2 SPHINGIDiE. 



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 larvae, 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 ins tar, 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, 

 Avith 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 in which they he, 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 pupae 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 S. 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 



