142 Mr. George J. Romanes [Feb. 8, 



This last head of evidence — that, namely, as to local and 

 specific variations of instincts — differs in one important respect from 

 all the other heads of evidence which I have previously adduced. For 

 while these other heads of evidence had reference to the theory 

 concerning the causes of transmutation, this head of evidence has 

 reference to the fact of transmutation. Whatever, therefore, we 

 may think concerning the evidence of the causes, it is quite distinct 

 from that on which I now rely as conclusive proof of the fact. 



I shall now, for the sake of fairness, briefly allude to the more 

 important cases of special difficulty which lie against Mr. Darwin's 

 theory of the origin and development of instincts. For the sake 

 of brevity, however, I shall not allude to those cases of special 

 difficulty which he has himself treated in the * Origin of Species,' 

 but shall confine myself to considering the other and most formidable 

 cases which, after surveying all the known instincts presented by 

 animals, I have felt to be such. 



First, we have the alleged instinct of the scorpion committing 

 suicide when surrounded by fire. This instinct, if it really exists, 

 would no doubt present a difficulty, because it is clearly an instinct 

 which, being not only of no use, but actually detrimental both to the 

 individual and the species, could never have been developed either by 

 natural selection or by lapsing intelligence. I may, however, dismiss 

 this case with a mere mention, because as yet the evidence of the 

 fact is not sufficiently precise to admit of our definitely accepting it 

 as a fact. 



There can be no such doubt, however, attaching to another 

 instinct largely prevalent among insects, and which is imquestionably 

 detrimental, both to the individual and to the species. I allude to 

 the instinct of flying through flame. This is unquestionably a true 

 instinct, because it is manifested by all individuals of the same 

 species. How then are we to explain its occurrence ? I think we 

 may do so by considering, in the first place, that flame is not a suffi- 

 ciently common object in nature to lead to any express instinct for 

 its avoidance; and in the next place by considering that insects 

 unquestionably manifest a disposition to approach and examine 

 shining objects. Whether this disposition is due to mere curiosity, 

 or to a desire to ascertain if the shining objects will, like flowers, 

 yield them food, is a question which need not here concern us. We 

 have merely to deal with the fact that such a general disposition is 

 displayed. Taking then this fact, in connection with the fact that 

 flame is not a sufficiently common object in nature to lead to any 

 instinct expressly directed against its avoidance, it seems to me that 

 the difficulty we are considering is a difficulty no longer. 



The shamming-dead of insects appears at first sight a formidable 

 difficulty, because it is impossible to understand how any insect can 

 have acquired the idea of death or of its intentional simulation. 

 This difficulty occurred to Mr. Darwin thirty or forty years ago, and 

 among his manuscripts I find some very interesting notes of experi- 

 ments upon the subject. He procured a number of insects which 



