Mr Jones , Some substituted ammonium compounds , etc. Ill 
Some substituted ammonium compounds of the type NR'R"Rf'X. 
By H. 0. Jones, B.A., Clare College (communicated by Mr Fenton). 
[ Received 4 March 1901.] 
The isomerism of substituted ammonium compounds has long 
been the subject of numerous researches; but until quite recently 
no very definite positive results had been obtained. The course of 
such investigations is much hampered and complicated, first, by 
the peculiar mobility of groups attached to the nitrogen atom, 
necessitating the use of ‘ heavy ’ radicals to render the compounds 
stable enough to shew isomerism, and further, by the great 
difficulty or even impossibility of producing the desired com- 
pounds in a crystalline state or even at all. 
Le Bel ( Compt . Rend. cxn. 724) found that an ammonium 
compound containing four different alkyl radicals acquired a small 
but fugitive rotatory power under the action of moulds. No 
isomerism was observed. This remained the only isolated obser- 
vation on the optical activity of the nitrogen atom until 1899. 
Le Bel also observed certain differences of crystalline form in 
compounds having three radicals the same. These were attributed 
to dimorphism. 
Schryver and Collie ( Chem . News lxiii. 174) prepared diethyl - 
meth}disoamylammonium chlorplatinate in the three possible 
ways, and found that the crystalline form of one compound 
differed from that of the other two. This difference however 
readily disappeared on recrystallisation. 
Wedekind ( Ber . xxxn. 517, 3561) found two definite and 
stable isomers of phenylmethylallylbenzylammonium iodide. All 
attempts to resolve these compounds into optically active portions 
were unsuccessful. 
Pope and Peachey {Jour. Chem. Soc. 1899, lxxv. 1127) using 
Reychler’s dextro-camphorsul phonic acid, succeeded in resolving 
one of Wedekind’s compounds into two portions of equal and 
opposite rotatory power. The optical activity is here due to the 
asymmetric nitrogen atom. 
Of the many configurations which have been proposed for the 
groups attached to the nitrogen atom, two seem to explain the 
phenomena observed satisfactorily, namely Bischoff’s ‘ pyramidal ’ 
configuration, in which the five groups are supposed to be dis- 
tributed around the nitrogen atom somewhat after the manner of 
9—2 
