ESTIMATION OF COLLOIDAL MATERIAL IN SOILS. 9 
been concentrated in a moderate volume of water it was taken down 
to a nearly air-dry condition on the steam bath. The amount of 
colloid thus isolated amounted to about 1 per cent of the weight 
of the whole soil. 
Samples were taken for the adsorption of malachite green when the 
material had reached a thin, pasty condition, 6 while the water adsorp- 
tion was determined on the air-dried material and the ammonia 
adsorption on the material dried at 110° F. 
Suspensions of colloidal material which passed through the super- 
centrifuge were examined microscopically. Practically no particles 
were visible in the ordinary microscope. Examination was also 
made, using a 3-millimeter objective, 12.5 X eyepiece, and a dark 
field illumination. Under these conditions the apparent size of the 
particles seldom exceeded 0.3 micron in diameter. 7 The unaggre- 
gated particles had a most active Brownian movement and were 
readily coagulated by electrolytes. The suspended matter in most 
cases did not settle out appreciably on standing for many months. 
As the concentrated colloidal material was dried, it shrank greatly, 
finally becoming a horny or resinlike mass that had a conchoidal 
fracture. It would polish readily and adhere strongly to the tongue. 
This material was doubtless just as pronouncedly colloidal as that 
extracted by Schloesing (34) or that described by Ehrenberg (8, 
p. 102). It also seemed to. have the same properties as the materials 
described by Hilgard (16) and by Williams (39). 
METHODS OF DETERMINING QUANTITIES OF MALACHITE GREEN, WATER, AND AMMONIA 
ADSORBED. 
The adsorption of malachite green was determined as follows: 
Approximately 0.25 gram of colloid or 1 gram of soil was shaken over 
night with 25 cubic centimeters of water in an end-over-end shaker. 
Enough N/10 sodium oxalate was then added to precipitate any sol- 
uble calcium present. After shaking for 15 minutes, a 0.2 per cent 
solution of malachite green oxalate was added, with repeated shaking 
until the depth of color in the supernatant liquid was about the same 
as that of a check solution containing 0.0004 gram dye per cubic 
centimeter. Shaking was continued for 1 hour to insure complete 
adsorption. About 5 cubic centimeters of normal sodium chloride 
solution was then added to coagulate the colloid, and after centrifug- 
ing to throw down the coagulated material, the supernatant liquid 
was read colorimetrically against the check solution. The quantity 
of dye adsorbed was calculated from the amount added and the 
amount left unadsorbed. 
The determinations reported in Table 1 were all made on the moist 
colloid and on the air-dried soil. Several tests showed that the 18- 
hour shaking of the sample preliminary to determining the adsorption 
was needed to bring the material in the most adsorptive condition. 
This was particularly important in the case of soils. Duplicate 
6 In subsequent work reported here the air-dried colloidal material was used for determining the adsorp- 
tion of malachite green. 
7 Of course, some aggregates of larger size were seen, but these were susceptible of resolution into particles 
of the size mentioned. Part of the material may well have been invisible. This refers not only to exceed- 
ingly small particles, but also to colorless particles of any size which had a refractive index "sufficiently 
near that of water. 
57580—24 2 
