12 REPORT 1859. 



well it may correspond to any particular stage in the development of science, 

 can never be more than a temporary expedient, which must be replaced 

 sooner or later by a classification framed according to more general prin- 

 ciples. 



The discoveries which have led to, or resulted from, the development of 

 the theory of polyatomic radicles, have caused a corresponding extension of 

 our notion of a chemical family or group. The principal relations of com- 

 position which have hitherto been observed among compounds of the same 

 natural family and deriving from the same type, may be expressed by the 

 following formulae, in which n and x are always whole numbers and n always 

 greater than x : — 



1. C n H s( " +1 -* } 0, 4. C"B* M) 0», 7. cfH^-^O 3 , 10. C'H^-^O 4 , 



2. C"H 2(n+1 - r) 2 ,5. C"H 2( "- r) 3 , 8. c'H**-*^ 0\ ll.C'H^-^O 6 , 



3. c"H 2(B+1-n) O 3 , 6. C"H 2(re "' r) O 1 , 9. c"H 2( *- J + T) 0% 12. C'H*"^ O 6 . 



As a special example we may take the tritylic (or propionic) group, in 

 which as yet the number of known terms is more numerous than in any 

 other. Here n = 3, x=0, and the above formulas become 



The substances which these formulae express are all hydrates, — alcohols 

 or acids. It will easily be understood that around each of them, considered 

 as a primary compound, a large number of derivatives, — ethers, anhydrides, 

 chlorides, nitrides, &c. — will arrange themselves. Thus, formula 1 repre- 

 sents, primarily, the monatomic alcohols. To the list of bodies belonging to this 

 class Berthelot* has recently added cholesterine, C 26 H 44 O (n = 26, x=5) (?), 

 and Borneo camphor or camphol, C 10 H 13 O (w=10, x=2). The first 

 of these substances is homologous with cinnamic alcohol, C 9 H 10 O ; the 

 second is as yet without homologues. Secondarily, it represents all bodies 

 which may be supposed to contain the same radicles as any actual or 

 possible monatomic alcohols. Here, therefore, come the simple and double 

 ethers, the chlorides, iodides and the like derived from monatomic alcohols ; 

 also the corresponding alkalies containing nitrogen, phosphorus, or arsenic ; 

 compounds containing metals, as kakodyl, zinc-ethyl, &c, and many other 

 bodies which these will serve to suggest. 



Formula 2 (derived from 1 by the addition of an atom of oxygen) re- 

 presents the diatomic alcohols or glycols, and such other substances as may 

 be supposed to contain similar diatomic radicles. The substances belonging 

 to this class are — 



(A) the glycols")-, of which there are already known ethyl-glycol, 



C 2 H° O 2 , propyl-glycol, C 3 H 8 O 2 , butyl-glycol, C 4 H 10 O 2 , amyl-glycol, 

 C 5 H 12 2 , probably also anisic alcohol |, C 8 H l0 O 2 , and saligenine, 



* Ann. Chim. Phys. [3] lvi. 51 (1859). 

 t Wurtz, Ann. Chim. Phys. [3] lv. 400. 



j Cannizzaro and Bertagnini, Ann. Chem. Pharm. xcviii. 188 ; Chera. Soc. Quart. Journ. 

 fat. 190. 



