1883.] Mineralogy. 1161 
prismatic mineral labeled topaz, which attracted his attention 
from the fact that it showed no trace of cleavage on the broken 
end, but had an irregular, conchoidal fracture. A trial of the spe- 
cific gravity gave 2.67, too low for topaz, and it was concluded 
that the crystal was iolite (cordierite). The crystal was 1.5 centi- 
meters long and r centimeter broad, larger than ever before 
known in Brazilian crystals. The absence of color in such a 
large crystal indicates that the usual deep bluish color is acciden- 
tal and due to some coloring pigment. 
SEPARATION OF MINERALS ACCORDING TO THE DEGREE OF Co- 
Lirsiopninite.—Mr. S. L. Penfield? has published two new 
analyses of lithiophilite, which are of interest in that they sub- 
stantiate the formula of an orthophosphate, Li (Mn, Fe) PO, for 
€ species, and taken together with his previously published 
analyses, complete a series showing a gradual transition between 
the two extremes of a lithia-iron phosphate containing little man- 
ganese (triphylite) and a lithia-manganese phosphate containing 
but little iron (lithiophilite). Lithiophilite is, therefore, a man- 
§anesian variety of triphylite. : 
One of the analyses was made on a pale bluish, transparent, 
brilliant lithiophilite from Branchville, Conn., and the other upon 
specimens from Norway, Maine, a new locality. The specimens 
from the latter place are blackened on the exterior by the oxida- 
tion to which this mineral is particularly liable, but in the inte- 
"or are of a light salmon color. 
SOME RECENTLY FORMED M1NERALS.—A. Lacroix has arpana 
uced crystals of gypsum by treating pulverized fluorite w! 
Sulphuric acid and then adding water. After standing for seve 
months at a low temperature until the liquid had evaporite 
Small but beautiful transparent crystals of gypsum were us 
upon the fluorite and on the sides of the containing flask. 
in: 
Brett Polytech. Fourn., 248, 112. 
Amer. Fourn, Se., Sept., 1883. 
