of Edinburgh, Session 1881-82. 
825 
Table III. 
100M. H 2 0 
NaCl 
KC1 
NaBr 
KBr 
Nal 
KI 
1. m. salt. 
•0061 
•0057 
•0059 
•0053 
•0054 
•0065 
2- 
•0117 
•0110 
•0102 
•0109 
•0120 
4. 
•0209 
•0191 
• • • 
•0183 
•0200 
6. 
•0280 
•0261 
8. 
*0338 
•0288 
- - - 
These numbers, though not complete, show clearly that the 
amount of contraction (expressed by the difference between the 
actual and theoretical specific gravity), produced by dissolving any 
one of these six salts, in the proportion of one molecule to 100 
molecules of water, is a number lying between *005 and ’006. In 
the case of two molecules between *01 and *012, and that the 
amount of contraction produced by the addition of each successive 
molecule, is a steadily diminishing quantity. These points may 
help to explain the abnormal specific gravity of ammonium 
chloride solution. 
Knowing that solutions of ammonium salts, e.g., chloride, 
nitrate, and sulphate, became acid on boiling, it struck me that the 
explanation was to be found in a partial dissociation on solution 
in water. In order to ascertain whether the dissociation already 
observed at 100° was due to heat, or whether it actually existed 
at ordinary temperatures, pure dry air was drawn for three days 
through a solution of ammonium chloride rendered alkaline with 
ammonia. At the end of that time the solution was distinctly 
acid ; but as this might be due to the formation of nitrous acid in 
small quantity, an alkaline solution of ammonium chloride was 
placed in vacuo over sulphuric acid, in less than twenty-four hours 
it was acid. 
The next point to be determined was, whether this abnormal 
specific gravity was common to all ammonium salts, if so, then 
the explanation by dissociation would be confirmed. I tried 
solutions of ammonium, bromide, nitrate, iodide, and sulphate. 
The specific gravity of the two former was abnormal, that of the 
two last normal. 
Table IV. gives the results of experiments with the solutions of 
