37 



in the solution at the end of the fourth washing approximately 0.0004 

 gram of the original nitrogen. Actual Kjeldahl determinations 

 were made on 250 cc. portions from the fourth washing in the case 

 of duplicates from the same soil, and the results indicated that 

 0.0006 gram of nitrogen still remained in the solution. Since this 

 was within experimental error of the theoretical value, the method 

 of washing by decantation was followed in all the subsequent work 

 with mineral soils, or those which had mineral soils added previous 

 to the analysis. 



The residue from the hydrolyzed soil was evaporated to dry- 

 ness on the steam bath in an evaporating dish, then further dried 

 at about 110 C. After cooling, this dry soil was passed through a 

 1 mm. sieve and after being thoroughly sampled, duplicate nitro- 

 gen determinations were made on 15 gram portions and the total 

 nitrogen remaining in the soil calculated. These results are listed 

 as ''insoluble humin nitrogen in the soil." The weight of the dry 

 soil divided by the average specific gravity (2.6) represented the 

 actual volume occupied by this soil residue. The total volume of 

 the hydrolysate minus the volume occupied by the insoluble resi- 

 due gives the actual volume of the soil solution. 



Since the analysis was made on 500 cc. of the soil solution it 

 was necessary to recalculate the total "insoluble humin nitrogen in 

 the soil" in order to determine the amount of this humin nitrogen 

 actually belonging to the aliquot analyzed. 



The total nitrogen belonging to the solution analyzed was 

 found by taking the sum of the total nitrogen in the solution and 

 the above insoluble humin nitrogen. Knowing the total nitrogen 

 content of the soil before hydrolysis and the total nitrogen in solu- 

 tion, the per cent of the total nitrogen in solution after hydrolysis 

 can be determined. 



The 500 cc. aliquot was concentrated under reduced pressure 

 until the hydrochloric acid was practically removed and the am- 

 monia nitrogen determined in the manner outlined under the peat 

 analysis. 



The "humin" fraction precipitated by the calcium hydroxide 

 was almost colorless or light yellow, due to the iron salts con- 

 tained in it. This bulky precipitate was always washed by de- 

 cantation after the method above described, except that distilled 

 water was used, the united washings being concentrated to 200 

 cc. or less for the precipitation of the basic nitrogen. 



It was found necessary to use 35 grams of phosphotungstic 

 acid for the precipitation of the bases. The remainder of the 

 analysis was carried out as directed under peats. 



3. The method for determination of "Jodidi numbers." A 

 100 cc. portion of the clear hydrolysate was concentrated under 

 reduced pressure and the ammonia nitrogen determined in the 

 usual manner. The residue remaining in the flask after this de- 

 termination was dissolved in concentrated hydrochloric acid and 

 phosphotungstic acid added.* After standing the usual length of 

 time the precipitate was filtered of! and the total nitrogen deter- 



*lt will be noted that no "humin" fraction is separated. In that respect 

 the "Jodidi numbers" differ from Hausmann numbers. 



