CHEMISTRY: J. STIEGLITZ 
197 
The recent development of electron theories of valence has given the 
investigation of the causes underlying molecular rearrangements an op- 
portunity for much greater precision in formulating the problems and 
in expressing the results of their study. Thus, the original electron 
theory of valence proposed by J. J. Thomson^ has been used with ad- 
vantage in interpreting the molecular rearrangements of hydroxylamine 
and halogen amine derivatives of certain organic compounds independ- 
ently and practically simultaneously by L. W. Jones- and myself.^ 
These interpretations differed in some important particulars, but agreed 
as to the fundamental relations. 
The electron theory will here be used in expressing the results of 
recent investigations of molecular rearrangements of triphenylmethane 
derivatives. Inversely, certain definite conclusions may be drawn from 
the study of the rearranging compounds in regard to assumptions made 
in the various theories of valence; and it is proposed to present this evi- 
dence first as a contribution to the greater problem of valence from 
the field of pure organic chemistry — the very field in which newer views 
are most urgently advanced by physicists and physical chemists. 
The Electron Theories of Valence and the Molecular Rearrangements of 
Halogen Amines and Similar Derivatives. When an acyl halogen amine 
R.CO.NH(Hal) is treated with bases, it undergoes the Hofmann rear- 
rangement, the radicle R (which represents an alkyl or acyl radicle or 
hydrogen) being torn from its union with carbon and carried to the 
nitrogen.^ We have, for instance: 
R.CO.NH(Cl) + NaOH->NaCl -|- HOH -f CO:NR. (1) 
Triphenylmethyl halogen amines, we have found, ^ undergo the same 
kind of rearrangement: 
(C6H5)3C.NH(C1) -h NaOH -> NaCl + HOH + (C6H5)2C :NC6H5 (2). 
Now, in both these classes of chlorine derivatives, the chlorine unques- 
tionably is present as the specific radicle CI of hypochlorous acid:^ for the 
compounds are prepared from hypochlorous acid ; they form the acid by 
hydrolysis; they form hypochlorous esters with alcohols, hypochlorous 
amides with amines; in all of these reactions the specific oxidizing power 
of the hypochlorous acid radicle remains intact; all attempts to replace 
the chlorine atom by negative groups (OH, OR, NH2) have proved 
futile; all attempts to prepare cbloroamines by means of the hydrogen 
chloride radicle CI and its derivatives (HCl, PCI3, etc.) have likewise 
proved ineffectual. On the other hand, in compounds such as the 
alkyl chlorides R3C.CI and acyl chlorides R.OC.Cl, the chlorine is un- 
