112 EXPERIMENT STATION EECOED. 



termining aliphatic amins as methylamin and trimethylamin, and for ascertain- 

 ing the purity of lithium carbonate and basic magnesium carbonate. 



The precipitation of phosphorus as ammonium phosphomolybdate in the 

 presence of sulphuric acid, K. G. Falk and K. Sugiuka (Jour. Amcr. Chcm. 

 Soc, 37 {1915), Xo. 6, pp. 1501-1515). — "The precipitate of ammonium phos- 

 phomolybdate formed in the presence of sulphuric acid (as in Neumann's 

 method) contains sulphate apparently as an essential part of the molecule, to- 

 gether with an excess of molybdic oxid, and no nitric acid. The composition of 

 this phosphosulphomolybdate precipitate may vary with the concentration of 

 the different constituents in the solution. These variations explain the different 

 factors found for the titration of the precipitate with alkali. The composition of 

 the precipitate for a certain set of conditions was found to be 4 [(NH4)sP04.12 

 M0O3] + (NH4)2S04.5Mo03." 



A new test for copper, W. G. Lyle, L. J. Crrr.TMAN, and J. T. W. Marshall 

 {Jour. Amer. Chem. Soc, 37 {1915), No. 6, pp. 1471-lJf81). — An aqueous solution 

 of normal amino-caproic acid is said to be an exceedingly sensitive reagent 

 which is more specific than other reagents composed for detecting copper. As 

 little as 0.004 mg. may be found by this means. Mercury and zinc interfere with 

 the test, but the precipitates caused may be prevented, in the case of mercury, 

 by the addition of sodium chlorid, and with zinc by adjusting the aciditiy of the 

 solution. 



The effect of grinding the soil on its reaction as determined by the Veitch 

 method, P. E. Bkown and H. W. Johnson {Jour. Indus, and Engin. Chcm., 7 

 (1915), No. 9, pp. 776, 777). — In experiments with ungi-ound samly loam soils 

 containing varying amounts of sand and the same soils ground so as to pass 20, 

 40, and SO mesh sieves, it was found that when acid soils are ground before be- 

 ing tested by the Veitch method the acidity is reduced and the reaction fre- 

 quently becomes basic. Tlie development of basicity increased with the degree of 

 grinding of the soil and the increase dependixl upon the amount of sand present, 

 being greater in coarse sandy soils than in fine sandy soils. These results are 

 taken to indicate that soils should be in their natural condition and xmground 

 when tested by the Veitch method. 



The determination of nitrates in soil, II. S. Potter and R. S. Snyder {.Jour. 

 Indus, and Engin. Chcm., 7 {1915), No. 10, pp. 86S, Sfi-J).— Comparative te.sts of 

 calcium oxid and calcium carbonate as flocculating agents in obtaining the soil 

 extract to be used for the determination of nitrates led to the conclusion that 

 with soils low in nitrates the use of calcium carbonate is to be preferred, and 

 when the colorimetric method is used is always better. 



Methods of determining iron and alumina in mineral phosphates, V. P. 

 KocHETKov and D. N. Kasatkin {I:: RczuVt. Yegct. Opjjtov Lnh. Rahot, 9 {1913). 

 pp. 71-SO). — Comparisons of methods of qxiantitative analysis of Viatka and 

 Smolensk phosphates and of artificial mixtures of the principal salts found in 

 natural phosphates for sesquioxids of iron and alumina are reported. 



The methods compared were (1) Glaser's method, (2) precipitation of iron 

 and aluminum phosphates by sodium acetate after neutralization of the acid 

 phosphate solution by sodium carbonate, (3) precipitation of the iron and 

 aluminum phosphates by ammonium acetate, and (4) Grandeau's method of 

 precipitation of the hydrates of the sesquioxids by ammonia after treatment of 

 the acid phosphorite solution with acetic acid for the elimination of a large 

 part of the calcium phosphate and by molybdate of ammonia for the elimina- 

 tion of the remainder of the phosphoric acid. 



The method of Grandeau (4) was found to give the best results. Glaser's 

 method (1) gave results slightly inferior which are considered suitable for in- 



