COMPOSITION OF AMALGAM QF AMMONIA. 581 



found, that this mode of analysis was Tery inaccurate : 1st, 

 because the amalgam is half destroyed before it is well dried : 

 2dly, because this amalgam displaces a volume of air, of 

 which it is difficult to take account: 3Jly, and lastly, be- 

 cause, on introducing it into the phial, the hidrogen and 

 ammoniacal gas evolved take the place of a quantity of 

 air, which cannot be estimated, and must necessarily occa- 

 sion great errours in the results. Hence the weights of all 

 differed from one another. One gave us on 3*069 gr. of 

 mercury an augmentation of 0*002; another, an increase 

 of 0003 ; a third, of 0-0045 ; and a fourth, of 0-001 only. 

 It is even possible, that a loss of weight might appear, 

 since the air of the phial is replaced by hidrogen and am- 

 moniacal gas. Such no doubt were the causes of Mr. Davy's 

 mistake, when he found that mercury, in forming an 

 amalgam, was increased only a twelve-thousand"th of its 

 weight. 



Impelled by these reasons to reject thi$ mode of analysis, Mode of ana- 

 we employed the following, which we consider as very 'ysis employed, 

 exact. Knowing the quantity of hidrogen contained in the 

 ammoniacal amalgam ; and not doubting, that the hidrogen 

 and ammonia were in a uniform proportion to each other 

 in this amalgam, we had recourse to this proportion, to de- 

 termine the whole quantity of the ammonia it contained^ 

 For this purpose we converted into amalgam 3*069 gr. 

 [47*403 grs.] of mercury; after the amalgam was well 

 dried with blotting paper, we introduced it immediately into 

 a small jar very dry, and a quarter iilled with mercury; 

 and immediately too clapping a finger 'on the mouth of the 

 jar, we shook the whole together for a few minutes. In 

 this way the portion of amalgam that still subsisted was de- 

 composed, the hidrogen and ammonia it contained return- 

 ing to the state of gas; for the moment the little jar was 

 immersed in mercury and unstopped, the mercury was seen 

 to sink. Three other similar experiments were made, in 

 order to obtain more decisive results; and after each ex- 

 periment the gasses were passed into one and the same very 

 dry tube filled with mercury. When they were thus all 

 collected in the tube, the quantity of ammonia they con- 

 tained was determined by agitating them with water. Then, 



to 



