SPECIES KV rNKIMTKSIMAL VAP! VTI>r;S. 5 



appi-eciably altered, while the monsoons whi'cli now blow 

 over the Island must have blown as at present for a vast period 

 of time, no sane individual will pretend that these species are 

 the last survivals of species, which were formerly widespread. 

 If they were so, there is no reason why they should not be 

 found upon some of the other hills of the island, whereas in 

 actual fact, as I shall show in a subsequent paper, nearly every 

 isolated mountain in Ceylon has its own particular species 

 endemic to it, so that, if we suppose all these to be survivals, 

 we shall be hard put to it to explain the former condition of 

 so small an island. 



It must therefore be admitted that the species peculiar to the 

 summit of Ritigala must have been evolved upon- that summit, 

 or near to it, and consequently, so far as the reply of the 

 infinitesimal variationists above-mentioned is concerned, the 

 argument that the characters may have been useful " some- 

 where " has the bottom knocked out of it, for, if the characters 

 were ever useful, they must have been useful on the summit of 

 Ritigala, and nowhere else. Not only so, but the argument 

 as to the " some time " is almost equally useless, for there is 

 no evidence to prove that the conditions on the summit of 

 Ritigala have altered within any reasonable period, in fact 

 within the period during which most, if not all, of our 

 existing species have evolved. If the characters themselves 

 were not actually useful, they must (on the theory of 

 infinitesimal variation) have been correlated with useful 

 variations. 



Now let us examine the endemics of Ritigala, beginning 

 with Coleus elongatus, Trimen. There are four species of 

 Coleus in Ceylon, but this species is by far most nearly allied 

 toC. barbatus, Benth. , which, be it carefully noted, also occurs 

 at the summit of Ritigala. From the other two species it 

 differs very much in the inflorescence, in the size of the leaves, 

 and in the habitat, as well as in those points in which it differs 

 from C. barbatus. No one can for a moment, however, main- 

 tain that these forms are merely varieties ; they are very 



