SOILS. 119 



"luiiiuis degree" of 15 to 25, lirh garden soils over 50 iiud woods earth 

 and peaty soils over 4^0. 



The soils examined were classified into 2 main categories, (1) ordinary 

 soils containing less than 50 per cent of humus, and (2) those containing 

 more than 50 per cent. Tlie first was divided into calcareous, clay, 

 and sandy soils; and each of these further divided into 4 other sub- 

 divisions depending upon the relative proi)ortions of lime, clay, and 

 sand. The scheme of classification is explained graphically and in notes 

 and tables. 



For determining moisture 5 gm. of soil which had passed through a 

 sieve containing 140 meshes to the square centimeter was dried at lOS*^. 



For the chemical analysis 50 gm. of soil was calcined for 10 to 15 

 minutes at a moderate red heat in a platinum crucible to destroy all 

 organic matter, and then treated with nitric acid by gently boiling in 

 a flask with condenser for 1 hour. The nitric acid was prepared by 

 diluting 50 cc, plus one half the weight in grams of the amount of car- 

 bonate of lime previously found in the soil, of nitric acid to a volume 

 of 250 cc. The solution was allowed to cool, filtered, and 100 cc. taken 

 for the determination of phosphoric acid and the same amount for lime 

 and potash. 



In the determination of phosphoric acid the molybdate precipitate 

 was obtained in the usual way, dissolved in ammonia, and precipitated 

 with magnesia mixture in the presence of a small amount of citrate of 

 ammonia. The ammonium-magnesium phosphate after thorough wash- 

 ing was dissolved in nitric acid (1 jyart of HjS^O;j to 9 of water), a slight 

 excess of ammonia added, and the solution finally rendered slightly 

 acid by addition of 1 or 2 cc. of acetic acid, using cochineal as an 

 indicator. Phosphoric acid was then determined l>y means of a standard 

 uranium solution by Halot's method. 



In the determination of lime 100 cc. of the origin;'<1 solution was 

 boiled in a flask with a slight excess of ammonia, the pre('ii)itate thrown 

 upon a filter and washed with boiling water, the washings being saved. 

 The precipitate was dissolved ui^on the filter with a boiling mixture of 

 20 cc. of nitric acid and 20 cc. of water, the solution precipitated again 

 with ammonia, boiled, filtered, and washed 3 times with boiling water, 

 the solution from this operation being added to the washings of the 

 first precipitate. 



The lime was precipitated in this solution by means of carbonate of 

 ammonia and weighed as carbonate. 



The potash was determined by precii)itating with platinic chlorid in 

 the usual way, and then decomposing the double chlorid with formate of 

 soda. The potassium-platinic chlorid was treated with a mixture of 

 9 parts of alcohol and 1 part of ether until the filtrate was perfectly 

 colorless, the crystals dissolved in boiling water, and the solution 

 gently boiled with 4 or 5 cc. of 15 j^er cent formate of soda for 15 min- 

 utes, adding more formate if the decoloration was not complete. After 



