vi DEFENCES OF INSECTS 233 



and 45). This extraordinary capability is only present 

 in exposed chrysalides.* Most of the chrysalides of 

 butterflies are freely exposed ; the majority of moths 

 pass their pupal period in burrows in the earth, or 

 enveloped in thick cocoons, and among these sensi- 

 tiveness to surroundings is probably entirely absent. 

 Up to 1 886 the fact of the power of adjustment 

 was really all that was known concerning this strange 

 phenomenon, and the manner in which the colours of 

 these pupae are determined was a mere matter of specu- 

 lative theory. In that year Mr. Poulton commenced 

 his elaborate series of investigations, which have 

 thrown a flood of light on this important subject. It was 

 then accepted by Mr. Wood, and by most naturalists 

 interested in the question, that the skin of the pupa 

 was photographically sensitive for a few hours only 

 after the caterpillar's skin had been shed. Comparison 

 was made with the sensitive photographic plate, which 

 darkens under the influence of light. Professor 

 Meldola, however, subsequently showed this photo- 

 graphic sensitiveness of chrysalides to be a mistake, 

 for ' the action of light upon the sensitive skin of a 

 pupa has no analogy with its action on any known 

 photographic chemical. No known substance retains 

 permanently the colour reflected on it by adjacent 

 objects.' This groundless, though very plausible, 

 theory was completely overthrown by Mr. Poulton's 

 simple experiment showing the animal's period of 

 susceptibility to surroundings to be in fact previous 

 to the development of the pupal form. From a small 

 number of freshly-found pupae he removed one, 

 immediately after the larva skin was shed, to a sub- 



