Shortly after this work was done there appeared a paper on this subject by Edie, Evans, Moore, Simpson, and Webster,78 who used a slightly modified method. The source of their material was yeast. This was extracted with alcohol, and the alcohol extract evaporated, cleared with lead acetate, and immediately precipitated with silver nitrate. This precipitate, when decomposed, gave a small quantity of crystalline organic substance that yielded ash. These investigators gave the name torulin to the substance isolated, but did not claim to have tested its curative power.

Later, in 1912, Funk endeavored to isolate the vitamin from different foodstuffs,79 e. g., yeast, rice polishings, ox-brain, milk,80 and lime juice.81 An extract of rice polishings was precipitated directly with silver nitrate and baryta. This process was chosen to shorten the procedure and to avoid the use of. a large quantity of alkali, which was supposed to destroy the vitamin. No vitamin was detected in this case. Allantoin was found, which, when phosphotung-stic acid is used, goes into the filtrate. The evaporated alcoholic extract of yeast was hydrolyzed with 10 per cent sulphuric acid solution for a short time previous to precipitation with phosphotungstic acid, and the hydrolyzate was worked up in a way similar to that for the first fractionation of the extract of rice polishings, with the difference that mercuric chlorid was not used. The silver-fraction when decomposed yielded a crystalline substance, melting at 233° C, that gave a precipitate with mercuric acetate, but not with mercuric sulphate or nitrate. No copper salt was formed in boiling water with copper oxide. The yield, 0.45 g. from 75 k. of dry yeast, was too small for recrystallization or analysis; a rather large dose of the substance (0.02 to 0.04 g.) proved, however, to be curative for beriberi pigeons.

In another fractionation of yeast the alcoholic extract was simply extracted with water and not hydrolyzed; in this case the substance described above could not be isolated but some pyrimidine bases, e. g. uracil and thymin, were obtained.

78 Edie, Evans, Moore, Simpson, and Webster, Bioch. J. 6, 234, 1912. 79J. Physiol. 45. 75, 1912. 80 Bioch. J. 7. 211. 81Ibid, 81.

In a later attempt 82 to obtain the vitamin from yeast, a crystalline product was obtained from the silver and baryta fraction, which melted at a temperature above 200° C, and which when administered to beriberi pigeons induced very quick recovery. When, however, maintenance experiments were performed with pigeons on a diet of polished rice given daily injections of the product, the birds could not be kept alive for more than a few days. The product of this first crystallization was divided into three substances, one of which was nicotinic acid, the other with the composition C24H10,O9N5 and C29H2809N5 respectively. The administration of these substances, individually or collectively, to beriberi pigeons did not show appreciable curative action.

From the vitamin fraction of rice polishings two crystalline products were obtained, one of which proved to be nicotinic acid and the other having the formula C26H20O9N4.

In the same year Suzuki, Shinamura and Odake83 published a paper on the chemistry of rice polishings. Their method of isolating the active principle was as follows: The rice polishings were first extracted with petroleum ether, to remove the fat, and then with boiling alcohol. The evaporated alcoholic extracts were diluted with 3 per cent sulphuric acid solution, and precipitated with phospho-tungstic acid. The precipitate, after decomposition with baryta, yielded a fraction to which they gave the name oryzanin. The crude oryzanin obtained by this process was purified by precipitation with tannic acid, and from this a picrate was obtained which was crystal-lized, by evaporation from acetone, in small yellow-brown needles. These, on decomposition, yielded a very active curative substance. After boiling two hours with 3 per cent hydrochloric acid, the crude oryzanin yielded choline, glucose, and two acids, C10H8NO4 and C18H18N209. Both acids are difficultly soluble in cold water, somewhat more soluble in hot water, readily soluble in alkali and alcohol. Both give an intense diazo reaction with p-diazobenzol-sulphonic acid, a deep indigo blue with phosphomolybdic acid and ammonia, a strong Millon reaction, and decolorize iodine-starch solution.

These reactions are also given by crude oryzanin.83a.

82 Punk J. Physiol. 46, 173, 1913.

83 Suzuki, Shinjunura and Odake, Bioch. Ztechr. 43, 89, 1912.

Drummond and Funk 84 repeated this work but were not able to confirm it. Neither were the above mentioned acids detected, nor was a curative picrate obtained. On the contrary it was found that the vitamin was not precipitated by picric acid. It was hoped that by using large quantities of rice polishings positive results could be obtained. In the first instance 380 kilos of rice polishings were used and the method carried out exactly as described above. In another case an extract from 620 kilos of rice polishings was worked up. This extract was hydrolyzed with acid previous to fractionation. In both cases a most careful inquiry failed to detect any new chemical substance that could be regarded as the vitamin itself or a decomposition product of it.

Funk85 concluded that the vitamin is extremely unstable and that methods employed in separating are likely to destroy it largely if not completely. Of the various destructive agencies he regarded the action of alkali as probably the most important and therefore endeavored to find a satisfactory method of separation, which would avoid the use of alkali. He found86 that the phosphotungstic precipitate from alcoholic extract of yeast can be divided by means of acetone into a small insoluble fraction which contains the bulk of the vitamin, and a large soluble fraction which is totally inactive. By using lead acetate instead of baryta to decompose the phospho-tungstates very clear solutions are obtained which facilitate further purification and the possible destructive action of the alkali is eliminated.