106 • [October, 



result was most interesting. Instead of a constant larva, I found 

 that scarcely any two were alike, and that not only did the ground- 

 colour vary extremely, but that the development of the red dorsal and 

 sub-dorsal lines was very different in different individuals. In some, 

 the suffusion was so complete that the whole of the dorsal area was 

 red ; in others, so faint that the larva was almost unicolorous green 

 or whitish, according to the ground-colour ; in some specimens, too, 

 the larvae had decidedly brownish, rather than reddish, stripes, in 

 others grey. The notes I made of the different varieties at the 

 time were as follows : — 



(1.) Ground-colour grey, with very dark reddish dorsal and sub-dorsal stripes. 

 In some specimens of this form the colour of the stripes suffuses the whole of the 

 dorsal area, while in others they exist simply as fine lines, often reduced to a series 

 of dots, by the breaking up of the lines in the neighbourhood of the abdominal 

 incisions. 



(2.) Ground-colour green, with the reddish dorsal and sub-dorsal stripes offering 

 almost as great differences as in var. 1, both as to depth of colouring, and area of 

 suffusion. 



(3.) Ground-colour whitish, with very faint reddish stripes, sometimes only a 

 faint reddish tinge in tbe dorsal and sub-dorsal areas. 



(4.) Ground-colour whitish, with indistinct, grey, narrow, dorsal and sub-dorsal 

 stripes, in fact, almost unicolorous. 



These differences were not due to different stages of maturity, as 

 these descriptions were written from larvae when perfectly full-fed. 



I believe it is generally supposed that all the Pterophorina simply 

 attach themselves by the anal segment when pupating, and do not 

 spin a web or other protection over them. Gonodactyla always does : 

 whether it spins up in the pappus of the flower-head, whether it spins 

 on a leaf, or on the side of a box, a slight web is spun over the larva 

 always. 



If I was surprised at the result of my observations with regard 

 to the range of variation in the larval stage, I was still more surprised 

 when these larvae commenced pupating. There seems as much 

 divergence in colour and markings in this stage as in the larval, a 

 very unexpected occurrence, for it is rare that any insect varies a 

 great deal in the pupal stage, and when it does (as in Papilio Machao'n), 

 it rarely exhibits intermediate forms. The varieties I noted were as 

 follows : — 



(1.) Ground-colour pinkish, with dark reddish dorsal and sub-dorsal stripes, 

 with the wing-cases darker and striated all over with fine longitudinal black lines j 

 generally, but not always, with distinct black spiracles. 



