352 CAENEGIE INSTITUTION OF WASHINGTON. 



still occupy areas of several square miles and serve to give a fairly- 

 complete picture of the original community. In wide stretches of the 

 Great Interior Valley, such as the vast Bakersfield plain, where over- 

 grazing was most serious, it was at first thought that Stipa had com- 

 pletely disappeared. A careful search during the past spring revealed 

 relict bunches and groups wherever there was some protection, while 

 the region about Tehachipi Pass showed it as a dominant still. This 

 is in accord with the statement of Fremont, who speaks of this as a 

 bunch-grass country. 



The question of the disappearance of the buffalo grass and the incom- 

 ing of the bluestems has been dealt with in ''Plant Succession, " and it 

 will suffice to point out here that this was purely an effect of grazing. 

 Much more important is the question of the effect of grazing on mixed 

 prairie which contains both tall-grasses and short-grasses. The fact 

 that pastures in the mixed-prairie associations were characterized 

 chiefly or wholly by short-grasses was first noted in 1916. Since that 

 time hundreds of observations have been made of the fact that graz- 

 ing reduces or drives out the tall-grasses completely, while it corre- 

 spondingly favors the short-grasses. Stipa is handicapped more than 

 Agropyrum or Andropogon; hence, it disappears first, and consequently 

 is the best indicator of the initial degrees of overgrazing. Bulhilis has 

 the advantage of Bouteloua, by virtue of its much denser mat and its 

 stolons, so that it is gradually increasing wherever present. A particu- 

 lar search has been made for tall-grasses in supposedly short-grass 

 regions during this spring. The results seem to leave no doubt that 

 there is no original short-grass north of the thirty-ninth parallel at 

 least, and that the extensive short-grass areas in this region are all 

 due to overgrazing, chiefly during the period when eastern Colorado, 

 for example, was a great lane for Texas cattle driven to Montana for 

 summer range. The short-grass of northeastern New Mexico is also 

 derived from mixed prairie, but whether this is true of the Staked 

 Plains and of northern Arizona is not yet determined. 



Where grassland touches sagebrush or scrub, grazing alone favors the 

 shrubs, though if fire enters regularly this advantage is more than off- 

 set, and the grasses encroach upon the scrub. All around the northern 

 and eastern edges of the great sagebrush mass, grazing has reduced or 

 destroyed the hold of the grasses and permitted the steady advance of 

 sagebrush. This is particularly true in Oregon and Idaho, where ex- 

 tensive areas appear to be climax sagebrush, and their true character 

 is only to be gained by a study of the grass relicts (Merriam, Univ. 

 Cal. Chron., 1899). Along the eastern edge of the sagebrush associa- 

 tion in Wyoming and Montana, the greater rainfall has enabled the 

 mixed prairie to hold its own in a large measure and to adopt the low 

 sagebrushes into the community as shrubby societies. Thus, the broad 

 ecotone of prairie and sagebrush in central Wyoming is really produced 



