186 THE MONOTYPIC GENERA [pt. ii 



are arranged in the same way with regard to the numbers of 

 genera contained in them. There are 54/1 (54 of one genus), 

 45/2-3 (45 of 2 or 3 genera), 40/4-6, 32/7-13, 28/14-23, 25/24-38, 

 22/39-63, 20/64-100, 15/101-200, 13/201-1143. The numbers 

 steadily decrease, while at the same time the number of species 

 included increases, being 1, 2, 3, 7, 10, 15, 25, 37, 100, 943, again 

 forming a hollow curve. 



But if the whole flora of the world show such a remarkable 

 grouping of its genera into sizes, then one will expect the same 

 type of arrangement, in a hollow curve, to hold for the individual 

 families, and in actual fact one finds that this type of grouping 

 into sizes holds for the genera of any single family, with a few 

 trifling variations among the very small families. For example: 



The families Contain 



Acanthaceae (206 gen.) 119/1, 32/2, 20/3, 9/4, 15 5, and so on to 300 



Aceraceae (6) 1/1, 1/3, 1/4, 1/5, and 7 and 115 



Aizoaceae (20) 8/1, 3/2, 1/3, 1/4, 2/5, and so on to 15 



Alismaceae (15) 5/1, 3/2, 3/3, 1/4, and so on to 33 



Amarantaccae (72) 29/1, 10/2, 7/3, 2/4, 2/5, and so on to 100 



Amaryllidaceae (94) 28/1 , 15/2, 10/3, G/4, 3/5, and so on to 100 



Commelinaccae (38) 15/1, 4/2, 3/3, 2/4, 2/5, and so on to 110 



Compositae ( 1 143) 446/1 , 140/2, 97/3, 43/4, 55/5, and so on to 1450 



Coniferae (45) 14/1, 8/2, 2/3, 5/4, 1/5, and so on to 70 



Saxifragaceae (96) 51/1, 12/2, 2/3, 5/4, 1/5, and so on to 225 



Scrophulariaceae (241) 88/1, 32/2, 18/3, 12/4, 8 '5. and so on to 250 



Simanibaceae (39) 17 1, 6/2, 2/3, 2/4, 3/5. and so on to 30 



The whole nimiber of families form similar hollow curves; the 

 Coniferae are one of the most aberrant families of the entire list. 

 As a general rule, the genera with one and tMO species make up 

 about half the total (cf. fig. on p. 187). 



This type of grouping even holds for families of lower type 

 than the flowering plants; for example, the Jungermanniaccae 

 acrogynae show 21/1. 6/2, 9/3, 4/4, Q/S and so on, the Rhodo- 

 melaceae 34/1 , 1 6/2, 5/3, 5/4, 6/5 and so on, the Hymenomycetineae 

 23/1, 10/2, 3/3, 8/4, 3/5, and so on. The numbers are more irregu- 

 lar, but the hollow curve is clearly shown. 



It is clear that this type of distribution of the genera by the 

 number of their contained species is a perfectly general phe- 

 nomenon. There are no exceptions, when allowance is made for 

 the lumping in my Dictionary. If endemic genera, or monotypes, 

 were really mainly relics or special adaptations, such distribution 

 as this would be inconceivable, obtaining as it does in every 

 locality, and agreeing with the distribution of genera about the 



