204 



DETEKMINATION OF rHOSl'IIOKlf (JXI])E. 



the last corresponds roiig-lily to the formuhi assio'iied to the 

 yeUow i)reeii)itate hy Gibhs* and by Gh\ddino-.t There is 

 abundant evidence to show that the reaction does not exactly 

 folio V,' theory v.hen jiractically carried out. This is so well 

 recog-nised that the IJ.S. Association of Ofhcial Ag-ricultnral 

 Chemists no longer prescribes a conversion factor, but directs 

 each chemist to determine liis own. On the other hand, we 

 have found no apjneciable dift'eiences between results obtained 

 by different chemists in this laboiatory, where all use the 

 same factor. The results g'iven on this page serve as an 

 illustration. 



Influence of Silica and of Iron.J 



It is generally held that the previous removal of silica 

 is essential to the success of the m lydate-magnesia process 

 for phosphoric acid. Silica is said not to interfere with the 

 A'olumetric method used by the Association of Official Agri- 

 cultural Chemists. But, if silica, be precipitated as silico- 

 niolybdate along* with the phospho-molybdate, it is difficult 

 to see why it should not affect the alkalimetric method. 

 Lorenz has showii that silica in amount up to 01 gram to 

 0'04 gram, of i)h()siihoric oxide has no effect on his gravimetric 

 process ; conseciuently it should have no influence in the 

 volumetric modification. A few results are given in support. 



1. If basic slag be digested with sulphuric acid, prac- 

 tically no silica g'oes into solution ; while if the extraction be 

 made with nitric acid, very appreciable quantities of silica 

 are dissolved. A sample of basic slag extracted by three 

 diff'erent methods gave the following results : — 



* Jour. Am. Chcni. Soc. iii, 317 (1881). 



f.J<nint. Am. Chem. Soc, xviii, 23 (1896). 



X The influence of citric acid and of ammoninni citrate was 

 discussed in the previous paper. 



§ Wiley, H. "W. : " ]*rinciples and Practice of Agricultural 

 Analysis," vol. ii, p. 199 (2nd edition). 



II Analyses by B. J. Smit, Assistant Chemist, Division of Chemistry. 



