REVERSIBILITY OF THE COLLOIDAL CONDITION OF SOILS 515 
the air-dry and the oven-dry condition, there is an increase with one 
surface soil and one subsoil, and a decrease with the other two; the differ- 
ences are significant, however, with one soil only, the Dunkirk surface 
soil. Between the oven-dried and the ignited condition there is a decrease 
which is significant In every case but one. The only point which stands 
out clearly is that ignition decreases the adsorption of this dye. By 
comparing the results with those given in table 17 (page 510-511), it is 
seen that all these soils adsorb less of this dye than of methylene blue. 
Suspension experiments 
The suspension method is described on page 499. The results of the 
experiments with this method are given in table 22: 
TABLE 22. Per Cent oF CrypE Soin REMAINING IN SUSPENSION IN DISTILLED WATER 
Turrty MINUTES AFTER SHAKING 
i In distilled | In 4-per-cent 
Soil treatment water ammonia 
WON tINUOUSIVAMOISb esis okies bh Se apegs, «basis ey dicds Oe 2 sas eleva 9.15 21.90 
ASrecabefedl, Tl Cra ies Gig epee ae eee etree tere ra 6.55 17.10 
TG. sab bag cay WCDI AAED RRQ abe cua ee Rr aera 3.40 (sp 
BOA 8 a aes a RN a aL Ai ae ae ea pera ee 3.70 Glo 
Riccumintoncome Ny 4.80 9.45 
~ In both distilled water and ammonia the moist soil was deflocculated 
the most, and the soil that was air-dried once was deflocculated more 
than those that were air-dried sixteen or thirty-two times. The defloccu- 
lating influence of the ammonia was very marked. These results show 
that the flocculated condition is not immediately reversible, even with 
shaking in water. 
Additional experiments 
Owing to the striking influence of drying on the colloidal matter of the 
Clyde soil, as shown by the experiments in water-vapor adsorption and 
'as mentioned in the literature, it was thought probable that differences 
‘in the amount of humus extractable by ammonia could be found, With 
