114 
colour forms in leaves, but the rapidity with which 
it is effected by the direct action of the sun, shew it 
to be accomplished by a power which acts locally and 
exteriorly to the plant ; and whose operations are car- 
ried on in a manner distinct from those which pro- 
perly constitute vegetation. These facts, therefore, 
supply another coincidence between the decomposi- 
tion of carbonic acid, and the formation of the green 
colour in plants, and still farther confirm the imme- 
diate relation that appears to subsist betwixt them. 
357. To account for the green colours of vegeta- 
bles, various hypotheses have been proposed. M. 
Humboldt, having observed some growing plants to 
retain their green colour in mines, from which light 
was excluded, and in which the atmosphere was im- 
pregnated with hydrogen gas, was led to ascribe the 
production of the green colour to the operation of 
hydrogen *. It is evident, however, that plants be- 
come green in the common atmosphere, where little 
or no hydrogen gas is present j and M. de Saussure, 
in various trials, could never observe that plants be- 
came more green from the languid vegetation which 
they experienced in vessels of hydrogen gas f. 
358. M. Humboldt farther attributed the white 
condition of the leaf to the operation of oxygen; 
and we have seen, that the leaf becomes green only 
when oxygen gas is expelled from it. But the non- 
expulsion of oxygen is no proof that the white co- 
lour is caused by the operation of that substance j 
* Thomson's Chemistry, vol. v. p. 362. 4th edit. 
t Recherches, p. 210. 
