210 



accordingly we can ultimately determine how universally this 

 relation holds good. It is interesting to note that in every case 

 above where both figures are given there is a smaller proportion 

 of monocotyledons among the introduced species than among 

 the natives. 



In applying this statistical method to other regions some 

 cautions must of course be observed. For instance, extreme 

 accuracy could not be expected where the number of species in- 

 volved is much less than a thousand. And it would hardly be 

 advisable to compare areas too widely separated, for the pro- 

 portion of monocotyledons may vary considerably on different 

 continents, or in different climatic zones. 



A similar method applied to different habitats in the same 

 region indicates roughly not the age of the flora of each habitat 

 but its affinities with other regions and its place in the order of 

 succession. In the Altamaha Grit region of Georgia for instance, 

 the flora of river-bluffs, which represent the extreme of meso- 

 phytic conditions for that region and have about 90 per cent, of 

 species in common with the Piedmont region and mountains, 

 contains only 13 per cent, of monocotyledons. On the other 

 hand the moist pine-barrens have only about 20 per cent, of their 

 species ranging beyond the limits of the coastal plain, and 44 

 per cent, of monocotyledons. 



Some other kinds of statistics may perhaps hereafter be found 

 equally useful for the same purposes. For example, the ratio of 

 Gamopetala; to Polypetala^, of grasses to sedges, or of woody 

 plants to herbs. In the glaciated region and coastal plain, sedges 

 seem usually to outnumber the grasses, while the reverse is true 

 in most other parts of the world ; and woody plants tend to be 

 more numerous in old regions than in new, if the climatic condi- 

 tions are not too different. 



