no SCIENCE PROGRESS 



the electrical conductivity and electrical potential of the blue 

 solution point to its being a colloidal solution of the metal 

 analogous to the " metallic fogs " formed by the electrolysis of 

 fused salts. Now the blue solution is formed by adding an 

 excess of ammonia to the " compound " and therefore it must 

 be supposed that the latter dissociates when an excess of one 

 of its products of dissociation is added. This is quite contrary 

 to the law of mass-action ; but as the absence of a compound in 

 the blue solution has not been definitely proved, the existence 

 of the compound NaNH 3 maybe provisionally accepted. The 

 foregoing statements apply to potassium as well as to sodium. 

 If the compound has the formula NaNH 3 or KNH 3 , the nitrogen 

 would function as tetrad, which is unusual ; and as ammonium 

 is a univalent radicle, we should expect the sodium-ammonium 

 to be also univalent and hence the free compound would pro- 

 bably be (NaNH 3 ) 2 . By determining the diminution of the 

 vapour pressure of ammonia containing some dissolved sodium, 

 Joannis (7) was able to calculate the molecular weight according 

 to the well-known method of Raoult. The results agree very 

 well with the double formula both in the case of sodium and 

 potassium. 



More recently Moissan (8) has obtained other metal- 

 ammoniums by the action of the dry gas on the metal. Those 

 of lithium and calcium are reddish-brown solids spontaneously 

 inflammable in air and less soluble in ammonia than the corre- 

 sponding sodium and potassium compounds. Their composition 

 corresponds to the formulae LiNH 3 and Ca (NH 3 ) 4 respectively. 

 Both lithium and calcium ammoniums gradually decompose at 

 •ordinary temperatures, liberating hydrogen and forming the 

 amide: LiNH 4 = LiNH 2 + H 2 , this decomposition being more 

 rapid in the case of the calcium than of the lithium compound. 

 The properties of the metallic amides will be considered later, 

 together with those of the acid amides. A great deal of work 

 has been done on the interactions of the metal ammoniums which 

 lies outside the scope of this paper. 



Attempts at the Isolation of Ammonium 



When an aqueous solution of an ammonium salt is electro- 

 lysed, ammonia and hydrogen are evolved at the kathode. This 

 is not surprising when regarded from the point of view of 

 the valency of nitrogen, as free ammonium would contain 



