280 THE VARIATION OF ANIMALS IN NATURE 



with the phenomena of protective coloration and mimicry. A 

 few suggestions may be added. 



It is really very difficult to estimate whether we ought to 

 call a colour-scheme protective, warning or neutral, except 

 in the limited number of cases in which there is striking resem- 

 blance to bark, rock, green leaves, etc., or in which the colours 

 are very unusually conspicuous. We are probably on safer 

 ground in affirming that a given pattern is cryptic than in 

 saying that it is conspicuous, not only because many apparently 

 conspicuous patterns really blend with their natural back- 

 ground, but because, so far as colour-pattern may have been 

 influenced by Natural Selection, it is much more likely that 

 cryptic rather than conspicuous patterns would have been 

 produced. 



Even so, it is difficult to believe that the colour of a large 

 number of animals is not neutral with a slight bias in the 

 cryptic direction. Few animals live in so well defined a 

 habitat that resemblance to any one conspicuous feature 

 would be serviceable, and actually cases of highly specialised 

 protective colours are not very numerous. Where the colours 

 are broadly cryptic, do we find that species differ in such a 

 way as to fit them for their particular habitat ? This question, 

 on our present knowledge, would, with very few exceptions 

 (see p. 236), have to be answered in the negative. But there 

 is another possibility. If we imagine two isolated populations 

 of a species, each under the action of selection in favour of a 

 generalised cryptic colour-scheme, it is quite possible that a 

 more or less successful pattern might be produced in both 

 cases ; but two patterns, not one, might result, since they would 

 have evolved in different ways, as the result of the various 

 mutations that happened to occur in the two populations. 

 Later, when the populations had become fixed as species, the 

 two might mix again, and then, though both would have a 

 generally cryptic pattern, the differences between the two species 

 would appear non-adaptive. Doubtless evolution has some- 

 times followed this programme, but it would be a big 

 assumption to refer the greater part of specific difference in 

 cryptic patterns to such a process. It would appear that 

 even on this explanation, where two cryptic patterns have 

 been built up independently under the action of selection, 

 we have to assume that each step in the evolution of pattern 



