6 



unprofitable. The likelihood of this being the true explanation is 

 increased when we enquire into the habits as well as the colours of 

 the species in question. Remarkable as is the change of aspect in 

 passing from one time of year to another, the change of behaviour in 

 the successive broods is perhaps even more astonishing. Thus in the 

 case of Precis antilope, Mr. Marshall, who has had great opportuni- 

 ties of studymg the habits of butterflies of this group in nature, 

 informs us that " the dry-season form only frequents the bush, 

 settling on the ground among the dead leaves, or very rarely on 

 small plants, the underside colouring affording it excellent protec- 

 tion. As the season advances the habits of the insect change, and 

 the latter form may be found .... frequenting open tops of 

 kopjes, flying boldly about within a limited area, and settling with 

 expanded wings on shrubs and bushes." With regard to P. archesia 

 he notes that " the dry phase generally frequents the wooded and 

 the wet phase the open country." The same correspondence of 

 habit with change of aspect is also observable in the species of 

 Precis from which we started, viz., P. sesaniKs. Here the underside 

 of the dry phase, though not specially leat-like, is well adapted for 

 concealment in shady places. These are the places which it most 

 affects, and, according to Mr. Marshall, " if alarmed it flies off with 

 a rapid, and often zigzag flight, settling abruptly among rocks or 

 herbage, when its greenish-black underside colouring is equally pro- 

 tective." The wet-season form, on the other hand, says Mr. Trimen, 

 " frequents open, grassy hills, especially their summit ridges or 

 highest points, and is very conspicuous, whether flying or settled." 

 From all this it appears that, speaking generally, not only the 

 aspect but also the habits of the cryptic dry- season forms are such as 

 to aid concealment ; while the behaviour of the wet-season forms 

 seems to court observation, the choice of locality and the mode of 

 flight being such as to display their conspicuous aspect to the best 

 advantage. And these last are just the characters which we should 

 look for in an insect protected by some unpalatable quality, or possibly 

 itself palatable but depending for its safety on its resemblance to 

 some unpalatable model. In the latter case the warning feature 

 would of course be false instead of true, inasmuch as it would give 

 a deceptive suggestion of a nauseous quality that did not really 

 belong to it. This leads us straight to the interesting subject of 

 mimicry, into w^hich I do not propose at this time to enter, further 

 than to remark that there is good reason to suppose that some of 

 these wet-season forms have been affected by contact with the great 

 group of distasteful butterflies known as Acra?as, either in the way 

 of actual mimicry, or in the way of forming with them an inedible 

 group making use of a common warning pattern. In the latter case 

 we should have to suppose that the dry-season phase, notwithstand- 

 ing its cryptic habit, is also distasteful; for it is very unlikely, 

 though perhaps not absolutely inconceivable, that the butterfly 

 should accompany its seasonal change of aspect and habits with a 



