CHEMISTRY. 615 



pyrophosphate with sulphuric acid. By this means a new body is obtained, ma»- 

 nesium pyrosulphophosphate, O.Tlogm. of majinesium pyrophosphate yielding 0.2305 

 gui. of the body. The content of MgO is calculated by the factor 0.265, of P^Oj by 

 the factor 0.47. 



The determination of lime, magnesia, and phosphoric acid in the presence 

 of considerable amounts of iron oxid, II. Pellet {Ann. Chhn. Analyt., 6 {1901), 

 ]h li>S;nhs. in ('hem. Ztjj., :25 {1901), Xo. 5S, Repert., p. 208). — The methods proposed 

 for the analysis of substances containing from 5 to 10 percent of iron associated with 

 small percentages of lime, magnesia, and phosphoric acid are as follows: For the 

 determination of lime, neutralize a portion of the nitric or hydrochloric acid solu- 

 tion freed from silica, corresponding to 2 to 5 gm. of the substance, with ammonia 

 to the verge of precipitation, and add a few drops of acetic acid, about 20 times the 

 amount of ammonium oxalate required to precipitate the lime, and a little more 

 ammonia. The operation is completed in the usual way. For the determination of 

 magnesia use an aliquot of the solution corresponding to the same amount of sub- 

 stance, and for each 50 cc. of the solution add 70 cc. of Petermann's ammonium citrate 

 solution, 10 cc. of saturated sodium phosphate solution, and 20 cc. of ammonia, and 

 proceed as usual. For the determination of phosphoric acid use ammonium molyb- 

 date to which 20 drops of ammonium citrate solution has been added to prevent the 

 separation of molybdic acid. Dissolve the ammonium phosphomolybdate in ammonia 

 and reprecipitate with dilute nitric acid in a warm solution. Dry at 100° C. and 

 weigh. One hundred parts of the precipitate correspond to 3.75 parts of phosphoric 

 acid. 



Estimation of magnesium by organic bases, W. Herz and K. Drucker 

 {ZAschr. Anorgan. Chem., 26 {1901), j^Ji- S47-S49; aim. in Jour. Chem. Soc. [Londoni, 

 80 {1901), No. 46S, II, p. 348). — The method of estimating zinc by precipitation with 

 dimethylamin is applicable to magnesium, giving accurate results. Magnesium salts 

 are also completely precipitated by a solution of free guanidin. This method is 

 especially recommended in the estimation of magnesium in the presence of alkalis 

 and in the analysis of mixed silicates containing magnesium. 



The determination of potash, O. ScnuMM {Ztschr. Analyt. Chem., 40 {1901), pp. 

 385-389; <tb.^. in Chem. Centlil., 1901, II, Xo. 4, p. 321).— The author recommends 

 Classen's electrolytic method, in which metallic platinum is separated from the 

 potassium platinum chlorid in sulphuric-acid solution and weighed. The factor used 

 in calculating potassium chlorid from the weight of platinum is 0.76689 for less than 

 0.2 gm. platinum and 0.764595 for 0.2 gm. or more. 



The relative value and limits of accuracy of various titration methods for 

 determining caustic soda in the presence of sodium carbonate, R. Lucion and 

 D. dePaepe {Bui. A.^^toc. Beige Chim., 15 {1901), No. 1, pp. 19-23; abs. in Bui. Soc. 

 Chim. Pari.% 3. s^er., 25 {1901), No. 12, p. 652). 



On various classic methods for the volumetric determination of small 

 quantities of sodium carbonate in the presence of large amounts of acid 

 sodium carbonate, P. T^rciox and D. de Paei'e (/>((/. A>fmc. Beige Chim., 15 {1901), 

 No. 1, pp. 23-25; ah.'<. in Bui. Soe. Chim. J'ari.'i, :!. i^er., 25 {1901), No. 13, p. 703). 



A method for the determination of the availability of organic nitrogen in 

 commercial fertilizers, J. P. Street {Jour. Amer. Chem. Soc, 23 {1901), No. 5, pp. 

 330-338). — The following method is proposed: To an amount of substance equiva- 

 lent to 0.075 gm. of nitrogen in a 300 cc. Erlenmeyer flask add 100 cc. of neutral 1.6 

 per cent potassium permanganate and digest the mixture on a steam l)ath for 20 

 minutes with occasional shaking, filter and wash with 125 to 150 cc. of water, and 

 determine nitrogen in the undissolved residue by the ordinary Kjeldahl process. 

 The results of a large number of tests of widely differing fertilizci-s sliow that this 

 method places them in al)out the same rank as that assigned by vegetation tests and 

 agricultural experience. 



