45 6 Willis . — The Distribution of Species in New Zealand. 
bution of species, so far as the area which they cover is concerned, the pre- 
sumption in favour of my hypothesis is very greatly increased. There would 
at least seem to be fair reason to accept it provisionally, and test it, as the 
theory of evolution itself, and many other theories, have been tested, by 
applying it to the solution of other problems. 
Finally, it may be pointed out that the facts of the flora of New Zealand 
agree so absolutely with those of that of Ceylon, that the whole argument 
against natural selection given in my last paper (pp. 6-16) might be repeated 
word for word with the necessary illustrations drawn from the New Zealand 
flora, as also might the argument against dying out of species other than by 
a geological convulsion, serious change of climate, or other important cause. 
There is no evidence to show that they die out under normal conditions. 
Even if my hypothesis of ‘ age and area ’ be not admitted, there is no 
denying the fact that the species living in any particular country have been 
shown to follow definite numerical laws in their distribution. It therefore 
almost of necessity follows that that distribution — so far as area covered 
is concerned — is determined by causes which act mechanically upon all alike. 
Such causes are not numerous, and if age, which is at least an obvious one, 
be rejected, it will not be found easy to discover a substitute. 
Geographical distribution is, it seems to me, removed from the realm of 
evolution proper, which is thrown back a stage. It is not a phenomenon of 
evolution, but a sequel to it. Once a species is evolved, its further history, 
i. e. its extension in numbers and in area occupied, appears to be, chiefly at 
any rate, a mechanical process, with which, so far as we can see, its past 
evolution has nothing to do. How it was evolved, or why, are other questions 
altogether. At the same time, as there seems to be an impression that this 
work is an attempt to destroy the theory of evolution, it will be well to point 
out that in reality, by showing that the species of a genus and the genera of 
a family are connected together by definite arithmetical relationships, it 
affords very great support to that theory, if support were needed for a theory 
established beyond cavil. 
In discovering the localities given for many species, I have been very 
greatly assisted by H. E. the High Commissioner for New Zealand, to whom 
and to the staff of whose office I desire to express my most grateful thanks. 
Summary. 
The flora of New Zealand is studied in this paper from the point 
of view of my hypothesis, that the geographical distribution of a species 
(i. e. the area which it includes within its outer localities) within a fairly 
uniform country not broken by serious barriers depends upon the age 
