- 162 - 



The loss from these cylinders was aided by gravity 

 which assistance was continued throughout the experiment, th 

 though in a diminishing degree as the experiment progressed. 

 When the first few weighings were made it was noted that 

 the cheese cloth extending beyodd the under edge of the 

 cylinders was moist under the samples from plats 2,3,7 & 

 10, and that this condition continued largest in the case 

 of the latter. This assistance to evaporation rendei&ed 

 by the cloth would serve to increase the variation between 

 the finer and courser soils. 



After the loss from sensible percolation had ceased, 

 the further loss rapidly declined till what seemed to be a 

 fairly uniform, level was usually reached. This fairly 

 uniform level gradually became less pronounced as the soil 

 became coarser till in plat 10 it was somewhat indefinite. 



Table "G" shows the loss of water for the first 

 hour to be very variable, plats 10, 25, and 9 being the 

 lowest. By referring to Table "C", Column VI^^ it will 

 be noted that where the pore space was determined by tak- 

 ing the sum of of the water retained from saturation and 

 the water in the samples as taken from the field (Formula 1), 

 these plate give the lowest determination (of pore space) 

 ojT any other plats except the two samples from plat 22. 

 Too much importance should not be attached to the variation 

 in these latter samples since the various data show that 

 thev often vary from what would be expected. 



The correllation between the relatively small 



