236 GEOGRAPHICAL DISTRIBUTION [pt. ir 



In the Mixed curves on p, 237, the fourth curve shows the Avhole 

 flora of Ceylon arranged in order, beginning with 573 genera of 

 one species each, and forming the usual hollow curve; the 6th 

 curve shows the flora of Cambridgeshire (Babington), the 9th 

 the flora of Italy. All the floras so far examined give similar 

 results, and the same is the case in local faunas, as the 10th 

 curve (Birds of British India) and the 15th (British Echinoderms) 

 illustrate. The curve is exactly like the curve given by other 

 combinations of animals or plants, as maj^ be seen by comparing 

 them with the other curves in the same figure, e.g. for the Com- 

 positae or the Chrysomelid beetles, the endemics of islands, or 

 those of Brazil. The tails in the figure are of course cut short: 

 their length depends in general upon the size of the flora; the 

 larger it is, the larger size, as a rule, do its genera reach to. 



Or if one take the flora of Ireland, one finds it to be, except 

 for a few Iberian plants in the south-west, a reduced copy of 

 that of Britain, and the Avay in which age alone has been the 

 chief determinant of what species shall occur there is very 

 strikingly shown by the following figures, extracted from ]\Ioore's 

 Cyhele Hibernica. 



The ]3lants of Britain in the Cyhele Britannica are grouped in 

 hundreds according to degree of frequency in Britain {i.e. the 

 number of Watson's vice-counties in which they occur). Of the 

 first hundred all occur in Ireland, of the second and third hun- 

 dreds all, of the fourth 98, fifth 97, sixth 93, seventh 84, eighth 

 74, ninth 63, tenth 66 (the only exception), eleventh 43, twelfth 

 26, thirteenth 16, and fourteenth 8, a steady diminution from 

 top to bottom. 



But if size also depends upon age, then it is clear that in any 

 local flora the genera, Avhich as a rule will not be endemic, should 

 be arranged in the same way. The genera arriving for example 

 in Britain will not all arrive simifltaneously, but some will arrive 

 sooner than others, and these will tend to be the larger genera of 

 the nearest source of supply, for the larger genera will usually be 

 the more widespread. The ultimate result will tend to be that 

 these genera will not only arrive first, but will tend to be repre- 

 sented by more species, so that one will expect the most Avidely 

 distributed species in the large genera {i.e. large for the country 

 in question, being represented there by many species) to be 

 more widely dispersed than those of the small. This we have 

 already seen to be the case in the most striking way (p. 114). 

 But one may push this arithmetical regularity further yet. If 



