CHEMISTRY. 523 



Boiled linseed oil; analytical constants, M. Kivv {('hein. Rev. Fed x. Ilarz-Ind., 

 8 {lUOl), Xn. 3, })}). .fO-4J; (ihs. in Jour. Sue. Vhrm. Ind., M {1901), No. 5, p. 484).~ 

 A table IS given in which the alterations in Unseed oil on boiling are represented at 

 6 different stages, from a thin oil to the consistency of India rubber. The changes 

 are represented by the variations in the acid value, saponification value, iodin 

 number, iodin number of fatty acids, acetyl acid value, and acetyl saponification 

 value. 



A comparison between the bromin and iodin absorption figures of various 

 oils, H. T. VuLTK and Lily Logan {Jour. Amev. Clion. Soc, £S {1901), No. ,?, pp. 

 156-159). — A revision of the bromin and iodin figures of various animal and vege- 

 table oils is offered. Of the 15 oils examined, only 6 were found to be at all capable 

 of forming substitution products. 



Kjeldahl method for the determination of sugar, E. Woy {Ztschr. Oefentl. 

 Chem., 6 {1900) , pp. 514-519; ahs. in Jour. Spc. Chem. Ind., 20 {1901), No. 4, p. 395). — 

 The author indorses the Kjeldahl method with the J essen-Hansen extension (E. S. 

 R., 11, p. 614) in the estimation of sugar. The different reducing sugars are reduced 

 under similar conditions, thus permitting comparisons by means of the copper values 

 determined. Again, the method employed is so simple as to admit of closely agree- 

 ing results by different analysts. A source of great error, however, lies in the 

 absorptive power of the alkali solution for carbon dioxid. The author suggests the 

 working out of similar tables to those of Kjeldahl, but using sodium carbonate 

 instead of caustic soda. 



Influence of salts on the rotary power of sugars, J. de Kowalski and P. 

 ToMARTSCHENKO {Arcli. Sci. Pkys. et Nat. Geneve, 11, No. 4, pp. 294-299; ahs. in 

 Chem. Centhl. 1901, I, No. 18, p. 984; Jour. Soc. Chem. Ind., 20 {1901) , No. 6, p. 623) .— 

 The authors found that the chlorids, bromids, and iodids of sodium, potassium, 

 and ammonium at different concentrations lowered the rotation of cane sugar. 



The coefficient 0.85 in the indirect analysis of sugar cane, L. Bonxin {Bui. 

 Assoc. Chim. Sucr. ef Distill., IS {1901), No. 7, pp. 465-467). — In a test of 14 varieties 

 of cane the author found that the coefficient for calculation to be exact should range 

 from 0.789 to 0.882 and to average 0.843. 



Scheibler's extraction method for the determination of the polarization of 

 beets, A. Herzfeld {Ztschr. Ver. Deut. Zuckerind. 1901, pp. 334, 335; abs. in Jour. 

 Chem. Soc. [london^, 80 {1901), No. 464, II, p. 426).— N description of the method 

 as carried out in the laboratory of the Verein der deutschen Zucker-Industrie. 



Notes on sugar beets, P. F. Trowbridge {Jour. Amer. Chem. Soc, 23 {1901), 

 No. 4, J>p- 216-223) . — The experiments reported were made in Michigan and repre- 

 sent the conditions at the various beet-sugar factories in that State. Tests were made 

 of 2 methods of finding the tare on beets, in practice at Michigan factories, the 

 sample in one case being well brushed with bristle brushes, and in the other washed in 

 a revohong washer. The tare by the method of brushing was 10. 14 per cent, while 

 the tare by washing and draining for a few minutes was 10 per cent. Owing to the 

 fact that there has l)een some controversy between the farmers and the factory man- 

 agers, tests were made to determine the proper factor for estimating the sugar in the 

 beet from the amount of sugar in the juice. From the analyses made during 2 

 seasons, the author found that the average factor should be 0.918, the range being 

 from 0.875 to 0.956. He concludes that the average factor lies between 0.91 and 0.92. 

 Attention is called to the necessity of taking similar portions of several beets for the 

 sample. 



A comparison was made of the German method of analysis by digesting with hot 

 alcohol, and the French method of digesting with hot w^ater. The percentage of 

 sugar obtained by alcohol digestion averaged 13.41; by water digestion, 13.43; and 

 in no case did the difference exceed the limit of error in ordinary duplicate analysis. 



