RECENT WORK IN AGRICULTURAL SCIENCE 



AGRICULTURAL CHEMISTRY. 



On the determination of phosphoric acid, U. Milone (Atti. 6. Cong. Inter- 

 nas. Chim. AppL, 1 (1006), pp. 63, 6^).— A comparison of the methods of Bon- 

 gartz, Leconte and Piucus (modified Neubauer), and Sonneuschein is reported, 

 the results indicating that the method of Bongartz « is quick and reliable. This 

 method is based on that of Thomson,* in which phosphoric acid is determined 

 by titration first with normal sulphuric acid using methyl orange as an indi- 

 cator, and then after boiling and cooling with the same acid using phenol- 

 phthalein as an indicator. 



Standardization of uranium solution for determining phosphoric acid, F. 

 Repiton iilonit. ScL, -',. scr., 21 (1907), II, Xo. 792, pp. 815, 816; abs. in Chem. 

 Zentbl, 1908, I, No.'S, p. 295; Jour. Chem. Soc. [London], 9// (1908), No. 5)7, 

 //, p. .'i28). — In applying the uranium method to the determination of phos- 

 phoric acid in calcium phosphate it is advisable to standardize the uranium 

 solution with a solution of calcium phosphate containing an accurately known 

 amount of phosphoric anhydrid. For the accurate determination of the phos- 

 phoric anhydrid the author recommends the iron method in which phosphoric 

 acid is precipitated by a solution of ferric nitrate of known strength and the 

 excess of iron precipitated by boiling with ammonium acetate. The combined 

 weight of the two precipitates less the amount or iron oxid used equals the 

 weight of phosphoric anhydrid. 



The determination of potash by the platinum chlorid method, H. J. F. de 

 Vries (Chem. WcekbL, .', (1907), pp. 231-242; 333-3^3; Ji55-462; 5 (1908), pp. 

 176-184; 261-269; abs. in Chem. Zentbl, 1908, I, No. 17, p. 1647; Jour. Chem. 

 Soe. [London], 94 (1908), No. 508, II, p. 534)- — In this article the author re- 

 views the various investigations which bear upon the causes of error in this 

 method of determining potash and gives his own views as to the best means of 

 avoiding them. 



He concludes that the best results with both pure potassium chlorid and mix- 

 tures of potassium chlorid with either sodium chlorid or magnesium chlorid, or 

 both, are obtained with the use of 96 per cent alcohol. With both SO per cent 

 and 96 per cent alcohol much better results were obtained by employing the 

 empirical factor 0.3056 than by using the theoretical factor 0.3071 based upon 

 the latest international atomic weights. The wide variation in results, some- 

 times amounting to as much as 4 per cent, can not, however, be explained by 

 the kind of alcohol used or the method of calculation, but is due primarily to 

 the presence of sulphuric acid. In the presence of sulphuric acid 96 per cent 

 alcohol gives uniformly too high results and 80 per cent alcohol to low. 



On the volumetric determination of potassium as the cobalti-nitrite, W. A. 

 Drushel (Chem. News, 97 (1908), No. 2520, pp. 124-126; abs. in Science, n. 

 sen, 28 (1908), No. 711, p. 219).— See a previous note (E. S. R., 19, p. 808). 



a Arch. Pharm., 3. ser., 22 (1884), p. 846. 

 *Chem. News, 47 (1883), p. 186. 



61717— No. 4—08 2 



307 



