CHEMISTRY. 617 



of the fat, dissolved in 10 volumes of carbon I )isulphid, twice as much concentrated 

 sulphuric acid as the fat is added drop by drop, with shaking;, to avoid rise of temper- 

 ature. The whole is rinsed into a sei)arati)ry funnel, the lower layer run (Uit, the upper 

 one sliaken with a little animal charcoal to remove tar and traces of acid, then 

 filtered, distilled, and the residue weij^hed. 



A new reaction of saccharin, A. Leys {Compt. Rend. AnuL Sci. PariK, 132 {1901), 

 No. 17, pp. lor>G-10r>S; (tb.'<. In Jour. Soc. Chem. Ind., 20 {1901), No. 6, p. 622).— This 

 colcr reaction with saccharin is given when even very dilute solutions are treated with 

 dilute copper sulphate or ferric chlorid and hydrogen peroxid. Five cc. of the solu- 

 tion containing saccharin, treated with 2 drops of ferric chlorid and 2 cc. of hydrogen 

 peroxid diluted, gixes a violet color in 30 or 35 minutes. 



This reaction may be used for determining the presence of saccharin in milk and 

 in butter. The milk is treated with potassium bisulphate and absolute alcohol and 

 the clear liquid filtered from the precipitated casein and fat. This licjuid is then 

 shaken with ether, evaporated, dried, taken up with Ijoiling water, cooled, and 

 tasted. If sweet, examine for saccharin as above. 



Butter is dissolved in equal volumes of chloroform and alcohol, water added, and 

 shaken. The separated chloroform retains the fats; tlie aqueous-alcoholic solution 

 the saccharin. The latter is tested as descrilied. 



Clarification of lovr-grade molasses for polarization, H. C. Laurence {Beet 

 Sugar Gaz., 3 {1901), No. 5, pp. 109, 110). — In the laboratory of the Los Alamitos 

 Sugar Company experiments were made of using permanganate of potash in clarify- 

 ing sugar solutions. After clarifying with lead subacetate tlie filtrates when dark 

 w'ere treated with a few crystals of permanganate of j^otash and again filtered, result- 

 ing in a brigiit and much clearer filtrate. In order to test the correctness of this 

 addition 50 cc. portions of the same samples were placed in 200 cc. flasks and treated 

 with a slight excess of lead solution. After 10 minutes 50 cc. of a strong perman- 

 ganate solution was added, the whole allowed to stand long enough for the air bub- 

 bles to escape, brought to the mark, filtered, and polarized. The polarizations were 

 made with ease, and agreed with those'obtained with the lead solution alone. The 

 filtrates from the lead alone were dark and hard to read; those from the ijerman- 

 ganate were colorless as water. 



These results were verified by A. M. Masser, chemist of the Wolverine C'ompany, 

 who also found the method ajjplicable to carmelized sugar. 



The action of carbonic acid in sugar solutions saturated with lime, J. 

 Weisbekg {Ihd. .Uxor. Clmn. Sun: et DhtilL, IS {1901), No. 7, pp. 457-4'J2) .—The 

 products formed in adding carbonic aciid to precipitate the lime in the carbonation 

 process in sugar making are described and discussed. In passing carbonic acid into a 

 sugar juice containing lime the alkalinity as denoted by phenolphthalein continues 

 to fall, while the polarization of the filtered juice falls and then rises again to the 

 original value. At the lowest j^oint of polarization there is the greatest thickening 

 of the liquid, the slowest rate of filtering, and this is the j)oint at which the decol- 

 oration of the juice begins. 



Estimation of sugar in vinous products, A. Bernard {Ann. Clilm. Analyt., 6 

 {1901),pj>.S!i-9:'>; <d)x. in .Jour. CJinii. Snr. [London], SO {1901), No. 403, II, p. 355). — 

 Ten cc. of the liquid to be examined is diluted in a conical flask, 20 cc. Fehling's solu- 

 tion added, and heated to boiling. The flask is cooled in cold water, the contents 

 filtered, and 15 cc. or more of the filtrate is titrated with a solution of potassium cya- 

 nid until colorless. The cyanid solution is made up so that 10 cc. will discolorize 

 10 cc. of Felding's .solution. The strength of tlie latter may be determined by means 

 of an accurately made solutif)n of invert sugar. 



Estimation of total extract and free acid in wine, F. Freyer {Oesterr. CItem. 

 ZUj., 4 {1901), p. 129; ahf. in. Aindi/.'it, 20 {1901), No. 303, p. i,57).— Owing to tiie many 

 sources of probable error in estimating the extract of wine by evaporation, the author 



