1920] AGRICULTURAL CHEMISTRY AGROTECHNY. 415 



chemists and two bactoriolojiists were made on each of over GO samples of 

 varied snides prei)ared under suiR'rvision from known materials, the results 

 being set forth in a series of tables. A weighed sum of the different determina- 

 tions Is proposed as a numerical criterion to determine whether or not a given 

 sample is edible. 



The (leterniiiiation of added water in meat products, 11. Lkdknt (Ann. 

 Falsif., 12 (JO JO), No. 133-134, PP- 356-367).— This is a criticism, based upon 

 laboratory analyses, of the methods in use in France for determining moisture 

 in meat pi-oducts and of the ollicial standards for detecting the presence of 

 added water. The requirement that meat products shall not contain more 

 than 75 per cent of moisture is considered too low, since meat which might 

 enter into the composition of such products has been found to contain as much 

 as 84 per cent of moisture. 



The composition, action, and analysis of chemical yeasts, M. A. Kling, 

 A. Lassikuu, and L. Veknekd (Ann. Falsif., 13 (1920), No. 135-136, pp. 0-17, 

 fig. i).— This is essentially a resume of recent German literature on tlie com- 

 position of baking powder and methods for its analysis. 



Chemical yeasts (baking powder), L. Weil (Ann. Falsif., 13 (1920), 

 No. 13.5-136, pp. 11-21). — This is a general discussion of the ingredients of 

 baking powder with suggestions as to limiting standards. 



The analysis of milk, C. Pohcher (Ann. Falsif., 13 (1920), No. 135-136, 

 pp. 35-37). — The author calls attention to two possible sources of error in the 

 analysis of milk. Data are given indicating that the addition of 1 cc. of 

 formol to 50 cc. of milk does not alter the reducing power of the milk, while 

 with increasing amounts of formol larger reducing values are obtained for the 

 same amount of milk. Data are also given showing appreciable differences in 

 the weight of the same volume of milk when measured with pipettes having 

 slow and rapid delivei-ies. This difference is accentuated with milk having a 

 high percentage of fat. 



A coniparative study of some methods for determining the fat content 

 of skim milk, T. J. McInekney and H. C. Troy (Neiv York Cornell Sta. Bui. 

 JfOl (1920), pp. 60-S5). — A study is reported of the accuracy of the Babcock 

 test for determining the percentage of fat in skim milk as compared with the 

 Adams gravimetric, the Roese-Gottlieb, the Mojonnier, and the modified Leff- 

 man-Beam methods. 



A preliminary study of the effect of various modifications of the Babcock 

 test upon the yield of fat from skim milk indicated that by increasing the 

 amount of acid, using excess speed, and centrifuging at a higher temperature 

 for a long time the percentage of fat obtained is considerably higher than that 

 obtained with the usual Babcock technique. 



The results obtained with the Babcock method thus modified compared favor- 

 ably with those obtained with the other methods tested. Tlie authors conse- 

 quently recommend the following modifications in the Babcock test when it is 

 to be used in determining the fat content of skim milk : 



"At least 25 cc. of sulphuric acid should be used. If the size of the bottle 

 permits, as much as 28 cc. may be used to advantage. The temperature of the 

 testing machine should be at least 180° F. A centrifuge having a disk 15 in. 

 In diameter, and strong enough to be run at a minimum speed of 1,800 revo- 

 lutions a minute without danger of breaking, is recommended. The diametei 

 of the disk is determined by measuring the distance between bases of the oppo- 

 site cups when they are in a horizontal position. The milk should be centri- 

 fuged for 10, 2, and 1 minute periods, instead of for 5, 2, and 1 minute periods." 



