276 JOHN ERNST WEAVER 



the stations cups were run in duplicate and careful record was 

 made between atmometers near each other to see that they 

 were changing in a parallel manner. The data obtained from 

 9 of the 31 cups used are not recorded in this paper. One or two 

 showed irregularities in their operation, while two disappeared 

 one week before final collecting. 



Fourteen stations were established in the various associations, 

 care being taken in all cases to select spots which possessed the 

 average amount of herbaceous, shrubby, or tree vegetation 

 characteristic of that specific association as a whole. To facili- 

 tate comparison of the evaporation between the various stations 

 and to exhibit the progress of evaporation during the season, 

 the. average water loss per day between the readings has been 

 calculated and shown in graphs with the ordinates representing 

 the number of cc lost per day by a standard atmometer; the 

 abscissae being the intervals between the readings. Because 

 of the difference of rainfall, exposure, and soil, it is deemed best, 

 at least in this preliminary report, to consider the plant associa- 

 tions and evaporation on the plains separately from those on the 

 adjacent buttes and mountains. 



The first group of stations was established near Colfax, Wash- 

 ington, fourteen miles northwest of Pullman. Here the south 

 fork of the Palouse River has cut a canyon in the basalt to a 

 depth of 90 m. Along the north bank of this canj^on the bunch- 

 grass-rimrock association has its best development in the region, 

 while the well protected south bank is covered with an open 

 yellow pine (Pinus ponderosa) association above which, still 

 on the north slope, lies the prairie.. 



The most characteristic plant of the bunchgrass-rimrock 

 association is Agropyron spicatum (including var. inerine). In 

 the shallow soil, often only a few centimeters deep, the bunches 

 are well developed, averaging 20 cm. in diameter at the base and 

 reaching a height of one meter. An actual count in twenty- 

 seven square meter quadrats in the vicinity of station No. 1 

 showed these bunches to average 6 per quadrat. Forty-five per 

 cent of the ground was bare or very sparsely covered with mosses 

 and lichens, relicts from a preceding association. Between the 



