IN a paper read at the Linnean Society under the above title on February 2, the statistical methods long employed in “Age and Area” were pushed to their final conclusion. Age and area (review in Ann. of Bot., October, 1921, p. 493) is the name given to a principle gradually discovered in many years of work upon the flora of Ceylon, which, in brief, affirms that if one take groups of not less than ten allied species and compare them with similar groups allied to the first, the relative total areas occupied in a given country, or in the world, will be more or less proportional (whether directly or not we do not yet know) to their relative total ages, within that country or absolutely, as the case may be. The longer a group has existed the more area will it occupy. Tens are compared in order to eliminate chance differences as much as possible, and allied groups to avoid as far as may be the complications introduced by different ecological habit, etc. Herbs, for example, probably spread much more rapidly than trees, but both will obey Age and Area. It is of course obvious that age of itself cannot effect dispersal, but inasmuch as predictions as to distribution of species, occurrence of endemics, etc., can be successfully made upon the basis of age alone, it is clear that the average rate of spreading of a given species, and still more of a group of allied species, is very uniform, and therefore affords a measure of age. The result of the work is to show that in general the species (and genera) of smallest areas are the youngest, and are descended from the more widespread species that usually occur beside them.
h/t Disconnected, fragmented, or united? a trans-disciplinary review of network science by César A. Hidalgo (Applied Network Science | SpringerLink)