Affinity and Fleat. 245 
chloric acid (this acid arising from a portion of hydro- 
chlorate already decomposed). Everything takes place as 
if the heat necessary for entirely decomposing hydrochlo- 
rate of alumina could never be entirely equal to that which 
extension furnishes to its molecules, within the limits of our 
experiments. 
Thus the phenomenon of dialysis is never complete; a 
little hydrochlorate of alumina always passes through the 
membrane, and the colloids cannot be separated in an abso- 
lute manner by means of diffusion. 
This principle has numerous applications. If we admit 
generally that all solutions are cooled on being diluted—as 
is shown by a great number of experiments—if, moreover, 
we assume, as my experiments seem to show, that in every 
change of state, accompanying solution there is a loss of 
vis viva, and. hence a concentration of latent heat in the 
substances dissolved, and in their menstrua, we can under- 
stand a great number of phenomena, the explanation of 
which escapes us, and bring them within the class of known 
phenomena. 
Thus, the sap of trees is, so to say, pure water at the 
moment it begins to move, and contributes to the increase 
of the plant. It is a solution so diluted with carbonic acid 
and organic or mineral substances drawn from the seed or 
from the soil, that its molecules may be considered abso- 
lutely free, or separated by the latent heat which is there 
accumulated. When this solution reaches the parenchyma 
of the leaves, where both its concentration and its altera- 
_ tion on contact with the elements of the air take place, it 
may be said that all molecular equilibria are successively 
possible; and if circulation raises them to concentration or 
to combination at a given moment, all the elements of car- 
bonic acid, of water, and of the mineral principles which 
the sap contains, may group themselves according to a 
formula determined previously by the rapidity of the circu- 
lation, the nature of the leaves, and the physical circum- 
stances necessary for the life of the plant. In this manner 
we can probably account for the diversity which the act of 
vegetation imprints on the nature of the products it forms 
with the elements of water, of carbonie acid, of ammonia, 
and of some minerals. | 
It is also in the heat stored up by solution that we must 
seek the principal element of the decomposition by vege- 
tables of carbonic acid into carbon and oxygen—a pheno- 
menon regarding which we must confess our total ignorance, 
