372 EXPERIMENT STATION RECORD. 



slight excess of siil]>liuric, acid and aiiiiiionia salts expelled. The 

 residue is dissolved in very dilute liydiochloric aeid, an excess of plati- 

 num chlorid added, and the solution is evaporated to a moist paste on a 

 wa4:er bath below boiling. The double salt is washed by decantation, 

 once with water (L'O cc.) containing a. few dro])s of hydroc^hloric acid 

 and 3 times with alcohol, dried, and weighed. 



Jones criticises this uietbod as being '^ unnecessarily long and essen- 

 tially faulty in more than one particular," and proposes the folloM'ing 

 substitute lor it: 



''Take 1 to 2 giii. of the sample, put into a ])latiuuin dish, a(hl half its weight of 

 calcium hydroxid (Ca (H0).2), mix well with a short stirring rod, then gently ignite. 

 This destroys organic matter and dissipates ammonia. Put the mass, when cool, into 

 a small glass mortar and triturate with a little water, rinse into a beaker, digest, 

 filter, and, of course, wash the easily washed residue. The filtrate is colorless and 

 only contains in solution what calcium sulphate and calcium hydroxid the amount 

 of water will dissolve, besides the alkaline salts. Add to this solution 2 or 3 drops 

 of hydrochloric acid till /ms/ acid, then, droj) by drop, barium chlorid solution to 

 very slight excess ; digest on hot plate, then filter. Add to this iiltrate, drop l)y drop, 

 sodium carbonate solution till no further precipitation occurs; digest and filter; 

 acidify the filtrate with hydrochloric acid, evaporate over water bath with platinum 

 chlorid solution, and proceed as usual." 



The determination of albumen in cows' milk, L. L. Van Slyke 



{Jour. Amcr. Chem. ;8'oe., JO (1^91), Xo. K)., pp. 71:^-715). — In the author's 

 method for casein^ about 10 gm. of milk is diluted with about 90 cc. of 

 water at 40 to 42° C, and then 1.5 cc. of 10 per cent acetic acid (by 

 weight) added. The casein precipitate is washed upon the filter, the 

 nitrogen determined by the Kjeldahl method, and the casein calculated 

 by the factor 0.25. 



In the method now described for albumen the filtrate from the casein 

 precipitate is heated in a boiling water bath for 10 or 1.") minutes, when 

 the albumen coagulates and settles out. Trials of heating from 5 

 minutes to 10 hours showed that the longer heating did not practically 

 change the results. The filtered and washed precipitate is treated by 

 the Kjeldahl method and the nitrogen nmltiplied by 0.2o to get the 

 albumen. 



The remaining nitrogenous compounds are determined by difference. 



Trials of BruUe's method for the recognition of margarin in 

 butter, C. A. Loury ue Bruyn {Chem. Ztg., IS {1804), No. 70, pp. ISil, 

 1S42). — This method' depends upon the degree of hardness of natural 

 and artificial butter which has been oxidized with nitric acid. According 

 to Brnlle, butter treated in this way remains very soft, while margarin is 

 usually very hard. The hardness is measured by the *'oleogrameter," 

 which shows the weight required to cause a rod to make an impression 

 upon the surface of the oxidized material. Brulle gives this as 0.25 kg. 

 for butter and about 5 kg. for margarin, but he does not give the size of 



'Jour Amer. Chem. Soc, 15 (1894), ]>. 635. 



-^Compt. Reud., 116 (1893), p, 1255; ami E. S. R. 5, p. 450. 



