THE THEORY OF MIMICRY 159 



Avliat unfortunately for tiie warning colour theory, they 

 provide only one genus (Mylothris) to which can be at- 

 tributed any distasteful properties, and in that case only 

 in a minor degree. This discrepancy of so important a 

 group is felt as a weak joint in the " Mimicry " armour, 

 and has led up to assumptions of the partially protected 

 characters of such genera as Belenois and Terias. Any 

 field naturalist of experience and of unbiassed mind will, 

 on the contrary, bear me out when I say there is no class 

 of butterflies more preyed upon than species of these 

 genera so far as predacious insects are concerned and 

 the records from birds appear equally damaging to these 

 assumptions. It is worth noting that during thr» dry 

 season in Natal, and probably elsewhere, the whites are 

 proportionately more numerous compared with butter- 

 flies of other colours than during the wet season. This 

 is the season when one might expect the attacks by birds 

 to be the most exacting. 



Mr. Eltringham even makes a model of Atella phalan- 

 tha, one of our most wary and least likely of insects, to 

 figure in that role; and, further, figures Mylothris tri- 

 menia as model to Phrissura lasta (female). As the My- 

 lothris is localized to parts of the Eastern Provinces of 

 the Cape and a few districts of Natal, and Phrissura lasta 

 is recorded from British East Africa, I fail to see what 

 purpose the resemblance serves in the mimicry sense. 

 There are several other examples given that have as little 

 right to consideration in this respect. 



In conclusion, to briefly summarize my view^s: — 



I consider that the evidence in support of Mimicry bv 

 natural selection as a protective agency is quite insuffi- 

 cient, and often misleading, except, perhaps, in those 

 cases that come under Procryptic disguises. In these 

 cases protection is as much afl'orded by absolute immo- 

 bility as by the nature of the disguise. 



That in cases of analogy in which the methods and 

 organization of non-related communities are of similar 

 nature, I am inclined to think the instincts involved are 

 partly due to psychological causes. 



