190 Willis . — The Relative Age of Endemic 
Mr. Ridley then goes on to suggest that the figures depend upon the 
number of specimens in the herbarium, which, if it were literally true in 
detail, and my Ceylon work were not supported by that upon New Zealand, 
would be a very damaging criticism. I was careful before publishing, 
however, to verify that Trimen’s figures ( 4 ) did, as a matter of fact, in most 
cases, indicate roughly the area occupied. He does not count duplicates 
from the same place, and goes, to the extent of perhaps 80 per cent., upon 
the results already obtained by Thwaites ( 3 ), who did the lion’s share of the 
investigation of the Ceylon flora, and that chiefly in the days before there 
was such extensive clearance for planting. Mr. Ridley complains that 
investigation stopped with Trimen, but if he will refer to my annual report 
for 1910 he will find an enumeration of the herbarium, showing that while 
Thwaites added 6,492 sheets, Trimen added 2,890, and I myself 1,603. 
But with all the additions since Thwaites’s time, the range in Ceylon of the 
various species has been comparatively little extended. The serious 
additions to localities made in my time are summed up in a paper by 
Mr. A. M. Smith and myself ( 19 ), but these were completely ignored in my 
papers under discussion, which are based solely on Trimen’s flora. Taking 
36 consecutive species at random from Thwaites, and giving them marks 
according to area as in Trimen, I found that they received 115 marks in the 
former, 100 in the latter case, and this I think fairly enough represents the 
general difference that Trimen’s work made in regard to the geographical 
distribution of the species of Ceylon ; 13 of the 36 are moved one place up 
by Trimen, and 12 remain the same ; 6 go down a place, and the remaining 
5 make larger moves. 
That Trimen’s figures do as a matter of fact very fairly indicate area is 
shown by the maps given on p. 12 (6) ; about 10-15 per cent, of the VR, 
5-10 of the R, and smaller proportions of the higher classes are in my 
opinion wrongly placed, and should be moved up, usually one class. But 
this movement would make no difference in the figures, as may be easily 
shown. Let us move up 15 per cent, of the VR species, then 10 per cent, 
of the increased R’s (to allow for the few VR’s that should go more than 
a class upwards), then 5 per cent, of the RR’s, and 2 per cent, of the RC’s. 
The result is to replace the table on p. 3 (6) by the following: 
Table I. 
Endemic . 
Ceylon - Penins . India . 
Wide . 
VC. 
19 
45 
221 
c. 
93 
120 
468 
RC. 
144 
105 
318 
RR. 
162 
86 
316 
R. 
191 
68 
163 
VR. 
198 
68 
122 
Rarity 
4-2 
3*4 
3.0 
