58 AGE AND AREA [pt. i 



shall go on. Cordyceps, with 60 species on insects, is a similar 

 case (73). 



Another popular theory about localised species like these 

 Ceylon endemics is still strongly held, though the one just con- 

 sidered (local adaptation) has suffered somewJiat of an eclipse 

 with the gradual decay of the hypothesis of natural selection. 

 Like the first, though completely at variance with it, this second 

 explanation is also founded upon natural selection, but some- 

 what less obviously. It is to the effect that species on very small 

 areas are really in process of dying out. It is evident that they 

 could not have arisen by aid of natural selection upon areas so 

 small, and therefore they are assumed to be moribund. This 

 hypothesis is supposed to be supported by the facts of fossil 

 botany, wliich unquestionably proves that many species have 

 existed in the past and no longer occur in the world to-day. 

 There is, however, nothing to show that the two cases are paral- 

 lel, except in a few instances where there is good evidence that 

 the present existing species once covered a mucli larger area. It 

 was simply assumed that such a species as Colcus clongatus had 

 once occupied more ground. 



Like the previous theory, however, this explanation breaks 

 down when applied to the very striking facts of the distribution 

 of endemics in Ceylon. How can species be dying out in a chain- 

 mail pattern, like the R and RR species given in the diagrams 

 above? And why were there so many more with the smallest 

 areas (VR) than with areas not q\iite so small (R and RR)? Had 

 one arrived in Ceylon just in time to see the disappearance of 

 a considerable flora? Was the dying-out becoming less and less, 

 and if so, why? This graduation of the areas of endemic species 

 from many small to few large was a most difficult point indeed 

 to explain upon this supposition of dying-out, just as it had been 

 for the theory of local adaptation. 



Again, why did so many of the "very rare" endemics choose 

 mountain-tops as a last resort? There were many widely dis- 

 tributed species, with very restricted areas in Ceylon, but these 

 did not choose such places, and why did they not? The "dying- 

 out" explanation supposes endemics to have ascended moun- 

 tains as the last refuge from the invading flora of the plains, but 

 in such a small and imiform country as the wet zone of Ceylon 

 it is hardly possible to suppose that there Avas, for example, a 

 separate Eugenia at every few miles; whilst some of the moun- 

 tains with endemic Eugemas did not even rise directly from the 



