in Butterflies of the Genus Precis. 31 



P. sesamus the dry-season form." This statement, coming 

 as it does from the pen of such an authority as Mr. Distant, 

 would at first sight appear to entirely refute my contentions. 

 But seasonal dimorphism is a curiously complex phenomenon, 

 and personally I venture to think that his experience is an 

 exceptional, and at the same time an extremely interesting, one; 

 for, apart from my own observations, it is quite at variance 

 with those of several thoroughly observant S.-African col- 

 lectors with whom I have discussed the subject and whose 

 acquaintance with the species is even more extensive than 

 that of Mr. Distant. Among these the testimony of Mr. C. 

 N. Barker is of particular importance, for he has observed 

 P. octavia (s. L) for many years from a dimorphic standpoint, 

 not only in Natal, but also in Zululand, Swaziland, and 

 Tongaland, and he has been firmly convinced that P. sesamus, 

 Trim., there represents its dry-season phase, having arrived 

 at that conclusion prior to myself and quite independently. 



It therefore gives me no little pleasure to be able to indicate 

 the accuracy of the observations of my genial friend, to 

 whom is due the credit of having first drawn attention to 

 the widespread occurrence of seasonal dimorphism among 

 African butterflies. For after not a few disappointments 

 I have at last succeeded in breeding typical P. sesamus, 

 Trim., from eggs laid by three separate females of P. octavia, 

 subsp. natalensis, Staud. — thereby establishing beyond doubt 

 what is certainly the most remarkable instance of seasonal 

 variation as yet known among the Lepidoptera. 



As a consequence of this, the synonymy of the species will 

 now stands as follows : — 



Wet-season form. Dry-season form. 



P. octavia, Cram. =[-P- umestris, Druiy.] 



tP. octavia, var. natalensis, Staud. ,~| „ , . , ... . 



= calescens, Butl. J = R octavia > subsp. sesamus, Trim. 



The following notes on the early stages of the species may 

 perhaps be of interest. On the 13th February of this year 

 I took five eggs laid by a female o -natalensis ; of these two 

 proved infertile and the remaining three hatched on the 19th. 

 One larva died when quite small, but the others thrived and 

 finally pupated, one early on the 20th, the other on the 21st. 

 The former emerged on the 4th April as sesamus, the latter 

 was unfortunately damaged by mischance and failed to 

 emerge. On the 27th February 1 obtained three eggs from 

 another female of the same form. Two of these hatched on 

 the 5th March ; the third hatched next day, but the larva 

 died. The surviving larvae developed somewhat unevenly, 



