478 Dr. T. A. Chapman's Contributions to 



Oxytropis campestris, the differences in the first six were 

 mere variations in the larva, and the remainder differed 

 owing to their being parasitised. 



It follows, by the way, that P. icarus in the locality at 

 Le Lautaret where I got these larvae does not affect Oxy- 

 tropis campestris, there are there plenty of other Papilio- 

 naceae for it, including its favourite Lotus corniculafus. The 

 six or eight larvae that proved to be parasitised " never 

 grew larger, and at about half the size (bulk, not length) of 

 the others retreated downwards into any available narrow 

 space and then took a cylindrical form, became fastened 

 by a little effusion from the mouth and dried up into a 

 brown cyclinder, in which state they remain, and are almost 

 certainly the hosts of some ichneumon, size 7"5 to 9'0 mm. 

 long, 2"5 to 2*8 mm. wide in different specimens. The 

 fastening by the head makes one suppose it is the tail 

 end till specimens are more closely inspected." 



The following notes made on the pupae of eros when it 

 was supposed more than one species might be represented 

 in my material, with a view to finding some distinctions 

 between the (supposed) species, will serve to show the 

 extent of variation in the pupae of P. eros. 



June 20. — No. 1, the largest and most striped (oblique) 

 larva, has a very transparent pupa, faintly greenish, 11 mm. 

 long, wings very glassy, showing tracheae ; there are a good 

 many very short hairs over head and prothorax. There is 

 a waist (seen laterally) about 3*5 mm. high, prothorax and 

 3rd abdominal about 3'8. 



June 26. — Wings are now satiny white, eyes brownish 

 and whole pupa more solid looking. 



June 20. — ^Pupa of larvae 2 and 3, extremely like No. 1 

 as to colour, transparency, etc., hairs rather more obvious 

 especially along abdomen, where they can hardly be seen 

 in No. 1. They are also a little smaller, 10'5 mm. long, 

 but about sam.e height. Larval skin covers last segments 

 in both. No. 1 is nearly free, this one has a " girth " 

 consisting of several threads on each side arising from 

 position of lateral pad, but taking separate directions — 

 one over prothorax, one over 1st abdominal and one 

 going right back to 6th abdominal segment; there are 

 also a few obhque and one or two longitudinal threads, 

 uniting these several strands, so that though the structure 

 hardly exists still it seems as much entitled to be called a 

 cocoon as a girth. The larval skin is quite free from any 



