420 



Journal of Agricultural Research 



Vol. X, No. 8 



soil and subsoil, respectively, had been saved from the previous experi- 

 ments and kept in tightly sealed jars and so were in the same moisture 

 conditions as when previously employed, we filled two tubes with each, 

 one by jarring and the other by tamping as above described. These six 

 pairs of tubes we allowed to remain in contact with the water for lo 

 days, as in Experiments I and II, the rate of rise being observed as 

 before and reported in Table XVII in the column headed "Later experi- 

 ment." 



Table XVII. — Concordance of data as affected by method of filling tubes and by paral- 

 lelism of experiments 



SOIL H (hygroscopic C0EFFICIBNT=7.6) 



SOIL D (hygroscopic C0EFFiaENT=I0.2) 



1 The soil column broke at end of third hour. 



* Tube cracked. 



With the soils in the driest condition no influence of the tamping versus 

 the jarring is shown, while in the moistest both showed a slower rise 

 during the first 24 hours when tamped, but the one a somewhat more 

 rapid rise during the following nine days, the tube with the other breaking 

 before the end of the second day. In the intermediate condition both 

 soils at first showed the more rapid rise in the tube filled by jarring, but 



