1903.] RICHARDS—SOLVENT IN CRYSTALS. 29 
solvent is water, an organic liquid, or a fused salt at high tempera- 
ture, and whether the crystallization is quick or slow, in small or 
large crystals, the effect is always traceable, although of course in 
varying degrees. The amount of the enclosure varies from perhaps 
0.01 per cent. to 0.5 per cent. of the total weight of the crystals.’ 
As a general rule, the clearer the crystal, the less the included 
mother-liquor ; but appearance is not a wholly safe guide, since 
sometimes the refractive index of the mother-liquor is not far from 
that of the crystal. In this case the included impurity is invisible. 
Moreover, the inclusion may, and undoubtedly does, often occur in 
cells so small as to be beyond the reach of the best microscope. 
It is not contended that the production of a perfectly pure crys- 
tal from a solution is impossible, but only that no evidence has 
been obtained proving that this end has ever artificially been 
reached, while much contrary evidence is available. 
In looking over the records of the determinations of atomic 
weights, it is surprising to see how many experimenters in even 
this most exact field of quantitative analysis have either entirely 
overlooked the danger, or have taken inadequate means to over- 
come it. This is especially true when the salts to be weighed con- 
tain combined water of crystallization ; indeed it is almost safe to 
rule out utterly all such results, without further consideration. 
Among other more or less vitiated cases, where salts supposedly 
anhydrous have been weighed, may be mentioned especially a 
number of the analyses of typical salts of osmium, iridium, 
platinum, and palladium. Often a painstaking but unthinking 
chemist has spent months in eliminating the hundredth of a per 
cent. of some foreign metal and finally ignored the ¢emth of a per 
cent. of water in his preparation. It is not, however, the purpose of 
this paper so much to seek the errors in individual instances of past 
work, as to point out the ways in which these errors may be avoided 
in the future. ; 
How then is this included solvent to be eliminated without de- 
composing the substance which we desire to weigh? 
It is usually considered as a sufficient precaution to powder the 
material finely and expose it to the air for a short time, in order 
to allow the undesirable water to evaporate. This crude proceed- 
ing involves a double uncertainty ; in the first place, the unwar- 
1 For examples of carefully obtained evidence, see Richards, Proc, Amer, Acad. 
Arts and Scicnces, 23, p. 177 (1887); 26, p. 267 (1891); 28, p. 11 (1893); 29, 
p. 60 (1893); 33, p- 299 (1898); 35, P. 139 (1899); 37, P. 434 (1902); 38; P. 
411 (1902). 
