1822.] M. Rose on Mica. 257 
single one was found, which might be properly considered as 
possessing only one axis of double refraction. lt was met with 
at Berlin in a collection of Siberian minerals, without any notice 
of the place where it had been found; it is highly probable that 
it was from Siberia. The analysis of this singular mica gave 
27] oa pion is ieee ae amnteneereniae tA 2.1 
FGronide Of OE ode sndese <.eheanndin suiblocnsest, Saas 
PUUTEANTND 85> 0. pai sae acayask, Hadbecaid adage tkass 16°05 
Magnesia......... et EINE : sn Ee 
| 21a I aS ERAS BE FORA See QS 
Fin Ggic ACI eg 5 oes sche eo A Beh Ae 0:68 
DRAB ESC i ae din in csiere t Neigh RR ES Trace 
97-19 
The great loss in this analysis depends certainly upon the diffi- 
culty of separating magnesia and potash so as to ascertain their 
exact quantity. All methods that have been made use of afford 
only approximation, and I know none which would give an exact 
result. The quantity of potash was ascertained in the following 
manner: Thin leaves of mica were carefully placed into a crucible 
with nitrate of barytes in alternating layers, and heated red hot. 
The ignited mass was dissolved in muriatic acid, the silica was 
separated by evaporation, the barytes by sulphuric acid, alumina 
and oxide of iron by ammonia. The liquid was evaporated to 
dryness, and the residuum after having been heated long enough 
to volatilize all the sulphate, and muriate of ammonia was dis- 
solved in water, and the solution was mixed with acetate of 
barytes. The liquid separated by filtration from the sulphate of 
barytes was evaporated, and the dry salt heated red-hot. Water 
poured onit dissolved the carbonate.of potash, which, after eva-, 
poration, was heated red-hot, and weighed. For further compa- 
rison, it was saturated with muriatic acid, and its weight again _ 
ascertained. In order to find the other component parts of this 
mica, the analysis was repeated with carbonate of potash instead 
of nitrate of barytes. Silica was obtained by evaporation to 
dryness in the usual way. After having been heated, the parti- 
cles adhered together; it was not, therefore, perfectly pure 
silica, which, when heated, forms an extremely fine powder. 
The liquid which had _ heen filtered from the silica was treated 
with ammonia, and the precipitate thus obtained boiled with 
caustic potash. The oxide of iron precipitated was dissolved in 
muzriatic acid, the solution neutralized by ammonia, and precipi- 
tated by succinate of ammonia. The alumina was precipitated 
from its solution in caustic potash by muriatic acid, redissolved 
by an excess of it, and again precipitated by carbonate of 
ammonia. The liquid which had been filtered from the precipi= 
New Series, vou, 1v. 8 
