to limitations of the Mitllerian Hypothesis of Mimicry. 129 



of inedibility, and the complications which would be intro- 

 duced by inequality in this respect have not been taken 

 into account. 



When we recognise however that there actually are 

 varying grades of unpalatability in butterflies, and that 

 certain enemies may adapt themselves to prey on the 

 less protected forms while avoiding those that are most 

 nauseous, it becomes clear that any mimicry which may 

 arise as a result of such selection could not possibly be 

 classed as Miillerian, but would be due to the simple 

 operation of the principle enunciated by Bates, in spite of 

 the fact that the mimic possesses qualities rendering it 

 distasteful to other animals. That there is no inherent 

 improbability in such a supposition is shown by some 

 experiments I made upon a tame ground hornbill (Bucorax 

 caffer) in Natal (Trans. Ent. Soc. 190i, p. 347). To this 

 bird, which roamed unconfined, I gave at different times 

 five species of Acrxa and one of Flanema, all of wliich 

 were readily eaten and with evident appreciation; but 

 when a Danaida ehrysipp2is was offered to it, it merely 

 crushed the thorax and dropped it at once, a second 

 specimen being treated in a precisely similar manner. 

 Again, through the kindness of Mr. C F. M. Swynnerton 

 I have obtained proof that wild Bee-eaters will prey upon 

 Acrseas; there is also some evidence that they avoid 

 Danaines. When fuller information is obtained on this 

 point it may be necessary to reconsider the current 

 supposition that the mimicry of Acrxa encedon for 

 Danaida chrysippnis is purely Miillerian. 



But there is another way in which Batesian mimicry 

 may have arisen among unpalatable butterflies. Professor 

 Poulton long ago pointed out that insectivorous animals 

 which, under normal conditions, would refuse insects 

 having an unpleasant taste, would yet eat them when 

 driven thereto by hunger. If we suppose that in such 

 circumstances a wild bird were compelled to feed upon 

 distasteful insects, upon making experiments it would 

 doubtless find that while some of these were extremely 

 unpleasant, others would be passable. It seems an en- 

 tirely fair and reasonable supposition to infer that this 

 bird would select the species which were less unpalatable 

 to the exclusion of the others, and if the necessary 

 variations arose, this selection would tend to cause the 

 former to become Batesian mimics of the latter. Such 



TRANS. ENT. SOC. LOND. 1908. — PART I. (MAY) 9 



