CH. xxii] GEOGRAPHICAL DISTRIBUTION 235 



families of which it is composed (fig. on p. 187). A very 

 httle examination of local floras suffices to show that this fs 

 indeed the case. 



If, for example, one take the fiora of Britain (37), one finds 

 that the families, by numbers of genera, are arranged in regular 

 order, diminishing as the number of genera increases, thusTss/l 

 (33 of 1 genus), 17/2, 9/3, 6/4, 3/5, 2/6, 2/7, 2/8, 1/9, 2/10, and 

 so on in scattered numbers to 46. The genera by numbers of 

 their species in Britain are 223/1, 90/2, 35/3, 32/4, 16/5, 15/6, 

 and so on. Until the numbers become small there is no break in 

 the regularity. The first two or three numbers contain the great 

 bulk of the total; 50 families out of 92 contain one or two genera, 

 and 313 genera out of 512 contain one or two species. This will 

 be found upon examination to be a general rule for all floras. In 

 New Zealand, for example, one finds the genera (total 329) to 

 be 155/1, 54/2, 29/3, 17/4, 12/5, 11/6, 11/7, 5/8. 5/9, 4/10, and so 

 on. In Ceylon (total 1027) one finds 573/1, 176/2, 85/3, 49/4, 

 36/5, 20/6, 19/7, and so on. In Vol. i (only) of the Flora of 

 British India one finds 173/1, 70/2, 33/3, 19/5, 7/10, and so on. 

 All form markedly hollow curves, with the great bulk of the 

 genera in the first tAvo figures, so that there is a very steep drop 

 until the third or fourth figure is reached, and then a gradual 

 tapering away to the larger genera. The larger the country, on 

 the whole, the larger the size of the biggest genera. 



One may push this type of distribution, shown in the hollow 

 curve, into yet more detail, and find that not only the whole 

 local flora of a country, say, for example, Britain, shoAvs this 

 curve, but also portions of that flora. The same curve is shoAvn 

 by the Monocotyledons and Dicotyledons of the British flora, 

 and even by the individual families, Avhcn of reasonable size, the 

 grasses for instance shoAving 24/1, 13/2, 1/3, 4/4, 1/5, 2/6, and 8, 

 11, 13. The line is Avavy, but the numbers are small, and there is 

 no doubt about the shape of the curve. 



Or one may take portions of the country inhabited by more 

 or less definite associations, or groups of associations, of plants, 

 and find the same thing. Thus if avc take Cambridgeshire, the 

 Wisbech division of the county (fen), and the very local Wickcn 

 fen, from Babington's Flora, of Cambridgeshire, Ave get the same 

 type of curves (cf. curve 6 in fig. on p. 237). One might 

 expect certain genera to prove unusually suitable, and to be 

 disproportionately represented, but this does not seem to be the 

 case. 



