8 WTLLIS : EVIDENCE AGAINST THE ORIGIN OF 



In the fourth place, the area of the summit of Ritigala is 

 very restricted, at the most being but a few acres. Not only 

 so, but the exposed rocks, upon which alone do the two species 

 of Coleus grow, do not occupy more than about a quarter of an 

 acre at most. Now, will any one pretend that the struggle for 

 existence is so keen upon so small an area that it can make 

 much difference to a Coleus what kind of calyx it has, or what 

 sort of inflorescence ? It might be possible to advance this 

 opinion if even the quarter of an acre were densely covered 

 with the plants, but in actual fact there are probably not a 

 dozen plants of either species on the whole summit of Ritigala, 

 and these are too widely separated for there to be much com- 

 petition between them. To get such wide differences between 

 the species, on the theory of natural selection of infinitesimal 

 variations, we should have to have had a strenuous competi- 

 tion between very many of them, and no one can pretend that 

 this is, or ever has been, the case on Ritigala. Even if the 

 whole area were bare rock, and were covered with these 

 species alone to the exclusion of others^ the competition would 

 be comparatively small. But this cannot be allowed for the 

 Coleus, for there are also endemic species of Trichomanes, 

 which lives in shady places. Bulbophyllum. which lives epi- 

 phytically, and Cyperus, living in open grassy places, to say 

 nothing of quite a number of endemic varieties. Can it be 

 honestly said that any of the characters of this species are of any 

 use to it in the struggle for existence ? 



Lastly, comes the difficulty that the species from which 

 C. elonyaMs is with almost absolute certainty descended, 

 C. harhatus, is also to be found living upon the summit of 

 Ritigala, and as numerously as C. elongatus. Now, if we 

 suppose that the characters of the latter were so advantageous 

 to it in the very small struggle for existence that went on upon 

 the top of Ritigala, we must surely suppose also that 

 C. harbatus would succumb in that struggle, but instead of 

 that it remains as numerous as C. elongatus. This would 

 therefore indicate that there was no serious struggle for 



