EXPERIMENT STATION BULLETINS. 593 



Kesults are averages of duplicate determinations and except as otlier- 

 wise stated are given as parts per million on soils. A complete report 

 of the experimental procedure is given in this paper. 



The chemicals used were KCl, commercial 14 per cent acid phos- 

 phate, commercial hydrated lime, CaSO^, KaCl, NaNO,, Ca, (P04)a, 

 Ca(H,P0J2 and CaCOg. 



Method of Analysis. 



SOILS. 



Loss on ignition — official method, U. S. Dept. Agr. Bur. Chem. Bui. 

 No. 107, rev. 



Total Nitrogen — official method. 



Total mineral elements — ^sodium peroxide fusion and official methods 

 for separations. 



Soluble mineral elements — solution made by provisional method for 

 the more active forms of phosphoric acid in soils. Official methods used 

 for separations. 



Soil Extracts. 



Chlorine, carbonates, nitrates — determined by methods given in U. S. 

 Dept. Agr. Bur. Soils, Bui. No. 31. Carbonates were estimated also by 

 Seyler's method for natural waters. Volumetric Analysis — Sutton., p. 

 100 and results reported in this paper as fixed carbon dioxide. 



Silica — on 300 c.c. of solution by Official Method, except soluble resi- 

 dues were taken up by nitric acid in place of hydrochloric acid. 



Iron, aluminum and phosphorus collectively, calcitim and magnesium — 

 determined after silica by treating the solutions with excess of sodium 

 and potassium carbonates, evaporating to dryness and igniting to de- 

 stroy organic matter. Separations were then made according to official 

 methods. Iron and phosphorus determinations were made first and when 

 phosphorus pentoxide exceeded in quantity the iron oxide a measured 

 excess of a standardized solution of ferric chloride was added to precipi- 

 tate all the phosphorus, and the requisite correction made. The calcium 

 oxalate was titrated by potassium permanganate. 



Sulfur and phosphorus — 200 c.c. of original solution was concentrated, 

 treated with excess of sodium and potassium carbonates, evaporated, 

 ignited at low temperature, taken up with dilute hydrochloric acid, and 

 sulfur estimated by the official method. The solution was then neutralized 

 by potassium carbonate, acidified with nitric acid and the official method 

 used to determine phosphorus. 



Iron oxide, potassium and sodium — 300 c.c. of original solution evapo- 

 rated treated with concentrated sulfuric acid, evaporated and ignited. 

 This process was repeated until residues were quite white. Residues were 

 dissolved in hydrochloiic acid, filtered, iron, etc., i»reeipitate(l by am- 

 monia and the iron estimated by the permanganate volumetric method. 

 The filtrate was then cleared by the usual methods and sodium and 

 potassium weighed as sulfates; then dissolved, changed to chloride with 

 barium chloride, excess of barium removed, and separation of the 

 chlorides effected by platinic chloride according to the official method. 



Aluminum oxid— estimated from iron, aluminum, and phosphorus 

 oxides bv differences. 



