Plants in their Relation to Others . 
503 
view that endemics were moribunds — at least in any serious proportion — 
but it is only during the last seven years that I have been able to bring 
proofs for what seems to me a more satisfactory explanation, viz. that they 
represent different early stages on the way upwards , not downwards. One 
cannot imagine species dying out in the regularly graduated way shown in 
the map of Ranunculus above, but if one imagine that these are new species 
formed in successive order, and just in the early stages of dispersal, one 
arrives at a perfectly simple explanation, which completely fits the facts so 
far as we know them. 
In view of the incisive figures which have been set forth in recent 
papers, a number of botanists are now beginning to admit that endemic 
species of small area of distribution are really young species, or at least have 
been formed in situ , but few are as yet prepared to admit this for endemic 
genera, and most people continue to regard them as survivals. 
If my view be the correct one to take, then it is clear that there is no 
difference between endemic genera and species, and others that occupy 
larger areas, except that in general they are younger ; so that while one 
may call Coleus elongatus endemic to the summit of Ritigala in Ceylon, or 
the genus Carpodetus endemic to New Zealand (it has one species, reaching 
to both ends of New Zealand), one may with equal justice talk about 
Hottonia palustris as endemic to temperate Europe and Asia, or the genus 
Senecio as endemic to the world. My aim in the present paper is to show 
that this is the more correct view to take, and that there is no appreciable 
difference but age. 
So firmly has the old view, that endemics are relics of old floras, held 
sway, that it never seems to have occurred to any botanist to try the simple 
test whose results I am about to set forth, and yet this test might have been 
made at any time in the last thirty years. Although it was suggested by 
a consideration of Age and Area, the results here set forth are simple facts 
which may be considered purely on their merits, without reference to any 
hypothesis whatsoever. 
One may test this question, whether endemic genera are or are 
not in general relics, in a very simple way, by making a prediction which 
shall at the same time be a crucial test, inasmuch as the result must be 
different on the two hypotheses, and then verifying it by comparison with 
the actual fact. The prediction, however, being as to what, in a broad 
general way, will be found upon the islands of the world, involves various 
complications. 
In the first place, if the islands have received their floras by casual 
oversea transport, it is clearly all but impossible to predict what will be 
found upon them. If a successful prediction is made, therefore, the fact 
speaks very strongly in favour of their having (in the mass) received their 
floras by land, not necessarily continuous, but with no very large gaps. If 
