472 Mr. VoiiMon's further experiments upon 



It is interesting to note the antagonistic objections 

 which Mr. Bateson and Mr. Beddard urge against the 

 protective efficiency of colouring, the one holding that 

 enemies are purelj' imaginary, and the other that they 

 are so supremely successful that no concealment is of 

 avail against them. 



No one feels more keenly than I the truth which 

 Darwin so constantly urges in his letters, that we are 

 profoundly ignorant of the conditions of existence of 

 almost every organism. But Darwin never used this 

 ignorance as grounds for the assumption that enemies 

 are " imaginary " for any part of the life of any animal. 

 He rather felt that the enemies were apt to be under- 

 estimated than over-estimated. I have great hope that 

 this part of the evidence for natural selection will be 

 tested as severely as possible by those who believe in 

 the doctrine ; for there seems to be little chance of such 

 work being forthcoming from those who attempt to 

 depreciate it. It is very much easier to assume that 

 enemies are imaginary than set about a searching 

 enquiry into the conditions of existence as they affect 

 any one animal. But such expressions of opinion have 

 their value in stimulating those who consider them to 

 be eminently unscientific to obtain direct evidence. 



I have for a long time wished to undertake such an 

 investigation myself, but one man alone cannot do much, 

 especially in the vast field of observation which must be 

 covered in order to obtain adequate direct evidence. 

 This paper and my other works are an indication that I 

 have not been idle. In the hope that others may be in- 

 duced to work at the subject, I will therefore mention 

 some lines which I think would lead to useful results. 



Larvffi of such species as A. betularia might be 

 liberated upon plants which harmonize and upon others 

 which do not harmonize with their colour. Only one 

 larva should be placed on each branch, not many 

 branches on the same tree should be employed, and 

 the trees should be widely separated. The larvic might 

 be liberated at the last ecdysis, so that their colours 

 would remain nearly constant. They should be observed 

 and noted twice a day. If they disappeared at once, 

 allowance would have to be made for wandering, but if 

 they settled down on the branch, there would be no 

 reason for suspecting them of this. 



