ECADS. 83 



sand on the side of a blowout. Fibrous roots, a millimeter or less in diameter, 

 arose in clusters from the rather short, shallow rootstocks. Some penetrated 

 rather vertically downward to maximum depths of 3.5 feet, and in fact the 

 roots were abundant to a depth of 3.3 feet. Many others spread out laterally 

 at all angles between the vertical and horizontal, some reaching distances 

 of from 1 to 2 feet from the base of the plant and ending in the surface 2 or 3 

 inches of soil. All parts of the roots are furnished with multitudes of very fine 

 absorbing laterals. While many of these are only a few millimeters long, 

 others have a length of 6 to more than 12 inches. All are profusely and mi- 

 nutely branched, the larger ones to the third and fourth order. Even the tips 

 are well supplied with branches. In a neighboring blowout other specimens 

 were not so deep-seated. Roots were not abundant below 2.4 feet and the 

 maximum penetration did not exceed 3.2 feet. 



Other plants were examined at Haigler. One clump was excavated on the 

 high, dry, rather vertical walled rim of a blowout. The bank was cut back 

 about 2.5 feet. Some of the roots were found at a maximum depth of 4 feet 

 and they were numerous at 3.3 feet. On the other hand, specimens examined 

 in the grassy carpet of a fairly stable old blowout, only a few rods distant, 

 were found to be shallower-rooted. None extended beyond 2.8 feet. The 

 soil in the former situation was considerably drier, and this may account for 

 the difference in root penetration. 



Further examination of this species at Seneca confirms its shallow, widely 

 spreading, profusely branching root habit. Here well-developed clumps 

 were excavated in the loose sand of an old blowout. Practically all of the roots 

 were found in the surface 1.8 feet of soil, only a few penetrating to a depth of 

 2.5 feet. The root habit of this species is fairly stable. 



Aristida purpurea. — Wire-grass characterizes large areas of grassland 

 throughout the Great Plains, especially where the soil is slightly sandy or 

 where overgrazing or other disturbance has given it a foothold in competition 

 with Bulbilis and Bouteloua gracilis. It is of very little grazing value, being 

 eaten by stock only when other forage is scarce. 



A complete description, accompanied by a drawing of this species as it 

 occurs in the hard loam soil at Colorado Springs, has been given in "Ecological 

 Relations of Roots," pages 46 and 47. Wire-grass has coarse, fibrous roots 

 which spread 5 to 8 inches laterally, but do not branch profusely in the surface 

 4 inches of soil. The soil is filled with well-branched roots to a depth of 3 feet, 

 at which level many terminate in much-branched tips, while others penetrate 

 to a maximum depth of 4 feet or slightly beyond. 



In the sandy loam at Limon the roots reached a depth of 4.2 feet and the 

 root habit agreed in all respects with that at Colorado Springs, except that the 

 roots in the surface layer of soil were somewhat better developed. 



At Ardmore a group of well-matured plants about 1.1 feet tall was excavated. 

 The Pierre clay soil was exceedingly hard and tenacious and underlaid at 4 

 feet with a layer of gravel and sand so closely cemented that it was picked away 

 with considerable difficulty. Here the wire-grass roots reached a working 

 depth of only 2.6 feet, although some penetrated 6 inches deeper. 



Further examinations were made at Burlington and at Sterling, Colorado. 

 At both stations only isolated groups of plants were found in the short-grass 

 turf and the tops were much dwarfed, being only about 6 inches high. At 

 Burlington the "hardpan" began at 1.9 feet and extended to a depth of 3.5 feet, 

 below which the soil was mellow but powdery dry. In the soil of short-grass 

 plains the "hardpan" is usually clearly demarked by its lighter color, due 

 apparently to an accumulation of calcium and magnesium compounds. Here 



