DAIRY PRODUCTS. 119 



Acidity. The samples examined showed under the microscope the 

 acetic ferment, and a portion of the acidity was therefore due to acetic 

 acid. It is the custom in giving the results of analyses of koumiss to 

 represent the whole of the acidity as due to lactic acid. If ordinary 

 yeast is used, and it generally is, it is possible that acetic acid may be 

 formed. This appeared to be the case with the samples in question, since 

 in distilling them a larger percentage of acid was found in the distillate 

 than could have been expected had lactic acid only been present. 



I made no attempt to separate these two acids, but estimated the total 

 acidity, and then represented it in terms of both acids. 



The direct titration of the lactic acid in the koumiss was attended with 

 such difficulty that the attempt was abandoned. Whatever indicator 

 was employed, the change in color was so obscured that no sharp reac- 

 tion could be obtained. 



To obviate this trouble the koumiss was mixed with an equal volume 

 of saturated solution of magnesium sulphate. After shaking the mixture 

 it was poured through a linen filter. The first portions running through 

 were turbid. After refiltering these the filtrate was quite clear. 



Better results were obtained by using with the koumiss equal volumes 

 of alcohol. The filtrate from this mixture was uniformly bright, In 

 this filtrate the acid was estimated by titration with standard sodic- 

 hydrate solution, making the proper corrections for dilution and using 

 phenol-phthalein as an indicator. I would recommend this alcoholic 

 method of clarification to all who may have occasion to determine acid 

 in milk. 



Alcohol. The alcohol was estimated by distilling 500cc. koumiss with 

 lOOcc. water until the distillate amounted to 500cc. This, being still tur- 

 bid, was redistilled with a small quantity of water. The final distillate 

 of 500cc. was used for the estimation of the alcohol in the usual way, viz, 

 by taking its specific gravity and calculating the alcohol from tables. 



Milk-sugar. The milk-sugar was estimated by the method I recoui- 



icnded in a paper read at the Philadelphia meeting of the A. A. A. S. 1 



Fat. Twenty grams of the koumiss were evaporated to dryness in a 



jhalchen, the whole rubbed to a fine powder, and extracted with ether 

 in a continuous extractor. The process of extraction lasted six hours. 



Albuminoids. The albuminoids were estimated by evaporating 5 

 grams of the material in a scbalchen, rubbing to a fine powder with 

 soda-lime, and burning with the same in the usual way. 



1 Am. Chem. Jour., vol. 6, p. 289 etseq. 



