50 The Study of Animal Life parti 



oiF and are replaced by white ones ; sometimes the 

 whiteness is the result of both these -processes. It is 

 directly due to the formation of gas bubbles inside 

 the hairs or feathers in sufficient quantity to antagonise 

 the effect of any pigment that may be present, but in 

 the case of new growths it is not likely that any pig- 

 ment is formed. In some cases, e.g. Ross's lemming and 

 the American hare (Lepus americanus), it has been clearly 

 shown that the change is due to the cold. It is likely that 

 this acts somewhat indirectly upon the skin through the 

 nervous system. We may therefore regard the change as 

 a variation due to the environment, and it is at least 

 possible that the permanent whiteness of some northern 

 animals, e.g. the polar bear, is an acquired character of 

 similar origin. There are many objections to the theory 

 that the winter whiteness of arctic animals arose by the 

 accumulation of small variations in individuals which, being 

 slightly whiter than their neighbours, became dominant by 

 natural selection, though there can be no doubt that the 

 whiteness, however it arose, would be conserved like other 

 advantageous variations. 



To several natutalists, but above all to Mr. Poulton, we 

 are indebted for much precise information in regard to the 

 variable colouring of many caterpillars and chrysalides. 

 These adjust their colours to those of the surroundings, and 

 even the cocoons are sometimes harmoniously coloured. 

 There is no doubt that the variable colouring often has 

 protective value. Mr. Poulton experimented with the 

 caterpillars of the peacock butterfly ( Vanessa io), small 

 tortoise-shell ( Vanessa urticce), garden whites (Pieris 

 brassicce and Pieris rupee), and many others. Caterpillars 

 of the small tortoise-shell in black surroundings tend to be- 

 come darker as pupae ; in a white environment the pupae 

 are lighter ; in gilded boxes they tend to become golden. 

 The surrounding colour seems to influence the caterpillar 

 " during the twenty hours immediately preceding the last 

 twelve hours of the larval state," " and this is probably the 

 true meaning of the hours during which the caterpillar 

 rests motionless on the surface upon which it will pupate." 



