736 Proceedings of Royal Society of Edinburgh. [sess. 
The development of organic chemistry was the chief cause of 
the extension and wide application of constitutional formulae. To 
organic compounds perfectly definite formulas have always, as 
far as ’ possible, been assigned. The artificial distinction between 
organic and inorganic compounds, the slow death of the Berzelian 
method of formulae-writing, and the fallacious notion of “ double- 
salts,” perhaps give the explanation why wp to the present such 
definite constitutions have not in all cases been assigned to 
inorganic salts. By means of these definite formulae the chemist 
has been enabled to comprehend more thoroughly the nature of 
organic compounds, to predict the existence of new substances, 
and in many cases successfully to prepare them. Recent develop- 
ments have shown, not the too great rigidity of such formulae, but 
that they are not sufficiently exact. 
In this paper I have endeavoured to show the close connection 
between a certain class of organic compounds and a considerable 
number of inorganic salts, and how a definite constitution can be 
given to the latter. It seems possible, when due weight has been 
attached to the existence of isomorphous mixtures and solid 
solutions, and such substances have been removed from the list 
of definite compounds, that definite constitutions can be ascribed 
to all salts, organic and inorganic, and that the notion of 
molecular union can be finally ruled out. 
I have appended a list of references to the more important and 
recently prepared compounds in the table. Particulars of most 
of those for which no reference has been given can be obtained in 
works of reference. 
In conclusion, my thanks are especially due to Mr J. Kerr, 
B.Sc., under whose direction this investigation has been carried 
on; also to Dr Hugh Marshall, F.R.S., for many valuable 
suggestions. 
Chemical Laboratory, 
Surgeons’ Hall, Edinburgh. 
