DAPHNIS NERII. 257 



artistic finish. IV. Boisduval (Ic. Chenilles, pi. iii., fig. i) : (8) A very 

 diagrammatic figure, tolerably uniform smooth green, spiracles with 

 a yellow line round each. Eye-spots very large, single; shagreen- 

 spots shown ; subdorsal line white, shown as seven oval (subseg- 

 mental) marks on each segment. V. Horsfield and Moore {Cat. 

 Lep. Ins. Ind., i., pi. x., fig. 3). (9) Rather diagrammatic, head 

 and prothorax green, meso- and metathorax yellow, venter and 

 hinder part of body yellow, rest blue!!, with white subdorsal line 

 bordered with yellow above, shagreen-dots diagrammatically indicated, 

 eye-spot in line of subdorsal line. This division of the 1st thoracic 

 into two is shown in several figures, no other shows so great 

 a division of the 3rd into two portions as this. VI. Hubner (Larv. 

 Lep., Sphinges, pi. iii., fig. 1) : (10) Head green, thorax and 1st 

 abdominal yellow ; body green, 8 subsegments shown, horn 

 yellow; white subdorsal line, uniform width, but with incisions of 

 subsegmentation lined through it from horn to anterior border of 

 2nd abdominal when it stops abruptly ; some lilac tinting on green 

 ground-colour below the line ; some yellow on prolegs and round 

 spiracles, which are black ; shagreen-dots shown, some also, in the 

 line of the subdorsal line, on the 1st abdominal. Eye-spots very 

 dorsal, lower margin level with the lower margin of the subdorsal 

 line ; also very double, each portion of about equal size. 



Variation of larva. — We have already given notes (sipra) of the 

 various forms of the larvae figured. Swinton records that, at Jerusalem, 

 two forms of the larva were to be obtained — green and brown ; 

 Jones also obtained a larva of the brown form at Riva on July 

 31st, 1900. Treitschke and Kollar note (Die Schmett., x., p. 128) 

 that each of them had had a larva that was of a bronze colour, the 

 front segments rose-red, a dark longitudinal stripe running to the 

 anus ; the anus and legs dull rose-colour. 



Cocoon. — The larva forms a long silken covering on the ground 

 in which to pupate (Moore). Similarly to most of the larvae of this 

 group, this species does not enter the ground to pupate, but makes 

 a sort of cocoon with the debris of leaves, which it unites with some 

 strands of silk (Boisduval). The larva pupates on the earth in a 

 loose cocoon, mingled with particles of earth and dry leaves (Bartel). 

 Larvae spun loose cocoons under moss lying on a layer of sand in 

 their breeding-cage (Klooss). 



Pupa. — A large pupa, fully longer than Manduca atropos 

 but not quite so stout, 2*5 inches to 3 inches in length, this latter 

 measurement would represent a large specimen moderately extended, 

 •625m. broad. Eorm much that of a Phryxid or Eumorphid pupa, 

 the 5th, 6th, and 7th abdominal segments narrowing gradually, the 

 8th, 9th and 10th rather abruptly. The colour is a very pale brown 

 (khaki, terra-cotta or nankeen) with some black markings and a 

 sprinkling of fine black dots. The most conspicuous black marks 

 are round the spiracles (thoracic and on the 2nd, 3rd, 4th, 5th, 

 6th, 7th and 8th abdominal), forming a row of conspicuous black 

 marks, fairly constant in size, the largest, on the 4th abdominal, 

 being about 2 -5mm. in diameter. The colour and markings 

 are very similar to those of the pupa of Hippotion celerio. A 

 fine black line runs narrowly down the maxillae, affecting in front 

 little more than the median suture, posteriorly the whole width of 



R 



