196 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, Vol. XV 111. 



butterflies for many years. There are few species that do not exhibit the 

 phenomenon in some degree and it is impossible to compare the various aspects 

 of it presented by different species in different places without inclining to 

 Mr. Young's opinion, that the cause will be found in the food of the larva, 

 Now the growth of vegetation does not depend on temperature, nor on humidity, 

 but on the combination of the two, and it is to this combination, I believe, that 

 we must look for the true explanation of the seasonal variation and the local 

 variation of butterflies and moths. Let us take the local first. If two 

 collections, one from the plains of India and one from the coast of 

 Canara or Malabar, are put side by side, the latter will be found to be 

 conspicuously darker and richer in colour. In some species the difference 

 i? striking, Specimens of Danais limniace from the warm and humid 

 southern coast are so different from Bombay specimens that I actually 

 sent some to the late Mr. de Niceville for his opinion whether they 

 were limniace or septentrionis. He decided, not without hesitation, that they 

 were limniace. A more remarkable case is that of Nepheronia gcca, a common 

 Bombay butterfly, which is found in Canara also, but only in the cold season, 

 its place being taken, in the monsoon, by N. pingasa, a very dark form not 

 extending to Bombay. I believe this to be merely a monsoon form of the 

 other, which the moisture and warmth of Bombay are nob quite sufficient 

 to produce. To come to seasonal variation, Hypolimnas bolina presents an 

 instructive illustration. Here there are not two forms, but a regular grada- 

 tion. The first specimens which appear in the rains, emerging from the latest 

 pupse of the previous November or December, which have hibernated or 

 sestivated through the dry months, are scarcely larger than H. mysippus, 

 dull in colour and marked with a broad whitish fascia on the underside. 

 The next brood is very different, but it is not until August that we get, in its 

 perfection, the magnificent butterfly that used to be distinguished as H. avia. 

 The food plant of this species is a monsoon annual, but Melanitis leda, or ismene, 

 feeds on grasses and is found all the year, in two well marked forms, of which 

 one displaces the other suddenly in Western India just when monsoon conditions 

 have ceased. Now, if the change is a result of the nutrition of the larva, and the 

 larval and pupal life extends to six weeks, as Col. Manders shows, then it is 

 evident that we must look back about tw r o months from the appearance of the 

 butterfly to find the conditions that produced it. Apply this to Col. Manders' 

 tables. The month in which most rain fell was January, the next March, the next 

 December : from January to March inclusive only wet-season forms were taken. 

 With the decreasing rainfall of April and May we have first intermediate and 

 then dry-season forms in June and July, which culminates in June with only 

 6*45 inches of rain, and August without a single specimen of the wet-season form. 

 All the months do not answer so unequivocally, however, and I would suggest 

 experiments in feeding larva?, some on abundance of the most succulent food 

 obtainable and some on dry fare. The latter must not be starved, or their 

 growth will be stunted, which is not the case, as a rule, with the dry-season 



