ON THE CRYSTALLINE SLAGS. 365 
_ potass-soda mixture and LO of nitrate of potass. The fused mass was digested 
as before with hydrochloric acid; but even then the silica did not appear 
_ perfectly white, and, accordingly, it was again fused and treated as before, 
_ when it was obtained beautifully white. The alumina was determined in the 
usual way. The iron and manganese were separated by the succinate plan. 
_ On redissolving both the oxides of iron and manganese, a small quantity of 
_ silica was left, which was collected and added to that previously obtained. 
_ The solution from which the alumina was separated, was precipitated by a 
little chloride of barium, and the sulphur determined from the sulphate of 
baryta: the excess of chloride of barium was removed from the filtrate by 
_ the addition of 20 grains of strong sulphuric acid diluted with 6000 grains 
_ of water; a liquid which does not precipitate lime. The lime was deter- 
mined, as usual, after separation of some oxide of manganese by acetic acid. 
_ The remaining manganese (the greater part having been precipitated with 
_ the alumina and iron by ammonia) was thrown down by hydrosulphate of 
ammonia. The magnesia was then determined by ammonia and phosphate 
of soda. 
1. Weight of slag employed 25:18 grains. 
2. Silica 5°73. 
3. Alumina 1°84. 
4, Sulphate of lime 2:07. 
5. Phosphate of magnesia 0°49. 
6. Oxide of manganese 0°98. 
7. Sesquioxide of iron 17:19. The iron must be estimated as protoxide, 
as will be evident from the results of the analyses. 
8. Sulphate of baryta 0°89. 
Analysis tabulated*. 
RIGA. eee ale Gs eet OnE So Sas 11°83 
Protoxide of iron........ Lo.) Sure Re et 13:95 
Protoxide of manganese .. 3°58 ...... 0°70 
it gt al Ie ae lela iD aati ino! ite Aa cla 3°41 
1 ii ial ed cl cageue We Rae ets ae Ory, 
Hs Co 2 a ae ee Lf ee 0°29 
21). ee I a ee OD 
Error of loss... /......... 0°45 
100-00 
From the identity in crystalline form of this slag with the one succeeding, 
_ we are inclined to regard it as a mixture of silicate of protoxide of iron with 
_aconsiderable amount of impurity, represented especially by the alumina. 
_ This view would also appear to receive confirmation from an inspection of 
_ the slag itself. We need scarcely refer to the numerous instances of well- 
formed crystals containing much foreign matter, so that there seems to be 
nothing improbable in this view respecting the composition of the crystals of 
4h: . . 
' this slag, which are certainly not perfectly formed. 
13. 
q ‘The following beautifully crystallized specimen was presented by Mr. 
Dawes. It was found in the flue of a puddling furnace, where it had pro- 
bably been exposed to a high temperature for a considerable time. 
_* T have no doubt that phosphoric acid and sesquioxide of iron existed in this slag in 
small quantity.—J. P. 
