112 A Study of the Vegetation of 



of the soil, the temperature of the soil and the evaporating power 

 of the air, and progressive stages in succession. The water con- 

 tent of the soil has been found to vary directly with the stages of 

 the sere from a minimum in the bunch-grass stage to a maximum 

 in the cedar community. Conversely, the soil temperature and 

 the evaporating power of the air have been shown to decrease 

 directly in proportion to the advancement of the stage of develop- 

 ment of the sere. 



In the forest communities the light values decrease progress- 

 ively as the climax community is approached. 



In addition to the xerosere, beginning with lichens and mosses 

 on the bare rocks of basalt, granite, or quartzite and ending ulti- 

 mately in the climax forest community, we have also traced the 

 stages of the subsere in the reestablishment of cut-over or burned- 

 over forested areas. We have seen how these more or less de- 

 nuded areas may finally come to support the same type of vegeta- 

 tion that characterized them before the disturbance of their 

 normal development. 



Lastly, we have considered the hydrosere. The communities of 

 this sere, although occupying a relatively small area geograph- 

 ically, show distinct stages from the submerged, floating and 

 swamp communities to that of the wet meadow. The latter is 

 usually replaced by a xerophytic grass land community, or infre- 

 quently by a deciduous flood-plain forest. 



In concluding this report on the study of the vegetation of 

 southeastern Washington and adjacent Idaho, the writer wishes 

 to state that perhaps no one feels more keenly than he the neces- 

 sarily preliminary nature of the work accomplished. 



The Snake River Canyon alone, with its peculiarly dry, mild 

 climate stands in marked contrast to the colder, more humid area 

 above, and forms a rich field for detailed ecological investigation. 

 Not infrequently has the writer spent days in late February or 

 early March among the abundant prevernal flora of the canyon, 

 while weeks later most of the same species came into blossom on 

 the plateau above, while yet another period elapsed before the 

 awakening of the same species on butte and mountain prairies. 



During the progress of this study a number of plant species 



I 12 



