34 W. Gibbs on the Constitution of Organic Compounds. 
posed contains two equivalents of free hydrogen, and it would 
therefore appear that lactic acid, even upon my view, should be 
bibasic. Ihave already stated that—as an empirical result— 
when an acid belonging to the type of four equivalents of water 
is derived from an ammonium or its hydrated oxyd, it is only 
the last equivalent of hydrogen which is replaceable by another 
radical to form a salt. ‘There must be a reason for this, and per- 
haps either or both of the following may be satisfactory. : 
the first place it appears probable that in all the salts of 
ammonium the fourth equivalent of hydrogen is differently com- 
bined from the other three, and if this view be correct it is réa- 
sonable to suppose that this peculiarity will exist also in the 
acids which are derived from ammoniums. Again it may be | 
that in glycosin, alanin, &c., and therefore in the acids derived 
from them, the two molecules constituting the aldehyds from 
which these bodies are derived, enter in connection and not sepa- 
rately, as I have all along represented them in the present paper. 
Thus alanin may be 
CiH302.H 
C2 0O+H0. 
} 
. 
| 
. 
In which case the lactic acid derived from it will be - 
‘ C1H302. A t 
OsH } Os, | 
and it is easy to see from this formula that there will be but one _ 
equivalent of replaceable hydrogen in the acid. This latter 
view is perhaps supported by the mode of formation of acetoni¢ 
acid which Stiadeler obtained by digesting acetone with hear t 
dric and chlorhydric acids, and which has the empirical formula 
CeHsOc. Its rational formula upon the above view will be 
CsH302.C2H3 
C2H } 01. 
H 
ae 
My reason for not adopting in the outset the slight modification — 
of the rational formulas just proposed is to be found in the fact 
that in glycosin, alanin and their congeners one equivalent of 
hydrogen—the last but one—is replaceable by an equivalent of 
silver or another metal. If now we suppose that these bodies — 
contain aldehyds, we must admit that it is in each case the hy- — 
drogen molecule of the aldehyd which is replaced by the metal, — 
or that there are compounds like CsH3O2. Ag, which is not sup 
—_ by any experimental evidence. The point is after all of | 
ut little importance, and does not affect the theory in any essen” 
tial particular. The so-called anhydrous lactic acid which has — 
the empirical formula CcoHsOs may be reduced—after Gounlie a 
its equivalent—to the type of siz equivalents of water, and wil 
then have the rational formula & 
