Apr. I, 1918 



Capacities of Soils for Irrigation Water 



31 



Each sample was placed on a weighed wire-mesh support, weighed 

 upon a solution balance, dipped into warm paraffin until fully coated 

 (usually three or four times), reweighed, and placed into a large desic- 

 cator of known volume.* 



The desiccator containing the soil, paraffin coating, and wire-mesh 

 support was filled with water from a i,ooo-c. c. graduate cylinder and 

 the volume of the soil, paraffin, etc., determined by displacement. From 

 the known weights of the 

 paraffin and wire-mesh sup- J* ^ 



port their volumes were ^^^*^ 



computed, using 0.9 as the 

 density of paraffin and 8.0 

 as the density of the wre. 

 After the volume measure- 

 ment the cubes of soil were 

 broken and a 200-gm. sam- 

 ple taken for the determina- 

 tion of the moisture content. 

 The results are presented 

 in the last column of Table 

 XII.- 



The results presented in 

 column 3 of Table XII were 

 obtaind by the use of a spe- 

 cially prepared iron cylin- 

 der, having a diameter of 6 

 inches, which could be. driven 

 into soils of ordinary com- 

 pactness only with great diffi- 

 culty. It was therefore not 

 attempted to drive it into this 

 very compact clay; but the top of a column of soil having a circular 

 cross section of about 8 inches in diameter was carefully smoothed and 

 leveled, after which two spatulas were inserted horizontally into the 

 column. The iron cylinder was placed vertically upon the column, as 

 indicated in figure 14. The column was trimmed very carefully with 

 a trowel until its diameter was slightly greater than that of the cylinder. 

 The weight of the cylinder, aided by very light tapping, caused it to 

 move slowly down over the soil column. 



' The cx)ver of the desiccator used has a i-inch diameter opening in its center, thus making it possible to 

 measure the desiccator volume accurately. 



2 The probable errors were determined by use of Peter's Formula Rm= °' " ^^ Where Rm=the 



wV«=i 

 probable error of the arithmetic mean of n determinations, and S V=the sum of the differences between 

 the arithmetic mean and each determination. In this case n=s- It is recognized that where n is so small 

 the formula is not strictly accurate. 



AurALFA 



Fig. 13. — Diagram showing plan of excavation for the 

 determination of the volume weight of soil by means of 

 the parafifin-inunersion method. 



