SATYRINJS. 149 



rainy day or two come, pup« formed during the wet interval would probably 

 produce the wet season form of butterfly. Again, as it takes butterflies some little 

 time to lay their eggs (after having completed this operation they die immediately) 

 it must frequently happen that the two forms overlap : — a dry season female not 

 having laid her eggs during the dry season would do so at the beginning of the 

 rains, and though caught in the rainy season, would still be a dry-season butterfly, 

 its worn appearance, if nothing else, proclaiming the fact ; and vice versa with a 

 wet season butterfly, not having completed her laying during the rains, and caught 

 in the dry season." 



On the subject of seasonal dimorphism in Indian butterflies of this subfamily, 

 Mr. W. Doherty (Journ. Asiatic Society of Bengal, 1886, p. 105) gives us the 

 result of his more extended observations, and says, " So far as my four years' 

 experience goes, there are four broods of Indian tropical and subtropical butter- 

 flies ; two in the wet season — in May or June and in August or early S eptember, 

 and two in the dry season — in October or late in September and in the first warm 

 weather of March, respectively. These periods vary in diS'erent localities, the 

 amount of rainfall being the chief cause of change. In Kumaon, where I spent 

 less than six months, I cannot be sure how many broods of butterflies occur there 

 or in what months they appear ; the second wet season brood, a numerous one, 

 appeared from the middle to the end of August, and the first dry season brood, less 

 important, especially in the drier valleys, came out in the last week of September, 

 my first specimen of Mycalesis visala having been taken September 22nd. In 

 Travancore, there was a small brood of dry-season forms early in March, and a very 

 large one in the second week of May. In Orissa and Ganjam, the first wet-season 

 brood did not appear till the end of July (the monsoon coming late that year), and 

 was poor in numbers. In the Chittagong Hill tracts, the last dry-season bi-ood, 

 including a vast number of species and of specimens, appeared in the middle of 

 March, while the first wet-season brood, both there and in Arakan, came out at 

 the end of May, and was a very small one. At Bassein, Burma, the first dry- 

 season brood, which, as I have said, appeared in Kumaon near the end of 

 September, was delayed till the middle of November. All butterflies do not have 

 four broods. A few seem to keep coming out at short intervals throughout the 

 year ; many are found only in the wet season, and some perhaps only in the dry 

 season. It is said that still others are found in but one month of the year, and so 

 have only one brood instead of four. Nevertheless, I think I may generalize my 

 experience into the brief statement that there are four broods, two of the wet, 

 two of the dry season, each of them simultaneous with or preceding by a month 

 the beginning and the end of the season after which I have named them. 



" Between the two broods of dry-season butterflies (October and March), and 



