194 BOTANICAL GAZETTE [march 



intervened, it would appear that the loss of moisture by transpira- 

 tion during the winter must have been very slight. 



It is of interest that in the spring of 191 1 the subsoil at depths 

 of 7-15 ft. in field E at Wauneta was found quite moist, showing 

 an average ratio of 1.6. The explanation of this will be discussed 

 in a later paragraph. 



On the last occasion samples were taken from near Imperial also. 

 While the soil and subsoil in all of the fields at McCook as deep as 

 sampled, and in all those at Wauneta except in the lower levels of 

 E and F, were derived from the loess, and showed hygroscopic 

 coefficients between 7 . 5 and 1 1 . o, all the soils and subsoils at 

 Imperial were residualin origin with hygroscopic coefficients vary- 

 ing all the way from 2.0 to 10.7. At Imperial, in contrast to 

 McCook and Wauneta, we sampled some fine sandy loams as well 

 as the more numerous fine textured soils, none of the latter, however, 

 being of loessial origin. Comparing the ratios it will be seen that 

 the prairies with the finer soil. A, B, and E, were as dry in the first 

 6 ft. as those at Wauneta and McCook, and that the one of the two 

 with sandy soil and subsoil, C and D, was but slightly more moist. 

 Below the sixth foot they were distinctly moister, but in none of 

 them was the ratio much above i . o. The reduction of the moisture 

 content to the hygroscopic coefficient was general and was inde- 

 pendent of the relative hygroscopicity. 



After a wet winter following a prolonged drought. — 

 Until the spring of 191 2 no more sampling was done at McCook, 

 Wauneta, or Imperial. The weather of the intervening months 

 had been unfavorable for any marked increase in the moisture 

 content of the deeper subsoil, although very favorable for moisten- 

 ing the surface foot. April and May of 191 1 had a rainfall some- 

 what below normal, June and July were very dry, while during 

 August and the first few days of September considerably more rain 

 than usual fell. The rest of September was dry, but the precipita- 

 tion of October was 3 times the normal and of such a character that 

 there was little chance for run-off; as vegetation had become dor- 

 mant the loss by transpiration must have been slight. November 

 was very dry, but the precipitation of December was twice the 

 normal, a heavy rain on December 20 further moistening the 



