Page:Proceedings of the Royal Society of London Vol 69.djvu/95

Rh

Mendeleef * from results, the source of which id not quoted, has put forward three equations to represent the densities d in terms of percentage p as follows :

From to 41 -2 per cent. (HN0 3 511,0) :

d =

From 41 -2 to 63 -63 per cent. (HN0 8 211,0) : d = 9570 + 84-18p-0-24;A

d =

From 63-63 to 100 per cent. :

d = 1 0652 + 62-08^-0-1 G/.

(Density of water at 4 = 10000.)

It was found, however, that the differences between the observed and calculated values were greatly in excess of those which could be accounted for by errors of observation. Pickering,! by the application of curves to the results of BerthelotJ for the heats of the dissolution and of Kolb for the specific gravities, deduced the existence of and isolated in a crystalline form the two hydrates HNOsSHoO and HN0 3 .H20. He also alludes to a minor alteration of properties at 18 20 per cent., but without drawing any conclusions therefrom, and to the fact that the maximum contraction coincides with the composi- tion of a definite hydrate.

The methods of purification have been fully described in our previous communications, and our experience has as yet pointed to no improve- ment, though we have had occasion from time to time to confirm the observations of others, working at other times and places, as to the persistent retention of some ammoniacal compound in ordinary distilled water.

Continued experience has shown that by the methods adopted for the purification of nitric acid, the amount of nitrous, sulphuric, and halogen acids did not exceed 1, 4, and 3 parts per million respectively, quantities which may fairly be presumed to have no effect upon the physical constants.

t ' Journal of the Chemical Society,' 1893 (Trans.), pp. 436-443. J ' Chimie Mecanique,' 1, p. 397.
 * 'Principles of Chemistry,' Eng. Trans. (2nd ed.), vol. 1, p. 272.