On Tribasic Phosphoric Add. 183 



phate of magnesia was carefully washed, and, without dissolv- 

 ing a second, time, dried and weighed. 



The ignited, mass in (b) was snow white, in (a) and (c) 

 slightly tinged with vellow. ' 



The fluid which was filtered off from the yellow precipitate 

 containing the phosphoric acid, was heated a second time with 

 more molybdic acid solution without coloring it in the least. 

 Ammonia was then added in excess and the precipitates washed 

 — (a) not quite as long as (b), and (1>) not quite as long as (c). 

 After drving and weighing, the residue was dissolved in 

 hydrochloric acid, and the Fe 2 3 separated from the molybdic 

 acid, by a twofold treatment with poly-sulphide of ammo- 

 nium. After washing the sulphide of iron, dissolving it in 

 HC1, oxidizing with X0 5 , filtering from separated sulphur, 

 and precipitating with NH S , this precipitate was, after careful 

 washing, dried and weighed. The results are as follows : 

 (a) 0.438 gr. (b) 0.2750 gr. (c) 0.3195 gr. 



PO- 5 36.74 36.73 36.71 



Fe,0 3 33.58 (10.44MO,) 33.40 (9.51 MO s ) 33.40 (8.35M0 3 ) 

 H() 29.68 29.87 ^M 



mean PCX =36.74 

 Fe 2 3 = 33.46 

 HO =29.80 



II. Method with Si0. 2 (Berzelius 1 by fusing 



I prepared for this method a mixture which I also think 



useful in the analysis of iron ores, and which consisted of 



lgr.-KO, NO, 



5 gr. — Si0 2 



20 gr.— KG, CO, 

 lOgr.—XaO, 00 



15 to 25 parts of this mixture are sufficient for :'. gr. of iron 

 ore. 

 The substance was fused with five parte of this mixture, and 



