A set of experiments was made with wheat germ extract treated with 2-4 per cent of sodium bicarbonate and autoclaved, in order to destroy as much as possible of the antineuritic substance. Addition of these extracts exerted a profound stimulating effect on the growth of yeast, but the extent of acceleration of growth was not proportional to the amount of extract added. This germ was used in a feeding experiment with rats to show whether it still contained any appreciable amount of B. The ration consisted of casein (18%), treated wheat germ (15%), butter-fat (5%), salt mixture (3.7%), and dextrin (58.5%). The animals failed to grow. In a subsequent period rapid growth took place when 3 per cent of untreated germ was included in the diet. All the antineuritic substance had been practically destroyed by heating with alkali.

In another set of experiments the effect of glucose, with and without wheat germ extracts, was studied. The results were not always consistent, but it was evident that glucose had a profound stimulating effect. In one case greater growth was secured by adding to 10 cc. of the control solution 1 cc. of 10 per cent glucose solution than was obtained by addition of 0.1 cc. of 75 per cent alcoholic extract of wheat germ without additional glucose. No modifications which were introduced were successful in inducing concordant results. It seems impossible to avoid irregularities due to the introduction of injured cells or cells of low reproduction power into the test specimens.

62 Sousa and McCollum, J. Biol. Chem. 44, 113, 1020.

In a set of experiments various extracts so treated (with heat under pressure in presence of sodium bicarbonate) that they were no longer capable of producing growth in rats were found to be still very effective in stimulating growth of yeast. Among these were extracts of fresh beef made by boiling the latter with water after it had been treated with 2 per cent sodium bicarbonate, moistened and allowed to stand for three hours, then autoclaved for one hour at 15 lbs. pressure. Extract of rolled oats similarly treated was also used. The oats so treated were fed to young rats in a diet of rolled oats, gelatin, casein, salt, butter-fat and carbonate of lime. The animals failed to grow. After 41 days 3 per cent of untreated wheat germ was added to the diet to furnish the antineuritic substance. The animals responded at once with rapid growth. 40 per cent of untreated oats furnished an abundance of antineuritic B, for growth. The treatment with alkali had, therefore, destroyed its content of this substance while leaving the yeast growth-stimulating power apparently unimpaired. Similar results were obtained with extracts of meat which had been digested either with alkali or acid.

Sousa and McCollum state that so far as their knowledge permits them to judge these digested meat extracts could serve only as a source of amino acids since the treatment was sufficient to reduce the proteins practically to the amino acid state, and, since muscle tissue is, even in the fresh condition very poor in the antineuritic, B, factor, the same meat after hydrolysis with strong sulphuric acid would scarcely contain a trace of it. Yet these meat extracts exerted a strong influence on the rate of development of yeast, approximately doubling the number of cells counted after 20 hours incubation.

If, then, glucose and amino acids can exert such stimulation on the growth of yeast it would seem to be conclusively demonstrated that the use of yeast as a test organism for determining the presence or absence of the antineuritic dietary factor is complicated by so many disturbing factors as to make it of little if any value.

Confirmatory evidence has been obtained by Eddy, Heft, Stevenson and Johnson 63 that the yeast test is not a quantitative measure of the content of vitamin B. The presence of B, however, possibly is one of the factors causing the stimulation of the growth of yeast by vegetable extracts.

63 Eddy, Heft, Stevenson and Johnson, J. Biol. Chem. 47, 249, 1021; J. 8. C. I. 1021, 713 A.

Moreover, evidence has been accumulated which appears to demonstrate conclusively that yeast can synthesize the vitamin B which it requires. Fulmer, Nelson and Sherwood 64 found that yeast would grow for months in a vitamin-free medium, at two-thirds the rate manifested in wort. Alcoholic extract of alfalfa stimulated growth, but heating the extract with alkali did not destroy this effect, and alcoholic extracted malt gave results like untreated malt. The growth was markedly influenced by the concentration of the ammonium salt used as nutrient, and by the presence of colloidal substances. They concluded that it is unwarrantable to state that any one substance is indispensable until the best synthetic medium for yeast has been developed. A study of the effect of alcoholic extracts of wheat embryo and lucerne led them to the belief that the growth-stimulant present in these extracts was not identical with vitamin B.65

MacDonald and McCollum 66 grew yeast in nutrient solutions free from vitamin, and found that a pure culture grew as well at the end of a series of 15 consecutive seedings in this solution as it did earlier, and concluded that the yeast plant must either be able to grow without vitamin or else to synthesize this factor in sufficient quantity for its own use. The latter assumption finds confirmation in the work of Nelson, Fulmer and Cessna,67 who made 180 consecutive seedings of yeast by adding 1 c. c. to 50 c. c. of a solution containing only carefully purified salts and sucrose. The concentration of the original constituents in the final solution was only 1 X 50 -180, yet it contained enough vitamin B to restore rats, fed on a vitamin-poor diet, to normal growth.

It seems highly probable that the growth-stimulant of the American investigators is identical with the "bios " of Wildiers, Amand and Devloo,68 but it is as yet impossible to say whether this is the same as vitamin B or distinct from it. Frankel and Schwarz69 endeavored to isolate the accelerating substance from rice and yeast sediment, but the attempt was unsuccessful.

64 Fulmer, Nelson and Sherwood, J. Am. Chem. Soo. 43, 186; 191, 1021.

65 See also Bmmett and Stockholm, J. Biol. Chem, 43, 287, 1020.

66 MacDonald and McCollum, J. Biol. Chem. 45, 307; 46, 525, 1021.

67 Nelson, Fulmer and Cessna, J. Biol. Chem. 46, 77, 1021.

68 Wildiers, La Cellule, 18, 313, 1001; Amand, lb. 21, 320, 1004; Devloo, lb. 23, 361,1006; Ide, Centr. Bakt. Abt 2,1007; 18,103; J. Biol. Chem. 46, 521, 1021. 69 Frankel and Schwars, Bioch. Z. 112, 203, 1920.