i9 
with reference to the Dying Out of Species. 
These figures go to show that the Dicotyledonous water plants are very old 
in Ceylon, as indeed one would expect, while the Monocotyledonous 
species are much more recent, even than the average of the whole flora. 
We have now to go on with the calculations from these figures, to 
determine whether any other information of value may be extracted from 
them, and first we shall deal with the question whether they show any 
indication that any of the angiospermous species of the Ceylon flora are 
dying out. The theory that about as many species are dying out as are in 
course of extension is still more or less implicitly held as an article of faith. 
We may refer to Table II, at the beginning of this paper, which gives 
the actual numbers of species under each of the different heads of rarity, 
with the percentages. 
As there exist in Ceylon 455 VR species, which are all confined to 
very small areas, often one single mountain top, it is evident that for the 
present the best classification of the flora is into six approximately equal 
groups, each containing about one-sixth of the total, or 468 species. But 
whether this grouping would always remain the same for all future time 
is another question, to which the observed facts give a clearly negative 
answer. The oldest group of species in the flora — those of wide distribution 
— show numbers increasing upwards in the scale ; the next oldest show 
numbers fairly evenly divided along it, and the youngest group, the Ceylon 
endemics, show numbers increasing downwards. But it must be remem- 
bered that the species of wide and of Ceylon-Indian distribution were cut 
off at a certain period, subsequent to which no more could enter the island. 
On the other hand, so far as we know, there is nothing to prevent new 
endemics forming to-morrow, and the number of endemics would in all 
likelihood increase with the increasing number of the other species ; hence 
the table with numbers increasing downwards. The view which I take 
of the history of the first two groups is that the numbers that first entered 
Ceylon would be small, and would increase as time went on and the 
number of species on the Indian peninsula increased. Then as Ceylon 
began to be cut off, the communicating land would narrow, and the 
number of species arriving across it would gradually decrease. This may 
be very roughly indeed represented by the following table (XIII). Having 
no knowledge of how rapidly the proportion of species crossing to Ceylon 
increased, of how rapidly the number of Ceylon-Indian forms grew, nor 
of how rapidly the communication was cut off ; and remembering that the 
early arrivals probably went more quickly up the scale than the later, 
and that there were probably such differences in adaptation, and in the 
chances that befell, that some would never rise above some definite level 
in the scale short of VC, it is evident that much greater mathematical 
skill than I possess is needed to construct any sort of diagram approaching 
accuracy, if indeed such were at all possible in the present very limited 
state of our knowledge as to the past history of Ceylon and its vegetation. 
C 3 
