1904-5.] Mr Cameron on the Constitution of Complex Salts. 723 
represented Gregory’s salt, while to the other he ascribed the 
constitution 
with ten molecules of water, instead of twelve as previously 
written. Rosenheim and his collaborators have shown that in 
most cases these doubled formulae are unnecessary, and have 
uniformly used the simple formulae, though lately they have 
applied A. Werner’s Co-ordination Theory, and have, for example, 
ascribed to Crofts’ “ red salt ” the formula 
Recoura, in his many papers on the chromisulphates, and latterly 
on the ferrisulphates, consistently uses the doubled formulae, as for 
example in the anhydrous alum acid 
and Klobb, in his researches on similar compounds, has used 
“ double-salt ” f formulae. To give but one other example, 
Kohlschutter, in his investigations on the complex uranium 
oxalates, has used invariably the simple formulae. 
There appear to be three chief reasons for this absence of 
agreement among investigators. In the first place, the old idea 
of “ double-salts ” which is still so largely upheld causes the re- 
tention of doubled formulae even when the substances come to be 
regarded as salts of complex acids. Secondly, the old formulae for 
the hydroxides of trivalent metals, e.g. Cr 2 (OH 6 ), have never been 
convincingly disproved. From the nature of these substances, 
constitutions can only be assigned to them by comparison with 
their derivatives, whereas hitherto in many cases the process has 
been reversed. 
Finally, in the case of the chromoxalates, Werner | has used the 
doubled formulae, on the ground that there exist perfectly definite 
compounds corresponding to the formulae 
* References, where not otherwise mentioned, will be found at the end of 
this paper. 
t In this paper by “double-salts” are meant molecular compounds of 
simple salts. 
X J. Chem. Soc . , 51, p. 383, and 53, p. 404. 
K 2 H 2 Cr 2 (C 2 0 4 )(0H) 2 + 8H 2 0, 
