34 THE VITAMINS 



with /?- and 7-hydroxypyridine, leading Williams to state: "The anti- 

 neuritic properties of these substances suggest that an isomerism 

 is at least partially responsible for the instability of the vitamin in 

 foodstuffs and that the antineuritic property may be inherent in the 

 potentiality of this type of isomerism. We may not conclude that 

 vitamins are necessarily hydroxypyridines since a similar isomerism 

 may exist in substances containing other heterocyclic nitrogenous nuclei 

 which are known to occur widely as constituents of animal tissues." 



With this conception in mind attention was directed to the natural 

 antineuritic substances of yeast. Starting with activated fuller's earth, 

 Williams and Seidell (1916) attempted to separate the vitamin from 

 it by extraction with acidified aqueous alcohol but failed in this at- 

 tempt on account of simultaneous extraction of aluminum compounds. 

 On shaking the activated fuller's earth with a five per cent solution 

 of sodium hydroxide in dilute alcohol and evaporating the extract, 

 they obtained a crystalline antineuritic substance the physiological 

 action of which was apparently not due to adhering mother liquor. In 

 attempts to purify this substance further by recrystallization they 

 found the resulting product to be identical with adenine and to have 

 no antineuritic properties. When, however, 10 milligrams of this 

 inactive substance was heated with one cubic centimeter of absolute 

 alcohol in a sealed tube at 180° C. for three hours it acquired anti- 

 neuritic properties and the power to give a blue color with the Folin- 

 Macallum phosphotungstic sodium carbonate reagent. These results were 

 thought to be in harmony with the conclusions of the previous paper 

 and to suggest the probability than an isomer of adenine is "the 

 chemical entity responsible for the characteristic physiological prop- 

 erties of the vitamin under investigation." 



Voegtlin and White on the other hand, reported (1916) that they 

 were unable to cause adenine to acquire antineuritic properties and 

 that there probably exists no relation between adenine and the anti- 

 neuritic vitamin. 



In a third paper Williams (1917) described further work with 

 synthetic substances from which he concluded that "the curative form 

 of a-hydroxypyridine is a pseudo base and that a structure conform- 

 ing more or less closely to the type of a betaine ring is probably an 

 essential characteristic of antineuritic vitamins." Attention was called 

 to the theoretical possibility of the existence of such a substance in 

 some of the simpler nitrogenous constituents of animal tissues, espe- 

 cially the nuclein bases. It was also suggested that nicotinic acid 

 may exist in a betaine form and that the curative properties of the 



