w 
162 On definite Proportions. 
parative state of saturation. And-that both these earthy salts 
are neutral, appears from the sufficiency of the force of cohesion 
in both to prevent the predominant actions of their very powerful 
bases. 
We may therefore, I believe, consider all those earthy and me- 
tallic salts as neutral, in which the acid is united with as much 
of oxygen in the base? as is found in another decidedly neutral 
combination of the same acid with an alkali or an alkaline earth, 
‘Thus, I consider as neutral all those sulphates, in which the base 
contains one-third as much oxygen as the acid ; and all arseniates, 
carbonates, muriates, and phosphates, in which the base con- 
tains half as much oxygen as the acid, Salts in which there is 
more oxygen than this in the acid, I call supersalts; and subsalts, 
those in which there is less. 
In the first series of my experiments on definite proportions, 
I have adduced two examples. of subsulphates. I thought that I 
had found that in the subsulphate of the oxide of iren the acid 
was combined with four times as much of the base as in the 
neutral salt, an opinion which appeared to be confirmed by a su- 
perficial examination of the subsulphate of the oxide of copper. 
But as I obtained more correct ideas on this subject, it appeared 
that the rule for the relation of the oxygen of the acid to that of 
the base by no means agreed with this determination ; for, if the 
result of the analysis had been correct, the oxygen of the acid 
would have been 4 of that of the base, and neither an integral 
multiple nor a submaltiple. I therefore repeated the analyses 
with greater ieedpee 
1. Subsulphate of the Oxide of Iron. 
I dissolved some red oxide of iron in concentrated sulphuric _ 
acid; I heated the mixture until the acid was completely sa- 
turated, then dissolved the salt in water, filtered the solution, 
decomposed it with caustic animonia, but so as not to precipi- 
tate the whole of the oxide, and digested the fluid for 24 hours 
upon the precipitate ; which was theri washed on a filter, as long 
as the presence of any sulphuric acid was indicated by the test 
‘of a salt of baryta. The salt, when well dried, had completely 
the appearance of the common precipitated oxide. When de- 
prived of its water over a spirit lamp, which expelled from it no- 
thing but pure water, it left a red pow der, exactly like the col-= 
cothar of vitriol. Ten grammes of this powder, strongly ignited, 
left behind 7:98 cf oxide of iron, and disengaged sulphurous acid 
during the operation: the remaining oxide, dissolved in muriati¢ 
acid, and tested with a salt of baryta, afforded no perceptible 
trace of sulphuric acid. 
In this subsalt, therefore, 20-2 parts of af tee acid were 
united 
