468 



THE COMMUNITY 



both floor and vegetation, in all stages of 

 humus formation. 



In comparison with forest floors, grass- 

 land floors have less bulky leaf mold Utter, 

 but may have as high or higher humus 

 content in the soil. This is partially a 

 result of the continental climate of grass- 

 lands over the world, especially the typi- 

 cally deficient rainfall in the latter part of 

 the hot season. This climatic limitation not 

 only influences the distribution of grass- 

 land ( Thorn thwaite, 1931; Carpenter, 

 1940a), but indirectly reduces stratification, 

 with the correlated reduction in bulk of 

 vegetation and consequent lowering of the 

 amount of litter. Low litter deposit on 

 the grassland floor, in comparison with 

 forest floors, is also partially a conse- 

 quence of increased erosion potential. Ero- 

 sion has an important eflFect upon the im- 

 mediate prosperity, as well as the long- 

 term stability, of the entire steppe com- 

 munity. If the vegetation is considered in 

 the role of a windbreak, the natural wind 

 erosion potential of grassland is greater than 

 that of forests. Wind-blown humus and 

 soil (Sears, 1935a), taken from one area 

 and carried as dust to another area, de- 

 nude the floor in the former instance and 

 may bury the original floor in the latter 

 case. Such dust storms, whether due to 

 drought or to intensive cultivation of the 

 soil by man, or by both these agencies, 

 seriously affect the grassland plants and 

 animals, including man. Water and water- 

 borne soil and humus are inevitably inte- 

 grated in the erosion complex, both 

 through direct and indirect effects. The 

 water-holding capacity of the soil (chre- 

 sard) of the subterranean stratum below, 

 and the windbreak action of the herba- 

 ceous stratum above, are affected by, and 

 affect, the floor stratum. To the chresard 

 must be added the mat of roots previously 

 alluded to, as well as the activities of bur- 

 rowing animals, as factors affecting the 

 amount of erosion in the grassland com- 

 munity. 



The animal population of the floor stra- 

 tum comprises (1) those tunneling in the 

 subterranean stratum, but obtaining all or 

 a part of their food from the floor; (2) pri- 

 mary floor inhabitants; (3) animals that 

 pass their inactive periods on the floor, but 

 feed upon the vegetation, either ascending 

 the latter (grasshoppers) or by grazing 



(antelope); and (4) animals that have a 

 seasonal change in their inactivity niche. 



In the tall grass prairie of parts of North 

 Dakota (Bailey, 1925; Hanson and Whit- 

 man, 1938) and Manitoba (Bird, 1930) 

 in recent years, the herbivore consuming 

 most grass is neither the large grazers 

 (prongbuck, bison) nor the summer-active 

 burrowing rodents and invertebrates, but 

 is a small rodent, Drummond's vole 

 (Microtiis drummondii). It measures be- 

 tween 110 and 145 mm. in total length, 

 but occurs in such numbers that several 

 runways are encountered over almost every 

 square foot of the prairie. It feeds wholly 

 upon the seeds and tender shoots of the 

 steppe herbaceous stratum, such as wheat 

 grass (Agropyron Richardsonii) , Koeleria, 

 Agrostis, and Stipa. Of these, wheat grass 

 keeps its flower heads throughout the win- 

 ter, giving a characteristic aspect to this 

 particular steppe community, and Drum- 

 mond's vole is active throughout the year, 

 feeding upon the rich herbaceous stratum, 

 but constructing summer nests in the sub- 

 terranean stratum and winter nests of 

 woven grass on the floor stratum. 



These investigations also emphasize the 

 point that composition of a stratum is more 

 easily determined than are the exact eco- 

 logical relationships of the several stratal 

 constituents. For example, the familiar cow 

 feeding in the pasture is in competition 

 with numerous herbivorous grassland in- 

 sects and such mammals as meadow mice. 

 We have just seen that the latter consumed 

 more grass than the large grazers. Wolcott 

 (1937) found that under certain conditions 

 cows ate less economic grasses and clovers 

 than did the grasshoppers in New York 

 pastures: 



"Expressing the data obtained in terms of 

 the weight of the insects themselves made 

 possible direct comparison with what the cows 

 were obtaining from the pastures. Surprisingly 

 enough, it was found that where there were 

 few cows in the pasture, they scarcely equalled 

 in weight the total of the wild life present 

 there, and the insects ate more of the grasses 

 and clovers than the cows did. Indeed, the 

 cows obtained a larger share of the pasturage 

 only where they kept the vegetation so short 

 that it afforded scantv protection for the 

 crickets, grasshoppers and leafhoppers, and was 

 more attractive to the robins, who foraged 

 there in greater numbers, and still further re- 

 duced the number of insects" (Wolcott, 1937, 

 p. 89). 



