On definite Proportions. 47 
bases, and salts; secondly, Laws for the formation of subsalts ; 
thirdly, Laws for the formation of double salts; and fourthly, 
General view of the results of my experiments. 
I. Laws FoR THE CoMBINATIONS OF WATER. . 
A. Combinations of Water with Acids. 
1. Tartaric Acid. 
Ten grammes of finely powdered dry tartari¢ acid were dis- 
solved in water, and heated with a solution of acetate of the prot- 
oxide of lead, as long as any precipitate was formed. The 
“mixture was evaporated to dryness, and the uncombined acetic 
acid was in great measure expelled. The mass that remained, 
when well washed on a filter, afforded 23-51 gr. of tartrate of 
protoxide of lead. ' 
Five grammes of this substance, when treated with dilute sul- 
phuric acid, afforded, in three different experiments, 4°2318, 
4-229, and 4-228 gr. of sulphate. Consequently the tartrate 
must consist of 
Tartaric acid...... 37°79 100-00 
Protoxide of lead ... 62°25 164°87 
Hence the 23°51 gr. of tartrate must contain 6°875 of dry 
acid, and 100 parts of the crystallized acid contain 11:25 of 
water, and 100 of dry acid unite with 12-7 of water, which con- 
tainl1-2 of oxygen. But the protoxide of lead saturated by 
100 parts of this acid contains 11-788 parts of oxygen ; so that, 
‘notwithstanding the slight difference in the quantities, it is pretty 
obvious that the crystallized tartaric acid must contain a quan- 
tity of water, in which there is as much oxygen as in the base 
saturated by the acid. In the experiments which I have under- 
taken, in order to investigate the laws of the formation of organic 
‘productions, and which I shall hereafter make public, [ have 
‘only found in the tartaric acid 56-384 per cent. of oxygen, with 
39-206 of carbon, and 4:41 of hydrogen, which is nearly five 
times as much as is contained in the water that saturates 100 
parts of tartaric acid. In two experiments on the tartrate of the 
protoxide of lead, I obtained from 3 grammes, with a very slight 
variation, 3°7555 gr. of carbonate of lime, and -425 of water. 
If now the compositions of this tartrate, of the carbonate of 
lime, and of water, are correctly ascertained, the deficiency in the 
computation must be made up with oxygen: but it must be al- 
lowed that this determination is subject, from various causes, to 
an error of as much as 1 per cent. I must here repeat the re- 
mark, that I may possibly have estimated the quantity of oxygen 
contained in the different bases somewhat too high, and that 
hence may arise the difference of the quantity as deduced from 
the 
