214 MR. E. B. POULTON ON THK PROTECTIVE [Mar. 1, 



which have been actually tested, and it is much to be regretted 

 that experimental investigations have not been further extended 

 and recorded in greater detail. The results of the tables of larvae 

 given above have been, in a very small proportion of cases, so 

 directly contrary to a priori expectation that I do not feel confident 

 in bringing forward any instances which have not been tested, although 

 I feel sure that the vast majority of them would yield favourable 

 results. I cannot, therefore, in this paper accept as satisfactory 

 the purely negative evidence that insect-eating Vertebrates have 

 been often seen to catch and eat insects of various kinds, but have 

 not been seen to catch at the same time and place certain highly 

 coloured species which were abundant and slow-flying. At the 

 time when Bates and Wallace first made public their most impor- 

 tant conclusions as to the meaning of conspicuous coloration and 

 the true significance of mimicry, it was quite right that evidence of 

 all kinds should be brought forward ; but after the lapse of twenty 

 years, we may fairly expect that conclusions which are so important 

 in Biology shall have received the most abundant and complete 

 experimental proof. And I know that lack of detail in the proofs 

 which have been afforded, and the fact that a large part of the evi- 

 dence brought forward is still founded on mere surmise (however 

 probable may be the result of an actual test), have prejudiced the 

 conclusions in the minds of many distinguished biologists, who 

 have come to look upon the whole subject with an undeserved 

 suspicion. 



I cannot find any record of actual experiments conducted upon 

 the well-known and conspicuous Heliconians and Danaids, and 

 therefore I do not include them in the following list. There is, 

 however, an observation of Meldola's which is of the nature of 

 demonstration, and which is so interesting that I quote it in his 

 words : — " It appears that tlie nauseous character of these .... 

 butterflies is to a certain extent retained after death, as I found 

 that in an old collection which had been destroyed by mites, the 

 least mutilated specimens were species of Danais and Euplcea, 

 genera which are known to be distasteful when living and to serve 

 as models for mimicry, see Proc. Ent. Soc. Lond. 1877, p. xii." 

 (Meldola's editorial notes to his translation of Weismann's Essays 

 above referred to, p. 337). This observation (since confirmed by 

 J. Jenner Weir, ' Entomologist,' vol. xv. 1882, p. 160) has the same 

 kind of interest as that of Butler upon spiders, drawing attention, 

 as it does, to the possession of a peculiar taste or smell which is 

 recognized as nauseous by animals as widely separated as the mites 

 and spiders are from lizards and birds. And such a consideration 

 enforces the conclusion previously arrived at from other evidence, 

 that when certain insect-eaters neglect the attributes which are 

 respected by others, we see the results of an "acquired taste" 

 produced in the first instance by hunger, and not by an obe- 

 dience to the dictates of an eccentric preference for what is very 

 universally regarded as disagreeable. 



Since the above was written, my friend and pupil Mr. E. A. 



