155 
to be correct, (ag= 101,4,) and there is no other subsulphate ; 
the pure turbith, no matter how prepared, possessing that 
constitution. By the action of ammonia on persulphate, or 
on turbith mineral, there is generated a white powder, 
which was submitted to most careful analysis, and gave the 
formula Hg0.so3 + 2ugo ++ Hg.NHy. was found. The proof that 
the fourth atom of mercury in this body is not an oxide, is full 
and positive, and hence Dr. Kane no longer retains the pa- 
rallelism of theories used in his former memoir, as the ab- 
sence of oxygen may now be considered as fully proved. 
The crystallizable nitrate of the red oxide of mercury 
has been found no, + 2g o +20, as the younger Mitsch- 
erslich had stated ; and Dr. Kane’ shows that there are two 
basic pernitrates, of which the one, yellow, is similar in com- 
position to the basic nitrates of copper and bismuth, that is, 
HO.NO; + Sug 0, and the second, of a brick red colour, he is 
disposed to consider as No; + 6 ug 0, though it was found 
exceedingly difficult to decide whether the compound -did 
not retain a trace of water. 
It is known that for the composition of the white precipi- 
tate, which is given by ammonia with pernitrate of mercury, 
different results had been obtained by Mitscherslich and Sou- 
beiran. These discrepancies have been reconciled by the 
discovery that there are at least two, perhaps three precipi- 
tates, almost identical in colour and properties, produced in’ 
this reaction, but which differ remarkably in their chemical 
constitution. When the solutions are.cold, and the ammonia 
not in excess, the white precipitate. has the composition 
NH3.NO; + 3ugo, the formula obtained by George Mitschers- 
lich, but if the liquor be warmed, it becomes ug.o.no; + 
2ugo +Hgnn. This modification is evidently that which 
was analyzed by Soubeiran, for he fourd' one atom of acid, 
one of ammonia, and four of mercury, which ratio is quite true. 
It will be at once’ seen thatthe former body corresponds to 
