92 



capitata is the most abundant perennial. In one blowout the orig- 

 inal bunch-grass association was being undermined : Viola pedata 

 and whole mats of Selaginella riipcstris were not only sliding down 

 the slope, but persisting at the bottom. Viola pedata holds a small 

 dune at its base until the erosion gets below the level of its roots, 

 when the whole miniature dune slides down with the plant. 



At the bottom of the slope the plants are usually undermined 

 completely and their dead remains are blown away. But if the erosion 

 of the basin is slow they may persist. One blowout (PI. XI, Fig. 

 2) in the Oc]uawka area showed a semicircle of established bunches 

 of Panicimi pseiidopiibescens and Leptoloiim cognatiun at the base 

 of the slope, and in their shelter numerous annuals were beginning 

 to colonize. 



The vegetation of the windward slope is very open, with at least 

 90 per cent, of the sand exposed. The individual bunches stand at 

 a much greater interval than in the association above the slope, be- 

 cause only a portion of them survive and the gradual settling tends 

 to separate the remainder. 



THE BLOWSAND ASSOCIATION 



The lee slope of the blowout, unless the rate of movement is 

 unusually rapid, is occupied by a variable group of annual plants, 

 most of which live also as interstitials in the bunch-grass and Fani- 

 cum pseitdopiibescens associations. As has been mentioned, the lee 

 slope is an area characterized neither by erosion nor deposition, but 

 by the mere movement of sand. Most of this movement affects 

 only the surface, or extends to but a slight depth. Every autumn 

 and winter countless seeds are blown across the blowouts. There 

 is virtually no chance of their being covered to the requisite depth 

 on the windward slope or in the basin, because there erosion is ac- 

 tive. Consequently both of these associations are almost entirely 

 without annuals. But on the lee slope, where the upper layers of 

 sand are almost always drifting, there is a good chance that some of 

 the seeds will be left covered to a depth of an inch (2.5 cm.) or 

 more. This seems to be the minimum depth at which germination 

 takes place, and marks the upper limit of moist sand during the 

 rainy season in June. It is quite probable that this minimum fluc- 

 tuates with the amount of rainfall, and may be much deeper in drier 

 years (cf. Britton, ipo^: 577). 



In late spring and early summer the seedlings appear, and the 

 frequent presence of thousands of dead stems of Aristida tuberculosa 



