440 M. A. DeCandolle on the Characters which distinguish 



we should probably find a still greater proportion of Dicotyle- 

 donous species ; for, without speaking of the secondary families, 

 there are many more Graminese and Cyperacese common to the 

 two extremities of this vast region, than Compositse or Legu- 

 minosse. It is true that, in taking the proportion from very 

 limited Floras, such as those of the environs of towns, we may 

 sometimes find the amount of Dicotyledones nearly as great, or 

 even greater, than in the entire province in which the town is 

 situated* ; but the environs of a town do not usually present 

 all the varieties of station which are indispensable to species, 

 and hence arise accidental causes which prevent the exemplifica- 

 tion of the law. A town surrounded by hills or mountains will 

 have more Dicotyledones, and one environed by moist meadows 

 more Monocotyledones, than the general conditions of the region 

 would lead one to expect. 



Of the two causes of error to which I have just referred, the 

 former, the imperfect knowledge of the Monocotyledones of 

 partially explored countries, is usually the most serious. The 

 second, the unequal extension of the species, is of less import- 

 ance ; and it may, moreover, be got rid of by taking care only 

 to compare countries of nearly similar extent. 



But there are more serious objections to the calculations in 

 question. The Monocotyledones are far from being homo- 

 geneous. What conclusions can be drawn from a number 

 which includes Orchidese or Iridese, Palms, Graminese, Cype- 

 racese or Juncacese, in very different quantities according to the 

 countries, to be afterwards brought into comparison with the 

 Dicotyledones ? Are the thousands of Orchidese, or the hun- 

 dreds of Palms of Brazil, analogous to the Cyperacese or Lili- 

 acese of our regions ? and nevertheless it is to these, under the 

 common name of Monocotyledones, that the Dicotyledones of 

 different countries are compared. The error is still further 

 augmented by the custom of regarding the number of Mono- 

 cotyledones as unity with regard to that of the Dicotyledones ; 

 for this apparent unity varies, and the elements composing it 

 in some regions have the value of plants with a simple, in others 

 with a complex organization ; in one place they are insignificant 

 herbaceous plants, in another woody plants, or even large trees. 

 I may also remark, that the frequency of the individual plants, 

 and their influence on the vegetation of a country, have no rela- 

 tion with the number of species in each group. 



From all these causes, therefore, the proportion of the species 

 of Monocotyledones and Dicotyledones is an abstract fact, 



* Round Ratisbon the proportion is 1 : 3*5 (Fiirnrohr) ; round Vienna, 

 1 : 3-6 (Neilreich); round Strasburg, 1 : 3*4 (Kirschl. in Flora, 1843, 

 vol. i. p. 196) J round Wurtzburg, 1 : 33 (Schenk, Flora, 1849, p. C>]). 



