26 Willis . — The Flora of Stewart Island (New Zealand) : 
many, but perhaps 400 or less. In actual fact the flora is 31 1, to which 
must be added 57 ferns. (Cf. Appendix, p. 42.) 
In my last paper we saw that New Zealand probably received the 
bulk of its flora by two great invasions, one northern, within 400 miles 
from the north end of North Island, and the other southern, with its centre 
somewhere a little south of the middle of South Island. If we assume, for 
the sake of simplicity, that Auckland and Dunedin were the actual centres 
of these invasions, 1 and draw circles round them, as done in the adjoining 
map, then it is clear that the older plants will be near the boundaries of the 
outer circles, in so far as actual barriers of sea have not interfered with 
their distribution, and the younger successively nearer to the centres. It 
therefore follows (this thesis 
will be more fully developed in 
a later paper on the outlying 
islands) that the Chatham Is- 
lands should have a good deal 
in common with the Kermadecs 
of the older families, genera, and 
species, which can have reached 
both, and similarly much in 
common with the Aucklands, 
while of families, genera, and 
species that are sufficiently old 
in New Zealand all three groups 
of islands should possess repre- 
sentatives. The circle for the 
northern invasion that covers 
the Kermadecs and Chathams 
does not include Stewart, while 
that for the southern invasion 
which covers the Chathams and 
Aucklands, or the Chathams and Campbell, does. 
The general result of this prediction, then, so far as concerns Stewart, 
is that one will expect to find there all families, genera, and species that 
occur in the Kermadecs, Chathams, and Aucklands, or in the last two only. 
But as Stewart is nearer to the centres, one may expect to find there also 
a good many that have not reached the Kermadecs and Aucklands, or the 
Chathams and Aucklands, or even any of the outlying islands. 
The Chathams, in their intermediate position, will probably contain 
a good many plants which do not reach the Aucklands. A straight line 
drawn from them to New Zealand reaches it at a point which is much 
1 Examination of the map will show that displacement of the centres of the circles by 100 miles 
or so makes no difference to the arguments in this paper. 
New Zealand and outlying islands. The dotted 
line is the 1,000 fathom limit. 
