ON THE STEREOCHEMISTRY OF NITROGEN. 183 
So far only a brief statement concerning these compounds has been 
made, and no full description of their properties or crystalline form has been 
given. The isomeric bromides differ in solubility in dilute alcohol (5:4 and 
8-2) ; the iodides in solubility in water and also in their temperature of 
decomposition. 
With the exception of these compounds described by Aschan, which, 
as will be seen later, must show isomerism, whatever view of their con- 
figuration and mode of formation be adopted, no compounds of the type 
under discussion have been shown to exist in isomeric forms, and it may 
therefore be concluded that under normal conditions stable isomerides 
cannot exist, the limit of ‘space-filling’ having been very nearly reached 
in some of the compounds used by the writer. 
iv) Compounds of the type NabcdX. 
ip y 
No systematic efforts had been made to obtain isomerides of com- 
pounds of the above type before Wedekind’s experiments were undertaken. 
Wedekind argued, from the experiments of Messrs. Schryver and Collie and 
others, that the groups in ammonium compounds were mobile, and that 
therefore isomerism could only exist when heavy groups were used and 
when the limit of ‘ space-filling’ had nearly been reached. 
The formation of phenyl-ethyl-methyl-allyl ammonium iodide in the 
three possible ways,! namely (a) combination of methyl-ethyl aniline and 
ally] iodide, (0) allyl-ethyl aniline and methyl iodide, and (c) allyl-methyl 
aniline and ethyl iodide, showed that only one product was obtained, 
though in the first case the compound was at once deposited in a 
crystalline state ; whereas the other two combinations gave an amorphous 
product which readily became crystalline on rubbing or on separating it 
from solution. Similar phenomena were observed by the writer in the 
formation of phenyl-benzyl-ethyl-methyl ammonium iodide.? The union of 
benzyl iodide with ethyl-methyl aniline took place very readily and 
yielded a gummy solid which, on separating from solution, became crystal- 
line and identical with that obtained by the addition of methyl iodide, 
or of ethyl iodide to the corresponding tertiary amines, which was 
crystalline from the first. 
The phenomena observed by Wedekind in the phenyl-benzyl-allyl- 
methyl ammonium salts? are, however, of quite a different kind. This 
compound was prepared by the union of (a) allyliodide and methyl-benzy] 
aniline, (b) methyl iodide and benzyl-allyl aniline, and (c) benzyl iodide 
and methyl-allyl aniline. Combinations (a) and (c) take place readily and 
give rise to the same product, the a compound, which crystallises in the 
prismatic system, melts at 140-142° C., and distils partly unchanged 
under reduced pressure. Combination (b) takes place very slowly and 
gives an oily product which is induced to erystallise only with great diffi- 
culty and yields a very small quantity of a crystalline solid, the /? com- 
pound. This compound crystallises in a different form (also of the 
prismatic system), melts at 158-159° C., and distils unchanged under 
reduced pressure without melting. The «a and 8 compounds could not be 
transformed one into the other. 
Isomeric a and (3 chlorides and bromides were also prepared, which 
! Stereochemie, 53; Ber., 1903, 36, 3791. 
2 Trans. Chem. Soc., 1904, 85, 224. 
3 Stereochemie, 33-52; Ber., 1899, 32, 517, 3561 
