On definite Proportions. 141 
after being dried with proper precautions to avoid its dis- 
sipation: and this proportion of sulpbur and iron is the 
same, with the exception of a small fraction, as that which 
is required for the sulphuret at a minimum. This ex- 
periment therefore demonstrates, that the quantities of sul- 
phur and oxygen, which saturate 100 parts of iron, are 
jn the same proportion as the quantities which saturate 100 
parts of hydrogen. But if 100 parts of ircn take up 29°5 
of oxygen, and if it is confirmed that 11-754 of hydrogen 
and §8-246 of oxygen constitute water, the 29°5 parts of 
oxygen imply that 100 parts of iron will set at liberty 3°929 
of hydrogen, which take up the quantity of sulphur united 
with iron at a minimum, that is, according to the experi- 
ments before related, from 58°75 to 59 parts per cent. But 
3-929: 59 = 100: 1501°54; and 100 parts of hydrogen 
must take up 1501°54 of sulphur; which is twice as 
much as the quantity of oxygen that we suppose required 
by 100 parts of hydrogen. Hence sulphureted hydrogen gas 
must consist of 6-244 parts of hydrogen and 93°756 of 
sulphur. 
These proportions, determinable by calculations, must be 
capable of being inverted in every possible way, if they are 
to be of any value. The number of analyses is still too 
small for ‘universal application; and they must’be diversi-~ 
fied in various direciions, in order to give sufficient oppor- 
tunity for more accurate tests. I will however adduce one 
example from the analyses reiated in this essay. 
Iron and hydrogen, saturating 100 parts of sulphur, 
must be in the same proportion as when they saturate 100 
parts of oxygen. According to the determinations given 
in the former part of this essay, 100 parts of sulphur are 
saturated by 170°2 of iron, and by 6-66 of hydrogen ; and 
100 parts of oxygen, saturated with iron at a maximum, or 
in the form of a protoxide, iake up 339 parts. 
Now, 170°2 : 6°66=339 : 13°265; whence 100 parts of 
oxygen should combine with 13°265 of hydrogen. Ac- 
cording to the result of the experiment above related, that 
is, 11°754 hydrogen to 88°246 oxygen, 100 parts of oxygen 
must be saturated by 13°32 of hydrogen. 
These computations may however happen to agree, even 
although the numbers may be crroneous, supposing the 
error to arise from a cause which is common to several 
analyses. I shall therefore consider this calculation of the 
composition of water and of [sulphureted Sw.] hydrogen 
gas as a proof of the truth of the Jaw of nature which I have 
laid down, without insisting on the perfect accuracy “4 
the 
