the Coloration of Eygs. 567 



markings do occur, they are most notably unstable, not only 

 in the same species but in the same nest." 



That the objection sounds more formidable than it really 

 isj we may realize at once by remembering how readily, 

 cases of close resemblance apart, we ourselves differentiate 

 by their appearance, most eggs that we know in a given 

 locality. A local collection in which the eggs of each 

 species were so numerous as not to give undeserved promi- 

 nence to relatively infrequent variations, some of these 

 iuseparable from a mode of pigment-deposition that is 

 sensitive to fright, ill-health, &c., would show the same. 

 A very great deal of variability would remain, but certain 

 considerations must be borne in mind : — 



(1) Insects are recognized by a definite pattern^ eggs by 

 an indefinite one, by the general effect. The principle is 

 more suitable for a rollable object which has to convey 

 sufficiently the same visual impression, whichever side is 

 uppermost, and it is probably correlated also with the mode 

 of deposition of the pigment. This very indefiniteness, 

 while it detracts (necessarily) in no way from ease of 

 recognition, gives a far greater latitude to unimportant 

 variation (even as between two sides of the same egg) and 

 quite likely, by the attitude of mind engendered in an 

 enemy, also paves the way for the survival in certain cases, 

 without too great loss, of variation that is somewhat 

 greater. 



(2) The very fact that variation often occurs within the 

 clutch will be of advantage to a variable species, by 

 acquainting an enemy. Avith the specific range of variation 

 at a minimum of loss. 



(3) A number of insects also, and some insect-groups, 

 are very highly variable, though the variability is often 

 easily explicable on selectionist lines. Even if eggs are 

 much more commonly variable, this merely means more 

 cases to be explained. It is very possible that the explana- 

 tions will in no way conflict with the theory. It is even 

 possible that important factors may operate in relation to 

 eggs which are not present in the case of insects at all, 



(4) Great as may be the variability of eggs, it is never- 



