4 BULLETIN 1122, U. S. DEPAETMENT OF AGRICULTURE. 



the soil into two fractions, colloid and noncolloid, and then deter- 

 mine the absorptions of these fractions. This is similar to the method 

 used by earlier investigators and is the one the authors first tried. 

 The work is described under "Estimation of noncolloidal absorp- 

 tion from soil fractions." The data obtained in this way were not 

 considered trustworthy, as it was exceedingly difficult to separate 

 the bulk of the colloidal matter from the larger soil grains and it 

 was practically impossible to tell whether or not a complete separa- 

 tion had been attained. 



It was thought that a more accurate estimation of the noncolloidal 

 absorption of the soil could be obtained by investigating the absorp- 

 tion of the individual soil minerals, since the noncolloidal part of 

 the soil is made up of mineral particles of various degrees of fineness 

 above the arbitrary limit of 1 micron. Each of the common soil 

 minerals was ground to the proper fineness and the absorption of 

 the mineral powder was determined. From the absorptive capacities 

 of the different minerals and from the mineralogical composition of 

 the soil, the noncolloidal absorption was calculated. Determinations 

 by this procedure are given under "Estimation of noncolloidal ab- 

 sorption from mineral particles." 



An indication of the nature of the colloidal material in soils was 

 also obtaiTied by comparing the absorptive capacities of finely ground 

 minerals and various gels with the absorptive capacities of the mixed 

 colloids which had actually been extracted from soils. This part of 

 the work is detailed under "Evidence concerning the nature of the 

 soil colloids, afforded by absorption data." 



METHODS FOR DETERMINING ABSORPTION. 



Determinations were made of the absorptive capacities of all 

 samples for malachite green, water vapor, and ammonia gas. The 

 methods followed in determining the absorption of water vapor and 

 ammonia gas have been given in detail in previous publications and 

 therefore need only brief descriptions. 



In the absorption of ammonia {17) a sample of material which has 

 previously been dried for 18 hours at 110° C. is exposed at 0° C. 

 to dry ammonia gas under atmospheric pressure until absorption is 

 virtually complete. The amount of ammonia driven off by evacuat- 

 ing the material at 100° C. is then determined. 



The absorption of water {21) involves a determination of the 

 amount of water which the sample will take up when exposed for 

 five days to the vapor of 2 per cent sulphuric acid maintained at a 

 temperature of 30° C. 



The method for the absorption of malachite green, which differs 

 somewhat from that described in. a previous publication {17), is as 

 follows: From 0.25 gram to 1 gram of material is mixed with 25 

 cubic centimeters of water and shaken over night. After shaking, 

 the suspensions are treated with enough N/10 sodium oxalate solu- 

 tion to precipitate any soluble calcium present. The amount of 

 sodium oxalate which it is necessary to use can be determined by 

 a preliminary test. After adding the sodium oxalate solution, the 

 whole is shaken 15 minutes to insure complete precipitation of the 

 calcium oxalate. Malachite green (a 0.2 per cent solution) is then 

 added with repeated shakings until the depth of color of the super- 



