42 On definite Proportions. 
melted into one substance with the glass of the retort. 
Long before the conclusion of the operation, the evolution 
of ammoniacal gas-had completely terminated. The retort 
had lost 1°6525 gr., the potass in the receiver had gained 
0525, and that in the tube ‘0025, which gives, for the dry 
ammoniacal gas that bad escaped, 1°5975 gr. and conse- 
quently for the sal ammoniac 31°95 per cent. of pure am- 
monia. Since in this experiment the mass was only melted 
at the bottom, it had retained 66 percent. of water; but 
the ammoniacal gas which escaped was the same in both 
experiments, within ‘002 of the weight of the salt. Hence 
these experinents confirm each other completely, and the 
more so, as the greater quantity of water, which passed over 
in the third experiment, had probably retained a little more 
of the ammoniacal gas in the potass. 
5.) The high temperature, which had been employed for 
driving over the last portion of water, might possibly have 
decomposed the ammonia, and formed some water from its 
oxygen and a part of its hydrogen, and this water remain- 
ing with the salt of lime might have lessened the loss of 
weight. In order to put this conjecture to the test, I re- 
peated the experiment with sal ammoniac, which I had 
mixed with three times its weight of finely powdered caustic 
potass, in a similar apparatus. The retort was heated over 
the flame of a spirit lamp, and the decomposition took 
place immediately upon the first operation of the heat, so 
that the ammoniacal gas rushed out with violence. As the 
heat was continued, no more ammoniacal gas was dis-. 
charged, but only the water of crystallization of the potass | 
and of the muriate of potass. During the expulsion of 
the aqueous vapours from the neck of the retort, it acci- 
dentally cracked, and a little water visibly evaporated 
through the fissure. But notwithstanding this circum- 
stance, and although the ammoniacal gas passed so rapidly: 
through the potass, that it could not be completely dried, 
yet the loss of weight amounted to no more than 33°5 per 
cent., or only 14 per cent. more than in the former experi- 
ment ;-a proof that my supposition respecting the decom- 
position of ammonia was unfounded. 
It was therefore established by all these experiments, that 
the muriate of ammonia consists of 
Muriatic acic .... 49°55 100°00 
Caustic ammonia. 31°95 64°48 
UP GETe st se 18°50 [37°34] . 
» And if we proceed on this foundation to calculate the 
/ quantity 
