494 
GEOGRAPHY. 
BOOK V. 
the same proportions as the number of species of a given 
family. 
" If from species or individuals of the same form, which re- 
produce themselves in conformity to certain fixed laws, we 
pass to those divisions of the natural system which are ab- 
stractions of different degrees of importance, we may either 
confine ourselves to genera, or orders, or sections of a still 
higher degree. There are certain genera and families which 
belong exclusively to certain zones, and a particular com- 
bination of the conditions of climate; but there is also a great 
number of genera and families, of which we find represent- 
atives under all zones and at all elevations. The earliest 
researches upon the geographical distribution of forms were 
those of M. Treviranus, published in his ingenious work on 
Biology (vol. ii. pp. 47. 63. 83. 129.), and the object of these 
was the stations of genera upon the globe. But it is more 
difficult to obtain general results from such a method than 
from that which compares the number of species of each 
family, or the great groups of a particular family, to the whole 
mass of phaenogamous plants. In the frozen zone, the variety 
of genuine forms does not diminish in any thing like the 
degree of decrement of species ; a greater number of genera, in 
a given number of species, is always to be found in such 
countries : and so it also is with the summits of high moun- 
tains, which are colonised by a great number of genera sup- 
plied by the more abundant vegetation of the plains. 
" It is very instructive to study the vegetation of the 
tropics and of the temperate zone between the parallels of 
40° and 50°, in two different ways : firstly, in determining the 
numerical properties of the flora of a large extent of country, 
including both mountains and plains ; and, secondly, in ascer- 
taining those proportions for the plains only of the temperate 
and torrid zones. As in our herbaria we have indicated, by 
barometrical measurement, the elevation of each plant in more 
than 4000 cases above the level of the sea in equinoctial Ame- 
rica, it will be easy, when the account of the species is 
completed (it is now completed), to separate those which 
grow at or above an elevation of 6000 feet from such as are 
inhabitants of a lower region. This operation will affect 
