BOOK V. 
GEOGRAPHY. 
503 
with Barbary plants ; the seeds of which are known to have 
been brought across the Mediterranean along with the Bar- 
bary wool, which is disembarked at that station. In like 
manner the various kinds of corn have been carried about from 
country to country for the service of mankind, until their real 
home has become doubtful. Medicago sativa is common in 
Chili, whither it has been transported by the Spaniards ; and 
instances in abundance of similar cases could be produced. 
But it must not thence be inferred, that all cases of species 
growing in places far away from their kindred forms, are to 
be referred to migration : for this, the agency of man, of 
animals, of seas, of wind, and of torrents, will doubtless have 
done a great deal; but none of these causes, nor any other with 
which I am acquainted, will explain the identity of the Calypso 
borealis. Orchis viridis, and Betula nana of North America 
and of Europe ; of the Potamogetons common to Europe and 
New Holland; of the Rose, already adverted to, as common 
to North America and China ; of the Osmorhiza of the 
Himalayas,with that of the United States ; of the wide diffusion 
of Samolus Valerandi ; and, most especially, of the identity of 
the cryptogamic plants of various countries, plants incapable of 
cultivation, unconnected with the purposes of man, and of all 
others, the most difficult of transport under any form. To us 
it appears that such plants must have been originally created 
in the places where they now exist ; the contingent circum- 
stances under which they were found having been favour- 
able to the particular mode of vegetable developement whi<jh 
was necessary for their formation. 
One rather important element in all calculations concern- 
ing the geographical distribution of plants is the actual 
number of species upon the surface of the earth. In the ex- 
isting state of herbaria, and with so many large districts 
either altogether or very imperfectly examined, there is no 
possibility of arriving at any thing more than an approxi- 
mation to the true number ; and even this may prove so very 
wide of the truth as to be really exceedingly fallacious. 
Nevertheless, some idea of it may be formed from the follow- 
ing data and conjectures : — 
K K 4 
