442 SCIENCE PROGRESS 



spread and occupied the area ABCD if a sufficiently long 

 interval had elapsed between the time of their arrival and 

 the time of the analysis of the flora, and if no great barrier 

 such as a mountain-range or great river intersected the area. 

 Similarly, the second group of arrivals would extend over the 

 area AklD and so on. The most recent arrivals would natur- 

 ally occupy the area AabD. Now, since the several areas all 

 overlap in the area AabD, that region should have the greatest 

 number of non-endemic species, i.e. species which also occur 

 outside ABCD, and the numbers should decrease regularly to 

 a minimum in kBCl. 



If, instead of arriving from outside regions, the species 

 were evolved within the region, not by a differentiating cause 

 such as natural selection but by chance mutations, then, since 

 the greatest number of species occurs m AabD, there will be 

 the greatest chance of species arising in this region. If the 

 laws of chance govern the origin of species the number of 

 species which are endemic, i.e. confined to ABCD, will be 

 greatest in the area AabD, and the numbers should decrease 

 regularly to a minimum in kBCl. New Zealand is an area 

 somewhat similar to ABCD, and Willis has proved that the 

 two predictions in italics hold good for the distribution of 

 species in these islands. For the sake of simplicity, the point 

 of arrival has been placed at the end, instead of somewhere 

 in the region df, as it is in New Zealand. It appears probable, 

 therefore, that the laws of chance do actually govern the 

 origin of species, and that the species spread more or less 

 mechanically as a rule. 



Considering further the number of non-endemic species 

 occupying the six areas, i, 1-2, 1-3, 1-4, 1-5, 1-6, on the age 

 and area hypothesis a species which occupies only area 1 or 

 AabD is a more recent arrival than one which occupies the 

 whole area ABCD, and the other species occupying the areas 

 intermediate in size are intermediate in age (in the country). 

 If the region ABCD was connected with the source of its non- 

 endemic flora a long time ago (geologically), many of the 

 species will have had time to occupy the whole area, and the 

 number of non-endemic species limited to the area AabD 1 will 

 be smaller than that of the species occupying the whole area 



1 This is to be distinguished carefully from the total number of species in the 

 area AabD. 



