Dr. Kane on the Compounds of Ammonia. 87 



not be successfully isolated, gave to the theory of organic chemistry great clear- 

 ness and consistency, and was indeed philosophically just, since from the facility 

 of decomposition of cyanogen in a variety of ways, we must infer that many 

 bodies of similar nature may be so much more easily decomposed, that in our 

 ordinary modes of operating on them their preservation becomes impossible, pre- 

 cisely as the existence of cyanogen had escaped the acuteness of Proust, of Ber- 

 thoUet, and others, who had experimented on prussic acid at former times. I 

 therefore do not hesitate to place the theory of compound radicals amongst the 

 greatest benefits which chemistry has lately received, and hope with confident 

 expectation for the addition of very many new examples to the list, hitherto 

 restricted to cyanogen and mellon. 



But what is the constitution of a compound radical ? does it consist of a 

 group, beyond which we cannot go without reducing it to its merely undecom- 

 posable constituents ? or has it, again, a symmetricity of constitution like the 

 whole mass from which it had been eliminated. I shall not touch upon this ques- 

 tion as affecting cyanogen, benzoyl, or similar bodies, limiting myself altogether 

 to the examination of how far our ideas of the nature of ammonium may be 

 affected by that point of view. 



In sal ammoniac, the chlorine is certainly united with a body which replaces 

 potassium, and if we could discover circumstances under which the chlorine 

 might be transferred to another substance, leaving all the hydrogen and azote 

 undisturbed, then the ammonium would be isolated ; but let us examine what 

 this ammonium should be. The sal ammoniac is chlor-amidide of hydrogen. 

 If the chlorine were removed, the amidogene should remain combined evidently 

 with twice as much hydrogen as constitutes ammonia, and this body, sub-amidide 

 of hydrogen, might well be able to represent in combination, and to combine 

 with, metals. This partial participation in metallic properties is found in other 

 sub-combinations, as in the sub-oxides of copper and of mercury, and hence the 

 generation of the ammoniacal amalgam, its low specific gravity, the sub-amidide 

 of hydrogen being probably gaseous : an extension of this view might illustrate 

 the condition of the isomorphism of two equivalents of one oxide with one of 

 another, (as pointed out in the alums and certain minerals in the last proposition,) 

 the former, perhaps, assuming the form o (ror) : the sub-oxide represented in 

 the brackets relating itself as a compound radical to the oxygen outside. Hence, 



