THE OLD AND THE NEW CHEMISTRY. 
29 
when multiplied into the specific heats of these elements, yielded 
an approximately constant number. Specific heat x combin- 
ing number = constant = 6*4. This statement was shown by Du- 
long and Petit to hold good for 1 3 solid elements. Since the time 
of its announcement the truth of the statement has been proved 
in the case of a considerable number of other elements. Che- 
mists are now, for the most part, convinced that a safe generaliza- 
tion may be made to the effect that combining number x 
specific heat = constant = approximately, 6*4. 
If this statement be accepted, it follows that 
approximately, combining number. 
If, therefore, we are unable to find a series of gaseous com- 
pounds of an element whose combining number we are desirous 
of determining, it is sufficient that we determine the specific 
heat of that element, and divide the number so obtained into 
6*4: the result will approximately represent the combining 
number required. By careful analysis of one or more compounds 
of the element in question we shall then be able to determine 
with very considerable accuracy the true combining number. 
The specific heat of silver was determined by Regnault to be 
0-057 : 
6-4 
•057 
= 112*28: the combining number of silver is 
therefore about 112. But careful analysis of silver chloride 
shows that this substance contains silver and chlorine in the 
proportion of 107*66 parts of the former to 35*46 parts of the 
latter. From analyses of gaseous compounds of chlorine, the 
combining number of this element has been determined to be 
35*46 : hence we conclude that the combining number of silver 
is 107*66. Had the specific heat determination yielded a num- 
ber half of that actually obtained, we should have adopted 
107*66 x 2 = 215* 32 as the combining number of silver. 
On the old equivalent system of notation, numbers were 
given to the elements, the products of which into specific heat 
were sometimes equal, approximately, to 3, sometimes to 6, and 
in a few cases to 2. Ho far-reaching generalization could be 
made expressing the relation between specific heat and equiva- 
lent number. 
In the year 1820 Mitscherlich announced, as the result of 
careful study, that there exists a close relationship between the 
crystalline form and the chemical composition of compound 
bodies. If two compounds have a similar crystalline form they 
have also a similar chemical composition. This fact was applied 
in the determination of the combining numbers of many elements. 
The combining numbers of the elements in a compound being 
