Chick and Hume52 criticize most of the research on vitamins on the ground that the consideration of quantity has been overlooked to a large extent in the experimental methods adopted, and that the conclusions drawn have been correspondingly untrustworthy.

In comparing the value of a series of foodstuffs as regards their value in content of some accessory food factors, the first step necessary, according to Chick and Hume, is to determine in each case the minimum daily dose which will maintain health in the experimental animal and to institute comparison between these amounts. These daily doses should be as large as can be tolerated before it is safe to deny to a foodstuff any value in respect of the particular accessory factor under consideration. Neglect of this simple and fundamental, but admittedly tedious procedure is evident in much published work upon the influence of various treatment upon the antineuritic and antiscorbutic and other properties of certain foodstuffs, and the results obtained are in consequence vague, and may even be erroneous. For example, if a foodstuff is rich in any particular accessory factor and is fed to an animal in large excess of the amount actually required, any destruction of the factor that may take place during heating cannot be assessed and may even remain undetected if the heated material is fed in similar large amount.

It is held necessary to determine the minimum required in both cases, and to institute a comparison between these two results before any satisfactory conclusion on this point can be drawn.

The methods used by Osborne and Mendel and by Hopkins for determining the food intake have already received mention in this volume. Where possible, it seems distinctly advantageous to feed the material to the tested separately from the basal ration, so that this can be estimated separately and no error may be introduced should the animal fail to consume the whole ration.

51 Osborne and Mendel, J. Biol. Ghem. 34, 623, 1918. See also Steenbock, Boutwcll and Kent, Ibid. 35, 518, 1918.

52 Chick and Home, J. Biol. Chem. 39, 203, 1919.

It is evident from the foregoing that the problems of vitamin research are far from simple, and that the correct interpretation of the results requires an extensive knowledge of all the detail of the experiment. It is extremely difficult, as Davis53 has pointed out, to secure uniform results in one laboratory, much more so to evaluate and correlate the results from a hundred different laboratories in all parts of the world.

In an attempt to develop a simple standard test for at least one vitamin, B, R. J.Williams and Bachmann have both made use of the fact that certain yeasts fail to grow satisfactorily in vitamin-free media, whereas on the addition of vitamin, growth takes place at a rate which is at least roughly proportioned to the amount of the vitamin present.

Bachmann 54 noted that Saccharomycetes grow better and ferment more readily in a medium containing small amounts of organic material but vary in their needs for traces of organic matter other than sugar. Some will grow in a solution of inorganic salts and sugar even when introduced in small amounts, as observed by Pasteur. A certain yeast isolated from fermenting canned pears was found not to grow in or ferment Nageli's solution unless introduced in very large amounts or some organic substance was added. The substances which cause fermentation are those which have been found to be rich in vitamins (especially water-soluble B), such as yeast extract, vegetable juices, milk, and cabbage extract, apparently containing the most efficient vitamins for yeast.

It is suggested that in a heavy inoculation some of the cells are dead, or that because the medium is unfavorable some may die, and that from these dead cells soluble substances that are stimulating to growth and fermentation of other cells diffuse out into the medium.

The liquid medium used for testing vitamin requirements of yeast was Nageli's solution, composed of 100 cc. distilled water, 10 g. dextrose, 1 g. ammonium nitrate, 0.05 g. calcium phosphate, 0.5 g. potassium acid phosphate, and 0.25 g. magnesium sulphate.

None of the salts were especially purified by recrystallisation. Durham tubes were usually used to show gas production and the amount of gas produced, approximately 8 cc. of the medium being placed in these tubes.

53 Davis, J. Home Ec. 12, 206.

54Bachman, J. Biol. Ghem. 39, 235-67, 1919.

Of two strains of yeast examined, one, an aerobic type, was able to cause fermentation in a medium containing little or no vitamins, while the other, which does not produce a surface film and is therefore anaerobic, will not ferment except in the presence of vitamins.

Williams used an entirely different technique, but came to the same general conclusions as to the importance of vitamins in the life-processes of the yeast cell. His method consisted in making a hanging drop culture of a single cell, and noting the rate of multiplication. A single yeast cell can be caused to produce from one to several thousand cells in 24 hours by varying the vitamin content, and it was found that the number of cells produced under given conditions in 18 hours is directly proportional to the amount of vitamin added. Inasmuch as several investigators have used this method or a modification of it, the experimental details are given, although Williams himself has now superseded it by a gravimetric method which he asserts to be more accurate and more easily applied.

The medium used was made up of purified reagents as follows: saccharose, 20 gm.; ammonium sulphate, 3 gm.; potassium dihydrogen phosphate, 2 gm.; asparagine, 3 gm.; calcium chloride, 0.25 gm.; magnesium sulphate 0.25 gm., all made up to 1 liter with distilled water. The media was sterilized at 10 pounds pressure for 10 minutes and kept in the refrigerator.

A suspension of yeast cells was made in about 30 cc. of sterile distilled water. Duplicate test solutions were prepared quantitatively by taking 25 cc. of the synthetic media and 1 to 5 cc. (depending upon the concentrations) of a definite solution of the unknown extract. This was diluted to 30 cc. (using sterile water if needed). The mixture was sterilized and 1 cc. of the yeast suspension added. After mixing, thirty-six drops were made with a fine pen point on a cover slip that had been coated with a very thin film of purified vaseline. This slip was inverted and sealed tightly on a hanging drop slide. Each drop was examined for single yeast cells, the observations were recorded, and the time was noted. The slide was then placed in the incubator at 30° C. and the readings were made again in 18 hours. In case the number of cells in each drop at the end of this period exceeded 75, the determination was repeated using less of the unknown. Similar tests were run on the synthetic media as a control. After correcting for this blank determination, the rate of growth in terms of yeast cells was calculated per gram of the original unknown material.

This method was used satisfactorily by Eramctt and Stockholm55 but Whipple found it unsatisfactory.56

55 Emmett and Stockholm, J. Biol. Chem. 43, 287, 1020. 56Whipple, J. Biol. Chem. 44, 175, 1920.