New Zealand Flora , with a Reply to Criticism . 355 
ground, and cannot spread in forest, others are forest herbs and cannot 
spread in open ground. 
Dr. Sinnott argues as if herbs could of themselves, without outside 
assistance of some kind, supersede and replace forest. There is little 
evidence for this,* though the fact that trees can replace herbs is familiar to 
every one who has lived in the midst of forest vegetation. So far as 
I am aware, some extraneous assistance is needed for the reverse to 
happen, such for instance as the operations of man, or a desiccation of 
the climate. 
So far as age and area is concerned, trees, shrubs, and herbs all behave 
in exactly the same manner. Plxcept in the position of the maximum there 
is no difference to be seen in the figures quoted for the genera in Tables V 
and VI ( 19 , p. 446). All show a gradual increase to a maximum, usually in 
the South Island, and a falling away again, or an increase to a maximum at 
the north, as in Pittosporiim. Yet Clematis is shrubby, Ranunculus and 
Lepidium herbaceous, Pittosporum is composed of trees and shrubs, 
Carmichaelia is shrubby, Tillaea herbaceous, and so on. 
Another point that Dr. Sinnott is apt to forget is that trees, shrubs, 
and herbs may come to a country from different sources, so that the tracing 
of their relative ages is rendered still more difficult. In this connexion it is 
worth while to see what can be learnt about New Zealand from a considera¬ 
tion of Tables IV, V, VI of my paper ( 19 ). Though the discovery of the 
facts contained in these tables was the result of a prophecy made by aid of 
age and area, the tables themselves contain nothing but bald facts. One of 
the first points that one notes is that whilst the majority of the families (and 
genera) show figures leading up to a maximum in the south, and falling 
away again (and that quite regularly for every family and genus in the 
flora), a very fair number, e.g. Pittosporaceae or Myrtaceae, commence with 
their maximum to the north, and taper away towards the south. Whatever 
be one’s views as to age and area, it is quite clear from the tables that 
the previous distributional history of these families was different from that 
of the others. Table IV gives only endemic species, but if we add to 
them all the families with northern maxima, we get the result shown in 
Table I, below. 
The first glance shows that all these families are markedly Indo- 
Malayan, though they contain in New Zealand a few genera whose southern 
location indicates a southern derivation, and indeed some of them, like 
Lauretta in Monimiaceae and Donatia in Saxifragaceae, are South American 
genera. But the overwhelming majority are Indo-Malayan. 
