280 JOHN ERNST A\ EAVER 



rhizomes, and thus it becomes a partial sod former. The differ- 

 ence between prairie vegetation occupying exposed and sheltered 

 slopes is so marked that it will be given special consideration 

 when comparing the results obtained at the base station where 

 evaporation was determined in both situations. 

 • The graph representing the evaporation at the prairie station 

 (fig. 2) shows a great similarity in its general course and in its 

 simultaneous maxima and minima to that of the preceding. 

 The markedly smaller average daily evaporation of 21 cc. (from 

 May 21 to September 10) is due to the slightly lower tempera- 

 ture and greater humidity, and the less wind velocity brought 

 about both by the position of the atmometer on the leeward 

 slope and by the vegetation which completely covers the ground. 



An open yellow pine forest occupies the more sheltered part of 

 the protected side of the canyon from which vantage ground 

 it is slowly invading the prairie. That these pines are relatively 

 recent invaders is shown both by the age of existing trees and the 

 almost entire absence of the remains of former generations. The 

 characteristic under-shrub is Opulaster pauciflorus. Important 

 herbaceous relicts from the prairie are Potentilla blaschkeana, 

 Siversia ciliata, Galium boreale, Trillium petiolatum, Viola adunca, 

 Synthyris rubra, and Heuchera glabella; while the presence of 

 such plants as Cytherea bulbosa, and Pterospora andromedea 

 indicate decidedly mesophytic conditions. The pine grass 

 {Calamagrostis suksdorfii) and Deschampsia calycina are also 

 important species here. The former often covers the ground 

 completely, while in other places a carpet of pine needles nearly 

 excludes vegetation. In carefully chosen typical areas, about 

 20 m. below station No. 2, and near the upper edge of the pines, 

 station No. 3 was maintained. An examination of the graphs 

 here obtained (fig. 2) shows again a general conformity to those 

 preceding; but the average rate of evaporation for the whole 

 period was only 11.8 cc. per day. 



A second set of stations was maintained on the basaltic soils 

 near Pullman, in the prairie formation. Preliminary investi- 

 gations in the spring and autumn of 1912 showed that the evapora- 

 ation on the two hillsides was markedly different; accordingly 



