11 



moisture enter as a factor into the conditions which regulate the 

 aspect of seasonally dimorphic butterflies. Thus Colonel Manders 

 conducted some experiments in Ceylon which showed that abnormal 

 conditions of dryness applied during the larval period, with no 

 alteration of temperature, were capable of causing specimens of 

 Cato/isiUa pi/ranthe, a dimorphic species somewhat resembling our 

 "brimstone" butterfly, to assume the spring instead of the autumnal 

 or winter aspect. 



Some very interesting investigations by Mr. G. A. K. Marshall, 

 in Mashonaland, point to a similar conclusion. There is a large 

 genus of butterflies, mostly African, known as Teracoliis. Many of 

 the species of this genus are the subjects of a well-marked seasonal 

 dimorphism ; so well-marked, in fact, that for a long time the 

 seasonal phases were treated as different species and received dis- 

 tinct specific names, just as in the case of Araachnia lerana and A. 

 prarsa. Mr. Marshall experimented on two of these species. Tera- 

 coliis oiiiphale, and T. achiiie, both white butterflies with orange tips, 

 as follows : — He reared some larvte which under ordinary conditions 

 would have resulted in dry-season butterflies, under artificial condi- 

 tions of warmth and moisture. The butterflies which emerged 

 during the height of the dry season were of the wet-season phase; 

 the transformation being a little more complete in some which had 

 been exposed to the experimental conditions both as larvfe and as 

 pup<T, than in others which had been so exposed as larva? alone. 

 On the other hand, some larvse which had been reared under normal 

 conditions, but subjected as pup^e to moist heat, gave rise to butter- 

 flies which showed little or no change from the dry-season forms 

 that were captured on the wing at the same time. These experi- 

 ments demonstrated that in these two TeracoU, as in the Araschnias 

 investigated by Mr. Merrifield, it is the larval and not the pupal 

 stage that is most susceptible to the climatic influence. Indeed the 

 pupal stage in these Teraculi appears to be almost entirely resistant 

 to experimental interference, while in the Araschnias it is by no 

 means inaccessible to such influence, though much less so than the 

 larval period. 



That this principle does not hold good in all cases was clearly 

 shown by some further experiments by Mr, Marshall on Beleiwis 

 severina, a white butterfly common in most parts of Africa south of 

 the Sahara. Here no perceptible effect was produced by exposing 

 the larvae to damp heat, but when the same artiflcial conditions 

 were applied to both larva and pupa, the resulting butterflies were 

 transformed into the full wet-season form, though the brood under 

 ordinary circumstances would have belonged to the dry-season phase. 

 Hence we must conclude that in Belenois severina it is the pupal and 

 not the larval stage that is most susceptible. A farther result of 

 great interest emerged from the experiments on this species. When 

 the larvfe and pupae were exposed to moisture alone without addi- 

 tional heat, the butterflies were found to have assumed the bright 



