260 THE EFFECT OF TEMPERATURE 



diminishes regularly with progress of development. It 

 is obvious, however, that both these capacities for re- 

 action are quite unusual, and have been specially ac- 

 quired for a special purpose. In all probability the 

 organisms are not more sensitive to environmental con- 

 ditions in general at these periods than they are at the 

 earlier ones; in fact, they are probably very much less 

 so, in that the growth has almost ceased. 



In certain caterpillars the existence of a variable pro- 

 tective resemblance has long been recognised, several 

 instances of the phenomenon being collected by Mel- 

 dola * in 1873. For example, the larvae of the genera 

 SmerintJius and Sphinx, which are green when feeding 

 on their respective food plants, become brown previous 

 to pupation, when the caterpillars are crawling over the 

 ground to find a suitable burying place. Again, the 

 Geometer Acidalia cleg ener aria is greenish brown in the 

 summer, but changes to a rusty brown in the autumn, 

 at the period preparatory to hibernation. Some years 

 later, Meldola recorded an observation by Mr. E. 

 Boscher,f relative to the larvae of SmerintJius ocellatus 

 (Eyed Hawk Moth). These larvae were noticed to be of 

 a whitish green colour when feeding on one species of 

 willow, and of a bright yellowish green when feeding 

 on another species, these colours being, on the whole, 

 protective. It was generally believed that such varia- 

 bility in the colour of caterpillars is due to the direct 

 chemical effect of the different kinds of leaves eaten, 

 but Professor Poulton,$ by his experiments on the 



*Proc. Zool. Soc. 1873, p. 153. 



f Weismann's " Studies in the Theory of Descent," 1882, p. 241. 



j: " Colours of Animals," p. 149. 



