292 



Such an intermediate detachment of the radicals round the 

 asymmetric atom, but under simultaneous formation of ethylene- 

 derivatives, was also previously proposed by Nef *) in his explana- 

 tion of phenomena of rearrangement in general. It still, however, 

 remains unexplained, why such a change of place does not always 

 lead to complete racemisation, as both antipodes have the same 

 mechanical stability. 



However, all these suggestions can hardly be considered other- 

 wise than made "pour besoin de la cause", and to be hypotheses 

 which in general appear quite arbitrary, although perhaps they are 

 useful in some special cases. Frankland, in his summary of 

 the facts observed up till now with respect to the remarkable and 

 highly important phenomenon of the Walden-inversion, is un- 

 doubtedly quite right, when he states: "it appears that there 

 does not exist at the present time any criterion by which the 

 relation between the configuration of an optically active compound 

 and that of its derivative, can be decisively ascertained". All 

 assumptions as to the occurrence or non-occurence of a change 

 of configuration during such a substitution, appear to be merely 

 arbitrary and hypothetical. 



Such assumptions for instance are: that the change of rotatory 

 power from positive to negative, and vice versa, if PC1 5 or PBr 5 

 act upon hydroxy-acids or their ethers, is really accompanied by 

 an inversion of the configuration round the asymmetric carbon- 

 atom; that eventually the same takes place when the halogen- 

 acids: HCl, HBr, etc., themselves act upon such hydroxy-acids 2 ); 

 that the same occurs when SOC1. 2 acts upon such molecules 3 ) ; 

 that NOBr and HN0 2 , acting upon ammo-acids, do not cause a 

 change of configuration 4 ); that, on the other hand, the substitution 

 of NH 2 by a halogen-atom is commonly accompanied by a change 

 of configuration; etc. etc. 



The question is undoubtedly enormously complicated. In truth, 

 every substitution of an atom or radical by some other, must 

 always have a distortion of the molecule as a necessary conse- 



!) J. U. Nef, Journ. Amer. Chem. Soc., 30, 746, (1908). 



2 ) A. MacKenzie and H. B. P. Humphries, Journ. Chem. Soc. London, 97, 121, 

 (1910); A. MacKenzie and H. Wren, ibid., 97, 1355, (1910). 



3 ) A. MacKenzie and F. Barrow, Journ. Chem. Soc. London, 99, 1923, (1911). 

 R. H. Pickard and J. Kenyon, Ber. d. d. Chem. Ges., 45, 1592, (1912). 



4 ) E. Biilmann, Ann. der Chemie, 388, loco cit., (1911). 



