242 THE VARIATION OF ANIMALS IN NATURE 



the lack of a pale moult in some northern mammals is as 

 much an argument against the general effect of environment 

 as a cause as it is against the adaptive origin. 



No one would attempt to deny that the white pelage is far 

 more frequent among arctic animals than those of warmer 

 climates. But even in the arctic region proper there are forms 

 which retain a dark coloration (Musk Ox, Reindeer, Mus- 

 tek zibellina), and it is not easy to explain this by reference to 

 special habits, etc. 



As for the presumed advantage of the white colour, we find 

 that there is little evidence to show that such animals are pro- 

 tected by the colour or that their habits render this feasible. 

 The question whether white forms on a white background are 

 not rendered conspicuous by the dark eyes, shadow and surface 

 modelling is usually disregarded. As for the heat-conserving 

 properties of the colour, this seems to be a negligible factor in 

 winter, and in high latitudes where the heat-losing properties of 

 dark and light colours are more or less the same (Cuenot, 192 1). 



It will be seen that the evidence on this subject is very 

 inconclusive, and in particular that the incidence of the white 

 moult in temperate regions and low altitudes (cf. Pulorius 

 nivalis monticola, supra) is of such a nature as to suggest that its 

 origin at least is non-adaptive. It may be noted that Hadwen 

 (1929) presents some evidence suggesting that white Reindeer 

 and cattle are more attacked by ectoparasites than are 

 normally coloured individuals. 



The problem of ' countershading. ' — The occurrence of ' counter- 

 shading ' was originally hailed as a remarkable demonstration 

 of the value of a particular type of coloration. The whole 

 subject has been very carefully reviewed by Roosevelt (191 1). 

 He points out that ' countershading ' can be of no value to 

 animals that are habitually attacked from above (e.g. by 

 hawks, etc.), nor to animals that are stalked along the ground, 

 for in most cases the carnivore which stalks in a crouching 

 position can see only the line of the prey's back and not the 

 line of the belly. ' Countershading ' can be effective only 

 when the prey is on a level stretch of ground, when the belly- 

 line is revealed and not concealed by irregularities of the 

 ground or by vegetation. 



The question is complicated, as Buxton (I.e.) has pointed 

 out, by the fact that in most cases not only is the under-side of 



