RADIATION AND VITAMINS 329 



tion or destruction of vitamin D (Bills, Honeywell, and Cox, 12). The 

 by-products which form by oxidation make difficult the crystallization 

 of the vitamin (Angus, Askew, Bourdillon, Bruce, Callow, Fischmann, 

 Philpot, and Webster, 1). In the experiments of Beard, Burk, Thomp- 

 son, and Goldblatt (6) extreme freedom from oxygen in a dry activation 

 did not increase the antirachitic potency of the product. As is usual 

 in the absence of a solvent, the potency attained was only moderate, 

 amounting in this case to roughly one-tenth of the value expected when 

 activation is performed in ether. The investigations of Windaus (89), 

 Angus, Askew^ Bourdillon, Bruce, Callow, Fischmann, Philpot, and 

 Webster (1) and Bills, McDonald, BeMiller, Steel, and Nussmeier (15) 

 show that vitamin D, either in crystalline form or as a resin, is decidedly 

 more stable to oxidation than some of the nonvitamin substances which 

 are formed with it in irradiation. Its stability is about the same as that 

 of ergosterol itself. 



Apart from any light-filtering action of solvents, and from any role 

 which they may play as carriers of dissolved oxygen, there appears to 

 be what Bills, Honeywell, and Cox (12) termed a specific solvent effect 

 on activation. This was revealed by parallel spectrographic and bio- 

 logical examinations on the course of activation in alcohol, cyclohexane, 

 and ether. The importance of the specific solvent effect was shown by 

 the fact that the time required for the attainment of maximum potency 

 in ether was longer than the time required for the entire sequence of 

 activation and destruction in alcohol. The maximum potency reached 

 in ether was higher than in alcohol or in cyclohexane, but the spectral 

 changes were most conspicuous in alcohol. 



THE INDIVIDUAL PRODUCTS OF IRRADIATION 



Attempts to characterize the individual products of irradiation and 

 to ascertain the number of them by interpretation of absorption spectra 

 have been numerous but not very profitable. The reason is easily under- 

 stood, for with mixtures of substances having overlapping absorption 

 bands, determinations of the position and height of the bands are unre- 

 liable. Precision methods of bio-assay and chemical methods of isola- 

 tion have been the principal means of progress during the later stages 

 of the study. 



A general survey of the irradiation products, with spectral absorption 

 curves, w^as given by Windaus, Liittringhaus, and Busse (99). Setz (77) 

 has brought the genetic relationships of the products up to date, and 

 confirmed the view that the changes involved are not reversible. It 

 appears from his work that when ergosterol is irradiated for not too long 

 a period, the product produced by wave-lengths below 2800 A is chiefly 

 tachysterol and vitamin D (calciferol), whereas the product of longer 

 wave-lengths is chiefly lumisterol and calciferol. 



