12 EXPERIMENT STATION RECOED. [Vol. 35 



A proposed new method for citrate-insoluble phosphoric acid, 0. H. Hunt 

 (Jour. Indus, and Engin. Chem., 8 (1916), No. 3, pp. 25 J-253 ) .—Confirming the 

 findings of previous investigators, the author has shown that lime, limestone, 

 and other calcium compounds increase the citrate-insoluble phosphoric acid. 

 The ratio between the lime, determined as such, and the citrate-insoluble phos- 

 phoric acid has been shown to be fairly constant. 



" In the proposed method the fertilizer is digested as for total phosphoric acid. 

 An aliquot is made alkaline with ammonia and the phosphoric acid is determined 

 in the precipitate thus formed. The ratio between the citrate-insoluble phos- 

 phoric acid and the phosphoric acid precipitated with ammonia is about 1 : 1.5. 

 By dividing the amount of phosphoric acid precipitated with ammonia by 1.5 

 we obtain a figure which is approximately equal to the citrate-insoluble phos- 

 phoric acid determined by the official method. The plus and minus errors in 

 the final results nearly balance each other. 



"The fact that the errors for bone meal are nearly the same, and that for 

 tankage and complete fertilizers are also nearly the same but of opposite quan- 

 tity, seems to suggest a possibility of using a different factor for different types 

 of fertilizers. The difference between the results obtained by the official and 

 proposed methods is no greater than the differences in the results obtained by 

 different analysts working on the same sample when the official method is used. 

 The proposed method claims to be much shorter and less expensive, doing away 

 W'ith the use and preparation of neutral ammonium citrate." 



The determination of citric-acid-soluble phosphoric acid in Thomas slag 

 by the iron-citrate method, Celichowski and F. Pilz (Ztschr. Landw. 

 Vcrsuchsw. Osterr., 18 (1915), No. 10, pp. 581-591). — It has been shown that in 

 the preparation of the iron-citrate solution a clear, fresh solution of iron chlorid 

 can be used. Old solutions which are partly decomposed and contain colloidal 

 iron oxid are to be avoided. The iron-citrate solution used in the determination 

 must not be too old, as it may influence the results due to partial dissociation. 

 The hydrogen-peroxid solution should be comparatively fresh and controlled by 

 testing from time to time. In the determination the reagents should be added 

 to the citric acid extract in the following order : Iron-citrate solution, hydrogen 

 peroxid, and magnesium mixture. To insure a rapid separation of the mag- 

 nesium precipitate the solution should be strongly agitated. 



The determination of potassium in fertilizers, F. Pilz (Ztschr. Landw. 

 Versuchsw. Osterr., 18 (1915), No. J^-5, pp. 117-108.) — From preliminary experi- 

 ments the author has developed a method for the determination of potassium 

 in separate (potassium chlorid, kainit, kieserite, etc.) and mixed potassium 

 fertilizers (pota.ssium superphosphate, wood ashes, etc.) similar to the per- 

 chlorate method. The procedure is described in detail and tables for the con- 

 version of KC1()4 into K2O appended. 



Note on the estimation of fat in food for infants, H. G. Chapman (Jour, 

 and Proc. Roy. Sue. N. S. Wales, JfS (191J,), No. 3, pp. -JCS-// 72). —Certain dis- 

 crepancies in the estimation of fat in infant foods by different procedures are 

 reported. 



The analysis of maple products. — VI, A volumetric lead subacetate test 

 for purity of maple sirup, J. F. Snell, N. C. MacFaklank, and G. J. Van 

 ZoEKEN (Jour. Indus, and Enyin. Chem., 8 (1916), No. 3, pp. 241-2^3, fig. 1). — 

 The volumetric lead subacetate method proposed by the authors consists in 

 diluting the sirup to 10 times its original volume and titrating with a standard 

 subacetate solution, the end i)oint being measured by ele<'trical resistance. The 

 volunu'fric lead number is then the abscissa of the p<»int of intersection of 

 two striiiglit lines on the plot, with volumes as abscissje and resistances as 

 ordinates. It is concluded that If future work corroboi-ates the past exi»erience 



