THE DISTRIBUTION OF BUTTERFLIES IN INDIA. 601 



are tho most isolated part of the oriental region, and these species belong- 

 ing to what is structurally by frr the oldest family must be relics of ono 

 of the earliest migrations. The great fixity of species shown after so 

 long a time suggests all sorts of theories on the problem of variation, such 

 as how the tendency of forms to vary is inherent or purely climatic 

 which must not be touched on here. 



It is a remarkable thing that this great peninsular island which 

 through countless ages was almost as isolated as Australia and with the 

 Lacadive and Maldive extension more than half as large produces 

 so few peculiar forms and so few of these really strikingly distinct. 

 Of the two peculiar genera only one is likely to have been locally 

 developed, the other being more probably a survival of forms elsewhere 

 extinct. The only solution of this difficulty I will offer for the moment 

 is that in cretaceous and earlier times the richest and most fertile part 

 which would be most likely to team with life of every kind was that 

 portion of the island which is now under the sea or buried in trap rock, 

 while the north-east of the island was comparatively sterile and badly 

 watered and only afforded a short resting place for immigrant species 

 on their wav from Assam to the south-west. It will be noted, too, that 

 of existing species the majority of immigrants from Assam have not 

 succeeded in perpetuating themselves in Gondwana and only found a 

 permanent new home when they reached the moist area of Malabar. 

 This is however a wild suggestion and not a solution. The older fauna, 

 if it existed, has become extinguished by the inroad of new, more 

 highly specialised forms since India became closely connected with the 

 rest of the world in that mysterious way in which older forms always 

 do before the march of younger races. 



