Grove,* in 1841, made a few experiments on tlie amalgam, 

 and advanced the idea that it is a chemical compound of 

 mercury and nitrogen, merely swelled up with hydrogen. 



In 1864, Dr. Wetherill-[- performed several ingenious 

 experiments on the amalgam, Avithout however attempting 

 any quantitative estimate of its composition. He concludes 

 that it is not an alloy of mercury and ammonium, and that 

 the swelling up of the mass is due to the retention of gas 

 bubbles by virtue of some unexplained action which he 

 somewhat vaguely refers to catalysis. 



In the Annalen der Chemie u. Pharmacie for 1868J is a 

 paper by Landolt, in which, after pointing out the discord- 

 ance of the quantitative results obtained by Davy, and by 

 Gay Lussac and Thenard, he describes a method by which 

 he attempted a new determination of the relative quantities 

 of ammonia and hydrogen. He prepared the substance from 

 a solution of sal-ammoniac, separated from the mercury, 

 which formed the negative pole, by a porous cell. The 

 amalgam, when removed from the circuit, was washed in a 

 stream of water to get rid of the adhering solution of sal- 

 ammoniac, which always contains free ammonia. It was 

 then immediately plunged into dilute hydrochloric acid of 

 known strength, and the hydrogen evolved was received in 

 a graduated cylinder placed over it, while the ammonia was 

 estimated by determining the amount of unneutralised acid 

 in the liquid. Two experiments gave results corresponding 

 respectively to 2*15 and 2*4 volumes of ammonia to 1 of 

 hydrogen. These figures of Landolt's cannot be considered 

 satisfactory, neither nearly agreeing with each other, nor 

 approximating to the ratio 2 : 1 sufficiently closely to justify 

 his conclusion that they "completely confirm the results for- 

 merly obtained by Davy." Indeed Landolt points out a 

 serious defect in his process, namely, that however rapidly 



* Phil. Ma(/., United Series, vol. xix., p. 97. 



t Silliman^s Amer. Journal [2], xl., 160. 



JSupp. Bd., Ti., p. 316. 



