84 Analyses of several Minerals. 
manner of a soluble silicate, the soda and potassa being separated 
by chlorid of platinum. Phosphoric acid was carefully sought 
for by several processes and repeated trials without success. 
The following are the result 
sae Phe ie ey ey ae , 
Alum ‘ : ' : 3°85 
Proictya of iron, . : : ; . 24:15 
Magnesia, ; : 1:10 
Lime, 1:73 
Soda, 1:60 
Potassa, ‘ ‘ ‘ . + ae 
Water, ; : ‘ . ‘ : 10:12 
101-12 
2. Vivianite from the green sand of Delaware, about four miles 
west of Cantwell’s Bridge. The beauty of the crystals, and the 
similarity of some of the specimens with Thompson’s mullicite 
induced me to investigate them. The mineral is perfectly crys- 
tallized in oblique rhombic prisms, with several terminal planes, 
and a brilliant cleavage parallel to a lateral end plane. Although 
the prismatic planes reflect a good image, yet their intervening 
es are rounded like the apatite from Rossie, N. Y. When f 
first obtained from the green sand, the crystals were perfectly col- 
orless, but in the course of some weeks changed to a light green 
color, still, “Neate ae their transparency. ‘T'he phos- 
Phosphoric oid, ‘ ; : é eo QFE 
Protoxyd of i ou ‘ ‘ : : 44-10 eis 
Water, . i. glk ene Be fiat 
Silica, A : ; : ‘ ‘ 0-10 | 
99:32 
The analysis shows that the formula is that given to vivianite. 
In a late number of his Annual Report, Berzelius calculated a differ- 
ent formula from a partial peroxydation of the iron; but the above 
well crystallized specimen indicated only a slight amount of oxy- 
dation, and the change in color of the mineral after removal from 
the green sand, indicated this oxydation. The true formula is 
therefore 3Fe0, PO, +8HO. 
3. Garnet, from ‘Franconia, New Hampshire. Part of the 
specimen exhibited the usual characters of colophonite, but the 
ge ee points of some of the coarse grains showed the bee 
of garnet in a combination of the 24-hedron 
