35S Willis.—-The Sources and Distribution of the 
The curves for the species of this northern invasion give one to imagine 
that the point of entry was comparatively narrow, perhaps not very much 
wider than one or two of the zones of 100 miles into which I have divided 
the islands. As New Zealand is mountainous throughout, it is possible that 
birds, which would be likely usually to land in the same hills, might have 
brought the plants, but many of them have seeds very much unsuited to bird 
transport. It would therefore seem probable that they must have arrived 
by some land bridge, and the same applies to the remainder of the flora, 
which shows a southern maximum. This is not denying that casual trans¬ 
port may occur, and indeed there are cases, like Ipomoea palmata on the 
northern coast, whose limited or peculiar distribution is more readily 
explained in this way than by a land bridge. As after 300 miles from 
North Cape both wides and en¬ 
demics begin markedly to fall off, 
it would appear probable that the 
bridge reached New Zealand some¬ 
where within the first 300, or at 
most 400 miles from the north. 
The next question naturally is, was 
it a bridge by way of the Kermadec 
Islands (see map) ? When one comes 
to look into this, one finds that only 
11 genera (representing 10 families) 
out of 60 in these 33 families occur 
in the Kermadecs, and of these it 
is practically certain that some, 
e. g. Sicyos, are genera which really 
entered by way of those islands. 1 It 
would therefore seem probable that 
the connexion by wfliich these families entered was not that way. But as 
they are all represented in tropical Australia, which is part of Indo-Malaya, 
it would seem likely that one of the two strips of shallow water shown in 
the map as running down from NE. Australia represents the remains of the 
bridge by whose means they arrived. 
Passing on now to the constitutional habit of these plants, which is given 
in the table, one notices at once that they are nearly all shrubs and trees, 
only 20 out of 130 being herbs. The proportion of trees and shrubs 
is no less than 84 per cent. And of the no no fewer than 4 6 are trees, 
out of a total of 72 trees in New Zealand. If we add the other families 
of probable northern origin, we get a total of 48 trees. In any case it 
is clear that the bulk of the trees in New Zealand arrived from the 
north, or were evolved from or in genera that arrived in this way. 
1 The question of the union of New Zealand and the islands will be discussed in later papers. 
k'LorcLHowelS p 
KermxxxLec !<? £ / 
/ / 
l ' 
S'be^vapt I.j& 
/ 'Snares 
oAiirkLaruL I. 
CamphelL 1 ^, ■ 
,-Boynty I? 
''Antipodes I 
New Zealand and outlying islands. The dotted 
line is the 1,000 fathom limit. 
