CH. xvii] AND GENERA OF LARGER SIZE 189 



Table showing, in each line, the percentages of genera confined to 

 the Islands, Australia, etc., and containing 1, 2, 3, or other nujnher 

 of species. The percentage is of the total number of genera con- 

 taining 1, or 2, etc., species, not of the total number of genera 

 confined to the islands, etc. 21ms 21 per cent, of all the mono- 

 types are found upon the islands, 49 per cent, of the genera -with 

 75-125 species occur in both Old and New Worlds. 



Percentages of Genera of different sizes (numbers of species) 



100 100 100 100 100 100 100 100 100 



This is a veiy remarkable table. In the case of Islands, Africa, 

 South America, and North America (with a slight exception at 

 the monotypes), the proportions of genera of different sizes 

 decrease regularly (allowing for the limiping of uncertain fours 

 and sixes as fives). This fact seems to me practically to exclude 

 the idea of local adaptation, as well as that of relic nature, for 

 the great bulk of genera, though there must of course be many 

 exceptions to this rule. But if this be so, then the idea that 

 plants have been guided in their evolution by natural selection 

 must also suffer something of an eclipse. One cannot imagine 

 natural selection producing genera in careful graduation of sizes 

 (and areas) like this. One would get distribution almost exactly 

 of this type by the simple operation of the "mechanical" prin- 

 ciple of Age and Area^ as expanded by its corollar}^ Size and Space. 

 If these two worked alone, and absolutely, one would get this 



1 As already several times explained, the general meaning of Age and 

 Area is simply that on averages and in the long run species and genera 

 spread at a more or less uniform rate, interfered with by barriers, physical or 

 ecological. On the older view it was imagined that distribution was do 

 rapid that all forms had already reached their limits, and that many were 

 in process of contracting their area of dispersal. 



