204 
CHEMISTRY: STIEGLITZ AND COLLABORATORS 
chloroplatinate gave 32.41% platinum, theory 32.74%) produced by 
the hydrolysis of the imido ketone (see equation 4). In the rearrange- 
ment of the chloroamines containing substituted phenyl groups, mixtures 
result: in a fraction of the rearranging molecules the substituted phenyl 
radicle migrates to the nitrogen, in the remainder the phenyl group 
itself leaves the carbon for the nitrogen. Thus, in the rearrangement of 
di-parachlorophenyl-phenyl methylchloroamine (the second compound 
listed above), the two substances (C6H4C1)2C:NC6H5 and (CIC6H4) 
(C6H5)C:NC6H4C1 are formed. This was established by the separation 
and identification of their respective products of hydrolysis (equation 
4), dichlorobenzophenone (C1C6H4)2C0 (m.p. 141°) and aniline (bleach- 
ing-powder test), and chlorobenzophenone (CIC6H4) (C6H5)CO (m.p. 77°) 
and chloroaniHne (m.p. 70°). In each instance the melting-point of a 
mixture of the substance obtained with the corresponding synthetic 
product was taken to complete the identification. The ratios in which 
the various radicles migrate to the nitrogen in the rearrangements was 
determined by quantitative examinations of the anilines formed: for 
the instance just discussed, the ratio of chloroaniHne to aniline was 
found to be 68/32, in molar terms — in other words, the two radicles 
share in the migration very closely in the same ratio as they are found 
in the original molecule. 
2. The reagents used to effect the rearrangements are those which 
the univalent nitrogen theory would postulate: bases (sodium and cal- 
cium hydroxides, sodium methylate) were used for the rearrangement 
of the monohalogenamines to facihtate the loss of halogen acid; and, 
characteristically, heat alone was sufficient to cause the loss of chlorine 
with explosive violence by the dichloroamines, the rearrangement 
ensuing: 
(C6H5)3C.NCl2 - CI2 (C6H5)3C.N < -> (C6H5)2C :NC6H5 (6) . 
3. A compound, triphenylmethylchloromethylamine (C6H5)3C.N(CH3) 
C1+, analogous to these rearranging bodies, with the exception that the 
well known stabihty of the union between nitrogen and the methyl 
radicle would interfere with the formation of a univalent nitrogen de- 
rivative, was found not to be rearranged, either by treatment with bases 
or by heat. A rearrangement would have produced a derivative of 
methylaniline (see below), which by hydrolysis would have given 
methylaniline. All the tests for methylaniHne, even the exceedingly 
sensitive chloride of lime color test, showed its absence in the products 
of hydrolysis of the compounds formed under the conditions mentioned. 
This important result shows that the tendency of a positive halogen 
