468 Proceedings Columbia Biochemical Association [April 



The iodin reagents for fat analysis cannot be used in the Volu- 

 metrie determination of cholesterol. The iodin number of pure 

 cholesterol has been determined by the methods of Hübl, Hanns 

 and Wijs. The Hübl method tended to give the lowest values, 70.3- 

 78.0, but even these values are higher than the theoretical value, 

 65.7, which is based on the assumed existence of one double bond 

 in the molecule of cholesterol. The Hanns method gave higher 

 results, 71.4-81.1. The highest values of all were obtained with 

 the Wijs method, 55.1-158.9, with an average value above 100. 

 There was a marked tendency, especially with the Hanns and 

 Wijs methods, for the iodin number to become higher, the greater 

 the excess of the iodin reagent. However, the iodin numbers were 

 not simple multiples of 65.7, hence the presence of a second double 

 bond in cholesterol is doubtful. 



The gravimetric determination of cholesterol as the free alcohol, 

 by modifications of Ritter's method, is unsatisfactory. When car- 

 bon dioxide was used to neutralize the excess of sodium ethylate, 

 divergent results were obtained : 99.90 per cent. and 92.43 per cent. 

 of the cholesterol taken was recovered. When hydrochloric acid 

 was used to neutralize the excess of sodium ethylate, the results 

 were still less satisfactory, 64.34 to 89.10 per cent. of the cholesterol 

 taken being recovered. The gravimetric determination as choles- 

 teryl benzoate is not quantitative; only 24.29 to 61.79 (average 

 42.86 per cent. ) of the cholesterol taken was recovered. The gravi- 

 metric determination of cholesterol as the free alcohol by Cappen- 

 berg's method gave excellent duplicates; 94.47 and 94.37 per cent. 

 of the cholesterol taken was recovered. The gravimetric determina- 

 tion of cholesterol as digitonin cholesteride was the most accurate 

 and most satisfactory of the methods studied. From 93.63 to 

 103.02, average 97.37 per cent., of the cholesterol taken was 

 recovered. 



85. A preliminary study of the resistance of fasting dogs to 

 hemorrhage. Paul E. Howe and William J. Gies. In con- 

 tinuance of the series of investigations in this laboratory on the 

 effects of changes in the volume of circulating blood in normal, 

 undernourished, and overfed animals,^" we have lately determined, 



^^ See Biochemical Bulletin, 1912, ii, p. 186. for the last of the series. 



