entirely dissociated into their acids and bases, the acids, owing to the presence of water, acquiring the high or abnormal values; the slight difference between the calculated and experimental numbers being due to the presence of a very small quantity of undissociated salt. In the case of triethylamine, the amount of dissociation is evidently considerably less than in the other salts, the third molecule of ethyl it contains adding to its stability. This is only what might be expected, as the addition of a fourth molecule yields the remarkably stable chloride of tetrethylammonium. This has also been examined, and probably does not dissociate when in solution, or only to a small extent. The rotations obtained from the chloride of tetrethylammonium and the hydrochlorides of ethylamine and ammonia have been plotted out side by side with the calculated numbers in Fig. 4. From this it will be seen that the calculated numbers form a curved line (similar to the ethylamines, Fig. 2, facing p. 730), but straightens a little between triethylamine hydrochloride and triethylammonium chloride, whilst the experimental form a straight line; the latter, owing to the dissociation of the salts, is considerably far apart from the curved line in the lower portion of the figure, but both meet at the top, at a point representing tetrethylammonium chloride. It will also be seen that the distance between the curved and straight lines is less at the points representing triethylamine hydrochloride than at those representing the other hydrochlorides, thus showing a smaller amount of dissociation.

From the different results obtained from solutions of hydrochloric acid in two kinds of solvents, it appeared desirable to examine some of these salts in solvents other than water. Solutions of diethylamine hydrochloride, and ammonium iodide in absolute alcohol were therefore employed, as no other suitable solvent could be found, but from the rotations obtained dissociation evidently takes place, though not to quite so large an extent as when water is used, as the following numbers show:

All these results made it important to study the behaviour of some other salts of ammonium when in solution, and for this purpose aqueous solutions of the nitrate, acid sulphate, and neutral sulphate of ammonium were employed. Nitrate of ammonium is not supposed to dissociate to any appreciable extent, nor is it likely that the acid sulphate would; the neutral sulphate, however, is known to do so to

To face page 744.

FIG. 4.

CURVES SHOWING ROTATIONS INDICATING DISSOCIATION
OF HYDROCHLORIDES WHEN IN SOLUTION.

6.0K NHẠCL

5.0

4.0

3.0

HARE S N & SONS LITH. STVARTINS LANE...

some extent, as it gives off some of its ammonia when its solution is boiled.

The calculated rotations for these salts compared with the experimental rotation are as follows:

From this it is seen that the nitrate and acid sulphate of ammonium give results very nearly identical with those calculated. This is valuable in two ways-first, inasmuch as it shows the method of calculating the rotation of these salts to be trustworthy; and, secondly,

that the very high numbers obtained by the halogen acids and their salts with ammonia and compound ammonias are abnormal.

The rotation of the neutral sulphate of ammonium is rather higher than the calculated, probably the latter is a little low, as the introduction of a second ammonia would most likely be attended with a slightly smaller amount of change than that represented by 0.7; it would not, however, nearly account for all the difference found, which is no doubt due to a certain amount of dissociation of the salt.

It is very probable that the determination of the magnetic rotations of salts in solution may be of value in distinguishing between those salts that do and those that do not dissociate in the presence of water, as well as to give an idea of the extent of dissociation.

With regard to the foregoing results, it was thought that the numbers obtained for the heat of neutralisation of the salts under consideration might in some way correspond with those of the magnetic rotations; they do not appear, however, to bear in the least upon this subject, chloride of ammonium, and the hydrochlorides of the ethylamines giving numbers close to those obtained for the nitrate and sulphate of ammonium. Triethylamine hydrochloride is also a further exception, as it gives a very low number for its heat of neutralisation; this is quite contrary to the results of the magnetic rotation, which show that this salt exists in a less dissociated condition when in solution than the other hydrochlorides, and therefore would be expected to give higher numbers.

The following table gives the numbers obtained for the heat of neutralisation of most of the salts examined :

=

:

2(NH, + HCI) 24,544 cal.
2(NH,Et+HCI) = 26,880
2(NHEt. + HCl) = 23,620
2(NEt + HCl) = 17,480

The principal results of this research may be briefly summed up as follows:

Nitric acid when diluted with water combines with it, forming an HO

HO acid, HO NO, corresponding to orthophosphoric acid, HOP:0. HO но

Nitric acid forms ethers in an analogous way to sulphuric and the fatty acids, that is, with condensation.

Unsaturated nitrogen, N'", acts in a manner analogous to unsaturated carbon, the rotations of compounds in which it exists being considerably higher than those containing the saturated element.