CHEMISTRY. 199 



The estimation of starch in meat products, J. Mayrhofer 



{Forsch. a. Lehcnsmtl. und Hyg. Ghem., 3 {1896), p. 141). — The method 

 is based on the power of alcoholic potash solution to dissolve the pro- 

 teids aud fat while the starch remaius behind. The substance is 

 treated on the water bath in a covered beaker with 8 per cent alcoholic 

 potash. The li(|uid is diluted with hot alcohol to prevent the gelatin- 

 ization of the soap, the insoluble residue collected on a filter, and 

 washed with alcohol. The starch is dissolved by means of aqueous 

 potash solution and reprecipitated with alcohol. It is then collected 

 on a filter, washed with alcohol and finally with ether, and the dry 

 starch weighed. It contains considerable ash which, according to the 

 author, can be avoided by acidifying the solution slightly with acetic 

 acid before precipitating. — w. h. krug. 



Composition of human fat, C. A. Mitchell {Analyst, 21 {1896), 

 July, p. 171). — Clievreul found that human fat consisted of olein and 

 stearin, while Heintz concluded that it was composed of olein and \)al- 

 mitin with small quantities of several other acids, one of them being 

 liquid. The author concludes from his work that human fat consists of 

 about 70 per cent of liquid acids, principally oleic, 30 per cent of solid 

 acids, probably palmitic, with small amounts of stearic and myristic 

 and traces of lower volatile acids. — b. w. kilgore. 



Determination of volatile fatty acids by the method of Leff- 

 mann and Beam, W. Karsch (Ghem. Zt<j., 20 {1896), No. 62, pp. 607, 

 608). — The author believes that this method — saponification with 

 glycerin-soda solution — is free from the objections of the other 

 methods, and commends it after 2 years of use. He gives comparisons 

 of the method with Wollny's method on 10 samples of butter, each 

 sample being saponified at least twice by both methods. The differ- 

 ence in volatile fatty acids found by the 2 methods ranged from 0.11 

 to 1.13 cc. and averaged 0.5 cc. decinormal soda solution. The more 

 closely Wollny's directions were followed the more closely did the 

 results approach those by Leffmann and Beam's method. The Wolluy 

 method invariably gave the highest results, which leads the author to 

 believe that the results by this method were affected by the absorption 

 of carbon dioxid. Aside from the greater ease and convenience of 

 Leffmann and Beam's method, he believes it is to be i^referable from the 

 point of accuracy. 



The highest Eeichert-Meissl number which the author has found by 

 the Leffmann and Beam method in the butter of the local creamery 

 since the previous October is 30.82 cc. and the lowest 26.53 cc. 



On the probable error of the rapid milk testing methods of 

 Babcock, Gerber, and Thbrner, compared v^ith the gravimetric 

 (sand) method, H. Schrott-Fiechtl {Milch Zty., 25 {1896), No.s. 12, 

 pp. 183-185; 13, pp. 199-201; 14, pp. 217-220).— The average results of 

 200 determinations by these methods are reported. 



The average differences between the results by the gravimetric and 



