284 Willi Sr — The Floras of the Outlying Islands of 
both invasions, and their species divided between them, the slight irregu- 
larities shown in the tables above would be removed. 
(26) The southern invasion of Monocotyledons being thus much larger 
than the northern, we shall expect the Aucklands to show a larger percent- 
age of this group in their flora than do the Kermadecs, and we shall 
expect the Chathams to show an intermediate percentage. 
Table XIX. 
A ucklands. Chathams. Kermadecs. 
Monocotyledons 54 45 % 49 31 % 15 21 % 
Dicotyledons 65 55 % 106 69 % 56 79 % 
(27) It follows incidentally from what has been said that Stewart 
Island should show a greater proportion of Monocotyledons in its flora than 
New Zealand as a whole. This comes out in 
\ . 
Table XX. 
New Zealand as a whole 
( including islands') . 
Monocotyledons 348 27 % 
Dicotyledons 954 73 % 
or, to put it in another way, 37 per cent of the New Zealand Mono- 
cotyledon flora occur in Stewart, and only 26 per cent, of the Dicotyledon 
flora (neglecting in each case the few Stewart endemics). The whole makes 
a very awkward problem for the supporter of Natural Selection ; why are 
Monocotyledons so much better suited to Stewart and the Aucklands than 
to the Kermadecs, and why are the Chathams intermediate in the propor- 
tion they bear? But if we simply recognize that the proportions are merely 
due to the size and time of the invasions of New Zealand by plants, we get 
a perfectly simple and straightforward explanation. 
(28) Examination of map-diagram 2 will show that if we draw a circle 
with its centre in the Auckland Islands, passing through the Chathams, it will 
also pass through Auckland City. We shall therefore expect that in 
general (unless of course there was not direct land communication in either 
of these directions), species which reach both the Aucklands and the 
Chathams will also reach Auckland City. Examination of the facts shows 
that of the 32 species that these groups of islands have in common, 2 6 are 
found at least as far north in New Zealand as Auckland City. Of the 
other 6, 2 are wides, both South American species, Tillaea moschata and 
Veronica elliptica , reaching, the one to the northern side of Cook’s Strait, 
the other to West Wanganui and Cape Foulwind, in South Island. 
Coprosma foetidissima reaches to the Thames goldfields, and Deschampsia 
caespitosa to the lower Waikato, so that these two reach within a very 
short distance of Auckland. Tillaea moschata and Rumex neglectus are 
Stewart. 
I 3° 34% 
252 66% 
