68 COLOR-RELATIONS OF CHRYSALIDS 



caused by the films of air between the thin plates of 

 glass which are formed on the surface of bottles long 

 exposed to earth and moisture. Both have the same 

 spectroscopic characters and the same transmitted colors 

 (complementary to those seen by reflection). The 

 brilliancy of tlie cuticle can be preserved in spirit for 

 any length of time ; it disappears on drying, but can be 

 renewed on wetting (this had been previously known), 

 and the colors are seen to change during the process 

 of drying and when the cuticle is pressed, for the films 

 are thus made thinner. The same lamellated layer exists 

 in non-metallic pupae of other species, and is us-ed as 

 a reflector for transparent coloring-matter contained 

 in its outer lamellae. Thus the structure which ren- 

 dered possible the brilliant effects due to interference 

 probably existed long before these special effects were 

 obtained, and was used for a different purpose." 



It has long been known that many of the chrys- 

 alids of the Vanessini which yield parasitic Ichneii- 

 monidae in place of their proper inhabitants are 

 frequently gilded ; which is to be explained, in 

 Poulton\s view, by the abnormal state of the cater- 

 pillar, which prevents the formation of pigment in 

 the chrysalis. In this instance the gilded appear- 

 ance is preservative not of the creature itself, but 

 of one of its foes, and does not appear a very wise 

 provision of Nature. 



