1917] AGRICULTURAL CHEMISTRY AGROTECHNY. 203 



after it has passed above the graduation mark is obviated; and that it is ex- 

 ceedingly easy to manipulate. 



The determination of aluminum as oxid, W. Blum (Jour. Amer. Chem. Soc, 

 38 {1916), No. 7, pp. 1282-1297, figs. 2). — From experiments reported it was 

 found with a hydrogen electrode and suitable indicators that the precipitation 

 of aluminum hydroxid by ammonium hydroxid is complete when [H+] ranges 

 from 10"°^ to 10"'". These points are approximately defined by the color 

 change of methyl red and rosolic acid. The presence of ammonium chlorid is 

 advantageous in controlling the alkalinity and in coagulating the precipitate. 

 Solutions of ammonium nitrate and chlorid are equally satisfactory for washing 

 the precipitated hydroxid. For the determination of aluminum the following 

 procedure is recommended : 



To the solution containing at least 5 gm. of ammonium chlorid per 200 cc. 

 (or an equivalent amount of hydrochloric acid) a few drops of methyl red are 

 added and the solution then heated just to boiling. Ammonium hydroxid is 

 carefully added until the color of the solution changes to a distinct yellow. 

 Boiling is continued for one or two minutes and the precipitate filtered at 

 once and washed thoroughly with hot 2 per cent ammonium chlorid or nitrate 

 solution. The hydroxid is ignited in a platinum crucible and after all the car- 

 bon is burnt off is blasted for five minutes. The crucible is then covered and 

 placed in a desiccator until cool. After weighing, a second blasting of five 

 minutes is desirable, as it permits of a more rapid weighing and, consequently, 

 more accurate results. 



It is indicated that the crucibles containing the ignited alumina should be 

 kept covered in the desiccator and on the balance. Five or 10 minutes blasting 

 is sufficient for precipitates containing from 0.1 to 0.2 gm. aluminum oxid. 

 The presence of ammonium chlorid during ignition causes no appreciable loss 

 of alumina. 



Notes on the determination of aluminum, C. F. Sideneb and E. Pettijohn 

 {Jour. Indus, and Engin. Chem., 8 {1916), No. S, pp. 7H-716). — Experimental 

 data submitted show that boiling the solution for one minute completely pre- 

 cipitates all aluminum present, and that longer boiling may lead to a re-solution 

 of part of the precipitate. The complete removal of ammonium chlorid from 

 the precipitate before ignition was found not to be necessary. A great ex- 

 cess of ammonium hydroxid should be avoided during precipitation and 

 should never be more than 1 or 2 cc. per 250 cc. of solution. When the pre- 

 cipitate is large it must be blasted for 40 minutes to insure its being reduced 

 to a constant weight, as the ignited alumina is strongly hygroscopic. See also 

 a previous note by Daudt (E. S. R., 34, p. 205) and that by Blum, noted above. 



A study of the silver arsenate test for arsenic, L. J. Curtman and P. 

 Daschavsky {Jour. Amer. Chem. Soc., 38 {1916), No. 7, pp. 1280-1282) .—The 

 authors have found that with pure solutions of arsenate the test with silver 

 nitrate is sensitive to 0.02 mg. of arsenic. Ammonium nitrate was found to have 

 no influence on the detection of 0.2 mg. of arsenic as sodium arsenate. A 

 procedure in which the test is capable of detecting 0.5 mg. of arsenate with cer- 

 tainty is outlined. 



Titrametric determination of nitrites, B. S. Davisson {Jour. Amer. Chem. 

 Soc, 38 {1916), No. 9, pp. 1683-1687, fig. i).— The author, at the Ohio Experi- 

 ment Station, reviews the earlier methods for the titrametric determination 

 of nitrites and outlines a procedure based on the nitrous acid-potassium iodid 

 reaction. A special form of aeration flask used in the method is described. 



Analytical data submitted indicate the accuracy of the modified procedure. 

 Organic matter in soil extracts does not materially affect the results. When 



