96 REPORTS ON INVE;STIGATI0NS AND PROJECTS. 



in the thin section can be measured: (i) with the graduated combination 

 quartz wedge a, the birefringence; (2) with the ruled plate b (interval of 

 cross-section ruling, o.i mm.), the optic axial angle, provided one or both 

 optic axes appear within the field of vision; (3) with the bi-quartz wedge 

 plate c, the extinction angle. These three features, and particularly the first 

 two, are usually estimated only roughly and not measured accurately in 

 petrographic work, because of the complicated apparatus hitherto required 

 for the purpose. The present ocular was designed to serve as a simple but 

 effective substitute for such apparatus and thus to facilitate the actual meas- 

 urement of these important properties. 



(8) The exact determination of sulphur in soluble sulphates. E. T. Allen and John 

 Johnston. Journ. Am. Chem. Soc, 32, 588. 1910. 



The errors in the determination of sulphur in soluble sulphates have been 

 investigated for those cases which most frequently occur in experimental 

 work. Solutions of sodium, potassium, ammonium, and magnesium sulphates 

 have been studied, both in a state of purity and in the presence of varying 

 quantities of hydrochloric acid and alkali chlorides. In a few systems the 

 influence of sodium nitrate has also been studied. 



The most important sources of error are three in number, one arising 

 from the solubility of the precipitate and the others from its composition. 

 Two minor sources may also be mentioned. 



(a) The error from solubility depends chiefly on the quantity of free acid 

 present. Contrary to the common belief, we find that the chlorides of sodium, 

 potassium, and ammonium exercise a hardly appreciable influence on the 

 solubility. Under easily regulated conditions (0.2 cc. 20 per cent HCl in 

 350 cc. original solution) it amounts to i to 2 mg. only. It is self-evident that 

 this error (almost negligible for precipitates of i gram or more) becomes 

 increasingly important, the smaller the precipitate. 



(6) All barium sulphate precipitates carry down with them quantities of 

 the alkali sulphates, varying with many conditions. In the case of pure 

 acidulated sulphates this quantity is not far from 0.5 per cent. It is espe- 

 cially affected by alkali chlorides and may be more than doubled in this way. 

 The correction depends of course on the atomic weight of the alkali metal. 

 About 0.75 per cent is the maximum quantity of ammonium sulphate ab- 

 sorbed by precipitates under conditions which have been investigated by us ; 

 the correction for it is, however, comparatively large (1.25 per cent), be- 

 cause ammonium sulphate is entirely volatilized when the precipitate is ig- 

 nited. Magnesium sulphate is scarcely absorbed at all by barium sulphate. 

 The peculiar selective absorption exhibited by barium sulphate suggests the 

 formation of solid solutions. The evidence on the subject is, however, too 

 meager for a proof. 



(c) Barium sulphate, when precipitated from alkali sulphates, always oc- 

 cludes a certain amount of "free" sulphuric acid, which is probably taken up 

 as acid sulphate of the alkali metal. It arises, of course, from the free acid 

 added to the original solution and increases with it up to a certain point. 

 Alkali chlorides decidedly increase the amount of it, and in the presence of 

 much of the latter it becomes the chief source of error. Aside from one or 

 two qualitative observations, this loss seems to have been thus far entirely 

 overlooked. It is greater for potassium than for sodium sulphate solutions. 

 In one case a solution of the former containing 10 grams KCl and 5 cc. 20 

 per cent HCl in 350 cc. lost 1.7 per cent in this way. 



