30 REPORT — 1875. 



The time allowed for precipitation varies from ten minutes (with vigorous 

 stirring) to twenty-four hours ; but most chemists are of opinion that six or 

 eight hours are sufficient. 



Very few chemists recommend re-solution and re-precipitation of the 

 double phosj)hate. 



Eut few precautions appear to be taken in the ignition of the precipitiitc. 

 The more careful analysts thoroughly dry the precipitate and remove it from 

 the filter, igniting it first gently, then intensely. 



Very different opinions are held as to the accuracy of the results obtained 

 by precipitation with magnesia. In many cases the obseiTers are merely 

 able to say that the process gives fairly constant results on repetition ; in 

 other cases they state that very concordant results have been obtained when 

 the same sample has also been analyzed by some chemist of repute. 



In some cases the observers consider that the process is liable to give 

 results somewhat below the truth, owing to the slight solubility of the 

 double phosphate in the mother-liquid and the loss of phosphate in the 

 oxalate-of-calciura precipitate. lu other cases the process is said to give 

 results in excess of the real amount, owing to the presence of other magne- 

 sium salts or of iron or aluminium in the double phosphate precipitate. 



A most elaborate series of experiments has been made by Mr. T. E. 

 Ogilvie on the magnesia process and the variations to which it is liable *. 

 His resiilts have been to a great extent confirmed by the researches of Mr, 

 E. M. Dixon. On the other hand, Professor Church writes, " My confidence 

 in this plan when carried out successfully, giving time for any oxalate to 

 fall after addition of citric acid and excess of ammonia, is not shaken by 

 Ogilvie's results reported recently in the ' Chemical News.' His experiments 

 seem to me to exaggerate the errors of the method greatly. Several times 

 have I got nearly identical results by the use of the molybdic method for 

 separating the P^ 0^ from the soluble part of a superphosphate, and by the 

 use of the oxalic method. I always use a measured quantity of the am- 

 moniacal magnesium chloride ; but considerable excess of this reagent often 

 produces but little influence on the result — sometimes none." 



Direct Citnc-Ackl Method. 



The method of Joulie is employed by some chemists of wide experience. 

 In this i^rocess the iron, aluminium, and calcium are all retained in solution 

 by citrate of ammonium, and no attempt is mado to separate the calcium as 

 oxalate ; but the phosphate is at once precipitated from the ammoniacal 

 solution by " magnesia mixture," the precipitate being cither ignited and 

 weighed or dissolved in acetic or nitric acid and the solution titrated with 

 uranium. This method is in 'many respects similar to that recommended by 

 rreseuius, Neubaucr, and Luck f. 



Iron-Acetate Ilethod. 



A few chemists employ a process of which the following is an outline : — 



The neutralized hydrochloric solution of an insoluble phosphate (freed 



from silica), or the aqueous solution of a soluble phosphate, is treated with 



acetate of ammonium (filtered from the precipitated phosphates of iron and 



aluminium if their separate estimation is required) and sufficient ferric 



* Chemical News, May 6, 1870; Proceedings of the Philosophical Society (Chemical 

 Section) of Glasgow, 1874 and 1875, &c. 

 t Zeit. f. anal. Chem. ix. 16, and Sutton's 'Volumetric Analysis,' second edition, p. 241. 



