ON THE CRYSTALLINE SLAGS. 353 
slowly heated to redness; the filter, with what adhered to it, was then intro- 
duced into the crucible and incinerated as usual. ; 
8. To determine the potass, the slag was digested as usual in hydrochloric 
acid, the iron was peroxidized by nitric acid, and the solution was then 
treated with excess of carbonate of ammonia. The filtrate was evaporated 
to dryness, and the ammoniacal salts were expelled by heat. The residue 
was treated with boiling water, and the solution filtered from the brownish 
residue. The filtrate was evaporated to dryness after the addition of excess 
of sulphuric acid. The residue was dissolved in water, acetate of baryta 
was added in excess, the sulphate of baryta was separated'with the usual 
precautions by filtration; the filtrate was evaporated to dryness, and after- 
wards heated to redness. The solution contained the potass as carbonate; 
hydrochloric acid was added, and from the amount of chloride obtained by 
evaporation and heating to low redness, the potass was estimated. Berze- 
lius’s method of separating potass from magnesia by oxide of mercury was 
also occasionally resorted to. 
9. The sulphur was determined either by oxidizing with strong nitrous 
acid, or by fusing with nitrate of potass and a mixture of carbonate of potass 
and soda. Chloride of barium was added to the acid solution. From the 
_ sulphate of baryta produced the sulphur was estimated. 
_ 10. The method resorted to for the detection of phosphoric acid will be 
described in each case. 
The actual quantities of the substances found by analysis will always be 
given, in order that the calculations may be corrected in the event of any 
errors in the received atomic weights being corrected by future observers. . 
The calculations have been made from the tables in the French translation 
of Rose’s work by Peligot (Paris, 1843). 
1. Anatysis. By J. P. 
1. Weight of slag employed 27°57 grains, after having been gently heated 
over a spirit-Jamp. 
2. Silica 10°49. 
8. Alumina 3°89. 
4. Sulphate of lime 24°35. 
5. Phosphate of magnesia (2MgO, P? O*) 5°74. 
6. Oxide of manganese (MnO + MnO? 03) 0:12. 
7. Sesquioxide of iron 0°39. 
8. Potass. Weight of slag employed 50°47 grains. Chloride of potas- 
sium 1°48. A minute quantity of precipitate was produced by the addition 
of antimoniate of potass to the solution. The quantity of chlorine was de- 
termined by nitrate of silver. The chloride of silver obtained weighed 2°814 
grains, which correspond to 0°694 of chlorine. 1°48 of chloride of potassium 
by the tables, contains 0°702 of chlorine. Difference 0°702—0°694=0°008. 
The chloride may therefore be estimated as nearly pure chloride of potassium. 
__ 9. Sulphur. Hydrosulphuric acid was liberated by the action of hydro- 
ehloric acid. Weight of slag 22°68. The nitrous acid process was em- 
ployed. The sulphate of baryta weighed 0°59. It was ascertained that the 
slag did not contain sulphuric acid. The sulphur is admitted to exist as 
sulphuret of calcium. 
10. Phosphoric acid was not detected. Weight of the slag employed 
51°38 grains. It was digested with hydrochloric acid, and the silica sepa- 
ted as usual. The precipitate obtained by the addition of ammonia in 
light excess was dissolved by hydrochloric acid; tartaric acid was added, 
