a Study in Taxonomic Distribution. 41 
Again it happens that (as predicted for the northern invasion) Stewart 
contains chiefly representatives of the larger genera — all the families of the 
southern invasion are to be found. It contains 56 genera of this invasion, 
represented in New Zealand by 451 species, against 5a unrepresented 
genera with 9 1 species. 
We have thus made quite a number of predictions, based solely upon 
the hypothesis of age and area, about the flora of Stewart Island, and every 
one of them has proved to be correct, with a large margin. It is clear that 
it is fairly safe to make predictions from age and area ; in other words, the 
operation of this law is by far the principal factor in determining the actual 
geographical distribution of species about the globe prior to the advent of 
man. Physical barriers of course have a much greater influence in deter- 
mining the actual areas occupied, but their influence is of purely negative 
kind, while age is positive. Local distribution, in a given district, on the 
other hand, is determined by the ecological conditions of that district, but 
unless the ecological boundary (beyond which a species cannot grow) is 
fairly broad and wide, it does not seem to affect seriously the total area 
occupied within the outer limits. 
It must again be made clear, for this is a point on which many of my 
critics fail to understand my position, that the law must not be applied to 
individual species, but only to groups of allied forms. It is a law which 
applies to taxonomic distribution, and its operations can only be clearly 
made out, and disentangled from the many other causes that aid in deter- 
mining geographical distribution, by taking a group of allied species. 
After a lapse of x years a group of 20 herbaceous Compositae will occupy 
an area z ; after a lapse of y years a group of arboreous Dipterocarpaceae 
will occupy the same area z in the same country, but we have no means 
of comparing x and y at present. Each, however, is governed by age 
and area. 
Another point whose misunderstanding seems to cause difficulty for 
some in accepting the law, is the enormous differences between one species 
and another in the area occupied. They seem unable to conceive how this 
can come about without the operation of Natural Selection in reducing the 
area occupied by one of the species. In thinking over the subject they are 
apt to forget that the area occupied per unit time increases as the age of the 
species increases. So long as one only takes the diameter of the area 
occupied, as I have done in New Zealand, a plant when it reaches a diameter 
of, say, 1,000 miles is perhaps twice the age that it was when it reached 500. 
But if the species were spreading every way in untouched continental areas 
of uniform conditions, then, while the diameter increased 
1234567 
2 
