422 



Journal of Agricultural Research 



Vol. X, No. 8 



Table XVIII. 



-Rise in tubes of soil with the lower end immersed in water kept at a 

 constant level, e/c— Continued. 



Time. 



SoilJ. 



Exper- 

 iment 

 I. 



Exper- 

 iment 

 II. 



Exper- 

 iment 

 III. 



Soil K. 



Exper- 

 iment 

 I. 



Exper- 

 iment 

 II. 



Exper- 

 iment 

 III. 



Soil L. 



Exper- 

 iment 

 I. 



Exper- Exper- 

 iment iment 

 II. III. 



SoilM. 



Exper- 

 iment 

 III. 



Hours 



1 



2 



3 



4 



24 



48 



72 



96 



120 



144 



168 



192 



216 



240 



Cm. 



18 



Ctn. 



84 



43 

 78 

 94 

 104 

 III 

 116 

 121 



Cm. 



102 

 103 



46 



96 

 9S 

 100 

 102 

 103 



46 



Cm. 



29 

 36 

 39 

 42 

 56 

 61 

 64 

 67 

 69 

 70 



Cm. 



The data in Table XVII in the columns headed "From Table XVIII" 

 serve to illustrate the concordance of data from the duplicates, the 

 averages only of which are reported in Table XVIII, Only with soils 

 D, F, G, and J in Experiment I and with soils F, G, and H in Experi- 

 ment II does the divergence approach in magnitude that shown in Table 

 XVII by D in the driest condition. 



RELATION OF RISE TO HYGROSCOPICITY 



In Table XIX the data are rearranged to show what relation the 

 rapidity and distance of rise bear to the hygroscopicity in each of the 

 three moisture conditions. At first the coarser soils showed the most 

 rapid rise, but the difference gradually lessened until those of inter- 

 mediate hygroscopicity led. After the first three or four days the two 

 subsoils G and H, with coefficients of 8.2 and 7.6, respectively, showed 

 the most rapid rise in all three moisture conditions. Closely following 

 these was the subsoil K, with a coefficient of 3.4, while L, with a coeffi- 

 cient also of 3.4, was still farther behind. This difference among the 

 three soils, alike in hygroscopicity, was shown in all three moisture 

 conditions. Soils A, J and K, differing widely in hygroscopicity, 13.3, 

 5.6, and 3.4, respectively, were much alike at the end of 10 days in 

 Experiments I and II, but differed much in Experiment III. In soil B, 

 and to a less extent in D, with coefficients of 12.9 and 10.2, respectively, 

 the rise was very slow in all three experiments. There thus appears no 

 definite dependence of the rate of rise upon the hygroscopicity (fig. 2). 



