STRUCTURE OF THE VITAMINS A 697 



The formation of stereoisomers of vitamin A could be expected to occur 

 when sources of natural vitamin A are exposed to ultraviolet light, iodine, 

 or other forms of catalysis which are known to bring about such ti'ans — > 

 CIS rearrangements. Smith and his collaborators^^^ found that, when vita- 

 min A concentrates Avere exposed to ultraviolet light, a marked decrease in 

 the absorption maximum at 328 mn obtained, but that these readings 

 tended to revert to the original values when the samples were allowed to 

 stand in the dark. However, the more pronounced the lowering in absorp- 

 tion maximum at 328 m^, the less efficient was the subsequent recovery of 

 the absorption when the samples were again placed in the dark. These au- 

 thors ascribed "these irregularities [as being] probably due to differences in 

 proportion of cis-trans isomers of vitamin A." Thus, ultraviolet radiation 

 produces not only a reversible photochemical isomerization, but also a con- 

 comitant photochemical destruction. The presence of variable proportions 

 of such stereoisomers of vitamin A in natural oils was suggested by Smith 

 et al^*^ as an explanation for the variations in absorption maxima of dif- 

 ferent vitamin A preparations in several solvents. Morton ^^° reached the 

 same conclusions on the basis of fluctuations in the E (1%, 1 cm.) (328: 620 

 m/x) (antimony trichloride) for vitamin A in various mineral oils. Baxter 

 et al.,^^^ in 1941, reported on a non-crystallizable vitamin A obtained from 

 the liver oil of the ling cod {Ophiodon elongatus) which appeared to be identi- 

 cal with crystalline vitamin A but proved refractory to crystallization. 

 The first recognition of a stereoisomeric form of vitamin A was the isola- 

 tion of "neovitamin A" from natural sources by Robeson and Baxter.^^'^'^ 

 In the case of some fish liver oils, the neo form constituted as much as 35- 

 39% of the total vitamin A. The neovitamin A which was prepared as a 

 crystalline product proved to be identical with "non-crystallizable vitamin 

 A." Recently, Cawley et al.^^^ found that synthetic vitamin A preparations 

 contained neovitamin A to the extent of about one-third of the total; this 

 would constitute a ratio comparable to that found in some natural fats. 

 It is believed that a catalytic conversion, vitamin A ;;=^ neovitamin A, may 

 take place both in vitro and in vivo. 



Robeson and Baxter"*^ have found that marked variations in properties 

 and reactions obtain between vitamin A and neovitamin A. These are 

 Usted in Table 9. 



>« E. L. Smith, F. A. Robinson, B. E. Stern, and F. E. Young, Biochem. J., S3, 207-212 

 (1939) 



i« E. L. Smith, B. E. Stern, and F. E. Young, Nature, I4I, 551-552 (1938). 



«» R. A. Morton, Nature, I4I, 552 (1938). 



'*! J. G. Baxter, P. L. Harris, K. C. D. Hickman, and C. D. Robeson, /. Biol. Chem., 

 141, 991-992 (1941). 



'" J. D. Cawley, C. D. Robeson, L. Weisler, E. M. Shantz, N. D. Embree, and J. G. 

 Baxter, Science, 107, 346 (1948). 



