COSYMBIA. By L. B. Prout. 141 



Notwithstanding its very different appearance, it shows but little structural deviation from Cosymhia, the chief 

 differences being in the cf antenna and in the somewhat stouter, more thickly-scaled palpus. Whether this 

 close structural resemblance is entirely due to close affinity, or in part to convergence, can only be decided 

 by the discovery of the early stages or of some important characters which have hitherto escaped the 

 observation of systematists. 



C. humifusaria Ev. (4a). White, glossy, the veins broadly (the 2"'' radial more narrowly) marked a^^J-'j-^^ 

 with brown; costal margin of forewing densely dusted with brown; lines brown, the postmedian somewhat 

 the darkest; antemedian bent outwards on the median vein; median rather broader, somewhat curved, closely 

 following the large, roundish cell-mark; postmedian outcurved between the 3'''' radia,l and the 2"'^ median 

 closely followed by a brown band; distal area with the brown vein-markings still further thickened, sometimes 

 leaving between them only quite narrow white wedge shaped markings; fringe chequered with brown and 

 white. Hindwing without first line. Underside similar, less glossy, without first line. Occurs from SE. Russia 

 and Transcaucasia to the Hi district, also in S. Spain and N. Africa. Said by Eversmann to fly by day, end 

 of May and beginning of June; in Murcia Korb took it over Artemisia. Dr. Seitz denies the day flight; he 

 tells me the species is not rare near Batna and sits among the bushes of Artemisia herba alba, flying off 

 when disturbed, like other Geometridae. 



23. Genus: Co^yuibia Hhn. 



Palpus rather short to moderate, shortly rough-scaled, terminal joint distinct, slender, smooth, cf 

 anienna bipectinate with moderately long branches, apical part simple. Hindtibia in cT with 2, in ? with 4 

 spurs. Forewing with areole simple, P' subcostal separating from 2"'* to 4"' considerably after 5"^. Hindwing 

 with 2°'^ subcostal connate or sometimes very shortly stalked with P' radial, discocellulars not oblique, 1=' 

 median arising close to posterior angle of cell or not infrequently even very shortly stalked. 



Egg longish oval, the surface covered with an irregular network of raised ribs. Larva moderately 

 elongate, nearly cylindrical, very slightly thickened posteriorly, head rather large, slightly bifid, body without 

 marked protuberances; on various trees, not hibernating. Pupa exposed, attached to a leaf, with cremastral 

 pad and central girth and showing several other remarkable resemblances to a Papilio pupa. 



An extremely natural genus in its structure and entire life-history. Even in shape the species agree 

 well. They are partially double-brooded and several exhibit marked seasonal dimorphism; the winter is 

 always passed in the pupal stage. The moths rest by day on tree-trunks, or more commonly among leaves 

 and are in general easily obtained in the day-time by searching, or by beating the bushes or trees in which 

 they hide. At dusk they are in general less observed than many other Geometrids, but later at night they 

 may sometimes be captured at sugar or attracted to a strong light. By far the best way of securing a 

 number, however, is by breeding. The ?? lay very willingly in captivity and the larvae are easy to feed up. 

 Moreover on account of their tree-frequenting habit they are unlike those of the preceding genera quite easily 

 oblained by beating. 



A very interesting and well established fact regarding the earlier stages is that the colour-dimorphism 

 which nearly all the known species of the larvae exhibit (green or brown) is always carried through to the 

 pupal stage. Not less interesting is the remarkable convergence, in the pupal structure and method of 

 suspension, towards certain other exposed pupae particularly of butterflies and Elachistids. So far as I know 

 this phenomenon has only been observed in two other cases among the rest of the Geometrid subfamilies — 

 the North American Euphanessa Pack (subfam. Larentiinae) and the Australian Terpna percomptaria Guen. 

 (subfam. Hemitheinae) — but in the Cosi/mhia-grow^ it seems pretty constant (Anisodes, Pleuroprucha, etc.). 

 It is said, however, that annulata often pupates among moss. 



The perfect insects also invite careful study, not only on account of their beauty and general 

 variability, but also on account of their adaptability to scientific investigations of the effects of temperature 

 and in hybridization. The accounts of the separate species must be consulted for detailed information on 

 these matters. 



The genus is not extensive, and is' almost, if not entirely, confined to the Palearctic Region (chiefly 

 its western part), Africa (2 or 3 species) and America (a few species). In the great Indo-Australian Region 

 and to a very large extent in South America it is supplanted by some specialized offshoots — Anisodes and 

 its allies. "Ceylon", given by Walker as the locality for his vusannana, is evidently a mistake, as already 

 pointed out by Hampson; his type-specimen is a typical pendularia CI. and this is, in its range, the most 

 northern species of Cosymhia. The genus has been called by many names {Cyclophora, Codonia, Leucoph- 

 thalmia, Ephyra, Zonosoma, etc.) but Cosymhia, as used by Warren, Dyar and Swinhoe, is clearly correct; 

 this was erected by Htjbner for nolaria Hhn. (= pupillaria Hhn. ab.) 



The genitalia of the European species have been carefully investigated by Bastelberger and indicate two 

 distinct groups: []) pendularia, orbicularia, alhiocellaria, annulata, puppillaria, qicercimontaria and ruficiliaria, with 

 forceps wanting or rudimentary: {2)porata, punctaria, linearia and supjnmctaria, with forceps powerfully developed. 



