ON THE STEREOCHEMISTRY OF NITROGEN. 189 
This configuration may therefore be left out of consideration, and since 
the ‘cubic’ has no advantages and some disadvantages, we shall confine 
ourselves to the ‘ pyramidal,’ which may conventionally be represented as 
a plane projection. 
7 
TEES 
The way in which this configuration can be made to account for 
(2) and (3) has already been discussed.! The argument, therefore, need 
only be briefly stated. 
In order to account for the existence of isomeric hydrindamine salts 
it is necessary to make one of two assumptions (5) or (c) (above) as to the 
manner in which the two new radicals, H and X, are attached to the 
amine, all three groups in the latter being in one plane with the nitrogen 
atom. 
First H and X may be placed at the two unoccupied positions at the 
base of the pyramid, the asymmetric group (a) being also at the base 
(since if it occupy the apex no isomerism arises) ; the two salts would be 
thus represented : 
m@ 
fom 
c 
Fig. 5. FIG. 6. 
He H x 
S< i >< 
a x a H 
This assumption requires that the compound produced by the com' 
bination of aX and N a 6 ¢ should have one of the configurations, 
Figs 7%. Fig. 8. 
Dd - Dd 
which are enantiomorphously related and should be optically active. 
This expectation cannot be experimentally realised. 
1 Trans. Chem. Soc., 88, 1404. 
