COLIINjE. 83 



Tliey select the sea-coast, I feel sure, simply to avoid obstacles. The road 

 between Trincomalee and Kandj, which runs through dense forest, is also 

 lai'gelj used by the migrating insects. When travelling south they have the N.B. 

 monsoon behind them, but when turning north tbey meet a stiff wind which really 

 seems to drive them to a faster flight. The breadth of the flight is usually not more 

 than a quarter of a mile. 



" The reason for these flights is at present very obscure ; it was probably 

 originally a question of food-supply. This instinct might have arisen from the 

 necessity for constantly seeking new feeding-grounds for the larvjB. As the species 

 increased, this tendency to expand would not only preserve the species, but would 

 cause in time its very material increase ; the necessity for constantly enlarging the 

 feeding-grounds would in time produce an inherited tendency to migrate. But in 

 due course, when all available feeding-grounds were occupied, as they soon would 

 be, in a small island like Ceylon, some check would be required to keep the 

 enormous number of resulting butterflies within due bounds, otherwise the species 

 would be in danger of annihilation from their very numbers. This appears to me 

 to be eff'ected in the following manner: — The insects of the wet-season migration 

 are mostly composed of females, and provided that the males can successfully 

 impregnate more than one female, the result would be an enormous number of eggs 

 laid, and this I have shown to be the case. The migratory instinct is so strong 

 that the females are precluded from taking any precautions for their future 

 offspring, as the females of most butterflies do ; and the result is that the struggle 

 for existence among the multitude of larvee subsisting on the food-plant, which is 

 quickly diminishing by their voracity, and also slowly by the heat and dry weather, 

 is so great that the larvae which would produce female butterflies succumb, and a 

 great majority of males are produced which form the dry-weather flights. This 

 majority of males would also be another factor in checking the increase of the 

 species. Daring the intervening portion of the year the species would gradually 

 increase, until the wet months at the fall of the year favour a luxuriant vegetation, 

 and all the female larvae then survive, and possibly being stronger crowd out the 

 male larvEe. These larvfe produce the overwhelming proportion of females in the 

 next tvet-season flight, with the result shown above. This migratory instinct, 

 originally due to a necessity for the increase of the species, is now become a means 

 of preventing its undue propagation." (Trans. Ent. Soc. Lond. 1904, pp. 701-6.) 



MiGEATORY Habits: in Bukjia. — Col. C. T. Bingham writes whilst "returning 

 down the Salween to Moulmein, on a hot steamy day in October, and when belou^ 

 Shwegon, I noticed clouds of butterflies, chiefly Gatoimlias, migrating, crossing 

 the Salween from East to West in a continuous stream" (Tr. Ent. Soc. 1902, 

 363). 



