Manchester Memoirs, Vol. Ixiv. (1920), No. 4 



R. G. Farg-her and F. L. Pyman (Trans. Chem. Soc, 1919, 

 115, 217, 1015) is covered by" the general statement and could 

 perhaps have been predicted with its aid. All true intra- 

 molecular changes can be similarly generalised and the 

 number of distinct types may be comparatively small. 

 Reference may be made here to "the dehydration of "pinacone 

 to pinacoline and that of borneol to camphene, two apparentlv 

 dissimilar reactions which may nevertheless be brought under 

 the same heading. In the expression 



ax y " R 3 



I 1/ 



R— C C^-R 2 



I 

 R 1 



R, R 1 , R 2 , R 3 are neutral groups, e.g., alkyl or aryl groups. 

 a is a divalent atom or group and x, y, are monovalent atoms 

 or groups of such a character that there is a strong tendency 

 to form the compound xy. For purposes of convenience (and 

 because it usually is so) x will be assumed to have electro- 

 positive character and y is electronegative. If x and y 

 become attached by a partial valency then conjugation occurs 

 and we have at once the conditions : — 



? •*— T # 



s — cr-fc *fcf* or p—'c 



1 ix-'V 1 " c \ 



R> 



(A) (B) 



since ring formation of the partial valencies is clearly impos- 

 sible unless there are an even number of atoms in the ring. 

 If now the process is continued in the same direction and the 

 compound xy is separated, the result will be in case B that 

 a three-membered ring will be formed : — 



R a R 2 



\ / \ / 

 C C 



/ \ 



R 1 R 3 . 



This occasionally happens, but in most cases the structure is 

 unstable and undergoes ring scission under appropriate condi- 

 tions reverting to the original substance or another compound 



