MODERN CHEMISTRY. 453 



which two atoms of carbon and one of nitrogen combined act 

 as a single atom or combining equivalent ; and the strange 

 substance called Kakodyle, already described, in which car- 

 bon, hydrogen, and arsenic combine to form a radical, singu- 

 larly marked and active in its affinities. Others of these 

 peculiar bases were at first known only conjecturally, but 

 their existence inferred from: the analogy of the compounds 

 they form. Ethyl, for example, the radical of the numerous 

 class of ethers, was defined to us through all its various com- 

 binations before it had actually been obtained in a separate 

 state. In naming the hydrate of oxide of ethyle, as the 

 equivalent of alcohol in the new chemical phraseology, we 

 at once illustrate the theory of these compound radicals, and 

 the nomenclature which is needed to express their presumed 

 character and relations. 



It must be conceded that in this abstruse part of Chemistry 

 various assumptions are made, which may be disproved by 

 future research ; and that among the numerous contingencies 

 of combination, furnished by the complex series of combining 

 bodies and proportions, the particular schemes now adopted 

 may not be those which actually exist in nature. This un- 

 certainty is shown by the various ways in which chemists 

 have represented the grouping of parts in the same compound. 

 It belongs, however, to the modes of union only ; and in no 

 way impeaches the truth of analysis, or the exactness of those 

 laws of definite proportion which form the foundation of the 

 whole. The doctrine of compound radicals occurs, indeed, 

 as a sort of corollary from these laws. Chemical affinity, 

 acting on the molecules of different bodies, with every degree 

 of force, produces combinations infinitely various in stability. 

 This is especially true as regards organic compounds ; the 

 simplest of which in aspect as sugar, starch, albumen, &c. 

 are composed in their smallest atoms, not of single ratios 

 of the organic elements, but of multiple proportions of the 



