AMMONIA NITEATE WITH AMMONIA. 
361 
withdrawn, the flask at once closed, and the stopper wired down. The flask was then 
removed from the ice, left to regain the atmospheric temperature, and again weighed. 
The delivery-tube on removal from the flask was washed, the washings were evaporated, 
and the nitrate left as residue weighed. By deducting the weight of this nitrate from 
that of the whole quantity put into the flask, the weight of this salt in the liquid was 
ascertained ; and by deducting this from that of the liquid, the weight of the ammonia 
condensed was obtained. 
The smallest quantity of ammonia sufficient to liquefy the nitrate at 0° is a little 
more than one third of its weight. This point was determined by boiling away some 
ammonia from the liquid containing more than sufficient of it, until on cooling to 0° 
the liquid just began to crystallize. This method, for several reasons, is not a very 
accurate one for the purpose, one reason being that the liquid is apt to exhibit to some 
extent, as already stated, the phenomenon of supersaturation ; but it was employed for 
want of a better, that would not have entailed trouble disproportionate to the import- 
ance of an accurate determination of this point at present. 
At 23°, the temperature at about which the nitrate and ammonia cease to unite at 
the atmospheric pressure, the ratio of the quantities that form the liquid is that of 
100 nitrate to 26 ammonia. This ratio was determined by first condensing, under the 
guidance of some previous experiments, about 27., parts of ammonia upon 100 of 
nitrate in the flask, dissolving up the whole of the nitrate by warming and agitating the 
contents of the closed flask, and then loosening the stopper and cautiously allowing the 
ammonia to boil off until on cooling the liquid to 20° the nitrate began to crystallize 
out. The stopper was then closely inserted again, and the temperature raised very 
slowly by means of a water-bath until the nitrate, by continued agitation, had all 
redissolved, save a very minute crystal. The temperature required to effect this re- 
solution exceeded 28°. On now allowing the flask to cool very slowly in the (glass) 
water-bath and watching the crystal, it was not observed to show any increase in size 
until the temperature had fallen to 2 2°*5, when its enlargement became distinct. The 
liquid was therefore regarded as saturated with nitrate at 23°. It was also about 
saturated with ammonia, because beyond this temperature ammonia ceases to be con- 
densed by the nitrate at the atmospheric pressure, and to this the liquid had been 
exposed. By now ascertaining the weight of the liquid and deducting that of the 
nitrate, the above ratio was obtained. 
5. In the conjoint liquefaction of ammonia nitrate and ammonia heat must be con- 
sumed by the nitrate in passing from the solid to the liquid state, and be generated by 
the ammonia in passing from the gaseous to the liquid state. The heat produced by 
the latter change proves to be in excess of that consumed by the former, even when, by 
keeping the nitrate in excess, the greatest relative liquefaction of this salt is ensured ; 
that is to say, the formation of the liquid is always attended with the evolution of heat. 
On the other hand, as already described, the destruction of the compound is attended 
with the production of cold. 
