n. .. fhe _ Earth incinerator in side view section .. i Fly cage made with wire gauze on a wooden framework A fly trap with poisoned bait... ae ds ILLUSTRATIONS IN THE TEXT. ‘Treatment of horse-manure by spraying Pa Beck Native latrine system for incineration of faeces ; > _ A portable latrine .. tc a, ; Portable incinerator made with galvanised i iron andi iron siokots oe 1x PAGE 14 ore se OZ ee | eS ve, wOZ -. 303 .. 3804 es §=~— SUS ow oh OOr ef ees ERRATA‘ Page 12, fig. 7, for baxtont read buatoni rh 13, line: 3, ,, Kantora ‘, (Rantara: of 16, . 4, » 2. @’Emmerez ,, .D. @inimerez. 5 36, ,, 44, ,, calacitrans » calcitrans = 44, ,, 5, .,, Haplogonotopus ,, Haplogonatopus. a 87, fig. Vi, ,, guern » guiarti. 5p OB ime 8g) Wook ti. rg ht) eee » 104, ,, 18. ,, Pseudoceus Pseudococcus. » 105, lines 13 & 19,for Pseudococcus sacchari, Newst., read Pseudococeus sacchari, Ckll. » 108, line 32, for Coccus » | Cocos, 29 118, 99 35, 29 C. 29 lags » 126, ,, JI, ., aegypnacum » aegyptiaca. » 129, ,, 27, , Aranthemum ,, Hranthemum. oo . obo, «ge. COL 5, hl eawem » Hevea. 5 Re » 38, ,, pseudaonidia ,, Pseudaonidia. 3, ta0,. ,, 19... Aédeomya , Aédomyia. Joie RE ype) Oe aay? chOmeeD sy a erzog. 5 185, , 53, » pimnformis » pinnaeformis » 187, ,, 13, 5, sulphurea » sulfurea. st 210, 4, 4, \5, suflezus » mflexus. wt holes 45 ESCH hh etsied » visited. See ere a. eee AOE » with. ae Teoh. VDD s5) WOEL » 243, ,, 21, ,, dipterologist » hymenopterologist. 3 206, 5, 26, 5 Tupells » rippells. BeeAO7, \ sat Hla wee «hoger, =. » moder, Ey von, » 2388, line 7; 244, line 36; 248, line 30; 249, line 20, and 250, line 7, for “"Zanry, Tread: Jabrh:” VOL. VIII. Part 1.—pp. 1-120. | AUGUST, 1917. BULLETIN OF ENTOMOLOGICAL RESEARCH ISSUED BY THE IMPERIAL BUREAU OF ENTOMOLOGY. EDITOR: THE DIRECTOR. : * _— rs Maal te * _ : Ft Nan ¥ ‘ at “ “tp, | r a . « ~ © ye. i 2 » 7 = nm x i iV ta} wT! tp, ® : ae ee Ei es ? - sag . P pis j e. _ ns < a ~i! Abe es “§ 3 P es * z - P as a a e F ae tO Te + tala t i act ’ o> * ‘ Ar bf jy28 } mary © ox, 5 ay a) ‘ : > » n é é o~ -"8 - , =o . my A , Py % Ag a ~ PA t r ‘ a a a ae a / “a be : a, ON} - if i a, Ly} by 10 he ae fr ' 7 Gate a LONDON: SOLD BY DULAU & Go,, Ltd., 37, SOHO SQUARE, W. 1. Price 4s. net. All Rights Reserved, IMPERIAL BUREAU OF EN TOMOLOGY. HONORARY COMMITTEE OF MANAGEMENT. THE VISCOUNT HAROOURT, Cbairman. Lrzvut.-Corone. A. ALCOCK, C.LE., F.R.8. Mr. E. E. AUSTEN. Dr. A. G. BAGSHAWE, C.M.G. Mr. E. C. BLECH, C.M.G. Sm JOHN R. BRADFORD, K.C.M.G., F.R.S8. Sura.-GenzraLt Sm DAVID BRUCE, C.B., F.B.8. Dr. 8. F. HARMER, F.R.S. Pror. H. MAXWELL LEFROY. Sm; JOHN McCALL. Dr. R. STEWART MACDOUGALL. Sm JOHN MoFADYEAN. Sm PATRICK MANSON, G.C.M.G., F.R.S Sm DANIEL MORRIS, K.C.MG. Pror. R. NEWSTEAD, F-.RB.S. Prov. G. H. F. NUTTALL, F.RB.8. Pror. E. B. POULTON, F.RS. Lrzvt.-CoLonet Sm DAVID PRAIN, C.M.G., C.LE., F.RS. Mr. H. J. READ, C.B., C.M.G. Tue Hon. N. C. ROTHSCHILD. Mr. HUGH SCOTT. Dr. A. KE. SHIPLEY, F.R8. Sm STEWART STOCKMAN. Mr. F. V. THEOBALD. Mr. C. WARBURTON. Director. Dr. GUY A. K. MARSHALL... Assistant Director. Mr. 8. A. NEAVE. Secretary. Mr. A. C. C. PARKINSON. IMPERIAL BUREAU OF ENTOMOLOGY, BULLETIN OF ENTOMOLOGICAL RESEARCH. Vou. VIII. 1917, OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. By Ropert Newsteap, F.R.S., The School of Tropical Medicine, the Unwersity, Inverpool. Liaveia abrahami, sp. nov. Female, adult. Ovate, strongly gibbose above and flattened beneath; segmen- tation both on the dorsum and venter distinct. Colour, in alcohol, buff-yellow ; with a thin deposit of white secretion in the hollows between the segments; legs and antennae yellowish brown. It is highly probable that much of the secretion had disappeared in the alcohol in which the specimens were preserved. Antennae of eleven segments, 2nd and 11th longest ; 3rd a little shorter than the latter, but much longer than the succeeding ones, which are submoniliform in shape; average length 0-8-0'°9 mm. Legs relatively short; tarsi a little more than half the length of the tibiae ; front pair shortest, measuring 0°85 mm. ; posterior pair longer, measuring 1-1-1 mm. ; hairs relatively long and slender ; claws simple, though in one example two of these organs have a well-marked tooth-like projection dorsally, just above the curved tip. Derm both dorsally and ventrally rather thickly set with very short fine hairs ; margin with a narrow band of long fine hairs, the longest of which measure 0-6-0'7 mm. Pores dorso-ventrally relatively very small, circular and with beaded rims ; these are less numerous than the fine hairs; those at the margins are slightly larger, much more numerous, and many have clear triangular-shaped centres. Anal segment of abdomen with three large ovate rings, the central one being slightly the largest and measuring about 0°2 mm. in its longest axis. Length, after maceration in KOH, 8-9 mm. Larva. Very like those of the genus Icerya. Marginal hairs of great length, those in the abdominal region measuring 0°7-0°8 mm. There are four pairs of very long caudal hairs, the median pair slightly the longest, measuring 11-12 mm. “Antennae of six segments ; terminal segment with many hairs, of which six to seven are much the longest, the longest measuring 0°5 mm. Derm rather thickly set with long fine hairs and numerous pores similar to those in the adult but relatively larger. Legs C365) Wt. P.7/21. 1,000 8.17. B.&F.,Ltd. Gp. 11/1. 2 ROBERT NEWSTEAD. hairy ; digitules long and slightly dilated, upper pairs wanting ; claws slender, with a distinct ventral tooth near the tip. Eyes prominent and somewhat hemispherical. Length, 0-9-1 mm. BritisH Guiana: Issororo, N.W. District ; “ inhabiting indentations in the bark of a rubber-producing tree (Sapwwm jenmant); attended by ants which construct coverings over the Coccids,” 3.vi.15 (A. A. Abraham, per G. E. Bodkin). Differs from L. primitiva (Towns.) in its larger size, the number of antennal seg- ments, and the absence of spines on the legs. Liaveia primitiva var. pimentae, nov. Female, old adult. Dried example (1 only) elongate-ovate, highly convex dorsally, hollow ventrally, the latter character almost as marked as in certain species of the genus Lecanium; segmentation of the abdominal region fairly distinct above; the cephalo-thoracic region with a faint median keel, not extending to the front margin. The whole of the body clothed with a dusky white, farinaceous secretion, which is denser in the hollows of the abdominal segments, the irregular depressions of the thoracic area, and the hollow ventral surface; the last-named also containing a small quantity of pure white flocculent matter, in which were found two dead larvae. Antennae of nine segments ; formula: 9, 1 (or 1, 9), 3, 2 (or 2, 3) (4, 5) (6, 7, 8), all the segments longer than wide; length twice that of the anterior femur plus tro- chanter, or slightly more. Legs (anterior pair) relatively short ; femora incrassate, equal in length to the upper portion of the tibiae, the latter nearly one and a half times the length of the tarsi and furnished ventrally with nine to ten long slender bristles ; claws smooth on the under surface. Integument dorso-ventrally rather densely clothed with long stout bristles, and, almost as thickly beset with circular “rosette? pores ; marginal hairs much longer and denser than those seen elsewhere, more especially so are they on the abdominal segments. Venter, near the distal extremity, with three large oval rings. Length (after maceration in KOH), 5:5 mm. Female, young adult. Similar to the preceding, but with the marginal hairs very long, the longest of them equalling the length of the antennae. The only example has the antennae asymmetrical: one of ten segments with a partial division of the 4th, the other of seven segments only. Female, second stage. Shape similar to that of the old adult, but less than half the size. Dorsum with or without a median ridge; completely covered with a dense, pale yellowish-white secretion; segmentation faintly indicated. Antennae variable, and in two examples asymmetrical, of eight or nine segments. Integument much more densely hairy than in the adult; marginal hairs as in the young adult female. Length, 2°8-3:2 mm. Larva (fully developed). Short ovate. Marginal hairs long and continuous in front ; the longest of those on the cephalic margin longer than the antennae. Caudal hairs in four pairs, of which the median pair is the shortest and less robust than the others, the longest hairs three-fourths the length of the body. Antennae with seven long hairs, the longest of which is nearly equal in length to the antennae. In the embryo larvae, taken from the body of the parent, the integument appears more OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 3 densely hairy than in the fully developed larva, owing apparently to the somewhat contracted nature of the integument. This applies also to the pores in the two stages respectively. JaMAIcA: on Pimento (Pimenta officinalis, Lindl.), 1916; ‘attended by the so-called ‘ stinking-ant,’ Cremastogaster sp., and to be found under the loose bark ; not of great importance.” (A. H. Ritchie). This may possibly prove to be Townsend’s Llaveia primitiva, the slight differences being possibly due to the age of the individuals. Probably the normal number of antennal segments will prove to be ten. More material is needed, Monophlebus ?hirtus, Brain. Brain* in his brief description of Monophlebus hirtus states that ‘‘ the dermis is closely crowded with slightly clubbed glandular hairs and occasional long hairs, the latter fitting into thickened sockets,” these characters being the salient features of the species. In the only specimen I have before me, the derm, more especially at the margin, is crowded with stout, blunt, curved spines (fig. 1, 6), which, though ™& Ses ee o; —=* 4 Tene “Ss gi AS A ORAV da Fig 1. Monophlebus ? hirtus, Brain, @¢; marginal spines and hairs. slightly narrowed proximally, can scarcely be said to be club-shaped. Before my example was passed through the various reagents and mounted in Canada balsam, the margin was seen to be broadly set with long and very stiff hairs (fig. 1, a); but these were so easilyjdeciduous and brittle that the majority of them broke away in the preparation. Length, 7:2 mm. NyasaLANnD: Mt. Mlanje, 1913 (S. A. Neave). * Trans. R. Sac. 8. Africa, v, p. 165 (1915). (C365) 4 - ROBERT NEWSTEAD. Aspidoproctus neavei, sp. nov. Female, adult. Ovate, sides elevated and more or less concave; dorsum in front relatively low; abdominal region highly convex. Cephalic or frontal depression well marked. Marginal dentate appendages long, rectangular, widely separated and ivory-like in colour and texture; there are about 25 of these altogether. Sub- marginal series of processes on the anterior half of the body (four to five in number) not quite so long as the marginal ones, but more robust and often irregularly dentate ; the abdominal series, in continuation, diminishing in size distally. In the median line in front is a pair of small processes and behind them two to three large single ones. Midway between the median and submarginal processes are three very deep pits or depressions. Cavities and hollows coated with creamy white, granular secretion. Colour pale to dark castaneous; dull orange-crimson in alcohol ; lower Fig. 2. Aspidoproctus neavei, Newst., sp. n., 9; a, spines of gland tracts; b, marginal spines of lip of marsupial opening; c¢, pores; d, pores at angle of marsupial opening. surface paler. Secretionary operculum relatively small compared with the size of the insect ; it is normal in shape and of a dark brown colour. Antennae of ten segments ; 3rd to 7th inclusive broader than long; 9th and 10th much the narrowest, being about half the width of the 6th, the 10th equal in length to the 6th, 7th and 8th, inclusive ; 2nd nearly twice the length of the 4th; formula: 10, 1, 2, 3, 9 (6, 7, 8), (4, 5); two kinds of hairs present, subspinose forms and simpler slender ones; the longest of the latter equal in length to the terminal segment, and measuring 1°75 mm. Spines of the gland-tracts (fig. 2, a) incrassated proximally, apex bluntly pointed or sometimes slightly dilated ; those of the proximal lip of the marsupial opening (fig. 2, b) generally curved and more or less spathuliform; elsewhere, both dorsally and ventrally, the spines are simple and more or less acutely pointed. Glandular OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 5 pores (fig. 2, c) circular ; some apparently with compound openings others with a tri- angular, quadrangular or circular openings. Length, 15°5-21 mm.; width, 12-15 mm.; height, 6-9 mm. Female, second stage. Ovate, sometimes slightly narrowed anteriorly ; posterior half of margin rather broadly produced or flattened ; dorsum convex. Marginal dentate processes similar to those in the adult but smaller. Dorsum without dentate processes and, with the exception of the rubbed portions, covered with granular secretion. Dorsal glandular pores circular, with four openings. Glandular pores at the angles of the marsupial opening (fig. 2, d) of three types; a relatively small form of somewhat variable shape, a large form with six openings, and a roughly quadrate form with four openings. These glandular orifices in the old adults become more or less obscure owing to the density of the chitin. NyAsALAND: Mt. Mlanje, on “ Mwange ” tree, x. 1913 (S. A. Neave). This Coccid differs from A. pertinax, Newst., in being less convex (especially in the abdominal region), and in the larger size and arrangement of the dorsal thoracic processes. Aspidoproctus verrucosus, sp. nov. Female, adult. Subhemispherical, produced in front, flat beneath ; secretionary operculum complete. Integument blackish; sparsely clothed with dusky golden pubescence, and punctate between the patches of secretion. Margin with short, tooth-like, contiguous appendages, dusky white in colour, with here and there faint traces of yellow, tips more or less pure white. Frons with two converging rows of similar, but smaller and confluent, processes. Above the last-named in the middle line are three widely separated, transverse rows of very small tooth-like appendages surrounded by dusky granular secretion. Leading from the third group towards the anal margin is a narrowly ovate series of scurfy patches and surrounding them an almost complete circle of similar patches of secretion. Between the various tooth-like processes are other irregular patches of secretion, which collectively give the insect a decidely warty or more or less mottled appearance. Derm rather densely clothed with relatively long spines, some of which are truncated, possibly owing to injury. The special gland-tracts at the margin of the abdominal area surrounded by a dense wall of spines taking a quadrangular or rectangular form ; within these are one or two small groups of circular glands. There are at Jeast three other types of glands on the venter, including a large quoit-shaped one of slightly varying dimen- sions, these being most conspicuous in the region of legs 1. and in. Length, 85; width, 7°3; height, 5°5 mm. Ucanpa: Ngamba Is., on trunk of a fig tree, 16. v.14 (Dr. G. D. H. Carpenter). In its general shape this Coccid resembles a very small example of Aspidoproctus pertinax, Newst.; but it is otherwise markedly distinct. The mouth-parts and antennae were absent in all the specimens, and were evidently detached when removing the insects from the tree. Palaeococcus bicolor, sp. nov. Female, adult. Ovate, narrowed in front, covered with short bright yellow or white waxen processes, which are arranged in seven more or less well-defined rows : 6 ROBERT NEWSTEAD. one short, median, and three on each side, the lateral rows continuous behind ; intervening spaces with somewhat granular wax; margins with outstanding hairs, which appear more numerous and much longer posteriorly. Venter, in the oldest examples, with a relatively thin layer of white wax, but this, in the specimens submitted, does not form a distinct ovisac or pad. Antennae of nine to ten segments; in two instances they were asymmetrical; terminal segment not quite so long as the two preceding ones together; there is a long slender spine on the second, and long stout hairs on the succeeding ones, in addition to which there are two to three very long slender hairs on the terminal segment, one of which is equal in length to the last two segments together. Legs very robust and long. Derm markedly hirsute; spinnerets large, circular and almost as numerous as the hairs. Stigmata relatively small. Margins with immensely long stout hairs, the longest being three times the length of the antennae. Many of the marginal hairs are dilated and frayed distally and sometimes present three distinct, lateral ridges, resembling somewhat the proximal half of the scapula of a mammal in miniature; these are clearly malformations due, possibly, to injury by abrasion. Length, 5-5 mm.-6°5 mm. Female, second stage. Waxen covering similar to that of the adult, but the median and submedian rows are not so pronounced; colour bright yellow, or rarely pure white. Antennae of seven segments, of which the third and seventh are the longest. Derm sparsely hirsute; pores much fewer than in the adult. Marginal hairs of immense length, the longest and stoutest arising from the abdominal segments; these are, when perfect, four times the length of the antennae, or slightly longer than the entire length of the insect. Length, 2-2-3 mm. Larva. Antennae with the terminal segment markedly incrassate, with five immensely long, stout, lateral hairs; the largest about twice the length of the antenna. Derm thickly set with fine long hairs; margins with similar but slightly longer hairs ; terminal segment of abdomen with three pairs of long hairs, of which the median pair is about half the length of the others; the two pairs of longest hairs three times the length of the antenna, or about one-fourth longer than the body. Gotp Coast: Aburi, on Thespesia sp. (W. H. Patterson). The general appearance and arrangement of the cereous coverings of the old adults resemble somewhat those of the young adult @ of Icerya seychellarum, Westw. ; but the strongly hirsute character and great length of the marginal hairs may readily serve to distinguish it. That the colour of the waxy covering should vary between bright yellow and white is rather remarkable, but the former, so far as one can judge from the few examples at hand, is the predominant colour. There is no admixture of the two colours in the same individual; all are either entirely white or entirely bright yellow. Palaeococcus caudatus, sp. nov. Female, adult. Apparently without any trace of an ovisac; dorsum covered with short stout waxen processes, white or dirty white in colour and thickly felted in texture; these processes are arranged in seven rows: the median row much the broadest and coalescing, more or less; the other rows almost uniform in size OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 7 and continuous in front, so that they appear concentric with the median row; the marginal row almost hidden by the submarginal row. A number of long hairs project beyond the waxen covering, especially at the margin in front. Posterior extremity markedly narrowed and with a single long stout waxen appendage or tail-like process. Integument, antennae and legs of dead examples pitchy-black. Antennae of eleven segments and of the usual form found in the females of this genus. Legs normal. Marginal hairs very long, the longest being a little more than one-third the total length of the body ; these hairs are much more numerous in front and also posteriorly ; at the sides of the body they are arranged in rather widely separated groups. Integument densely studded with relatively large circular spinnerets, interspersed with fine short hairs, the latter much more numerous ventrally. Length, inclusive of waxen appendages, 7-8 mm. ; length, exclusive of caudal appendage, 5 mm. Female, second stage. Dorsum with five rows of white waxen appendages ; median row coalescing ; submedian and marginal rows much shorter than in the adult female; caudal process similar to that in the adult. Antennae of nine segments; marginal hairs and spinnerets as in the adult, but the former are relatively longer, being about one-half the total length of the body. Ueanpda: Entebbe, on crotons, 13.vii.12 (C. C. Gowdey). The somewhat pyriform outline of the female, with the long stout caudal appendage, should readily serve to distinguish this insect from any of its allies. Palaeococcus cajani, sp. nov. Female, adult. Ovate and highly convex. Waxy covering pure white, the arrangement very like that of Icerya euphorbiae, Brain,* but more or less distinct processes are traceable; these are arranged as follows: margin with 13-14 short and bluntly rounded processes packed closely together and coalescing behind ; immediately above them is a well defined constriction, and over this a ridge with similar but more irregular processes ; cephalic extremity with an irregular cluster of blunt processes; the rest of the dorsum denuded by pressure in packing. Colour of integument black. A number of fine hairs protrude through the waxen covering. No ovisac, but a little flocculent secretion beneath the body. Length of dried example, 3°9 mm. When first placed in chloroform the marginal appendages show up very distinctly and the outstanding hairs are regularly arranged so that a pair of them (one dorsal, the other ventral) form the dividing line between the waxen processes. Colour in this medium, dull orange-crimson. Antennae (fig. 3, @) of nine segments; the terminal equal in length to the second and third together ; all, with the exception of the last two, broader than long; hairs of two kinds: a relatively short form, which is somewhat spinose, and a very long slender one; the longest on the last three segments, one of which is twice the length of the distal segment. The antenna on the opposite side has the articulations of the third and fourth segments very faintly indicated. Legs with the femora very broad; tibiae and tarsi with slender hairs of varying lengths. Derm scantily clothed with hairs; pores (fig. 3, 0) * Trans. R. Soc. 8. Africa, V, p. 167, pl. xxu, fig. 48. 8 ROBERT NEWSTEAD. circular, presenting the characteristics shown in the illustrations according to the plane in which they are focussed. Marginal hairs very long; one of the longest, just behind the antennae, about one-fourth longer than the latter, or about half “ Fig. 3. Palaeococcus cajani, Newst., sp. n., 2; a, antenna: b, gland pores ; ¢, base of dermal hair. the width of the body; these structures are readily deciduous and very few are left intact in mounted material. S. Nigerta: Agege, on pigeon pea, 1914 (Dr. W. A. Lamborn). Icerya nigroareolata, sp. nov. Female, adult. Secretionary covering and appendages very like those of Icerya aegyptiaca, Douglas, but the flocculent matter beneath the appéndages is thicker. When macerated in KOH the females assume a very elongate form, thus differing materially from J. aegyptiaca, which is broadly ovate. Antennae of eleven segments ; fourth the shortest and broader than long; eleventh longest and twice the length of the tenth ; hairs very long and for the most part arranged in whorls, the longest hairs (terminals) being slightly longer than the two last segments together; formula A1.(2,, 3),.(7,85,95 10), 4ky,4,( 0841 {2,,,3, 7, 85;9,,10), Ligds prdaewe roluene hairs on the upper surface of the tibiae very long, the longest being equal in length to the tarsi (fig. 4, a); digitules simple. Marginal hairs relatively short, the longest arising from between the antennae and also at the anal extremity; some of the former about half the length of the antennae. Derm rather sparsely hirsute. Pores or spinnerets (fig. 4, b) of two types: one large, with an inner hexagonal OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 9 border ; the other less than half the diameter of the former, with a beaded margin ; the large ones are generally distributed over the dorsum and also form large triangular groups along the margin of the abdomen. Venter with three large vaginal discs or areoles, the laterals (fig. 4, c) reniform, the median one ovate, with a faint central transverse division or construction; these, if maceration is not carried too far, are seen to be completely covered with opaque black pigment ; when the pigment is partly removed a finely reticulated surface is visible, but when completely removed the reticulation may disappear almost completely. Length of macerated specimens, 5-6 mm. ; width, 2°50-3 mm. Larva. Very elongate. Gland pores or spinnerets as in the adult; arranged in regular transverse rows, the larger ones preponderating, the smaller ones confined chiefly to the median line. Terminal segment of the antennae with three long hairs, the longest being one-fourth longer than the antennae; the remaining hairs are also very long, but less than half the length of the longest. Marginal — ——— 2) "4 Pe % a Fig. 4. * Icerya nigroareolata, Newst., sp. n., 9.; a. tarsus; b, gland pores; ¢, lateral areole, with portion of pigmented surface. hairs very long, the longest of the abdominal ones nearly as long as the antennae ; the long caudal hairs six in number, three on either side of the anal orifice; two are very long, being almost as long as the body, the third a little more than half the length of the others. Length, 0°9-1 mm. Length of antennae, 0°'4 mm. Ucanpa: Kampala, on coffee, 4.xi.13 (C. C. Gowdey); Jinja, on croton, 1914 (Dr. R. Van Someren). icerya sulfurea var. pattersoni, nov. Female, adult. Waxy covering bright pale yellow; covering of the dorsum imperfect, but that which remains is suggestive of short stout plates; marginal series of waxen appendages long, broad, curved downwards and slightly curled distally. Ovisac well defined ; colour, white or white with a faint tinge of yellow. Derm faintly hirsute, but thickly studded with minute hairs interspersed with long slender ones. Margins with irregular groups of Jong outstanding hairs, the 10 ROBERT NEWSTEAD. longest varying between one-third and one-half the length of the antenna. Pores circular and very numerous; these are of two sizes: the larger (fig. 5, a) bemg twice the diameter of the others. Antennae (fig. 5, 5) of eleven segments ; terminal segment much the longest but scarcely wider than the preceding one. lntie p = Fig. 5. Icerya sulfurea var. pattersoni, Newst., nov., 9; a, pore; b, antenna; ¢, leg. Legs (fig. 5, c) long; femur very short; claw faintly bidentate, digitules simple hairs; underside of tibia very hairy. Eyes truncate. Stigmata large, slightly longer than the tarsus. Gotp Coast: Aburi, on Tectona sp. (W. H. Patterson). The form and arrangement of the waxen appendages are very like those in Icerya aegyptiaca, Doug., but the bright pale yellow colour is distinctive and the structural details differ. Margarodes buxtoni, sp. nov. Female, adult. Form rather elongated and very tumid. Derm almost completely covered with a layer of white, mealy secretion. Colour in alcohol, dull brownish- crimson. After maceration in KOH the body assumes a broadly ovate form (fig. 6, a). Derm, on both surfaces, finely but markedly pubescent, the individual hairs relatively long and exceedingly slender. Pores (fig. 6, f) circular, with a central opening and numerous minute ones surrounding it; these organs are almost as numerous as the hairs, more especialky so on the terminal segments of the abdomen. Antennae (fig. 6, b) short, stout, and widest proximally, composed of eight segments, each (with the exception of the last) with a well-defined chitinous band; distal segment with several minute stout spines, and two groups of long fine slender hairs; the remaining segments are apparently nude, but of this one cannot be quite certain, as the innumerable hairs on the surrounding integument are lying superimposed over a great portion of the antennal segments. Anterior leg (fig. 6, c) large, stout, and more than twice the size of the other limbs; claw densely chitinised and distinctly articulated with the exceedingly short tarsal segment; tibio-tarsal articulation well defined, and immediately below this is a OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. ll distinct group or tuft of long fine hairs. Posterior legs (fig. 6, d) not differing materially from the intermediate pair; claws distinctly articulated; digitules represented by slender hairs. Eyes well developed, truncate, and placed considerably below the antennae (see fig. 6, a). Thoracic stigmata (fig. 6, e) large and somewhat reniform in outline; outer lateral lobe-like extension with a group of circular pores; abdominal stigmata present on the first and second segments only ; these organs are represented by minute circular orifices, external to which is a roughly horse-shoe-shaped thickening of the body wall. Anal orifice circular ; Ws ee > Ny Y YZ) Yj a —L7 © Zz VLE xe Fe < e leh 77 AS ave QIAN 4IZEZZET 1s ; a \ LAA 11, \ ) Ny \ , "\\) Sea ZG LL LPL e 7 UVEN SN a » gill $ RALOSSSN Fig. 6. _Margarodes buxtoni, Newst., sp. n., 2; a, ventral view of a specimen cleared in potash; b, antenna; c¢, front leg; d, posterior leg; e, stigma; f, pore. ring strongly chitinised, with a disc-like structure at the distal margin. Vaginal orifice nearly opposite the analfopening. Length, 57 mm.; width, 43 mm. ; length of front legs, 1 mm. Male (fig. 7, a). Colour of dried specimens dark piceous ; when placed in potash, rich dark crimson. Wings hyaline; costa pale crimson, darkening distally ; veins, of which there are traces of three, faintly indicated ; outline variable in the two examples before me; in one the wing is slightly longer and narrower than the other. Subterminal segment of the abdomen with a dense tuft or pencil of fine white silky filaments. Antennae (fig. 7, b) of nine segments, and finely pubescent ; distal segment 412 ROBERT NEWSTEAD. much the longest, and furnished near the tip with a minute circular sensorium ; articulations of the last four segments with relatively broad and clear integument. Hyes large, compound ; a single prominent ocellus on the proximal edge of each of the compound eyes. Sculpturing of thorax doubtful, but apparently with prominent rounded gibbosities. Anterior legs (fig. 7,d) much shorter and more markedly robust Fig. 7. Margarodes baxtoni, Newst., sp. n., ¢; a, dorsal view, from a dried specimen; b, head and antenna; c, hind leg; d, front leg; . . . a €é, pygidium, lateral view. than the others; tarsus and claw united, the latter strong, stout and densely chiti- nised ; femur, tibia and tarsus with fine hairs. Posterior leg (fig. 7, c) well developed ; the tibia a little more than twice the length of the tarsus ; claw distinctly articulated ; femur, tibia and tarsus clothed with fine hairs. Terminal segment (fig. 7, e) broad and somewhat quadrate, with a distinct median angular projection ; intromittent organ slender and cylindrical. Owing to the density of the chitin the special groups of i ; \/ } i OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—Iy. 13 glands from which the tuft of silky filaments arise are not traceable. Length, inclusive of the anal tuft of filaments, 4°5 mm.; exclusive of the filaments, 2°5 mm. AtGcERIA: El Kantora, 13.iv.13, 19, 2¢¢ (P. A. Buzton). Mr. E. E. Green, to whom I forwarded the above description and drawings, con- siders that “‘ this Margarodes is undoubtedly a new species. In its hairy integument it approaches my M. niger, but that species has limbs of a very different form.” Stictococcus gowdeyi, Newst. SouTHERN NicERIA: Agege; on young shoots of cacao, invariably protected by the ant Oecophylla, 1915 (Dr. W. A. Lamborn). These examples differ from the type lot of specimens by being of a rich dark translucent crimson or of a translucent olivaceous yellow, sometimes with a trace of crimson. They are, in all probability, young adults which had not acquired the permanent colour of more mature examples. 7 Stictococcus intermedius sp. nov. Female, adult. General facies as in L. sjéstedti, Ckll. Pseudo-margin with a regular series of short, deep, sharply defined and slightly radial grooves, giving the extreme edge a finely crenulated appearance, as if marked by a roulette; midway between Fig. 8. Stictococcus intermedius, Newst., sp. n.; marginal spines of: a, adult 9; b, second stage 2; e, larva at period of ecdysis. the median line and the roulette border is an irregular row of deep, clearly defined pits—seven to eight on either side. Colour pale to dark castaneous or piceous, with the integument polished. Marginal spines (fig. 8, a) of two distinct forms: (1) broad and irregularly digitate, and (2) very long, somewhat contorted, broad and 14 ROBERT NEWSTEAD. flat, geniculated, and finely spinose ; just within the margin is a series of long, slender, ~ simple spines. The young adults are reddish buff in colour, with the pits and hollows of the roulette border piceous or dark castaneous. Length, 3-4 mm. Female, second stage. Short ovate, flat, with the sides slightly raised. Colour pale reddish-buff. Marginal spines (fig. 8, b) broadly digitate and similar to those in the adult, but usually with one of the lateral projections more pronounced than the rest ; and a broad, flat, falciform and spinose one alternating more or less with the others. Larva (at period of ecdysis) with the marginal spines as shown in fig. 8, c. Gotp Coast: Aburi, on cacao, ? 1913 (W. H. Patterson). Separable from S. sjéstedti, Ckll., by the form of the marginal spines (in all stages) and by the sculpturing of the dorsum in the adult female. Stictococcus multispinosus, Newst. Male puparium. Like that of Kermes quercus (Linn.) ; a white, felted, elongated sac, and very brittle. Two examples were found wedged in between the old adult females. Neither was quite perfect ; one contained a propupa, the other was empty. Fig. 9. Stictococcus multispinosus, Newst., sp. 0. : a, terminal segment of g; b, head of g; ec, marginal spines of larva. OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 15 Propupa. Turned to dull crimson in cold KOH. Very robust. Frons with one pair of bristles. Genital sheath short, obtuse. Abdominal segments with a few stiff hairs. Length, 0°38 mm. S. Nicer1a: Agege, on pigeon-pea stems, 1914 (Dr. W. A. Lamborn). Male (imperfect). Head slightly longer than broad; ventral surface (fig. 9, b) with two divergent narrow sclerites, and four pairs of bristles, which are separated by a median longitudinal suture, the distal pair on the frontal margin. One pair of ocelli near the apex formed by the sclerites. Thoracic apodeme normal. Legs long, sparsely clothed with bristles; claws rather strongly falcate ; lower digitules with slightly longer knobs than those of the tarsii Abdomen very elongate; the segments with submedian and lateral bristles, the terminal segment with a pair of very long marginal bristles. Genital armature (fig. 9, a) with three pairs of long bristles ; stylus in two parts, with the tip of one bluntly rounded, the other truncate. Length, 1:3 mm. Larva. Marginal spines (fig. 9, c) of two forms ; minute club shaped ones, alter- nating with immensely long, stout, curved and barbed ones. Ucanpa: Kampala, on Markhamia platycalyx, 2.1x.15 (C. C. Gowdey). Asterolecanium spectabile sp. nov. Female puparium. Lanceolate and convex dorso-ventrally ; margins broadly flattened, with slight constrictions opposite the stigmata; slightly narrowed and produced posteriorly. Surface strongly punctate; a definite narrow median suture extending nearly the whole length of the puparium, the suture faintly interrupted at iegular intervals along the abdominal region, suggesting segmentation. Colour brilliant orange-crimson, with black suffused markings anteriorly, the latter due to the under-lying shrivelled body of the female; the narrowed posterior end opaque, glassy and colourless ; the cephalic margin narrowly orange-crimson. Fringe com- posed of closely adjacent, straight, glossy filaments of a pinkish colour; in very young individuals this lies flat upon the surface of the leaf, but as the ventral portion of the test gets more and more convex (exteriorly), the fringe drops and becomes vertical ; as a rule, it is either entirely wanting or can be traced only in isolated and abraded patches. Length, 1-3-2 mm. Female, adult. Form somewhat doubtful, but clearly very elongate and probably filling the test at gestation. Rostrum approximately central. Paired or 8-shaped pores disposed in a single row, not extending to the posterior extremity or the extreme cephalic margin. A short series of small circular glands on the ventra margin opposite the thoracic stigmata only. Tubular glands geniculated, with the proximal end attenuated; these are distributed over the whole of the dorsal area; their length equal to the greatest diameter of the rudimentary antennae, approximately. Abdominal extremity with two minute setiferous lobes; setae short and with their tips broken away in all the preparations (eleven). Anal ring hairless, its diameter not much greater than the basal attachments of the setae of the lobes. Genital orifice relatively very large, transverse, and placed well away from the anal orifice. Rudimentary antennae with three to four extremely minute spines. 16 ROBERT NEWSTEAD. Male puparium. Similar to that of the female, but slightly narrower; the ends equally rounded. Colour as in female. Length, 15; width, 0°5. The puparia of the young females (2nd stage) are amber-yellow in colour. Mauritius; Botanic Gardens, on palm trees, 1915 (Z. d Emmerez). A really handsome species and the only one known to me which has such brilliantly coloured puparia. Lecaniodiaspis tarsalis, sp. nov. Female test. Colour warm buff; narrowed slightly posteriorly ; dorsum convex, with a median interrupted longitudinal ridge, and about twelve transverse ones on each side. Orifice terminal, circular; projecting from it, in some individuals, is a short white waxen filament. Texture dense; surface with exceedingly minute whitish particles. Length, 2-22 mm.; width, 1:2-1°3 mm. Female, adult. Legs (fig. 10, a, a) well developed; anterior tarsi unusually long, with a deep and clearly defined dorsal constriction; tibiae relatively short, from Fig. 10. Lecaniodiaspis tarsalis, Newst., sp. n., 9; a, a, legs; 0b, antennae; c, marginal spine ; d, anterior stigmatic spine; e,, tubular glands; e,, cribriform and 8-shaped pores; /, anal segment. one-half to one-third the length of the tarsi; digitules long, those of the tarsi very faintly knobbed. Antennae (fig. 10, 6) normally curved, of nine segments; 8th with a short hair, 9th with several; relative length of individual segments varying slightly, 2nd and 3rd usually the longest, but sometimes the 6th is as long as the 3rd. Marginal spines (fig. 10,c) stout and rather bluntly pointed ; these are widely separated. Anterior stigmatic spines (fig. 10, d) two, both spathuliform, one-half the length of the other; posterior pair both short and very widely separated. Tubular glands (fig. 10, e,) generally distributed over the whole of the derm, and so numerous as OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 17 sometimes to touch or slightly overlap each other; about two-thirds the length of the marginal spines; 8-shaped pores generally distributed but most numerous towards the margin. Two divergent rows of cribriform plates (fig. 10, e,), four on each side of the median line; these are circular, with labryrinthiform gratings, the diameter about three times as great as the length of the 8-shaped glands. Anal segment (fig. 10, f) with rather prominent lobe-like extensions and a broad bilobed plate below the anal aperture, each with a pair of very short subapical spines and two long marginal hairs. Pre-anal plate with a very deep emargination. Anal ring with ten hairs. Larva. Elongate. Antennae stout, of six segments; 3rd and 6th longest, but both relatively short. Marginal spines minute. Anterior stigmatic spines similar to those of the adult; posterior pair absent. Legs robust; anterior tarsi much longer than the tibiae. The 8-shaped pores in well-defined longitudinal rows. Anal lobes similar to those of the adult, but the terminal bristles are much longer. Anal ring with six hairs. South Arrica: Pretoria, 1914 (H. d’Emmerez). This insect is related to Lecamiodiaspis africana, Newst.,* but is still more remark- able in having the legs normally developed. It differs moreover in its smaller size, the presence of marginal and stigmatic spines, the extraordinarily long tarsi, the antennal formulae, the character of the “ cribriform plates,” and in other details. Phenacoccus ballardi sp. nov. Female, adult. Dorsum, in dried example, covered with a felted mass of white secretion, completely obscuring the segmentation ; margin with a conspicuous fringe of extremely long, white appendages, which are shortest in front; those arising from the cephalic margin about three times the length of the felted dorsal area. Length, inclusive of the fringe, 10 mm. Form, after maceration in potash, slightly elongate-ovate. Antennae of nine segments, the articulation of the 3rd and 4th less pronounced than the rest. Hyes small, but prominent. Legs slender, longer than the antennae; tarsal digitules simple bristles ; those of the claw very long and dilated ; two strong apical spines on all the tarsi. Margin with a complete series of inconspicuous spinose tubercles the spines short and truncate; between them and’ surrounding them a number of large circular pores ; two or three long hairs accompany each of the abdominal groups of spines. Anal lobes similar to the marginal ones. Anal ring with six hairs. Inte- gument with a few minute scattered spines and small circular pores, the latter most numerous on the terminal abdominal segments. Length of macerated specimens, 1-7-2 mm. Female, second stage. The marginal appendages similar in their arrangement to those in Pseudococcus (Dactylopius) longispinus (Targ.); but they are much longer and arranged radially; posterior pair longest. Form, after maceration, elongate. Antennae of six or seven segments. Marginal series of spinose tubercles not so clearly defined as in the adult. Integument with numerous minute spines ; and there are two or three very long hairs arising from near the anal spinose lobes. * Bull...Ent. Res., 11, p. 100, fig. 13 (1911). (C365) 4s 18 ROBERT NEWSTEAD. Very like P. eceryordes, Green (Mem. Dep. Agr. India, ii, p. 26) in having a con- spicuous fringe of white filaments; but in Green’s species the 2 has a “ fringe of short, stout, conical spines which extend into broad clusters on the abdominal segments”; whereas in P. gracilis the spines are truncate and their arrangement similar to those in P. insolitus, Green (l.c.), but in the latter spinose tubercles are present also on the dorsum. S. Inp1a: Coimbatore, on mango, 10.11.14; S. Kanara District, on an unnamed plant, xx.1913 (7. V. Ramakrishna, per E. Ballard). Tachardia bodkini, sp. nov. Female test. Obconical; centre with a bluntly pointed prominence, rarely with two; margin with low, blunt, and usually bifid processes, clearly the remnants of the rays of the test of the young female; surface smooth ; a nipple-like prominence over the anal opening. Colour bright subtranslucent orange-red, or in certain lights faintly orange-ruby. Texture very hard and brittle, similar to that of the lac of commerce (7. lacca). Average length, 5-4 mm.; height, 3°3 mm. Female, adult. Form doubtful; specimens macerated in KOH become broadly ovate when mounted under pressure. Antennae about equal in length to the greatest width of the anterior spiracles, composed of four ill-defined segments, with no trace of hairs. Post-anal spike about two and a half times the length of the antennae. Compound circumgenital glands or “ tubercles”? in four groups, each group con- sisting of over one hundred individual pores. Margin of flattened specimen with three pairs of large bilateral gland-tracts, the pores in each group of three distinct kinds: (qa) four or five relatively large ones, with strongly chitinised subcutaneous tubes, arranged more or less in line and in the centre; (b) minute, circular and rimless pores occupying a large and roughly ovate area; (c) a complete outer ring of circular chitinous pores, resembling the circumgenital glands in the Diasprnag, arranged on one side of the circle in a single row, on the opposite side in an irregular band, two or threedeep. The large pair of spiracles shaped somewhat like a cardinal’s hat, the irregular chitinous plate surrounding them beset with circular pores. “ Lac tubes ”’ circular, the chitinised rim convex, the greatest diameter about equal to the length of the antennae; interior with relatively large pores. Anal chitinous tube shagreened, 7.e., set with minute, closely adpressed spines; apical plates cuneiform or notched ; anal ring with (?) ten hairs. Female test, second stage. Five-rayed, the anterior and posterior rays longest ; central area with a nipple-like prominence. Colour slightly darker than that of the adult. Length, 18-2 mm. Male puparvum. Subcylindrical, each extremity with a pair of short lateral rays ; dorsum with a central nipple-like extension, on either side of which is a distinctly segmented, convex, median ridge. Length, 1°83 mm. BRITISH GUIANA: near Repos, Georgetown, on Sapium jenmani, xii.1915 (G4. E£. Bodkin). Froggatt* has described a species (7'. angulata) the test of which, he says, comes to a blunt point at the apex, its general appearance resembling a large blunt *Proc, Linn. Soc., N.S.W., xxxvi, 1911 OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 19 rose-thorn. This applies, in a general way, to Bodkin’s specimens; but the apical extension of the test in angulata is fluted, whereas in bodkini it is perfectly smooth. The structural characters of the two species are, however, quite distinct. I have pleasure in dedicating this species to its discoverer. Pulvinaria aristolochiae, sp. nov. Female, adult. More or less cordate in outline and rather flattened; dorsum almost completely covered with a well-defined layer of flake-like wax which varies in colour from dirty grey to greyish brown. Antennae (fig. 12, a) of eight segments ; 3rd, 4th and 5th unusually long, the two first-named markedly swollen distally ; 3rd about twice the length of the 4th; three long hairs on 2nd, one of which lies (in three examples) close up to the succeeding segment; there is also a long distal hair on the 3rd ; two on the 5th ; a single slender spine on the 6th, 7th and 8th, the last-named also with a few very short hairs. Legs (fig. 12, b) stout, long; tarsus relatively very short, less than one-third the length of the tibia ; lower digitules long eo Fig. 12. Pulvinaria aristolochiae, Newst., sp. n., 9; a, antenna; b, leg; c, marginal spines ; d, submarginal tubercle ; e, anal lobes. and very broadly spathuliform. Marginal spines (fig. 12, c) stout, pointed and placed rather closely together; stigmatic spines broken away in all the specimens, their points of attachment being continuous with the marginal series. Submarginal pores (fig. 12, d) very large, continuous, but rather widely separated. Anal lobes (fig. 12, e) with four stout spines near the apex, on the inner edge. Anal cleft short, usually a little less than one-sixth the entire length of the body. Anal ring with eight hairs. No derm cells present; but there are numerous circular spinnerets (? ventral), each having an inner concentric ring. Length, 5°7-7°6 mm. Ovisac pure white and closely felted, long and generally tortuous. Length, 10-20 mm. (C365) 20 ' ROBERT NEWSTEAD. Gotp Coast: Aburi, on Aristolochia sp., (W. H. Patterson). The affinities of this insect are with P. maxima, Green, and P. jacksont, Newst., but it differs from both in having pointed marginal spines, and from the latter species by the much denser and shorter ovisac. Pulvinaria elongata, sp. nov. Female, adult (fig. 13, e). Extremely elongate or elliptical, extremities more or less equally attenuated ; dorsum convex, cephalic region flattened. Colour, in life, varying from very pale crimson to bright rosy flesh-colour, with two more or less distinct, longitudinal lines of bright crimson, the latter frequently interrupted by a black wavy loop-like line, due apparently to a portion of the alimentary tract Fig. 13. Pulvinaria elongata, Newst., sp. n., 9; a, normal 8-segmented antenna; b, 7-segmented antenna; ¢, anal lobes; d, stigmatic cleft; e, adult 9. (2 Malpighian tubules). Dry cabinet specimens are pale buff, with a broad median suffused area of smoky brown or red-brown, and often with a well-defined median keel. The ovisac is very short indeed, projecting but slightly beyond the margin ; it extends anteriorly, usually as far as the region of the eyes. Antennae (fig. 13, a) normally of eight segments, of which the 3rd is slightly the longest; in two individuals the antennae are asymmetrical, one being composed of seven segments (fig. 13, 6) the other of eight ; in the former the 4th is twice the length of the 3rd, OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 21 but it shows a partial subdivision. Legs long and somewhat slender; tarsal digitules nearly as stout as those of the claw, the latter with disc-like terminals. Anal lobes (fig. 13, c) with four stout spines; one apical, the others arranged in a triangle; beneath each scale are four to five very long stout hairs, which are apparently attached to the eversible sac; their number and arrangement is somewhat exceptional. Abdominal hairs very long, in two pairs. Circumgenital pores numerous, extending beyond the anal plates. Anal ring with six unusually long, stout hairs. Stigmatic cleft (fig. 13, d) practically obsolete ; spines three, the middle one about twice the length of the laterals; margin with relatively long and very stout hairs. No derm cells present. Length, 35-6 mm. Larva. Narrowly elongate. Lateral margins with widely separated, stout hairs ; frons with nine to ten similar but much longer and stouter hairs or spines; these are packed closely together and give the insect a strikingly characteristic appearance. Anal lobes each with an immensely long hair. Stigmatic spines well developed, but the longest of the three is less than half the length of the marginal hairs. BritisH Guiana : Georgetown, “ on blade of sugar cane,” 12.11.13 (G. £. Bodkin). In its general external appearance this Coccid bears a rather striking resemblance to a member of the genus Aclerda, owing to its elongated form and the small amount of white secretion beneath the insect, but it has no real affinity with this genus, and although the ovisac is but slightly developed, it cannot be considered as generically distinct from Pulvinaria; neither can it be placed, so far as one can judge, in either of the allied genera Tectopulvinaria or Protopulvinaria. Pulvinaria ? flavicans, Mask. Female, adult. Dead examples shrivelled and resembling a shapeless mass of dirty, amber-coloured beeswax ; dorsum in some examples with detached fragments of glassy secretionary matter, with here and there particles of a whrte felted substance. Fig. 14. Pulvinaria ? flavicans, Mask., 9; a, b, antennae; c, stigmatic and marginal spines; d, leg; e, anal lobes. 22 ROBERT NEWSTEAD. Marginal spines visible under a two-inch objective. Ovisac very short, slightly overlapping the sides of the female, and most pronounced posteriorly. Specimens attacked by Chalcidid parasites are very elongate, highly convex and of a red-brown colour; no ovisac was formed by them. Length of 9 and ovisac, 2-33 mm. After treatment in KOH the form is more or less ovate or broadly pyriform. Antennae (fig. 14, a) normally of eight segments, rarely of seven ; in the latter case the 3rd and 4th are either completely fused or only partly so (fig. 14, b); hairs on the terminal segments relatively short. Marginal spines (fig. 14, c) long, stout and pointed ; they are placed closely together, the space between them being much less than their length, more especially so are they on the frons and the distal margin. Stigmatic clefts obsolete ; stigmatic spines (fig. 14, c) usually six in number, of which two pairs are much larger than the others, broad, curved and flattened. First pair of legs (fig. 14, d) with the tarsus bearing a well-defined dorsal constriction. Anal lobes (fig. 14, e) with the distal portion thickened, convex, and with many fine hairs; beneath the tips are two pairs of long slender hairs, one pair being a little more than half the length of the lobe. Anal cleft slightly less than one-fifth the length of . the insect. Anal ring with ten hairs. Circumgenital glands small, but very clearly defined. Venter, just in advance of lobes, with two pairs of hairs, one pair much longer than the other. No derm glands or pores traceable. Length, 2-25 mm. ; width, 1°5-2°6 mm. Britisa Guiana: Rockstone, on “ blood-wood ” plant, 30.xii.13 (G. #. Bodkin). A species of ant had built little shelters or “‘ tents” over some of the females. Pulvinaria subterranea, sp. nov. Ovisac. White and felted and, when complete, very elongate, with traces on the proximal portion of 3-4 longitudinal groves; incomplete forms more robust and ' decidely more convex. Length of elongated forms, 35-45 mm.; short forms, 2—2°5 mm. Fig. 15. Pulvinaria subterranea, Newst., sp. n., 9; a, a, antennae ; b, stigmatic cleft and marginal spines ; ¢, anal lobe. OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 23 Female, adult. Tilted and shrivelled at gestation; after treatment in KOH broadly ovate. Colourpale brown. Antennae (fig. 15, a, a) usually of eight segments, rarely of seven; in the former case, the 4th is shorter than the 5th, and the 3rd the longest ; in the example with seven segments the 4th, 5th and 6th are equal, the 3rd and the 6th are also of equal length and the longest. Legs very robust ; lower digitules stout and strongly dilated. Stigmatic clefts (fig. 15, b) barely indicated ; spines three, the laterals about half the length of the central one and much the stoutest. Marginal spines (fig. 15, 6) long and for the most part faintly frayed distally ; those at the anal margin slightly the longest ; they are somewhat irregularly arranged and in places are two deep. Anal cleft about two and one-fourth times as long as the lobes; the latter (fig. 15, c) somewhat variable, with one apical and generally two subapical bristles. Derm glands scarcely traceable, even in well stained specimens, and then only in one or two places as small and approximately oval spots. Length, 15-2 mm.; width, 0:95-1°5. Uaanpa: Entebbe, on roots of Chrysanthemum, 10.i11.14 (C. C. Gowdey). Pulvinaria africana, sp. nov. Female, adult.—More or less ovate, narrowed anteriorly. Colour (in formol) dull amber-yellow ; surface of many individuals with a faint black network formed by the mycelium of a fungus. Ovisac closely felted, elongate, with the sides more or less parallel, straight or curved; surface usually with four well-defined, rounded, longitudinal keels, and transverse conchoidal wrinkles. Antennae (fig. 16, @) normally >. Fig. 16. Pulvinaria africana, Newst., sp. n., 9; a, antenna; b, anal lobes ; c, stigmatic cleft and spines. of eight segments, of which the 3rd is slightly the longest; formula: 3 (2, 8) 1, (4, 5) (6, 7); m one example the antennae are asymmetrical :—one being normal, the other of six segments only, of which the 4th is much the longest. Legs somewhat robust ; anterior tarsi with a median constriction. Anal cleft short, about twice the length of the lobes or one-eighth the length of the body. Anal lobes (fig. 16, 6) 24 ROBERT NEWSTEAD. with four or five apical bristles. Stigmatic clefts (fig. 16, c) clearly defined but small; spines three, all of them stout and bluntly pointed, the central one curved and a little more than twice the length of the laterals. Marginal spines (fig. 16, c) in two irregular rows ; these are of two forms: one long and stout, with fine lateral hairs ; the other much shorter, usually simple, but here and there one may be found with a divided tip. Arising from near the insertion of each antenna is a group of three bristles, each successively longer than the other, the longest being nearly equal in length to the 2nd and 3rd segments of the antennae togerther. Derm cells irre- gularly oval or circular, but these are quite obscure in the majority of the specimens. Length, 1:8-2°2 mm. Length of ovisac, 35-4 mm.; width, 15 mm. (average). GoLp Coast: Accra, on guava, 2.vii.16 (Dr. J. W. Scott Macfie) ; heavy infesta- tion. Clearly this is not Pulvinaria psidi, Mask., as defined by either Maskell* or Green. It differs in a marked degree in the form of the ovisac, and also in the character of the marginal spines, which are not dilated and dentate distally and are moreover, so far as I can judge, relatively very much larger, more irregularly arranged and not nearly so uniform in length. It is also not specifically identical with the specimens of P. psidw, Mask., recorded by me in this Bulletin (Vol. i, p. 67, and Vol. i, p. 94). Many of the examples were attacked by a parasitic fungus, and they were also preyed upon by a predaceous Lepidopterous larva, of which there were several examples. A description of the cocoons of the latter is appended below. Larva with the whole of the upper surface protected by a tough silken cocoon, into which is incorporated the remains of the Coccid (Pulvinaria africana, Newst.) and its ovisac ; the bodies of the female Coccids often more or less intact ; the remains of the ovisacs are sometimes arranged in narrow spiral-like bands, or they may completely cover the semi-cocoon. Plant hairs and the mycelium of a fungus also present. Ventral surface of the larva nude. Cocoon, containing a parasitised pupa, complete, but thinner on the ventral surface than elsewhere. The general form is slightly more elongate than the semi-cocoon of the larva. Comminuted remains of the Coccid and its ovisac incorporated, as in the cocoon of the larva. Ceropiastes avicenniae, sp. nov. Female test. More or less hemispherical, smooth and more or less evenly rounded, with very faint traces of plates, apparently arranged as follows: one dorsal, one anterior, two bilateral and one anal; in each of the first three is a dark brown or almost black nuclear spot with a white centre; the anal plate with three similar spots placed closely together, the central one, formed by the tip of the caudal pro- cess, being without a white centre. ‘Two examples gave the following measurements : (1) length, 7° mm., diameter, 7 mm., height, 6 mm. ; (2) length, 7 mm., diameter, 6 mm., height, 6 mm. Female, adult. Semiglobular in shape, but slightly longer than broad ; integument highly chitinised but thin ; submarginal tubercles well developed and five in number ; * N.Z. Trans., XXV, p. 223 (1892). t Coceidae of Ceylon, p. 264 (1909). OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 25 one cephalic and two bilateral, each of the latter placed just above the stigmatic clefts; median dorsal tubercle prominent and keel-like. Caudal process long, stout, measuring 1'2-1°3 mm. long. Frontal margin slightly produced ; posterior margin, immediately beneath the caudal process, with a similar projection but wider and with a median division (anal cleft). Stigmatic clefts quite small and shallow. Antennae of eight segments, of which the 3rd, 5th, and 8th are the longest. Stig- ' matic spines minute, conical, arranged three to four deep, centrally, but diminishing in number bilaterally at the extreme margin. Legs normal, as far as can be traced. Gland pores of the integument simple, somewhat thinly and irregularly scattered. Marginal spines not traceable in the three females examined. British GuIANA: Mahaica Creek, on courida (Avicennia mtida), viii. 1916 (G. E. Bodkin). The old adults when denuded of the test are very like those of Ceroplastes ceriferus (Anderson), but the caudal process is slightly shorter and the number and arrange- ment of the submarginal tubercles differ in a marked degree. Only one antenna has been studied, and this, so far as one can judge, consists of eight segments, but the articulations are ill-defined and difficult to trace ; clearly however it is markedly different from that of C. ceriferus. Ceroplastes bipartitus, sp. nov. Female test. Colour, in dried specimens, very like pale dirty beeswax. In the young adults the test is broadly oval, somewhat hemispherical and divided into nine plates: three bilateral, one cephalic, one anal and one dorsal, the last-named with a conspicuous dark brown or blackish, oval spot, with a central elongated patch of pure white wax; the nuclear spots to the lateral plates are smaller and generally much less conspicuous than the dorsal one. Margin over the stigmatic areas with a pair of laterally compressed and somewhat disc-shaped extensions, each extension carrying on its edge a narrow strip of opaque white wax, the tip of which sometimes reaches the dark nuclear spot of the lateral thoracic plate. In very old examples. the test has increased in thickness considerably, but this has been so much damaged in transit as to render it useless for descriptive purposes; however, one can trace the curious marginal extensions, which are somewhat like a narrow-waisted and dis- torted bobbin, or the toy used in the once popular game “ diabolo.” Average length of young adults, 3 mm.; height, 16-2 mm.; average length of old adults, 6 mm.; height doubtful. Female, adult. Denuded of wax, hemispherical ; caudal process very long, varying in length from one-half to a little less than one-half the length of the remaining portion of the insect. Submarginal tubercles small, but generally clearly defined : one cephalic and three bilateral, the two over the stigmata slightly more pronounced than the rest. When examined under a high power lens, by transmitted light, these tubercles are seen to be traversed by clear cell-like tracts forming an irregular reticu- lated pattern. It is reasonable to assume, therefore, that these may be the special set of glands which secrete the nuclear spots in the centre of the plates in the test. Derm relatively thin, but strongly chitinised. Pores minute, separated over a large portion of the dorsum by slightly varying distances equalling the length of one of 26 ROBERT NEWSTEAD. the short segments of the antennae or two of them together. Stigmatic clefts (fig. 17, b) relatively shallow, but very clearly defined; spines short, obconical, those at the extreme margin very minute and stud-like. Marginal spines or hairs not trace- able. Antennae (fig. 17, a) of six segments, the 3rd longer than the last three together. Legs normal. Length of denuded female, inclusive of the caudal process, 45-4°6 mm.; length of caudal process, 1:3-1‘5 mm. Male puparium (fig. 17, e,e,). Consisting of two distinct parts; the lower half boat-shaped, and of a glassy, vesicular texture, as in those typical of the genus. Lecanium; the upper portion opaque, low, convex, and of a dirty beeswax colour, with nine narrowly rectangular, submarginal patches of snow-white secretion. Anal Fig. 17. Ceroplastes bipartitus, Newst., sp. n.° a, antennae of 9; b, stigmatic cleft of 9; ¢, derm of Q; e, e1, puparium of ¢, dorsal and lateral views. cleft apparently obsolete. On the emergence of the male the whole of the upper portion falls away, leaving the ventral half attached to the food-plant. The line of cleavage between the upper and lower portion is clearly defined in those puparia from which the imprisoned male has not escaped. Length, 1-6 mm. SoutH Arrica, 1914 (de Charmoy). Ceroplastes destructor, sp. nov. Ceroplastes ceriferus (Anderson) Newstead, Bull. Ent. Res. 1, pp. 66, 195 (1910). Female test. White, creamy white or dirty white ; exceedingly soft and containing an excess of moisture. Form irregular, with large but ill-defined gibbose protru- berances ; sides usually with two narrow opaque lines of secretion from the stigmatic clefts. No trace of lateral plates. Length, 5-8 mm. Female, adult. More or less hemispherical, with the sides often slightly compressed ; caudal process long; integument castaneous and highly chitinised, smooth and OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 27 shining, and without fovea or lateral tubercles. Antennae of six segments, the 3rd being as long as the 4th, 5th, and 6th together; the last three segments with stiff and bluntly pointed, spinose hairs. Legs small; hind femora very short and often distinctly incrassate ; hind tarsi equal in length to the tibiae, or sometimes slightly longer. Claw very short; lower digitules very long and stout; upper digitules normal. Stigmatic clefts well defined, but relatively small; stigmatic spines very small and pointed, bases not constricted ; basal attachment (disk) very large. Caudal process (after maceration) transparent and somewhat flexible ; sides with an irregular double row of short spinose hairs, and in addition to these there are two pairs of longer hairs (one pair of which is twice the length of the others) slightly ventral to the row of short ones and towards the distal extremity. Anal lobes short and highly chitinised. Dorsal pores very small, rather widely separated. Ventral integument opposite the caudal process, with rather extensive groups of circular pores, many of which, in well cleared specimens, are linked together with lines of dark chitin. Length, 4-7 mm. Female, second stage. Pyriform, the sides without tubercular projections ; caudal process long. Eyes relatively large. Antennae of six segments and very similar Fig. 18. Ceroplastes destructor, Newst., 3; a, antenna; 6, puparium. to those of the old adult female. Legs small; tibiae and tarsi of equal length. Stigmata large ; parastigmatic’ glands extending in a broad band to the stigmatic clefts. Stigmatic spines 40-50 in number, one of which is considerably larger than the rest; there is also a single long simple spine on either side of the group, near the extreme margin. Ventral group of circular pores slightly larger than those in the adult, and the chitinous strips connecting them are absent. Dorsum thickly studded with spinnerets, which are interspersed with short stout spines dilated distally. Caudal process absent. Length, 2-3-7 mm. In the second larval stage of the female the waxen appendages are arranged in a very similar way to those in Ceroplastes ceriferus, Anderson. Male puparium (fig. 18, 6). Elongate, faintly yellowish white or faintly greenish yellow; texture hard, glass-like and opalescent ; dorsum with a central boss-like _ patch of white secretion ; margins with three relatively large, opaque, white appen- dages, which are often sub-divided. Anterior extremity with three similar but much 28 ROBERT NEWSTEAD. smaller appendages, the outer ones diverging from the central one; these processes are however often wanting and seem to be readily abraded; many of them are also strongly recurved over the dorsum. Length, 1°5-1‘7 mm. Male. Dried examples dull red-brown or dusky brownish-crimson. Ventral ocelli in a single pair. Antennae (fig. 18, a) short and stout and of ten segments. Genital armature styliform and very long. No visible trace of caudal filaments, though the usual organs for the attachment of these are present. Wings strongly and beautifully iridescent, and when seen by direct light the membrane appears distinctly and regularly punctate. This Ceroplastes has twice previously been recorded (loc. cit.) as being doubtfully referable to C. ceriferus, Anderson; and this chiefly because of the striking simi- larity of the old adult females to those of the latter. With a more extensive series of specimens, in all stages, one has been able to gather that the Uganda insect is in many ways markedly distinct; more especially so are the young adult females, the male and its puparium. Ceroplastes egbarum, Ckll. Male puparium. Narrowly elongate; ventral half shaped like a shallow boat, opaque glassy white; upper half also glassy beneath, but covered with a thick, opaque, creamy-white layer of wax; with or without a median depression, in the Fig. 19. Ceroplastes egbarum, CkIl., 3 pupa, xX 130. centre of which is a very short elongated process of white faintly segmented secretion, which is broadly divided transversely in the centre. Marginal processes similar ; three on each side and two cephalic. The dorso-ventral scales are rather widely OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 29 separated distally from a point near the centre of the puparium, so that the terminal portion of the abdomen of the enclosed pupa is more or less exposed. The examples submitted are grouped together in long narrow colonies, two or three abreast, one overlapping the other with, generally, the lower half of the puparium only exposed. Length, 2-22 mm.; width, 1 mm. Pupa (fig. 19, a). With well developed wing-pads and three large anal tubercles. Length, 2 mm. Pro-pupa. Antennae of six segments, of which the 3rd and 6th are longest, the former longest of all. Stigmatic spines three, narrowly conical, the central one the longest. Marginal hairs few and widely separated. Just within the margin is a series of exceptionally long subcutaneous gland-tubes, which are placed rather closely together. Gotp Coast: Tamale, on Pithecolobium saman, 31.1.16 (C. Saunders). The discovery of the male puparia is of much interest. They were attached to the same branch as two of the females, but were slightly separated from the latter. Ceroplastes lamborni, sp. nov. Test of old adult female. Conical, more or less vertical in front and sloping upwards and forwards from the caudal process; sides with faint traces of three stigmatic plates. The four white stigmatic processes extending beyond the margin. Colour, in old dried examples, dusky red-brown and somewhat oily in appearance. Texture hard and almost as brittle as resin. Length, 3°5mm.; height, 35mm. A slightly larger example was also submitted, but this was imperfect. Female, old adult. Dorsum obconical, with a deep constriction at the base in a line with the caudal process ;_ sides below the constriction bulging and irregular ;_ cephalic lobe well defined. Stigmatic clefts deep, but relatively small. Caudal process very short, piceous. Integument shining, dull castaneous; the conical projection of the dorsum with obscure black and more or less confluent spots. Antennae of six segments, the 3rd much the longest and occasionally with a partial sub-division, the Jength equalling the three succeeding segments; formula: 3, 2, 6, 1 (4,5). Legs normal. Stigmatic spines small, conical and arranged in a relatively large, compact group; one of them, near the centre, is slightly larger than the rest. Dorsal pores small, with a minute cylindrical external projection resembling a truncated spine ; intervening spaces faintly reticulated. Marginal spines traceable only in places, minute and hair-like. Female test, second stage. Somewhat rectangular in outline; dorsum with a well- marked and somewhat angular prominence, behind which, immediately over the anal orifice, is a broad tongue-shaped plate of dull white secretion ; sides with three well- defined plates; stigmatic processes pure white and projecting considerably beyond the margin, when perfect. Spaces between the plates and elsewhere translucent, the colour of the sublying insect showing faintly through. Texture hard and brittle. Female, second stage. Elongate, narrowed and produced in front, truncate pos- teriorly ; dorsum with a strong keel or, in more advanced specimens, a fusiform swelling, surrounded by a deep channel; sides more or less vertical; venter flat ; eyes prominent; cephalic lobe strongly produced. Stigmatic clefts well defined. 30 ROBERT NEWSTEAD. On maceration in KOH the integument is thin and transparent, with the exception of a small portion surrounding the minute caudal process. In very young females the test is ovate, opaque and glassy, with a very elongated plate of bufi-yellow secretion, the posterior end of which rests against the naked caudal process. Stig- matic processes pure white and projecting beyond the margin. SouTHERN NicErtA: Ibadan, on cacao and “on a large climber on bush tree ” (Dr. W. A. Lamborn). The tests of the old adults bear some resemblance to those of Ceroplastes coniformis, Newst., but are markedly distinct. There isa striking, indeed a remarkable, difference between the young and old examples, but their structural details are specifically identical. Ceroplastes subdenudatus, sp. nov. Female test. Incomplete, covering only the central portion of the dorsum of the 9 ; this is more or less bluntly conical, the apex with an irregularly pointed patch of dull white wax, sometimes faintly stained with yellow or yellowish brown, with a central © nuclear spot ;_ the rest of the body of the 9 with a thin transparent coating of varnish- like secretion ;_ stigmatic processes white, thread-like and curly, resting upon the sides of the 2. Greatest diameter, 2°5-3°4; height, 2°3-3°6 mm. Female, adult. Somewhat hemispherical, but the height is shghtly greater than the diameter at the base, the latter circular in outline. Cephalic lobe and stigmatic clefts rudimentary. Caudal process rudimentary and placed, as a rule, just above the middle line between the margin and the centre of the dorsum. Surface shining, faintly wrinkled at the sides. Antennae relatively small, of seven segments; 4th much the longest ; formula, 4, 2, 3, 7, 1 (5, 6), or the 7th may equal the 3rd; long stout spines on the last three segments. Stigmatic spines of two types; those at the margin stout and pointed, forming a group of 15-16; the inner group obconical and numbering 19-22. Marginal spines similar to the pointed ones in the stigmatic clefts ; very few at the sides, but they are placed closely together on the cephalic margin. Postanal arch of chitin with a well-marked series of circular pores. Two specimens gave the following measurements: (1) diameter, 2°4; height, 3°3; (2) diameter, 3-2; height, 3°5 mm. UGANDA: Entebbe, on Acacia sp., 7.iv.14 (C. C. Gowdey). Easily distinguished by the apical patch of white wax and the unusually long 4th segment of the antennae. Ceroplastes vuilleti, Marchal.* Puparium of male. Hlongate. Colour of the dorsal half somewhat like that of dirty beeswax, with seven relatively large protuberances partly covered with dirty white, granular wax; one anterior, three bilateral and one central. Lower half shaped like a shallow boat; posterior extremity often tilted upwards, with a narrow transverse slit between the dorsal and ventral portions. Inner: surface dirty grey. Length, 14-155 mm.; width, 0°7-0°8 mm. * Bull. Soc. Zool. France, 1909, p. 68; Mem. Soc. Zool. France, xxii, p. 165 (1909). OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—IV. 31 The specimens are badly weathered and discoloured, so that too much importance must not be attached to the colour of the puparia as givenabove. Itis highly probable that the protuberances of the dorsum were pure white when quite fresh and that the texture was also more or less glassy. Female, adult. Marchal (I. c.) has described the female and its test, in detail ; but there are some structural points which need elaboration and these are appended below. Antennae (fig. 20 a, a) of eight segments, the articulations of the last four segments appearing very broad owing to the thinning down of the surrounding chitin ; in one example the 4th and 7th segments show partial but irregular and incomplete articulation. Another example has the normal number of segments on one side, and on the other only six; the latter is clearly malformed, owing possibly to the presence of Chalcidid parasites. Legs normal; lower pair of digitules not so markedly dilated as is usual in the members of this genus; tarsi shorter than the tibiae. Dorsal pores (fig. 20, b) very numerous; the intervening spaces finely and faintly reticulated in the more highly chitinised portions. Stigmatic clefts deep and Fig 20. Ceroplastes vuilleti, Marchal, 9; a, a, antennae; b, derm glands. | broad ;” the spines at the outer edge relatively long and pointed ;_ the broadly ovate, | Inner group very short, conical and placed closely together; one in the centre of the group is larger than the rest. Marginal spines similar to those at the edge of the stigmatic clefts, but smaller and widely separated. A number of minute spines are scattered over the dorsum. Marchal found one antenna with nine segments, but states that the additional segment was probably due to the sub-division of the 8th. His description and figure of the stigmatic cleft and the marginal spines is correct, but he evidently failed to determine the true character of the group of minute conical spines on the interior of the clefts. Apart from these details my examples agree with the rest of Dr. Marchal’s description and figures. S. Nicer1a: Agege and Ibadan, abundantly on pigeon-peas, 1913 (Dr. W. A. Lamborn). Dr. Lamborn is to be congratulated on the discovery of the male puparia, as little is known, comparatively speaking, of the males or their metamorphoses in this genus. 32 ROBERT NEWSTEAD. Ceroplastes zonatus, sp. nov. Female test. Broadly ovate in outline, highly convex; marginal plates very faintly indicated, but apparently without nuclear spots; dorsal plate very large, with a central nuclear spot of white wax ; cephalic margin slightly clypeate ; lateral: margins in very old examples with a pronounced foot-like extension from each of the stigmata, from which there extends a thick white waxen appendage. In the younger forms the foot-like extension is wanting, but the white waxen appendages are present and always porrected. Colour creamy white, suffused with very pale brown; dorsal plate surrounded by a shaded wavy zone of dark brown and brownish black, with here and there a suffused patch of dull flesh-colour. In very old examples the zone } of colour extends to the margins and is of a shining madder-brown to piceous colour. On the removal of the outer surface of the test with chloroform, it is seen to be divided Fig. 21. Ceroplasies zonatus, Newst., sp. n., 9; a, antenna; b, stigmatic cleft, with spines; ¢, portion of anterior group of spines ; d, portion of proximal group of spines. into seven areas by pale orange-coloured lines ; a central polygonal area, corresponding to the area occupied by the dorsal plate, from the angles of which radiate to the margin single lines marking off the areas of the lateral and cephalic plates. Female, adult (denuded of the test). Ovate; cephalic margin clypeate; dorsum low and wrinkled ; twe large, bilateral, submarginal extensions, both longitudinally striated; the space between these extensions of the body-wall and the margin markedly constricted. Dorsum with a large keel-like process. Caudal process very short and conical. Stigmatic clefts deep. Antennae (fig. 21, a) of eight segments ; the articulations relatively very broad; 3rd about equal in length to the 7th and 8th OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)——IV. 33 together ; a very long hair on the 2nd and 5th, and a slightly shorter one on the 8th ; there are two spines on the 8th and one on the 7th. Legs normal. Stigmatic spines (fig. 21,6) covering a large and somewhat pyriform area, the length of which is nearly equal to twice the length of the antennae ; the spines, with the exception of a small roup near the stigmata, are obconical and the space between them with bands of Task granular bodies, which collectively form a polygonal reticulation (fig. 21, ¢) ; the small proximal group of spines (fig. 21, d) are longer than the others and pointed. No trace of marginal spines. Derm thin and transparent after maceration ; rather thickly set with minute pores and minute scattered spmes. Caudal process surrounded by a porose zone of brown chitin. Length, 3-9-4°6 mm. Soutu Arrica: 1914 (EZ. d@ Emmerez). Inglisia theobromae, sp. nov. Female test (fig. 22,a). Strongly bilobed, the lobes forming two cones, which aré confluent at their bases; each cone is rounded ventrally and almost flat dorsally, with the ends truncated. In the younger forms the cones are fixed at an angle of, approximately, 45°; in the old adults they are much more diverted and in some 1 a iit Fig. 22. Inglisia theobromae, Newst., sp. n., 2; a, test; b, anal appendages of test; c, c, anteunae; d, stigmatic and marginal spines. instances assume an almost horizontal position. The cones are strongly fluted vertically and very finely striate transversely. Colour pinkish buff with a faint pearly lustre. Anal aperture (fig. 22, 6) inthe young adults with a pair of thin, trans- lucent, waxen plates, upon which there rests a pencil of thick, white, waxy filaments with truncated ends; the latter are held in place by the anal lobes and their accom- panying bristles. Above the lobes a small irregular mass of dusky secretion. (C365) o 34 ROBERT NEWSTEAD. Length, between the extremities of the cones, 45 mm.; height of middle line, 2mm. Young adults measure: 2°2 mm. between the extremities of the cones, and 15 mm. along the middle line of the dorsum. These may be taken as average measurements. Female, adult. Shape similar to that of the test, but shrivels on drying, so that the test readily falls away in two parts. Antennae (fig. 22 c, c) of seven or eight segments ; in the former case the 4th is the longest, but invariably bears one or more partial divisions ; in those having eight segments the 3rd is the longest. Stigmatic clefts obsolete, the position occupied by a single long spine (fig. 22, d) which is equal in length to two-thirds of the antenna ; it is sometimes strongly curved at the end. Stigmata just within the margin and between them three to four circular pores. Marginal spines (fig. 22, d) in an irregular row, and of two sizes; all are conical and sharply pointed. Anal orifice surrounded by a very long loop-shaped band of dense chitin. Anal lobes pointed ; inner edge with two tubercular projections; the tubercles and the tip each with a spine, and there is also a very large spine attachment in the middle of the apical projection. Anal ring with six hairs. A narrow line of dorsal pores extends from the anal lobes to the region of the proboscis; the latter relatively small and the mentum monomerous. ' Ucanpa: Nagunga, on stems of cacao pods and flowers, 18.ix.15 (C. C. Gowdey). In the form of the test this species bears a close resemblance to I. castalloae, Green,* but differs in having the ends truncated, in the absence of the “deep transverse furrow extending downwards from the apex of each cone’ and the “ rounded lips ” forming the border behind and in front. Furthermore, the female of I. castilloae has no stigmatic spines, and the antennal characters are also different. * Jour, Econ. Biol., vi, p. 29, pl. i. (1911). NATURAL TRANSMISSION OF TRYPANOSOMIASIS (T. PECORUM GROUP) IN THE ABSENCE OF TSETSE-FLY. By Rupert W. Jack, F.E.S., Government Entomologist, Southern Rhodesia. (Maps I & II.) The opinion that trypanosomiasis of the T. pecorwm type is, under favourable conditions, transmissible amongst domestic stock through the agency of blood- sucking flies other than Glossina has been expressed by several observers.” Observations in Southern Rhodesia during the past seven years have convinced the writer that, given the necessary conditions, transmission of this disease may take place in the absence of tsetse-fly, and that such transmission has occurred more frequently than has been generally recognised. It may be of interest therefore to give the facts concerning several outbreaks in the territory which form a chain of evidence pointing to the above conclusion. The trypanosome was in each case diagnosed by Mr. Ll. EK. W. Bevan, Veterinary Bacteriologist, 8. Rhodesia :— I. Ts cattle. Farm on railway about 3 miles N.E. of Gatooma. Season 1909-10. Ten cattle were brought to farm about 26th October 1909, from Victoria, but had been by rail to Salisbury and back to Gatooma. Two were brought to farm about 6th December. At the time of my visit, 3lst January, the two latter were healthy ; but the remainder, of which one had already died, were all “ fly-struck,” the first case having been noticed during the first week in January. The range of the cattle was confined to the farm, the road between the farm and Gatooma and across the railway to the Cam mine. Fly is known to exist 3 to 5 miles to N.E. (see Map J). The writer spent a full week looking for fly on this farm and in the vicinity with negative results. A few specimens of Tabanus and a few Stomoxys calcitrans were seen about the cattle. Karly in March the two remaining cattle fell sick and were diagnosed by Bevan as suffering from trypanosomiasis. Since that year cattle have lived in a healthy state on this farm. Il. 4H.’s cattle. Farm within 2 miles of Gatooma to S.E. Season 1909-10. Forty cows were on the farm for sixteen months and out of these nineteen showed symptoms of trypanosomiasis (diagnosed by Bevan). The only occasion on which any left the farm was from August to November, 1909, when all but six went to the Inez Mine (opposite direction to the fly belt). The first case was recorded in July 1909. Stomoxys calcitrans, Lyperosia sp. and two species of day-feeding mosquitos were common about the range of the cattle. No tsetse was seen on or near this farm, * (1). Kinghorn & Montgomery: ‘“‘A Report on Trypanosomiasis of Domestic Stock in N.W. Rhodesia.” —Ann. Trop. Med. Parasit., June, 1908. (2). G. E. Owen: ‘‘ Mechanical Transmission in Trypanosomiasis.’—J1. Comp. Path. Therap., September, 1914. (The species of Trypanosome in this instance is not certain.) (3). H. Llewelyn Jones: ‘“‘ The Treatment of Trypanosomiasis in Cattle caused by the Trypanosoma pecorum.”’—Jl. Comp. Path. Therap., June, 1915. (C365) 2 36 RUPERT W. JACK. and there was no subsequent reason to believe that “ fly’ occurs in the vicmity. Many cattle are used on the roads in and out of Gatooma, and if fly occurred the fact could not remain hidden. III. Chicago Mine. 8. bank of Msweswe River and about 5 miles HE. of railway line. Season 1909-10. Thirty-six oxen had been at work near the mine for two years, drawing crushing material from a shaft about a mile from the mine. The only time when any of them were away was when four oxen drew a Scotch cart to Gatooma and back about a month before Christmas 1909. The first case was noticed in March 1910 (diag- nosed by Bevan). At the time of my visit, 6th April,-five cattle had died and at least ten more were “struck.” SS. calcitrans and Lyperosia sp. were extremely numerous about the cattle. The writer could find no sign of tsetse in four days, and there has been no subsequent reason to believe that fly occurs at this spot. Cattle have been working on other mines near by for years, but early in 1909 fly- struck cattle were brought to the Lydia Mine, which is only a short distance across the river from the Chicago, and may have initiated the outbreak. The nearest known fly belt is 12-15 miles distant in a straight line to the north. IV. M.’s swine. Farm on N. bank of Umfuli River, on E. side of and adjacent to railway line. Visited on 15th November 1911. A number of sows had been on the farm for two years. The first death occurred on 15th October 1911, and fifteen had died up to time of the visit, as well as one boar. A large number of young pigs on the farm were all healthy, and a number of cattle there also remained unaffected. Tabanus taeniola was very abundant on the cattle at the time of the visit, but did not pester the swine much. S. calcitrans was abundant on the swine. This farm is about 10 miles from any part in which one would expect to see tsetse, but the fly was at that time probably carried occa- sionally close to Hartley, which is about 3 miles distant from the farm. The immunity of the cattle, apart from other facts, practically proves that something other than tsetse was responsible for the spread of the disease amongst the pigs. - (N.B.—Bevan is doubtful as to the trypanosome in the swine being identical with that affecting cattle in the district, but states that morphologically it belongs to the T. pecorum group.) V. S.’s & G.’s swine. Adjacent farms on the Umfuli River. S.’s farm is also adjacent to M.’s. Date of visit, 29th January 1916. Of 8.’s swine, three pigs were bought from M. and had remained on the farm since January 1914. On 18th December 1915, 32 sows arrived from East London. A sow from Kast London died on 15th November 1915, and a second on the 27th, both being heavy in young. On 21st December, 40 pigs arrived from East London (35 sows and 5 boars) To that date, 19 heavy sows and 1 boar had died and two more died during the writer’s visit, whilst several young pigs showed symptoms of being “struck.” Other pigs died subsequently. Cattle ranged on the farm over the same ground as the swine and all remained healthy. Tabanus fuscipes and Haematopota sp. were and had been very troublesome to the pigs since the commence- ment of the rains. S. calacitrans was moderately abundant, and the pigs were, of course, much infested with lice (Haematopinus). NATURAL TRANSMISSION OF TRYPANOSOMIASIS. 37 G.’s pigs were obtained from 8. and were exposed to similar conditions, and the outbreak took a similar course. Cattle also remained healthy on this farm.* VI. M.’s Cattle. Sikombella River; Sebungwe District. In this outbreak the sequence of events is so clearly defined and the distance of the farm from any fly belt so great that it constitutes in itself a definite proof of the transmission of this form of trypanosomiasis in the absence of Glossina. The situa- tion of the farm in relation to the fly belts in the district is shown in the accompanying map (Map II). The outbreak occurred in the 1915-16 season and the history of the affected cattle is given by the owner as follows :— “In September 1915, four spans of oxen were taken to Sileya’s kraal and were employed in transport work between that kraal and Sikombella until 17th December. During October eight of these were taken by a Dutchman into the fly belt on the Sassame River—it may have been the Ensengwe River fly belt in the vicinity of the Sabala River (Bandala’s kraal)—and on and off after that were in contact with the remainder of the four spans until 17th December.f After this they were running not only with the remainder of the four spans, but with all the cattle on Sikombella farm; the eight infected oxen were all dead by the end of December, one having died with the first rain at Sileya’s on or about 15th November. At the middle of February some twenty head showed sick, but the true cause was not, suspected at that time. About 25th February a team of healthy-looking oxen, made up of 11 oxen which had been to Sileya’s and 5 young oxen which had not previously left the farm, were inspanned and sent to Gokwe with a load in company with another team of 16; which had been to Sileya’s and looked fit ; two of these died on the road up, including one of the young oxen, and some six others became sick, including three of the young animals (these three subsequently recovered and are now working‘) I now removed all sick animals to a camp about a mile away, but from time to time I had to keep adding an animal which showed sick amongst the‘healthy herd. Towards the end of March two old cows and my bull (none of which had left the farm) became sick ; the two cows quickly died, one leaving a very young calf, which is alive to-day ; the bull has entirely recovered since. From the end of February the infected cattle were dying at the rate of two or three a week for two months, after which the death rate would be about one a fortnight until the end of September ; some 12 head of infected cattle remaining at that date have recovered and are now working. Some 36 head have died, including the 8 fly-struck animals; all these showed typical fly sickness. With the exception of the eight oxen taken into the. fly belt, none of my oxen have ever been in the vicinity of a fly belt, having only travelled the main road between Sileya’s kraal and Queque; also the main road between Sikombella Farm and Pondo Mine (years ago)—Mr. Luxat’s.” Cattle had been living in a healthy condition on this farm and near by for years. The writer visited the locality in August 1916, and after exploring the vicinity for three days on the possibility of fly having “‘ come down” to the farm from the belt * The location of the preceding five outbreaks is indicated on Map I. + The oxen returned with the waggons from Sileya’s kraal to the Sikombella River in December. t The disease was diagnosed by Bevan from smears taken from these three oxen. 38 RUPERT W. JACK Map I. SX > - New Milky Way *Golden Valley Sketch-map showing the position of the Gatooma tsetse area. 39 NATURAL TRANSMISSION OF TRYPANOSOMIASIS, Map II, rs e \Bululai K's py fa Tie taken by the 8 ano % 4 N ~—s> ahs, So \ 1 he Guwe2* oa SE Sm Mumbwe egetela ANS Ml i a Ae My 3a 2=©0s Nyamararo oa ie) A oe Wunu ar 38Gadoma &. a Gorodema. ” Vp \ con = ——— \e sie] Senyeniti / yy Qwtany, is ae r a you 2 Mope iss % {yt vIri ae 'd Kraal/) / a Robbs Drift “ally Messy, i ‘Ny, NA img " Cine e % : 4 SiN. \ .MANVoOnI _pSilika I > Stay wfeMarume Mkulu rses Nijelele a < Or;, 4 Cattle) , Wo. Little Sidoma. " R. Ad a latives keep Seng attle here EAS nym = wy . me *€ySidoma, Mkul c , Z “Nhiiys,, Fp R.Mbumbusi Ona sr >’ ae e & eos R.Iattope att Am ms orn, fF S arm 36S rcs ~ 80 Scene of outbreak No.Vi "S et } Suna Vag COMMIT ES Mine &Do a Ane bella Sketch-map showing the position of the fly areas in the Sebungwe District. 40 RUPERT W. JACK. to the north, travelled down the Gwanyika River and down the Umniati to ascertain if any movement of the fly had taken place. No fly were encountered until the spot was reached where its southern limit on the Umniati River was marked in 1914. From the Umniati the return journey to Gokwe was made via the Mtanka River, all the country traversed, as on previous occasions, proving free from fly. The following points are noteworthy in connection with the cases cited above :— (1). In all cases the outbreak attained its height during the spring and summer months (October—April). (2). In all cases, in connection with which sufficient time has elapsed for such observation, there has been no recrudescence of the disease in the following season, although susceptible animals have still been kept in the same locality.* (3). In all cases infection occurred only amongst animals that were herded together. From the foregoing it appears probable :— (a) that the transmitting agent or agents are abundant in the spring and summer and disappear or decrease greatly in number in the winter ; (b) that the disease is not readily transmissible under Southern Rhodesian conditions by these agents from chronic cases that live over until the rains or recover ; 7 (c) that the agents concerned in this method of transmission are not capable of perpetuating the disease indefinitely, or are only capable of doing so under exceptional circumstances ; (d) that- the method of transmission is of a mechanical and not of a cyclical nature ; (e) that the segregation of infected animals (on showing temperature) would in the absence of tsetse effectively check the spread of the disease. Allowing for the moment that the deduction given tentatively under (a) is correct, very few species of blood-sucking Diptera are eliminated by this statement, as, apart from Glossina morsitans and certain species of Hippobosca, all the common blood- sucking flies are very much scarcer in winter than in summer. Moreover, observa- tions as to the prevalence of possible transmitting agents in the locality of the out- breaks furnishes a number of different species as likely subjects for experiment. The most probable agents are included in the following :—TZabanus fuscipes, T. taeniola, Haematopota pertinens, Haematopota spp., Stomoxys calcitrans, Lyperosia and mosquitos. Of these it must be admitted that S. calcitrans and mosquitos are the only agents that are everywhere abundant throughout the season over which infection extends. It is, however, by no means necessary to infer that the power of mechanical transmission is confined to one species or to one family ; and in view of Mitzmain’s;failure to associate S. calcitrans with the transmission of surra in the Philippines, it appears unlikely that this species is solely responsible in regard to the present.disease. The ‘whole matter calls for careful experiment and investiga- tion. ; * This is, of course, contrary to the experience in Barotseland and Portuguese East Africa NATURAL TRANSMISSION OF TRYPANOSOMIASIS. 41 With regard to (b) and (c), these deductions are fully borne out by experience in this territory where fly-struck cattle have been freely moved into fly-free areas in the past. These movements have not resulted in establishing trypanosomiasis in any area away from the fly belts. If this method of transmission can continue freely and indefinitely, it is difficult to understand why the disease has not by this time become generally spread throughout the continent. In no instance as yet recorded in Southern Rhodesia, however, can Glossina be definitely disassociated from the inception of the outbreak. The main point is, that the herding of “ fly-struck ” cases with healthy animals is a practice attended with danger, especially in the spring and summer months. eek | AIR ATMOROMATEMR DIO i ait fb PSH O AEG XD vd. tua aero elut 6 eis ecoion had Odt ci, anete aoqiytt odak bavoiv vise} ada 3% ack HOG, thie BiRA HOLOBENRE: pats de s\cintesy: ib hebland di yy leer! exci aod, aeQlwolesinninest Jo bedi ada ira nidd, = JOuk eadkonessils ¢ odd. Vile Devas po s, Asmcityoa, att die . od | pargqeosy) map save won ,aleobod st sip d 08 i bab f sad? al ni aa ant dead add to. nos iret ie aD f tort bedaloodin F, elation t diiv holiignte gars A tl al Laser lees yetlaed Agia Ae fy : | | ald roan OAL i Y yee an, saith — = 10. yalbyed: * ies t ties StH ey, 10% ; , Li F * T Fee Let : fit 13- j ) , ; »* a if ‘ j ty | bets f trunnnisaon 40 of « Mmechetticah Awe, ae ; - . ‘ é in TFA & recauiatt) ty : 4 rl Mis dA) hare =p | t elentivey chook the wie al thas love Shak sf | i thent At weG pee pee v) baht su Javely anor ta hos ; vec Of Dn aS I Aree e olbnimatedl hie thie gtate won weleni and Certa ypewies 0) \ Hepat all. vi aomanen fo Ab 5 nv jst Sateen Li wit ; n “A56 it eu, "Mocevret, 0 ie Lone of peghitlle thu. ving agenda ae tie doanliey et he a <i mbar of eeedeont spies i Viale subjects fir ex, pputvse. he, tgertocic ro the following tn Tobanes, Doty ; 7 a | 7 ern Saat hail aes etornta, Up Stomyrys calevtrans, . ba Ei ecquicd OF shea le penne ee i a a Ree el Soca tien & é + i ; ; if | be is , . mt (2° coon are? Mite oxteuk irik So . eet T , meiry “5 to: tnloy ti ‘o' be Dy of sab. woos Hae lee ta Lage Baits ow ive tie rechaneae % ao \ibaly Sabian: Te EAs | Fe pan "s "an Africa 43 A NEW SPECIES OF PARAPHELINUS, PERK.,- FROM BRITISH GUIANA, WITH A DISCUSSION OF THE GENUS AND THE ALLIED APHELINUS, DALM. By James WaterstTon, B.D., B.Sc., Imperial Bureau of Entomology, London. During the past summer (1916) Mr. C. B. Williams has, on behalf of the Depart- ment of Agriculture, Trinidad, been engaged in investigating the natural enemies of Tomaspis and its allies, in the hopes of finding some effective means of controlling the froghopper, Tomaspis saccharina, Dist., which in recent years has caused consider- able damage to the sugar plantations of Trinidad. With this object Mr. Williams has made collections of minute parasitic Hymenoptera in the Island and the adjacent mainland of British Guiana, and this material has from time to time been forwarded to me for study. Mr. Wilhams has from the first been fully alive to the importance of securing any specimens of the egg-destroying Paraphelinus, Perk., and I am glad to find species of this interesting genus in the gatherings from both Trinidad and British Guiana. The literature dealing with Paraphelinus is scattered and perhaps inaccessible to those to whom it is most likely to prove serviceable. I have therefore, in some introductory notes, recapitulated the main facts in the life-history and host attachment and added a bibliography to the present paper. As an economic insect Paraphelinus owes its importance to the fact that while most APHELININAE attack Aphids, Coccids, and Aleurodids, this genus destroys the eggs of Xiphidium and Tomasms. Although the genus is of comparatively recent erection, it is likely, as Dr. Perkins pointed out in his original description, that Agononeurus locustarum, Giraud, bred from the eggs of a grasshopper, Xiphidiwm fuscum, is really a Paraphelinus. P. xiphidii, Perkins, is an efficient controller (Swezey, 2, 4) of Xiphidium varipenne, Swez. The parasite was first observed in March, 1905, in the grounds of the Hawaiian Sugar Planters Association’s Experi- ment Station, Honolulu, where it had almost certainly been introduced. The eggs of the Xiphidium are laid in clusters just beneath the leaf-sheath of the sugar-cane, where they are attacked by the parasite ; 7-14 examples of Paraphelinus may emerge from a single egg and occasionally the whole cluster of eggs (2-15) is affected. Like many species of which several imagines emerge from a single host, all the individuals in a brood of P. xiphidii use the same hole of emergence, which is gnawed by the first to hatch out. The aperture is circular, and when the host egg is closely pressed against the leaf-sheath, the latter as well as the egg-shell may be gnawed through. If the Xiphidium egg is examined in balsam after the parasites have emerged, the empty pupal envelopes of the Paraphelinus may be seen, together with numerous darker granules within the shell. These granules are the accumulated foecal matter which is discharged by the insect shortly after emergence. _ Para- sitised eggs are bluish-black in colour. The life-cycle of Paraphelinus varies from 20-31 days, that of the host being three months; the parasite is thus placed at a distinct advantage, and shortly after it was first noted, proof of its beneficial work was given by the rapid disappearance of Xiphidium in some districts. 44 JAMES WATERSTON. About three years later Swezey (4) recorded a disconcerting departure from the usual habit of Paraphelinus. In February 1908, P. xiphidi was bred as a hyper- parasite from cocoons of a Dryinid, Haplogonatopus, Perk.* Experiment showed that Paraphelinus reared from the Dryinid readily attacked Xiphidium eggs and the resulting imagines quite as freely reverted back to Haplogonotopus, attacking the larva just before it pupated. Whether the host was an egg or larva the number of Paraphelinus was the same, viz., 12 or 13 commonly. Apparently P. xiphidii is a vigorous parasite, attacking whatever pabulum most conveniently offers itself; and the unfortunate feature of this particular case of hyperparasitism lies in the fact that it was desirable to protect the Dryinid because it destroyed leaf-hoppers. X¢phidium varvpenne itself is partly a noxious insect, in so far as it attacks tender shoots, flower stamens, etc., and partly useful as a devourer of Aphids. The material which I have had before me is as follows :— (2) A single female—‘ St. Clair, Trinidad; 12.iv.16; swept from grass ” (C. B. Willams), which I have determined as P. tomaspidis, How. (see notes p. 96) ; (b) two lots of a distinct new species from British Guiana: (i) Issororo, N.W. Dist., 6.vii.1916, Lot 7, ““ Emerged from base of grass, Imperata caudata, containing Tomaspis ? carmodyi” (C. B. Williams), 2 gg, 8 92; (i) ib. et id., Lot 8, during July 1916, 83g, 2699; (c) a slide containing a “ parasitised egg of Xiphidium g and 6 QQ of Paraphelinus, Perkins ”—Lahaina, Maui, 28. ix. 1905 (O. H. Swezey) ; possi- bly the seven examples referred to by Swezey (2, p. 215). This was generously placed at my disposal by Dr. Perkins to treat as might seem best. The seven examples have been remounted after careful potashing, the male and three females being partly or completely dissected. These preparations with others from Mr. Williams’ material are deposited in the British Museum. Only the most careful manipulation will produce satisfactory preparations of Paraphelinus. The wings may be detached from either dried or spirit material and mounted in the way already suggested (Bull. Ent. Res. vii, p. 231-2) ; but after potash and acetic acid, transference to oil should be gradually attempted. Three or four graduated layers from oil to absolute alcohol in a narrow 1 inch tube are not too many when dealing with such delicate organisms; the actual transference to absolute should be effected as quickly as possible, for even a momentary pause will permit the acid to evaporate and the preparation will be ruined. The object of having the specimen charged with acetic acid when passed into absolute is to regulate transfusion and to prevent collapse. The slide of P. ziphidii was treated as follows :—The hard balsam at the edge having been removed, a little xylol was run round the cover slip; when the xylol had acted on the balsam fora short time two needle points were inserted a little apart and the cover broken by pressure between. In this way the cover was gradually chipped off without at any time bringing pressure to bear upon the brittle mounts. * “ H. vitvensis. . . . No repetition of this hyperparasitism has since come to bend notice.” —O. H. Swezey, in litt., Dec. 1916. A NEW SPECIES OF PARAPHELINUS. 45 It was interesting to find that after ten years the balsam had hardened only for a short distance from the edge. After being detached the specimens were thoroughly soaked in xylol and transferred to spirit, after which they were manipulated in the usual way. External Morphology of Paraphelinus, Perk. In the following account of Paraphelinus some notes of more than generic appli- cation have probably been included. The value of particular characters, whether tribal, generic, or specific, is at present very imperfectly understood in the Aphelinines. In particular, too little attention has been paid (a) to the clypeus and mouth-parts, especially the labrum, (b) to the sternal apophyses, and (c) to the general chaetotaxy ; and in respect of these and some other features, I have tried to place the genus on a sounder basis. : Although the elongate facies, the antennae, and the head, distinguish the group at a glance, it is perhaps worth noting that one of its most striking features—the course and ending of the genal keel—might quite easily arise by an exaggeration of conditions obtaining in other allied genera. In the not very closely related Cocco- phagus group, é.g., 1m some quite thin-headed forms, the keel is well forward, though the genae are normal. A little further shifting would cause the lower extremity of the keel to fuse with the sides of the clypeus and the condition found in Paraphelinus would result.. Similarly the antennae with their distinctive “ elbow ” and terminal “horn ” might result from an inconsiderable modification of the antenna in A phelinus itself. The cumulative effect of such shght differences however is decided, and (apart from their host attachment) the species showing them form a compact genus. The two known males are easily separated, but the females are extremely uniform, presenting differences of a rather critical nature. In discussing the antennae I state that 6 (7) joints are present ; 7.e., of these six are obvious, but the first funicular has, I believe, the lamina or disc of the ring joint fused to it basally. With Perkins and Howard I assign the penultimate joint to the funicle and consider the last joint to be a solid club; Girault treats the penultimate as the first of a two-jointed club. In its natural position (seen best in a specimen mounted without pressure in a cell or under a supported cover-slip) the antennal “ beak” curves gently outwards and the sense-organ appears in its normal position on the inner aspect. When, however, the antenna is mounted separately, the weight of the cover-slip, without distorting the general shape, brings the surfaces into the same or parallel planes, when the tip of the antenna appears to be decurved, and the sense-organ is dorsal. All the figures of the antenna have been drawn from this position. Besides the usual linear sensoria the club bears two peculiar organs each with a central stouter bristle (?); focussing down a little, these organs seem circular, but when the central bristle is distinct the surrounding rim is vaguer. The head is particularly difficult to examine, owing to the inequalities of surface it presents, and to its liability to collapse during the process of mounting. The chaetotaxy and shape of the clypeus may be made out in shrunken specimens, but notes on the proportions are worthless unless the head is naturally swollen. In the thorax noteworthy features are the slight advance of the axillae, structure of the sternopleurae, etc. One peculiar abnormality of the dorsal chaetotaxy 46 JAMES WATERSTON. deserves mention, viz., the very frequent failure of the second anterior medio- dorsal bristle on the left side of the mid lobe. The lateral ridges of the propodeon are really folds more or less internally strengthened. If preparations of Paraphelinus are made too quickly, the thin mid-dorsal region of the thorax is apt to collapse longitudinally, forming a more or less distinct furrow. This appearance can very easily be produced by plunging a specimen into oil after incomplete dehydration. Whether in life this line of weakness in the chitin has a sulcate appeareance or not is uncertain, but I think the point worth mention, as Girault (6, p. 74) says of Para- phelinus australiensis, ““ There is a narrow median grooved line down the thorax.” P. speciosissimus, Gir., is much more isolated, and it was hardly advisable to recon- struct the genus, as Mr. Girault (5) has done, from such a form. In four of the five described species the forewings are hyaline or faintly tinted, not conspicuously banded; the general coloration pale, not black and yellow; and the eyes hairy, not bare. Later Mr. Girault (9, p. 180) mistakenly, as it seems to me, sank Para- phelinus: “ The forms named Paraphelinus, Perkins, intergrade with the forms of Aphelinus, Dalman, as the Australian species show, so that obviously there are not two genera represented by them. Paraphelinus must therefore fall as a true synonym of Aphelinus.” In Aphelinus (sens. str.), which preys upon various Aphids, the head is short and thin, normal, lenticular, with large eyes, behind which there is no temporal develop- ment. The occiput is sharply margined where it joins the vertex. The genal keel passes ventrally outside and behind the toruli ; the clypeus is practically in the same plane as the frons, and its edge is straight. The mid lobe of the mesonotum is densely covered with short sharp bristles, the parapsides, which are deeply invaded by the axillae, bear 4 bristles. There are short, stout bristles on the metapleurae. The fore wings are triangular and the hind wings broad, with a short fringe; the abdo- minal tergites decrease in length posteriorly, and the ovipositor and its sheath are short. In all these characters Aphelinus, Dalm., offers a precise contrast to Para- phelinus, Perk. ; and there are other differences, less obvious but equally important ; e.g., in Aphelinus instead of a backward-sweeping parabolic incrassation the pleurae show a short nearly verticle anterior rib. The prepectora are medianly fused and, in conjunction, band- or coilar-like, reminding one of the genus Coccophagus, to which Aphelinus also approaches in its mandibles (cf. also the mesonotum). The genotype of Aphelinus is A. abdominalis, Dalm. (Svensk. Vet.-Akad. Handl., xl, p. 181, t. 8, £. 55, 56, 1820), of which A. basalis, Westw. (Ann. Nat. Hist., vi, p. 122, 1833) is, I believe, correctly regarded as a synonym. ‘The foregoing remarks on the generic characters are based on an examination of two very old specimens of A. basalis in the British Museum Collection ; one bears the data—‘ stood under this name in old B. M. Coll.—C. Waterhouse”; the other is labelled “ basalis”’ in the handwriting of F. Walker, to whom the determination of both examples is due. The first is now completely dissected on two slides, the second mounted whole, except that the fore wings are under a separate cover-slip. But while Paraphelinus is distinct from Aphelinus, it is closely approached by the small yellow-bodied group (at present placed in Aphelinus) which attack Coccids. Whether Perkins’ genus should include the latter or not I cannot say; it is one of the Coccid-destroyers, A. diaspidis, How. (Ann. Rept. Dept. Agric., p. 355, 1880), A NEW SPECIES OF PARAPHELINUS. 47 which Howard has figured as representative of Dalman’s genus (U. 8. Rept. Agric. Tech. Series, No. 1, p. 23, fig. 7, 1895). Genus PARAPHELINUS, Perk. Q Head (fig. 1, a, 6, c): In profile, long antero-posteriorly, the vertex, which hardly rises above the level of the eyes, being flattened, lke the upper frons. Face inflated, widening ventrally and curving in sharply above the perpendicular clypeus ; genal keel short; temples and genae swollen, the latter widening ventrally. Eye subtriangular, occupying about two-thirds of the depth, both upper angles broadly, SEPT md “ SS Fig. 1. Paraphelinus aiphidii, Perk., Q: (a) head, thorax, and propodeon ; (b) head, etc., in profile. Paraphelinus perkinsi, sp. n., Q: (ce) head, front view ; (d) pronotum ; (e) presternal apophysis; (f) mesosternopleurae—all flattened. the lower angle hardly at all, rounded, set with numerous bristles arranged in straight rows. From above, the vertex is both broad and long, the orbits nearly parallel, but slightly divergent posteriorly ; ocelli in a wide-angled triangle, the lateral pair about two diameters from the edge ; supra-occipital edge broadly rounded ; temples swollen distinctly backwards. From in front, the face proper contracts ventrally, the most remarkable feature of the head here being the running of the broad genal keel forward below the toruli to the upper angles of the clypeus, and the shortening 48 JAMES WATERSTON. of the mouth-edge. The swollen out genae are carried forward beneath the toruli also, and are largely visible from in front. In the middle of the clypeus is a minute notch flanked on each side by a more or less distinct short lobe or tooth. The toruli are elongate, sub-triangular, broadest ventrally, with their adjacent sides sub-parallel. Antennae (fig. 2), 6 (7)-jointed ; scape, pedicel (ring-joint), three in funicle, club solid. Bulla long, apex of pedicel oblique, funicular joints gradually expanded. Second funicular joint set at an angle to the oblique apical dorsal surface of the first ; second and third funicular in the same straight line, but the suture between them is more or less oblique. Third joint with one sensorium; club wider than last funicular, with several sensoria, and ending in a short outwardly curved beak. On —————— Se Fig. 2. Antennae of females of: (a) Paraphelinus perkinsi ; (b) P. xiphidvi; and (c) P. tomaspidis. the inner apical aspect is a small sense-organ consisting of 4—5 sensory hyaline bristles, stouter than the elements of the general chaetotaxy of the antenna, standing above definite passages through the integument of the club. Mouth-parts: Labrum triangular, apically rounded, with 2 central preapical bristles. Mandibles broad basally (8:7), tridentate. Trophi nearly circular; cardo small and narrow, stipes swollen and broad ; maxillary palpus with the second joint sometimes presenting a false appearance of being divided in two; labial palpus longer than either joint of the maxillary ; ligula with two median setigerous cells. Thorax (fig. 1, a, b, d, e, f) flat dorsally ; including the propodeon, half as long again as broad, widest at the level of the fore wings. The pleurae are very little A NEW SPECIES OF PARAPHELINUS, 49 convergent towards the rounded, nearly rectangular, postero-lateral angles of the propodeon. The integument weakly chitinised and smooth or faintly patterned along a narrow middle line ; the pattern beyond the middle line more raised towards the sides and the cells generally longer than broad. Pronotum descending, developed into two weakly connected triangular sclerites ; when flattened, deeply emarginate anteriorly, straight posteriorly, the hind and lateral margins meeting in rather over a rounded right-angle, befcre which is the deep semicircular spiracular emargination. The longest and stoutest bristle 1s placed above the spiracle and there is always another fairly stout bristle before the edge half-way from the postero- lateral angle to the middle. Prosternum bare, rectangular, truncate posteriorly. Episternite large, with a minute bristle anteriorly and half a dozen on the anterior articulating projection. Mesonotum distinctly longer than the scutellum, mid- lobe large, the width at the anterior extreme of the parapsidal furrows about equal to the length ; 10 stout coloured bristles in two widely separated rows (4 : 4) about the mid line and one inside each parapsidal furrow anteriorly. Side lobes narrow ; two bristles on outer edge above the tegula, which also bears two bristles. The suture between the mid lobe and scutellum broad, a little curved antero-laterally where the small axillae (one bristle) shortly invade the rather narrow parapsides Scutellum with 4 bristles (2: 2), wider apart than those on the mid lobe. At che level of the anterior pair but a little nearer the mid line are a pair of clear pustules. Mesophragma long and narrow, reaching back to below the hind margin of the first abdominal segment. Metanotum and propodeon not carinate, but as both, espexially the second, are declivous, about the mid line the latter shows up in a more or less distinct ridge. Metanotum band-like (the smooth lateral hollows small and shallow), present mainly as a broad short reticulated post-scutellum, which is widelv angled over the propodeon. The latter as a rule nearly smooth, except medianly ; two posteriorly slightly divergent faint lateral folds, so far apart as to be nearly in line with the sides of the abdomen. Beyond each fold there is a narrow trisngular area with a short anterior sulcus ; at the inner anterior angle lies the broadly oval moderate sized spiracle. Sternopleurae: sternum broader than long, pattern various, but rather large, very finely raised ; two minute widely-spaced bristles on anterior edge, and 4 (one near each postero-lateral angle, and one on each side of the mid line) near posterior edge. Between sternum and pleura is a weakly defined edge, obliterated posteriorly (without internal support) and represented by a slight crease in flattened- out preparations. At the anterior end of this crease, and confluent with it for nearly half its length superiorly, there is a strong parabolic internal rib which, beginning at the antero-lateral sternal angle, sweeps backwards, enclosing about four-fifths of the pleural surface, and ends superiorly at the insertion of the forewing. The area thus enclosed is a little more prominent than the posterior (epimeral) region, which is delimited ventrally by a ridge, sloped from above the mid coxa diagonally towards the upper anterior pleural angle and fading out (at about one-fourth) in front of the parabolic rib already described. The pleural surface bears one minute bristle antero-ventrally. The episternite is indicated only by an internal chitinised suture. Prepectus large, antero-posteriorly striate, the cross reticulation nearly obsolete. The internal chitinous rib very near the hind edge. The metapleurae are absolutely bare. (C365) D 50 JAMES WATERSTON. Wings elongate, at least three times as long as broad, subspatulate in shape. Neuration reaching to one-half or just beyond, and very little bent at the junction of the marginal and submarginal. Submarginal cell narrow, with one minute bristle on the upper surface at one-third from the radix, and 2-4 apically. Sub- marginal vein with four short stiff bristles, one stronger on the marginal before that vein reaches the costa, 6-8 major bristles frmging marginal and postmarginal combined, while dorsally these veins bear 9-12 short appressed bristles. One bristle on radius, which is acutely inclined towards the costa. Basally the wing is narrowed, showing a few bristles below the submarginal. Across the wing, at the junction of the marginal and submarginal, is a sloped indistinct bare area, and another broad “hairless line,” much better defined and roughly parallel to the first, crosses the wing from the end of the radius. This second track does not reach the posterior edge, before which one or two rows of the usual discal ciliation intervene. It is similarly cut off from the small clear spot outside the radius, ea. Bl > Ne a ma AAA Se <= Fig. 3. Paraphelinus xiphidii, wings. generally by one, sometimes by several, rows of bristles. The clear path shows scattered black dots, which are not the bases of broken-off bristles, but under an oil immersion are seen to be excessively minute prickles (rudimentary bristles 2) set in the wing membrane. Beyond the bare track the discal ciliation is regular and dense, coming evenly up to the margin everywhere, but tending to form an isolated isoclinal row parallel to the costa on the apical half. Marginal cilia on costa very short and dense, sparser and longer apically, longest on the distal two-thirds of the hind margin and shortened again after the frenulum. Hind wings not much shorter than the fore wings, with no submarginal cell, narrowed basally and very elongate (up to eight times as long as broad). Posterior fringe long, equal to the extreme breadth (at the hooks). Costal fringe moderate. There is a row of minute bristles along the costa and a distinct isoclinal row subcostal in position. Discal ciliation otherwise sparse, varying specifically. Legs. Fore legs: femur with no strong ventral bristles, except a (coloured) longer anterior one and a shorter (hyaline) posterior one, subapical in position ; A NEW SPECIES OF PARAPHELINUS, Hl tibia with the superior apical angle narrowed, chitinised, the apical comb hardly developed, with 2-3 spines. Mid legs: tibiae long, spur long and thin. Hind legs: coxae elongate, flat, oval; tibiae elongate, exceeding those of mid legs, spur short and not half the first tarsal joint. Tarsal proportions: in the fore legs the first and second tarsal joints are subequal (9:8), whilst in the mid and hind legs the first joint is much longer than the second (4:3); joints 3 and 5 are subequal in all the tarsi, 4 being a little shorter. Abdomen elongate, acuminate, seen from above at least twice as long as broad, from the petiole to the base of the ovipositor sheath. Dorsal surface smooth, broadly in the middle, except for a narrow scaly reticulate band across the middle of the first tergite. Towards the side this band becomes normally reticulate, while at the extreme sides and on the overlaps of all the tergites the cells are drawn out antero-posteriorly. Posterior tergal edges straight, on 5 and 6 a little convex posteriorly. Tergites 1-5 equal, 6-7 longer. (The first tergite being anteriorly concave appears shorter, but at the sides it is even longer than its successors.) Sternites 1 and 5 are subequal and long, 2-4 much shorter and subequal. Normal abdominal chaetotaxy :— 1, Id, Thi, IV. V. VI. VII, Tergites « Hin es 2,2 2, 2 ee ae 8 2, 2 4.4 Sternites .. 0 oe a, 6, 6 — — The bristles on tergites 1-3 are wide apart, one at the edge closely accompanied by the second. On tergite 5 the eight bristles form a continuous row with one in front and sometimes one behind at the sides. There are a few (3-5) minute post- spiracular bristles besides the longer transverse ones on tergite 6. The setigerous process on tergite 7 bears four bristles, one shorter. This tergite extends narrowly backwards above the base of the ovipositor sheath, but the bristles borne here are probably of specific value. On the fifth sternite the bristles are arranged 1, 2 and 3 on each side of the mid-line. The free portion of the sheath is rather long, and projects distinctly with the ovipositor. Male similar to female. The antennal formula the same, with the scape flattened more or less. Wings shorter and broader. Fore legs a little stouter; second tarsal joint relatively shorter than in the female. List of Species described in the Genus Paraphelinus, Perk., with Data of Host Attachment, original Collector, and Location of Type. 1. P. xiphidui, Perk. (1906) (Le., p. 264, pl. xx, f. 6). From eggs of Xiphidium varipenne, Swez.; Hawaii (O. H. Swezey). Type (female) in Mus. of Haw. Sug. Pl. Assn., Honolulu, Hawaii. 2. P. speciosissimus, Gir. (1911) (l.c., p. 181). Host unknown; Urbana, Illinois (A. A. Girault). Type (female) in U. 8S. N. M., No. 14,122, Washington, D.C. 3. P. australiensis, Gir. (1913) (l.c., p. 74). Host unknown; Nelson, Queensland (A. A. Girault). Type (female) in Queensland Museum, No. Hy. 1712, Brisbane. (C365) D2 52 JAMES WATERSTON. 4. P. tomaspidis, How. (1914) (Le., p. 82). From eggs of Tomaspis varia ; Trinidad (P. L. Guppy) Type (female) in U. 8. N. M., No. 18,321, Washington, D.C. 5. P. perkinsi, sp.n. From Tomaspis t carmodyt; British Guiana (C. B. Williams). Type (female) in British Museum, London. To the above should possibly be added Agonioneurus locustarum, Giraud (1), bred from eggs of Xiphidiwm fuscum in galls of Lipara spp. on Arundo phragmitis (Europe). Since sinking Paraphelinus, Perkins, Mr. Girault has with extreme brevity described the undermentioned yellow or pale-coloured species of Aphelinus from Australia. In 1913 (9) the following :—haeckelr, grotiusi,* darwini, minutissimus, newtoni ; in 1915 (11) he added miltoni* and ruskint. Of these the two asterisked species are probably referable to Paraphelinus, but only a re-examination of the types, which are deposited in the Queensland Museum, Brisbane, can enable one to place them exactly. No remarks have been offered in the notes just given on the general coloration of ~ Paraphelinus, but it is probable that the genus when more extensively collected and described will be found to include only pale forms like perkinsi, xvphidii, and tomas- pidis. Detailed colour notes have been given only for the first of these. The infus- cations of the abdomen, etc., are exceedingly faint, and when the body contents have been cleared away by potash the whole insect is perfectly transparent. Only in aiphidit are the genae distinctly darker ; in tomaspidis they are faint. The wings in xiphidii and perkinsi are hyaline, in tomaspidis faintly embrowned. The following keys may serve to separate the species :— Key to the Males of Paraphelinus, Perkins. First two funicular joints very short and broad, the third elongate, cylindrical, nearly equal to the club, which bears the longest bristles of the antenna xiphidi, Perk. First two funicular joints normal, cylindrical, the third normal, two-thirds of the club, which bears the shortest hairs of the antenna perkinsi, sp. 0. Key to the Females. 1 (2). Yellow and black species; forewings with a conspicuous median black band as wide as the marginal and stigmal veins... -. specrosissimus, Gir. 2 (1). Concolorous, the head and thorax reddish yellow, abdominal tergites at most a little infuscated laterally ; wings hyaline or uniformly faintly tinted. 3 (4). Deep orange yellow, immaculate; hing wing near tip with about 5 lines of | discal cilia .. oe fe Re australiensis, Gir. 4 (3). Paler species; genae, Oe aan ‘nid haeie of abdomen more or less infuscated. | 5 (6). “ Hairless line” of forewings Oe on anterior half, with anteriorly convergent sides .. Bs a's tomaspidis, How. 6 (5). “ Hairless line”’ crossed at most By one or Bed (is of discal cilia at level of the radius knob, with sub-parallel sides. | A NEW SPECIES OF PARAPHELINUS. 53 7 (8). Club just longer than pedicel, third funicular joint about twice (5 : 3) as long as the second dy ' xiphidiw, Perk. 8 (7). Club distinctly shorter Hoan ok ay shi Aecibedlar not ss need ia twice (4 : 3) as long as the second so : 7° perkinsi, sp 0.. The following notes refer chiefly to Hi more areas Iptagienisic differences between the females. It will be noted that the new species is more hirsute than the others. The males are so easily separated by the characters given in the key and figures that minor differences are not now described, but the dimensions of the antenna and wings have been stated. The legs of three of the species are compared in a concluding paragraph. Paraphelinus xiphidii, Perk. (1906). 9. Head: in profile (fig. 1, 6) the vertex and upper frons are about four-fifths of the depth ; where the frpns is swollen out to its greatest extent the length and depth are about equal. On the gena the reticulation is drawn out, and there are 12-15 short scattered bristles up to the insertion of the episternites. From above (fig. 1, a) about one-half as long as broad, the proportions across the vertex at the posterior ocelli are 6 : 11, and in front of the anterior ocellus, where the eyes are nearest, 5 : 12 ; wider than thorax (22:19). MReticulation of vertex transverse, more regular anteriorly, three minute bristles in ocellar triangle with 2 (1 : 1) stronger ones between each pair of ocelli, the bristles close to the ocelli. Behind the posterior ocelli the vertex bears six stouter bristles (2, 2, 2), the middle pair in advance of the lateral pairs, and there are also 6-7 minute bristles at each side on the rounded edge above the temples. Between the orbits, a lateral and an anterior ocellus are 4-5 bristles. The frons bears along the orbits (between the keel and the level of the anterior ocellus) about a dozen bristles, the lowermost fine hyaline, 4-5 of the uppermost dark and - stouter. Between each upper row of stronger bristles and the mid line are about 9 similar bristles (5,4). The greater part of the face below is bare and faintly reticu- late. Between and above the toruli 6-8 pairs of bristles, one pair above the clypeal edge and 3-4 (single) inside the genal keels. Antenna (fig. 2, 6): length, 37 mm.; bulla (3:1) over one-third (5: 14) of the scape. Scape + bulla equal to pedicel + funicle. Pedicel just over half (23 : 42) the scape; Ist funicular (8 : 6), 2nd (12:7), 3rd (20: 10), club (44:12). The club distinctly longer than the funicle (11 : 9), equal to (or if carefully measured, just longer than) the scape, and bearing 4-5 long sensoria. The scape with 4 ventral bristles, otherwise like perkinsi, except that there are only about a dozen bristles posteriorly. Thorax: pronotum with the spiracular emargination occupying about two-ninths of the lateral edge; between each stout bristle on the posterior edge and the mid line are two weaker ones, 2-3 minute on each side of the mid line, and 6-8 obliquely across each half. Pattern everywhere distinct, traceable to the mid line. Mesonotum (14: 15) half as long again as the scutellum (3:4). Mid sternal region distinctly separated from prepectus. Sternal pattern very faint but the cells large and sharply outlined when finely focussed. Propodeon nearly twice as long as the metanotum (11 : 6) which is coarsely trans- versely reticulate; pattern very faint medianly, more distinct towards the lateral ridges. Spiracle circular, with two minute bristles at the outer edge of the rim. 54 JAMES WATERSTON. Wings (fig. 3) about three and a third times as long as broad; length, ‘7 mm. ; breadth, 21 mm. (excluding fringe). Hind wings over eight times as long as broad ; length, ‘6 mm.; breadth, ‘07 mm. (excluding fringe) ; longest hairs of fringe as long as, or just exceeding the breadth. Abdomen twice as long as thorax and propodeon, and nearly thrice as long as broad (8:3). The 7th tergite long and narrow, much longer (10:7) than the preceding (spiracular) tergite. The ovipositor does not exceed the sheath, which projects shortly (one-third the length of the seventh tergite), the free portion of the sheath being about one-half the base. Above the fixed part of the ovipositor sheath there are on , ot $477! re ~ ~~ i ae re Cat nes - = -— [SA tae ce iy tag Ce TEHZI— Vig. 4. Paraphelinus xiphidu, 3: (a) antenna. P. perkinsi, 3: (b) antenna; (c) forewing. the apical half about 4 longer bristles with 3 above placed below and behind the setigerous process ;_ the free portion of the sheath bears a ventral row of 4—5 longer bristles with as many more (shorter) above and one or two (minute) at the apex. Length, about 1 mm.; alar expanse, 1°6 mm. | g.—Antenna (fig.4,a): length,°37 mm. Forewings: length, 64 mm.; breadth, ‘2mm. Hindwings: length, 54 mm.; breadth, ‘07 mm. Paraphelinus perkinsi, sp. nov. Vertex, upper two-thirds of frons, eyes, and entire dorsal thoracic surface, pale rusty yellow; the dark inner chitinous socket of the eye shining through ; ocelli, tips of mandibles, and ventral half of orbits (very narrowly) dark reddish brown ; genae infuscated a little. Wings faintly tinted, veins hardly darker than the membrane. Otherwise colourless transparent, with the following marks: Anterior A NEW SPECIES OF PARAPHELINUS. 55 tarsi faintly darker, all claws brown ; sterno-pleural sutures narrowly darkly outtined, a pale median dorsal line from behind the head to above the ovipositor ; pronotum and propodeon each with two or more or less defined triangular darker spots about the mid line. The tergites (except medianly) are faintly infuscated, so that the abdomen is indistinctly banded; on the 7th tergite (bearing the setigerous process) the band is darker, narrow, and distinct. Sternites a little infuscated ; ovipositor reddish brown. ©. Head from in front (7:6); frons narrowest (2 of the width) at two-thirds above the clypeal edge. Toruli well below the base-line of the eyes, three-fourths their own length from the mouth-opening and nearly twice (11 : 6) their length apart. Pattern oe Ses? i iene a Fig. 5. Paraphelinus perkinsi: (a, b) wings of 9; (c) left mandible of 9; (d) lett mandible of ¢. P. xvphidii: (e) labrum of 9. on vertex and frons (except just between the scapal grooves, where the surface is nearly smooth or finely mamillated) large, raised on vertex and upper frons, but very faint on the face; clypeus smooth. Along each lower orbit are 5-6 bristles and about the same number (strong) superiorly ; nearly a dozen smaller bristles between the orbital row and the mid line, on the upper angles of the frons ; 5-6 minute bristles inside the genal keel and 9-10 pairs of bristles between and above the toruli; the two clypeal bristles are well above the edge. Antenna (fig. 2, a): length, -42 mm.; bulla (2:1) five-eighteenths of the scape. The latter narrow (5:1), with 6 bristles along the ventral edge, 6-8 on the outer 56 JAMES WATERSTON. apical half, about 10 on the dorsal edge, and 20 on the inner surface evenly but sparsely disposed. Surface, especially on the inside, with large-celled little-raised reticulation. Pedicel (25: 11) scarcely half the scape. Funicle shorter than either club or scape, being about four-fifths of the former or three-fourths of the latter. First funicular (9: 7) (measuring to the ventral apical angle) equal to one-sixth of the scape. In the same scale, the 2nd and 3rd funicular joints are (12 : 9) and (16 : 13) respectively. The club (45: 15) is five-sixths of the scape and bears 5-6 sensoria. Bristles of pedicel sparse and evenly set, as on the scape; on the first funicular, round the insertion of the second and along the short dorsal edge are 6-7 bristles, and one on the inner ventral angle; from the 2nd funicular onwards the bristles are more closely set, shorter and fewer; on the club they are numerous and more appressed. Thorax: pronotum with the bristles weaker than in xiphidir, and an extra one on the posterior edge near the middle. Proportions much as in zvphidi, scutellum longer (5 : 6), mesophragma to scutellum as 17 : 10, and longer than mid lobe (17: 15). The four pairs of median bristles are in two nearly parallel lines, converging very slightly anteriorly. In xvphidi the first pair immediately behind the pronotum are a little nearer to one another, and the second pair (and sometimes the third) more outwardly displaced, so that the lines are not parallel. Propodeon: the pre-spiracular propodeal fold is distinctly shown only for a short distance anteriorly, i.e., beside the spiracle. Posteriorly there are only some indistinct longitudinal rugae ending at the side of the petiole. Spiracle relatively larger and broadly oval, with two minute bristles outside. Wings (figs. 5 a, 6): Forewings (3:1), length, °8 mm.; breadth, 26 mm.; broader than in xiphidw with two more macrochaetae on the marginal and 1-2 additional minute bristles below the submarginal distally (cf. also thé disposition of*the 6-7 rows of longer bristles below the marginal to that vein). The bare area beyond the radius smaller. Hind wings broader, not eight times as long as broad; length, ‘68 mm.; breadth, ‘09 mm.; numerous bristles at apex. Longest hairs of fringe about three-fourths the breadth. Abdomen: free portion of sheath much as in xiphidiw, but the long bristles are nearer the apex. Above the sheath the short bristles are very numerous, 14—16, the last (below the setigerous process) being longer. Length, ‘85 mm.; alar expanse, 1°8 mm. 3. Like the female, paler if anything on head and thorax, mandibles apically darker. The fore legs darker than in the female. Antenna (fig. 4, b): length, ‘28mm. Forewings (fig. 4, c): length, ‘54 mm. ; breadth, 18mm. Hind wings: length, 44 mm.; breadth, ‘07 mm. Paraphelinus tomaspidis, How. (1914). . Head: from in front, ciliation of eyes very short and sparse. Clypeal teeth distinct. Between and above the toruli three pairs of bristles, 2-3 bristles inside each genal keel; between the orbits and mid line at the level of the anterior ocellus merely 2-3 bristles. At the same level the frons is one-half the width (just wider than in xuphidir). A NEW SPECIES OF PARAPHELINUS. 57 Antenna (fig. 1, c): length, 3mm.; bulla over one-third of the scape (4: 11). The latter only a little over four times as long as broad (33:8). Bulla + scape longer than pedicel + funicle. Club (nearly 4:1) elongate, three times as long as the last funicular, and exceeding either funicle or scape. Bristles of scape much as in ziphidn, except that there are only 3-4 bristles on the anterior apical half, and about 10 posteriorly. Wings (fig. 6). Forewings: length, ‘5 mm.; breadth, ‘142 mm.; narrowed basally, narrower than in either xiphidw or perkinsi, about 34 times as long as broad. The oblique hairless tract with convergent sides and obscured on about the anterior half, with 10-11 bristles below the submarginal. Hind wings (9: 1) elongate, narrow and pointed; length, 45 mm.; breadth, 05 mm.; almost as long as the forewings, and nearly bare apically. Longest bristles of fringe exceeding the greatest width. Fig. 6. LP. tomaspidis, How., 9, wings. Abdomen not twice as long as thorax and propodeon. Ovipositor distinctly pro- jecting and exceeding its sheath. The free portion of the latter with two longer ventral bristles, two shorter above, and a number of minute ones at the apex ;_ there are 9 short bristles in a patch above the distal end of the base of the sheath. Length, ‘8 mm.; alar expanse, 1:1 mm. Mr. Williams’ example is a little smaller than the genotype, with which however it agrees in having the “ hairless line’ incomplete. The following is a comparative description of the legs of the three species here dealt with :— Fore legs: coxae of xvphidii with 10-11 bristles anteriorly (about 15 in perkinsi, 7 in tomaspidis); femur about 4:1 in perkinsi, narrower (5:1) in tomaspidis ; comb of first tarsal joint 13-14 in perkinsi, 9 tomaspidis, 10-11 xiphidw. Mid legs: 4—5 stiff bristles on coxa ; first tarsal joint with 34 spinose bristles on apical posterior edge. Hind legs: coxae bare, but for two bristles above the trochanter and one antero-ventrally ; femur (a¢phidir) anterior subventral row of 6-7 widely spaced bristles and two irregular subdorsal rows of 6-7 each, on apical two-thirds; in tomaspidis only 7-8 subdorsal bristles in all, none below one half; tibial comb in all three species with 8-9 spines. In the chaetotaxy of the legs xiphidii and perkinsi agree more closely, tomaspidis being a barer species. The first tarsal joint of the mid and hind legs is also rather 58 JAMES WATERSTON. shorter, the first and second being nearly 5 : 4, as opposed to 4 : 3 in the other species. In the figures of the wings special attention should be given to the shape of both fore and hind pairs and to the distribution of the minute bristles on the basal half. It is noteworthy that in these respects 2 tomaspidis approaches somewhat to 3 perkinsi. fone Bibliography. Giraud, J. Verh. Zool.-Bot. Ges, Wien, xiii, p. 1278. Swezey, O. H. Report of the Work of the Experiment Station of the Hawaiian Sugar Planters’ Association, Division of Entomology, Bull. no. 1, pt. 7, p. 214, December 1905. . Perkins, R. C. L. Ibid., Bull. no. 1, pt. 8, pp. 245 & 264, January 1906. . Swezey, O. H. Proc. Haw. Ent. Soc., ii, pt. 1, p. 21, October 1908. . Perkins, R. C. L. Fauna Hawaiiensis, ii, pt. 2, suppl., p. 658, 17th December 1910. . Girault, A. A. Jl. New York Ent. Soc., xix, p. 181, September 1911. . Mercet, R. G. Los Afelininos, pp. 32, 35, 45, 107, 108, 110, 290, 30th December 1912. . Girault, A. A. Arch. f. Naturg., Ixxix, Abt. A., pt. 6, p. 74, September 1913. . Girault, A. A. Mem. Queensland Museum, pp. 181-184, 10th December 1913. . Howard, L. O. Proc. Ent. Soc. Wash., xvi, p. 81, June 1914. . Girault, A. A. Mem. Queensland Museum, pp. 45-47, 4th June 1915. All the above (except 1) contain explicit references to Paraphelinus, but every described yellow or reddish species of Aphelinus will ultimately require to be studied before the limits of Paraphelinus are understood. 59 A NEW THRIPS DAMAGING ORCHIDS IN THE WEST INDIES. By C. B. Witxiams, M.A., F.ES. The thrips described below has recently been a source of considerable damage to orchids grown for ornamental purposes in Trinidad, West Indies. Up to the present it has not been found on wild orchids and it is possible that it is not indigenous to this island. A large number of orchids are brought over from Venezuela and it may have been imported with them from that country. Order TH YSANOPTERA. Family THRIPIDAE. Physothrips xanthius, sp. nov. Q. Measurements.—Total length, about0‘9mm. Head, length, 0-076 mm., width, 0-132 mm.; prothorax, length, 0-088 mm., width, 0°160 mm.; pterothorax, tf, OP ET = é YZ ote a Vf, VELL FES 3 thy GItttHtt} - greatest length, 0-180 mm., width, 0:200 mm. ; abdomen, width, about 0-180 mm. ; wing, length, 0580 mm., width of middle, 0-028 mm. Antennae 7, 1 2 3 4 5 6 7 8 length (uw)... 22 5 34 48 40 52 9 14 width (1) 2 29 24 18 7 15 13 6 4 ’ Total length of antenna, 0:290 mm. 60 C. B. WILLIAMS. Coloration.—Body and legs almost uniformly pale yellow, the fore part of the head very slightly darker. Antennae a little darker brown, except for the bases of the third and fourth segments, which are very pale. Eyes black; ocellar pigment reddish orange. Wings brown, distinctly darker than the body, with the central portion of the outer half a little paler than the margins. , Head not quite twice as broad as long; the cheeks not arched. yes large, very slightly projecting ; distance between the eyes greater than the width of the eye; distance from the eye to the back of the head about half the length of the eye. Ocelli on a level with the middle of the eye, the three forming an obtuse triangle ; the posterior pair about twice as far apart as they are distant from the margin of the eye; the anterior ocellus directed forward. Ocellar and post-ocular spines and a few smaller ones on the vertex, pale and indistinct; the ocellar spines arise from the sides of the ocellar triangle. Mouth-cone bluntly rounded, reaching about . two-thirds across the prosternum. Mazillary palps short, three-segmented ; the basal segment short, the two others subequal. Labial palps slender, apparently two-segmented. Antennae eight-segmented, relatively long, being three and a half times the length of the head: the first segment short and broad; the second barrel-shaped ; the third with a distinct pedicel and narrowed to a slight neck at the apex; the fourth widest at about half its length, but narrowing rapidly apically to a long neck; the fifth with a distinct ring joint at its base, truncate apically ; the sixth long and gradually tapering; the eighth half as long again as the seventh. Forked trichomes on the third segment dorsally and the fourth ventrally, that on the third segment reaching just beyond the base of the fourth, and that on the fourth reaching to about one-third the length of the fifth segment. Prothorax about one and two-thirds as wide as long, all angles rounded. Two pale spines at each hind angle of the pronotum each about half as long as the dorsal length of the pronotum ; short spines at the front angles, other small spines very inconspicuous. Pterothorax normal; at the front angle there is, more or less distinct, a small double tooth-like projection in front of the base of the fore wing, one portion of this is long and slender, the other short and broad. Legs normal. Wings fully developed ; veins on the fore wing almost invisible, only indicated by the position of the bases of the spines. On the costa there are 20-21 spines ; on the fore vein two groups of 3 near the base and 2 near the apex; on the hind vein 11-13. The anterior fringe commences just distal to the base of the hind vein; the posterior fringe extends a little further towards the base of the wing. Hind wings paler than the fore wings, with a rather broad dark, central vein distinct almost to the apex of the wing. The minute setae on both wings are dark and conspicuous. Abdomen normal; the ninth segment longer than the tenth; spines on the last two segments pale, those on the ninth segment about as long as the tenth segment, those on the tenth segment a little longer. No trace of comb-like structure on the hind margin of the eighth tergite. 6 unknown. Described from about a dozen females collected on Cattleya orchids in Port of Spain, Trinidad, on various dates during 1915 and 1916 by F. W. Urich and C. B. Williams. Type in the author’s collection. A NEW THRIPS DAMAGING ORCHIDS IN THE WEST INDIES. 61 This species is chiefly recognisable by its yellow body colour, brown fore wings and the relatively long antennae. The larvae and pupae are found with the adults on the upper surfaces of the leaves, particularly near the mid rib and towards the base of the leaf. They cause injury by piercing the surface, resulting in brownish spots or patches on the leaves. In severe cases the plant may be entirely killed. 8 . BTA Teaw gute ny 9. uM egaiw.cvol oword | soaldo yhod walloy, a, ydrel tle eee ss aa Wit toi bao! ‘ae soqpin edly xy: 2! lobal mht tone bgmsio} oY 6 fpar'yodT sol Seite wien aif ebiewod Bodedrebion add wovect ost sto > vodoiug WH Bloque Hsioword ab yalblganm balsa 9 : | he fist “Covites ads ean, ‘wah sl a ured arepsher Por, ti oleae ol m Wie so tey of i hogs Battin kis chm or me dae. | Newer abn at carols ah ive tee}. faze oraie lee “< c | nie be ails talons: yeent | Ge, Sn ee aft 7 dh Toe hs grew = te ; Teg th nes i pee aerate atu wrt réintively. lowe pe ate fant “ Z } dre) Amerowead pigivos dh Chan fenigt Shrine, am outiw 9k thir Penk aie hat ah vpecit t frout of thet ‘ubepslen, the , tht beiciielee rw ‘pine in chs genta . f the tee phen re ; G0 afi owns de » ic boka eee see be id | ee nee, Sh te mata oma deck, <t Abt » uci, Sie <= Goth wings A oe Abdomen toteaal He aengk ther the senth a sere: eae Ven ” 20 A peut, itp ev hu + : lop A ye / ides 3 > Ulam tend sb erty, Mage & utes a . rr) ye uted — al , See agin ss Gere Lipson ae Snes ihiahn, Aer teacln oma a ais aged Ol Viull aad ‘* Ps Type in. ibe » | i. u 4" ba ‘ + iG ies . ay st 7 e A ~ - Peas ‘ a ” alas 74 ; ' i . r - e wee « ee ‘4 a ' a we os NEW ETHIOPIAN FRUIT-FLIES OF THE GENUS DACUS. By Prof. M. Brzzi. Since the publication of my previous paper on the Ethiopian fruit-flies of the genus Dacus,* I have received some additional species, the descriptions of which are given in the following pages, with the purpose of completing our knowledge of this very important genus of Diptera. For the material here illustrated I am indebted to Dr. Guy A. K. Marshall, Director of the Imperial Bureau of Entomology, and to Dr. Alberto Mochi, of Ghinda, Erythraea: to whom I have to offer my best thanks. I. Subgenus Trrpacus, Bezzi, 1915. 1. Tridacus d’emmerezi, sp. nov. (fig. 1). Very distinct from any other known species of the present group, on account of the broad fuscous patch on the middle of wings, around the small cross-vein. $2. Length of body, 7-9 mm.; length of wing, 6-7 mm. Body of a uniform reddish colour, with less striking yellow and brown markings ; thorax and abdomen with short yellowish pubescence on the disk ; all the bristles of head and thorax black. Head of a more yellowish colour, opaque, but shining on the occiput, which has a paler border near the eyes; ocellar dot dark brown, on a brown transverse band; frons with a less developed fuscous middle spot, with almost indistinct dark dots on the sides and with two pairs of lower fronto- orbital bristles, which are not approximated; lunula reddish brown, shining ; face with shining antennal grooves, in the middle of which are placed the very small and rounded black spots; peristomial spot dark brown. Antennae much elongated, with the first jot about as long as the second; they are entirely reddish, the third joint being sometimes a little infuscated at the distal end; palpi very broad, pale yellowish, almost bare. Thorax with some irregular, blackish markings on the disk, forming two more or less distinct longitudinal stripes, which do not reach the scutellum, or two less distinct broader interrupted stripes along the dorsocentral lines. Yellow markings as follows: humeral calli entirely ; a narrow stripe along the transverse suture, broadly interrupted in the middle, and continued on the pleurae with the broader mesopleural band, which ends with a small spot on the upper border of the sternopleura; two broad contiguous hypopleural spots, which are margined inwardly, in front of the insertion of the halteres, with a conspicuous blackish patch. The anterior supra- - alar bristle is well developed, like the two posterior ones; the four scapular ones are long and strong; the pteropleural is well developed. Scutellum yellow, with the base narrowly reddish brown, and with a pair of long apical bristles. Mesophragma reddish, with the above-mentioned dark spot on the sides. Halteres whitish yellow. Abdomen elongate, rather constricted near the base; hind border of the second segment with a broad yellowish band, which is more or less interrupted in the middle; sides of the first. sometimes of the second and middle of the third segments provided with more or less distinct, but not sharply defined, dark brown * Bull. Ent. Research, vi., Sept. 1915, pp. 85-101, 14 figs. 64 M. BEZZI. or blackish markings ; venter reddish, the second segment with a yellow band. Third segment of the male with 8-10 long, black cilia at the sides of the hind border; ovipositor not longer than the last abdominal segment, reddish yellow, with the basal segment depressed. Legs of a prevalent pale reddish colour, but the coxae and the trochanters dark reddish-brown or blackish; the basal half of the femora and all the praetarsi pale yellowish. Wings (fig. 1) of teehee large size, with the veins reddish yellow; last portion of fourth vein gently bisinuous; second longitudinal vein long, the distance of its end from that of the first being not much shorter than the distance from that of the third; small cross-vein long, its lower end placed much after the middle of the discal cell; lower prolongation of the anal cell longer than the distance of its end from the hind border of the wing. They are hyaline, with a broad, pale yellowish fore border; which includes the stigma, but excludes the costal cells, and is not extended over the third vein, or only a little ; in contact with this border is a broad and more infuscated patch, surrounding the small cross-vein and encroaching more or less on the upper end of the discal cell, Anal stripe very broad and more infuscated than the marginal one. es, P -_— ee ae menses 2” Fig. 1. Tridacus d’emmerezi, Bezzi, sp. n. Type g and type Q, and some additional specimens, in the collection of the Imperial Bureau, from Mauritius, Reduit, 1916, collected by Mr. D’Emmerez de Charmoy, in whose honour this interesting species is named. II. Subgenus Dacus, Bezzi, 1915. In my Key to the species (l.c., p. 89) the section 21 (22) must be amplified as follows :— 21 (22). Face without black spots. a (6). Two contiguous hypopleural spots; third abdominal segment of male ciliated; wings with a brown anal stripe .. .. wmmaculatus, Coq. 6 (a). A single hypopleural spot; male with the third abdominal segment not cihated; wings without an anal stripe. c (d). rons with three pairs of blackish dots near the eyes; abdomen entirely black along the middle line; wings with the submarginal cell hyaline at base over the small cross-vein. .. annulatus, Becker. d (c). Frons with the basal half brown, the rie nae yellowish, and without black dots; abdomen with a broad, reddish yellow, longitudinal middle band, which is divided by a black line; wings with the sub- marginal cell broadly infuscated at base and with the upper end of the small cross-vein margined with fuscous .. a 2. mochu, sp. 0. NEW ETHIOPIAN FRUIT-FLIES OF THE GENUS DACUS. 65 2. Dacus annulatus, Becker, 1903 (fig. 2). An elegant species, well distinguished by the entirely black yellow-marked body, by the unspotted face, by the single hypopleural spot, by the unciliated third abdominal segment of the male, and by the lack of the anal stripe on the wings. The face is unspotted, but usually there is a more or less developed black or blackish spot on the upper edge of the clypeus, just below the base of antennae. All the bristles are black ; two pairs of distant lower fronto-orbital bristles; only the external scapular bristles are well developed, the middle ones being rudimentary ; no anterior supra-alar; pteropleural well developed. The sutures . of the abdominal segments are partly fused in the middle. The legs are pale yellowish, as described by the author; but in some ¢ specimens the apical third of the hind femora, and sometimes the whole of the hind tibiae, are black. Wings (fig. 2) with the second longitudinal vein short, the fourth segment of the costa being about one-third the length of the fifth; prolongation of the anal cell not longer than its distance from the hind border. Fig. 2. Dacus annulatus, Becker. The present species was collected by Ehrenberg in Egypt about a century ago and described rather recently by Becker from the original specimens; I have before me numerous examples from Ghinda, Erythraea, December 1916, collected by Dr. A. Mochi, who has found the species very common there. This species has already been recorded from Hrythraea (Asmara, leg. Kristensen) by Dr. Enderlein and by Dr. Speiser, both in 1911. 3. Dacus mochii, sp. nov. (fig. 3). Nearly allied to the preceding species, but at once distinguished by the very differently coloured frons and abdomen, and by the peculiar wing-pattern. g. Length of body, 55-6 mm.; length of wing, 45-5 mm. Occiput shining black above, with a narrow yellowish border, and pale yellowish below; ocellar dot black; frons with the basal half infuscated, the apical half pale yellowish, destitute of blackish lateral dots; lunula shining brown; face pale yellowish, immaculate, shining only along the antennal grooves; peristomial spot wanting. Antennae of moderate length, the first joimt not elongated; they are dark yellowish, with the third joint infuscated at the distal end; palpi and proboscis pale yellowish. The bristles of the head are black; there are two pairs of lower fronto-orbitals, one very near the dividing line between the face and frons, the other at the level of the dividing line between the dark and the clear portions of the frons. Thorax entirely black, even on the pleurae, with short whitish (C365) E 66 M. BEZZI. pubescence on the back. ‘The wax-coloured markings are as follows: humeral calli entirely ; a short and narrow stripe on each side of the transverse suture; a very broad, spot-like patch (about double as broad as the corresponding band of annulatus) occupying almost the whole of the mesopleura and the adjoining part of the ptero- pleura, ending with a small spot on the upper border of the sternopleura ; a single rounded hypopleural spot. Mesophragma entirely black. Scutellum wax-coloured, with a very narrow, black, basal stripe. Halteres whitish. All the bristles are black ; no trace of anterior supra-alars ; pteropleurals weak ; scapulars not distinguishable in the type; two scutellars. Abdomen ovate, punctate and pubescent lke the pronotum, a little constricted near the base; it is black, but has a broad yellowish band at hind border of second segment, like annulatus, and besides a broad yellow middle band extending from hind border of second segment to the end, where it is dilated and occupies almost the whole hind half of the last segment; along the middle of this yellow band there is a narrow black stripe, dilated at base and attenu- ated before reaching the tip; the sutures of the segments are partly fused and the third is not ciliated at the sides. Venter yellow, with dark or blackish sides ;_ geni- talia reddish. Legs with the coxae entirely of a pale yellowish colour, the ends of the femora and the four apical joints of the tarsi being a little reddish. Wings (fig. 3) = . ase" Oe es OT ec ere neon” Fig. 3. Dacus mochit, Bezzi, sp. n. hyaline, with brownish veins; second vein short; small cross-vein a little beyond the middle of the discoidal cell ; last portion of fourth vein gently curved at base ; prolongation of the anal cell proportionally short. There is a narrow blackish fore border, which leaves the costal cells hyaline, but is more intensively coloured on the stigma; below this, the fore border is dilated, filling broadly the base of the submarginal cell and surrounding the upper end of the small cross-vein; at its end the marginal band is a little dilated, but without forming a distinct spot at the end of third vein. There is no trace of an anal stripe, the anal cell being only a little yellowish on its upper half. Type 3, and an additional specimen of same sex, in the writer’s collection, taken near Ghinda, Erythraea, December 1916, by Dr. Alberto Mochi, in whose honour the species is named. 4, Dacus woodi, sp. nov. Exceedingly close to the preceding species (D. mochii), and resembling it in its unspotted face and unciliated third abdominal segment of the male, but differing in the colour of \ody and in the not infuscated upper end of the small cross-vein. NEW ETHIOPIAN FRUIT-FLIES OF THE GENUS DACUS. 67 $9. Length of body, 55-6 mm. ; length of wing, 45-5 mm. The species differs from M. mochii in the following points :— (1). The basal half of frons is yellowish, and therefore there is a very distinct dark transverse band on the middle; the lunula is pale yellowish, not brown. (2). The thorax is prevalently of a reddish, not black, colour, chiefly in front and on the pleurae; all the markings and the scutellum are of a bright yellow colour; the mesophragma is distinctly reddish on the sides, not entirely black. (3). The abdomen has the second segment more broadly yellow; the longitudinal yellow band is broader and has no middle black line in the male. (4). The wings have the base of the submarginal cell broadly infuscated, but this infuscation is not con- tinued over the upper end of the small cross-vein. The female is very like the male, but the abdomen has the second segment more broadly black, and there is a middle black line in the longitudiual yellow stripes. Ovipositor entirely reddish-yellow, shining, with the basal segment swollen, about as long as the last abdominal segment, but not much prominent. Type 3, type 9 and additional specimens of both sexes in the collection of the Imperial Bureau, from Nyasaland, Chiromo, Ruo. R., 22.1x.1916, collected by R. C. Wood, in whose honour the species is named. 5. Dacus hamatus, sp. nov. (fig. 4). A very distinct species, which in my key falls into section 25 (p. 90), but is dis- tinguished from imornatus and from any other known species by the exceedingly peculiar wing pattern, and by the longitudinal keel of the abdomen. 62. Length of body, 6-65 mm.; length of wing, 5-5-5 mm. Head entirely of a shining black colour, even on frons and face ; occiput with a narrow and short, spot-lke stripe on each side on the orbits; frons with a pair of yellow dots on the basal halt; lunula yellowish brown; the frons shows anteriorly a rounded prominence, which is peculiar to this species. Antennae dark yellowish, with the two basal joints very short, the second with some black bristles above, the third very elongate, blackish on the apical half ; palpi broad, of a reddish colour ; proboscis black. Cephalic bristles black ; two pairs of distant lower fronto-orbitals. Thorax entirely black, punctate, with a short dark pubescence on the back, which is paler at the sides. The very striking yellow markings are as follows: a rounded humeral spot, which covers only the hind half of the humeral calli; the fore half being black ; there is no sutural stripe, only the notopleural calli at the beginning of the suture being yellow; an oval spot in the middle line of the back, behind the suture; a broad mesopleural patch, extended on to the pteropleura and ending with a small spot on upper border of sternopleura; a single hypopleural spot Mesophragma entirely black, punctate, a little shining. Halteres pale yellowish. Scutellum yellow, with a narrow black base and with a pair of bristles. All the bristles are black; there is no trace of anterior supra-alars; pteropleurals well developed; only the external scapulars are strong. Abdomen rounded, convex, constricted at base ; along the middle line there is a very peculiar obtuse keel, which on the middle segments is rather prominent; it is clothed with dark pubescence, which at the sides of the base is greyish, erect and rather long. The abdomen is. (C365) E2 68 M. BEZZI. entirely black, opaque, punctate ; only the last segment is more or less reddish in the middle of the hind border; the segments are fused together in the middle, and the third of the male is not ciliated ; ovipositor shining black, smooth, with the basal segment swollen and of conical shape, not longer than the last abdominal segment ; apical segment short, reddish yellow. Venter entirely black; male genitalia dark brown, surrounded with yellowish. Legs with coxae, trochanters and femora shining black, but the middle pair with a narrow base, and the hind pair with more than the basal half, yellowish; tibiae yellowish, with blackish or darkened basal half; tarsi pale yellowish, with whitish basal joints. Wings (fig. 4) hyaline, with yellowish veins, which are darkened on the dark markings; second vein long, its distance from the end of the third being only 14 times longer than that from end of the first; third vein straight at the end and parallel with the last portion of the fourth, which is only a little curved near the base ; the lower end of the small cross-vein is placed at the middle of the discal cell; the prolongation of the anal cell is shorter than the remainder of the anal vein.* The dark pattern consists of the black stigma, the infuscated bases of the marginal and submarginal Fig. 4. Dacus hamatus, Bezzi, sp. b. cells, and a very peculiar, hook-shaped band, which passes over the small cross-vein, crosses the middle of the discal cell and ends sometimes with a short prolongation in the third posterior cell. There is no trace of an apical spot or anal stripe. Type g¢ and type 9, and some additional specimens of both sexes, in the collection of the Imperial Bureau, from Nyasaland, Chiromo, Ruo. R., 11.v.1916 (R. C. Wood). . The present species, with its characteristic wing pattern, seems to approach to the Oriental species with banded wings, for which I have preserved the genus Bactrocera in a restricted sense; these last species however are always provided with a pair of praescutellar bristles. 6. Dacus blepharogaster, sp. nov. (fig. 5). eee Falling into section 28 of my Key (p. 90), but very different from=riufus on account of its prevalently black colour ; it is besides distinguished from annulatus and mochit, to which it is allied in general appearance, by the black-spotted face and by the ciliation of the third abdominal segment in the male. g. Length of body, 7 mm.; length of wing, 6 mm. * All these peculiarities of the venation are to be found also in the nearly allied inornatus, Bezzi, which however has no trace of the longitudinal keel on the abdomen. NEW ETHIOPIAN FRUIT-FLIES OF THE GENUS DACUS. 69 Head yellowish ; occiput shining, with the vertex reddish with two black spots and with a broad blackish patch on each side; orbits yellow, broadened below ; ocellar dot black ; frons opaque, with three pairs of blackish dots on the sides and an undetermined fuscous patch in the middle; lunula shining black ; face entirely shining, with a broad rounded black spot on lower half of each antennal groove ; peristomial spot broad and black. Antennae entirely reddish yellow, the third joint a little infuscated at the end, the first joint much shorter than the second ; palpi and proboscis reddish yellow. Cephalic bristles black ; two pairs of distant lower fronto-orbitals. Thorax entirely black, slightly reddish on the sides in the hind half and along the pleural sutures, chiefly on the front half. The bright yellow markings are as follows: humeral calli entirely; a short triangular stripe along the sides of the transverse suture; a rather broad mesopleural stripe, ending with a spot on upper border of sternopleura ; a single hypopleural spot. The back is punctate and clothed with short greyish pubescence; mesophragma entirely black. Scutellum bright yellow, with a narrow black basal line. Halteres whitish. All the bristles are black ; no trace of anterior supra-alars; the middle seapulars are well developed, like the external ones; pteropleurals rather weak; a single pair of scutellar bristles. Abdomen in shape, colouring, punctuation and pubescence like that of annulatus; but it is less elongate, more oval; the hind one” Ae ews wt wae ereern” Fig. 5 Dacus blepharogaster, Bezzi, sp. n. border of the third segment has also a narrow yellowish band{; third segment with 6-7 black cilia at the sides of the hind border. Legs with the coxae pale yellowish, the middle and hind femora being dark reddish at the end; all the tibiae are broadly darkened, the posterior ones being entirely black ; | tarsi reddish brown, with all the praetarsi whitish. Wings (fig. 5) hyaline, with the fore border and the apical spot as in annulatus, but there is a distinct greyish spot before the end of the anal vein, which is entirely wanting in that species. ‘The veins are yellowish and their disposition is like that of annulatus, but the prolongation of the anal cell is much more developed, being longer than the rest of the anal vein. Type 3, a single specimen in the writer’s collection, taken near Ghinda, Erythraea, 7.xi.1916, by Dr. Alberto Mochi. 7. Dacus erythraeus, sp. nov. (fig. 6). Allied to the preceding species, but distinguished by the smaller facial black spots and by the non-ciliated third abdominal segment of the male; and also differmg from it and from annulatus and mochii by the broader black base of the scutellum and by the isolated blackish apical spot of the wings. 70 M. BEZZI. In my Key the present species falls ito section 27 (p. 90) together with oleae, which also occurs in Hrythraea,* but has no real affinity with it. $9. Length of body, 55-65 mm.; length of wing, 5-6 mm. Occiput black, rather shining, with narrow yellowish orbits, which are broadened below; ocellar spot shining black; frons opaque, with the basal half reddish brown and the apical half pale yellowish, but with three distinct pairs of broad blackish dots on the sides; lunula shining red or brownish; face pale yellowish, shining, the black spots being more stripe-like and placed on the inner border of the antennal grooves near the mouth-border; peristomial spot less darkened, but distinct. Antennae reddish, with the first jot short and the third infuscated at the end; palpi pale yellowish; proboscis reddish. Cephalic bristles black ; two pairs of strong and distant lower fronto-orbitals. Thorax entirely black, even on the pleurae, or only with a reddish stripe near the prothoracic stigma. Yellow markings as follows: humeral calli entirely, but with a black border above and in front; a narrow, triangular and short sutural stripe on each side; a rather narrow mesopleural stripe, ending with a very small and less distinct sternopleural spot; a single hypopleural spot, which is broadly margined with black below. Scutellum bright yellow, with a basal black band, which is much broader than Fig. 6. Dacus erythraeus, Bezzi, sp. n. in all the preceding species. Mesophragma entirely black. The back of the thorax is punctate and clothed with short whitish pubescence. The bristles are black ; no anterior supra-alars; scapulars short, the middle ones not distinct; ptero- pleurals well developed; two scutellar bristles. Halteres whitish. Abdomen elongate in the male, more rounded in the female, convex, punctate, with whitish pubescence, which on the sides is rather long. It is black; the second segment has a rather broad, pale yellowish, transverse band at the hind border, which is broadly interrupted in the middle ; fourth and fifth broadly reddish in the middle, forming a triangular patch with the base behind and provided with a middle longitudinal black line, which does not reach the hind border of the last segment. Venter yellowish, irregularly blackened towards the middle. Male with non-ciliated third abdominal segment and with shining red genitalia. Ovipositor short, as long as the last abdominal segment, but not prominent, reddish, with the basal segment rather inflated but not conical. Middle segments of the abdomen in both * F. Silvestri. Viaggio in Eritrea per cercare parassiti della mosca delle olive (Boll. del Labor. di Zool. gen. e agr., Portici, 1914, ix, pp. 186-226, 24 figs.). F. Silvestri. Contributo alla conoscenza degli insetti dell’ olivo dell’ Eritrea e dell’ Africa meridionale (l.c., 1915, ix, pp. 240-334, 78 figs.). NEW ETHIOPIAN FRUIT-FLIES OF THE GENUS DACUS. 71 sexes with the sutures partly obliterated. Legs pale yellowish; front femora brownish at the end; four posterior femora broadly black at the end; middle tibiae black at the base exteriorly, hind tibiae entirely black; praetarsi paler than the other joints. Wings (fig. 6) hyaline, with blackish veins; second vein short, the fourth segment of the costa being about one-third the length of the fifth ; third vein rather straight and parallel with the last portion of the fourth, which is less curved near the base; lower end of the small cross-vein placed after the middle of the discal cell; prolongation of the anal cell longer than the remainder of the anal vein. The very reduced pattern consists only of the black stigma, the infuscated marginal cell and an isolated blackish spot at the end of the third vein; there is no distinct grey spot at the end of the anal cell. Type ¢ and type Q, a single pair of specimens in the author’s collection, from Ghinda, Erythraea, 3.xu.1916, collected by Dr. Alberto Mochi. ; = s 4 is - . | (ee eee cea oe Poy?” UDA ROME a mol aR A IAT 102 overs . 4 2 , “Fs A ' pall: ap 2 G1 eis ‘, . f ‘ ) As &IGMIOT tf GT . {th reli J a. ay won | Jetarenitdo Yidredy ett O33 by oC". ite ome ce Isa id yap ord “TORT it. fOr A EME set t io oid pad Pat itetse oead edt = ) .» a er ae , ' ais S7ibu Eeerh (85 ae i . i “ee <that BY ban ies So Ci, poe @ pe Pres se ) 4 alti mer ir a | ae L a > is, wm, ¢ 7 aa’ 7 ” a ‘ NOTES ON SOME DISTINCTIVE POINTS IN THE PUPAE OF WEST AFRICAN MOSQUITOS. By A. Incram, M.D., C.M., and J. W. Scorr Macriz, M.A.,. D.Sc., West African Medical Staff. Wesché employed the relative length of the trumpets to the length of the thorax and the shape of the openings of the trumpets in drawing up a Key to the Pupae of West African Culicidae (Bull. Ent. Res., i, pp. 18-19). Howard, Dyar and Knab in their valuable monograph ‘“‘ The Mosquitoes of North and Central America and the West Indies” (Vol. i, pp. 102-103) state that in their experience too much has been made of the difference between the pupal trumpets of Anopheles and Culex; the variations in the shape, size, and length of the breathing trumpets between different species are numerous, they say, but “furnish no characteristics which are diagnostic of genera or larger groups.” They consider that there are intermediate forms and that differences between species are frequently more striking than between genera themselves. They remark also that this may equally well apply to the differences in shape of the pupal paddles, admitting, however, that “there is a striking difference between the two tribes CULICINI and SABETHINI”’ as regards the paddles for ‘‘ In the Cutictni the paddles are large, broad and rounded in outline; they are strengthened by a stout longitudinal mid- rib which bears a spine or seta apically. In the SaBeTurni the paddles are much smaller, narrower and tapered to a point; the midrib is either absent or poorly developed and there is no spine or seta.” It may be remarked here that in Eretmopodites, the only genus of the tribe SABETHINI up to the present known in Africa, the midrib is well developed in the pupal paddles and there is a long terminal seta, much longer than in the Cuicint. Howard, Dyar and Knab believe that “the greatest diversity among the pupae of mosquitoes will be found in the number and arrangement of the setae on different parts of the body. These, we are sorry to say, have not been adequately studied, but we are convinced that they will furnish both generic and specific characters”; and they point out that “there is a well-marked difference in the two tribes above-mentioned. In the SABETHINI the seventh and eighth abdominal segments each bear a pair of ample fan-shaped tufts at the apical angles; these are absent in the CuLIcINI.” Bacot takes up the question of the “ possibilities of the pupal paddles (anal _ plates of Wesché) for purposes of classification and the separation of ill-defined species’ of mosquitoes in his “Report of the Entomological Investigation undertaken for the Yellow Fever (West Africa) Commission ”’ (pp. 140-146), stating that “very marked differences may occur in the pupal paddles of closely related species.” Bacot qualifies this statement, however, by remarking that “there seems .. . no likelihood, so far as.the scanty material examined goes, that these characters could be made to afford the basis for an independent scheme of classification.” In the opinion of Bacot those appendages of the pupa which are obviously likely to be adaptive rather than ancestral in character may form a means of discrimination. 74 A. INGRAM AND J. W. SCOTT MACFIE. Following Bacot we have described the pupal paddles of a few mosquitos which, so far as we know, have not yet been illustrated in this particular manner. It seemed to us, after reading Bacot’s paper, that, while helpful in the differentiation of certain species, a scheme of classification based solely upon the characters of the pupal paddles with their appendages would break down sooner or later. We must admit, however, that this break-down has not occurred so far as the examination of our materials has gone, yet the points of distinction are sometimes so slight that we believe that not only the numbers and arrangement of the setae on different parts of the pupal case should be taken into consideration, as suggested by Howard, Dyar and Knab, but also in certain cases the length and shape of the pupal trumpets, in attempting to draw up a scheme of classification based upon pupal characters alone. Such a provisional scheme we have outlined below for the pupae which have been figured by Wesché and Bacot, and for those we ourselves have had the good fortune to examine. Unfortunately we have not been able to avail ourselves of all the descriptions given by Wesché, as some of them are incomplete and his drawings, which were made at a low magnification, do not reveal certain minute points on which we have had to rely for the differentiation of species; this is especially the case in the genus Culex, in which we have had reluctantly to abandon the attempt to include his species in our scheme. In drawing up this scheme we have relied as far as possible on the characters of the paddles and the setae on the last two abdominal segments. These structures furnished sufficient points for the differentiation of genera. In dealing with species, however, we have had to have resource to other characters in order to distinguish even the small number of pupae discussed, and it is probable that when further ’ materials are available the setae on all or most of the abdominal segments and every other structure capable of giving assistance will have to be included in devising a complete scheme of classification. Table of Tribes. 1. Paddles small, breadth less than half the diameter of the 8th segment; tufts on the 8th segment as long as the paddles or nearly as long a ny “i bs ire SABETHINI. Paddles relatively large; tufts on the 8th ay a shorter than the paddles z ag ees ee 2. Large species; long dorsal tent ee on Sep ith aM sth segments .. a - ” af .. MEGARHININI. Smaller species; without such et on = ; a bi iia aces 3. Dendritic seta on the posterior angle of the 8th Socaechy is ANOPHELINI. Tuft or differentiated seta on the Pee is angle of the 8th segment .. Ki . CULICINI. The only species ae to mn tebe ‘Bacamecen: the pupa of which has been. described is Toxorhynchites brevipalpis, which has been figured by Bacot. The large size of the pupa is at present sufficient to differentiate it from any other, but the insignificant size of the tuft on the 8th segment, the absence of terminal setae on the paddles, and the presence of long serrated dorsal setae on the 7th and 8th segments are also characteristic. THE PUPAE OF WEST AFRICAN MOSQUITOS. 75 The pupae of Eretmopodites, the only genus of the tribe SaBeTHINi at present known to occur in Africa, differ from those of the SABETHINI described by Howard, Dyar and Knab, as has already been pointed out, in having well-developed midribs in the paddles, and long terminal setae or tufts of setae. The five species described, including one described by Bacot which is known only in the pupal stage, can readily be distinguished, as shown in the following table. Genus ERETMOPODITES, Theo. 1. Single stout seta on the end of the paddle .. f . mi ¢ 2 Tuft of about a dozen setae on the end of the paddle .. HF. quinquevittatus. Tuft of about half a dozen setae on the end of the paddle .. E. inornatus. 2. A single stout seta on the posterior angle of the 8th segment. . H.?(undetermined). Tuft of long setae on the posterior angle of the 8th segment .. :. ‘3 3 3. Tuft of four setae on the posterior angle of the 7th segment .. —-E.. chrysogaster. Tuft of six setae on the posterior angle of the 7th segment .. EH. dracaenae. Genus ANOPHELES, Mg. The pupae of only four West African species of this genus have hitherto been described in this way, namely those of A. costalis, A. pharoensis, and A. mauritianus by Wesché, and A. funestus figured by Bacot. The latter author also figured A. costalis. The pupa of A. marshalli is described here for the first time, and further details are given of those of A. pharoensis and A. mauritianus so as to make the descriptions of all the five species comparable. Anopheles marshalli, Theo. The pupal paddles are somewhat narrow, with a marginal fringe on the outer side. The thickening of the outer margin of the paddle or buttress, described by Wesché as present in the Anopheline pupae he examined, is also present in this species, but is less well developed than in A. costalis. There are two setae near the distal end of the midrib, the more distal one being the longer and stouter. The latter seta is not bent into a hook, as in the case of A. costalis and A. funestus in Bacot’s figures, but is gently curved with the convexity inwards. The seta at the posterior angle of the 8th segment consists of branches coming off a main stem ; the main stem is, however, much shorter than that of the lateral hairs of A. costalis. The 7th segment carries no lateral tuft, but bears a single seta at its posterior angle, which is simple, and internal to it there is a longer but more slender seta, also simple. On the posterior border of the 5th and 6th segments there are similar setae, but the inner setae are usually double, and between them and the setae at the angles there are smaller double hairs, which, however, are larger than the similar hairs on the 7th segment. At the posterior angle of the 4th segment there is a small knob-like seta. On the posterior margins of the 2nd, 3rd, and 4th segments there are several single, double and triple hairs, none of which however are strongly developed, 76 A. INGRAM AND J. W. SCOTT MACFIE. Anopheles pharoensis, Theo. (fig. 1). The paddles are pyriform, with a slight external fringe, and have well developed midribs; they have also an external buttress, which is less developed than in A. costalis. There are two terminal setae on each paddle, a large and a small one, as in the other species of this genus. The large hair is not bifid, though the small one, which is poorly developed, may be split at its end. These hairs lie further apart than do the similar hairs in A. mauritianus, and the larger hair is longer than the larger hair on the paddle of A. mauritianus (9 units to 7). There is a stout irregularly branching dendritic seta, much like that of A. costalis, at the posterior angle of the 8th abdominal segment; it is about one-fifth the length of the paddle. At the posterior angles of the 5th, 6th, and 7th segments are stout, curved and sharply pointed setae, longer than the similar setae on the segments of A. mauritianus. At the posterior angles of the 3rd and 4th segments are curiously stunted and blunted setae. Internal to the seta at the posterior Fig. 1. <Ancpheles pharoensis, Thec,* angle and at the posterior border of the segments on the dorsum is a tuft of hairs upon the 3rd to the 6th segments ; these hair-tufts have fewer hairs than the similar tufts on A. mauritianus ; on the 7th segment this hair-tuft is replaced by a single or bifid hair. Nearer the middle line of the dorsum on either side upon the posterior margins of the 5th, 6th, and 7th segments js a single stout hair extending to the distal margin of the segment over which it projects ; this hair is replaced by a tuft on the posterior border of the 3rd and 4th segments. The trumpets have wide apertures, which are not fluted. Anopheles mauritianus, Grp. (fig. 2). The paddles are large and pear-shaped, with a slight fringe on their outer margins; they have well-developed midribs and also buttresses, which are slighter than those of A. costalis. The terminal hairs on each paddle are a large hair, which is usually bifid, and a small one, which may have its end split into three or four pieces. The hairs are slightly curved, but not hooked. At the posterior angle of the 8th abdominal segment is a stout seta giving off delicate branches only upon its inner aspect; these branches all extend to about * Unless otherwise stated the magnification of all the drawings is x 50. THE PUPAE OF WEST AFRICAN MOSQUITOS. 77 the same distance, namely to the termination of the seta itself, so that the branches coming off near the base are relatively long, while the others diminish in length progressively towards the apex—the appearance produced being suggestive of a harp. There are short single stout setae with pointed ends at the posterior’ angles of all the segments proximal to the 8th, except the Ist. These setae gradually increase in size as the posterior end is approached. On the dorsal surface and at the posterior border of the segments immediately internal to the seta at the posterior angle there is a tuft of simple hairs on the 3rd to the 6th segments; on the 7th segment this tuft is replaced by a single or double hair. On each side nearer the middle of the dorsum on the posterior border of the 6th and 7th segments is a single or double stout hair extending slightly beyond the middle of the segment over which it projects; on the posterior border of the 3rd, 4th, and 5th segments this single hair is replaced by a tuft. The trumpets have wide apertures, which are fluted. Fig. 2. Anopheles mauritianus, Grp. With such a small amount of material it is scarcely worth while making out a table for identification, especially as there is no difficulty in separating the pupae, but a few of the many points of distinetion may be briefly mentioned. The buttress on the outer margin of the paddle is conspicuous in A. costalis, less well developed in A. pharoensis, and in A. mauritianus, as Wesché states, it is less _ well marked than in A. pharoensis. In A. marshalli it is only slightly developed. Unfortunately Bacot’s figures do not show the buttress in either A. costalis or A. funestus. The longer terminal seta on the paddle is “reminiscent of a boot-hook” in A. costalis and A. funestus, according to Bacot ; in A. marshalli it is gently curved ; in A. pharoensis it is almost straight, and in A. mauritianus it is usually bifid, The dendritic setae on the posterior angles of the 8th abdominal segment also differ in these species. In A. funestus they are “larger, with less spread of the branches” than in A. costalis, according to Bacot. In A. marshalli they are smaller than in A. costalis, their length as compared with the length of the paddle being as 7 to 35 in the former, and as 12 to 36 in the latter. In A. mauritianus they are harp-shaped. 7& A. INGRAM AND J. W. SCOTT MACFIE, Tribe CULICINI. Table of Genera. 1. Paddles of unusual shape. . os ets f. if “ hei #2 Paddles of usual shape, oval or eee we) sé $ wT 3. 2. Paddles narrow, without fringe; end of oar, meta- morphosed into a spine Sh Mansonioides. Paddles “ knife-like,” inner lobes — chink bis ate: 1g Uranotaena. 3. End of midrib thickened and tube-like Es .. Humelanomyia. End of midrib not thickened, terminal setae Hacialy ate in phe 4. Terminal setae long is iy om P? ye ~ “as ‘emcaidee Terminal setae Peart e 3 oe me pee 5. Terminal setae very long; stout chore wre Fess a very long central branch a little internal to the tuft on the 7th segment .. eG a oe ce eh a“ st Aédomyia. Terminal setae shorter; no such dendritic seta on the 7th segment .. ne Aédes. 6. Tuft oh the 8th segment nell no > tuft on the ‘Ith; ‘pda more or less infuscated. . ve prs Mimomyia. Tufts on the 7th and 8th segments stall aaa we bs ener 7. Long double hairs arising near the eyes <n = of Culiciomyia. No such hairs = io am s a.) oP es ss Culex. In the above table no attempt has been made to separate the genera Stegomyia and Ochlerotatus, but this will be referred to later when dealing with Aédes. Culiciomyia nebulosa possesses the two conspicuous hairs arising from a point near the eyes described by Wesché as present in the species he knew as C. freetownensis and C. cinerea, both of which have been sunk by Edwards. We do not know if these hairs are present in other species of Culiciomyia, but as we have not observed them in any species of Culex it is possible they may prove to be a distinguishing feature. . Genus Apres, Mg. The only species of Ochlerotatus that we have had the opportunity of examining differ from Stegomyia in having a small dendritic tuft on the 2nd segment similar to, but much smaller and more delicate than, that on the Ist segment. It may be possible therefore to differentiate these genera as follows :— Tuft on the Ist abdominal segment dendritic, almost no stem; similar but smaller tuft on the 2nd segment .. #3 .. Ochlerotatus. More decided stem to the dendritic tuft on the lst segment; no similar tuft on the 2nd .. ‘i ee > se aty Stegomyia. Genus SteGomyi1A, Theo. ¢ 1. Tuft of setae on the posterior angle of the 7th segment .. te a ieee Single or double seta on the 7th segment nA Pr me ie ae at 2. Paddles without a fringe .. oi Aes 20 ‘i ps S. sugens. Paddles with a fringe ne “fe y = bas G a oe THE PUPAE OF WEST AFRICAN MOSQUITOS. 79 3. Stout branched seta at the posterior angle of the 6th segment, fringe shorter... t? Ay re S. metallica. No such special seta on the 6th segment ; ier ea midrib more highly chitinised .. ne ey .. _S. luteocephala. 4, Paddles narrow; single seta or a poorly developed tuft on the 8th segment oe .. 8. unilineata. Paddles broader ; tuft on ‘is sth Perv well Gorioped es Be eee 5. Long fringe on the paddles ~ a oe Sp ce S. africana. Short fringe on the paddles = s a “i op oe 6. Paddles with distinct notch an 4 = af ee S. simpson. Paddles without distinct notch .. : ap Roe. Ak 7. Single seta on the 7th segment ; tuft on the sth ‘i Ech aay hairs +9 ist .. S. apicoargentea. A single or Reaslly a ‘datible seta on the 7 th get tuft on the 8th with only one or two hairs subplumose aa i S. fasciata. The tufts and setae of Stegomyza are variable both in different individuals and on the two sides of the same pupa, so that it is necessary to examine more than a single specimen in those cases where these structures are used in differentiating the species. We have not had an opportunity of examining the pupae of S. sugens, S. africana, and S. apicoargentea but have relied on the descriptions by Wesché for the differ- ential points. Stegomyia luteocephala, Newst. The paddles are oval and rather narrow, but not so narrow as those of S. unilineata, they have a long fringe and a highly chitinised midrib, and the terminal seta is well developed. There are tufts at the posterior angles of the 7th and 8th segments, those ofthe 8th segment being the larger. These tuits vary in size considerably, and those on the 7th segment may be reduced to stout bifid hairs in some specimens, but as a rule they are well developed on one or other side of the body. On the 8th segment there is a single simple hair internal to the tuft, but the segment is otherwise almost bare. All the other segments carry a number of small setae, especially at their posterior margins, which are mostly simple, although a few are bifid; none of these hairs is conspicuously or strongly developed. The dendritic tufts on the first segment are not very strongly developed and have only a small number (about 7) of primary branches. Stegomyia metallica, Edw. (fig. 3). The paddles are oval, with a short fringe, a well-developed midrib, and a long terminal seta. At the posterior angles of the 7th and 8th segments there are setae differentiated as tufts of subplumose hairs; those on the 8th segment being small but definite tufts, those on the 7th being little more than strong double or triple branched hairs. The 8th segment is bare but for the tuft at each angle and a small simple hair internal to it. On the 7th segment there are, in addition to the tufts, _a number of simple hairs, mostly situated on the posterior margin, and a pair of 80 A. INGRAM AND J. W. SCOTT MACFIE. triple hairs. The 6th segment has at each posterior angle a strong bifid seta, on its posterior margin a number of simple and one pair of bifid setae, and a few small hairs further forward. The 5th segment has a single stout hair at the posterior angle and a number of small hairs near the posterior margin, one of which is larger than the others. The setae on the 4th segment are similar to those on the 5th, but the long hair is more highly developed. On the posterior margin of the 3rd segment, in addition to small hairs, there are on each side of the mid line a double or triple << SS 2S a Ne <—S cs Fig. 3. Stegomia metallica, Edw. and a long simple seta. On the 2nd segment these setae are more highly developed and the inner one is composed of four branches. All these setae on the 2nd to the 8th segments are more highly chitinised and better developed than those of S. luteocephala. The dendritic setae on the lst segment are well developed and have about a dozen primary divisions. Stegomyia simpsoni, Theo. The paddles are broader than those of the two species just described and show a distinct notch at the point where the midrib ends; they have a slight fringe; the midrib is well-developed, and they carry the usual long terminal seta. On the 8th segment there is a small tuft at the posterior angle, some of the hairs of which are subplumose, and a single or bifid seta internal to it. The 7th segment bears a few quite small setae, mostly at its posterior margin, and at the posterior angle two larger setae, one of which is larger and stronger and is situated a little anterior to the other. The 6th segment shows similar hairs and one strong seta a little above the posterior angle. On the 5th segment the small hairs are mostly bifid and sometimes triple, the strong seta above the posterior angle is present, and in addition there is a long, strong seta on the posterior margin a little internal to the angle. The 4th segment bears similar hairs, but the strong seta above the posterior angle is smaller, and the long seta internal to the angle is more highly developed. On the 3rd segment the seta above the posterior angle is reduced still more, and on the posterior margin internal to the angle there is a small tuft and then a seta of moderate size. On the 2nd segment there are a few simple and divided hairs, none of which is specially developed. The dendritic hairs on the lst segment are well-developed and show more than a dozen (about 16) primary divisions. THE PUPAE OF WEST AFRICAN MOSQUITOS. 81 Stegomyia unilineata, Theo. (fig. 4). The paddles are narrower than is usual in the other members of this genus, the ratio of length to breadth being on the average as 1°6 is to 1; they have a stout midrib, a well developed fringe, and a long terminal seta. The abdominal setae are very variable in this species both in different individuals and on the two sides of the same specimen. *The 8th segment is almost bare but carries at each posterior angle a stout seta, and internal to it a small hair. The seta at the posterior angle is well developed and may be single, double, or composed of three or four elements : it is subplumose. At the posterior angles of the 6th and 7th segments there are similar stout setae, but anterior to this there is little or no differentiation of the setae in this position. The seta at the posterior angle of the 7th segment is pubescent and may be single or double, that on the 6th segment is much smaller, simple, and either single or bifid. All the segments from the 2nd to the 7th bear numerous setae, the majority of which are situated on the posterior margins. Most of these Fig. 4. Stegomyia unilineata, Theo. setae are small, but two series are rather more highly developed. The one series is situated on the posterior margins about half-way between the mid line and the posterior angles and is represented by a long, single seta on the 6th and 7th segments, a long single or double seta on the 4th and 5th segments, and a long single seta on the 2nd and 3rd segments. ‘The second series lies slightly internal to the first, and the setae composing it are developed into tufts, which gradually become smaller as the posterior extremity is approached. On the 2nd segment the tufts are usually well developed, on the 3rd and 4th they are smaller and consist of only four or five hairs, on the 5th and 6th they are reduced to triple hairs, and on the 7th they are represented by single or double hairs not specially differentiated from the other small setae on the segment. The dendritic setae on the Ist segment are well developed and have a large number (over a dozen) of primary divisions. The above description is based on the examination of five pelts of pupae of this species. Genus OcHLEROTATUS, Arrib. 1. Two small dark spots on the margin of each paddle... di O. sudanensis, No such spots de a sig $y. rm she i’ iv iat 2 (C365) ‘ 82 A. INGRAM AND J. W. SCOTT MACFIE. 2. Single seta at the posterior angle of the 8th segment .. .. O. domesticus. Tuft on the 8th segment .. FP are 3. Tuft on the 8th segment long, half fhe sesigth of chs nsidille ormore .. thky a Tuft on the 8th segment short, less than half the length of the paddle .. 5 4. Paddles with a fringe and a single long terminal seta .. * .. O. apicoannulatus. Paddles without a fringe, short bifid terminal seta... as O. simulans. 5. (Paddle plates “run at sharper curve from the ribs ’’) .. O. mgricephalus. Paddles of usual shape .. # :% Fy oh ue Be ENG 6. Paddles with a fringe “at ge ny =: .. O. albocephalus. Paddles without a fringe .. ?. 4a Ae * ng O. minutus, O. punctothoracis, O. caliginosus. This table is given with considerable diffidence, as we have ourselves examined only two of the species. Bacot figures three, namely, 0. simulans, O. aprcoannulatus and O. minutus, O. sudanensis was briefly mentioned by us in a previous paper (Bull. Ent. Res. vii, p. 7), and the others were described by Wesché. 0. albocephalus, Theo., has not previously been described, or figured. Ochlerotatus albocephalus, Theo (fig. 5). The paddles are oval in shape and show a well marked midrib, a slight external thickening or buttress, a slight but distinct external fringe, and a long terminal seta. The terminal seta is one-fifth the length of the midrib and in this respect resembles that of O. minutus as figured by Bacot. The 8th segment bears at each posterior Fig. 5. Ochlerotatus albocephalus, Theo. angle a small tuft of about four simple or pubescent hairs, which is just about the same length as the terminal seta on the paddle, namely, one-fifth the length of the midrib. Internal to the tuft, close to the outer margin.of the paddle there is a long, slender, single or bifid hair. At the posterior angle of the 7th segment there is a bifid seta and a few small hairs, and along the posterior margin of the segment are a number of delicate and rather long hairs. On the 4th, 5th and 6th segments one of the setae on the posterior margin and a little internal to the angle is especially de- veloped, so that it reaches right across the following segment. This seta may be single or double. The dendritic setae on the lst segment are well developed. On the 2nd segment there is also a pair of dendritic setae, which are, however, much smaller than those on the 1st segment and but poorly chitinised. THE PUPAE OF WEST AFRICAN MOSQUITOS. 83 Genus MANSONIOIDES, Theo. The early stages of Mansonioides uniformis are not known. The pupa of M. africanus can readily be distinguished from all other known West African pupae by the structure of the trumpet, which is developed at its distal end mto a spine suitable for piercing the roots of the water-weed Pistia stratiotes, but it can also be separated from pupae of other genera of the tribe CuLicrni by the characters of the paddles alone. The structure of the paddles has been figured by Ingram (Bull. Ent. Res. in, p. 377) and it is only necessary here to add that the ends are slightly thickened and that apparently there are no terminal setae. At the posterior angles of the 7th and 8th segments there are large single setae in place of tufts. Genus AipomyiA, Theo. Aeédomyia africana, N.L. (fig. 6). The paddles are pyriform, slightly notched, and without a fringe ; the outer margin is a little thickened and the midrib is well developed. There is a single stout terminal seta at the end of the paddle about as long as the paddle itself. In the middle line and at the distal end there is a slightly darkened patch. Fig. 6. Aédomyia africana, N.L. The tufts at the posterior angles of the 8th segments are well’ developed and consist of four branched hairs. On the 7th segment there is a stout double seta at the posterior angle, and internal to it another tuft formed by the branching of a stout seta the median element of which is extremely long. Similar long tufts, or more correctly dendritic setae, are also present on the 4th to the 6th abdominal segments. Internal to these, and in a similar position on the Ist, 2nd and 3rd seg- ments also, there is a tuft of slender hairs. The dorsal tuft on the lst abdominal segment is poorly developed. There are several tufts on the thorax, one of which lies just behind the trumpets and another at the posterior margin. The trumpets are not very long, but are highly chitinised. Genus CuLEex, L. 1. Tuft at the posterior angle of the 8th segment containing branched hairs... cm aid - - extordl Tuft on the 8th segment of cn Aiea ee aie ai te ceive 2. Tuft at the posterior angle of the 7th sal pe et j branched hairs 3 ce Tuft on the 7th segment of sea ttt or Taal ed és - fee (C365) F2 84 A. INGRAM AND J. W. SCOTT MACFIE. 3. Infuscated spot on the paddles... r + ai .. C. quasigelidus. No such spot .. i ws bie 52. 4. A triple hair at the — sneie of ‘the bth aint 6th segments .. ..C. thalassvus. A tuft of five or six haat a Hide paseenen sable: of she bth and 6th segments .. .. _C. tritaenorhynchus. ‘5. Large species; length of thi pivians also 0: 9 mm. (70 units) es Ah a mA .. C. tugripes var. fuscus. Small or medium- ined pesies * 6. Stout pupa, well chitinised; tubal aperture large (19 units to a trumpet length of 42 units)... : : ” C. fatigans. Slender pupa, poorly chitinised ; tubal apnea smaller (10 units to a trumpet length si 35 units) .. s .. C. wnvidoosus. 7. Tuft on the 8th segment long, more than half the length of the paddle (19 to 34); trumpets short ate Cyathomyva fusca. Tuft on the 8th segment shorter, less than half the length of the paddle (8:42); a pair of long hairs on the distal margin of the 6th segment ; trumpets larger .. NB C. quart. In drawing up this table it has been necessary to take into account certain minute characters which unfortunately has made it impossible to include some of the species described and figured by Wesché in insufficient detail. It has also been necessary to re-describe C. fatagans and C. invidiosus in order to bring out these points. The branched or subplumose character of the hairs composing the tufts at the posterior angles of the 7th and 8th segments is not a very satisfactory differential feature and has been adopted in the table only because it was employed by Wesché. Four of the species have not previously been described. It may be noted as a point of possibly generic importance that none of the species of Culex that we have examined has had a fringe on the pupal paddles. Culex fatigans, Wied. (fig. 7). The pupa is stout and well chitinised. The paddles are large and slightly pyriform, with a well developed midrib and a slight thickening of the external border, but without a fringe. At the distal end of the midrib there are two small setae situated close together, the one being rather longer than the other. At the posterior angle of the 8th segment there is a tuft of about 6 to 8 branched hairs. At the posterior angle of the 7th segment there is a group of three setae ; the one nearest the angle is a small tuft of 4 or 5 delicate hairs, just above itis a larger tuft of 3 to 5 stout hairs, which are usually subplumose but may be branched, and internal to this seta and a little above it is a small double or triple hair. Several other hairs are situated along the posterior margin of this segment, most of which are small and double, but one particularly long single hair may be found near the angle. The 6th segment bears just above its posterior angle a long delicate triple hair, on the posterior margin a little internal to the angle a very long double hair, and just internal to this a tuft of 4 hairs. Similar setae occur on the 5th segment, but the tuft internal to the angle 1s composed of rather more hairs, 5 usually. On the 4th segment the long lateral seta is sometimes double, THE PUPAE OF WEST AFRICAN MOSQUITOS, 85 sometimes quadruple and the tuft internal to it comprises 5 to 7 hairs. On the 3rd segment there is a tuft of 7 or more hairs on the posterior margin a little internal to the angle. There are only a few small setae at the posterior angle of the 2nd, 3rd, SS S\ IS Fig. 7. Culex fatigans, Wied. and 4th segments. The dendritic setae on the Ist segment are well developed, there is also a pair of small dendritic setae on the 2nd segment. The pupal trumpets are relatively short, but have a large aperture. Culex invidiosus, Theo. (fig. 8). The paddles are similar to those of C. fatigans, that is, they are pyriform and have a well developed midrib, a slight external thickening, but no fringe. At the free end of each midrib are two small setae, set close together, one of which is a little larger than the other. The pupa is, however, much more delicate than that of C. faligans and Fig. 8. Culex invidiosus, Theo. is but poorly chitinised. The 8th segment bears a tuft at its posterior angle, and a delicate double hair close to the insertion of the paddle. The tuft is composed of 7 to 9 branched hairs. There are several delicate setae near the posterior angle of the 7th segment the most conspicuous of which is a tuft composed of 2 to 4 hairs, which may be either simple or subplumose. This tuft lies a little above the angle and close to the lateral margin of the abdomen. On the posterior margin of this segment and about half way between the angle and the mid line, there is a delicate tuft of about 4 hairs. The 6th segment has a delicate tuft of about 5 hairs in the posterior angle, an extremely long double seta on the posterior margin a little internal to the angle, 86 A. INGRAM AND J. W. SCOTT MACFIE, and internal to this a tuft. of about 7 delicate hairs. The 5th segment bears similar setae to those on the 6th segment. On the posterior margin of the 4th segment there are two tufts a little internal to the angle, the outer composed of about 6 hairs and the inner of about 10. On the 3rd segment there are two similar tufts, the outer one of which is however much reduced. On the 2nd segment there are two delicate dendritic setae with long peduncles. The dendritic setae on the Ist segment are well developed. The pupal trumpets have a smaller aperture than those of C. fatigans. Culex thalassius, Theo. (fig. 9). The paddles are broad, with a well marked midrib, and an external thickening, but without a fringe. There are two small terminal setae, the larger of which is sometimes bifid. On the 8th segment there is a large tuft of 10 to 12 branched hairs at the posterior angle, and a small double hair near the external root of the paddle. On the 7th segment there is a small tuft of about 7 delicate hairs close to the posterior angle, Fig. 9. Culex thalassius, Taeo. and a little above this a larger tuft of 4 to 6 stout branched setae. On the 6th segment there is at the posterior angle a delicate triple or double hair, but the most con- spicuous seta lies a little internal to this on the posterior margin, and is long, strong, . bifid, and pubescent or subplumose. The setae on the 5th segment are similar to those on the 6th. On the 4th segment the seta at the posterior angle is sub-divided into a tuft of about 5 delicate hairs, and the conspicuous seta internal to it is sometimes triple. On the 3rd segment the tuft near the posterior angle consists of about 6 elements, but the stout seta internal to it is no longer conspicuous. On the 2nd seg- ment there is a small dendritic seta. The dendritic seta on the 1st segment is large and shows about 8 primary divisions. Culex tritaeniorhynchus, Giles. (fig. 10). The pupal paddles are similar to those of C. thalassius ; they are without a marginal fringe, and their terminal setae are paired but very small. The more conspicuous abdominal setae are as follows. On the 8th segment there is a lateral tuft of about 6 branched hairs. On the 7th segment there is a minute tuft in the posterior angle and just above it a larger tuft of 4 branched setae, and on the posterior margin about half-way between the angle and the mid line there is a tuft of 4 hairs. On the 4th, THE PUPAE OF WEST AFRICAN MOSQUITOS, 87 5th, and 6th segments there is at the posterior angle a tuft of 5 or 6 hairs; just internal to this a stout double seta, which is not so well developed as in C. thalassius; and again internal to this and on the posterior margin a tuft of 6 to 8 hairs. The latter tuft is larger on the 4th and 5th segments than it is on the 6th, but on the 3rd segment Fig. 10. Culex tritaeniorhynchus, Giles. there is a similar tuft which is rather larger than that on the 4th segment. On the 2nd segment there is a small dendritic tuft. The dendritic tufts on the Ist segment are well developed and have about a dozen primary branches. Culex guiarti, Blanch. (fig. 11). The pupa as a whole is very poorly chitinised. The paddles show no marginal fringe; they have paired small and slender terminal setae. Most of the abdominal setae are poorly chitinised, the only conspicuous ones being the lateral tufts on the 7th and 8th segments, and the very long double setae on the 5th and 6th segments. Near the posterior angle of the 8th segment there is a tuft of 7 or 8 subplumose hairs ; this tuft is small, being less than a quarter the length of the midrib. In a similar Fig. 1l. Culex guirti, Blanch. a position on the 7th segment there is a tuft of 3 or 4 subplumose hairs, and on the more anterior segments delicate tufts of about 5 fine hairs. On the posterior margin a little internal to the angle there is on the 5th and 6th segments a strong double or triple seta of great length; in a similar position on the 4th segment there is a tuft of 5 hairs. A little internal to these setae there are delicate tufts on the 3rd to the 7th segments. The dendritic setae on the lst segment are large and much branched. The trumpets are long, poorly chitinised, and have a narrow aperture. 88 A. INGRAM AND J. W. SCOTT MACFIE. Genus CyatHomytA, Theo. Cyathomyia fusca, Theo. (fig. 12). The paddles are rather small, with a slight external thickening, a midrib, but no fringe; they carry paired terminal setae, which are small and very delicate. On the 7th and 8th segments there are conspicuous lateral tufts; those on the 8th segment are long and are composed of 7 subplumose hairs, those on the 7th of 4 or 5 hairs, subplumose towards their extremities and branched at their tips. No such tufts Fig. 12. Cyathomyia fusca, Theo. are present on the other segments, which carry at their posterior angles only a few minute setae. On the posterior margins of the 4th, 5th and 6th segments there is a specially large seta a little less than half-way between the angle and the mid-line ; this seta is single on the 6th segment and double or triple on the 4th and 5th. The | trumpets are short, highly chitinised and with a moderately wide aperture. Genus EKEUMELANOMYIA, Theo. Eumelanomyia inconspicuosa, Theo. (fig. 13). The pupa of this mosquito possesses pear-shaped paddles with a distinct external buttress ; the midrib is well formed and has a peculiar expansion at its end; no terminal seta was visible in the single specimen examined. At the posterior angle of the 8th abdominal segment there is a fan-shaped tuft of 8 subplumose hairs; near Fig. 13. Humelanomyia inconspicuosa, Theo. the posterior angle of the 7th segment there is a fan-shaped tuft of 7 less well developed subplumose hairs, but actually nearer the posterior angle there is a slender triple hair; near the posterior angle of the 6th segment there is a slender triple hair. The trumpets are rather long and narrow, slightly expanded at the ends and with small apertures. The whole pupa is dark and strongly chitinised. THE PUPAE OF WEST AFRICAN MOSQUITOS., 89 Genus MimomytiaA, Theo. Edwards in his “‘ Revised Keys to the Known Larvae of African Culicinae” has given a table by means of which the pupae of M. splendens, M. plumosa, M. mimomyia- formas, and M. hispida can be distinguished. In the following notes we give certain additional points of importance with regard to two of these species, Mimomyia mimomyiaformis, Newst (fig. 14). The pupa as a whole is dark, and the paddles and the middle parts of the abdominal segments are covered by minute spines. The paddles are pyriform, the inner distal angle, however, is characteristically squared; they are infuscated with a white Fig. 14. Mimomyia mimomyiafcormis, Newst. comma-shaped marking on their outer sides; there is a fringe of large sparsely scattered points; the midrib is well developed, and there are single short terminal setae. The tufts at the posterior angles of the 8th segment are very small and consist of only 2 or 3 short hairs. On the 7th segment there are only a few minute setae near the posterior angle, and on the posterior margin a little internal to these two rather larger bifid setae. The latter setae are better developed on‘the anterior segments, and are further subdivided. The dendritic setae on the Ist segment are poorly developed and but little branched. The trumpets are long and pale at the tips. Mimomyia splendens, Theo. (fig. 15). The paddles are narrow, ending in a small knob-like projection, and slightly darkened at the tips. There is a well developed buttress both on the inner and outer margins of the paddles, but the midrib is poorly developed distally. The fringe on the paddles is long. The terminal setae are small, and there is a small seta close to — Fig. 15. Mimomyia splendens, Theo. the midrib a little distance from the end. The tuft at the posterior angle of the 8th segment is poorly developed and consists of only one or two short hairs; the pos- terior angle itself is almost tooth-shaped. At the posterior angle of the 7th segment 90 A. INGRAM AND J. W. SCOTT MACFIE. there are only a few minute hairs. On the 2nd to 7th segments there are on each side two long powerful setae on the posterior margin a little internal to the angle, and a small stout seta internal to them; these setae are more highly developed on the anterior segments, and on the 7th segment may be poorly represented. The dendritic setae on the lst segment are small and sparsely branched. The trumpets are long and pale for the apical third, but the actual end is somewhat darkened. Genus URANOTAENTA, Arrib. The only species of this genus that we have examined is U. balfowri. The paddles are peculiar in shape, as described by Wesché, that is the plates on the inner sides of the midribs are larger than those on the outer sides, an arrangement we have not observed in any other pupae. The paddles of U. balfouri appear to be more pointed at the apex than those of U. ornata, which has been figured by Bacot, and the pupa differs also in having a fan- shaped tuft of simple hairs near to, but not actually at the posterior angle of the 7th abdominal segment, and the small tufts at the posterior angles of the 8th segment composed of only two hairs. In the specimen of U. balfouri examined by us there was also a dark ring at the base of the paddles. Sub-family CHAOBORINAE. Genus CHaosporus, Lichtenstein. Chaoborus ceratopogones, Theo. (figs. 16, 17). The pupa is less curved than the pupae of the Cuticrnar. It has very large paddles of a peculiar shape ;_ these paddles have a well developed buttress on their inner side and their internal margins are fringed, the midrib is broad and lies nearer the inner == Ss S Fig 16. Ohaoborus ceratopogones, Theo. Fig. 17. Pupal trumpet of Chaoborus ceratopogones, X 75. than the outer side of the paddle, the inner lobe being thus narrower than the outer. There is a single slender terminal hair at some distance to the outer side of the end of the midrib. The pupa seems to be devoid of stout hairs, but there are delicate tufts near the posterior angles of all the segments and along the posterior margins. These tufts might easily be overlooked. The trumpets bear but a slight resemblance to the trumpets of the CuLIcINAE; they are flask-shaped or pear-shaped, with the narrow end outwardly directed: their surface is beautifully reticulated, suggesting an THE PUPAE OF WEST AFRICAN MOSQUITOS. 91 alvéolar-structured organ possibly functioning as a primitive lung; the trumpets have an exceedingly narrow slit-like opening distally and the proximal ends are connected with well marked tracheae, which narrow at their junction with the trum- pets and become broader and more convoluted as they are traced into the interior of the pupal case. Miall (Aquatic Insects, p. 119) states that “The pupa floats in an upright position at the surface of the water.” This statement is incorrect so far as the pupa of Chaoborus ceratopogones is concerned, as it appeared to spend the whole of its time at the bottom of the water in the glass vessel wherein it was placed. _ The pupal stage lasted about 36 hours in the two specimens bred. In conclusion, we desire to express our thanks to Dr. Guy A. K. Marshall, Director of the Imperial Bureau of Entomology, for his kind assistance in deter- mining several of the species here dealt with; and to Miss Rhodes for the great care and trouble she has taken in the preparation of the illustrations. a 7 trae ‘¥ Cie Se at ‘ { ; | a ies -" 20, RA” Oy. hee 5 ee ee \ : 1 ; } ~ : 7 . - 2 ms é : AOTigtom, ¥ carn’ ey 4O Saas vo Pe i x ; j vie ; i i? ' dimaog : + It ne ( } ii i } ‘ ce. OL : uit; fi ij i ’ ‘ i ry ff j 1 wf) : i ai bf iJ . me saul f . : p Te rT i a 9F et | pevTtht LehiaeAy GaBF ; of 4 a4 1% y » Lore ree ‘yuh —, ‘5... - 7 os " erie @ 502 Fi) j 4 EB sicemete me hr, odie | ‘pea o —. 7 r - ne) eG . 93 NOTES RELATIVE TO THE IMPORTATION: OF TIPHIA PARALLELA, SMITH, FROM BARBADOS TO MAURITIUS FOR THE CONTROL OF PHYTALUS SMITHI, ARROW. By D. > EmMMEREZ DE CHARMOY, Government Entomologist, Mauritius. The existence in Mauritius of the Melolonthid beetle, Phytalus smithi, Arrow, was detected for the first time in July 1911 on those lands of Mon Rocher Estate which adjoin the Royal Botanical Gardens of Pamplemousses. It was at once suspected that this new pest of the sugar-cane must have been conveyed into the Colony by some shipment or other of sugar-cane cuttings from abroad, but no clue could be found as to its country of origin until February 1912, when it was detected in Barbados by Mr. Guy A. K. Marshall, Director of the Imperial Bureau of Entomology, who was then on his way to the Agricultural Conference held at Trinidad in that year. The first results of the campaign against the insect were recorded in a report published by the Government of Mauritius in 1912. The methods therein prescribed for combating the pest are those that are still employed and consist mainly in :— (1) Digging out larvae by means of hoes from those fields which show an average infection of 15,000 to 20,000 larvae and over per arpent. ‘This operation is as a rule carried out during the months of May to September and is extended to October in very exceptional cases, for by this time of the year most of the larvae have burrowed deep into the ground for their ultimate transformation. (2) Capturing adult insects at night from 7 to 9 p.m., which is carried out during the months of November to April and sometimes until June, The annexed statements show the months of greatest emergence of the adult insects, the number of insects destroyed from July 1911 to June 1916, and the annual expenditure incurred. It will be noticed that a considerable reduction in the number of insects has taken place in that zone which was at first most heavily infested, thus showing the possibility of effectively bringing the pest under control wherever, as at Mon Rocher, - the prescribed measures are carried out with persistence and energy. From the accompanying sketch map (see page 96) it will also be seen that the insects, though not increasing’in numbers, have steadily gained ground during the last five years, the radius of the infected zone being now greater by one kilometre. The quarantine measures, however, that were imposed from the very beginning of the outbreak have succeeded in localising the pest within the limits of the District of Pamplemousses and in preventing its spread throughout the Colony. As soon as the origin of the pest was ascertained, suggestions were made relative to the introduction of its natural enemies, which it was presumed might exist in the country of origin. Mr. Marshall during his stay in Barbados having ascertained that the insect, although distributed throughout that island, did not cause any serious 6ZI‘SHZEF | OS8‘F69'9 | DOFLGE‘TS | OGO‘SOS‘L | O€L‘F8F‘9R | GEP‘9EO'S | NES*SFOOT | SI8‘IIS | LET‘O9F‘9Z ae 9FG‘LIT OGL‘6S 666 = 2 = = ce LZ6‘09S‘8T | OOS*ZIO'S | SIS‘LSZS‘LI | OOL‘EST O88‘°S8ES | OS1'Ss LIG‘9OSPhF = OZL°SI9°SI | 000‘OT9'S | 986‘°618‘II | OSO‘OLS 888 ‘PEPrs = L98°609°Z —# ELEGSS'0T | OOSESO'T | SBSTFE'ST | OOT*FOT'D | EBF*LLE“OS | OOSFZS‘T | S8E‘LLL‘S = C96‘SSFI | 0086 IZ8°L98°S | OOS‘LFE 686'L86°F | SFOSLST L90°ZZZ'S | SI8‘I1S ‘S0[}09 “OVAIVIT "SO[}00 | "OVAIV'T *SsoT}00q ‘OVAIV'T *S0]}00q “OBAIV'T ‘9I6I-SI6I “CI6I-FI6I ‘PIGI-SI6I “EI6I-ZI6I 918816 a. SE6'T 62E*60S‘T | SZL*Lgs 9S 698'F = PIPEST‘OL | LFL‘9OT ZE1‘8c6‘6 | 999°EGE ‘sop00 gq ‘OVAIV'T ‘ZIGI-II6I SpOOSUT JO IJOqWINN [eIOT, IiUAANOY pu OI[VSOYy opuery felesoy ojt0g ‘ oouviods puv si1oj}uelg Teurg “i uvi[q nvog qUuno wz. Iayooy uopy a EE EEE EE ee SS ee eee eee ee | eee ee eee ee eee eee ee eee ‘OI6T 8UNL O71 TIGL Ang wouf pruysig sassnowodung oy) ur pohousap mys snpeydyg fo avamwT pun sajoog fo saqunyy ay) Gurmoys 2190, THE IMPORTATION OF TIPHIA PARALLELA INTO MAURITIUS. 95 Table showing the Total Numbers of Phytalus smithi destroyed in the Pamplemousses District from 1911 to 1916, with the Cost. _ Amount Spent. Total. Larvac. Beetles. Rs. Cs, Mon Rocher 1,309,029 | 22,929,442 | 18,041 15 Mount 9,117,847 65,379,783 31,920 88 Beau Plan 3,180,050 24,734,003 17,282 74 Small Planters and Esperance 3,478,475 | 47,817,895 | 21,824 15 Petite Rosalie Grande Rosalie and Souvenir 41,688 118,545 193 13 Total Number of Insects 17,127,089 | 160,979,668 | 89,262 | 05 Table showing the Number of Larvae and Beetles of Phytalus smithi captured at Mon Rocher during the Year 1914-15. July August September October November... is Bh December January February March .. April Total .. i we Larvae. 330,700 11,200 300 — —- — | 347,200 Beetles, —————EE 40 30 3,150 48,017 2,286,757 74,110 636,830 246,910 38,262 3,367,321 96 D. D’ EMMEREZ DE CHARMOY. damage to crops such as it did in Mauritius, at once concluded that it was being kept in check by some natural enemy. Upon Phytalus larvae which he brought back to the British Museum, Mr. Gilbert Arrow detected the presence of the larva of a Scoliud wasp, suspected of being that of Dielis (Campsomeris) dorsata, which Mr. Marshall had observed in fields infested with Phytalus and Lngyrus tumulosus. 5 es FO =) La Ag & if) pel SB ‘ len > “AI OR aS MISTRICT 5 |. \ a i ee m \ Fig ~~ / DP \ / \ 7 mt 7a P.M. Powders Mills SY \ Hy R.O. Royal Observatory % \ of B.G. Botanical Gardens “S ‘XS Le? L. Langlois Sketch map in which the spiral black line shows the gradual increase in the distribution of Phytalus smithi during the past five years. This fact led them to believe that this Scoliid might be one of the chief parasites that kept the Phytalus in check. It has since been shown, however, by Mr. W. Nowell, then Mycologist to the Local Department of Agriculture, Barbados, in a report published in the Annals of Applied Biology (1915) that the ratio of this para- sitism to that due to Tvphia parallela at the same time and place was calculated to be THE IMPORTATION OF TIPHIA PARALLELA INTO MAURITIUS. 97 about one to a hundred. Mr. Nowell further states that towards the end of 1911 a large number of coccons had been found by him in cane-fields infested with white grubs, which cocoons were identified as being those of T7phia parallela, the existence of which had already been recorded in Brazil. At about the same time the writer of this article recommended that enquiries be made from the Government of Barbados as to the natural enemies of the Phytalus with a view to their introduction into Mauritius. From the labels attached to the specimens existing in the British Museum previous to its discovery in Mauritius, it was supposed that Phytalus smith was indigenous to Trinidad. It was only in 1912 (March) that the nature of one of its enemies was indicated to the writer by Mr. Gilbert Arrow, and soon after definitely identified by Messrs. Bovell and Nowell of the Agricultural Department of Barbados, as Tiphia parallela. The Government of Barbados was at once approached with a view to obtaining a few of these valuable insects and arrangements were kindly made by Mr. John R. Bovell, Superintendent of Agriculture, for the despatch of several shipments of Tiphias to Mauritius under different conditions from April 1913 to August 1915 as follows :— (1) 2nd April 1913.—First consignment shipped from Barbados, a short time before the inauguration of the Department of Agriculture in Mauritius, and landed 26th June. It consisted of cocoons picked up in the fields and packed with some earth in a small wooden case. None of the cocoons hatched out. (2) 8th July 1913.—Second consignment sent by parcel post, and reached Mauritius 15th August. This shipment too consisted of cocoons which did not hatch out. (3) 25th June 1913.—Third shipment despatched, and arrived 15th August 1913. It consisted of four Wardian cases prepared under the care of Messrs. Bovell and Nowell and containing larvae of Phytalus and of Tiphia as follows :— Case.No. I... No.2... No, 3... No. 4. .« Total. Tiphia larvae... “e 50 40 50 41 181 Phytalus larvae .. oe nee 160 180 120 650 Tiphia cocoons .. eines — ~— — 170 The contents of these four cases were placed on the day of receipt in the Insectary at Pamplemousses, which had been amply stocked with Phytalus larvae of different stages during the previous six weeks. As far as could be ascertained without breaking up the earth mass, these cases contained, besides 170 cocoons—31 of which had hatched out during the voyage, while the rest never hatched out—a fair number of cocoons in situ, numerous non- parasitised Phytalus larvae, five parasitised ones and two male Tiphia. From 15th to 28th August, one male emerged daily; from 28th August to 14th September none was found; and from the latter date to 2nd February 1914, a few insects of both sexes emerged daily, the greatest number in a single day being four males and two females on 29th November and six males on 9th December. As a rule these insects came out after 10 a.m. and buried themselves in the earth towards 3 p.m. (C365) G 98 D. D EMMEREZ DE CHARMOY. On 9th October the first parasitised larva was found, on the 20th a second, and a third on 11th February 1914. In April an examination of the Insectary was made and at the bottom were found but a few cocoons showing emergence holes. The writer was absent from the Colony from October 1913 to March 1914 and was _ therefore unable to ascertain personally the number of insects that hatched out during that period ; but as his instructions were strictly complied with, he has good reasons to believe that these were very few and that the development of the insects in the Insectary was impaired by causes as yet to be determined. With regard to this consignment there are a few facts that deserve to be noticed. Mr. Nowell had ascertained the probable egg capacity of a female to be about 70, and further that the insect can reproduce parthenogenetically. Supposing there were a minimum of 100 females in the cases out of 212 adult insects therein contained (181 originally introduced and 31 that hatched out from the cocoons en route), we see that 5,000 Phytalus larvae could have been parasitised by them, if we assume an average laying of 50 eggs per female Tiphia. As it was, the cases contained only 530 Phytalus larvae; many of those that outlived the voyage were non-parasitised, and nearly a hundred Phytalus cocoons were found ;_ so that it is clear that the egg-laying power of the Tiphia had been considerably reduced under the prevailing conditions. As the life-cycle of a Tiphia ranges, according to Mr. Nowell, between 50 and 60 days, it is evident that the two larvae found parasitised on arrival had been attacked by wasps of the second generation that emerged during the voyage; and here again it is a matter of surprise that only two of the Phytalus larvae in the cases were parasitised. There is, further, no room for doubt that the reproduction of the Tiphias was continued in the Insectarium up to the fourth generation, with a gradual reduction in the egg-laying power of the females. (4) 10th June 1914.—Fourth consignment despatched, and reached Mauritius 15th August 1914. This shipment was made under the same conditions, but owing to the scarcity of adult Tiphias the cases were stocked with 29 cocoons and 32 adults only, while Phytalus larvae were placed therein to the number of 1,064. Mr. Sydney Dash, Assistant Superintendent of Agriculture, Barbados, was able to ascertain that nearly 15 per cent. of the cocoons had hatched out and that a certain number of Phytalus larvae had already been parasitised before the despatch of the cases. One of the cases was received in a soaked condition and from the three others 22 male and 10 female Tiphias were obtained. There were numerous non-parasitised larvae, just as in the previous shipment. These cases were left undisturbed, the insects as they emerged being removed and placed in the Insectary so as to allow of the exact number of emergences being known. Hatching began on the 4th September and continued up to 24th January 1915, when 3 males were found; 5 females and 10 males hatched out in September, 2 males “and 1 female in October, 7 males and 1 female in November, and 3 males in January, or a total of 32 insects, which constituted the 3rd generation of the 58 adult insects that emerged in May 1914, not taking into account the 4th case that was flooded, THE IMPORTATION OF TIPHIA PARALLELA INTO MAURITIUS. 99 As already stated, the adult insects were transferred from the cases to the Insectary as soon as they hatched out ; by 20th November, however, no parasitised larva could be found (so far as could be ascertained without disturbing the insectaries), and it was decided to set free some insects. Two couples were therefore liberated at Mon Rocher to the east of the Botanical Gardens ;_ two further couples escaped from the Wardian cases on being transferred to the Insectaries and flew away to the west of the Gardens. (5) The fifth consignment was despatched on 18th June 1915, and reached Mauritius on 25th August; it consisted of cocoons spun in the case. After the parasitising of the Phylalus larvae had taken place in Barbados, the case, which had been fitted with four small feet, was closed exactly at the level of the contained earth, thus guarding both the earth and cocoons from being moved about. |The case travelled via Southampton in the cold storage hold of steamers as far as Durban, but from the latter place to Mauritius it had to be transhipped and was placed directly upon ice blocks, which caused the death of the Phytalus larvae and the Tiphia pupae in the cocoons. On arrival there were found only fifty cocoons that had been spun in situ, that is to say, a number far inferior to that originally placed in the case. The recurrence of the same facts on three different occasions, and the ever-increasing difficulties that were experienced in obtaining new shipments from Barbados, induced the Department of Agriculture to consider the question of obtaining another closely allied species from a nearer source than Barbados. This question was carefully gone into and the Chamber of Agriculture passed a resolution to the effect that steps should be taken for the introduction of Tvphia bisinuata from Madagascar; and in the meanwhile, in case this attempt should not prove successful, that more effective measures be taken to secure further shipments of Tiphia parallela from Barbados or Brazil under the best possible conditions. In view of a great reduction in the number of Phytalus to be found at Mon Rocher, very careful and thorough searches were made amongst the larvae. Not only was a large percentage of larvae found to be affected with bacterial or cryptogamic diseases, but also during this search parasitised larvae came to light. Two parasitised larvae were detected on the 8th and 9th May 1916 in the very field where Tiphias had been liberated in 1915, and by the 26th of the same month 7 other parasitised larvae had been found in various fields that were somewhat distant from each other. The first adult Tiphias were met with on the 23rd of the same month on Herbe Condé (Cordia interrupta) in the immediate vicinity of the field where Tiphia larvae had been found, and some days later 6 females and 10 males were found at Esperance. In the month of July parasitised larvae and adult Tiphias were found in greater numbers, more than 40 adults being met with in one instance on a single plant of Herbe Condé. In the following months they were found less frequently and became scarce in September and October. . The presence of parasitised larvae in the very field where adult Tiphias had been liberated 6 months previously left no doubt as to their orig n, as well as that of those found elsewhere, 7.e., on the leeward side of the Botanical Gardens, in which direction those adults had flown that escaped from the Wardian cases as before stated. (C365) G2 100 D. DEMMEREZ DE CHARMOY. In view of the number of the adult insects (40) met with at one time on a single shrub of Cordia interrupta and the area over which they spread, it may safely be concluded that there existed several hundreds of adult couples scattered over a surface of about 400 to 500 arpents. This fact is a clear proof of the influence exercised by a state of captivity upon the longevity and egg-laying power of certain species. At this stage, a question naturally arises which it would be difficult to answer for the present. Granted that four couples were able to generate such a large number of insects within a period of 8 months, and that these insects became scattered over such a large area, how is it that in the same locality and under apparently similar conditions a much greater number of couples could not, on two different occasions, increase to a proportional extent ? At first sight one can only ascribe this to their having been kept in captivity ; and with a view to investigating the different effects that-may be produced by such unnatural conditions, we shall here briefly review the observations we were able to make both as regards Tiphia parallela and Elis rufa, a species of Scoliid peculiar to the Mascarene Islands. It is known that captivity brings about a state of sterility in certain vertebrates, and that this sterility may be temporary or permanent, according as certain conditions are temporarily or permanently absent. Captivity entails restricted space, deficient nest-material and, it may be, deficient food and unsuitable temperature ; and it may well be responsible for the sterility and lesser vitality of certain animals. As regards Tiphia, climatic conditions may be excluded, as they were the same for both captive and free insects; nesting-materials may also be set aside, as they consisted of Phytalus larvae, which in the Insectaries were available in greater numbers and were of easier access. Insufficient space does not appear to be a check on fertility, because from unfertilised (parthenogenetic) females that were kept in such small vessels as flower-pots Nowell succeeded in obtaining fertile eggs. Further- more, reproductions, although on a considerably reduced scale, actually took place in Wardian cases and in the Insectary at Pamplemousses. It is possible, however, that insufficient space, which necessarily impedes the movements of insects, coupled with the absence of direct solar rays, may have a certain bearing on their longevity, if not on their fertility. It will be seen further on that the egg-laying habits of Tiphia and Elis are such as _ to necessitate a sufficient length of time to allow of all eggs being laid; and it may be as well to mention here certain facts which may somehow account for the considerably reduced fertility of Tiphias when in a captive state. The writer of this article, whilst studying the life-history of Elis rufa, had been struck by the invariable number of parasitised larvae or cocoons met with in cane-stools infested by Lachnosterna. Whatever the number of larvae found in a stool, there were hardly ever more than two—exceptionally three—para- sitised larvae or cocoons to be found; if larvae, they were of the same age; if cocoons, they hatched at a few days’ interval. In captivity it was observed that larvae were parasitised by twos and at intervals varying from 6 to8 days. The same observations were recorded in the case of Phytalus larvae parasitised in open fields by Tiphaa. Examinations of the ovaries of several female Elis showed that the eggs could only THE IMPORTATION OF TIPHIA PARALLELA INTO MAURITIUS. 101 be laid in this manner owing to their different degrees of development; for in no case could more than two eggs be found in the same stage of development, namely one in the right oviduct and one in the left ; and furthermore the interval between one laying and the next is explained by the marked difference in the sizes of the successive eggs in the ovaries. The following observations will confirm this statement : On 21st June two female Tiphias were captured at Esperance and placed in the Insectary“at Pamplemousses together with numerous Phytalus larvae and were fed on honey and water. On 26th June a hundred of the larvae were examined and one of them was found to be parasitised. On 10th July three other larvae were parasi- tised-; on the 17th two, and on the 27th three; or a total of 9 eggs laid by two female Tiphias during a period of more than one month. On 5th August two other females were submitted to the same conditions and by the 12th August had para- sitised five larvae. It is quite possible that some other parasitised larvae had burrowed too deep in the soil to be found, but it is clear from what precedes that Tiphia and Elis lay their eggs only by twos, and exceptionally three at a time, and that an interval of at least seven days takes place between one laying and the next in captivity. It may be well to add that none of these eggs hatched out ; and furthermore Nowell states that he could never obtain more than six eggs from one captured female. It would seem therefore that it is the longevity of the insect that is affected by captivity rather than its egg-laying power, and we are then led to enquire whether the food question does not play here an important rdle. Nowell has observed that in Barbados Tiphias when in a free state lived on the secretions of Aphids, and that when in captivity they willingly fed on sugary liquids and on honey, but that neither in the captive nor in the free state did they resort to flowers for their food. The writer, who had made the same observations, was therefore afraid that it would be difficult to establish them at Pamplemousses for lack of proper food material, seeing that the few existing species of Aphids appear only at a certain time of the year and are then rapidly destroyed by Syrphids, Braconids, Coccinellids and fungi. Now it is a noteworthy fact that Tiphias have established themselves at Pam- plemousses and have made good the scarcity of plant-lice by having recourse to the sweetish content of the vesicular hairs of Cordia interrupta. This plant is a native of British Guiana, which was introduced about 15 years ago; it is now a regular pest and infests all uncultivated fields. This unexpected change of diet might at first sight lead one to believe that the quality of the food is immaterial to these insects, and it had no evil influence on the captive ones. A closer observation of other facts, however, leads to a different opinion; namely, that.this change of diet was the result of a deliberate choice, most probably due to the constituents of the vesicular hair content of Cordia interrupta being more closely related to the secretion of Aphids than to the nectar of flowers. It is indeed well known that ants are as a rule very fond of the secretions of Aphids, while they are only exceptionally met with on flowers. The red ants of Pample- mousses (Solenopsis sp. ?) show a peculiar liking for the secretions of Coccids and are 102 D. D EMMEREZ DE CHARMOY. found in great numbers on trees infested with Icerya seychellarum, Westwood. Now these ants sometimes occur in such large numbers on leaves of Herbe Condé that one might believe this plant to be infested with Aphids, the belief being strengthened by the fact that the ants circle round a central spot, just as if they were surrounding an Aphid or a Coccid. This further tends to prove the similarity that apparently exists between the secretions of the Aphids and the sweetish content of the vesicular hairs of Cordia interrupta. It would be premature to conclude from the preceding observations that the ‘quality of the food was chiefly responsible for the attempts at Tiphia importation having been nearly a failure, as there may exist other causes which more systematic searches might bring to light. The writer’s aim in recording these observations and in trying to interpret them was to point out the difficulties experienced in introducing insects from a long distance and the necessity of a closer study of the food requirements of the economic species utilised by mankind as auxiliaries. The importation of Tiphias and their establishment in Mauritius have brought to light several interesting facts which may serve as guides in the future for the intro- duction into other countries of species of the same family. It is the pleasant duty of the writer to avail himself of this opportunity for acknow- ledging the valuable assistance received from all those who contributed to this importation. To Mr. John R. Bovell, Superintendent of Agriculture, Barbados, and his assistants, Messrs. W. Nowell and J. Sydney Dash; to Dr. Guy A. K. Marshall, Director of the Imperial Bureau of Entomology ; to Mr. C. P. Lounsbury, Chief of the Division of Entomology of the Union of South Africa ; and to Mr. C. W. Mally, Entomologist of the Cape Province, are due the grateful thanks of the planters of this Colony for their untiring efforts in securing the importation into Mauritius of this valuable parasite. [Since the foregoing paper was received Mr. d’Emmerez has written to say that at that time (5th May 1917) specimens of Tiphia parallela were to be found in thousands on Corda wnterrupta all over Esperance Estate and Maison Blanche. He estimated the females on the plants at about 5°%, but was not certain whether this was due to the earlier emergence of the males, or to the fact that the majority of the females were engaged in oviposition underground. He states that in his insectary the females have laid fertile eggs on larvae of Adoretus versutus, and on young larvae of Oryctes.—KD. | 103 NOTES ON THE. COCCIDAE OF BRITISH GUIANA.* By G. E. Bopxrn, B.A., Dip. Agric. (Cantab.), F.Z.S., F.E.S., Government Economic Biologist, British Guiana. The object of this paper is to supplement an article which appeared early in 1914 in the Journal of the Board of Agriculture of British Guiana (Vol. iii, No. 3.) At that time it was proposed to publish a further account of these insects as soon as more species had been collected and identified, and fuller information gained. Also, taking into consideration the great economic importance of the Coccrpag, it has been deemed advisable to disseminate as widely as possible all information which has been acquired concerning them locally. In the previous article, some fifty-one species were enumerated and notes recorded concerning their habits and economic importance. In this paper information appertaining to sixty-eight species is given, a number of which are new to science. Further knowledge with regard to the para- sites and predaceous enemies has been gained, as well as their economic status. The following short account of previous work on the Coccrpak of British Guiana is of interest. Their study may be said to have commenced in 1888 when a number of common species of Coccids were collected in the Botanic Gardens, Georgetown, and forwarded to Mr. J. 8. McIntire, of Shepherds Bush, London. These species were examined both by this gentleman and also Mr. J. W. Douglas, and the results of their studies were published in the Entomologists Monthly Magazine, the Journal of the Quekett Microscopical Club, and in two papers published in Timehri, the Journal of the Royal Agricultural and Commercial Society of British Guiana, the one written and illustrated by Mr. McIntire and the other by Mr. R. Ward. Between the years 1888-1894 a series of articles dealing with Guiana CoccIDAE appeared in the Entomologists Monthly Magazine, written by Mr. J. W. Douglas, Mr. A. C. F. Morgan, Professor R. Newstead and Mr. E. E. Green. The material on which these articles were based was collected in the botanic Gardens, Georgetown, by Mr. G. 8. Jenman, F.L.S. Thus some twenty-two species of scales were described either for the first time or identified as previously described species. During more recent years, with the exception of a small but interesting collection made by Mr. A. W. Bartlett, who was Government Botanist for some years, no further work has been done on the Coccrpaz of this part of Tropical America. The collection of CocciDAE in this laboratory was commenced by the writer in 1911 and may now be said to be very fairly representative. Collections have been made in all the various cultivated parts of the Colony, and species have been obtained from many of the wild and uninhabited portions. I have to thank the following gentlemen who from time to time have forwarded interesting specimens, Mr. R. Ward, Mr. L. D. Cleare Jnr., Mr. P. M. de Weever, Mr. A. A. Abraham, Mr. E. M. Morgan, and Mr. H. W. Moore. The determinations have been made in most cases through the agency of the Imperial Bureau of Entomology by Professor Robert Newstead, F.R.S., the well-known authority on Coccrpar. With the exception of * Published by permission of the Director of Science and Agriculture,. British Guiana. 104 G. E. BODKIN. a few species, either dried or microscopical preparations, and in many cases both, are preserved in this laboratory. For the sake of convenience the classification adhered to throughout the paper is that adopted by Mrs. Fernald in her well-known catalogue. In British Guiana, as elsewhere, scale-insects are among the worst insect enemies with which agriculturists have to deal. The cost of their annual depredations must be enormous, yet little is done by the smaller farmer in the way of controlling them. Systematic spraying is carried out on some of the large coconut estates with beneficial results. The CoccINELLIDAE are among the most important enemies of scale-insects in this country and the following species have been observed and identified. Pentilia insidiosa, Muls. Observed to be predaceous only on Asterolecanium bambusae, Bdy. It is a common species where this Coccid occurs; few bamboos are free from them. Hyperaspis festiva, Muls. Predaceous on Pseudococcus sacchari, Ckll. Occurs commonly, being found beneath the leaf-sheaths of sugar-cane with the scale. Hyperaspis orthopustulata, Muls. Another common enemy of P. sacchari, CkIl. Brachyacantha 10—punctata, Melsh. Predaceous on a species of Pseudoccus.. An uncommon species. Neda dilychnis, Muls. Predaceous on Aspidiotus destructor, Sign. An uncommon species, but found in this one instance occurring in large numbers on a heavily infested coconut palm. Cryptognatha nodiceps, Mshl. A common species preying on A. destructor, Sign. Azya trinitatis, Mshl. Another common enemy of A. destructor, Sign. Azya pontbrianti, Muls. Frequently occurs as an enemy of Savssetva hemi- sphaerica, Targ. The Chrysopid, Chrysopa claveri, Navas, is occasionally found as a predaceous enemy of the sugar-cane mealy bug (Pseudococcus sacchari, Ckll.).. The following Hymenopterous parasites have been bred from CoccIDAE at various times. Considerable difficulty has been experienced in obtaining identifications. A large number of species still await determination. c Arrhenophagus chionaspidis, Auriv., bred from Hemichionaspis minor, Mask. Leptomastix dactylopu, Howard, bred from Pseudococcus citri, Risso. Lecaniobius cockerelli, Ashm., bred from Saissetia nigra, Nietn. Three species of coccophagous Lepidopterous larvae occasionally occur, namely Blastobasis lecamella, Busck (BuastoBastpak), Vitula bodkini, Dyar (PYRALIDAE) and Vitula toboga, Dyar (PyRALIDAE). These three species attack indiscriminately the following Coccids: Saissetia nigra, Nietn., Saissetia oleae, Bern., Saissetia hemi- sphaerica, Targ., and Ceroplastes floridensis, Comstock. Two species of entomogenous fungi are at times found attacking colonies of scale- insects. Both species are well known elsewhere in the West Indies. They are the red-headed fungus (Sphaerostilbe coccophila, Tul.) and the shield-scale fungus (Cephalosporium lecanii). The red-headed fungus principally attacks Chionaspis citrt, Comstock ; while the shield-scale fungus attacks such Coccids as Saissetia NOTES ON THE COCCIDAE OF BRITISH GUIANA. 105 mgra, Nietn., Saissetia oleae, Bern., Saissetia hemisphaerica, Targ., Coccus mangiferae, Green, Coccus hesperidum, Linn., and Coccus viridis, Green. Both of these fungi are more effective in the interior districts, where a more humid climate prevails. On the coast lands they are only operative during the rainy season. The following species of ants have been observed to attend certain species of CoccrIDAR. Daceton armagerum, Latr. P. citri, Risso, on cacao pods. Cryptocerus atratus, L. P. citri, Risso, Coccus hesperidum, Linn., Saissetia nigra, Nietn. Cryptocerus minutus, F. Pulvinaria pyriformis, Ckll., Coccus hesperidum, Linn. Ectatoma tuberculatum, Oliv. Saissetia ngra, Nietn., on Hibiscus. Tetramorium guineense, F. Pseudococcus sacchari, Newst., on sugar-cane ; P. citri, Risso, on cacao pods. | Azteca schimperi, Km. Lecanium aequale, Newst., on Courida; Cero- plastes avicenniae, Newst., on Courida. Dolichoderus (Hypoclinea) Pseudococcus citri, Risso, on cacao pods. bidens, L. Solenopsis pylades, Forel. Pseudococcus sacchari, Newst. A species of Cremastogaster fosters Akermes quinqueport, Newst. Family COCCIDAE. Subfamily MoNopHLEBINAE. Llaveia abrahami, Newst. A rare species. First observed by Mr. A. A. Abraham at the Government Experimental Station, Issororo, N.W.D. It was only found in thissinstance in an indentation of the bark of the native rubber tree, Sapvum jenmant. Subfamily DIASPINAE. Chionaspis citri, Comstock. A very common species; to be found on the bark of most species of citrus trees and also on the castor oil plant. Howardia biclavis, Comst. Fairly common on the branches of jasmin (Jasminum sp.). Howardia biclavis detecta, Mask. A very uncommon form. Only once collected on the branches and twigs of the native rubber tree, Sapiwm jenmani. Diasyis boisduvali, Signoret. Occasionally found on the stem of plantains. Common on the leaves of some orchids, especially the Cattleyas. Diaspis echinocactt opuntiae, Ckll. Recorded in 1893 from British Guiana by T. D. A. Cockerell and in 1889 by Newstead. I have personally never collected this species in the Colony. Aulacaspis rosae, Bouché. Very common on the leaves of mango trees; also found on cinnamon (Laurus cinnamonum). The males are always more conspicuous. Hemichionaspis minor, Mask. Fairly common on cultivated cotton and also on a species of wild ipecacuanha. 106 G. E. BODKIN. Pinnaspis buxi, Bouché. Common on a number of ornamental plants and also on garden plants. Aspidiotus cydoniae, Comstock. Occasionally occurs on the stem of the egg-plant (Solanum melongena). Aspidiotus destructor, Signoret. The commonest and most destructive Coccid in British Guiana. Occurs on a large variety of cultivated and wild plants. Its com- monest cultivated host-plants are the coconut palm, plantains (Musa paradisaica), bananas (Musa sapientium) and avocado pear tree (Persea gratissvma). Aspidiotus diffinis, Newstead. Recorded from British Guiana in 1893 by R. Newstead as a new species. It has not been collected here since. Aspidiotus sacchari, Ckll. A common species, usually found on the stem of sugar- cane just beneath the surface of the soil. . Aspidiotus rapax, Comst. An uncommon species. Up to the present only found on the twigs of the Oronoque tree (Lrythrina glauca). Professor Newstead remarked on examining this species. “The 29 in this colony have remarkably malformed squamae ; I thought it new at first, but am of the opinion that the organs in question are in some way diseased, they are so inconstant.” Morganella longispina, Morgan. Recorded from British Guiana by A. C. F. Morgan in 1889; careful search has failed to reveal its presence since. Pseudaonidia fossor, Newst. An uncommon species; only once collected on the twigs and branches of a large Muscatel grape-vine in Georgetown. Selenaspidus articulatus, Morgan. Occurs commonly on all varieties of citrus plants and also on the leaves of Liberian coffee (Coffea liberica). Chrysomphalus aondum, Linn. Not a common species. Occurs on the leaves of citrus plants. Chrysomphalus aurantu, Mask. A very occasional species on citrus playpts. Chrysomphalus biformas, Ckll. Of frequent occurrence on most orchids, especially Schomburgkia crispa, and also on sisal plants (Agave rigida var. sisalana). Chrysomphalus dictyospermi, Morgan. Recorded from British Guiana in 1889 by A. C. F. Morgan. It has not been collected since. Chrysomphalus dictyospermi pinnulifera, Mask. Recorded from British Guiana in 1892 by Maskell as occurring on crotons (Codiaeum spp.). Not observed since. Chrysomphalus dictyospermi arecae, Newst. Recorded from British Guiana by Newstead in 1893. Not found here since. Chrysomphalus personatus, Comstock. A common species. Occurs on leaves of star apple (Chrysophyllum cainito), many ornamental palms, mangoes, and also on the leaves of Para rubber trees (Hevea brasiliensis). Chrysomphalus erythraspidis, Newst. A comparatively rare species. Collected once on the twigs of the Oronoque tree (Lrythrina glauca). It was found in abundance in this particular instance. Lepidosaphes beckii, Newm. One of the most widely distributed and commonest species of Coccids in the Colony. Citrus plants, crotons, Barbados cherry (Malpighia glabra) are a few of its common host-plants. NOTES ON THE COCCIDAE OF BRITISH GUIANA, 107 Lepidosaphes pinnaeformis, Bouché. Recorded from British Guiana in 1892 by Maskell. It has not been taken here since. Ischnaspis longirostris, Sign. Commonly occurs on mango (Mangifera indica), Liberian coffee, and many species of ornamental palms. Parlatoria ziziphus, Lucas. To be found at times on citrus plants, though it is not a common species. Subfamily ORTHEZIINAE. Orthezia insignis, Douglas. An exceedingly injurious and common species. Found on a large variety of plants, chiefly belonging to the order Compositae. This and the following species are well-known garden pests, especially in Georgetown. Orthezia praelonga, Douglas. Commonly infests a number of plants such as crotons, mangoes, Barbados cherry, etc. ; it has also been known to attack sugar-cane and a species of rubber tree (Sapiwm jenmant). Subfamily DacTYLOPrINAE. Asterolecanium bambusae, Bdv. The ordinary species of bamboo are seldom without this Coccid. It is widely distributed over the Colony. Asterolecanium fimbriatum, Fonsc. Recorded from British Guiana in 1889 by Morgan on leaves of Cupania sapida. I have been quite unable to find this species despite a vigorous search. Asterolecanium pustulans, Ckll. A not uncommon Coccid on the leaves of some orchids and the akee (Blighia sapida). On several occasions I have observed it on the twigs of Hevea brasiliensis, where it was causing considerable damage. Lecaniodiaspis dendrobu, Douglas. Recorded from British Guiana by Douglas in 1892. I have not been able to find this species on the host-plant mentioned by Douglas. Pseudococcus citrt, Risso. Of frequent occurence on citrus plants, cacao, ferns, crotons, occasionally on rice plants and several species of grasses, rarely on sugar- cane. Pseudococcus nipae, Mask. Recorded from British Guiana in 1893 by R. Newstead on coconut palms. Not a common species. Pseudococcus virgatus, Ckll. On one occasion collected from the leaves of Boulanger - oregg-plant. It occurs also on some garden plants, such as violets. Pseudococcus sacchari, Ckll. The common sugar-cane mealy bug of British Guiana. Found wherever sugar-cane is grown. It is much more prevalent in dry weather, when it flourishes ; wet weather causes it to be attacked by a green parasitic fungus which causes great mortality. Subfamily TACHARDIINAE. Tachardia lacca, Kerr. Collected here in 1890 by Robert Ward in the Botanic Gardens. I have not observed the species personally. Tachardia bodkini, Newst. Collected on the twigs of a native rubber-producing tree, Supium jenmam. <A rare species. 108 ‘+ @ E. BODKIN. Subfamily Coccrnae. Pulvinaria pyriformis, Ckll. Of common occurrence on the leaves of guava (Psidium guayava), avocado (Persea gratissima) and a few species of ornamental plants. Pulvinaria simulans, Ckll. Recorded some years ago by A. W. Bartlett. I have not collected the species myself. Pulvinaria flavicans, Mask., var. formicicola, Newst. An uncommon species. Collected from a wild species of plant with a deep red sap, which has earned it the name of “ blood-wood plant.” This species was attended by a species of Cremasto- gaster ant. Pulvinaria elongata, Newst. A rare species. Occurs on the leaf-blades of sugar- cane. Ceroplastes denudatus, Ckll. Recorded in 1893 by Cockerell as occurring in British Guiana on crotons. It has been collected here recently on a species of wild Sola- naceous plant. Professor Newstead remarks on examining this species “ these are the first examples I have seen with the tests complete.” Ceroplastes dugesii, Towns. Recorded some years ago by A. W. Bartlett as occurring on Schinus terebinthifolius. Ceroplastes floridensis, Comst. A common species, occurring especially on citrus plants and ferns. Ceroplastes cirripediformis, Comst. An uncommon species, occurring on Ipomoea sp. in Georgetown. Ceroplastes avicenniae, Newst. An uncommon species, occurring solely on the maritime plant known locally as Courida (Avicennia nitida). Vinsoma stellifera, Westw. Recorded from British Guiana since 1888. An exceedingly common species on a great variety of plants, principally coconut, citrus plants, ornamental palms, and orchids. Occurs also on several species of wild plants found in the interior districts. Eucalymnatus chelonoides, Newst. A rare species. Collected from leaves of Pachira insignis in the Botanic Gardens, Georgetown. Eucalymnatus perforatus, Newst. Found only occasionally on the leaves of coconut (Coccus nucifera) and French cashew (Eugenia jambolana). Eucalymnatus tessellatus, Sign. A common species on ornamental palms. Coccus aequalis, Newst. A small species and common in certain districts on Courida (Avicenna nitida). Invariably associated with the species of ant mentioned previously ; occasionally a species of Cremastogaster ant is found in attendance. Coccus hesperidum, Linn. A common species, especially on citrus plants and Liberian coffee (Coffea liberica). Coccus mangiferae, Green. Recorded from British Guiana since 1888. An exceedingly common species on the leaves of mango, also on jasmin (Eugenia panicu- lata). Coccus viridis, Green. A common species on Liberian coffee in certain districts. Of considerable economic importance. Coccus wardi, Newst. A rare species. From leaves of Malacca apple, Georgetown. NOTES ON THE COCCIDAE OF BRITISH GUIANA. 109 Akermes quinquepori, Newst. A rare species. From beneath the bark of certain trees. On this occasion taken from Macrolobium acaciaefolium in the Botanic Gardens, Georgetown. Never more than one or two are found together in one position. The species is invariably attended by a small black species of Cremasto- gaster ant. When alive it is of a dull pinkish colour closely adhering to the bark. Coccus impar, Ckll. An uncommon species. Name of host-plant unknown. Saissetia begoniae, Douglas. Recorded from British Guiana in 1892 on Begonia sp. It has not been met with since then. Saissetia hemisphaerica, Targ. An exceedingly common species. Its commonest host-plants are Liberian coffee, citrus plants, guava, and several species of ferns, Occasionally found on crotons. Saissetia nigra, Nietn. The commonest species of Saissetia in British Guiana. Recorded since 1891. Host-plants: Hevea brasiliensis, Sapium jenmani, Okra (Hibiscus esculentus), Solanum melongena, and many others. Saissetia oleae, Bern. A common species on citrus plants, crotons, Duranta, etc. Saissetia hurae, Newst. A rare species. A thick infestation was however dis- covered on one occasion on the twigs of a sand-box tree (Hwra crepitans), attended by a species of Cremastogaster ant. Sarssetia scutata, Newst. A rare species. A large colony was discovered in one instance on a cannon-ball tree (Cowroupita guianensis). The infestation covered the curling branches which bear the flowers and fruit near the ground. A species of Cremastogaster ant was in attendance. “hace Ab, ‘ig? an | mw Ga of piauwts, pera: aOR ws t ie a =, = at | A AETS orm, 39, 49000 ba nfatind to hg ait diumend arog, oe 7, 3 Gets fy 8 piopiott off at ‘agerelo\ pasion pie as an soda Ade. oi 140) anit ‘Kauel’ wine a » ait trees “ote <y vy Uy Oe ‘alge MOREE Havas # hee Hebisita otis vee | fs pi nt of waltedtua efsaals win lard q (lint a to 4 he on Ah ve ' adiigy Sc pada 5 aia iv, Bain sf isha be . yet. a0, a Py pani) ‘dative mort bah iin gt nl wiry qT ery bit? noha TPitee 48/06 diane Wom JEL OD Prin ml Mom Apt Ih ah eek bores at Hive) Te eotonay le e¥ee hi .evany elite i Pesta} anes rea OTH 7 atsotirny ay ree rahe F . h4e ry ein danke 7+ as GY. a i) BOIDaTA Tes (OTTO oat ATPat A. ‘ardt) ae init, a lep MANE BT v. fj get, See: “Mott, tear pet RTA Last etl RY gti pe a merit) wing ; if ritJ on 2 at 4S ;. As, | DOT eet) Atoeliy suv oo as cam epee isch Fe rth. Bee ae “gh. sreanad air got MRICALI( Meu? fh ay 948, vein oY pe Awad 4 : Py oe re Bobcat tt ATL Wo must\) sot? 2od-bese e ly ener ad oo cove: LAO £0 etait a 2) bs are 1 SGU ; : . Ain moalebiqoleiantant)’ im. eetiy nd ie ae 4, ete al boy Wopaid. GW Vilod o f ai aif vr SF. Senet cishyshuray ;' bewsveo mobteteatat 4 ot {8 pry Bigs orb Mase ie fet nie ate Oh a aaioaips x -bitiiony ait durGb Dae ets wont t ahi teat doiitve rita wire , | ae — Ts we ma . 1h) Pen, comer tA ne mid < ver mila 2 S43) iin ‘ana jac 2) fae tes i td wich Gat "ae “astine take seertl ap % ty a) Cabbhoneed en ¢ 4 oe ee . ‘ _ ay ; a the Di ¢ ; . Be ‘ A pot “Porno gated nina ta Rertaty Gol yews Y ae Co ely the “speci oe el “alt a ror ee ib od lll Vv SOME INSECTS INJURIOUS TO CACAO PLANTS IN THE BELGIAN CONGO. M. Raymond Mayné, the Government Entomologist in the Belgian Congo, has recently made a special study of the insects that live on cacao in that country, and the Belgian Ministry of Agriculture is shortly publishing his report on the subject. The material collected by him comprises a number of species new to science, some of which are described below by various specialists. All the species dealt with were taken in the Mayumbe district, a region of dense tropical forest, not very far from the coast and lying due north of Boma. Order COLEOPTERA. Family MELOLONTHIDAE. By GILBERT J. ARROW. Aserica variegata, sp. nov. Ferruginea, subopaca, fronte, pronoti macula magna tripartita elytrorumque plagis tessellatis viridinigris ; sat longe ovalis, corpore supra et subtus irregulariter albo-setoso, scutelli lateribus elytrorumque basi densius setosis, partibus obscuribus fere nudis, hic et illic dorso squamis nonnullis majoribus ornato ; clypeo grosse et rugose punctato, antice leviter emarginato, fronte et pronoto parce punctatis, hujus lateribus leviter arcuatis, angulis anticis acutis, posticis rectis, elytris profunde striatis, intervallis convexis; pygidio parce punctato et setoso, linea mediana laevi; pedibus haud brevibus, tibiis anticis fortiter bidentatis, posticis modice latis. 3, oculis magnis, clava antennali ad stipitem longitudine aequali. | Long., 6mm.; lat. max., 3 mm. This insect, according to M. Mayné, devours the young and tender leaves of the cacao. I have seen only a single pair. It is one of the group of species called Lepiserica by Brenske, but included in Autoserica by Péringuey, who has pointed out that no practical means of separating them has been devised. The latter name is also redundant, being synonymous with Aserica. A. variegata is rather elongate in form and decorated like the species of Ewphoresia, but without a sternal process. The elytra are deeply striated and the alternate intervals are decorated with greenish-black patches, almost bare of the setae with which the remaining surface is sprinkled. The posterior part of the head and the middle of the pronotum are also greenish-black, the latter dark area trilobed behind. In addition to the minute setae there are also broad scales upon the scutellum, in the intervals longitudinally dividing the dark patches upon the elytra, and a small cluster on each side of the middle of the pronotum. Pseudotrochalus concolor, Kolbe. This is undoubtedly the insect so-called by Brenske, but the original description is quite inadequate for its certain identification. Its typical form has been well described by Gyllenhal (Schonherr’s Synonymia Insectorum) under the incorrect name of Melolontha versicolor, F. It is remarkably variable in coloration, but normally red-brown, with narrow black margins and five vague black spots upon the 13g INSECTS INJURIOUS TO CACAO PLANTS pronotum, frequently reduced to two placed at the base. The five spots may also coalesce and cover part or the whole of the pronotum, and the elytral border may extend from behind forwards to cover the whole hinder part of the elytra (var. nigromaculatus, Brenske), or there may be an intricate dark network. In one specimen the scutellum alone remains pale, in another this and a red spot on each side of the pronotum, and finally, the entire surface may become black. Triodonta procera, Lansb. I think this is probably Lansberge’s species, in spite of certain inaccordant phrases in the description. It is certainly not nitid and the hairy clothing of the lower surface is distinctly longer, instead of shorter, than that of the upper side, but these discrepancies may perhaps be due to hasty writing. It is strange that the real significance of the remarkable modification of the mentum described by Lansberge has never been recognised. It is certainly not, as supposed by him, a possibly generic feature characteristic.of a few of the species, but is dis- tinctive of the males of nearly all. Like the equally remarkable claw-development in the same sex, it is a prehensile device, its essential feature being the backwardly- directed and extremely strong, close and regularly-arranged spines (generally cleft at the end) with which the enlarged mentum is provided on its outer face. Family CURCULIONIDAE. By Guy A. K. Marsnatiz, D.Sc. Systates ramosus, sp. nov. (fig. 1). Colour varying from black to red-brown, rather thinly and unevenly ee with grev or buff scales so that the integument is largely exposed. 3S. Head dull and shagreened, the forehead almost flat and with a very short deep central furrow, the furrow separating off the rostrum strongly angulate in the middle ; along the inner edge of the eye is a row of 5-7 erect flattened white setae. Rostrum longer than its basal width, sculptured like the forehead, tricarinate, the central carina rather more distinct than the others and bifid in front, the two lateral ones strongly convergent anteriorly, the sides of the rostrum not falling steeply below them, but forming a continuous transverse curve with the median area ; this lateral con- vexity causes the scrobes to be more sharply delimited behind than usual; the genae broadly dilated. Antennae with the scape moderately stout, subcylindrical, of almost the same thickness throughout, slightly curved downwards only at the apical fourth, with rugose longitudinal wrinkles and clothed with subrecumbent broad dark setae and sparse pale scales; the funicle with joints 1 and 2 equal, 3 longer than 4, and 4 to 7 equal. Prothorax much broader than long, the sides strongly rounded, broadest well behind the middle, without any apical constriction, but markedly constricted at the base, which is not much broader than the apex, the latter being shallowly sinuate ; the upper surface comparatively flat, finely shagreened, usually with a short bare smooth central line, and set with very small and widely separated granules, each bearing a short erect scale-like seta. Hlytra ovate, broadest at about one-fourth from the base (being there obtusely angulated) and rapidly narrowing thence to the apex ; the basal margin abruptly raised and forming a sharp transverse ridge, each external angle being produced obliquely forwards into a broad process, 1:2 mm. long, which is IN THE BELGIAN CONGO. 113 broadly truncate at the apex, the external angle being extended downwards to a point, while the basal half of the process is dilated into a broadly trianglar sharply pointed tooth, the greatest width of the process being 1 mm.; the upper surface with regular rows of large, bare and rather tfansverse punctures; the intervals scarcely broader than the punctures, smooth and shiny, and each bearing a row of long erect scale-like setae, which are deeply emarginate at the tip; the scales are small, convex, fluted and almost circular, the apical edge being somewhat flattened. Legs stout, without any long silky hairs ; the front tibiae shallowly bisinuate internally Tassart. Fig. 1. Systates ramosus, sp. n., Mshl., 3g. and having at the apex on the lower surface a broad deep cup-like cavity, the inner basal margin of which is produced into a short lobe bearing four stout spines, with two others above it; the mid tibiae with a shallow longitudinal impression on the lower surface at the apex with three short spines on its anterior edge ;_ the hind tibiae not distorted, the inner surface smooth and flattened almost to the base, its lower edge with small denticulations. Length, 74; breadth, 34 mm. . Differs only as follows :—The elytra are broader and shorter, not nearly so much narrowed posteriorly, the dorsal outline being much more convex and steeper behind ; the apical excavation on the front tibiae is much shallower, and the hind tibiae are not flattened internally, except narrowly in the apical third, both edges being distinctly denticulate. Length, 6-7; breadth, 31-34 mm, (C365) H 114 INSECTS INJURIOUS TO CACAO PLANTS This remarkable species may be readily recognised by the striking basal prominences on the elytra and the cup-like cavities on the front tibiae of the male. This insect is stated by M. Mayné to occur but rarely, the adult attacking the leaves of full-grown trees. Systates maynei, sp. nov. (fig. 2). Colour black, piceous or red-brown, thinly covered with small sub-circular convex fluted grey scales, often having a pale greenish or coppery reflexion. 3. Head separated from the rostrum by an angulated furrow ; the forehead almost flat, shiny and faintly alutaceous, with a deep central furrow, which is about equal to one-half the anterior furrow; along the inner orbit of each eye a row of 5-7 erect flattened pale setae. Rostrum longer than its basal width, gradually narrowing from the base to the middle and broadly dilated at the genae; the upper surface = =H iy & © S55 : SESE 3 oe — PRES ie 2 “i (v, > XS Fig. 2. Systates maynet, Mshl., sp. n., 3. tricarinate, the median carina bifid at the apex, the outer ones strongly convergent in front; the lateral areas not perpendicular, but sloping gradually outwards, the upper portion of the scrobe being therefore sharply delimited behind by an oblique carina. Antennae as described for S. ramosus, except that joints 4 and 7 of the funicle are each slightly longer than 5 or 6. Prothorax a little broader than long (6: 5), the sides rather strongly and regularly rounded, broadest about the middle, the apical margin truncate and only slightly narrower than the base; the upper surface, shagreened, usually with an abbreviated shiny central line, and set with sparse small granules, each bearing an erect scale-like seta. Hlytra ovate, broadest not far from the base and gradually narrowing from there to the apex, the sides being quite straight from the base to the point of greatest width (above the middle of the metasternum) ; IN THE BELGIAN CONGO. 115 the basal margin forming a low straight transverse carina, the outer angles of which are produced into a very short sharp point ; the upper surface with rows of large bare transverse punctures, the intervals narrower than the punctures and hardly broader than the interspaces between them ; each interval with a regular row of dark erect broad apically-emarginate setae. Legs stout and without long hairs; the anterior pairs of tibiae normal, shallowly bisinuate internally ; the hind pair with the inner face strongly flattened in the apical half, its upper edge being there carinate and the lower edge finely denticulate, and the external apical angle not produced. Length, 43-53; breadth, 24-2? mm. ®. The elytra are much more rotund and less narrowed behind, and the very convex dorsal outline is steeper near the apex ; the hind tibiae are not flattened internally and are more conspicuously denticulate. Length, 43-53; breadth, 23-3 mm. In general form like a small specimen of S. fossulatus, Klb., which has the base of the elytra similarly constructed ; but the latter species differs in the following respects :— the rostum is gradually widened from base to apex, the outer dorsal carinae are almost parallel, the first joint of the funicle is distinctly longer than the second, the granules on the prothorax are much larger and shiny, all the tibiae of the ¢ are fringed beneath with long silky hairs, the outer apical angle of the hind pair is produced into a broad acute-angled process, ete. This weevil is abundant in cacao siiabtons being especially injurious to young plants in the nurseries. The injury is caused by the mature insect, which devours the leaves, eating large portions out of them along the edges. Nothing is known as to the early stages of this or the preceding species. Alcides theobromae, sp. nov. (fig. 3). : 9. Colour shiny black and almost bare above; the prothorax duller, with a transverse lateral band of pale recumbent hairs at the base and near the apex ;_ the elytra with the following markings formed of similar hairs: a very small spot at the shoulder and near the base of interval 2, a spot on interval 4 a little before the middle, an irregular transverse row of spots behind the middle on intervals 2, 4, 6 and 8, and a patch in the pre-apical depression ; the lower surface somewhat densely clothed with tawny feathered scales. | The markings are evidently very easily abraded. Head dull, finely aciculate on the vertex, the forehead with shallow confluent punctation and shallowly impressed between the eyes. Rostrwm about as long as the front femur, moderately stout and almost straight, the dorsal outline straight for three-fourths the length and then sloping gently ; the upper surface fairly closely punctured throughout, but less so towards the apex, and without any furrows or carinae, the punctures much larger and longitudinally confluent at the sides on the basal half. Antennae stout and black ; the funicle with the two basal joints short, 1 longer and thicker than 2, 3-5 subequal and strongly transverse, 6 much longer than 5 and nearly quadrate, 7 hardly as long as the first two joints of the club, the first joint of which is bare and shiny on its basal half. Prothorax a trifle shorter than its basal width, broadest at the base, rather strongly narrowed in front, distinctly con- stricted near the apex, the sides moderately rounded ;_ the dorsal outline very convex, the downward slope of the front half being much steeper than any part of the elytral 116 INSECTS INJURIOUS TO CACAO PLANTS slope; the upper surface closely and evenly set with small low shiny granules of various sizes. Scutellum transverse, smooth and impunctate. Hlytra broadest at the shoulders, where they are only slightly broader than the prothorax, and gradually narrowing from there to the apex ;_ the dorsal outline rising very slightly from the base to one-fourth the length, then gradually sloping backwards and becoming steeper in the apical third ; the basal lobes longitudinally wrinkled, and a shallow depression involving the bases of striae 1 and 2 behind the scutellum ; the striae very shallow on the disk, rather deeper at the sides and apex, the punctures moderately large, ine Fig. 3. Alcides theobromae, Mshl. sp. n., 9. their septa narrower than the intervals, which are smooth and shiny and bear very small scattered punctures, except on interval 1 and the juxta-apical area, which are rather rugosely punctate; a distinct epipleural carina from the hind coxae to the apex, the epipleurae being oblique in front, but quite horizontal at the apex and there bearing coarse confluent punctures. Legs moderately stout, all the tibiae with a sharp internal tooth near the apex, the front pair with a broad triangular tooth about the middle, which is represented by a feeble angulation on the middle pair. Sternum: the metasternum granulate and with a large prominence on each side in front of the hind coxae. Length, 8; breadth, 34 mm. Very closely allied to A. cultrirostris, Thoms., which differs in the following points :— The elytra are mahogany red, the tibiae lack the inner apical tooth, the anterior slope of the prothorax is less steep, and the punctures on the disk of the elytra are much larger, so that their septa are scarcely narrower than the intervals. The larvae of this weevil bore longitudinal galleries in the small branches of cacao trees; the leaves subsequently turn yellow and the branch dies. The beetle is not as yet sufficiently numerous as to be regarded as an important pest. IN THE BELGIAN CONGO. 117 Family LAMIIDAE. By C. J. Ganan, M.A., D.Sc. Tragocephala maynei, sp. nov. T. nobili (Fab.) affinis, sed differt capitis medio haud vittato, elytris prope basin non transversim fasciatis. Capite inter antennas sat lato fere plano, margine antica frontis genarumque et macula pone oculum coeruleis; prothorace utrinque ante tuberculum lateralem ochreo-fulvo fasciato, basi lateraliter coerulescente ; elytris nigro-velutinis, utrisque maculis quatuor ochreo-fulvis notatis: macula prima ad latus paullo pone humerum, secunda transversa et tertia parva rotunda obliquiter positis pone medium, quarta transversa paullo ante apicem; macula infra humerum, puncta postero-laterali, et macula minuta ad suturam vix ante apicem coerules- centibus ; corpore inferiore pedibusque coerulescentibus, nigro fasciatis et maculatis ; antennis (2) paullo pone medium elytrorum transeuntibus, scapo ad apicem angustim cicatricoso ; prosterno sat lato, antice verticali et transversim truncato, mesosterno antice late rotundato ; femoribus brevibus subclavatim incrassatis. Long., 20; lat., 7 mm. This species is nearly related to Tragocephala nobilis, Fab., with which it agrees in general form and structure, the only difference noticeable in this respect being the somewhat broader intercoxal process of the prosternum in the present species. It differs however in markings ;_ the head is without a median band, the genae and front are bordered below with greyish blue pubescence and there is a patch of the same colour behind each eye; the prothorax is banded at the sides as in 7’. nobilis, but with the band in front of the lateral tubercle more ochreous or brownish in colour ; the elytra are marked each with four ochreous tawny spots—one at the side a little behind the shoulder, two placed obliquely a little behind the middle, and the fourth transversely a little before the apex ; each has, in addition, a bluish spot below the shoulder, a blue lateral speck at about one-fourth from the apex, and a speck on the suture just before the apex. Body beneath and legs glaucous blue, banded with black. Exocentrus ortmansi, sp. nov. Niger sat dense griseo-pubescens, antennis dense setosis quam corpore paullo longioribus, articulis ab 3° cinereo anguste annulatis ; prothorace transverso, spinis lateralibus pone medium positis, retro directis ;_ elytris nigro-setosis, setis in seriebus octo utrinque ordinatis, interstitiis bi- aut uni-seriatim sed haud regulariter punctatis, punctis post medium evanescentibus. Long., 6; lat., 2°5 mm. | Black, with a rather dense covering of grey pubescence. Antennae a little longer than the body, black with a rather dense fringe of black setae beneath, the joints from the 3rd narrowly ringed with grey at the base. Prothorax with a few scattered black setae, lateral spine placed behind the middle, and directed towards the shoulder of the elytron. The elytra, seen in certain lights, exhibit a faint brownish band behind the middle ; each has eight longitudinal series of black setae, and on the basal half the intervals between the rows of setae are punctured in rather irregular rows, the rows in some parts being double and in others, especially farther back from the base, single. 118 INSECTS INJURIOUS TO CACAO PLANTS IN THE BELGIAN CONGO. Order RHYNCHOTA. By W. L. Distant. Family CoREIDAE. Pendulinus devastans, sp. nov. Above with the head, pronotum, scutellum and corium ochraceous, thickly darkly punctate ; narrow margins to the pronotum and scutellum and the margins and venation to cortum pale ochraceous and impunctate; membrane shining bronzy- brown; body beneath and legs pale ochraceous; head beneath and sternum thickly but not darkly punctate, a small shining black spot on each side near outer margins of posterior coxae ; tibiae sometimes moderately roseate in hue; antennae reddish-ochraceous, apical joint either paler in hue, or testaceous with the bases and apices paler, basal joint moderately thickened and a little curved outwardly, its apical area somewhat incrassate, first and second joints almost subequal in length, second and third joints slender, second longer than third, fourth incrassate, subequal in length to third; head above centrally longitudinally incised; lateral pronotal angles subprominent ; venation of corium robust. : Long., 13 ; breadth between pronotal angles, 4 mm. Immaiure forms. Ist stage-—Head and pronotum fuscous-brown ; abdomen above more or less roseate in hue; body beneath and legs ochraceous; antennae strongly incrassate, the third joint distinctly dilated. Long., 5-7 mm. 2nd stage-—Above more or less pale castaneous, on anterior and posterior margins of fourth abdominal segment, a smal] rounded _pro- truberance each containing two small shining black tubercles, and another black tubercle on each side near the lateral margins of the segment ; antennae as in first stage. Long., 9 mm. The perfect insect is allied to P. carmelita, Burm., from which it differs by the more robust and differently coloured antennae, absence of the black fasciate markings beneath, etc.* * I add a description of another a.lied species found in Natal, which is probably also a destructive insect :— Pendulinus nigromarginatus. Allied to P. carmelita, Burm. and C. devastans, Dist., from both of which it differs in having the lateral margins of the pronotum and more than basal half of the sub- late:al margins of the corium, black ; antennae ochraceous, basal joint stoutest and darkest, distinctly curved and considerably longer than the second joint, which with the third joint is slender, fourth jot mutilated ; body beneath pale ochraceous ; legs and rostrum dark ochraceous ; other characters generally as in P. devastans. Long., 11-12 mm. | Hab. Natal, near Durban (Bell Marley). 119 COLLECTIONS RECEIVED. The following collections were received by the Imperial Bureau of Entomology between Ist January and 3lst March, 1917, and the thanks of the Managing Com- mittee are tendered to the contributors for their kind assistance :— Dr. W. M. Aders :—386 Culicidae ; from Zanzibar. Mr. G. E. Bodkin, Government Economic Biologist :—2 Culicidae, 5 other Diptera, 2 Siphonaptera, 31 Hymenoptera, 12 Coleoptera, 2 Trichoptera, 3 Mallophaga, 5 Rhynchota, 5 Orthoptera, 1 Mayfly, 1 Dragonfly, 9 Ticks, 1 tube of intestinal worms, 1 Centipede, 4 Scorpions, and | Spider; from British Guiana. Mr. J. R. Bovell, Superintendent of Agriculture :—2 Diptera, 4 Coleoptera, 3 Lepidoptera, 2 Rhynchota, 1 Cockroach, and 1 Spider; from Barbados. Mr. D’Emmerez de Charmoy, Government Entomologist :—98 Trypetidae and about 50 pupa-cases, and 3 other Diptera; from Mauritius. Mr. E. C. Chubb, Curator of the Durban Museum :—44 Culicidae, 2 Hippoboscidae, 187 other Diptera, and 65 Coleoptera ; from Natal. Dr. J. B. Davey, Medical Officer :—68 Culicidae, 5 Tabanidae, 4 Glossina, 36 other Diptera, 23 Hymenoptera, 13 Coleoptera, 3 Planipennia, 4 Trichoptera, 2 Isoptera, 4 Mallophaga, 10 Rhynchota, 17 Orthoptera, 2 Odonata, 100 Ticks, 1 Centipede, 2 Scorpions, and 4 Entozoa; from Nyasaland. The Division of Entomology, Pretoria :—15 Diptera, 1 Moth, 14 Hymenoptera, 90 Coleoptera, 27 Rhynchota, and 1 Grasshopper ; from South Africa. Mr. P. R. Dupont, Curator of the Botanic Station :—4 packages of Coccidae and fungi; from Seychelles. Mr. W. W. Froggatt, Government Entomologist, New South Wales :—126 Diptera, 32 Chalcidids, 6 other Hymenoptera, and 14 Coleoptera ; from various Australasian localities. Mr. C. C. Gowdey, Government Entomologist :—3 Culicidae, 4 other Diptera, 2 Lepidoptera, 60 Chalcidids, 98 other Hymenoptera, 243 Coleoptera, 20 Thysanop- tera, 6 packets of Coccidae, about 100 Aphididae, 117 other Rhynchota, and 17 Orthoptera ; from Uganda. Mr. E. Hutchins, Chief Veterinary Officer :—1 Tabanus and 14 Haematopota ; from Uganda. Mr. R. W. Jack, Government Entomologist :—8 Diptera and 13 Moths; from Rhodesia. Mr. F. P. Jepson, Government Entomologist :—About 100 Diptera ; from Fiji. Mr. H. H. King, Government Entomologist, Anglo-Egyptian Sudan :—3 Ants ; from Khartoum. 120 COLLECTIONS RECEIVED. Dr. W. A. Lamborn :—51 Tabanidae, 405 Glossina, 14 Hippoboscidae, 270 other Diptera, 1 Moth, about 80 Chalcidids, 117 other Hymenoptera, 52 Coleoptera, 48 Rhynchota, 4 Harwigs, 1 Stone-fly, 8 Ephemeroptera, and 2 Odonata ; from German Kast Africa. Mr. 8. Leefmans, Laboratorium von Plantenzeikten, Buitenzorg :—27 Coleoptera ; from Java. Prof. R. Newstead, F.R.S.:—About 50 Ants, from Jamaica; and 4 bred Chalcidids. Capt. J. E. T. Phillips :—50 Haematopota ; from Uganda. Dr. J. J. Simpson :—360 Tabanidae, 1,211 Glossina, 497 Hippoboscidae, 1,012 other Diptera, 338 Hymenoptera, 247 Coleoptera, 1,158 Rhynchota, 9 Odonata, and 3 Orthoptera ; from the Gold Coast. Mr. R. Veitch :—38 Diptera, 1 Moth, 1 Caddis-fly, 20 Chalcidids, 21 other Hymenop- tera, 47 Coleoptera, 3 Planipennia, 6 Rhynchota, 10 Orthoptera, 1 Dragonfly, 2 Arachnida, and 1 Centipede; from Fiji. Mr. C. B. Williams :—7 Tabanidae, 4 Hippoboscidae, 9 other Diptera, 1 flea, 1 moth, 10 Chalcids, 235 other Hymenoptera, 60 Coleoptera, 5 Mallophaga, 106 Rhynchota, and a tube of Entozoa; from British Guiana. The Wellcome Bureau of Scientific Research :—2 Hymenoptera, 18 Coleoptera, 77 Lepidoptera, 2 Rhynchota, 7 Orthoptera, and 1 Dragonfly ; from Venezuela. Mr. R. C. Wood :—14 Tabanidae, 1 Asilid and its prey, 132 Trypetidae, 179 Hymenoptera, 3 Coleoptera, and 3 Spiders; from Nyasaland. —_ Aer i ee EN Yt ge ca IMPERIAL BUREAU OF ENTOMOLOGY. "HONORARY COMMITTEE OF NAN ACEMENT THe VISCOUNT HARCOURT, Gaieman, -Laeut.-Coronzt A. ALCOCK, CLE, iia | Me. E. E. AUSTEN. apne ay Dr. A..G, BAGSHAWE, CLG. i ae Mp: C. BLECKS OMG) ena. Sm JOHN R.: BRADFORD, K.C.M.G., FERS. Sune.-GenrraL Sir DAVID ‘BRUCE, C. Bs ‘F. R. 9. Ma’ J; CF. FRYER.» Dr. §. F. HARMER, F.RS. Pror. H.. MAXWELL LEFROY. SE Sir JOHN MoCALL. ops . Dr. R. STEWART MACDOUGALL. $tr JOHN McFADYEAN. -~ Sp PATRICK MANSON, G.C.M.G.,. PRS ‘Sm DANIEL MORRIS, KOMG Pao, RB. NEWSTEAD, F.RSS souseran ie ae - Prom. G. H. F. NUTTALL, ERS. Pe RNC Elie Pror. E. B. POULTON, F.RS. ) oat. Lirur.-Corowe. Sm DAVID PRAIN, Cue. OLE., BRS. te Mr. H. J; READ, C.B., C.MLG. 7” | Tae Hon, N. C. ROTHSCHILD. Mn HUGH SUOTT Shien ns a Dr. A. BE. SHIPLEY, F.R.S. RT a - $m STEWART STOCKMAN. ‘sig i Mr. ¥, V. THEOBALD. a Ge eS “Mz. C. WARBURTON. . is yos Vara ‘ ye maviodigen te cous’ UO BC otON. Som teal ie ce eee | ; “Dr. GUY A,.K. MARSHALL. as 8 bh Ontos: if “Assistant Director. By ee AN ‘ ENS Mr. ay NEN Heh ‘Secretary; ‘Mr, AL vor Cc. ‘PARKINSON. oy / b Na He 121 ON THE SUPPOSED FIRST STAGE LARVA OF LEPTOHYLEMYIA COARCTATA, FALL. By D. Kerry, D.Sc. (From the Quick Laboratory, University of Cambridge.) It has been well known since the studies of Taschenberg (1864-1872) that the larvae of Leptohylemyia coarctata, Fall., attack wheat and rye. The damage due to this fly has been observed many times in almost all European countries, and many papers have been devoted to its life-history. Of these papers the most important are those of E. Ormerod (1882-1895), 8. Rostrup (1905-1911), T. Hedlund (1906— 1907), P. Marchal (1909) and finally the recent work of Kurdjumovy (1914). We owe to Hedlund (quoted by Kurdjumov, p. 25) the most complete account of the development of this fly. According to this author, copulation takes place with this species at the beginning of July. The examination of the ovaria of females captured in the field at different times has led him to the conclusion that the deposi- tion of eggs takes place between the Ist and 22nd of August. Hedlund has also observed that eggs of this fly, which were laid in earth in the breeding vessels at the beginning of August, contained in September a completely formed larva which in December had not yet hatched. In this way it seems that he has been able to confirm the supposition of Ormerod and of Rostrup that L. coarctata has only one generation per annum ;_ that it deposits its eggs towards the end of summer or at the beginning of the autumn; that the eggs, which are always laid in the earth, hibernate; and that the larvae do not hatch till the spring. These results were accepted by Kurdjumov (1914) in his recent publication on this fily.* According to this author’s personal observations the newly hatched females are never sexually mature and they pass the whole summer in this state of immaturity, so that oviposition takes place only at the beginning of the autumn. Kurdjumov’s paper contains also the description and the figures of the egg and primary larva of L. coarctata. This is of special interest, because of all the life-history of this insect the part that is most disputed and the most difficult to observe, but which is of great economic importance, is precisely the beginning of the cycle, 7.e., the oviposition and the first-stage larva. From his description and figures it would seem that Kurdjumov produces definite data relating to the beginning of the cycle. We shall now consider under what conditions this primary larva was discovered by Kurdjumov. During the whole summer he kept in his breeding vessels both males and females of L. coarctata. On 28th August all the flies that he had bred were, with one exception, killed by a fungus. All the females which perished had undeveloped ovaries. The | * The reader is referred to this paper (in Russian) for full references and for a critical examination of many papers dealing with this insect. Kurdjumov, N.V.: Adia genitalis, Schnabl, and Leptohylemyia coarctata, Fall., Reports of the Agricult. Exper. Station of Poltava, Branch of Agric. Entomol., 1914, No. 21, part ix., pp. 1-43. [Abstracted in Rey. Appl. Ent. ii, 1914, p. 250.] (C394) Wt.P7/121. 1,000. 11.17. B.& F.Ltd, G.11/1, A 122 D. KEILIN. single female still living had a rounded abdomen, which showed that its ovaries were well developed. This female was put in another breeding vessel, and the dissection of this fly, done three days later, showed that its ovaries were perfectly matured. The breeding vessel (terrarium) was well examined. Kurdjumov found no eggs on the plants, but in the earth he found at no great depth two eggs similar to those discovered in the lower chambers of the ovary of the dissected fly. In one of these eggs the black mouth-parts of the larva were visible through the egg-covering. On 9th November this egg was taken to the laboratory and on 12th November the larva protruded the anterior part of its body from the egg and died in this position. Kurdjumov inferred from the facts (a) that the frost did not kill the egg, and (6) that as the larva hatched as soon as it was transferred to a warm place, it is the egg which represents the hibernating stage of L. coarctata. He next describes the first stage larva; but his description, and especially his figure 26, convince me that neither eggs nor larva belong to the developmental cycle of L. coarctata. Fig. 1. Mouth-parts of Kurdjumov’s primary larva. {It is quite sufficient to look at Kurdjumov’s fig. 26, which I reproduce here (fig. 1), and which represents the mouth-parts of his primary larva, to establish from all the details of the structure he delineates that it pertains to a carnivorous larva.* If we compare the mouth-parts of this supposed first-stage larva with those of a full-grown larva of Leptohylemyia, we can easily see that the difference is not one which we usually find between the primary and third stage larva of the same species. The difference is such as can only exist between two different ethological groups. As figure 2 shows, the mouth-parts of a full-grown Leptohylemyza larva are very like those of one of the phytophagous Anthomyid larvae (Pegomyia ngritarsis, or even those of Adia genitalis, discovered by Kurdjumov) ).7 The first stage Leptohylemyra larva should be then very similar to the peunery phytophagous larva and it probably resembles closely that of Adia genitalis (fig. 5 of Kurdjumoy). Kurdjumov’s description contains two other points which show that his primary larva does not belong to Leptohylemyia, for he states: (1) that the last abdominal segment is round and without the protuberances which are very characteristic for the full-grown larva of Leptohylemyia ; (2) that the posterior spiracles of his primary larva have three clefts and closely resemble those of a third stage cyclorrhaphous *See: D.Keilin. Recherches sur les Anthomyides a larves carnivores. Parasitology, x, nos, + Iam much indebted to Messrs. C. Warburton and F. R. Petherbridge for the larvae of Leptohylemyia coarctata herein referred to. THE LARVA OF LEPTOHYLEMYIA COARCTATA. 123 larva. This demonstrates the existence in this insect of development with almost obliterated trimorphism, such as characterises carnivorous Anthomyid larvae. We may therefore conclude from (1) the structure of the mouth-parts, (2) the shape of the posterior end of the abdomen, and (3) the structure of the posterior spiracles, Fig. 2. Mouth-parts of a full-grown larva of Leptohylemyia coarctata. that the primary larva found by Kurdjumov does not belong to the developmental cycle of Leptohylemyva coarctata, Fall. On the other hand it is quite easy to suppose that the two eggs found by Kurdjumov in his breeding vessel were brought in with the earth from outside. (C394) A2 otal" My ee a rears i) y ee a ets 4, an ? J os i. @ i, es as xh 7 Oe ee 10 ARGH, dyysitie diver ; eee zah, Ly toynad aide mb onaysel nad a SR TIAL bre uw fata ugescaciin a Saver'tead: a | peed orth (2) asimpalticans 20 90 tuned adh Ch) airvat af Galnctiqa, 19 (tien, ah So eredourle old) Dag o suri a a ay Ashe cn! . i, pes Way ‘on ‘Oy ahha wir he ay ea if’ y 4%, ates see fe naps ay ni) Ge - a a tind Goll Fr ais ee ¥ ea" , we. yp Peetsesy cers Gy a) we ai Fo | Wes karst} nih OL! 7 esa? . Pee es iepy Edt dinneet waite sd ie hen avn ened ie ‘ 7 ; — farw BL FARO TR wie? wie rit aicl om now nag, ma heap agge owe : | gbisiue CO aN Vin j ; : 2, ee pean ey Shae wpnuiry deerg, i eete hehe | . yi ‘2 a nr ne phere dl its is “gcse | . neler peairts Vv; rhs a ‘hae » . eleng “4 aa “4 eee ‘bo bili Ata -. Sy cutis | ath Mii oy erga ers eas ee oie hae? va! ie “eed “"y : > 7 OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—V. By Rosert Newsteap, F.R.S., The School of Tropical Medicine, The University, Liverpool. Aspidoproctus armatus, Newst. PortucuEsE Conco: San Salvador; on tree of unknown species (Dr. Mercier Gamble). Aspidoproctus ? glaber, Lind. NyasaALANp: Mlanje; on “ mwanga” tree, x.1913 (S. A. Neave). A small undersized female, unfortunately imperfect. I have placed it here, as it agrees, apart from its size, with Lindinger’s description. Aspidoproctus pertinax, Newst. NyasaLtanp: Mlanje; on “ kalati” tree, x.1913 (S. A. Neave). Female, young adult: Broadly ovate and highly convex above; covered with a relatively thin layer of very pale yellowish-buff secretion, which is Somewhat granular in texture. Margin with a closely set series of large waxen rectangular appendages, a few of which are completely divided longitudinally ; dorsum, in the middle line, in front with three pairs of short blunt waxen processes, each surrounded by a fringe- like series of relatively large, granular pieces of wax. Midway between the short dorsal series of processes and the margin is another complete row of tooth-hke pro- cesses, all of which have a granular fringe. Length 7°5 mm. The above description was drawn from a specimen which had been preserved in alcohol. Fortunately Mr. Neave collected examples which show the intermediate stages between the quite young adults and the fully matured female, so that there can, I think, be no doubt as to the specific identity of his specimens. Aspidoproctus ? pertinax, Newst. SourH Inp1a: Bangalore, Lal Bagh; on a wild plant, vi.1908 (7. V. Rama- krishna, per HK. Ballard). A single and apparently young adult female, which differs externally from typical African examples in having the posterior marginal processes slightly longer and so arranged that they partly overlap. The larvae are inseparable from the cotypes of A. pertinax in the author’s collection. More material is needed for an exact specific determination of the adult female. Aspidoproctus ? tricornis, Newst. NyasaLanp: Mlanje; on “ mwanga” tree, x.1913 (S. A. Neave). A single immature female which apparently belongs to this species. It differs from the adult in having three relatively small, dorsal, horn-like processes, of which the median one is much the longest and faintly bifid at the tip. Marginal tubercular projections bifid along the abdominal area; the flat tooth-like appendages arising from these are, when present, well separated and divergent; some of the appendages on the thoracic margin are broad and flat, with truncated ends; the four cephalic ones bluntly rounded, and with more or less pointed processes. 126 ROBERT NEWSTEAD. Icerya aegyptiacum, Doug. ZANZIBAR: on young date palm, 15.iv.13 (Dr. W. M. Aders). This is the first record of the occurrence of this pest in Zanzibar. The examples were nearly all young females. Icerya maxima, Newst. Ucanpa: Ngamba Is., Lake Victoria; on tree trunk, 28.v1.14 (Dr. G. D. H. Carpenter). These specimens have a larger number of glands than typical examples ; otherwise the structural characters are specifically identical. Unfortunately the external coverings had almost completely macerated in the preservative, so that nothing can be added regarding these. Icerya seychellarum, Westw. UcanpnA: Entebbe; on mango, 10.11.14 (C. C. Gowdey). Icerya sulfurea, Lind. Ucanpa: Entebbe; on Eranthemum, 4.1.13 (C. C. Gowdey). Icerya ?sulfurea, Lind. Ucanpa: Kampala; on Castilloa, 15.vin.13 (C. C. Gowdey). Stictococcus formicarius, Newst. Ucanpna: Entebbe; on Ficus sp., 4.xu.12 (C. C. Gowdey). Stictococcus multispinosus, Newst. Gotp Coast: Aburi; on kola (W. H. Patterson); a small percentage of the females were attacked by Chalcidid parasites. Ucanpa: Kampala; on Mark- hamia platycalyx, 2.1x.15 (C. C. Gowdey); 29, 2-3 J puparia and an imperfect J, the last two being new. Stictococcus sjostedti, Cll. GoLtp Coast: Aburi; on cacao (W. H. Patterson). Pseudococcus citri, Risso. British East Arrica: Nairobi, Old Government Farm, 5,500 {t., 3.1.13 (Geo. W. Evans). Ucanpa: Entebbe, on coffee, 13.vi.12; Tero Forest, on an un- known shrub, 13.vu.12 (C. C. Gowdey). Pseudococcus sacchari, Ckll. Mapras Pres.: Megapatam; onrice,xi.1914(7. V. Ramakrishna per E. Ballard). British Guiana: Georgetown; on sugar cane, 27.vu.16 (G. #. Bodkin). The distinguishing features of the adult females are: the presence of a pair of dark brown or piceous plates between the 2nd and 3rd free abdominal segments on the venter ; the great length of the transverse coxal sclerite on the anterior legs ; and the long bristles on the segment above the anal lobes. The second-stage females are similar, but the anterior coxal sclerites are slightly longer relatively, as compared with the width of the coxae; and there is an additional group of long abdominal hairs above the anal lobes. ~ OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—V. 127 Pseudococcus hymenocleae, Ck]. South Arrica: Onderstepoort, Transvaal; food-plant not stated, 1914 (D. d Emmerez). ft “ Pseudococcus virgatus, (kl. Ucanpa: Nama Konkoni, Chagwe; on coffee (C. C. Gowdey). ZANZIBAR: on cassava and sugar-cane, 4.iv.13 (Dr. W. M. Aders). Gotp Coast: Aburi, on French beans, Jatropa curcas and Colocasia, 1.1.13 (W. H. Patterson); Accra, on oleander and a creeper called “Salonica,’ 2.vin.16 (Dr. J. W. Scott Macfie). SouTHERN Nigeria: Ibadan, on cacao seedlings, 5.xu.11 (A. D. Peacock). Phenacoccus insolitus, Green. Mapras: Saidapet; on egg-plant (Solanum melongena), 13.vu.07 (TL. V. Ramakrishna, per HK. Ballard). Mr. EK. E. Green in his original diagnosis (Mem. Dept. Agric. Ind., ii, p. 26, 1908) states that the 8th and 9th segments of the antennae are “fused together.” In several of my examples this character is very pronounced ; in others, however, the articulation is as clearly defined as the rest. Tachardia decorella, Mask. : Ueanpa: Entebbe; on Anona muricata, heavy infestation, 20.11.13 (C. C. -Gowdey). SoutH Arrica: Pretoria, 1914 (D. d@ Emmerez). Asterolecanium coffeae, Newst. British East AFrrica: on coffee plants seven miles N.E. of Nairobi, 9.xii.12 (Geo. W. Evans). Ucanpa: Nakasanja, Chagwe, on coffee, 10.vi.13; Banda, on coffee, 28.vu.16 (C. C. Gowdey)—heavy infestations. Mr. Evans, in a note attached to his specimens, states that this Coccid infests every part of the tree, causing the leaves to droop. He found twenty plants attacked by it. A small percentage of the specimens are infested by Chalcidid parasites. The general colour of the tests of the females is pale yellowish-green. In Mr. Gowdey’s specimens the colour of the test of both old and young adults is pale bottle-green, shaded with bright yellowish-green or golden green ;_ the filaments forming the fringe, crest, etc., dull pink. In the adult females the anal lobes are furnished with a single long bristle; these had evidently been broken away in the type material described by me in 1911.* Asterolecanium bambusae, Bdv. Gotp Coast: Aburi; on bamboo (W. H. Patierson). UGanpa: Entebbe; on bamboo, 7.11.14 (C. C. Gowdey). Soutn Arrtca: 1914 (D. d Emmerez). Cerococcus hibisci, Green. C. hibisci, Green, Mem. Dept. Agric. India, ii, no. 2, p. 19, pl. 1 (1908). S. Inpra: Guntur; on egg-plant, 20.11.10 (7. V. Ramakrishna, per E. Ballard). The tests of the females agree best with those forms which are described by Green (l.c.) as having a “ golden-yellow scurfy tomentum,” but the tomentum in my * Mitt. Zool. Mus. Berlin, V. 2, Pp; 161, fig. 4a. 128 ROBERT NEWSTEAD. examples has almost entirely disappeared, owing apparently to detrition, and has assumed a distinct scurfy appearance, which is also patchy and irregular, or arranged in ill-defined transverse bands. Ceroplastodes cajani, Mask. S. Inp1a: Coimbatore; on red gram; on Zizyphus, Feb. 1909; on Ocimum sanctum, Feb. 1910; Anantapur Dist., on wild indigo, April 1907 (7. V. R., per EK. Ballard). Inglisia chelonioides, Green. S. Inp1a: Coimbatore; on Parkinsonia aculeata, 27.vi.14 (T. V. Ramakrishna, per E. Ballard). Inglisia conchiformis, Newst. Gotp Coast: Aburi; on Gliricidia maculata, 12.1.16(W. H. Patterson). ‘‘ Native host could not be traced. All available material forwarded. Difficult to dry. No parasites found. No economic damage, as only one plant infested.” (W. H. P.). Ceroplastes actiniformis, Green. S. Inp1a: Coimbatore; on coconut, x.1908 (7. V. R., per KE. Ballard). Ceroplastes africanus, Green. SoutH Arrica: 1914 (D. d@Emmerez). Ucanpna: N.E. of Lake George; on acacia, 20.vi.11 (C. C. Gowdey). The nipple-like process on the dorsum of the waxen test, characteristic of typical examples of this insect, is wanting in all Mr. Gowdey’s specimens, the position of these processes being indicated by a rather deep depression. Ceroplasies ceriferus, Anderson. 8. Inpia: Coimbatore; on “ wild elm,” x.1909 (7. V. &., per E. Ballard). Ceroplastes cirripediformis, Comst. British Guiana: Berbice, on Hura crepitans, in association with Lecanium (Akermes) sp., 27.x1.13; Georgetown, on [pomea sp., 3.1.14 (@. #. Bodkin). Geroplastes denudatus, Ck1l. British Guiana: Demerara; on a wild species of Solanaceous plant, x.1915 (H. W. B. Moore). Ceroplastes ficus, Newst. SoutH Arrica: Wonderboom; on Ochra pulchella, 1914 (D. @ Emmerez). Goup Coast: Aburi; on Anona (W. H. Patterson). Mr. Patterson’s 9 9 are larger than those of the type lot, which came from Uganda (Bull. Ent. Res., i, p. 191, fig. 5). Ceroplastes personatus, Newst. Gotp Coast: Aburi; on Coffea lberica, 3.x11.15 (W. H. Patterson). This is the only additional record since its discovery in 1898. OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—V. 129 Ceroplastes quadrilineatus, Newst. Uaanpa: Kimi Is., Lake Victoria, v.1914 (Dr. G. D. H. Carpenter). Ceroplastes rubens, Mask. S. Inpra: Ganyon District, Chicacola; on mango, iv.1910 (7. V. R., per K. Ballard). Zanzripar: Maruhubi; on young orange trees, 16.vu.13 (Dr. W. M. Aders). Ceropiastes ugandae, Newst. Ucanpa: Mount Mubendi; on an unknown tree, 1.x.10 (C. C. Gowdey). It is necessary to call attention to the fact that this insect so closely resembles Ceroplastes egbarum subsp. fulleri, Ckll., in the form and colour of the waxen test that the two insects are apparently inseparable. But the 9 of C. ugandae may at once be distinguished by the presence of three large bi-lateral tubercles, all of which are distinctly larger than the rudimentary caudal process. “‘ The broadly ovate group of circular pores,’ described by me* as occurring immediately above the stigmatic clefts, are really not pores but minute conical spines. In a well cleared specimen which is now before me, these structures are very distinctly displayed. Ceroplastes vinsonioides, Newst. Ucganpa: Mabira Forest; on coffee, 28.xu.12 (C. C. Gowdey). Pulvinaria burkilli, Green. 8. Inp1a: Coimbatore; on Zizyphus sp., 19.11.13 (1. B. Fletcher). Two imperfect females and many male puparia. One example has the normal number of segments to the antennae (eight), while the other presents only seven, of which the 4th is nearly as long as the 3rd. Pulvinaria cupaniae, Ckll. JAMAICA; on Ficus sp., 1916, heavy infestation; “on mulberry imported last year from U.S.A.,” v.1916 (A. H. Ritchie). Pulvinaria jacksoni, Newst. Ucanpa: Kampala; on Aranthemum bicolor, 4.vin.15 (C. C. Gowdey) ; chiefly young adult 2° without ovisacs. British East Arrica: Nairobi; on granadilla, 11.1914 (7. J. Anderson). ‘ Fulvinaria psidii, Mask. Ucanpa: Mabira, on coffee, 15.x.10—heavy infestation; Mpumu, Chagwe, on coffee, 12.xu.12 (C. C. Gowdey). British East Arrica: Kikuyu, on coffee, 1.1914 (T. J. Anderson); Bukoba, on “ nsambyia,” 8.vi.12 (C. C. Gowdey). S. Inpta: Coimbatore, on mango, Feb. 1914; Koilpati, on guava, x.1907 (7. V. R., per K. Ballard). Ten of the Mabira females were prepared for microscopical examination ; all of these had their bodies completely traversed by the hyphae of a fungus, which I believe to be a primary parasite. * Bull. Ent. Res., ii, p. 94, fig. 8¢ (1911). 130 ROBERT NEWSTEAD. The derm cells in these examples from East Africa appear somewhat larger and less. oval than I have seen in other specimens from Africa ;_ possibly due to the age of the individuals. Lecanium (Saissetia) cuneiformis, Leon. British Hast ArricA: Mua Hills; on Acocanthera sp., 1.1914 (1. J. Anderson). Lecanium (Saissetia) hemisphaericum, Tare. British East Arrica: Limosa, 1.1914 (7. J. Anderson). UGanpa: Nagunga, on Aristolochia, 20.1x.15; Entebbe, on Adiantum, 10.i1.14—curiously stunted forms, such as are not infrequent on the stems cf Adiantum; Mabira, on coffee, 29. vu. 13 (C. C. Gowdey). Lecanium (Coccus) hesperidum, Linn. British East Arrica: near Mombasa, on banana,1.1914; Mua Hills, on lemon, 1.1914 (T. J. Anderson). Lecanium (Saissetia) nigrum, Nietn. 8. Inpia: Coimbatore; on Lawsonia alba, iv.1910, and on cotton, very heavy infestation, vii.1910 (7. V. R., per E. Ballard). UGanpa: Mwera; on coffee, 29.vu.13 (C. C. Gowdey). British East Arrica: Government Farm, Kabete ;. on ornamental shrub, 1.1914 (7. J. Anderson). Lecanium (Eulecanium) somereni, Newst. Ucanpa: Nagunga; on Hrythrina excelsa, 20.1x.15 (C. C. Gowdey). The specimens are of an unusually dark colour and heavily parasitised. Lecanium (Eucalymnatus) tessellatum,- Sign. BritisH Guiana: Botanic Gardens, Georgetown; on Malacca apple, 28.vu.15 (R. Ward, per G. E. Bodkin); associated with Lecaniuwm wardi, sp. n. These examples show a considerable variation in the arrangement of the plates which form the tessellated pattern of the dorsum. Some agree with Signoret’s description (the first I have seen); others (two) with the var. perforatum, Newst., though these are not quite typical. Lecanium (Coccus) viride, Green. British Guiana: Onderneeming, Essequibo, 60; on Liberian coffee, 29.vu.13 (G. E. Bodkin). SourH Arrica: 1914 (D. d@Emmerez); all immature 99, differmg from typical examples of L. viride only in that some have eight instead of seven segments to the antennae. Hemilecanium imbricans, Green. S. Inpr1a: Southern Mysore; on Cedrela tuna, heavy infestation, x1.1903 (H. Ballard). Aspidiotus (Chrysomphalus) auranti!, Mask. British Hast Arrica: Kabete; on orange tree, 7.1.14 (If. J. Anderson). SOUTHERN RuopeEsiA: Mt. Chirinda, Melsetter (C. #. M. Synnerton)—very heavily OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—V. 131 infested by a fungus, apparently a species of Nectria ; Chalcidid parasites also present. SoutH Arrica: Pretoria; on Acacia, 1914 (D. d Emmerez)—a few examples living in association with Chionaspis capensis, Newst. JAMAICA: on citrus imported from India in 1913, 15.11.16 (A. H. Ritchie). Fist: Taveuria; on bananas, 1913 (Chas. H. Knowles). Aspidiotus camelliae, Sign. British East Arrica: Nairobi, 5,500 ft., on apple, 3.1.13, and on rose-tree, 6.vi.12 (Geo. W. Evans); Kabete, on trunks of black wattle, 7.1.14, and on apple and fig, 11.1914 (7. J. Anderson). Aspidiotus cyanophylli, Sign. Ucanpa: Entebbe; on peach, 4.xu.12 (C. C. Gowdey). Fist: Taveuria; on bananas, 1913 (Chas. H. Knowles). Aspidiotus cydoniae, Comst. British East Arrica: Kabete, on mango, 7.1.14 (7. J. Anderson). UGANDA: Entebbe, on Bauhima, 4.11.14, on guava, 10.11.14, and on rose shrubs, 24.v.12; Kampala, on “Cape lilac,” 4.vii.15 (C. C. Gowdey). Gotp Coast: Accra, on oleander, 12.1v.16 (Dr. J. W. Scott Macfie). Jamaica; on Nectandra coriacea, 1916 (A. H. Ritchie). S. INpta: Coimbatore, on fig fruit, v.1912, and on vine, xi1.1909; Bangalore, on pear, 11.1910 (7. V. Ramakrishna, per E. Ballard). Aspidiotus destructor, Mask. | S. Inp1a : Cochin State, Kimbalengua ; on cocoanut leaves, 8. vii.08 (£. Ballard)— heavy infestation, but about 70 per cent. destroyed by a small Coleopterous larva. ZANZIBAR: Marahubi; on mango, 24.iv.13, in association with Lecanium adersi, Newst.; on stems of castor oil plant, 16.vi.13 (Dr. W. M. Aders)—the examples on the last-named plant have very remarkable puparia, being highly convex and exceedingly thick and robust. SoutH AFRICA: on an unnamed plant, 1914 (D. d@ Emmerez). British East ArricA: Mombasa, on cocoanut palm, 7.1.14 (T. J. Anderson). UGANDA: Kisube, on banana, 10.vili.13: Entebbe, on mango, 13.1.13, on guava, 10.11.14, on screw-pine, 16.i11.14, and on Ceara rubber, 10.vui.13; Kampala, on banana, 8.x.13, and on Heavea brasiliensis, 26.x .13 (C. C. Gowdey). Gotp Coast: Accra, on mango, 6.xi.16 (Dr. J. W. Scott Macfie) ; Aburi, on Pandanus (W. H. Patterson). Aspidiotus (Chrysomphalus) dictyospermi, Morgan. S. Arnica: Fort Beaufort, 1914 (D. d@ Emmerez). Aspidiotus (Chrysomphalus) ficus (Riley), Comst. 8. Iyp1a: Coimbatore, on mango, 11.1914; Pemkonda, on Ficus sp., 11.1907 (7. V. R., per E. Ballard). Aspidiotus (pseudaonidia) fossor, Newst. British GuIANA: Georgetown, on grape-vine, 10. viii.13 (@. E. Bodkin). 152 ROBERT NEWSTEAD. Aspidiotus orientalis, Newst. British East Arrica: no data (7. J. Anderson). 8. Inp1a: Coimbatore, on tamarind fruit, 1.1911 (7. V. R&., per E. Ballard), very heavy infestation ; Guntur, on egg-fruit, 20.11.10 (7. V. R., per E. Ballard), heavy infestation. A large proportion of the African adult females have the margin of the cephalo- thoracic area chitinised, a character not noted in the type lot, but present also in examples from Southern India. The incorporation of the epidermal layer of the fruit in the Coimbatore puparia has given them a distinctly russet-red colour. The three large bilateral squamae in nearly all of the examples examined are almost entirely free from the minute lateral serrations which may be clearly seen in typical examples ; otherwise the structural characters are specifically identical. The puparia of the Guntur females are much more robust than those of the co-types, and bear a striking resemblance to those of A. cydoniae, Comst. Aspidiotus (Chrysomphalus) rossi, Mask. S. Arrica: Modderfontein; on Hucalyptus sp., 1913 (D. @ Emmerez). Aspidiotus (Pseudaonidia) tesseratus, d’ Emm. Jamaica: on Matyba apetala, Nectandra coriacea, Gr., Trophis racemosa, Urt., 1916 (A. H. Ritchie); all heavy infestations. Aspidiotus (Pseudaonidia) trilobitiformis, Green. S. Inpra: Coimbatore, vin.1912 (7. B. Fletcher). Ucanpa: Kampala, on oleander, 20.vu.15, on mango, 20.vu.15 (C. C. Gowdey). Aspidiotus (Chrysomphalus) triglandulosus, Green. S. Inpra: Bangalore: “attached to scabs on Jack leaf,” vi.1912 (7. B. Fletcher). Aspidiotus (Selanaspidus) silvaticus, Lind. Ucanpa: Entebbe, on Citrus aurantium, 7.1x.13 (C. C. Gowdey). Female puparium: More or less circular or somewhat deltoid when arrested by the prominent veins of the leaf; thin and flat. Exuviae central in the circular forms, and submarginal in the deltoid ones. Those on the upper surface of the leaves russet- buff, with broad pale margins and slightly darker exuviae; on the underside more opaque and brownish buff, with obscure buff-yellow exuviae. Greatest diameter 226mm. Female adult: Very broadly pyriform; thoracic constriction slightly: less than one-third the entire length of the body ;_ thoracic tubercle apparently absent. Pygidium roughly triangular in outline; its margin with two pairs of well developed lobes, the median pair deeply emarginate on both sides; the curiously forked squamae beyond the usual prominent tooth-like process seem to be characteristic of the species. 7 Lindinger’s* examples of the puparia were evidently very much smaller, but the structural details given by him (I.c.) of the adult female agree in all essentials. (Juite 50 per cent. of the females were attacked by a parasitic fungus. * Jahr. der Hamb. Wissen. Anst., xxvi, p. 10, taf. ii (1909). OBSERVATIONS ON SCALE-INSECTS (COCCIDAE)—V. 13. Vey Aspidiotus (Selanaspidus) articulatus, More. Jamaica: Kingston; “ on citrus imported from India,” 15.11.16 (A. H. Ritchie). Chionaspis (Hemichionaspis) minor, Mask. SouTHERN NicreriA: Ibadan; food-plant not stated (? cotton) (Dr. W. A. Lamborn). Gotp Coast: Aburi, on Jatropa curcas (W. H. Patterson); the female puparia are larger and much more dilated than typical examples ; otherwise they do not materially differ. SourH Arrica: Drakensburg, Cape Province; on willow trees (D. d’ Emmerez). Cooley* has described a variety of this species, under the name strachani, from material collected on an unknown plant from Abeokuta, West Africa, and states that it differs from H. minor “‘ in having the exuviae of the female scale rather darker and contrasting more strongly with the secreted portion, and in having the median lobes broader in proportion to their length and more finely and less distinctly crenate.”’ Dr. Lamborn’s examples of the females do not exhibit any morphological characters by means of which one can separate them from typical H. minor; but the puparia of the females show a somewhat marked departure from Maskell’sf description in having the secretionary portion much more widely pyriform, dusky white, and relatively thin and smooth. Chionaspis dentilobis, Newst. Uceanpa: Entebbe, on an unnamed plant, 6.v.10 (C. C. Gowdey). A large percentage of the females were parasitised, evidently by Chalcidid Hymenoptera. Chionaspis (Phenacaspis) lutea, Newst. Gotp Coast: Aburi, on Funtunia (W. H. Patterson). - The male puparia associated with the females in this colony are very strongly tricarinate and not at all like those in the type lot described from East Africa. Possibly the West African examples belong to a species of Diaspis. The 29 are specifically identical with the co-types of C. lutea. Fiorinia proboscidaria, Green. JAMAICA: Kingston, “ on citrus imported from India,” 15.11.16 (A. H. Ritchie). This species has not hitherto been reported from the West Indies. Parlatoria pergandei, Comst. JAMAICA: Kingston, “on citrus imported from India, 1913,” 15.11.16 (4. H. Ritchie). Parlatoria ziziphus, Lucas. Jamaica: Kingston, “on citrus imported from India, 1913,” 15.11.16 (4. H. Ritchie). * Special Bull. Hatch Exp. Stn., Mass. Agric. Coll., 10th Aug. 1899, p. 54. 7 N.Z. Trans., xvii, p. 33, pl. vi, fig. 4 (1884). + Mit. Zool. Mus. Berlin., v, 2, p. 169 (1911). 134 ROBERT NEWSTEAD. Ischnaspis filiformis, Doug. JAMAICA: on Ficus sp., 1916 (A. H. Ritchie). UcGanpa: Mabira Forest, Chagwe ; on coffee, 28.x11.12 (C. C. Gowdey). Mytilaspis (Lepidosaphes) citricola, Pack. Goup Coast: Accra, ona sickly lime-tree, 12.11.16 (Dr. J. W. Scott Macfie). Lepidosaphes gloverii, Pack. Ucanpa: Entebbe; on croton, 10.vu.13 (C. C. Gowdey). THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. By A. Ineram and J. W. 8. Macrin, West African Medical Service. (PLates I—IV.) Howard, Dyar and Knab in their monograph “ The Mosquitoes of North and Central America and the West Indies,” in discussing the rearing of mosquitos, emphasise the importance of larvae in specific determination. They write (Vol. I, p. 181) “* Many species of Culex are of uncertain determination without the associated larvae, while some Aédes have identical adults, yet dissimilar larvae. The characters of the larvae reside in the modifications of the chitinous appendages of the skin and the arrangement ofthe hairs. As these are fully retained by the cast skins, it is possible to preserve both the larva and the adult of the same identical specimen, thus assuring absolutely correct associations.” This plan has been carried out so far as possible in the rearing of the mosquitos, the larvae of which—obtained during 1916 in Accra and its vicinity—we have attempted to describe below. We take this opportunity of pointing out that so far as our experience of the mosquitos of Accra goes, Howard, Dyar and Knab are in error in controverting the statement of the late Sir Rubert Boyce that mosquitos living in crab-holes “ are the chief nuisance in those houses which are situated near the sea.” Howard, Dyar and Knab write : “ Our observations prove that the crab-hole mosquitoes do not bite nor annoy man and no consideration need therefore be given to their destruction.” In Accra Culex thalassius, Ochlerotatus writans and Anopheles costalis are commonly found flourishing in crab-holes, all of these mosquitos are vicious biters and invade houses ; their destruction therefore is one of the problems which must be faced by Sanitary Authorities on the West Coast of Africa, especially as A. costalis is a proved carrier of malaria. We have here to express our indebtedness to Dr. A. C. Parsons, Junior Sanitary Officer at Accra, and to Dr. J. B. Alexander, Medical Officer of Health at Accra, for their untiring zeal in procuring for us specimens of larvae. Our thanks are again due to Dr. G. A. K. Marshall, of the Imperial Bureau of Entomology, and to Mr. F. W. Kdwards, of the British Museum, for much help and advice and for identifying species. Anopheles marshalli, Theo. The head is small as compared with the thorax. The brushes are prominent. The antenna possesses no hair-tuft and its surface is covered with minute spines, which, however, are almost absent externally. At the end of the antenna there are stout spines and a short branched hair; this hair shows at least three branches. The anterior frontal hairs (clypeal hairs) are all simple and appear to be placed in a triangle, one being situated anteriorly and two posteriorly ; of the two posteriorly placed hairs the external is only slightly in advance of the internal. The internal posterior hair is nearly twice as long as the others. The mental plate is small and carries a median tooth with four teeth on either side. 136 A. INGRAM AND J. W. S. MACFIE. The thoracic plumes are fairly developed, the median hairs overlap the occiput. There are no rudimentary palmate hairs on the thorax or the first two abdominal segments. There are long lateral feathered hairs on the first three abdominal segments and all the segments bear short branched hairs laterally. Palmate hairs are present on the fourth to the seventh abdominal segments. ‘The leaflets of which these hairs are composed show no shoulder and scarcely any filament. Upon an average there are nine hairs on each side of the main stem (in A. costalis the average number is seven) when the structure is looked at in profile. The comb differs from the comb of A. costalis in that the long teeth are, relatively to the short teeth, more elongate and more numerous. Wesché, describing the comb of A. costalis (Bull. Ent. Res., i, p. 21) in the various stages of development of the larva, says of the final stage, ‘‘ the character of the comb remains the same, four short spines followed by a long one.” In the comb of A. marshall: there appear to be three short spines followed by a long one. This larva falls into the same group as A. pretoriensis and A. rufipes in Edwards’ key (Bull. Ent. Res., iii, p. 374), but is distinguished from them by having no palmate hairs on the second abdominal segment. Breeding place.—This larva was taken in a Pistia-covered pool at Christiansborg, near Accra (see Pl. II, fig. 1). From the same pool larvae of A. costalis, Aédeomyia africana, Culex quasigelidus, Mimomyia splendens and Mansonioides africanits were obtained. Stegomyia simpsoni, Theo. The head is small and moderately chitinised. The antenna is short and heme in place of the tuft a single hair at about its middle or a little beyond. There are several small tufts of hairs and one tuft of three long hairs arising between the base of the antenna and the eye and projecting laterally. The mental plate has a median pointed tooth of moderate size and ten or eleven smaller teeth on each side, the two outermost teeth being much smaller than the others. The brushes are small. The thoracic plumes are not very strongly developed and are composed of simple hairs. The hook-like spines at the bases of the ventral plumes are about as strongly developed as those of S. fasciata. The abdomen is not conspicuously hairy and is without true stellate hairs. All the hairs are simple. The hairs forming the siphonal, subsiphonal, and anal plumes appear also to be simple. The comb consists of a row of seven or eight spines barbed at their bases, which resemble those of S. fasciata but are rather less strongly barbed. The siphon is less than twice as long as its basal diameter. The pecten is composed of about eight slightly barbed spines. The tuft is situated just beyond the middle of the siphon and is composed of three simple hairs; it lies at about the same level as the last pecten spine. The beard is poorly developed. The dorsal hairs on the anal segment are three above and one below on each side. Theanal papillae are long and blunt at the ends. The above description was made from a single specimen, but unless it was an abnormal one, this larva can be distinguished from the other known larvae of this genus with barbed comb spines and without conspicuously hairy abdomens by its pecten of only eight spines. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 137 Breeding place.—The larvae of this species were found in rot-holes in trees and were associated with larvae of Culicuomyra nebulosa, Stegomyra fasciata, S. metallica, S. unilineata, and S. luteocephala. Stegomyia unilineata, Theo. The head is small, though well chitinised. The antenna is short, and there is no hair-tuft, a single or bifid hair being, however, visible at about half the length of the shaft. In one of the specimens examined a distinct constriction occurred in the antennae at a fourth of the length from their bases. Several tufts of stout hairs may be seen on the head. The mental plate carries a substantial median tooth with ten more slender teeth on each side, the lateral teeth becoming broader and more widely separated towards the base of the plate. The brushes are small. The thoracic plumes are short and are formed of hairs which are subplumose at their bases. The hook-like spines visible at the bases of the ventral plumes on the thorax of some other members of this genus are scarcely perceptible. Numerous hair-tufts are present on all the abdominal segments, these are not true stellate hairs, but resemble the hair-tufts on the abdominal segments of S. luteocephala. A few of the constituent hairs are pubescent, the majority are simple. The hairs forming the subsiphonal, siphonal and anal plumes show slight pubescence. The comb consists of about eight spines with fringed bases set in a single row. The siphon is about twice as long as the diameter of its base, and there are 8-12 barbed spines in its pecten; distal to the pecten is a tuft of four or five simple hairs, situated a little beyond the middle of the siphon. The anal segment carries a scanty beard; the hairs on the end of the dorsum are three above and one below on each side, and there is a tuft of five stout hairs—which are simple— on the lateral aspect of the segment. The papillae are long and blunt-ended. This larva may be distinguished from other larvae of the same genus which have numerous hair-tufts on the abdominal segments by the absence of hook-like spines on the thorax. Breeding place.—The larvae were found in association with the larvae of S. luteocephala, S. metallica and S. simpsoni in a rot-hole holding a small quantity of water in the trunk of a Flamboyant tree (Poinciana regia) in the laboratory compound at Accra (Plate IV, fig. 2). Mansonioides africanus, Theo. (Plate I, figs. 1, 3; text-fig. 1). The larva of Mansoniordes africanus has already been described by Ingram (Bull. Ent. Res. iii, p. 76), and has also been figured by Edwards (Bull. Ent. Res. 10, p. 377). The siphon tube is complex and would appear to resemble that of Man- soma titillans, a South American mosquito, the larva of which also attaches itself to the roots of Pistva. A general description of this tube has already been published, but a more detailed account of the structure, especially that of the apical third, may be of interest as indicating the high degree-of specialisation of the breathing apparatus of the larva of this species. The apical third of the siphon tube is more highly chitinised than the basal two- thirds and appears to be movable on the latter portion. It is an incomplete tube composed of two lateral chitinous lamellae with a ventral membrane connecting (C394) se 138 A. INGRAM AND J. W. S. MACFIE. them, but dorsally it 1s open to accommodate the saw-like element to be referred to immediately. The lateral chitinous lamellae extend to the tip of the siphon, where they come into relationship with the ends of the internal structures, and they bear at their extremities two pairs of slightly chitinised hooks. At the base of each lamella there is a single bristle, which lies therefore at about the same level as the Fig. 1. Siphon of Mansonioides africanus; (1) outer case, lateral view; (2) apical third of the outer case, ventral view; (3) inner elements: Tr, trachea, 8S, saw, L, the lateral pieces, V, ventral pieces, Ax, axial rod; (4) ventral pieces flattened out and seen from the ventral aspect: L, the lateral pieces in profile; (5) saw flattened out, dorsal view. larger dorsal bristles, but more ventrally and not as figured by Edwards. These bristles are simple and are not furnished with a fringe or membrane. The ventral membrane is feebly chitinised and is irregularly ribbed or folded transversely, so that in profile it looks like the edge of a file. It is attached laterally and basally to the chitinous part of the siphon by a delicate membrane. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 139 Within the siphon tube a number of complicated structures can be seen. Running from end to end of the tube and extending for some distance into the abdomen there is an axial rod of chitin, a narrow flat structure with its edges directed towards the dorsal and ventral aspects of the siphon, and with a hinge or joint at its lower end. The two tracheal trunks pass upwards inside the siphon to the right and left of this axial rod; in the apical third they lose their spiral structure and unite with the rod so as to form the “ common hollow space ”’ described by Raschke. This arrangement is usual in mosquito larvae, but in the case of Mansonioides africanus the margins of the space do not form simple valves, but are highly specialised so as to enable the larva to penetrate the roots of Pistia stratiotes and to attach itself firmly to the plant. Laterally the outer walls of the tracheal tubes are continued as highly chitinised plates, which rapidly taper to points. On the ventral aspect the axial rod is continued as a stout, highly chitinised structure, on which a trough- shaped body composed of a mesial and two lateral parts is jointed, each of the lateral pieces bearing at its distal end a triple hook of great strength. A delicate flexible membrane connects the bases of these hooks with the edge of the siphon. On the ventral margin of the “ common hollow space” a highly chitinised trough- shaped structure with a saw-like dorsal keel is hinged. Howard, Dyar and Knab state that in Mansonia the “outer portion of the tube is . . . furnished with serrations,’ but in Mansonoides africanus the saw appears to be attached to the margin of the “common hollow space” and to be separate from the tube, and for these reasons we conclude that it probably represents a modified valve. These structures may perhaps be correlated with the primitive elements in the closing mechanism above the spiracles of Anopheline larvae, corresponding respectively to “the lateral flaps, the posterior plate, and the anterior fan-shaped plate. The saw-like structure has about fourteen teeth and appears to be attached a little above its lower end to the dorsal margin of the “ common hollow space.” Its lower end divides, a strong median process passing down into the siphon tube and two knob-like bodies projecting dorsally so as to he between but in close apposition to the bases of the two stout curved bristles situated at the junction of the apical third of the siphon with the basal portion. These bristles have a delicate fringe or membrane on their lower surfaces, which suggests that they may have a sensory function and may be of service to the larva when feeling its way between the rootlets in search of a suitable point at which to attach itself. The larvae of Mansonioides africanus after the first instar attach themselves to the tap-roots of Pistia plants ; they swim about in the water until they encounter the rootlets and then, tail fore- most, follow these inwards to the root itself. In this manoeuvre vision can be of little assistance to them ; but if the large bristles are sensory, as we suggest they may be, they might help the larvae to reach the root, and, once arrived, might transmit a stimulus enjoining the larvae to attach themselves. The actual process by which the larvae penetrate the tissues of the root, appreciate the fact when they have reached the air-containing tubes, and fix themselves firmly, must be a matter of surmise, but as extremely powerful muscles are attached to the axial rod, it may be conjectured that this structure plays an important part both by approximating the terminal elements, so as to form a sharply pointed cone for thrusting into the root, and by putting into action the saw-like structure attached to it, which, (C394) B2 140 A. INGRAM AND J. W. S. MACFIE. as we have stated, is poised on the dorsal margin of the “ common hollow space ” as its fulcrum. The six sets of terminal hooks no doubt suffice to anchor the larvae once they have penetrated the plant tissues. In specimens killed and fixed in situ and subsequently cleared, the siphon is seen to be embedded in the root up to a point about half-way between the tip of the tube and the bases of the dorsal bristles, the hooklets at the end are spread out in a cluster and appear to be firmly fixed in the. tissues, and the dorsal and ventral bristles of the siphon, are widely extended with their tips in contact with the surface of the root. The siphon of the larva in the first instar bears but a slight resemblance to that of the fully-developed larva, it is a bottle-shaped structure, which is made up of a conical basal portion (8°5 units) and a more highly chitinised and slender apical portion (7°5 units). The basal ring is hardly visible, the movable junction between the apical third and the basal portion is indistinguishable, but the chitinised axial rod is well developed and has powerful muscles attached to it in the lower half of the tube. About half-way up the lower part of the siphon and slightly on the ventral aspect, that is, in the position occupied by the tuft in the fully-developed larva, there is a single, very long, simple hair. At the shoulder of the bottle, that is, at the point where the narrow part of the siphon commences, the tracheal trunks are constricted, and from this level run up the distal end of the siphon as two chitinised tubés separated. by the axial rod. At the tip of the siphon on the ventral aspect there is a pair of powerful highly chitinised double hooks, corresponding with the similar triple hooks of the fully developed larva ; dorsally there are two pairs of less highly chitinised hooks, which are toothed; and a pair of relatively stout bristles mounted on large bases can be distinguished, which appear to represent the dorsal bristles of the later stages. The saw-like structure on the dorsal aspect is not developed at this stage. The relatively poor development of the terminal structures is no doubt associated with the habit of the larva in this instar of attaching itself to the rootlets instead of to the tougher tap-root. The larva at this stage, the first instar, differs in many other respects from the fourth phase larva. ‘The head is large, considerably larger than the thorax; the antennae are relatively huge, possessing the two prominent bristles which are conspicuous in the fully-grown larva, and have in place of the tuft a single stout branched hair. The mental plate is small and has three slender lateral teeth on each side of the prominent mesial tooth. The eyes are prominent and rounded; the “ main” eye is not yet developed. The lateral abdominal hairs are two on the first, and one on each succeeding segment. The comb consists of three to five fringed spines, quite unlike the curious elements of the comb of the fully-developed larva. The tuft on the siphon is represented by a simple stout hair, and all the plumes of the eighth segment are in this stage simple hairs. The dorsal hairs on the anal segment are conspicuous, arranged in pairs, the ventral pair being shorter than the dorsal; there isno beard. The anal papillae are equal, and have rather less sharply pointed ends than those of the fully grown larva. After the first moult the siphon assumes the form described as being characteristic of the species. The dorsal saw-like structure is well developed in the second instar, but has rather fewer teeth than in the fully developed larva. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 141 The egg.—The eggs are laid on the under surfaces of the leaves of Pistia stratiotes in clusters or cushions about 2-3 mm. in diameter, containing about 150 eggs apiece. Each egg is rounded at its attached end, but prolonged into a snout at its free end, is dark brown in colour with the surface covered by a cellular reticulation, and measures about 0°96 mm. in length and 0°17 mm. in breadth at its widest part. At the blunt end of the egg there is a layer of clear, colourless material to attach it to the surface of the leaf. The pointed end projects freely and bears at its extremity the micropilar apparatus consisting of a cap-like mass of clear, colourless material similar to that at the rounded end. This cap is readily detached, revealing the presence of four blunt processes at the end of the egg. The egg-masses lie on the under surfaces of the outer leaves. As the plant unfolds, the leaves become nearly horizontal, and in this position the cushions of eggs come into contact with the water, the pointed ends of the eggs being directed downwards. When the larva hatches the egg breaks horizontally at its widest point, detaching a cone-shaped piece about 037 mm. long consisting of the end of the egg and the micro- pilar apparatus. This cone falls to the bottom of the water. The larva on escaping also descends to the bottom and attaches itself to the root of a Pistia plant, selecting a delicate rootlet and not the main tap-root as the older larvae do. Breeding place.—The larvae of Mansonioides africanus are found in pools and swamps in which Pistva stratiotes is growing (see Plate I, fig. 1; UI, fig. 1), but up to the present we have not found them associated with any other water-plants. At Accra we obtained specimens regularly from a particular pool from April to November. In April and at the beginning of May the larvae were difficult to find, in June they were very numerous, in July-and August they diminished in numbers, and in September and October they were again very few. They were most plentiful therefore during the rainy season, which at Accra reaches its height in June. Ochierotatus albocephalus, Theo. (fig. 2). The larva has a small and dark head, slightly more than half as wide as the thorax ; the antennae are slender, curved and covered with spicules, the tuft is placed about the middle and consists of a few subplumose hairs. The midfrontal hairs are multiple (5-8) and are plumose ; the brushes are prominent, being composed of coarse hairs. The eyes are large. The mental plate has a large median tooth and nine smaller teeth on each side of it. The thoracic plumes are poorly developed, the constituent hairs being plumose. The lateral abdominal hairs are multiple on the first segment, triple on the second, and single on the remaining segments. The comb consists of 12-15 barbed spines placed in an irregular row. ‘The subsiphonal plume is composed of subplumose hairs, the siphonal plume and the anal plume consisting of pubescent hairs, and there is a single independent hair adjacent to the anal plume. The siphon is about three times the diameter of its base; the pecten comprises 9-11 spines, which become slightly larger and are more widely separated distally ; a tuft of simple hairs is situated just beyond the middle of the siphon (35-60). The anal segment is broader than it is long (22 x 19); the anal papillae are almost equal in length, stout and with sharply pointed ends, being longer than the anal segment (28:19). The beard is well 142 A. INGRAM AND J. W. S. MACFIE. developed, extending all the length of the ventral surface of the segment. The hairs on the dorsal end of the anal segment are quadruple above and single below on each side. This larva may be distinguished from the other known African Ochlerotatus larvae (Edwards’ key, Bull. Ent. Res., ii, p. 376, and Bull. Ent. Res., vu, pp. 5-6) by its possessing multiple mid-frontal hairs with a pecten of only 9-11 spines. Breeding place.—The larvae were found in a variety of situations, namely, in small cavities washed out by the sides of cement drains running across an open and wind- swept golf-course, in an empty grave in the Accra cemetery, in broken pipes, earth drains, pools and crab-holes. Associated with them, A. costalis and C. thalassius were frequently found, and once C. fatigans. Fig. 2. Larva of Ochlerotatus albocephalus, Theo.; head and end of abdomen ; a, mental plate. The larval stage appeared to be passed through very rapidly, as some of the samples in which pupae predominated were taken from pools that could not have been in existence more than three or four days. The attitude of the larvae when at the surface of the water was characteristic, the body and the siphon tube being almost in line or forming an exceptionally obtuse angle. - Ochlerotatus minutus, Theo. (fig. 3). The head is small, narrower than the thorax (52 : 80), and not so highly chitinised as the head of O. albocephalus. The antenna is curved, slender and covered with spicules ; it carries a tuft of subplumose hairs at its middle. The mid-frontal hairs are peculiar in that they are delicately branched; they are multiple, numbering 6-8. The ante-antennal tuft is composed of 9-10 plumose hairs. The mental plate has a small median tooth and 9 lateral teeth on each side. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 143 The thoracic plumes are formed of plumose hairs and are poorly developed. The Jateral abdominal hairs are multiple on the first segment, triple on the second segment, and single on the remaining segments. The comb consists of 12-15 barbed spines set in a triangular patch. All the hairs forming the siphonal, subsiphonal and anal plumes appear to be subplumose. The siphon is slightly swollen about its middle and is rather more than three times as long as the diameter of its base, its pecten shows nine spines, which increase in length and are more widely separated distally ; there is a slender tuft of simple hairs just beyond the middle (34/58). The anal segment is longer than it is broad (23 : 19); the beard does not appear to Fig. 3. Larva of Ochlerotatus minu'us, Toeo.; head and end of abdomen; a, mental plat>. extend the full length of the ventral surface; the hairs on the dorsal end of the segment are a tuft of 5-6 hairs above and of one stout hair below on each side. The anal papillae are short and bluntly pointed. This larva may be distinguished from other known African Ochlerotatus larvae (Edwards’ key, Bull. Ent. Res., i, p. 376, and Bull. Ent. Res., vu, pp. 5-7) by the fact that its multiple frontal hairs are delicately branched. Breeding place.—The larvae were found in April at the margin of a lagoon in crab-holes that had been enlarged by children digging for crabs (Plate II, fig. 2). The water in which they were living was dirty, thick with deposit, and quite salt. In the same holes larvae and imagines of Culex thalassius were also found. The occupants of the house neighbouring the lagoon were much troubled by mosquitos, and a small collection made on the same day as that on which the larvae were found contained specimens of both O. minutus and C. thalassius. 144 — A. INGRAM AND J. W. S. MACFIE. Ochlerotatus apicoannulatus, Edw. (fig. 4.) The whole larva is of a dark brown colour and very opaque, so that it is difficult nan uncleared specimen to distinguish details. The head is a little more than half the width of the thorax, being well chitinised. The antenna is of a light brown colour, curved, and covered with spicules. The whole antenna measures about 25 units and carries a tuft of five or six subplumose hairs at 11 units from its base. The mid-frontal hairs are multiple (8-10) and are plumose ; the ante-antennal tuft is plumose. There are on the lateral aspect of the head three tufts of hairs, a small one immediately in front of the eye and a large one shghtly posterior to the eye; the hairs constituting the tufts are simple; a short distance anterior to the small tuft of simple hairs is a well developed tuft of plumose Fig. 4. Larva of Ochlerotatus apicoannulalus, Edw.; head and abdomen; a, mental plate. hairs situated more towards the ventral aspect of the head. The mental plate shows a stout median tooth with nine teeth on each side; these lateral teeth, which are pointed, increase in size towards the base of the plate. | The thoracic plumes are small. The lateral abdominal hairs are multiple on the first segment and paired on the other segments. The subsiphonal plume is formed of plumose hairs; the anal one apparently consists of simple hairs, and the siphonal seems to be represented by a single hair, but this is difficult to determine. The comb consists of 11-12 spines set in an irregular row. The siphon is about three times as long as the diameter of its base (55 units x 19) and carries a pecten of 18-20 barbed teeth, the last tooth being situated at about 23 units from the base of the siphon, THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 145 ‘and a strongly developed tuft of six pubescent hairs is inserted at 28 units from the base. The valves are prominent. The anal segment is slightly longer than it is wide; it has a voluminous beard extending the whole length of the ventral surface, the hairs on the end of the dorsal surface being three above and one below, and there is a slender tuft of simple hairs on the lateral aspect of the segment immediately beneath the insertion of the dorsal hairs. Of two larvae examined, one showed no anal papillae and the other only one; this papilla was of extreme length, measuring about 48 units and having a pointed end. The larva may be distinguished from other Ochlerotatus larvae (Kdwards’ key, Bull. Ent. Res., iii, p. 376) having multiple mid-frontal hairs and a pecten with about 18 teeth, by the number of teeth in its comb. Breeding place.—The larvae were obtained from a hollow between the branches of a Flamboyant tree at Christiansborg, near Accra. The water collected in the hollow was dark brown in colour and contained rotting leaves and twigs. Larvae of S. fasciata, S. metallica and S. luteocephala were flourishing in the’ same situation. Cyathomyia fusca, Theo. (fig. 5). The head is about as wide as the thorax, and the frontal hairs are plumose. The antenna is rather slender, being covered with spicules, and carries its hair-tuft of simple hairs at five-sixths of its length. The eye is large. The mental plate has one very large mesial tooth, upon each side of which are four square-ended teeth and five sharply pointed teeth, these latter being nearest the base and the two most basal being larger than all the others except the mesial tooth. The thoracic hairs on the dorsum are long and stout, but are scantily plumose. The lateral abdominal hairs are quadruple on the first segment and paired on the succeeding segments. The comb consists of about 35 long scales, which are fringed and are set in a triangular patch. The siphonal plume is formed of a few scantily plumose hairs with an independent hair adjacent; the subsiphonal plume consists of 6-8 plumose hairs, and one or two simple hairs constitute the anal plume. The siphon is about 12 times as long as the diameter of its base;* the pecten is composed of sharply pointed spines, 12-15 in number, and extends to about a quarter of the length of the siphon; beyond the pecten are 7-8 tufts of simple hairs, which are scanty, short, and irregularly placed. The anal segment measures about 20 units in length by 12 in width ; its dorsal hairs are paired, two above and two below ; the beard is rather poorly developed, and there is a small tuft of simple hairs on the lateral aspect just below the dorsal hairs. The anal papillae are unequal, the dorsal pair (41 units) are nearly twice as long as the ventral pair (24 units). This larva may be distinguished from other Culex larvae possessing a siphon about 13 x 1 (Edwards’ key, Bull. Ent. Res., 11, p. 381, and Bull. Ent. Res., vii, pp. 11-12) by its numerous comb scales (30-40) and by the absence of distally scattered pecten Spines. *In specimens dehydrated and mounted in balsam the length, however, is not more than ten times the diameter of the base. 146 A. INGRAM AND J. W. &. MACFIE. Breeding place.—The larvae were found flourishing in the dark brown water that. had collected in a hole partly filled with decaying vegetable matter in a Flamboyant tree (see Plate IV, fig. 1). With them were associated larvae of A. costalis, A. marshalli, S. fasciata and S. luteocephala. | \ ar i au | | | A ips | ; yy}! "a NN | WAVG yyy } fd LQ WN pia X Us Fig. 5. Larva of Cyathomyia fusca, Theo.; head and end of abdomen ; a, mental plate. Eumelanomyia inconspicuosa, Theo. In our former description of the larva of this species (Bull. Ent. Res., vu, p. 12) certain gaps had to be left owing to lack of sufficient material for examination. As we have been fortunate enough to procure further specimens, we are now able to complete the description. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 147 The tuft on the antenna is usually situated at a point about four-fifths of the length from the base, and not about the middle as formerly stated. The mental plate, which resembles that of Cyathomyia fusca, has a very large median tooth, on each side of which are 8-9 teeth. These lateral teeth are larger towards the base of the plate, the four teeth nearest the central tooth are square-ended, the fifth tooth is in a transitional stage, and the four basal teeth are sharply pointed. The lateral abdominal hairs are multiple on the first segment, triple on the second segment and double on the remaining segments. The dorsal hairs on the anal segment: are one above and one below on each side. The anal papillae are unequal, the dorsal pair being nearly twice as long as the ventral, and as a rule the longest pair do not greatly exceed the length of the anal segment. Culex thalassius, Theo. (fig. 6). The head is large, as wide as the thorax. The antenna is slightly curved and has spicules on its basal half and on the outer aspect of its apical part ; it carries a tuft of plumose hairs just beyond the middle (17/28). The mid-frontal hairs and the tuft at the base of the antenna (ante-antennal plume) are strongly plumose. The mental plate has a substantial median tooth and eight sharply pointed teeth on either side. The eye is large, and the “ larval” eye is in most specimens distinctly separate from the “imaginal” eye. The thoracic plumes are plumose and their insertion is marked by a chitinous plate. The longer lateral abdominal hairs are subplumose ;_ they are quadruple on the first and second segments, and paired on the remaining segments. There are 60-70 scales in the comb; these are small, with a fringe of hairs, and are collected into a somewhat irregularly triangular patch. The hairs forming the siphonal and subsiphonal plumes are plumose, the anal plume appears to be formed of three or four pubescent hairs, which occasionally branch at their apices; in addition there is an independent hair between the anal and subsiphonal plumes. The siphon is nearly five times as long as the diameter of its base (83 units to 17) ; the pecten contains 12-15 barbed spines, beyond it being 4—5 pairs of strongly developed hair-tufts, while a smaller tuft is situated a short distance below the end of the siphon, and another smaller tuft is also to be seen laterally at a slightly lower level. The hairs in the larger tufts are pubescent apically. The anal segment is longer than it is broad and has a well developed beard ;_ the hairs on the dorsal end are usually four above and one below on either side. A few small spines may be found on the end of the segment immediately beneath the dorsal hairs. The anal papillae are very short and oval in shape; they measure in length 10-12 units, as compared with the anal segment itself, which measures 30 units. Eggs.—The eggs of C. thalassius are laid in a raft of the usual shape and size; they are a little more than four times as long as their greatest breadth. The surface of the egg shows cellular or reticular sculpturing. Breeding place.—The larvae appear to be unusually adaptable. They have been found flourishing in a variety of situations, such as a brackish lagoon, crab-holes, foul-smelling water holes, earth drains, pools of various sorts, an iron pot, and a spring. The first sample collected was found in a lagoon (see Plate III, fig. 2), the water of which contained 680 parts of chlorine per 100,000. The larvae were distributed along the edge in the shallow water, but did not frequent the deeper parts. They 148 A. INGRAM AND J. W. 8S. MACFIE. occurred singly, drifting with the wind-swept water, and were very readily alarmed, when they sank at once to the bottom. The water in which they flourished was exposed to the full glare of the sun and was quite hot. one + = no ee eee ee \\ \\\ WA\VAASN IY \\\\ \t VN \\ HUNKS VANS, Fig. 6. Larva of Culex thalassius, Theo.; head and end of abdomen a, mental plate. Another sample was collected from a pool into which the sea poured at each high tide (see Plate IIT. fig. 1); a small crab found in this pool was a most voracious feeder on the larvae. Other samples, however, were obtained from fresh-water pools, collections of rain- water, etc., so that a high degree of salinity is not essential for this species. No THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 149 difference in the length of the anal papillae or “ tracheal gills” could be detected between the larvae found in the brackish and the fresh water. The larvae were commonly found together with larvae of A. costalis and occasionally with those of O. minutus, O. albocephalus, C. invidiosus, C. tritaeniorhynchus, and -C. fatigans. Larvae of C. tigripes var. fuscus were found preying upon them. The imago is one of the most troublesome mosquitos in the bungalows at Accra and is a severe biter. Culex tritaeniorhynchus, Giles. This larva, which occurs in the same pools as the larva of C. invidiosus, resembles it closely in configuration. When the two larvae are examined side by side, however, that of C. tritaeniorhynchus seems to be more robust, to have a longer siphon and to be more strongly chitinised. The head is large, almost as wide as the thorax. The antenna is covered with spicules, and its outer third beyond the insertion of the hair-tuft is much darker and has more prominent spicules than the basal two-thirds. The tuft of very plumose hairs is placed at two-thirds of the length of the antenna from its base. The mouth- brushes are prominent. The mid-frontal hairs are plumose, and the ante-antennal tuft is formed of dark and plumose hairs rising from a pedicle. The eyes are large. The mental plate shows a moderate-sized median tooth with 6-8 pointed teeth on either side, its posterior border being deeply incised mesially. The mandibular teeth are more powerfully developed than those of C. invidiosus. The thorax carries the usual long plumes.. The lateral abdominal hairs are sub- plumose and are multiple on the first segment, triple on the second segment and paired on the remaining segments. All the abdominal segments show stellate hairs, which are longer than the similar and similarly placed hairs on the abdomen of the larva of C. invidiosus. The comb comprises 35-40 fringed scales, which appear to be darker than the comb-scales of C. amvidiosus. The siphonal and subsiphonal plumes are formed of plumose hairs, the hairs of the anal plume being simple. The siphon is about ten times as long as the diameter of its base; there are 10-12 spines, which are barbed inferiorly, inthe pecten. The siphon also carries 6-8 pairs of tufts of simple hairs distally to the pecten; these tufts are longer than the tufts on the siphon of C. invidiosus, and those visible ventrally are more regularly placed in line. The anal segment is almost as wide as it is long (23 units : 21); the anal papillae are equal, have pointed ends, and are longer than the anal segment. The beard is fairly developed, and the dorsal hairs on this segment resemble the similar hairs on the anal segment of C. invidiosus. This larva, as stated above, bears a striking resemblance to the larva of C. invidiosus, the apparently distinctive points have been tabulated below :— C. tritaenvorhynchus. C. invidiosus. Larva as a whole .. More robust. Less robust. Antenna me .. Apical portion above origin Apical portion _ scarcely of hair-tuft darker and darker, spicules not more with more prominent spi- prominent. cules. 150 A. INGRAM AND J. W. S. MACFIE. C. tritaeniorhynchus. C. invidiosus. Lateral abdominal hairs Paired after the first two Single after the first two segments. segments. Scales of comb.. .. Darker in colour. Lighter in colour. Siphon .. e .. Ten times the length of its Hight times the length of its diameter at the base. diameter at the base. Hair-tuftsbeyondpecten Hairs longer and tufts more Hairs shorter and tufts ir- or less regularly placed in regularly placed in a a line. spiral. Anal papillae .. .. Nearly equal in length. Ventral pair much shorter : than dorsal. Anal segment .. .. Almost as broad as itislong. Longer than broad. Breeding place.—The larvae were found in large numbers in a swamp formed by the heavy rains in June and were associated with A. costalis and C. fatigans, and after a day or two larvae of C. tagripes var. fuscus were found preying upon them. They were particularly numerous at one part of the swamp where the water had submerged a narrow path through the grass. On re-examining the swamp a week later none of the larvae could be found, but on this occasion natural enemies (Ephemerid larvae, dragon-fly larvae, Notonecta, larvae of Hydaticus and other species of water-beetles, etc.), which had not been observed on the first occasion, were extremely numerous. On another occasion larvae of C. tritaeniorhynchus were found together with those of C. invidiosus, C. thalassius, and A. costalis in a large water-hole by the road-side. The water in this hole contained 18 parts of chlorine per 100,000. ‘Culex guiarti, Blanch.* (fig. 7). The head is as wide as the thorax and is strongly chitinised. The antenna is slightly curved and stout; on its surface a few spicules occur, most conspicuously at its base. The antennal tuft, which is placed at about three-fourths of the length of the antenna from its base, consists of dark and plumose hairs. The apical portion of the shaft beyond the hair-tuft is considerably darker than the basal three-fourths. The mid-frontal hairs are plumose and paired, the ante-antennal tuft being formed of dark and plumose hairs. The brushes are strongly developed and the palps are large. The eye is large and band-like, extending on to both the dorsal and ventral surfaces. The “larval” eye is small and round, lying below and separated from the main eye. The mental plate is small, consisting of one stout median tooth with six or seven more slender teeth on each side. The dorsal thoracic plumes are formed of subplumose hairs, those on the anterior border being very long; the ventral thoracic plumes are formed of plumose hairs. The lateral abdominal hairs are multiple on the first segment, quadruple on the second and triple on the remainder. Plumose hairs form the siphonal and sub- *The adults associated with two other larvae have also been identified as C. guiarti (Bull. Ent. Res., vu, p. 7, footnote). Itis not possible to decide at present whether this is due to some error in association, or to the name OC. guiarti including more than one species of mosquito. THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. 151 siphonal plumes, while three or four pubescent hairs and an independent hair detached towards the subsiphonal plume represent the anal plume. The comb consists of 7-8 stout spines placed in an irregular row. The siphon is nearly nine times as long as the diameter of its base and carries a pecten of 8-9 rather slender x AN A Wy \ NN Fig. 7. Larva of Culex guiarti, Blanch.; head and end of abdomen; a, mental plate. spines, fringed at their bases. Four or five tufts formed of two to three simple hairs are visible beyond the last spine of the pecten. The anal segment is nearly twice as long as it is broad (30 : 16 units) ; the beard is somewhat scanty, but the dorsal 152 A. INGRAM AND J. W. S. MACFIE. hairs are well developed, consisting of three hairs above and one below on each side:. The anal papillae are slightly longer than the anal segment (35 : 30 units) and are. subequal with rather pointed ends. This larva may be distinguished from other Culex larvae having siphon tubes 8 x 1 by the possession of a comb consisting of 7-8 spines (Edwards’ key, Bull. Ent. Res., i, pp. 380-381). Breeding place.—These larvae were found at Nsawam (Plate IJ, fig. 1), 25 miles north of Accra, ina swamp (covered with Pistia stratvotes) which lies on both sides of the railway embankment, and were associated with the larvae of Mansonioides africanus and Mimomyia splendens. Mimomyia splendens, Theo. (fig. 8). The larva resembles those of Mimomyia hispida and M. mamomyiafornis in the shape of its antennae and in the hirsute appearance of its head. The head is dark and large, though not so broad as the thorax. The antennae are curved, with a tuft of plumose hairs at about two-thirds of the basal portion ; they are possessed of divergent apical pieces similar to those of MW. hispida and M. mimomyiaformis, these pieces being devoid of the spicules present on the basal portions, and at their junction with the basal portions are two long bristles. The apical piece in this larva does not appear to be lighter in colour than the rest of the antenna. <A bifurcated subplumose hair extends laterally from the head in front of the eye, as in M. hispida, and there is a small plume of simple hairs on the dorsal surface of the head immediately internal to the eye. The mental plate is small and flat, with a central tooth and six teeth on each side, the outermost of which are the smallest. All the frontal hairs are strongly plumose, the ante-antennal plumes resembling feathers. The long hairs on the thorax are plumose. The lateral abdominal hairs on the first two segments are multiple, large and plumose, those on the remaining segments being plumose, small and paired; and on all the abdominal segments there are conspicuous stellate hairs. The comb consists of a single row of large spines, 6-7 in number. The siphon is about three times as long as the diameter of its base, the pecten being formed of two spines, which are placed vertically and not horizontally asin M. hispida. There is a hair-tuft of subplumose hairs about the middle of the siphon and a pair of hooks on the dorsal valves. The subsiphonal plume is formed of subplumose hairs, the hairs of the siphonal plume being pubescent, while the anal plume consists of simple hairs; there are also two independent hairs, which are branched, on each side of the subsiphonal plume. The anal segment is longer dorsally than ventrally, has its posterior border edged with spicules, long and short spicules alternating, and possesses a single long and subplumose hair on its lateral aspect. The dorsal hairs are collected into tufts and are more strongly developed than the beard. The anal papillae are short, nearly equal in length (the ventral pair being slightly longer), and have pointed extremities ; they are about half the length of the anal segment. This larva may be distinguished from the larva of Mimomyia mimomyiaformis,. which also possesses an antenna with a divergent apical piece and a comb of 6-7 spines in a single row, by its pecten of relatively strongly developed spines, THE EARLY STAGES OF CERTAIN WEST AFRICAN MOSQUITOS. Too Breeding place.—The larvae were found in a small pool covered with the water- weed Pistia stratiotes (see Plate II, fig. 1) and were repeatedly obtained from the same pool from February to July. With them were associated the following larvae :—A. costalis, Aédeomyia africana, Culex quasigelidus and Mansonioides africanus. Fig. 8. Larva of Mimomyia splendens, Theo.; head and end of abdomen; a, mental plate. The larvae were very active in their movements, but when stationary lay practically parallel with the surface of the water. They appeared to be definitely associated with the weed Pistia stratiotes, as they were generally found under the leaves of this plant. (C394) 154 A. INGRAM AND J. W. S. MACFIE. The leaves of P. stratiotes have an unwettable surface and are strongly ribbed on the underside, so that as the plant expands the outer leaves come in contact with the surface of the water and carry down with them a film of air. It was to the under- surface of these outer leaves that the larvae of M. splendens attached themselves, lying horizontally with their siphons inserted into the film of air, and in maintaining this position the hooks described above were no doubt of great assistance. In this manner they were able to live for a long time (in one experiment 24 hours) without direct access to the air and may moult whilst still submerged. Chaoborus ceratopogones, Theo. The head is small and laterally compressed, as romp with the head of a larva of the CuLIcINAE, and carries very few hairs. Two delicate tufts of simple hairs are visible, one in front of and one behind the prominent eye. The mouth-parts de- scribed by Miall (“‘ Aquatic Insects” pp. 116-117) are very well seen in the larval pelt. There is a distinct neck. No plumes are present on the thorax, but on its lateral aspect short tufts of simple hairs occur and dorsal to these hair-tufts a single longer hair. The abdominal segments have laterally a single long hair and a tuft of ieee hairs ventral to this hair similar to the arrangement of the lateral hairs on the thorax. The usual paired and pigmented air sacs are present in the thorax and the seventh abdominal segment. The ventral brush on the ninth segment is composed of about 20 curved and elegantly plumose hairs. The anal papillae are short, the ventral pair being slightly longer than the dorsal. The hairs on the dorsal extremity of the ninth segment are one above and one below; they are nearly equal in length and are strongly plumose. BULL.” ENT: RESEARCH. -Vot.*VIIT, Parr 2. PDATE |. Fig. I. Swamp at Nsawam partly covered with Pistia plants; breeding place of Culex guiarti and Mansonioides africanus. Fig. 2. First stage larva of Fig. 3. Larva of M. africanus Mansonioides africanus. attached to roots of Pistia. ‘suapuajds erduioullp pue snuBolife saproruosuey ‘snpijabisenb xajng ‘eueodid fe ‘SNJNUIU SN}ZBJOLATYOO pur snisseyey}Z xXarng jo vod BrAulopay ‘I//eysdeul ‘vy ‘sije}soo sayaydou yp jo vox|d gulpsedq SuIpeddq ‘: UOOsR] B JO UlsdBW JY} 3B B[OY-qvdD ‘°% ‘sI4 £$a]017Bd]S BIZSIG paaMm-dayeM 9yy Aq pdd9AOD [OOg ‘| ‘SI - Fest * + > ye ek ‘ on "I SLwig 7 Lavel <DiAcIOA “HOUvasay “lie Wang ’ py as : en, « FUG: Not y ; mt Pat ‘ Set 0 Mies = ee oF" ae, ; Petey elas Wea F nt hie Pastel aay Oo dots - an VA te . | . “wy val ott “ ; . . ‘ s . - = ty | | | ‘s > . > Pa wa ; | i | ES ' . met rot i $A - a = ~ a | 7 j J Fe J A ae ; | - . 7 t = 4 : ce . ; ve) a Ceti NE *y ra iy ) 7 g cy r ¥ a s a a Aa ‘ 7 - | | ii 3 ( be 8 . ats : = 7 9 “4 pr’ ni i ce wa 7 (Pa any FE r a 7 a4 _ 7 ‘a ; , . a BULL, ENT: RESEARCH. Vot. VIII. Parr 2; PLATE III. Fig. 1. Pool flooded by the sea at high tide, in which larve of Anopheles costalis and Culex thalassius were found. Te pea Si ‘yy fe a ese 4 \ Bee NSS AE Sel ee a Fig. 2. Lagoon at Christiansborg, near Accra; breeding place of Culex thalassius. > oé a] = * - “3 =- = ‘ > : : d nae ea: ce oH PO) fae ee i ‘@ . ® oy — i et Sc i i ‘Bosnf ‘Byeydas0ajny “Ss pues Berduoyzedg pus wBleydascajn, ‘Ss ‘ezeresef erdurohazsg Borjyejaul *S ‘ruosduirs *S ‘ezeaurzrun BIhfulobajS jo BAdey] ‘Tyeysdeul “yp ‘sije}zsoo sajaydouy jo Bade, dd9M YOIyUM poeddnooo YOIYM UI ZJOY B YM dod} JUBAOqUIR[ ‘Z% ‘SI Ul YdOJ UIBLU BY} UI 9[OY B YIM ood} YuBAOqUIBI ‘| ‘“sI4 ‘Al ALv1d % Levey TA “10K “AOuvaSas INZ: “Ting fost vi nae ett wiges A SUMMARY OF OUR KNOWLEDGE OF INSECT VECTORS OF DISEASE. By MatcoLm Evan MacGrecor, B.A. (Cantab.), Wellcome Bureau of Scientific Research. So much has been said about insects since the War began that it is, | think, advisable that some attempt should be made to summarise our knowledge of the more important insect-borne diseases and their vectors. While insects have long been suspected of being responsible for the transmission of serious diseases, it may be said that practically the whole of our knowledge of insects in this réle has been acquired within the past twenty years. So rapidly, however, has the charge on this offence been made out against them that, although it is common knowledge that they have been proved guilty, it is not generally realised upon how many counts the verdict rests. Tt has lately been my good fortune to give class instruction for the War Office to officers of the R.A.M.C. who are proceeding to the East, and in order to bring home to my audience the importance of the connection between insects and disease I have compiled the tables which I now publish. These can in no way claim to be complete, but merely present the more important insect-borne diseases, including human diseases that on certain grounds are suspected of having insect vectors. With these tables I also publish one (Table VI) which includes the chief insects and Acarina that are directly the cause of disease in man and his domestic animals. To complete the list of insect-transmitted diseases would demand the consideration not only of other mammals as hosts, but also of avian and reptilian hosts. In the present instance this would be to carry the subject beyond general interest, but it must be remembered therefore that, long as the present list of charges is, insects are not here arraigned on all the counts that might with justice be preferred against them. During the last few years medical entomology has been rapidly establishing itself as an Invaluable branch of preventive medicine, and with the outbreak of the present war a great deal of interest and study has been devoted to this subject in Europe, notably in connection with the transmission of typhus fever by lice, and the dissemina- tion of bacteria and other organisms by flies. Centuries ago insects were suggested as being possibly concerned in the spread of disease, and from time to time such logical hypotheses were advanced that it is sur- prising that the establishment of the truth was not sooner forthcoming. In 1577 Mercurialis, an Italian physician, suggested that plague, which was then ravaging Kurope, was spread by flies feeding upon the diseased and dead, and later depositing faecal matter on food consumed by healthy persons. This idea held ground, and various suggestions occur as to the spread of disease by flies in the literature of the XVIth century. Edward Bancroft in 1769 advanced the theory that yaws was transmitted by flies feeding on diseased subjects and carrying the contagion by settling on open wounds or scratches on healthy persons. In 1848 Dr. Josiah Nott, of Mobile, Alabama, published a remarkable article in which he gave reasons for supposing that yellow fever was an insect-borne disease; but although he mentioned many insects, he did not specify any one as the particular vector. (C394) c2 156 MALCOLM EVAN MACGREGOR. The connection between malaria and the mosquito had long been held, it is said, by the Italian and Tyrolese peasants, and even by the natives of East Africa, but the first charge brought against the mosquito in the spread of disease, by scientific authority, was in connection with yellow fever. In 1853, Dr. Daniel Beauperthuy, a French physician, wrote ably arguing that yellow fever and other fevers were trans- mitted by mosquitoes, but in those days there being no accepted belief in a contagium vwum, he concluded that the virus was obtained from decomposing material that the mosquito had consumed and that was later accidentally inoculated into man. Raimbert, in 1869, showed by experiment that anthrax could be disseminated by flies. Kpoch-making in the history of our knowledge of insect vectors was Manson’s discovery, in 1878, that Filaria bancrofti was spread by mosquitoes; but at first he thought the Filaria escaped from the insect into water and reached man in this manner. Later work by Manson and his colleagues determined the exact means of transmission. It was not until 28 years after Beauperthuy’s theory that Charles Finlay, an American of Havana, definitely attributed, in 1881, the transmission of yellow fever to a mosquito of a definite species. He had noticed the connection that seemed to exist between the presence of large numbers of Stegomyia fasciata and the prevalence of yellow fever. He then attempted to transmit the disease experimentally by the bites of this mosquito, and whilst his experiments are open to criticism, there is no doubt that he did succeed in doing so. Three years later, in 1883, another American, A. F. A. King, advanced the first well-formulated mosquito-malaria theory, and in 1898 Ross, in India, demonstrated beyond doubt the important réle played by mosquitoes in the transmission of malaria. In 1899 the American Yellow Fever Commission (Reed, Carroll, Lezear and Agra- monte) was sent to Cuba and was there able to demonstrate with certainty that yellow fever is transmitted by S. fasciata. It is interesting thus to note the almost parallel development in time of our know- ledge of two of the most important insect-borne diseases. To deal even briefly with the historical aspects of our knowledge of other diseases tabulated below would be to consume a large amount of space, and the foregoing account will have indicated the path that has led to subsequent discoveries whose histories are readily available. I will pass therefore to a few notes on each of the Tables. Notes to Table I.— Diseases of Unknown Origin. The majority of these diseases are doubtless caused by living viruses, often organisms of ultra-microscopic size, and commonly referred to as “ filterable viruses.” In the case of pellagra it would appear, however, from the most recent work that although it is still considered by many persons to be possibly an insect-borne disease (and according to Sambon having a likely vector in either the CERATOPOGONINAE or SIMULIIDAE) Goldberger in America considers it a disease now certainly attributable to vitamine-starvation through an unbalanced diet. If this is the case, there is no causative organism and no vector, and pellagra should be ruled out of present con- sideration. The question nevertheless is by no means sett!ed. OUR KNOWLEDGE OF INSECT VECTORS OF DISEASE, 157 The virus of acute anterior poliomyelitis is still not isolated with certainty. Flexner and his colleagues have been able to cultivate a filterable micro-organism which produced the disease in experimental animals, and more recently Rosenow and his fellow-workers have isolated a polymorphous streptococcus with which they were also able to produce paralysis in certain animals. Nuzum and Horzog were able to do likewise by a gram-positive micrococcus isolated from the brain and spinal chord of persons dead from the disease. Poliomyelitis has been very generally suspected of being transmitted by insects, particularly by Stomoxys calcitrans (the stable fly), fleas, and TaABANIDAE (gad-flies). Nevertheless it appears more likely that it has an aerial transmission, infection being acquired through the buccal and nasal mucous membranes, | Wolbach claims to have discovered the causative organism of Rocky Mountain spotted fever in the bodies of infective ticks (Dermacentor venustus). TABLE I. The more Important Insect-borne Diseases of Unknown Origin. (N.B.—Names in these tables marked with an asterisk are certain vectors; names not so marked are probable vectors ; names followed by (?) are possible vectors. ) | Organism. "Host. Disease. Vector. | Man. | Dengue, or breakbone | Sandflies (Phlebotomus),* mosquitoes fever. (Culex fatigans).* 2 | 4, | Three-day fever (sandfly | Sandflies