250 APPENDIX. 



countries, for in a fragment of the ' Flora of Argentina ' Grisebach enumerates 375 



species * 



Whether the other orders collectively have increased in the same ratio, since the 

 publication of the first part of the 'Genera Plantarum' in 1862 up to the present 

 time, as the Composite, is not readily ascertained ; but, judging from a cursory exami- 

 nation of the number of new genera intercalated in a copy of the ' Genera Plantarum' 

 in the Kew Herbarium, and of a number of recent monographs of orders, it is probable 

 that tbey have. To give an example, C. B. Clarke's monograph of the Cyrtandreae has 

 considerably more than doubled the number of species previously known f- Never- 

 theless, it may be assumed that the Compositse still maintain the proportion of at least 

 10 per cent, of the phanerogamic flora of the world. In Mexico and Central America 

 the Composite constitute 13 per cent, of the phanerogams, and they represent 28 per 

 cent, of the genera of the order retained by Bentham and Hooker, and 15-5 per cent, 

 of their computation of the species. There are no available data for comparisons with 

 the whole phanerogamic flora of other parts of America; but thanks to Dr. A. 

 Gray's marvellously sustained energy in descriptive botany, we have now, in the work 

 cited above, all the figures up to date for the Gamopetalse. In America, north of 

 Mexico, the Compositse constitute 42 per cent, of the genera and 47 per cent, of 

 the species of the Gamopetalae, while in Mexico the percentages are respectively 

 33 and 37-8. Turning again to Bentham's essay we find that his total of species for 

 the Old World, excluding Insular Regions, is about 400 in excess of that for America 

 and of genera thirty-three fewer, and his Mediterranean region is also numerically the 

 highest in species ; yet the number of genera is 100 less than in his Mexican region, 

 which, however, does not correspond to ours, but includes the countries to the north 

 from Texas to the Pacific, embracing the southern half of California. As thus circum- 

 scribed, this region undoubtedly possesses the densest aggregation of Composite, both 

 in relation to area and total number of phanerogams. It is true that in many smaller 

 areas in various parts of the world the percentage of Compositse is much higher, 

 Mr. A. DeCandolle $ tabulates the dominant orders for a large number of areas, insular 

 and continental ; most of them, however, are of small extent, and the figures for the 

 larger ones are drawn from collections representing only fragments of the floras. 

 Thus a portion of Hartweg's collections in Mexico and Guatemala yields 16 per 

 cent, of Compositee ; Humboldt and Bonpland's plants from the region of Quito 22 

 per cent.; and Gay's plants from Central and Northern Chili 21 per cent., and 

 so on. In insular floras like that of Juan Fernandez it is even higher. DeCan- 

 dolle makes it 25 ; but we make it 31*4 per cent, of the certainly indigenous 



* Symbol® ad Floram Argentinam. 



t Monographic Phanerogamanim, v. 1883. 



$ Geographie Botanique Raisonnee, ii. pp. 1189-1233. 



