tT Og ing ae ad : ey PI Og IN samen Tag a be 2 SE man ase hk Vay ofan aeey at = ~ ~. 2 > ae — a a Se om) x os . OF A Bante ae Se SPE NSE AS SNS Qe PRO age Pe ey ata es —, wast» yy ah oe! -—s 42 — x . es all - 5 r . — <> Se a oe et a epee or = ee oe = aor Te mS a a ahi SB NR ea a me OS ae re se = RISES oS oC Vitis pete ae ae a. ie; rf ed . - % Sees es at, aGe 4. (easy 130. Pi, A! Be: Neale nove stl iin Ne lage” ree oe - fg PR ees : wee eu he Gi 4 “ wy tes, . , yy + 7 2133 vot h ATA 1d 2 gor ee aon Ap. =. 09 Po Woe , fs #4 au EY fats! tthe? Hil i o Toes é P's rs PS 0G bal ath: ‘s +e = — a » > _ ; > ) ~ o Loree {Tse A Nec letce . a i heabar'e Tere eos 0) 0 at vin ea OE aay Ag in Set Sa ae. SS ne Pt ae oe ——, - =. ~ 7 - =e AS ay ates ie, — ~* — iS a) aS ae AO on = ; Nie : at oe a, . - a 7 - * ~ = Re Ee ON ae ee eee A : ee : i cae &: “— 5 , ~ ee = ; <<. e : P “eT - ~ . es ~ u ” . oa re = rns OOS LTTE * soda tiedtelnen t “rs res ‘ Pe - 7 ‘ : : 7 5 4 . 7 z 7 2 4 sd Ne . = 2 . Pm Oe . * ar a ot, © : - ad 2% ” a ta 1 7 7 of - eet 7 ra a he 7 ae 7 7 -% : = : oF 5, ae - : po > F ; : eee: os . = s ora Poul . 7 > . - : rte Seog : oA ’ 7 ’ 7 : = z + 4 r] Sy 5 © 7 adh : = Bn on . . -. , ; m ; < ; : - 7 rm = . ane ® 2 7 > 7 ee ee ¥ < wt 7 - . Li % 2 - 7 . : 7 os 7 + a 7 . * . Ml - ‘ 7 "7 A a ane eS oe f- ; - : fi iy 2 es ‘ . y 7 ’ - i mee - . 7 - me : : * 2 ‘ se. 7 “ aly ae : net my ; : x * , A 7 ; a ; i ; . ba fe. A Ez - ; . + _ 5 a be oe yoe 7 7 ™ - J Py ib ow: = ¥ Mg ' 7 2 . t ; = 7 ‘ ; - : : a “a ” « 7 . + ; - + - : : —_ : : . - - . he © . a - . 7 7 ; : : a ’ H ~~ i - , - . : - : - . . e « . Z a2 x ; : . A : . ¢ “e : se . fo 7 : : 7 - . = : m - ‘ 7 ae : : a“ é ; ‘ cd a ™ . ‘ t ' » i 2 i 7 . - . s = ss - . : ¢ ’ « , ' : : - 7 ro. 7 : : ag 3 7 : ‘ . Pe 7 - - 4 cand ~ - = q 7 - 7 . . ate : = te | : : : et - ’ 2 P5 oN ‘ bi) Mom > a Fe bal ° ‘ . + boa : xs - : ‘ : 7 Pare ~ . as fee Ae P ’ et z* ry o 7 Z . : io ~ 2. ° ~ : * ate U = “i a , v9 ay ae ae F Ne an F % + F ais : : >t ute - 7 a 7 5 bes Z sent 3 DI vee or Ry eerte ky 7 ‘ s 2 We al ot Oe gx 5 eel eS oe a te bi . Ys - - 1 7 ; a #). hg . = - - . * 7 : : & : fut: « on nt i 7 : oe : 7 . 2 7 - = *, ’ if me 7 . 7 a ' 1" d - : ‘ ; i . Ms ¢ - » te we? 4 a * = F 7 Y hd * . .* & yy ma 2 f 3 ; aH ; > erm ; yy 7 7 > . 5 * - ’ ; 7 - cam ” = + 7 . EN, THE FAUNA OF BRITISH INDIA, INCLUDING CEYLON AND BURMA. PUBLISHED UNDER THE PATRONAGE OF THE SECRETARY OF STATE. HZDITED BY LT.-COL. R. B. 8S. SEWELL, C.I.E., Sc.D., F.R.S.. 1.M.S. (reé.). ASSISTED BY G. TALBOT, F.R.S.E. VOL. V. SPHINGIDSE. BY fh): BELL, CEES LES: (re), PRES: AND Lirvut.-CoLonet F. B. SCOTT, I.A., F.R.E:S. TAYLOR AND FRANCIS, LTD., RED LION COURT, FLEET STREET, LONDON, June 15, 1987. GM \THSONIANy oct 2° 1986 Ligranics ALIEKIc FLAMMAM. PRINTED BY TAYLOR AND FRANCIS, LTD., RED LION COURT, FLEET STREET. CONTENTS. Page TEE TRTRIAUT STA? TSC 0 a a Vv SCUTESP TES) 1 STE OTN 0) a Vil eOSoARY, OF TERMS ...........5 0% PED een het deat 5. ae xi RSME VE WAICOMDEN DHE St. 5S) credo buals ole creole sc beaid cen we XV Serra RES aT LONI OU CUTIOIN & Sasityoy ee a) ced dh wie ee dle lev eos a Porat g I IMEI OU OC VE) (HOO) 5) Mea cee asters ac. ticles: o seestins oe a 5 a (LESTE Ae Pa Aas dh 3 8 a gee te dey 55 (LEAD OR) PRG oie ee cach Aig emer eee 15 Fe (linn aio ees sis ee Tee nee aries Isat 2. 18 TE UATECTIRSY UE a ARS oe ANY, STEN 8 0c mg arta pn a ea 28 SR WS CER OING Pore ier Nee ere a ae ee ee a ae 33° 2 EEC TEUTUL TET ON es Ee ET As Se Rd eae ee OL 42, 2 PEILU D-GIRUAIETEING aa OW SARE Ss At er ats Cn 493 AppenpDix A.—The Kanara District of South India: Its Topography and Flora in relation to the Distribution of the SpHinGcID& (7. R. Bell) APPENDIX B.—Food-plants of the Indian Hawk-Moths.. 507 JA TEL ETE DDE OU ML OVOTE: 1 OSHS T-< lege el me RO orto 533. a 2 Pier ATOR Y* NOTE. Wir the completion of the present volume our knowledge of the Hawk-Moths of the Indian Empire is now so advanced. that a considerable time must elapse before any important additions can, be made to it. The authors and the editors have tried to make the volume as complete as possible, and the importance of structure, as well as pattern, is emphasized by the inclusion of full morphological details and illustrations. To this is added, for most species, an account of the early stages and illustra- tions of them, much of which is here published for the first. time. By kind permission of Lord Rothschild and Dr. K. Jordan, free use has been made of their ‘ Revision of the Sphingide,’ published in 1903. The text-figures of structure, excepting figs. 1-4, selected for the present work, were all reproduced from Rothschild and Jordan’s book by Mr. W. H. T. Tams, of the Natural History Museum. The opportunity is taken here to acknowledge the useful and ready help received from the Editor, Mr. Tams and the authors in the task of preparing the work for publication ; a special debt is owing to Mr. Tams for also reading through the page-proofs. Thanks are tendered to the printers for the very careful marking of the proofs. Acknowledgement is tendered to the India Office for permission to use twenty of the blocks illustrating Hampson’s work (1892). The sources. of other figures are acknowledged by the authors. vi PREFATORY NOTE. The following figures are from, photographs and drawings py the authors :—1, 2, 3, 4, 9A, 18 A, 29, 34, 44, 66, 69, 75, 79, 92, 96, 104, 107, 108, 112, 113, 115, 121, 122, 124. The following figures are from photographs by W. H. T. Tams :—15, 16 A, 20, 24, 25, 26, 27, 30, 36, 38, 39, 40, 41, 46, 47, 48, 50, 52, 54, 55, 56, 61, 62, 67, 71 B, 73, 77, 78, 81, 82, 86, 93, 95, 97, 98, 99, 100, 102, 103, 106, 116, 117, 118. The following figures are from Hampson (1892) :—9C, 10 A, 14 A, 19, 32, 37, 51 A, 53 A, 64 A, 71 A, 72 A, 75 A, 80 A, 84, 85 A, 89 A, 94, 109, 111. The following figures are reproduced by permission of the Bombay Natural History Society :—Text-figures 30 and 44; Pl. XIII, figs. 2, 6,8; Pl. XIV, figs. 2 and5; Pl. XV, figs. 4 10, 11. Text-figures 77 and 78 are taken from the Ent. Zeitschr. Frank., 1930 and 1931. THE SuB-EDITOR. AUTHORS’ PREFACE. THe Hawk-Moths of India were first classified by Sir George Hampson in ‘The Fauna of British India.—Moths,’ vol. i, published in 1892. In 1903 Rothschild and Jordan published ‘A Revision of the Sphingide,’ dealing with the Hawk-Moths of the World, and in 1904 Sir George Hampson published supplementary papers to ‘The Fauna of British India ’ volumes, in which he adopted Rothschild and Jordan’s classifi- cation as applied to the Hawk-Moths of India, in the ‘ Journal of the Bombay Natural History Society,’ vol. xv, p. 630. Since that date nothing has been published dealing with the Hawk-Moths of India as a whole and separate from those of other regions. | In this volume we attempt to remedy this deficiency, and to bring together information which has been published by other authors, supplemented by the results of our own studies, in a form suitable for the use of students in India. Our work covers the whole Indian Empire, including Burma and the Andaman Islands, and Ceylon. Our own studies have been concerned chiefly with the early stages, which have been hardly touched upon by previous authors. We have bred a large proportion of the Indian Hawk-Moths, and have made careful descriptions of the early stages, life-history and habits, supplemented by paintings of the early stages of many species made by Miss E. M. Bell and by Lt.-Col. F. B. Scott, Indian Army, who has also taken photographs of the caterpillars and moths of many species. Vill AUTHORS’ PREFACE. We have also used some paintings made by Indian artists, some under the supervision of the late J. Davidson, EKsq., I.C.S., and some under our own supervision. A Bibliography is included giving a list of the principal literature of the subject. This should be consulted for author and year to which reference is made under each species. in the body of the work. The principal works consulted by us, and repeatedly cited,. are as follows : F. Moore, F.Z.S., ‘ The Lepidoptera of Ceylon,’ vol. ii (1882). Sir George F. Hampson, ‘ Fauna of British India.—Moths,’ vol. i (1892), vol. iv (1896). Professor Dr. Ernst Hofman, ‘Die Raupen der Gross- schmetterling ’ (1893). The Hon. Walter Rothschild and Dr. Karl J ordan, “A Revision of the Lepidopterous Family Sphingide.” ‘ Novitates Zoologice,’ vol. ix, Supplement (1903). Richard South, ‘ The Moths of the British Isles ’ (1907). The Rev. A. Miles Moss, ‘“‘ Sphingide of Para, Brazil.” ‘Novitates Zoologice,’ vol. xxvii, pp. 333-424 (1920). Rudolph Mell, ‘ Biologie und Systematik der Sitidchine- sischen Sphingiden ’ (1922). Adalbert Seitz, ‘The Macrolepidoptera of the World,’ vol. 11 (1911), vol. x (1928). Also numerous papers in ‘The Journal of the Bombay Natural History Society.’ Hampson’s work was extremely useful in giving the first description of Indian Hawk-Moths in a form available to. the public in India, but was entirely superseded on the publication of Rothschild and Jordan’s ‘ Revision’ in 1903. This outstanding work is a complete monograph of the whole family known to exist at the time of publication, and _ has. many plates of the moths and their structure ; later volumes of ‘ Novitates Zoologice ’ give descriptions of some of the AUTHORS’ PREFACE. 1x species discovered since 19038. Little was known of the early stages of any Hawk-Moths when the ‘ Revision’ was published, and Mell’s work, which is of real scientific value, is the first to give an account of the early stages of any Indian Hawk-Moths. It deals with the species occurring in South China, and contains many excellent plates of species common to 8. China and India. Permission was granted to quote freely from Mell’s work, and this we have done in respect of the larve and pupz of species occurring in India but not bred by us. Seitz, in his great work on the ‘ Macrolepidoptera of the World,’ follows the ‘ Revision ’ very closely. He gives many excellent plates of the moths. Miles Moss, in his work on the ‘Sphingide of Para,’ has very interesting notes on the life-history and habits of the Hawk-Moths of South America, and some excellent plates of the curious larve and pupz of that region. All our specimens have been identified by Lord Rothschild and Dr. Jordan, at the Tring Museum. New species and sub- species discovered by us have been described and named by the same authorities, to whom our best thanks are due for the great assistance and advice they have given us in this and other matters. The work of putting the manuscript into shape for publica- tion, of preparing the bibliography, glossary and indices, and of correcting the proofs, has devolved upon Mr. G. Talbot as sub-editor. We thank him for all the trouble he has taken, and for the efficient way in which he has carried out the task. We most certainly could not have coped with this work in anything like the manner in which Mr. Talbot’s experience has enabled him to do. The arrangement of the figures in the plates and text has been done jointly by Messrs. Tams and Talbot. We are very greatly indebted to Mr. Tams for undertaking the self-imposed task of making the large number of photo- graphs used to illustrate the text. These comprise not x AUTHORS’ PREFACE. only the structural figures copied from Rothschild and Jordan’s ‘ Revision,’ but also figures of types of species described by Rothschild, Jordan, Hampson, Butler and Walker. To this end some specimens were kindly loaned from the Tring Museum, and others were made available by permission of the ‘Trustees of the British Museum. Mr. Tams’s help was freely given in his desire to render the book as complete and. useful as possible. We also desire to acknowledge the valuable help given in collecting and breeding by Brigadier H. L. Scott, D.S.O., M.C., brother of the junior author, and by Colonel Campbell, D.S.O. In an appendix we give a list of the 135 species and sub- species of which the food-plants are known, together with the names of these plants; also an appendix containing a list of the food-plants, with the species which feed upon them. T. R. D. BELL. F. B. Scort. London, June, 1937. GLOSSARY OF TERMS. Aciculate.—A surface that appears as if scratched with a needle. Acuminate.—Tapering to a long point. Aideagus.—The outer chitinized sheath of the membranaceous penis. Bevels.—Lateral (sometimes ridged) basal slopes of the movable segments of the abdomen in the pupa. Carinate:—Keel-shaped. Cariniform.—tIn the form of a keel. Chextotaxy.—The arrangement of the set or bristles on any portion of the exoskeleton. Cilia.—Series of fine hairs arranged in tufts or single lines; the fringe which edges the wing. Cilium.—PI. cilia, q.v. Clasper.—A chitinized plate or flap, being a process attached to the ninth sternite in the imago, and serving to hold the female parts during copulation. Also applied to the pair of prolegs situated upon the tenth segment of the larva. Clavate.—Clubbed ; thickened gradually towards the tip. Clypeus.—That sclerite of the head to which the labrum is attached. Comb.—A row of long bristles, often prominent on the mid-tarsus of SPHINGIDZ. Costa.—The thickened anterior margin of the wing. Coxa.—The basal segment of the leg, by means of which it is articulated to the body. Coxal piece.—A small, bipartite, diamond-shaped plate on each side of the middle line of the pupal thorax, lying between the fore leg and the tongue-case. Cremaster.—A specialized process on the tenth (the last) segment of the pupa. Crenulate.—With small scallops. Cross-veins.—Applied to the veins closing the discal cell of the wing. Also called the discocellulars. Dentition.—Arrangement of the teeth, or the form of the teeth, on a chitinized part. Dextrad.—Extending or directed towards the right. Discocellulars.—The veins which close the discal cell of the wing. Distal.—F arthest from the body, opposed to proximal. 7 . ‘ 4 / & Xil GLOSSARY OF TERMS. Emarginate.—Notched ; with an obtuse, rounded or quadrate section. cut from a margin. Epicranium.—A dorsal plate of the fend forming laterally the sockets. for the antenne. Episternum.—A ventral thoracic plate lying below the parasternum. Epistome.—A plate of the head covering the base of the tongue. Falcate.—Curved like a sickle. False clypeus.—Area between clypeus and lobes of head; a very thin band on each side of true clypeus, not seen in the early instars. Fasciculate.—Bundled ; clustered as in a bundle; tufted. Femur.—tThe third segment of the leg. Filiform.—Slender, and of equal diameter ; hair-like. Frass.—The excreted pellets of the larva. Frenulum.—A set of fused or closely contiguous bristles arising from. near base of costa of hind wing, and used as a link with the fore wing in flight (see retinaculum). Frietion scales—Modified scales found upon the male clasper, and supposed to assist in stridulating. frons.—The front or anterior portion of head lying between the bases. of the antenne and the clypeus. Fusiform.—Spindle-shaped ; tapering gradually to each end. Genal process.—A more or less triangular projection between the pilifer and eye. Glabrous.—Smooth, not hairy. Harpe.—The ventral armature of the clasper. Horn.—A fleshy process on the dorsum of the eighth abdominal sepment of the larva. Incrassate.—Thickened ; rather markedly swollen at some one point, especially near the tip. Instar.—The period or stage between moults in the larva. The first instar is the stage between the egg and the first moult; the final instar the stage preceding the change to a pupa. Irrorate.—Marked with minute points ; freckled. Labium.—tThe lower lip ; a compound structure which forms the floor of the mouth. Labrum.—The upper lip ; covers the base of the mandible and forms the roof of the mouth. i Ligula.—The part of the labium in front of the mentum. ~ Mentum.—A transverse arched stripe of chitin between the labial! palpi; the second segment of the labium. Merum.—tThe lateral area of the sternites of the thorax. Mesial.—At the middle. Mesonotum.—The upper surface of the second or middle segment of the thorax. GLOSSARY OF TERMS. Xlll Mesothorax.—The second or middle segment of the thorax ; bears the middle legs and anterior wings. Metathorax.—The third segment of the thorax ; bears the hind legs and second pair of wings. . Notum.—The dorsal or upper part of a segment of the thorax. ‘Obsolescent.—Becoming obsolete. Obsolete.—Nearly or entirely lost. Onychium.—The tarsal claw. Paramerum.—A small plate in the lateral ventral part of the mesothorax. Parapleurum.—A lateral plate of the first abdominal segment. Parasternum.—A large thoracic ventral plate extending obliquely dorsad and mesiad from the meral suture, separating the meral and sternal parts of the sternite, to the membrane connecting meso- and prothorax. Paronychium.—Two (or sometimes one) slender lobes arising from the tarsal claw ; the so-called false claws. Pectinate.—Comb-shaped ; with even branches like the teeth of a comb. Penis-funnel.—An aperture with raised and chitinized edges, situated between the tenth abdominal sternite and the ninth, from which protrudes the penis-sheath. Penis-sheath—The outer chitinized covering of the membranaceous penis ; also called the edeagus. Peristernum.—An anterior ventral plate of the mesothorax. Pilifer—aA small sclerite, in the form of a hairy process, at each side of the clypeus. Postscutum.—The third posterior dorsal plate of the anterior thorax. Prescutum.—The small anterior plate of the thorax. Pro.—Anterior. Prolegs.—The fleshy unjoined abdominal legs of the larva; those of the tenth segment sometimes called claspers. Prothorax.—The first segment of the thorax; bears the anterior legs but no wings. Protomerum.—A small plate in the lateral ventral part of the thorax lying below the paramerum. Proximal.—Nearest to the body. Pulvillus.—A pad-like structure between the tarsal claws. Retinaculum.—A catch or bar on the underside of the fore wing, formed by stiff bristles, scales, or projecting membrane, and serving to engage the frenulum in flight. Rugose.—Wrinkled. Sclerite.—Any piece of the body-wall bounded by sutures. Scutum.—A dorsal area of the thorax. Seriate.—Placed in a row. Setiferous.—Bearing sete or bristles. Setiform.—In the form of a bristle or seta. ‘Shagreened.—A surface roughened with minute tooth-like projections. X1V GLOSSARY OF TERMS. Sinistrad.—Towards the left. Spinulose.—Set with spines or spinules, Spiracles.—The lateral pores in the insect body through which air enters the trachee. Sternite.—The ventral piece of a ring or segment. Stigma.—aA spiracle or breathing pore; a patch of modified scales on the wing; a small discocellular spot. Subspiracular.—Applied to the lateral surface of the larva; the longi- tudinal area just below the spiracles. Sulcate.—Grooved. Tarsus.—The distal portion of the leg, consisting of five segments. Tegula.—A small lateral sclerite of the mesonotum, situated just in front of and usually covering the base of the fore wing. Tegumen.—The ninth abdominal tergite in the ¢. Tergite.—The dorsal part of a segment of the abdomen. Termen.—The outer margin of the wing. Tibia.—The fourth segment of the leg, situated between femur and tarsus. Trochanter.—The small second segment of the leg lymg between the coxa and femur. Tubercle.—A little solid pimple or short fleshy process, which may or may not give rise to a seta. Tuberculate.—Covered with tubercles. Tumid.—Swollen ; enlarged. Turgid.—Tumid. Uncus.—The tenth abdominal tergite in the 3. Vaginal plate——The part of the vaginal area in front of and behind the orifice. Venter.—The belly or under surface of insects in any stage. SYSTEMATIC INDEX. Page A. Sphingide—Asemanophore 47 Subfam. ACHERONTINZ Buil. 48 Wmibe Acherontiiini .........-. 51 Gen. Acherontia Lasp. ..... 52 ielsechesis (Far)! ....... 55 2. styx styx (Westw.) ... 58 sen. Horse Ofen 0... b. 60 3. convolvuli convolvuli ‘(LITEOL OS) Weg ORS ae oa 61 Gen. Megacorma fRoths. & PUI PROS AN cote ea Seo « 65 4. obliqua (Walk.) ...... 66 | PUB E MS PINGING 2.005.206 s sss 67 Gen. Meganoton Boisd. ..... 69 parame (Held) 95 ..... 5. 70 6. nyctiphanes (Walk.) .. 73 7. rufescens rufescens IPED A Ne OR aks eel tg 76 Gen. Psilogramma Roths. & JC HOLE LE SES ee nae 7 8. menephron menephron (OTRO DA OS IPRE ORE amen isin Gen. Apocalypsis Butl. ..... 81 ORRWCIOKMUSUE. 0s os eae 82 Gen. Pentateucha Swinh. ... 85 10. curiosa Swinh. ...... 85 Gen. Pseudodolbina Roths. .. 86 11 @.. fo.celator Jord. ..... 87 b: foto (Wale) «sine. 0% 88 | 12. zqualis Roths.& Jord.. 91 Gen... Sphinx Linn. 2.0... 92 I3. ‘uniforms: (Buitl.) .\..... 92 Tribe Sphingulinta ........:.- 93 Gen. Dolbinopsis Roths. & TOTO, Mer ceteeusis 6 shes es 93 14. grisea (Hamps.) ...... 94 fen. Dolbina, Sgr, «2. a + ex 94 | 15. inexacta (Walk.) ..... 95 Page Subfam. AMBULICINZE Buitl.... 98 Gen. Compsogene Roths. & LI OAT ick PA eR ra er oP 103 16. panopus panopus COREL DAA Syst eee eRe 104 Te ATAS OTN Ebest, 6 ehs 0 108 Gen. Oxyambulyx Roths. & ORD re Nae eh hens 109 18a. sericeipennis _ sericei- pennis (ST) nies n, 6 115 6. sericeipennis agana POT cert ieee. Meum p eel ya 117 19. placida (Moore) ...... 118 20. maculifera (Walk.) .... 118 Zier EUG ION. & ae ery tis. 119 27 leanOrey CUE.) weg. 5 aces 123 23. ochracea (Buil.)...... . 124 24. liturata liturata (Butl.). 125 25 a. substrigilis auripennis (LOGON sehiis Sy ove. 127 6. substrigilis aglaia HOLM es euleeive. cosets 128 c. substrigilis substrigilis (RACES POD tein ae ae hae 131 ZO. WAU Gl OnDas ee 6 eae. p sc 132 27. canescens canescens (fA 071 DE) a eee an 135 28. subocellata (Feld.)..... 136 29. cyclasticta Joicey & K.. 139 Gen. Clamis: Witbi.c 2... 6. :. - 139 30. phalaris (Cram.) ...... 142 31. undulosa undulosa CVECORE) Rit. spage ofa a0? « 146 32. deucalion (Walk.)..... 149 33. bilineata bilineata CVE aie acim oo Stam 150 34. titan titan Roths. & 1) Di See a ae 153 Gen. Leucophlebia Westw. .. 156 30. lineata Westw. ....... 7 36. 6mittens Walk......... 159 Xvi Page Gen. Polyptychus Hiibn..... 160 37 a. trilineatus luteatus LOUUS (Gl J OTA. |. «cai 163 6. trilineatus sonanthis SON. BR Pic sce sree fe venatens 163 c. trilmeatus trilineatus MVEOOTE 257. Ais o 51 cited 167 d. trilineatus undatus Roths. & Jord. 5. 5. 168 e. trilineatus mincopicus Ki R0 een” C'alGrr nec 168 38. dentatus (Cram.) ..... 169 Gen. Marumba Moore....... 172 39. gaschkewitschi fortis POR. oc: Gee cee chess 175 40. cristata cristata CEUEL Ne einer sperickede tens 176 41. spectabilis spectabilis (Butliye See eee ee 178 42. dyras dyras (Walk.) ... 179 43. nympha Roths. &. HOD: os 2 ckotgee es etras 182 44q. sperchius albicans (BU i eos, home ecto ree 185 6. sperchius gigas (Butl.) 185 45. poliotis Hamps. ...... 187 46. andieusi( Wallen) ee sien: 189 47. bengalensis Hamps. ... 190 48. decoratus (Moore) .... 191 “Gen. Daphnusa Walk........ 192 49. ocellaris ocellaris Walk. 193 ‘Gen. Langia Moore ........ 194 50. zenzeroides zenzeroides Mi OOPe 3 ixcs cies Sains wit 194 ‘Gen. Rhodoprasina Roths. & DONA.) ccs Cesccaleeegees 198 ble Hloralis (Buti) mes - ee 199 52, callantha Jord.-. 3... - 200 ‘Gen. Clanidopsis Roths. & Ord. th. er ree renetoe 206 ESy ep-qblyaey (WoW oacdace 207 ‘Gen. Agnosia Roths. & Jord. . 209 54. orneus (Westw.) ...... 210 55. microta (Hamps.)..... 213 ‘Gen. Parum Roths. & Jord... 213 56. porphyria (Buil.)...... 214 57. colligata (Walk.) ..... 215 “Gen. Oypa Walk... «sc 217 58. decolor decolor (Walk.). 218 59a. pallens pallens Jord. . 219 b. pallens enodis Jord. . 220) 60. ferruginea Walk. 222 Gen. Smerinthulus Huwe ... 223 61. perversa (Roths.) ..... 223 Gen. Degmaptera Hamps. 224 62) amirabilis (Roths.) 2.7... 224 “Gen. Callambulyx Roths. & PUORG pretersiornii oie. vis, HAE 228 SYSTEMATIC INDEX. Page | 63. rubricosa rubricosa i (Wealk:) hee 229 64. junonia (Buil.)........ 230 65. poecilus (Roths.) ...... 231 Gen. Anambulyx Roths. & J OT... Lee eee 231 66. elwesi (Druce) ........ 232 Gen. Smerinthus Latr....... 232 67. kindermanni obsoleta Stor... setae eee 233 Gen. Phyllosphingia Swinh. . 234 68. dissimilis perundulans SSWUIN sx. .<'e epee 235 B. Sphingide—Semanophore. 236 Subfam. Sestina Roths. & JORT a Flies see ace lense 238 Uribe Ses7ini ee eee eee 238 Gen. Hemorrhagia Grote & ROD. Oe eee 239 69. fuciformis fuciformis (LAnn)) ee ee ee 242 70. saundersi(Walk.) ..... 243 71.. rubra (Hamps.) ....>- 244 Gen. Cephonodes Hiibn. .... 244 72. hylas hylas (Linn.) ... 247 7a. pleus (Crain) ane eee 250 Gen. Sataspes Moore ....... 251 74. infernalis (Westw.) .... 253 f. infernalis Westw. .. 255 f. uniformis Buil. 256 15, tagalica Botsd-eee 256 f. tagalica Boisd..... 256 f. thoracica Roths. & J OTE: eee 256 f. collaris Roths. & J OF ha sae oe ak 257 f. hauxwelli de Nicév. 257 76. scotti Jord... a seeieeme 257 Subfam. PHILAMPELIN2 Koths. & J0rd... oe eee 259 Tribe Nephelini.... - er ee 259 Gen. Chromis Bvibn= =ao-ee 264 77. erotus erotus (Cram.) . 264 Gen. Deilephila Lasp. ...... 266 48. ners (Linn) Greece 268 79. hypothous hypothous (Cram.) 95. eee 271 80. layardi (Moore) ...... 273 81. placida placida (Walk.). 273 82a. minima _ernestina (Moore) \ eer 274 6b. minima minima (Butl.) 274 Gen. Dahira Mcore eee 276 83. rubiginosa Moore ..... 277 SYSTEMATIC INDEX. | | Page Gen. Ampelophaga Brem. & GEE Ns CL ces 2 278 84a. rubiginosa fasciosa MEOOTCN Tei w Peeks ortts 280 84 6. rubiginosa harterti LOLS SM RO NAG Oe Eke ree 282 85. khasiana khasiana 5 LOL NS Ns dante Wd, 5. 283 86. dolichoides (Feld.)..... 284 87. obliquifascia Hamps. .. 285 Gen. Elibia Walk. ......... 285 88. dolichus (Westw.) ..... 287 Gen. Acosmerycoides Mell... 287 89. leucocraspis leuco- craspis (Hamps.) .... 288 Gen. Acosmeryx Boisd. .... 289 WOenata (Moore)... 2... 291 91. anceus subdendata Moths. ee Jord. oi. -- 294 92. socrates Boisd. ....... 395 7..cinerea) Butl, <.... 295 93. sericeus sericeus (Walk.) 297 94. omissa Roths. & Jord. . 298 Gen. Lepchina Oberih. ...... 299 95. tridens Oberth.. ....... 299 Genus Panacra Walk. .e:.... 299 96 a. busiris atima Roths. & 2) Oi lec SM atone eee 302 6. busiris busiris Walk... 304 c. busiris marina Roths. BNO ee a SIN Noe eas vs 305 97. automedon Walk. . 305 98. moseri Gehlen ........ 306 99: dohertyi Roths. ....... 307 HOO: variolosa Walk. ....... 308 101. sinuata Roths. & Jord.. 308 | 102 a. metallica anfracta eC RICM ein cta te Ae ais Ss 309 6. metallica metallica LED AL, «ap eg ae te a et 312 103. perfecta: Buil. ........ 312 104. mydon mydon Walk... 313 Gen. Angonyx Boisd. ...... 315 105. testacea testacea (Walk.) 316 Gen. Enpinanga Roths. & SCC ee 318 106. assamensis (Walk.) 319 107. Jabuana oceanica Roths. GUT OLUE Reeser: | ss 3 320 Gen Cizara Wales f.2 5.5 3: 320 108. sculpta (Feld.) ....... 321 | Gen. Nephele Hitbn. ........ 324 | 109. didyma f/f. didyma CPGDT 2) sae ratee eee 3. 325 didyma f. hespera Ch abr.) 2 eee Ss S2On | Gen, Gurelea Kby. 22.2: .:.. 328 5 | 110. hyas hyas (Walk.)..... 331 WO Lae V.: Page 111. masuriensis (Buil.) .... 334 112. himachala himachala PACED Nae ie eet 228 336 113. montana ordiin i... 338 Gen. Sphingonepiopsis Wllgr. 338 114. pumilio (Botsd.)....... 339 Gen. Eurypteryx Feld....... 34] 115. bhaga bhaga (Moore) .. 342 Gen. Rhodosoma Buil. ..... 343 116. triopus (Westw.) ...... 344 Gen. Macroglossum Scop. ... 345 117. stellatarum (Linn.) ... 352 118. bombylans (Boisd.) ... 353 119. regulus (Bovsd:) Ve se 355 120. gyrans (Walk.) ....... 356 21h paikiebitian(B Uti) fee 360 122. particolor Roths. & Jord. 362 123; belis, (Lan). 12°25 4. 365 124. assimilis Swains. ..... 368 125. pyrrhosticta pyrrhosticta CUTIE Niwas | Biel a Se 370 126. troglodytus (Boisd.) .. 372 127. insipida insipida (Buil.) 373 28. viene Ords 2s. «ies 376 129. sitiene (Walk.) ....... 378 130. fringilla (Bowsd.) ..... 379 131. divergens (Walk.)..... 382 132. prometheus prometheus CEGUSO aN iin sie a raes oes 383 133. variegatum Roths. & SN OKUR « Buspteekcton Eis theme 384 134. saga (Buil.) .....4.... 384 135. glaucoptera (Buil.) 385 136. semifasciata semifasci- - ata (Hamps.) ....... 386 137. (aquilas(Boisds), ates ..% 386 138.) sylvia (Boisd.) 2240.6. 387 139 a. corythus corythus GUase erctve ceo 5 387 6. corythus luteata (Butl.) 140. hemichroma (Buil.) ... 390 141. passalus rectifascia (TAGIEIS) sin Pee aaa oe 391 1425 taro (Oramay ye sc... 392 143. mitchelli imperator CEU RE SA ea tala 393 Gen. Rhopalopsyche Buil. 393 144 a. nycteris nycteris (S260) | a eee et ead an 394 b. nycteris bifasciata Shi eaes Ae ae he aS 396 Subfam. Co@rocampinz Butl. 397 Gen. Celerio Oken ......... 400 145a. euphorbize robertsi (EIT | terion Cae eka 402 6. euphorbie nervosa Roths. & Jord. ...... 404 b XVill Gen. Celerio (cont.). Page 146. gallii gallii (Rott.) ..... 407 147. nicea lathyrus (Walk.). 408 148. lineata livornica (Hsp.). 408 Gen. Pergesa Walk. ........ 409 149 a. elpenor rivularis CE OSEAN ie wie etl cuae 410 6b. elpenor macromera (Gril) ein asieee 4]1 Gen. Hippotion Hiibn....... 413 150... velox (Habre se 415 151. celerio (Linn:) ..:.<... 417 152. echeclus (Botsd.) ..... 420 153. rafitesi (Guill) see ae 422 154. boerhavie (Fabr.)..... 424 Gen. Theretra Habn. ..°.... 427 155. nessus (Drury) ....... 430 156. boisduvali (Bugn.) .... 433 157. clotho clotho (Drury) .. 434 158. gnoma (fabr.) 159. latreillei lucasi (Walk.). 438 | 160. alecto alecto (Linn.) .. 440 1Gl. smamnsonr Clark 3... .. 443 162. suffusa (Walk.) ....... 444 163. lycetus (Cram.)....... 445 164. oldenlandiz oldenlandize CB GORV USES RAB See hsie 448 165. pinastrina pinastrina (Mart.) SYSTEMATIC INDEX. Page 166. insignis insignis (Buil.). 454 167. griseomarginata (Hamps.) Ae eae 454 168. pallicosta (Walk.)..... 455 169. castanea (Moore)...... 459 Gen. Rhyncholaba Roths. & Jord. 3: See 461 170. acteus (Cram.)......-: 462 Gen. Rhagastis Roths. & Jord. 465 171. velata (Watk:) 3222225 467 172. acuta (Walk) J2oeeee 470 173. aurifera aurifera (Buil.) 471 174. confusa Roths. & Jord.. 473 175 a. lunata_ sikhimensis Roths. & Jord. ...... 474 b. lunata lunata (Roths.) 475 176. olivacea (Moore) ...... 475 177. gloriosa (Buil.) ....... 478 178. albomarginatus albo- marginatus (Roths.) .. 478 Gen. Cechenena Roths. & SOEs Witte 481 179. mirabilis (Buil.)....... 483 180. segrota (Buil.) -2522:2: 485 181. helops helops (Walk.) . 486 182. minor minor (Buwil.) ... 487 183 a. lineosa scotti Ri ths... 489 6. lineosa lineosa (Walk.) 490 Family SPHINGID-A. I.—GENERAL INTRODUCTION. Morus of the family SpHINGID@, or Hawk-Moths, can, with few exceptions, be distinguished from other lepidopterous insects by their general appearance. The long, narrow, pointed fore wing, the short, triangular hind wing, the large eye, powerful thorax and sharply-pointed abdomen, the grace- ful, high-bred appearance of the whole creature, can hardly be mistaken. These characters are very constant throughout the family ; although the end of the abdomen appears in some species, such as the Humming-bird Hawk-Moths (Macroglossum), to be broad instead of pointed, this broadening is only apparent, being caused by the expansion of lateral scales. When these have been removed the abdomen is found to be pointed as in other species. The venation of the wings, the smooth and appressed hair or scales of the body, and other external characters are also very constant. The antenna is filiform or setiform in many species, but is more or less strongly clubbed in others. There are usually spines on the abdomen and the tibie. The moths vary greatly in size, the largest being equal in bulk, but not in expanse of wing, to the largest moths, and the smallest with an expanse of only 20 mm. from wing-tip to wing-tip. The length of the tongue also varies greatly in different species, from being the longest tongue found in any insect (Cocytius, 250 mm.) to two tubercles barely 2 mm. in length. In the Indian HAerse convolvuli it may reach a length of 130 mm., or nearly 5 inches, while in other species it is short and functionless. The labial palpi are usually large, but, like the tongue and other organs, may be much reduced. There is, throughout the family, a tendency to the modification and reduction of many organs. The colour of the moths is usually sober and cryptic, resembling the bark of trees and other natural objects, but some of the species are brightly coloured, and most have pleasing tones and markings. Some species mimic very closely bees and other insects, but there does not appear to be any Hawk-Moth which mimics another Hawk- Moth. The colouring of the male and female moths is usually the same, but the sexes are differently coloured in some species. VOL. V. B SEP 14 1937 No SPHINGIDA. The eggs are either nearly spherical or more or less oval in shape. The surface is usually smooth and shining, and the colour some shade of green or yellow. The larve, when full-fed, are nearly cylindrical in some subfamilies, but taper more or less strongly towards the head in others, the head being either rounded or triangular. There is always a horn on the twelfth segment, straight and bifid in the first instar, of various shapes and more or less strongly chitinized in the later instars, while it may be reduced to a short tubercle or a knob in a few species (e. g., Langia zenzeroides). The surface of the larva is usually naked, with only a few scattered hairs. Some species have tubercles and, rarely, fleshy spines. The colour of a great many is green with pale-coloured oblique stripes ; the colour-scheme, combined with the position m which they le, causes them to resemble closely a leaf with its side-veins. Others are brown, yellow, or variegated, and many species have large eye-like spots, or ocelli, which give them somewhat the appearance of a snake’s head. The pupe are usually short, cigar-shaped, rounded in front and pointed behind. In the species which pupate under- ground the colour is usually brown without any markings, while in those which pupate on or near the surface the colouring is cryptic, pale with darker streaks or dots. Rothschild and Jordan state, in the Introduction to the ‘Revision’ (1903), that they were very much hampered in their attempts to classify the Hawk-Moths by a lack of ‘material regarding the early stages. Only a few of the more common species had been bred, and the published descriptions of these were most inadequate. This lack of material has been remedied to some extent by the publication of Mell’s work on the Hawk-Moths of 8. China. We have now bred 105 species and subspecies of the Indian Hawk-Moths belonging to 40 genera; 14 of these species were new to science at the time, and we give descriptions of the early stages and habits in this volume. Mell has described 12 species which extend from S. China to India, but which we have not bred ; incomplete descriptions of 15 more species are available from other sources. We therefore know something of the early stages of about 66 per cent. of the species and 85 per cent. of the genera which occur in India. The early stages of 69 species and of 8 genera are still quite unknown, and it is very desirable that as many as possible of these should be discovered and properly de- scribed and figured, or preserved in spirit, so that they may be properly described by others. Hampson and other authors, except Mell, confined their descriptions of the caterpillars to the colour, which is individually very variable, and to the shape of the head and horn, characters too constant throughout GENERAL INTRODUCTION. 3 whole subfamilies to be of much use in identification or classification ; they scarcely mentioned the pupz and habits. The Hawk-Moths are essentially a tropical family, the number of species existing in the temperate regions being comparatively small. Very few species extend into the Arctic Regions, and then only as occasional stragglers. Only seventeen species, some of them rare stragglers, occur in the British Isles. With the exception of the Arctic and Antarctic Regions the family is found throughout the world. The Hawk-Motks were first classified by Linnzus in the year 1758, under the name “ Sphinx,” this name having been first used by Reaumur in 1736 for the English Privet Hawk-Moth, on account of the Sphinx-like attitude adopted by the larva when it is alarmed. Linnzeus included in his genus Sphinx forms other than the true SPHINGID&, and Samouelle adopted the name SPHINGID for the family in 1819. The SpHINGIDz of the world and of separate regions have been classified, revised or catalogued by many other authors. The Hawk-Moths of India were first classified by Hampson in Blanford, ‘ Fauna of British India—Moths,’ vol. i (1892). In 1904, after the publication of Rothschild and Jordan’s ‘Revision ’ of the family, he published papers supplementary to the ‘ Fauna of British India ’ volumes, in which he adopted Rothschild and Jordan’s classification and applied it to the Hawk-Moths of India (‘Journal of the Bombay Natural History Society,’ vol. xv, p. 630 (1904). Hampson’s original classification of the family, like that of other authors whose works were published before the ‘ Revision,’ was _ based chiefly on external and easily visible characters. Rothschild and Jordan found that these characters alone could not be relied on for purposes of classification, as some (colour for instance) were so variable individually, and others (shape and venation of wings) so constant among nearly all sphingid forms. Their classification was based on characters revealed by a minute examination of the structure, both external and internal, of practically every Hawk-Moth known to exist when the ‘ Revision’ was published, and has been accepted by all later authors. In 1903, when the ‘ Revision’ was published, there were 722 species of Hawk-Moths known throughout the world. In 1911 some 850 species were known, and the number has now risen to over 1,000. Of this total about 250 species, or one quarter, occur in the Oriental Region. Hampson (1892) recorded 121 species from India and Ceylon, and in 1904 the number of known species had risen to 163 (J. Bombay Nat. Hist. Soc. xv, 1904, p. 630). Omitting two of Hampson’s species, Rethera kamarovt and Celerio zygophyll1, which occur in Afghanistan but have not been B2 4 SPHINGID A. found in India, the number of species and subspecies now known is 204. The Hawk-Moth fauna of India is therefore very rich. , We have followed, with slight modification, the system of classification, the nomenclature, and the general arrange- ment of the ‘ Revision.’ In describing the larva and pupa we number the segments from 1 to 14, segment 1 being the head and segment 2 that segment of the body’ lying immediately behind the head. The length of the larva is measured from the front of the head when held in the normal position for the species to the end of the anal claspers. The hairs which are always present on the head and body are not mentioned unless they present some peculiarity. The lengths of the various parts are measured along the dorsal line, the breadth at right angles to it. In the pupa measurements are taken from the most frontal part of the head; the length is the distance from this point to the tip of the cremaster. The measurements of the tongue, fore and mid-leg, and antennal case are made with reference to the distance from the front of the pupa to the tip of the wing-case ; so that the statement that “ the antenna is equal to the fore leg, which reaches to the middle of the wing-case,’’ means that the tip of the fore leg case and that of the antennal case both reach to half the distance from the front of the head to the tip of the wing-case—the word “case ’ being omitted except when referring to the wing-case. The wings being folded up, their true length and their tips cannot be seen. The descriptions and keys refer to the full-fed larva unless any earlier instar is mentioned. The imaginal characters of the family SPHINGIDZ and of its subfamilies, tribes, and genera have been taken, with slight modification in some cases, from the * Revision’ and. from later volumes of “‘ Novitates Zoologice,’ and the imaginal characters of the species and subspecies have in most cases been taken from the same sources, rearranged and, where necessary, supplemented by the descriptions given in the ‘Fauna of British India—Moths,’ vols. i and iv, and other works. In describing the imago we have therefore used the same system of numbering the segments as that used in the ‘ Revision ’—that is, the head, three thoracic segments and the abdominal seements—and we use the same system as that used in the ‘ Revision’ in describing the venation. This system, and also that used in the ‘ Fauna of British India,’ is shown in fig. 8. Descriptions refer to the upperside except where the underside is specifically mentioned, and to both sexes unless. the sex is specified. GENERAL INTRODUCTION. 5 In accordance with the latest practice we have used the termination “INI” instead of “‘1cm” for the names of tribes. The food-plants of the larve have been given, so far as they are known, but in the case of the more common species some have been omitted for want of space. IJ.—_MORPHOLOGY. The Egg. Sphingid eggs are either nearly spherical or more or less oval in shape; they are broader than high. When oval the egg lies with the longer axis parallel with the surface to which it is attached. They vary from about | to 3 mm. in length. The size is not always proportionate to the size of the moth which lays it, the egg of the Convolvulus Hawk- Moth, for example, being about the same size as those of most of the Humming-bird Hawk-Moths, or about 1 mm. in length, though the Convolvulus Hawk-Moth is nearly as large as the Death’s-head Hawk-Moths, whose. eggs are about 2 mm. in leneth. The surface is smooth, usually shining, though sometimes dull to the naked eye, but under the microscope slight shagreening or an indication of reticulation can be seen in those of some species. When first laid the colour is often whitish ; it may remain so, but usually turns some shade of yellow or green, more rarely brown. ‘There are seldom any markings, though reddish bands and patches may appear in those of the genera Marumba, Degmaptera and Panacra ; such markings are possibly due to coloured parts of the developing larva showing through the shell. Mell (1922) discusses very fully the morphology of the sphingid ege and the number and proportion of embryos which reach maturity. Apparently Herse has the greatest number of embryos, all of which may reach maturity. Oxyambulyx ocellata Jays only a small number of comparatively large egos, from which, according to Mell, 88 may develop out of a possible 371. He obtained’ a maximum of 132 embryos out of 282 eggs of Clanis bilineata in China, but in many cases in 8. India specimens of that moth laid more than 280 eggs *. When the larva is about to emerge the head lies near one end of the egg, and the body stretches back to the other end, * Mell considers the number cf eggs laid by Herse convolvuli as ‘“oigantic,” and also mentions as something extraordinary the 1,100 eggs laid by a Ghost-Moth (Heprarip™). We once removed the abdomen of an apparently defunct female Ghost-Moth (Phassus malabaricus), and the detached abdomen laid 11,500 minute black eggs ! 6 SPHINGID 2. then bends sharply forwards again, venter against venter, to the head and under it, the horn continuing in the line of the body, but often not visible from the outside. The shell is thick and opaque in some species, translucent in others, and as the larva develops it is absorbed to some extent from the inside, till, just before the emergence of the caterpillar, it is nearly transparent. The axial lme of the developing caterpillar becomes visible first ; the eyes appear as minute dots before the outline of the head is visible, and they become darker and finally black; the dark tips of the mandibles are the first part to show any movement, opening and closing for quite a long time before they commence to bite their way through the shell. Larva (fig. 1). The sphingid larva is subcylindrical in shape when full-fed, tapering forwards slightly from segment 7 or sharply from 5. In the former case the head is usually large and in the latter small. These are, broadly speaking, the two types of larva, the former being common among the Asemanophore and the latter among the Semanophore. There are three pairs of true legs, one pair on each of segments 2, 3 and 4, a pair of prolegs on each of segments 7 to 10, and a pair of claspers on 14. There is a spiracle on each side of 2 and on 5 to 12. Whilst so far agreeing with many other lJepidopterous larve, it is further characterized by the possession of a horn on the dorsum of segment 12. This horn is chitinized in the last instar and is present in all Indian species, though some- times much reduced. A horn similar to that of the SpHrn- GID is found in a few species of NoTODONTID. The head (fig. 2) consists of two lobes, together with the mouth-parts, and varies considerably in shape, not only generically and specifically but individually in different instars. It is invariably rounded in the first instar, and may remain so until the last mstar, but In some genera it becomes triangular in the second instar ; it may remain triangular or become rounded again in the last instar. The vertex is some- times more or less conical, and a long process may arise from the apex of each lobe of the head in the second instar ; these processes usually become shorter in proportion to the length of the head in succeeding instars, and are often represented in the final instar by a tubercle at the apex of each lobe. The two processes are closely appressed till near the tips, where they diverge shortly. The triangular shaped head characterizes the subfamily AMBULICIN&, and is never found in the tribe Acherontini (though appearing again in the genus Dolbina of the tribe Sphingulini), nor in the subfamilies CHHROCAMPINA and PHILAMPELIN2; but it occurs again GENERAL INTRODUCTION. ig in the genus Sataspes of the tribe Sesiini, subfamily SESIINz,. connecting the Asemanophore, Ambulicine insects, with the Semanophorx, Philampeline forms. The highest region of the head is called the vertex ; behind the vertex is the occiput, with a small triangular sinus situated dorsally on the hinder margin and called the occipital sinus ; the front of the head is called the face, the side the cheek, the underside of the head behind the mouth-parts the gula or throat. Big A. A sphingid larva. The segments are numbered 1 to 14. a, antenna; b, true legs; c, prolegs; d, claspers; e, horn; f, spiracle; g, secondary rings ; h, anal flaps. . Segment 8, enlarged. f, spiracle; 7, the four main hairs. Spiracle, enlarged, showing central slit. . One of the true legs. a, base; }, first segment; c, second segment ; d, third segment, bearing a claw. . Proleg. a, base; b, shank, with terminal fringe of hair; c, ankle ; d, foot, with a fringe of hooklets. He FOn The true clypeus (fig. 2 A, b) is roughly triangular, but the apex may be rounded; the sides may be straight or curved inwards or outwards; the base is generally somewhat emar- ginate and the basal angles are frequently occupied by low 8 SPHINGIDA. rounded tumidities. The length of the clypeus is rarely more than half that of the head, and often considerably less than half. The false clypeus (fig. 2 A, d) is a narrow strip A Fig. 2. A. A diagrammatic representation of the larval head, seen from the front. a, lobes; 6, true clypeus; c, basal tumidities; d, false elypeus; e, labrum, with lateral bristles; jf, ligula; g, man- dibles ; 7, antenna, with special bristles ; 7, eyes. B. Ligula and part of labrum of Polyptychus trilineatus sonanthis, seen from the front (greatly magnified). a, ligula; 6, sinus, showing bevelled edge: c. bristles; d, labrum, lower part; e, labial bristles. GENERAL INTRODUCTION. 9 lying outside the true clypeus; its apex may be acute or rounded, and the sides may extend downwards to the base of the true clypeus. The false clypeus is not present in the earlier instars, but is always present in the last instar, though sometimes it is difficult to trace, as in the genus Macroglossum. The labrum (fig. 2 A, e) is a transverse plate, the proximal part chitinized, the distal part generally membranous ; measured along the dorsal line it is between one-half to one-third the length of the true clypeus. The front or lower margin is straight and the hinder margin arched towards the base of the clypeus; the hinder margin is connected with the base of the clypeus by membrane which allows it a certain amount of movement. In many species the labrum is sculptured by longitudinal ridges, as shown in the figure, but in other species it is smooth; at each side there is a prominent bristle (fig. 2 B, e), which -is directed forwards, downwards and inwards. The ligula (figs. 2 A, f; 2B, a) lies below the labrum, and serves to direct the edge of the leaf to the cutting surfaces of the mandibles when the larva is feeding. It is apparently developed from the lining of the gullet, being rooted under the labrum and independent of it, as is shown both by dis- section and by the fact that it can be seen to be protruded and retracted when the larva is feeding. It is also capable of being altered in shape at will, so that it is either transversely convex or flat. It is very generally kidney-shaped. with a frontal triangular sinus which varies in depth and width in different species. In some of the Ambulicine larve (genus Oxyambulyx) the ligula is very long and the sinus very deep and narrow; the structure then resembles in shape two sausages joined at the base. The surface is smooth and set with stout bristles, the position of which in Polyptychus trilineatus sonanthis is shown in fig. 2B. We are unable to say if these bristles occupy the same position in all genera and species. The mandible (fig. 2 A, g) is a strong, truncated, hollow wedge, the base fitting into a socket in the skull near the base of the antenna. The basal half is curved gently away from the antenna, and the distal end is flattened, bevelled and more strongly curved in the same direction, so that the outer edges nearly face each other inwards and form ‘the cutting edges. In the first instar the cutting edges are definitely toothed, but the teeth become less prominent in later instars until, in the last instar, they have almost disappeared, and are often represented only by transverse grooves. The antenna (fig. 2 A, 2) is composed of three segments. Its base is set in a cavity of the skull into which the whole 10 SPHINGID®. organ can be more or less retracted. The basal segment is the thickest, and is usually equal in length to the third ; the second segment is shorter and is of less diameter than the basal segment; the third is the thinnest, and has at its tip two bristles of varying (in different species) and markedly unequal length. The colour is uniform in different colour-forms of one species. A Fig. 3.—Larval head. ~ A. Lower portion of the right lobe, seen from the front, showing position of eyes (greatly magnified). B. A diagrammatic view, seen from below. a, antenna; c, clypeus ; la, labrum; lM, ligula; lp, labial palpi; md, mandibles; mp, max- illary palpi; 0, eyes; s, spinneret. The eyes (figs. 2A, 7, 3 A) are situated partly on the cheeks and partly on the gula. They are six in number on each side, in a group just above the base of the antenna. They are simple, round and convex, and are often raised on a chitinous base. They are vitreous in appearance, often nearly GENERAL INTRODUCTION. iia colourless, with a black pupil, and the base often reddish- brown or black in colour. We call the uppermost eye no. 1, and number the rest downwards. Eye 2 is a little below and on the inner side of 1; 3 is below and slightly on the imner side of 2; 4 is well below and slightly on the outer side of 3; 5 is slightly below 4, and on the outer side of all the rest ; 6 is the lowest, and is situated on the under surface of the head near the base of the antenna, directed downwards. Eyes 1 to 4 lie on a curve convex to the base of the mandible, 3, 4 and 6 always more or less in a straight line, with 6 at a greater distance from 4 than the latter is from 3; 5 is always on the outer side of 4 and 6; 4, 5 and 6 often lie at the corners of an equilateral triangle. Within these limits the arrangement of the eyes varies in different species, and the size of the eyes also varies. The surface of the head may be dull or shining and polished ; it may be smooth or covered more or less closely with seti- ferous tubercles, and between these there may be superficial corrugations or reticulations; in addition to the bristles on the labrum, ligula and antenna already mentioned, there are main hairs arranged very similarly to those on the body segments—one on each side of the apex of the false clypeus, one on each side at its middle, and one on each side near the base of the true clypeus. The body in the first instar has the same shape in all subfamilies. It is always cylindrical, with a straight bifid horn, the diameter of the body and the length of the horn varying with reference to the length of the body in different species. In the full-fed larva the shape varies with the sub- family. In the ACHERONTUNZ# the body is nearly cylindrical, tapering only slightly from segment 7 to the large rounded head; in the AmBuLicina the body tapers more sharply from segment 5 to the large triangular head; the SEsimnm larve either resemble those of the AMBULICIN2 or have bodies nearly cylindrical; the larve of the subfamilies PHILAMPELIN2Z and CH@ROCAMPINZ taper sharply forwards from segment 5 to a small rounded head, and segments 4 and 5 are sometimes tumid and, more rarely, laterally flanged. Except for the horn, important processes of any kind are rare in Hawk-Moth larve ; in the genus Polyptychus the newly- hatched larva has a dorsal prominence on segment 14, but it disappears in the later instars; Pseudodolbina fo has long fleshy tubercles on segments 3, 4 and 14; in a larva recently discovered by us (Apocalypsis velox) some of the tubercles are developed into long fleshy spines ; Meganoton nyctiphanes has a fleshy hump on 3. The proportional lengths of the 1 SPHINGID. segments are very similar to those of other lepidopterous larve, the divisions between them being always visible, though more clearly defined in some species than in others. Each segment from 2 to 12 has more or less deeply depressed transverse limes which divide the segment into secondary rings (fig. 1, g); on segment 2 there are three rings, on segment 3 there are usually six, on segment 4 seven, and on segments 5 to 11 eight each; on segment 12 there are three or four complete rings near the front margin and others which do not reach the dorsum. In the first instar the horn is always straight, tapering gently from a truncate-conical base to a shortly bifid tip, each arm of the fork bearing a seta. It is usually long, and may be very long. In succeeding instars it becomes proportionally shorter, except in a few species, and often changes in shape. In the full-fed larva it may be straight, or curved downwards or upwards, or curved. first downwards and then upwards, as in the genus Acherontia (but not in Psilegramma, as stated by Rothschild & Jordan, 1903, p. 42). It may be long or short, stout or thin ; in Langia zenzeroides and in some of the species of Clanis it is reduced to a large conical tubercle, in the genus Hlibia to a rounded tubercle; in Khodosoma and Degmaptera it is flattened laterally, and in Rhagastis and Cechenena it may be flattened laterally or clubbed. It usually ends in a simple point, but may remain bifid to maturity. The surface of the body may be dull or shining, with or without tubercles on some or all of the segments, or with tubercles in some instars only. In the first instar the surface of the horn is either smooth and shining or dull, and in some species it is covered with small sete which may arise from granules. In the last instar the surface of the horn may be rough, with pointed or rounded setiferous tubercles, — or smooth as in most of the genus Macroglossum and the subfamily CHa@ROCAMPINa ; it is usually strongly chitinized and shining, but may be dull; the bifid tip usually disappears, being replaced by a simple point. In the flattened horn of Degmaptera mirabilis the setee become spinous on the under surface. The legs (fig. 1 D) consist of a base a and three segments 0, c and d, the last of which bears a simple claw. The proleg (fig. 1 E) consists of a base a, shank 6, ankle c, and the foot d, set with curved hooklets. The proleg of segment 10 is invariably larger than that of segment 9, and the latter is sometimes larger than the co-equal pairs of segments 7 and 8. The clasper is similar in construction to the proleg, but the shank is larger and more conical. The upper edges of the claspers are covered by the triangular anal flap. GENERAL INTRODUCTION. 13 The horn appears to have developed from a pair of tubercles on segment 12. The tubercles are most strongly developed in the subfamilies ACHERONTIIN2 and AMBULICIN2%, and are largest on the dorso-lateral line and on the oblique stripes : they are least in evidence in the subfamily CH@ROCAMPINA, and their presence or absence is of considerable phylogenetic importance. They are often present on segment 2 and on the anal flap and claspers when entirely wanting elsewhere. They are frequently wanting in the newly-hatched larva, but may develop in later instars; more rarely the reverse occurs. They develop rarely into long fleshy spines or processes, as mentioned above. The spiracles are always oval in shape, with a central longi- tudinal slit and often a raised chitinous rim. Those on segments 2 and 12 are usually larger than the rest, and that on 12 is placed obliquely, the long axis roughly on a line drawn from the base of the horn to the lower edge of the front margin of the segment. The remaining spiracles have the long axis at right angles to the dorsal line of the larva. Chetotaxy—We have, unfortunately, not studied the hairs of the newly-hatched larve of many of the species we have bred, but in the case of Clanis phalaris they are arranged in small groups ; in Cephonodes and Gurelca they are branched ; in Sataspes infernalis they are bifid; in Rhopalopsyche nycteris they are simple; the compound hairs in each case becoming simple in the second cr third instar. The secondary hairs are sometimes wanting. The body of the larva in the last instar is covered with minute hairs, amongst which, on each segment, there are some much larger ones which we call the main hairs. The minute secondary hairs are arranged, usually in a single row, along the secondary rings. On segment 2 there are two transverse rows of main hairs, one row close to the front margin and another row at about the middle of the segment; on segments 3 and 4 there is one row at about the middle of the segment; each row is composed of four hairs on each side of the dorsal line—one subdorsal, one dorso-lateral, one supra- and one subspiracular. On the remaining segments the hairs are not in a row; on 5 to 11 the subdorsal hairs are in front, usually on the third secondary ring, the dorso-lateral hair being behind on the sixth ring; the supraspiracular hair is just above and in front of the spiracle, and the subspiracular below the spiracle : the two upper pairs of hairs form what is called the dorsal trapeze. On 12 the dorsal trapeze is arranged differently, the dorso-lateral pair of hairs being in front and the subdorsals behind ; on 14 the trapeze is similar to that on 12 except that the subdorsals are at the very end of the anal flap; iL: SPHINGID 2. the hairs which on this segment take the place of the supra- and subspiracular hairs on the, other segments are placed one behind the other, also on the edge of the anal flap, in front of the subdorsals. The distance between the subdorsal and the dorso-lateral pairs of hairs is about one-third the length of the segment, and the subdorsals are about the same distance apart from each other; the subdorsals are the shortest and are hardly traceable among the secondary hairs ; the dorso-laterals and supraspiraculars are about equal in length and slightly longer than the subdorsals; the sub- spiraculars are the longest, and may in the last instar be about twice the length of the spiracle. The main hairs are usually simple (the secondary ones are always so), but there are species in which the subspiracular is compound ; it is pal- mately branched in a horizontal plane in Psilogramma and Cizara, as well as in a few species of other genera ; it is always very fine and delicately inserted, and appears to be always moving from the base; it is very easily broken off or removed by slight friction ; when simple it is usually erect, but may be decumbent. In addition to the hairs described above there are always some ventral main hairs, and, in the adult larva, there are some hairs on the segments of the true legs and a fringe of eight to twelve comparatively stout hairs starting from the iower margin of the shank and subtending the ankles of the prolegs, three more on the base of the proleg and some along the hind margin of the clasper face. The fringe at the end of the shank is never developed in the newly-hatched larva, in which there is a single hair. Each hair arises from a circular spot, which often develops into a granule and in many genera into a rounded or conical tubercle of a horny or fleshy consistency, which may be simple or multiple. : Coloration.—Commonly some shade of green or bluish-green, with various markings, but is sometimes brown, black, yellow, reddish or variegated. Colour dimorphism is not rare, and in some species there is trimorphism or even polymor- phism. The head may be immaculate or dotted and longi- tudinally striped. The markings of the body take the form of longitudinal and oblique stripes, patches and bands with dots or transverse lines. Eye-ltke markings (ocellz) are common in some of the subfamilies, and in a few species the spiracle on segment 5 is surrounded by an ocellus-like spot. The horn is usually black in the first instar, but may be yellow, green or parti-coloured (Rkhyncholaba). In later instars the colour is very varied. The colour and markings of the larvee are characteristic of whole genera and even of some of the subfamilies. GENERAL INTRODUCTION. Ld Pupa (fig. 4). We number the segments of the pupa from 1 to 14 as in the larva, segment 1 being the head and segment 14 the anal or cremastral segment. Segment 2 is the prothorax, 3 the mesothorax and 4 the metathorax, the three together forming the thorax ; segments 5 to 14 form the abdomen. Head.—The front is called the frons, the lower part the Fig. 4.—Pupa. A. Ventral view, showing a, appendage to clypeus, consisting of labrum and ligula; b, coxal piece; c, fore leg; d, antenna; e, mid-leg ; f, wing-case: g, bevels of movable segments ; h, cremaster. B. Lateral view, showing segments 1 to 14. a, frons and clypeus region of head; 6, eye-crescent; c, eye;-d, fore leg; e, antenna ; f, mid-leg; g, wing-case ; h, spiracle ; 7, cremaster. €. Female pupa of Acherontia lachesis, showing segments 12, 13, 14. a, clasper scar; b, cremaster. D. Male pupa of Acherontia lachesis, showmg segments 12, i3. a, clasper scar. 16 SPHINGID. clypeus. The larval clypeus is altogether wanting in the pupa; the name has, however, been retained and applied by most authors to the lower portion of the frons. In the process of the transformation of the larva into the pupa the head appears to undergo a bilateral screwing motion which obliterates the larval clypeus and raises the eye-region and antenne towards the vertex of the pupal (and imaginal) head. Attached to the pupal clypeus is a composite piece which appears to correspond with the labrum and ligula of the larva. The top of the head adjoming segment 2 is called the vertex ; on each side of the head is a large eye-case; the base of the tongue-case is next to the composite appendage of the clypeus ; its tip sometimes does not reach beyond the anterior junction of the wing-cases, as in many AMBULICINA, but more commonly continues as a narrow ridge entirely separating the wing-cases. In those species of which the imago has a very long tongue some special arrangement has to be made for its storage in the pupal case. This is provided by the extension of the head of the pupa into a laterally flattened hollow sheath, as in Cechenena lineosa, or by the development of a free tongue-sheath in the form of a tube projecting from the front of the head and bending backwards towards the venter, the closed end being bulbous to allow of the tongue turning back without too sharp a bend. The free sheath is of varying length in different species, and the distal portion may be curved into a semicircle or a spiral. This free sheath is found in some ot the genera of the tribes Acherontiini and Sphingini of the subfamily ACHERONTIINS, and again in Rhyncholaba acteus of the subfamily CH@ROCAMPINZ. oe Thorax.—The case of the fore leg lies nearly parallel with. the tongue-case (fig. 4); in some species there is a small, bipartite, diamond-shaped piece between the tibia and the tongue-case; Mell calls it the “shin,” but it is doubtful what it represents. We have reason to suppose it to be the coxal trochanter of the fore leg, and we call it the “~ coxal piece.’ In many species it is hidden under the femur and tibia of the fore leg. The mid-leg (fig. 4 A, e) lies immediately outside the fore leg and the antenna outside the mid-leg, with its base near the top of the head next to the eye-case. The hind leg is hidden beneath the wing-case, but the extreme tip may in some instances just appear between the apices of the wing-cases, which always end at or near the hind margin of segment 8; the fore wing lies over the hind wing and covers all but a narrow band of it which is visible along segments 4 to 7. Segments 9, 10 and 11 of the abdomen GENERAL INTRODUCTION. 17 are usually movable, and the margins are more or less deeply bevelled. Markings—The surface of the pupa is dull or polished and shining, and either smooth or pitted, wrinkled or corru- gated. Sculpturing is often present on the head, segment 4, the antespiracular region of 9, 10 and 11, the ventral surface of 13 and 14, and the cremaster. Head-sculpturing is found in the genera Marumba, Parum and Sataspes in the form of a pair of frontal, very rugose ridges ; sculpturing on segment 4 in Acherontia and Meganoton as subdorsal pear-shaped, raised and roughened surfaces, and in Psilogramma and Herse as a pair of subdorsal weals or ridges. The ante- spiracular sculpturing consists of a specifically varying number of roughly parallel ridges on the front bevels of the movable segments 9, 10 and 11; this is found in most of the Acheron- tiine pupe and in Compsogene, Clanis, Marumba and other Ambulicine forms, with slight indications of it in some of the CH@ROCAMPIN2E. The sculpturing on segments 13 and 14 consists of the clasper scars and sex marks, and the cremaster is often corrugated, pitted or ridged. The sex of a pupa can, in most cases, be determined by examining the ventral aspect of segments 13 and 14. The-scars of the anal claspers (fig. 4 C, D) are on segment 14 in front of the base of the cremaster, the scar on each side of a central depression forming a somewhat sausage-shaped thickened “lip,” the two scars together forming a longitudinally-placed, mouth-shaped organ. In the 3 pupa (fig. 4 D) there is on 13 a similar but smaller mark and no mark on 12, and the margins of 12, 13 and 14 are not emarginate. In the 2 pupa (fig. 4 C) these margins are distorted by the middle anterior portion of segment 14 being produced in a tongue-like or triangular process right across 13 and forwards to the middle of 12, carrying, so to speak, the middle part of 13 with it, as shown in the figure. The large “mouth” in the figure is formed by the clasper scars ; the black median line on 13 is the posterior of two marks which always occur in 9 pup; it probably represents the anus of the imago. This mark may be shorter or it may be broader than in the figure, or may even be more or less circular, and it is often connected by a depressed line with the generally better defined and usually broader and rounder mark in the middle of 12, in the very apex of the tongue or triangle; in some cases this latter, presumably sexual, depression is also mouth-shaped. Sometimes these “ sex ”’ marks are faint and difficult to distinguish, but a single mark on segment 13 and no mark on 12 always indicates a J pupa ; a blurring of the segment margins and a mark in the middle of segment 12 always indicates a Q. VOL. V. 9) 18 ; SPHINGID. Spiracles (fig. 4 B, h).—These are situated on segments 2 and 5 to 12, as in the larva, though those on 5 are generally hidden beneath the visible inner margin of the hind wing ; those on 2 are placed in the division between segments 2 and 3 and are often covered by a lobe extending from the front margin of 3 which fits into a corresponding emargination of the hind margin of 2. They are oval in shape as in the larva. The cremaster is a chitinous extension from the dorsal surface of segment 14; it may be broadly or narrowly conical, or flattened, long or short, with the tip usually bifid ; in some species there are additional lateral teeth or spines, and these, like the terminal ones, may be branched or end in hooks, or may be simple. In Langia the cremaster is reduced to a minute spine or is absent. Coloration.—In those species which pupate in a cell under- cround the colour is either black, brown or chestnut, without any markings, except in Degmaptera mirabilis, which is chest- nut with the eye-cases cream-coloured ; the surface of these pupe is usually highly chitinized and shining. The pupa of Psilogramma menephron is exceptional in being covered with a plum-like bloom. The pupez of those species which pupate in a cocoon on the surface are usually pale in colour with darker stripes and spots, or parti-coloured, but the pup of the subfamily Sxsun# are of the uniform dark type, though they pupate on the surface. In the subfamily PHILAMPELINZ the pupa of Cizara sculpta is prettily marked and those of the genus Panacra are variegated, resembling green and erey lichens ; that of Angonyx testacea is an exception in being nearly uniform black in colour. Pupz of the subfamily CHEROCAMPINA are usually pale with darker stripes and dots. Imago * (figs. 5, 6, 7, 8). Head (fig. 5).—The dorsal skeleton is divided by two transverse sutures into the clypeus (cl), epicranium (ecr), and occiput (occ). The epicranium forms laterally the sockets for the antenne, which stand nearer the eye in some Hawk-Moths than in others. The clypeus is the largest plate of the three ; it is more or less strongly convex, especially mesially. It bears at the anterior margin the labrum (Ir). The labrum is in most instances raised to a large, transverse, cariniform tubercle, which is generally vertical in front. It projects sometimes frontad over the base of the tongue, concealing the Bn pam of the epistome (ep). 7 ‘This account is taken ae Fn bail he jouer Paes ision,” 1903. GENERAL INTRODUCTION. 19 The epistome covers the base of the tongue. When normal it has a thin mesial lobe and a large process at each side. The lateral processes are designated “ pilifer’’ by Kellogg. The normal pilifer (p) is a curved obtuse process, concave and flattened on the inner side, and is beset on the inner surface with a great number of long stiff bristles which project over the base of the tongue, which they touch. ee a Fig. 5.—Morphology of head. A. Head of Psilogramma menephron, denuded, dorsal view.