William J. Gies 



191 



gauze over a very low flame until all the moisture is driveti off. The 

 crucible should then be placed in a large platinum dish and heated to 

 a high temp. over a Bunsen flame. Most of the filter paper is burned 

 off at this stage. The crucible should be finally heated to constant 

 weight over a free flame with a Meker burner. During the final igni- 

 tion the crucible should be nearly covered with a crucible lid. 



2,. An insufficient qiiantity of ammonium chlorid in the sol. A 

 large quantity of ammonium chlorid would increase the solubility of 

 calcium phosphate and decrease the solubiHty of iron and aluminum 

 phosphates. I happened not to mention this fact in my original paper. 

 The ammonium chlorid was introduced as f ollows : After digesting the 

 organic matter with nitric and sulfuric acids, the mixture was boiled 

 nearly to dryness to drive off all the NO2 and the excess of sulfuric 

 acid. This procedure always resulted in the production of a sediment 

 which would not dissolve in water. It was found necessary to treat 

 this Sediment two or three times with hot hydrochloric acid to effect 

 complete Solution. This sediment always contained more or less iron, 

 as could be observed from the color of the sol. after boiling with 

 hydrochloric acid. In view of the fact that so much acid was used 

 in dissolving this sediment, it was deemed advisable to nearly neutralize 

 the portions taken for analysis with ammonium hydroxid prior to 

 starting with the A. O. A. C. method. When working with pure salt 

 Solutions, I have always added large quantities of a sat. ammonium 

 chlorid Solution. 



I tested my modified A. O. A. C. method with regard to its effi- 

 ciency when large quantities of iron and aluminium are used. The 

 bulky precipitates were cumbersome to wash, but my results (sub- 

 mitted below) show that the method is reliable even when the com- 

 bined phosphates weigh as much as 400 milligrams. 



These results, coupled with my previous findings, convince me 

 that the method is absolutely reliable, if followed with care. 



