ECOLOGY. 349 



a major climatic cycle, they have been able to maintain themselves against 

 the climax forest, aided to some extent to-day by cultivation and fire. A 

 somewhat similar condition exists in the true prairie; but the bunch-grasses, 

 Stipa spartea and Kceleria cristata, have been much reduced in consequence of 

 their habit, early growth in spring, and early flowering, and all the dominants 

 have suffered from the encroachments of the vigorous Andropogon furcatus and 

 Poa pratensis. Wherever the buffalo-grass occurred in scattered mats, graz- 

 ing has caused it to become the chief and often the only pasture grass in the 

 western half of the true prairie. 



Because of its great extent the mixed prairie has undergone several types 

 of modification. The most widespread of these is its conversion to short- 

 grass, consisting of buffalo-grass and grama, or the former alone, and giving 

 rise to a peculiar community long thought to be climax in character. While 

 the short-grass subclimax thus produced covers thousands of square miles, 

 the tall-grasses, Agropyrum, Stipa, and Sporobolus, are rarely entirely lacking 

 in it, and they usually exhibit their proper value whenever grazing or other 

 conditions permit. No section of this climax is without convincing evidence 

 of its tall-grass members, and the short-grass plains should no longer be 

 regarded as a climax, though it will always need to be taken into account as 

 an outstanding modification of the mixed prairie. The latter has also been 

 greatly changed by the entrance of sagebrush, as well as by mesquite. Where 

 the rainfall has been sufficient to compensate, the grasses have held their 

 own, and a savannah containing much sagebrush results. Elsewhere the 

 grasses disappear largely or completely, and the sagebrush takes on the 

 appearance of a real climax, as in northern Utah, southern Idaho, and 

 eastern Oregon, only to yield in turn to the grasses present when these are 

 afforded protection by fencing. The change wrought by mesquite is less 

 significant, as this rarely becomes sufficiently dense to affect the grasses 

 adversely. 



In the warm valleys of southern Arizona and New Mexico the desert- 

 plains grasses have been so completely replaced by desert scrub that the 

 latter has until recently been regarded as the climax. The levels above 3,000 

 feet are mostly mesquite savannah, but below this the scub appears to be in 

 complete possession. The similarity to sagebrush in several respects led 

 to a comprehensive search for relict grasses and other evidences of change. 

 Each year has increased the list of relict species and areas, while the excep- 

 tional grass summer of 1921 yielded convincing evidence as to the structure 

 and extent of the former grassland of the valley at Tucson. The most 

 illuminating discoveries were finding Bouteloua eriopoda in 12 relict areas, 

 though it had been known only near the mountains 30 miles to the north and 

 south, and locating the 7 dominant species of Bouteloua in a foothill canyon, 

 several of them 40 miles away from any known stations. Perhaps the 

 most satisfactory proof of the former grassland has been afforded by the 

 widespread development of Bouteloua rothrocki, Aristida divaricata, and 

 Sporobolus cryptandrus over the Larrea plain during wet summers. Some 

 of the dominants have been traced to the eastern edge of the Colorado 

 Desert in a rainfall of 6 inches, and one or two still persist in sand under a 

 rainfall of 3 inches. 



