173 



scribed, it becomes necessary to look upon the so-called 

 compound radicals in a new manner. 



Assuming as proved, that NH3. ho replaces K0,'and NH3 iicl 

 replaces kg/, hence, NH4 replaces k. ; but NH4 is NH2-I-H-I-H. 

 That is, a sub-amiduret of hydrogen, compounding to certain 

 suboxides and sub-chlorides. It may be isolated, but all that 

 the author asserts is, that NHa-f-Hc/acts as K.cl. He considers 

 it impossible to avoid giving to NH3 -f- cu cl and NH3 + zw cl 

 the same rank, and hence the transition to h^ NH2 -f h^ cl, 

 and similar combinations. He conceives that we cannot, in 

 the present state of our knowledge, assign to the bodies 

 NH3 cu, or H^ NH2. H^ the title of compound radicals, or give 

 to them specific names, and, therefore, whilst he retains the 

 word ammonium as convenient, and looks to the isolation of 

 it, and to its resembling a sub-oxide, and not a metal, he 

 considers the oxide of ammonium as more properly an oxam- 

 ide of hydrogen, and sal ammoniac as a chloramide of hydro- 

 gen, as the white precipitate is chloramide of mercury, and 

 so with various other bodies. 



In the development of the theory of compound radicals, 

 which arose from the author's investigations, two conse- 

 quences were obtained, viz., that the amide replacing oxide 

 of hydrogen, the bodies HO.SO3 + ho, or cuo, SO3 -f- ho, 

 assimilate themselves to H0.SO3 4- NH3, or CMO.SO3 -{- NH3, 

 and thence the water designated saline in Graham's me- 

 moir, forms, with the metaUic oxide, a compound base, to 

 which the theory must eventually be applied ; and, secondly, 

 the further extension of the investigation shews the difficulty 

 of drawing a line between these and the proper basic salts, 

 of which a great number has been examined by Dr. Kane, 

 for the purpose of obtaining evidence on these points, and 

 the result has been, that such basic salts are constituted on 

 the same type as the neutral salts of the same family, the 

 water being replaced by oxide of a metal, and in many cases 

 the metalHc oxide becomes likewise hydrated by combined 



