the Coloration of Eggs. 553 



the models could be found (if necessary) in the suitability 

 of white as a warning colour for dark places. It is the 

 warning colour par excellence of night-flying Lepidopteru. 

 And Mr. Pycraft has suggested that white in holes, for 

 the same reason of visibility, is useful for the avoidance 

 of accidental breakage by parents *, and, it might be added, 

 by other animals sheltering in burrows. Last, and not 

 least, we may add to these various possible contributing 

 factors (as an alternative or crown to variability) the loss 

 of pigment when it becomes useless and unselected, as in 

 an egg that is now inaccessible (through position or an 

 effective guard) or in darkness. 



5. A Reservation. — The possibility of mimicry in eggs must 

 be treated with caution, as pure coincidence in their colora- 

 tion is so general a phenomenon. As I said at the meeting : 

 " The coloration and pattern of eggs is so simple as to lead 

 us to expect the same scheme of colouring to crop up over 

 and over again quite independently of advantage. I have 

 lately been working at eggs in the British Museum and at 

 Tring, and nothing has struck me more than the fact that 

 this continually occurs — very often, it is true, obviously 

 through affinity, but very often again not. And this state 

 of affairs waiuis us to be very wary about attributing a given 

 resemblance to mimicry." 



In support of this warning, I speciully exhibited several 

 African eggs that in appearance are indistinguishable from 

 certain unrelated English eggs. It matters little whether 

 we call such accidental resemblance *' coincidence " or the 

 result of "parallel evolution/^ 



" At the same time,*' I went on, " it is a state of affairs 

 under which mimicry may often have taken place — for what 

 is mimicry, at best, but selected coincidence? — and the 

 material for selection, the coincidences, are here abundant. 

 So that if we rigidly confine ourselves to a few highly 



* Believed by a speaker at the meeting to have been suggested 

 originally by Seebohm, but Mr. Pycraft tells me that, so far as lie is 

 aware (and be has looked up tbat author), the suggestion originated 

 with himself. 



