the Life History of Agriades tliersites. 305 



the end of the true pocket entered it, since the i:)ocket is so long as 

 to overlap it. In A, thersites the extremities or extreme tips of the 

 antennae also enter the pocket. This is probably the case in other 

 pujDae of this group in which the ends of the antemiae terminate, 

 not in a point, but by a square end as seen on the pujial exterior. 



I do not know how the imaginal skin over the venter of the fiftli 

 abdominal segment frees itself from this pocket, no trace of this 

 condition exists in the imago, unless the circumstance that the 

 ventral plate of the fifth abdominal segment is the smallest of the 

 series has some relation to it. 



Explanation of Plates XXVI, XXVII. 



[See Explanation facing the Plates.] 



Explanation of Plates XXVIII-LIII. 



Plate XXVIII. Top of the Valley of the Guisane, just below Le 



Lautaret, showing ( + ) habitat of 

 A . thersites at about 6400 ft. 



Plate XXIX. Looking down on Le Lautaret from the west, where 



A. thersites occurs at about 7000 ft. 



Plate XXX. One of the low ridges west of Le Lautaret where 



A . thersites occurs at about 7500 ft. 



Plate XXXI. Photographs of larva by Mr. H. Main. 



Fig. 1, P. icarus ; figs. 2, 3, 4, 5, A. thersites ; x 2. 



The most striking difference between the last stage larvae of 

 thersites and icarus, as seen by the naked eye or with a low power 

 hand lens, is that the dorsal hairs on the abdomen of thersites are 

 colourless, and often glisten like silver, whilst those of icarus 

 include many black hairs. 



In this and the following plates, the white hairs distinguish 

 A. thersites, and give P. icarus a dull dingy aspect in comparison. 



Plate XXXII. Photograph by Mr. H. Main. 



Tigs. 1 and 2, as also 3 and 4, are stereoscopic 

 of A. thersites larva (full grown) X 2. 



