35 
a Study in Taxonomic Distribution. 
Of the 31 1 species of the flora of Stewart, 206, or 66 per cent., belong 
to the comparatively few genera that possess in New Zealand 5 or more 
species in the genus, the average number in a genus in New Zealand 
being 4-2. 
On the hypothesis of Natural Selection, as usually interpreted, one 
would expect to find Stewart Island — an outlying island, small in area, 
simple in geological structure, with small flora and peopled by old species — 
occupied largely by plants unsuited to life in the crowded South Island, in 
fact a kind of refuge for the destitute. This expectation is vehemently con- 
tradicted by what we have just indicated. The genera of Stewart, as 
a glance at the list of the flora will show, are in reality what it has been the 
custom to call the ‘ successful ’ genera of the neighbouring island, from which 
it must have received its flora. These genera are in reality simply those 
which were the first to arrive of their various affinity groups. The first 10 
genera of the Stewart flora are represented in New Zealand by 103 species, 
or an average of 10-3 species per genus, against an average for New Zealand 
of 4-2. The last 10 have 70 species in New Zealand, or an average of 7-0. 
The 154 genera of Stewart contain in New Zealand 1,067 species, against 
325 for the 175 unrepresented genera. 
(4) The Stewart Island plants being old, we shall expect to find 
‘wides’, which are the oldest forms, best represented among them. 
Table VII. 
r ir i . Endemic to Neiv Zealand 
/! or A r ew Zealand and Islands. 
New Zealand as a whole 301 1,000 
Stewart Island 1 1 1, or 36 % 187 \ or 18 % 
(5) One will expect to find the Stewart Island plants, as very old, very 
widespread in New Zealand : 
Table VIII. 
Endemic to 
Endemic to 
Class. 
Range in Neiv Zealand, 
Wides. 
New Zealand 
A T ew Zealand 
and Islands. 
only. 
1 
1,001-1,080 miles 
641-1,000 
81 
37 
48 
2-4 
23 
18 
40 
5-10 
1-640 
7 
8 
36 
hi 
63 
124 
Thus 166 plants, or 55 per cent., belong to the first class, which ranges 
New Zealand from end to end. Even the species endemic to New Zealand 
alone show more in the first class than in classes 2, 3, and 4 put together, 
though in New Zealand as a whole the numbers in the first four classes are, 
52, 60, 59, and 61. 
If we calculate out the rarity in the usual manner, we find that while in 
1 The remaining thirteen include the ten local endemics, and Aralia Lyallii, Urtica australis , 
and Poa foliosa , not found in New Zealand proper, and perhaps, or probably, water-borne. 
