DIMORPHIDI. 233 



to be remarkably near the Sphinges." As to the latter point, he 

 considers (loc. cit.) that the Sphingidae probably sprung from a 

 form like Endromis {Dimorpha) rather than Aglia, and states that 

 taxonomically Aglia is by no means so closely allied to the Sphin- 

 gidae as Poulton infers ; in its neuration, Endromis {Dimorpha) is much 

 nearer, and it is also a more generalised or synthetic form than Aglia. 

 Dealt with seriatim, the various stages of Dimorpha suggest to 

 us the following alliances : — Egg. — The egg, with its transparent shell, 

 has most in common with the Sphingid and Citheroniid eggs, 

 less in common with the opaque and porcelain-like eggs of the 

 Lachneids and Saturniids. It appears to differ widely from the 

 Sphingid egg, in that the development of the embryo darkens it, 

 but this is possibly of no very great importance. Larva. — The 

 newly-hatched larva is distinctly Lachneid in the arrangement and 

 character of its warts, viz., i larger than ii, iv and v many -haired, 

 but separate as in Bombyx (mori) and the more generalised Lachneids, 

 and not united as in the specialised Lachneids (anted, vol. ii., p. 439). 

 [In the larva of B. mori, iv and v are separate and subspiracular 

 (See also Grote, Die Satumiiden, figs. 3-4)]. When larger, too, 

 the larva resembles Bombyx {mori) in the development of 

 the caudal horn, in which character, perhaps, that of B. mori 

 approaches more closely to the Sphingid caudal horn. Buckler also 

 notes the head of the larva as being very small, with the thoracic 

 segments tapering to it, and retractile, as in Eumorpha (C/ioerocampa) 

 though to a less extent ; but it fails entirely in the peculiar development 

 of the prespiracular tubercle of the Sphingids (generally referred to 

 v*), whilst Bacot connects it more closely with the larvae of the 

 Lachneids, noting that, in the 1st stadium, the larval tubercles i 

 and iii bear 3 or 4 or more hairs in Dimorpha, while in the Sphingids 

 they are small, and single-haired. These become atrophied later, in the 

 Sphingids, whilst in Dimorpha, on the contrary, the change in the 

 tubercles is apparently similar to that which takes place in the 

 Lachneids, i.e., that at the 1st or 2nd moult, i and ii seem almost 

 (or quite) to disappear and to be replaced by scattered hairs • 

 but this is rather apparent than real, for, in Dimorpha, the hairs do 

 not really disappear, but are reduced in size and so are lost among the 

 secondary hairs that arise at this stage, whilst in the Lachneids they 

 do not even become smaller, but are almost inextricably confused 

 among the mass of secondary hairs developed f. Cocoon. — The 

 cocoon reminds one somewhat of those of certain Saturniids, and the 

 gummy substance that makes it waterproof, being very similar to 

 that used in the cocoons of Saturnia {sens, strict.), aids the 

 resemblance. Bacot also says that it resembles that of Aglia tau 

 considerably, and that of Caligula japonic a very markedly ; it 

 resembles somewhat the cocoon of Eumorpha elpenor which is 



* We suspect this to be not v but a tubeixle homologous with the "supple- 

 mentary prespiracular" of the Lachneids fantea, vol. ii., p. 439). 



t In the Sphingids one finds the single primitive tubercular hairs reduced in size 

 and lost among the secondary hairs (shagreen hairs). In Dimorpha, the duplicated 

 primitive tubercular hairs are reduced in size, and lost among the secondary hairs 

 (shagreen hairs). In the Lachneids we get the duplicated primitive tubercular hairs 

 obscured among a crowd of secondary hairs, only shagreen in character to the 

 extent that they have coloured bases, usually black. They are simply obscured ; 

 there is no reduction in size (Bacot in litt.). 



