Recognizing the same difficulty in the use of alkali, Vedder and Williams87 and Sullivan and Voegtlin88 sought to avoid it by use of barium acetate and lead acetate respectively to decompose the phosphotungstic precipitates, and report some measure of success.

Pol89 separated a crystalline acid which he assumed to be the active principle from a kind of bean, the katjang id joe (Phaseolus radiatus) which was used by Grijns90 for the cure of polyneuritis.

83a Kaoliang (Andropogon sorghum, Brot.) is reported by Kimura (J. S. C. I. 1921, 672 A) to be rich in orizanin. 84 Drummond and Punk, Bioch. J. 1914, 8, 598. 85 Funk, Bioch. Bull. 5, 1, 1916. 86Funk, Bioch. Bull. 5, 1, 1916.

87Vedder and Williams, Phil. J. Sci. 1913, 8, 175, 183. 88 Sullivan and Voegtlin, J. Biol. Chem. 1916, xxiv. 89D. J. Hulshoff Pol, J. Physiol. 51, 432, 1917. 90 Grijns, Geneesk, Tyds. v. Ned. Indie, 41, 1901.

For the isolation of this acid the beans were boiled in excess of water, the fluid was decanted and basic lead acetate added. The precipitate was collected on a filter, well washed with water, and then water added to it till it formed a thin pap. Through this hydrogen sulphide was passed to remove the lead, and carbon dioxide was bubbled through the filtrate to remove the hydrogen sulphide. The fluid thus obtained was administered to four out of a number of beriberi patients, all of whom recovered. To isolate the active principle further, the purified extract was evaporated to dryness and crystals obtained. Both the extract and the crystals were acid. The constitution of the acid was not determined.

Suguira91 prepared a colorless, crystalline substance capable of exerting a curative action on polyneuritic pigeons from dried, powdered brewers' yeast by treating it with ten times its weight of 5 per cent sodium chloride solution, and subjecting it to air dialysis. By this method, 22 mg. of colorless, crystalline substance, almost free from sodium chloride, can be obtained from 10 grams of dried yeast. A solution in water of from 2 to 5 mg. of these colorless crystals injected subcutaneously into a pigeon may be sufficient to remove completely the symptoms of polyneuritis within a few hours. Its constitution was not determined.

Hofmeister 92 isolated an antineuritic substance from rice to which he gave the name oridine. It was found to be isomeric with betaine and valine and from its properties would seem to be related to the pyridine and piperidine groups.

The hydrochloride was prepared as follows: Rice meal was extracted with 80 per cent alcohol, the alcoholic extract evaporated to a syrup in a vacuum, and made up with hydrochloric acid to a solution containing 3 per cent of the acid. After extraction of impurities with ether, the acid solution was concentrated again to a syrup and made feebly alkaline with sodium carbonate, and precipitated with bismuth-potassium iodide solution. After acidifying the filtrate, the precipitation was repeated, the precipitate being triturated with silver carbonate and decomposed with hydrochloric acid. The substance was then obtained in crystalline form on drying the filtrate. Doses of 5 to 10 mg. of the hydrochloride were found to be active. The aurichloride, C5H11O2N,HAuCl4, formed platelets and flat prisms, m. p. 277°. Oridine hydrochloride, C5H1102N,HC1, forms colorless prisms, m. p. 240°. The free base was a white, hygroscopic powder, soluble in water. On purification the compound lost its activity.

91 Suguira, J. Biol. Ghem. 36, 191, 1918. 92 Hofmeifter, Bioch. Ztsohr. 1920, 103, 218.

Myers and Voegtlin93 have recently obtained an active crystalline compound from autolysed yeast extract by a modification of Funk's method which eliminates purines, histidine, proteins and albumoses. Their procedure is as follows:

Air-dried yeast is ground to a fine powder and twice extracted for three hours with boiling methyl alcohol, containing 0.01 per cent hydrochloric acid, in the proportion of 2 c. c. of alcohol to each gram of yeast. The alcohol is removed in vacuo at 35° C. and the residue repeatedly extracted with ether and 0.1 per cent hydrochloric acid, keeping the volume as small as possible. From this acid extract the purines are precipitated by hot aqueous silver acetate, very little of the active principle being carried down in this fraction. To the filtrate, containing a large excess of silver acetate, saturated barium hydroxide solution is added to precipitate the vitamins along with other extraneous material. This precipitate is suspended in water, acidified with sulphuric acid, and decomposed with hydrogen sulphide, the filtrate treated with slight excess of lead acetate to remove the sulphuric acid, and again with hydrogen sulphide to remove the lead, then concentrated in vacuo. Up to this stage, according to Myers and Voegtlin, practically none of the activity is lost. On addition of mercuric sulphate to the concentrated vitamin solution an inactive histidine fraction is precipitated.

The filtrate containing an excess of mercuric sulphate is treated with absolute alcohol until precipitation is complete. The pale yellow precipitate formed contains the bulk of the active material. " The degree of separation in this last procedure is influenced by two factors, (1) the concentration of excess mercuric sulphate present and (2) the final concentration of ethyl alcohol. If these adjustments are right, none of the active material remains in the alcoholic filtrate." The alcohol-insoluble material is suspended in water and the mercury removed by means of hydrogen sulphide. The filtrate from the mercuric sulphide is freed of hydrogen sulphide in vacuo and the sulphuric acid removed by means of lead acetate and the excess of lead by hydrogen sulphide. By concentration in vacuo over soda lime a definitely crystalline product was obtained by Myers and Voegtlin, which crystallized in spindle-shaped crystals and was active as long as the crystals were surrounded by the mother liquor. When washed with absolute alcohol and dried, the crystals lose their activity and change from spindles to prisms, but on recrystallization from water, they take on the spindle shape again.

The investigators believe that there are at least two substances in the final solution, both possessing a distinctly basic character.93a.

93 Myers and Voegtlin, J. Biol. Chem. 42, 199, 1920.

93a Yeast extract does not cause secretion of pancreatic juice as does secretin. Secretin can be extracted from the intestine of a cat showing the so-called polyneuritic condition to a very marked degree. Vitamin B has a curative effect on polyneuritic animals when given by the mouth, whereas secretin has no action on the secretion of pancreatic juice when so administered. The suggestion of Voegtlin and Myers that vitamin B and secretin are identical is not supported is the conclusion reached by Anrep and Dmmmond (J. Physiol. 54, 249, 1021).