II. CHEMISTRY 203 



Vitamin D3 was obtained in crystalline form by Schenck^^^ in 1937. The 

 isolation procedure was essentially the same as for calciferol, i.e., purifica- 

 tion via the 3,5-dinitrobenzoate, followed by saponification. Vitamin D3 is 

 more difficult to crystallize than D2, and the yields are inferior.^^^ • ^^^- ^^^ 

 There has been described a double compound of the vitamin with cholesterol 

 (or with cholestanol or coprosterol), which crystallizes in good yield and 

 represents the most economical way to obtain a preparation of uniform 

 purity.^^'^' ^^' The constants of the esters of vitamin D3 are shown in Table 

 VIII. 



Pure crystalline vitamin D3 is slightly more stable than D2, and much 

 more stable than its precursor, 7-dehydrocholesterol. Considerable deteri- 

 oration occurs in 72 hours when the vitamin is exposed to air at room tem- 

 perature, even in brown bottles. Deterioration is negligible after 12 months' 

 storage in evacuated amber ampules at refrigerator temperature.^^'^ The 

 complex with cholesterol is less stable than the pure vitamin. ^^^ As with 

 calciferol, the various nitrobenzoic acid esters of vitamin D3 are remarkably 

 stable. They remain unchanged for at least 5 years with no special pre- 

 cautions other than the avoidance of light. Solutions of vitamin D3 in edible 

 oils or in propylene glycol and dispersions of it in canned milk have excel- 

 lent keeping qualities, even under severe conditions. ^^'^ 



Vitamin D3 is sensitive to hydrogen chloride. -^^ It does not give a pre- 

 cipitate with digitonin,^^^ It gives a yellow color in the antimony trichloride 

 reaction.^" In these respects, as well as in optical properties and chemical 

 reactions, it reveals its similarity to calciferol. The similarity in most 

 non-biological properties and reactions of these two forms of vitamin D 

 supports the quip that only a bird can tell them apart. 



5. Minor and Obscure Forms 



The number of steroids which exhibit antiricketic action is considerable, 

 but many of them are so vaguely characterized that an exact count is im- 

 possible. In attempting to enumerate them, one is confronted with the in- 

 adequacy of many reported animal tests and wdth the probability that, if 

 tests were conducted differently, the number of recognized vitamins D 

 would be larger. There is also the problem of contamination, which has 

 confused investigators ever since the first crude cholesterol was irradiated 

 and which now is involved in the so-called vitamin D5, or activated 7-de- 

 hydrositosterol. Lastly, there is the factor of the occasional bad assay, 

 which, according to the statisticians, appears even in the best-regulated 

 laboratories at unpredictable times. 



"^ F. Schenck, N aturwissenschaften 25, 159 (1937); A. Windaus and F. Schenck, 



U. S. Pat. 2,099,550 (1937). 

 3'8 A. Windaus, F. Schenck, and F. von Werder, Hoppe-Seyler's Z. physiol. Chem. 



241, 100 (1936). 

 31^0. Linsert, U. S. Pat. 2,264,320 (1941). 



