472 SCIENCE PROGRESS 



and the hexahydro-/'-diethyl-benzylamine carboxylic acids : 



NCC.Hs), 



In the foregoing paragraphs we have been concerned with 

 compounds which owed their isomerism to some pecuHarity 

 either in the arrangement of radicles round a carbon atom, 

 or in the mode of linkage between atoms. We must now 

 turn to the question of an isomerism which is due to other 

 types of asymmetry, and we may commence with the case of 

 asymmetric nitrogen. 



Without going into any intricate details, it may be said that 

 when a nitrogen atom is united to three separate radicles, no 

 isomerism is to be found. Of course, as has been mentioned 

 already, a nitrogen atom doubly linked to a carbon atom may 

 give rise to isomerism, but in the case of a compound such as : 



a 



/ 



N— b 



no isomerism occurs ; and it is therefore assumed that the 

 groups <7, b, and c and the nitrogen atom lie in one plane or 

 are free to move about in space. Were this not so, we should 

 expect to find enantiomorphism shown by such substances. 



When we allow a tertiary amine to react with an alkyl 

 halide, a quaternary ammonium salt is formed, just as ammonia 

 and hydrochloric acid unite to form ammonium chloride : 



H H H R'" R R'" 



/ \X / \/ 



N - H + HCl = N - H N - R" + R . X = N - R" 



\ /\ \ / \ 



H H H R' X R' 



When the four alkyl groups attached to the nitrogen atom differ 

 from each other, as in the case of NRR'R"R"'C1, the nitrogen 

 atom is asymmetric, and the compound is found to exist in 

 two enantiomorphous forms of equal and opposite rotatory 

 power. The case is parallel to that of the asymmetric carbon 

 atom. It is evident that in passing from the trivalent to the 

 pentavalent condition of the nitrogen atom we have passed from 

 a plane configuration of the molecule to one of three dimensions. 



