826 
Proceedings of the Royal Society 
various strengths of the three salts, chloride, bromide, and 
iodide. It shows that in the case of the first two salts, solution 
is attended by increase of volume, hut that when the last is dis_ 
solved contraction takes place as mentioned above. 
Table IV. 
A. 
B. 
C. 
D. 
E. 
NH 4 C1 
20-68 
1-06003 
1-07515 
+ -01512 
1*53 
14-84 
1-04408 
1-05219 
+ -00811 
8-64 
1-02603 
1-02872 
+ -00269 
NH 4 Br 
21-28 
1-12976 
1T408 
+ -01104 
2-379 
15-31 
1-09048 
1-0973 
+ '00692 
10-81 
1-06265 
1-0669 
+ -00435 
nh 4 i 
18-58 
1-12631 
1T241 
- -00221 
2-464 
10-92 
1-07118 
1-0694 
- -00178 
6-71 
1-0394 
1-0381 
- -00130 
The arrangement is the same as in Table II. 
The question naturally arises, Why, if ammonium chloride and 
bromide dissociate to some extent on solution, does ammonium 
iodide remain intact, as these results seem to show ? There is 
good reason to suppose that dissociation really takes place in this 
case also, for when these results are treated in the same way as 
those in Table II., it is found that when one molecule of ammonium 
iodide is dissolved in 100 molecules of water, the amount of con- 
traction is *0015. How from Table III. the amount of contraction 
in the case of a solution of that strength of one of the six salts 
there given is ’005 — *006. Unless dissociation has taken place, I 
fail to see why ammonium iodide should not give the same amount 
of contraction. 
While searching in Berthelot’s Chimie Mechanique for some 
data that might explain this smaller amount of dissociation in the 
ase of ammonium iodide, I found that the result of Berthelot’s 
experiments on the thermal effect of diluting solutions of the 
sulphates, chlorides, and nitrates of potassium, sodium, and 
ammonium is, that we may conclude that these salts are not 
decomposed by water at the ordinary temperature, to an appreciable 
