188 



THE MONOTYPIC GENERA 



[PT. II 



again a hollow curve. One may even find the sections of this 

 curve regularly arranged. The families with one genus show 12/1 

 (twelve with one species), 8/2, 6/3, 3/4, 3/5, 2/6-7, 3/10, 3/15, 

 and so on to 290 [Symplocos). This result agrees absolutely with 

 what has been said under Size and Space in Chapter xii. 



There is no demonstrable difference between monotypes and 

 genera of larger size, except in the smaller number of species, 

 and (usually) smaller area occupied. Of the 4853 of them, 1037 

 occur only upon the islands of the world, usually only upon one; 

 887 occur in South America, usually only in small portions of it; 

 835 in Africa, 612 in North America. These four divisions of the 

 world contain in all 3371 monotypes, or 69 per cent, of the total 

 mmiber in the world. But if one count up the number of ditypes 

 they contain, one finds it to be only 59 per cent, (or a much smaller 

 proportion) of those in the world. The number of tritypes con- 

 fined to these portions of the world is only 51 per cent., or a 

 lesser proportion again, that of genera with 5 species 46 per cent., 

 of genera with 10 33 per cent., of genera with 50 20 per cent., of 

 genera with 100 species 10 percent., and of genera with more than 

 100 species they contain only 3 per cent, of those in the world. 

 It is clear that what Avas said above under Size and Space is in 

 general correct, and that the larger genera tend to occupy larger 

 areas in proportion to the number of species that they contain 

 (for the proportions decrease with perfect regularity). Distribu- 

 tion about the Avorld, and number of species, go mainly with 

 Age. It is inconceivable that natural selection should group 

 genera like this. 



This regular curve for the occurrence of genera not only shows 

 with a large number, such as those just considered, but also with 

 much fewer. If we divide the world into continents and larger 

 areas, and enumerate for each region the genera confined to it, 

 we get: 



