Explanation of form and colouring. 209 



objections, you can hardly claim that any results you may have obtained from, 

 at any rate, captive animals are necessarily in the least degree reliable/' I would 

 agree ; and it is only because T believe that these objections can be amply 

 met in the way required by my critic that I have, as a matter o£ convenience, 

 adopted the present order. I think it right, however, in the meantime to 

 indicate the lines on which I would be prepared to meet the criticisms : — 



1. ApiJarent reluctance of birds to feed on butter/lies. [The following was 

 written long before my recent short paper for the Entomological Society 

 (Proc. 1915, p. xxxii), and should be read in conjunction with it, also with my 

 paper on the subject in 'The Ibis' (Oct. 1912, p. 635), with Mr. Marshall's 

 paper referred to below, and my own remarks at the end of Section 2 of the 

 present paper.] I cannot help feeling that those who have believed that birds 

 do not feed to a very appreciable extent on butterflies have, at any rate^ had a 

 good deal of apparent justification for their opinion. The evidence supporting 

 that opinion is of a negative character, but it is bulky. As for its quality, I 

 should suppose that some at least of the very extensive and admirable 

 American stomach investigations that failed to reveal much in the way of 

 butterfly c^d'^ns were microscopical : I do not know how completely exhaustive 

 in relation to the very finest debris they were, or to what extent specially 

 directed towards the finding of Lepidopterous remains. So far as actual 

 attacks seen and recorded are concerned, Mr. MarshalFs painstaking and 

 highly resultful investigation showed (Trans, Ent. Soc. Lond. 1909, pp. 329- 

 383) that these have been vastly more numerous than was generally believed : 

 his total was close on six hundred, and very many of the records were 

 multiple. At the same time, butterflies being as a whole the most conspi- 

 cuous and distinctive of insects, and attacks by birds on less recognizable 

 insects being a matter of such daily observation (I speak, at any rate, of my 

 own experience), it would certainly seem significant that attacks on butter- 

 flies too are Jiot witnessed with the greatest frequency. Mr. Marshall has 

 suggested one or two excellent and cogent probable reasons for this lack of 

 evidence, and I am able, from my own observations, both to confirm these 

 and to add one or two others. 



Butterflies are, in any case at Chirinda and in many other parts of the 

 world, by no means abundant relatively to the members of certain other 

 orders. At a time when butterflies were particularly abundant I have 

 watched more small Diptern, etc., cross a shadow in half a minute than I 

 have seen butterflies in the whole day ; each year I have watched the great 

 grass-fires drive out the butterflies as single spies, but the grasshoppers in 

 battalions ; I have roughly calculated that probably more driver-ants passed 

 a certain spot in one hour, and that more wild hive-bees could be seen any 

 day in April at a single one of my Eucalyptus saligna gum-trees in a very 

 few hours, than the number of butterflies ever seen by me in a year — this 



