THE AGE AND AREA LAW 443 



ABCD (supposing the rate of immigration to have been more 

 or less constant) and the number of species occupying inter- 

 mediate areas will be intermediate also, forming a decreasing 

 series from ABCD or 6 unit areas to AabD or 1 unit area. This 

 has been found to be the case for both Ceylon and New 

 Zealand. 



Considering now the case of the endemic species, there is 

 no reason to suppose that new species have ceased to be pro- 

 duced so that we can postulate a relatively constant supply. 

 If the laws of chance govern the origin of species we can 

 picture the production of new species from an immigrant at 

 a constant rate in the following way : while occupying area, 

 it gives a species, a, while occupying areas 1 and 2 it gives 

 a species, £, in area 1 and another, 7, in area 2, and so on, 

 until, while occupying the areas 1-6, it gives 6 species. Species 

 a will meantime have spread (possibly over areas 1-5) and 

 so will /3 and 7 and other intermediate species to a smaller 

 degree. There will, therefore, be one endemic species occupy- 

 ing 5 unit areas, two occupying 4 unit areas, three occupying 

 3 unit areas, four occupying 2 unit areas, and five occupying 

 1 unit area, while six will be just beginning to spread. From 

 this it will be seen that the number of endemic species occupy- 

 ing a small area is naturally larger than the number occupying 

 a large area, while the intermediate areas possess intermediate 

 numbers of endemic species forming an increasing series from 

 5 unit areas to 1 unit area. 1 This is exactly the opposite of 

 what occurs with the non-endemic species (see above) and the 

 gradation in opposite directions of the numbers of species 

 occupying the various sizes of areas is exactly what Willis 

 finds by comparing the endemic and non-endemic floras of 

 Ceylon and of New Zealand. 



Among the many predictions made and confirmed by 

 Willis about the flora of New Zealand were three which may 

 be given thus : if the area ABCD was separated from the 

 source of its non-endemic flora at a much earlier period than 

 a similar area A'B'C'D', then the average area occupied by 

 the species in ABCD would be greater than the average area 

 occupied by those in A'B'C'D'. This would be the result of 

 more non-endemic species having had time to spread and occupy 

 most of the areas 1 to 6 in ABCD, while new species would 

 1 The actual area might be AabD, a&dc, cdfe, ejhg, or ghlk. 



