( 10 5 
means’ of their leaves,)(1) the moisture nécessary 
for their well being.. Du Hatnel (and after him 
‘Sennebier) has shown, that the filaments that sur- 
round the roots of plants, and which have been call- 
ed their hair, perform for them in the earth the office 
that leaves perform in the atmosphere, and that if 
deprived of these filaments the plants die. 
‘Tt would be easy, but useless, to multiply facts. 
of this kind, tending to establish a doctrine not con- 
test-d, but which after all does not assert, that wa- 
ter makes any part of the food of plants. On this 
point two opinions exist—tbe one, that this liquid 
is a solvent and conductor of alimentary juices; 
the other, that it is itself an aliment and purveyor 
of vegetable fond at the same time. The first opin- 
ion is abundantly established. Water, when charg- 
ed with oxigen, supplies to germinating seeds the 
want of atmospheric air, and saturated with animal 
or vegetable matter in a state of decomposition, or 
slightly impregnated with carbonic acid, very per- 
ceptibly quickens and invigorates vegetation. The 
second opinion is favored by some of De Saus- 
sure’s experiments. On these, Chaptal makes’ the 
following remark, which expresses very distinctly 
an approbation of the doctrine they suggest: * : he 
enormous quantity of hydrogen (which makes so 
large a part of vegetable matter) cannot be ac- 
counted for, but by admitting (in the process of 
vegetation) the decomposition of water, of which 
(1) Bonnet's experiments shew, that it is the under surface of the leaf that per- 
forms this function. The upper surface has a different office, 
