202 A. F. GUSTAFSON 



following quantities: coarse sand, 10 cc; medium sand, 10 cc; fine sand, 

 15 cc; very fine sand, 17 cc; white silt loam, 18 cc 



According to Mosier and Gustafson (68) the surface per gram of the different 

 grades, assuming perfect spheres of average diameter for the grades, is as 

 follows: coarse sand, 30.2 sq. cm.; medium sand, 60.4 sq. cm.; fine sand, 

 129.3 sq. cm.; very fine sand, 302.1 sq. cm.; white silt loam, 2270.7 sq. cm. 

 Analysis^ of this latter soil shows (Bureau of Soils sizes) 1.5 per cent medium 

 sand and coarser grades, 1.7 per cent fine sand, 7.5 per cent very fine sand, 

 70.6 per cent silt, and 18.3 per cent clay. The calculation of the surface per 

 gram of white silt loam was based upon these percentages, the surface areas 

 given above, and the assumption that 824.8 sq. cm. is the surface of 1 gm. 

 of silt and that 9090.2 sq. cm. is the surface of 1 gm. of clay. 



The figures for total surface per gram are undoubtedly more nearly accurate 

 for white silt loam than for the quartz since the latter is angular and of all 

 conceivable shapes with no spheres. The surface of the quartz must be 

 considerably greater than the figures shQw, whereas in the white silt loam soil 

 the angles have been worn off the particles to some extent, bringing them 

 somewhat toward the spherical shape — yet the figures at best are but an 

 approximation which may be of some value for purposes of discussion. 



Solutions of potassium nitrate were then made up of such strength that 

 1 cc of the first solution contained 0.1 mgm. of nitrate and the other, 0.5 

 mgm. of nitrate. The hygroscopic capacity of quartz is so low that this factor 

 was ignored. A quantity of potassium nitrate solution equal to the water- 

 holding capacity was added from an accurate burette to each of four 50-gm. 

 samples of the grades of quartz and of white silt loam. These moistened 

 samples were covered to prevent evaporation and set aside over night to per- 

 mit of any reaction or adjustment in the moist mass, after which two samples 

 of each were placed in the oven at 105°C. for 8 hours. Nitrates were deter- 

 mined immediately on the other two samples. Each was placed in a 500-cc. 

 beaker and distilled water added to make a total of 250 cc; for example, to 

 the coarse sand containing 10 cc. of KNO3 solution, 240 cc. of distilled water 

 was added and to the very fine sand having 17 cc. of the solution, 233 cc. of water 

 was added. Thus the relationship throughout was 1 part of soil or quartz 

 to 5 of water. 



This experiment was run in two parts, 6a and 6b. In the first part, the 

 nitrate solution used contained 100 parts per million, or 0.1 mgm. per cubic 

 centimeter of nitrate and in the second, the solution contained 500 parts per 

 miUion or 0.5 mgm. per cubic centimeter. In all other respects the two trials 

 were identical. The stronger solution was used in the second trial because a 

 dilution of 10 cc. of the first solution to 250 cc was considered too great for 

 securing results of the degree of accuracy desired. 



The results of the first set of determinations are given in table 8. 



The results of the second set of determinations are given in table 9. 



* This analysis was kindly furnished by Dr. Bizzell and Dr. Buckman. 



