186 § 104. ANALYSIS OF SOILS AND ROCKS. 



To 1000 grms. of the soil add water enough to make 

 2000 c.e. with that already in tlie soil, shake the mixture 

 frequently in the course of 48 hours, decant and filter 

 1000 c.c. through a dry filter, add some sodic carbonate 

 to the filtrate, evaporate the solution to a small bulk on 

 the water-bath, and divide the residue into two equal 

 parts. Determine nitric acid in each portion, represent- 

 ing 250 grms. of soil, in the usual manner (§ 62 ci). 



d. Chlorine. — To determine this, add enough water to 

 300 grms. of the air-dried soil to make 900 c.c. with what 

 is already contained in it, shake the mixture frequently 

 in the course of 48. hours, decant and filter 450 c.c. of the 

 liquid, add a little sodic carbonate to the filtrate, evapo- 

 rate to about 200 c.c, filter again, supersaturate the fil- 

 trate, which represents 150 grms. of soil, with nitric acid, 

 and precipitate the chlorine in the acid solution with 

 argentic nitrate (§ 63). Treat the precipitate as one pro- 

 duced in the presence of organic matter. 



e. Sulphur* — It often happens that a much larger 

 amount of sulphuric acid is found in the soil after ignition 

 than before, indicating that a notable quantity of sul- 

 phur exists there as sulphuret, or in some organic combi- 

 nation. To determine the total amount of sulphur in the 

 soil, mix with 50 grms. of it, 1-2 grms. of pure saltpetre, 

 moisten the mixture in a platinum dish with a solution of 

 pure potassic or sodic hydrate, free particularly from sul- 

 phates, dry, and heat gradually to a red heat ; when the 

 mass is cool, boil it with dilute hydrochloric acid to 

 which a little nitric has been added, evaporate to dryness, 

 and eliminate silica in the usual way, but without weigh- 

 ing it ; add water to the filtrate from the silica, and pre- 

 cipitate sulphuric acid with baric chloride. 



/. Hydratcd aluminic and ferric oxides.— To deter- 

 mine the quantity of these substances, that, according to 

 Knop, play so iin2)ortant a part in the absorbent action 



