McCollum and Sim-monds 111 do not agree that the needs of the animal are for a.specific type of labile isomerism rather than a specific chemical complex, and suggest that the temporary relief of polyneuritis by so many unrelated substances, as noted above, may be the result of the pharmacological action of these substances resulting in the stimulation of certain of the cells of the body rather than renewed function of cells which have been subjected to a selective fast and later have been supplied with the missing food complex. In support of this view they point out that histological methods have shown polyneuritic animals to suffer a degeneration of the motor cells of the cord, and that this change is progressive, some cells presenting the normal appearance while others in the same field are degenerated.

111 McCoUmn and Siinmonds, J. Biol. Chem. 33,55, 1918.

"It would appear plausible that when the motor cell changes have reached a certain point loss of function supervenes and paralysis results. There still remain, in the cord of a paralyzed animal, motor cells which appear normal when stained and which may be capable of restoring the motor functions of the muscles when influenced by substances which stimulate them to heightened sensitiveness. ... If this line of reasoning could be shown to be valid, it would follow that experiments with pure chemical substances of known constitution, with a view to finding by good fortune the one playing an important physiological role, might be entirely misleading unless it were shown that the 'cure' was permanent. For such complete proof it is necessary to demonstrate the resumption of growth and maintenance of health as long as the substance is supplied in the food mixture. . . . Sustained normal function is indispensable to adequate proof that the dietary essential in question is being adininistered".

While we are still far from a definite conclusion as to the chemical structure of the vitamin, there are certain properties in addition to those already referred to which are invariably associated with the physiologically active substance. It appears to be basic,112 although Osborne, Wakeman and Ferry113 suggest that the persistence with which it is retained by edestin may indicate that it is chemically combined with the edestin, and hence that it has acid properties. It is completely precipitated by silver nitrate and barium hydroxide,114 by phosphotungstic acid in 5 per cent sulphuric acid solution,115 giving a phosphotungstate which is insoluble in acetone,116 and almost entirely by ammonium molybdate.117 It can be precipitated by tannic acid,118 but this method when applied to rice polishings does not yield large amounts.119 Drummond120 found that upon fractionation of yeast dialysate with silver nitrate and baryta, the precipitate produced with silver nitrate, which contained the purine fraction, had no effect on growth, while that produced on the further addition of baryta and containing the pyrimidine contained small traces of vitamin.

112 Punk, Trans. Soc. Tr. Med. 5, 86, 1911; J. Physiol. 43, 395, 1911; Ved-der, Phil. J. Sci. 7B, 415; Osborne and Wakeman, J. Biol. Chem. 40, 383; Myers and Voegtlin, J. Biol. Chem. 4a, 199, 1920.

113 Osborne, Wakeman and Ferry, J. Biol. Chem. 39, 35, 1919.

114 Funk, J. Phys. 43, 395, 1911; Cooper, Bioch. J. 7, 268, 1913; Edie, Evans, Moore, Simpson and Webster, Bioch. J. 6, 234, 1912.

115 Cooper and Funk, Lancet, 1911, ii, 1266; Funk, J. Physiol. 43, 395, 1911; 45, 75, 1912; Funk and Macallum, J. Biol. Chem. 27, 63, 1916; Eustis and Scott, Bioch. Bull. 3, 466, 1914.

116 Funk, Bioch. Bull. 5, i, 1916; Drummond, Bioch. J. 11, 255, 1917.

117 Cooper, J. Hyg. 12, 436, 1913.

118 Suzuki, Shinamura and Odake, Bioch. Ztschr. 43, 89, 192.

119Vedder and Williams, Phil. J. Sci. BS, 175.

120Drummond, Bioch. J. 11, 255, 1917.

It is partially precipitated by mercuric chloride121 but not at all by basic lead acetate,122 platinic chloride,123 or picric acid.124

With the phosphotungstic acid solution introduced by Folin and Macallum125 as a test for uric acid it gives a deep blue color,1" and a similar color is obtained with the phosphomolybdic-phosphotungstic acid reagent of Folin and Denis.127 Eddy128 believes from his experiments that this color reaction offers a practical method of standardizing dosage.

The "uric acid reagent" has been found to give a reaction with di- and polyhydric phenols, monohydric phenols containing an NH, group in the benzene ring, and certain. purine and tyrosine derivatives, while the phosphomolybdic reagent appears to be specific for purine derivatives and polyphenols. Amino acids, polypeptides, and diketopiperazines are negative to both reagents. The replacement of one hydrogen atom in the purine ring lessens or abolishes the uric acid reaction. In the case of phenol reagent this occurs when two hydrogen atoms are replaced.129 Referring to the color produced by the Folin-Macallum reagent, Williams and Seidell130 state:

It has been our experience that a negative test for this oolor indicates the absenee of the curative substance in crude solution. A positive test can with less assurance be taken to indicate the presence of the vitamins.

121 Punk, J. Physiol. 43, 305, 1911; Bioch. Bull. 5, 1, 1916.

122 Cooper, J. Hyg. la, 436, 1913; Vedder and Williams, Phil. J. Sci. B8, 175, 1913; Funk, Bioch. Bull. 5, 1, 1916.

123 Funk, 1. c.; Susuki and Shinamura, J. Tok. Chem. 80c. 32, Zentr. Bioch. Biophys. 12, 11; Suzuki, Shinamura and Odake, 1. c; Steenbock, Proc. Am. Physiol. 80c, Am. J. Physiol. 42, 610, 1917.

124 Funk, 1. e.; but cf. Susuki, Shinamura and Odake, 1. c.

125 Folin and Macallum, J. Biol. Chem. 11, 266; 13, 363; Benedict and Hitchcock, Ibid. 20, 619, 1915.

126 Funk and Macallum, Bioch. J. 7, 356, 1913.

127 Folin and Denis, J. Biol. Chem. 12, 239; Suguira, J. Biol. Chem. 1918, 36, 191.

128 Eddy, Proc 80c. Exp. Biol. Med. 14, 164, 1917.

129 punk and Macallum, Bioch. J. 1913, 7, 356.

130 Williams and Seidell, J. Biol. Chem. 26, 431, 1916.