with reference to the Dying Out of Species . 1 7 
Peninsular-Indian species), whose ancestors only came from, say, Hyderabad 
or Poona? If the plants at Tuticorin were in the habit of ranging over 
the whole area occupied by the species this might be conceivable ; but as 
things actually are, this hypothesis, which at present holds the field (though 
sadly injured), involves the bold assumption that the plant living at 
Tuticorin carries in itself the accumulated experience of many different 
climates and conditions that have been passed through by its ancestors. 
And for this we have absolutely no warranty whatever, whilst, as pointed 
out elsewhere, the actual conditions, even for the endemics with the most 
limited area of distribution — an acre or two — are never the same for two 
plants of the same kind, nor for the same plant in two consecutive seasons. 
If we do not make this assumption, we have to make the almost 
equally bold one that the greater distributional areas of some species 
depend upon the fact that they were born with greater ‘adaptability’. 
And here we meet at once with the difficulty of explaining why the species 
endemic to South India as well as Ceylon — often only to a small part 
of South India, e. g. the Nilgiris — should be more adapted to Ceylon than 
those endemic to it only. As Table II shows, 45 out of 492 such species 
have become very common, against only 19 out of 809 of the Ceylon 
species proper. And of the ‘ wides ’ no less than 221 out of 1508 have 
become very common. Such facts would go to show that there was 
nothing to be gained by having local endemic species, which could not 
become so common as those which came from a distance, and would also 
indicate that adaptability was in some way bound up with origin upon 
a large area. The wides are the most, the Ceylon-Indias the next, and 
the Ceylons the least adaptable to the Ceylon conditions. 
If it be assumed that the ‘ wides ’ were originally developed over the 
whole of large areas, we have to explain why such a fact should make 
the descendants of any one of a given species more adaptable to Ceylon, 
especially when we remember that in the case of widely distributed species 
we usually find that there are many local varieties. Does each local 
variety carry in itself the adaptability to spread into new areas with great 
success ? 
The advocates of the old view are faced by a very difficult question 
when the matter is reduced to figures as in my previous paper. Why 
should a species that ranges over Ceylon and Peninsular India be commoner 
in Ceylon than one that only ranges over Ceylon ? Because it has a wider 
range, they reply. And why should one that ranges over a larger area 
be commoner yet? Because it has a still wider range, is the only reply 
they can make. This is simply an appeal to ignorance. They cannot 
suggest any convincing reason why mere wide range should involve greater 
commonness, unless it be simply greater age within the country, as 
I maintain. If mere wide range involve greater commonness, why are so 
c 
