386 Dr Gmelin on a Black Mineral from Candy in Ceylon ; 
cipitated by succinate of ammonia ; the succinate of iron left, 
when heated, gave 0,26814 gr. of oxide of iron, == 20.514 per 
cent, protoxide of iron. 
(5.) The solution was then boiled with carbonate of potash, 
which precipitated carbonate of magnesia. This precipitate, to- 
gether with that obtained in No. 2. was put upon the same fil- 
ter. It assumed, after ignition, a reddish colour, which an- 
nounced some manganese. When dissolved in muriatic acid, 
there appeared bubbles of carbonic acid gas, and the smell of 
chlorine was perceived. The solution, therefore, was mixed with 
sulphuric acid, evaporated to dryness, and exposed to a heat 
sufficient to drive off the water of crystallization. The dry sul- 
phate of magnesia weighed 0.63 gr., containing 0.214326 gr. 
magnesia, — 18.240 per cent. 
This mineral is therefore composed of 
Contain Oxygen, 
Alumina, 
57.200 
26.712 
Protoxide of iron, 
20.514. 
6.219 
Magnesia, with a trace of Manganese, 
18.240 
6.767 
Silica, - - 
3.154 
99.108 
It appears that alumina contains 
nearly twice 
as much oxygen 
as magnesia and oxide of iron together, and that the composi- 
tion of this mineral may be represented by MA 2 -}- /A 2 . 
Silica cannot, I think, be considered an essential ingredient 
of the mineral ; its quantity is very small, and, considering the 
hardness of the stone, the greater part of it may fairly be de- 
duced from the agate mortar. 
The specimens of this mineral, for which I am indebted to 
the liberality of my friend Mr Heuland, were massive, and there 
is sufficient ground, on account of the chemical composition, and 
the external characters of this mineral, to consider it as Spinel; 
it may be ranged in the system as Massive Pleonast *. 
* We are informed that Count Bournon, In a late memoir, describes this mi- 
neral under the name Candite, Mr Heuland also furnished Laugier with speci- 
mens, but his analysis has not reached this country.— -E d. 
