(980 ) 
TAMMANN found the first cases in 1896 in magnesium-platinous cyanide 
and in a whole group of silicates (zeolithes). LÖWENsTEIN *) found 
in 1909 a series of new examples: neutral oxalates, like those 
of cerium, thorium, erbium, and lanthanum, strychnine sulphate and 
basie zirconium oxalate (Zr (C,O,), . 2 Zr (OH), . aq.). In the course of 
my investigations on swelling in 1907 I myself found new examples 
in the flavone derivatives quercetin, quercitrin and sophorin.*) Quite 
a series of examples is therefore, known and by a systematic search 
this number is certain to increase rapidly *). 
The hygrometric lines of four of these substances: quercitrin, 
thorium oxalate, the zeolithe calcium chabasite, and basic zirconium 
oxalate have been reproduced in the illustration for a comparison of 
their form with those of the swelling crystals. The straight horizontal 
initial part is in some of these substances considerably longer, in 
others such as quercitrin and basic zirconium oxalate of nearly the 
same order of magnitude as in the swelling crystals, but in principle 
they have the same form: an S-shaped line which commences with 
an almost horizontal part, turns the convex side downwards, then 
at a still larger water content — 
gets a point of inflection and 
turns the concave side downwards *). 
The behaviour of basic zirconium oxalate, which has been inves- 
tigated by Lowenstein *), is particularly interesting. This substance 
consists of pyramidal doubly-refracting crystals which are mixed 
erystals with water and swell to double their size when placed over 
a 5°/, solution of sulphuric acid under a glass bell jar; for instance 
a crystal, which, when in equilibrium with a 380°/, sulphuric 
acid (4 = 0.75), had a length of 17 scale-divisions, had a length of 
32 scale divisions after having been placed over a 5°/, sulphuric 
1) Zeitschr. f anorgan. Chemie 63, p. 69—139. 
2) These experiments have not yet been published; this will happen shortly. 
3) Such discoveries have, I think, a great significance for the criticism of the 
water determinations in preparative, and particularly in organic, chemistry. For, the 
habit of rounding off the analytical figures obtained in determinations of water of 
crystallisation to whole figures seems no longer permissible. That the crystal 
investigated is a hydrate of constant composition cannot be proved by the fact 
that the analytical figure found is practically a whole number, but only by the 
determination of the form of the hygrometric line, or that of the melting point line. 
4) In order to save space in the reproduction, the lines shown in the illustra- 
tion for calcium chabasite and thorium oxalate are drawn only for values of ¢ larger 
than 0.20. 
5) Loeser polly. 
