40 Willis. — The Flora of Stewart Island ( New Zealand ) ; 
The Relationship of Stewart Island to Invasions 
of Plants. 
We have now to go on to consider the relationship of Stewart Island 
to the great invasions of plants into New Zealand which were considered in 
my last paper ( 9 , p. 355). And first let us deal with the northern invasion, 
which probably entered, we saw, at some point (or more likely space) in the 
North Island not more than 400 miles south of North Cape (the island is 
320 miles long). 
It is at once evident that the northern invasion was very early, or that 
it found New Zealand at the time of its entry comparatively unobstructed 
with vegetation, for although it had to travel the whole length to Foveaux 
Strait, 1,000 miles south of North Cape, none the less a very fair proportion 
of its species arrived there in time to cross to Stewart before it was too late. 
There reach the strait, of the plants of this invasion, 5 wides and 32 
endemics, and of these 3 wides and 14 endemics cross the strait. Those 
which cross belong, as one would expect from predictions 2, 3, to the largest 
genera and families of the invasion. As a matter of fact they belong to 
Pittosporaceae (1 genus, 19 species), Saxifragaceae (6/8), Myrtaceae (4/18), 
Araliaceae (5/15), Cornaceae (2/5), Lentibulariaceae (1/6), and Urticaceae 
(5/10), a total of 7 families, 24 genera, and 81 species, i. e. an average per 
family of 3-4 genera, and per genus of 3-3 species. The remainder of the 
invasion consists of 2 6 families, with 38 genera and 52 species, averages of 
1*4 genera per family and 1-3 species per genus. 
To go on to the southern invasion, it is at once evident that this was 
either much later in New Zealand or found the ground more troublesome 
to traverse, as although it was mainly herbaceous, and we have some 
reason to suppose that herbs may travel faster than trees, of which the 
northern invasion was chiefly composed, and although it had its centre 
about 400 to 600 miles south of the other invasion, and was therefore so 
much the nearer to Stewart, none the less it is not so well represented there 
as the northern invasion. Only 51 wides out of 91 (at Foveaux Strait) 
occur in Stewart, or 56 per cent., against 3 out of 5, or 60 per cent., of the 
northern wides ; and of the total of plants of the invasion, and its result in 
endemics, only 115 out of 286 cross Foveaux Strait, or 40 per cent., against 
17 out of 37, or 45 per cent. 
Had the southern invasion entered New Zealand by way of Stewart 
Island, one would not expect such a result, but would expect to find more 
species in Stewart than on the northern side of the strait. One may without 
any further discussion, it seems to me, come to the conclusion that the 
southern invasion of plants entered Stewart from the north side of Foveaux 
Strait, and consequently that Stewart did not lie in the direct track of that 
invasion. 
