the Outlying Islands of New Zealand. 333 
only occurs near to Kawhia, on the west coast of the North Island. Cheese- 
man remarks that ‘ the Maoris assert that it sprang from the rollers or skids 
that were brought in the canoe Tainui when they first colonized New 
Zealand ’. It is fairly evident, from the figures, that this legend is quite 
probably correct, and therefore, as this tree only occurs elsewhere in 
Australia and the Chatham Islands , 1 that the Maoris came immediately 
from one or other of these places, perhaps most probably the former, as 
Kawhia is on the west coast. The origin of the Maoris remains a problem, 
but their route is perhaps made a little more clear. 
Summary. 
The distribution in New Zealand of the plants which also reach 
the outlying islands is here dealt with. Starting from my hypothesis of 
age and area, the prediction is made that the most widespread plants in 
New Zealand will be those that reach the islands also, and that those which 
do not reach them will be less widespread. This is confirmed by the facts 
in the most striking way, both in the case of wides and endemics. 
The endemics which reach the islands are more widespread in New 
Zealand than the average of all the wides, a fact impossible of explanation 
by Natural Selection ; the hypothesis of youth and area is placed in 
a difficult position to explain why the younger a species is, the more 
.islands it should reach. The species endemic to the islands only are also 
grouped in a way very difficult of explanation on any other hypothesis than 
age and area — the more wides, the more endemics. Other facts are also 
brought up which render these hypotheses untenable. 
It is shown that submergence may overtake spread, even to the extent 
of killing out a species. 
Incidentally, a strong point in favour of my hypothesis is the almost 
certain way in which it picks out those species where there is doubt about 
identification or nativity. 
1 Eleven ‘wide’ species only occur on the islands without occurring in New Zealand, and of 
these four are Kerguelen species occurring only on the southern islands, and six are tropical forms 
only found in the Kermadecs. One only, Lcucopogon Richei , occurs in the Chathams and 
Australia, and may be classed with Pomaderris as a case for which there must be some special 
explanation. 
A a 2 
