912 EXPERIMENT STATION RECOED. 



prevent errors due to ammonia in the air are described in detail. It is 

 claimed tliat with the small quantities of material and ai)i)aratus required 

 complete combustion is much more easily carried out. 



When the Kjeldahl process is used 0.2 to O.G gm. of the soil is digested with 

 about 5 cc. of sulphuric acid until transformation of the nitrogen into am- 

 monia is complete. The ammonia is distilled and Nesslerized in the usual way. 



Tests of both methods are reported showing that a high degree of accuracy is 

 possible. 



On a method of determining small amounts of nitrogen and the applica- 

 tion of the method in the detection of organic impurities in water, S. 

 KoKSCHUN (Arch. Hijg., 62 (.1907), No. 1, pp. U2-106; abs. in Ghcm. ZcntbL, 

 1907, II, No. 18, pp. 1548, 1549; Chem. Ztg., 31 (1907), No. 101, Repert. No. 

 93, p. 631). — In the method proposed, from 200 cc. to 1 liter of the water is 

 evaporated to 20 cc. after the addition of from 2 to 3 cm. of dilute sulphuric 

 acid. From 5 to 8 cc. of concentrated sulphuric acid and about 1 gm. of potas- 

 sium sulphate are then added and the solution boiled until colorless. The 

 acid is neutralized and the ammonia determined by Nesslerization. The nitro- 

 gen in suspended matter in the water is determined by precipitation witli fresh 

 basic ferric acetate, the precipitate being collected and treated as described 

 above. 



Notes on volumetric determination of magnesium in water, G. B. Frank- 

 FORTER and Lillian Cohen {Jour. Amer. Chcm. Soc, 29 {1907), No. 10, pp. 

 1464-1467; abs. in Chcm. ZentbJ., 1907, II, No. 26, p. 2080).— The authors call 

 attention to an error in the calculations in the original description of Meade's 

 modification of Stalba's arsenic method for determining magnesium in water. 

 They also report tests of the corrected method in which better results were 

 obtained without the use of starch as an Indicator. 



Contribution to the determination of oxygen in water, W. Cronheim 

 {Ztschr. Aiigew. Chem., 20 {1907), No. 45, pp. 1939-1942; abs. in Chcm. 

 Zcnibl., 1907, II, No. 26, p. 207. 'i). — This article deals with sources of error in 

 the Winkler method for the determination of oxygen in water and with correc- 

 tions which must be applied in order to obtain reliable results. 



The polarimetric determination of sucrose. Ill, The Clerget method, 

 F. Watts and H. A. Tempany {West Indian Bui., 8 {1907), No. 1, pp. 111- 

 119). — A critical study of the Clerget method is reported and some modifica- 

 tions are suggested. According to the authors, the results obtained " emphasize 

 the very great accuracy of the process as an analytical method, and the indis- 

 criminate use of factors regardless of the conditions for which they were de- 

 termined can not be too strongly deprecated, as detracting from the intrinsic 

 accuracy of what is undoubtedly one of the most accurate processes employed 

 in organic analysis." 



The polarimetric determination of starch, E. Ewers {Ztschr. Offentl. Chem., 

 14 {1908), No. 1, pp. 8-19). — The article describes a polarimetric method of 

 determining the starch content of such substances as grain, meal, bran, feeding 

 stuffs, and potatoes. 



To the sample in a flask acetic acid, hydrochloric acid, and hot water are 

 added successively, with digestion in a hot water bath for a stated period at 

 each addition. Potassium ferrocyanid is added to the cooled solution, which 

 is then diluted to a given volume, filtered, and tested in a polarimeter. A blank 

 test, similar to the above, with the omission of the hydrochloric acid, is also 

 made, the data being used as a correction to those from the preceding test. 



The reported results by this method agreed very closely with those by the 

 gravimetric method. The author believes that it is specially applicable for 



