AGPJCULTUKAL CHEMISTEY AGROTECHNY. 611 



Estiination of cacao shells in cocoa powders, Louise Kai,tjsky {Zischr. 

 Untersuch. Nahr. u. Genmsmtl., 23 {1912), No. 12, pp. 654-661).— Thin method 

 is based on the difference in specific gravity noted between the outer and inner 

 coats of the bean. It was found that the 2 tissues when present in a mixture 

 could be separated by means of a mixture consisting of chloral hydrate, glycerol, 

 and water. The specific gravity of the inner portion, after the removal of the 

 fat and treatment with diastase and later with hydrochloric acid, was found to 

 vary between 1.1131 and 1.3503. In most instances, however, it was less than 

 1.25. The specific gravity of the husk on the other hand lies between 1.234 

 and 1.9337. The inner portion of the bean of commercial cocoa powders had 

 a higher specific gravity than the products prepared by the author. 



Bacteriological and chemical methods for determining the quality of milk, 

 E. B. Fbed and G. W. Chappeleae, Jr. ( Virginia 8ta. Bpts. 1911-12, pp. 206-239, 

 flgg^ ^), — As the plate bacterial count method fails to show the type of organ- 

 ism present in milk, requires too long a time to execute, fails to give a true 

 index of the actual number of organisms present, and finally necessitates an 

 expensive equipment and skill, it can not be adapted to general usage. In these 

 experiments the value of the reduction test in comparison with the plate, acid, 

 fermentation, sediment, and catalase tests was studied. For the work with 

 whole milk, which was obtained from the city of Roanoke, Va., and the 

 creamei-y of the Virginia Polytechnic Institute, beef peptone agar was used as 

 the nutrient medium for the plate and similar work, as it was found to give the 

 best results. For the experiments with pure cultures in some instances beef 

 peptone agar was employed, but for the Streptococc-ws lacticus and Bacterium 

 lactis addi milk whey agar was used. 



In the stain-reducing experiments much was found to d^iend upon the kind 

 of stain used. The pure stain usually employed for medical purposes was found 

 the most applicable. In all probability the controlling factor in stain reduction 

 by micro-organisms is the toxic properties of the dye rather than the difficulty 

 encountered in reduction by chemical means. lodin green, methyl violet, 

 gentian violet, and safranin were not reduced, and in the final experiments 

 methylene blue, litmus, indigo carmin, and neutral red were used. The stains 

 were also titrated against a solution of titanium trichlorid, and neutral red was 

 found not to be reduced by this reagent 



For the reduction test a methylene blue solution consisting of 1 gm. of pure 

 methylene blue in 1,000 cc. of physiological salt solution, diluted with an equal 

 volume of water, was found to be best suited. Next came litmus, while neutral 

 red and indigo carmin were found to be quite toxic. The temperature influences 

 the time of reduction. " The various organisms have different powers of reduc- 

 tion which vaiy widely among themselves, and, as a rule, the lactic acid organ- 

 isms reduce first; then follows B. vulgatus, B. aerogenes, B. denitriflcans, B. 

 coli communis, and B. vulgaris, all of which are strong reducers." " In pure 

 cultures time of reduction and growth of cells show a strong relationship and 

 this relationship takes the form of an inverse proportion. As the number of 

 cells approaches infinity the time of the reduction approaches zero, and vice 

 versa." 



Fi-om the experiments with whole milk it is noted that the percentage of 

 add in a general way has some relation to the number of micro-organisms 

 present in the milk, but the reliability is an uncertain quantity and consequently 

 the acid test is of little practical value. " [The fermentation test] is an easy 

 test for finding out to a certain degree the type of flora present in milk. A com- 

 parison of the sediment and fermentation tests shows that a high dirt content 

 is usually associated with a gaseous curd. Its r^ation to the other test is 

 slight. Of the several methods studied, it proves the best criterion for determin- 



