216 NATURAL HISTORY BULLETIN 



the limits of the latter determined by the accident of seed-dis- 

 persal. Alexander Winchell (1864) considered the seeds of 

 plants throughout the region preglacial/'^ and that when the gla- 

 cier receded the seeds on higher grounds germinated, and the 

 flats, flooded and covered with sediment, remained barren until 

 the lighter seeds of herbs were introduced, and herbs, especially 

 grasses, took possession of these areas forming prairies. Harvey 

 (1908) noted (p. 86) the difficulty of seed germination on the 

 prairie "either because of a dense sod or a lack of soil moisture", 

 and concluded that ''the question of non-invasion upon the 

 prairie proper is primarily and initially one of pre-occupation 

 and the inability of seedlings successfully to withstand the ex- 

 tremely severe conditions of the first winter's exposure." Un- 

 fortunately for this view many seeds do germinate even on the 

 prairie, broken prairie when allowed to lie fallow usually goes 

 back to the normal prairie condition excepting where the blue- 

 grass invasion has succeeded, and there are in the sections in 

 which prairie predominates many groves in places protected 

 from the dry winds of summer but fully exposed to the cold 

 northwesterly winds of winter. ^Moreover where shelter would 

 favor the densest sod we find groves most frequently. Pre- 

 occupation cannot explain the consistent relative distribution of 

 prairie and forest in the western part of Iowa. 



The accident of seed-dispersal no doubt largely determines the 

 grouping of plants in both forest and prairie, but it cannot ac- 

 count for the complete change which takes place in the flora as 

 we pass from the prairie to the forest. 



The Bison 



The bison, or American buffalo, formerly roamed over the 

 western prairies and plains in countless numbers, and has been 

 regarded as an important cause of the treelessness of the prairie. 

 Aughey (1880) thought that the hardness of the surface, which 

 prevented tree-growth, was due to these millions of buffaloes, 

 and Mayr (1890), Marsh (1898), Gilbert and Brigham (1902) 

 and Channing (1908) expressed the opinion that Indians set fire 



1' On p. 338 he stated: "The drift deposits became the vast granary in 

 which nature preserved her store of seeds through the long rigors of a 

 geological winter ' ', — an utter impossibility ! 



