336 NATURAL HISTORY BULLETIN 



climatic conditions which are impossible in view of the fossil 

 fauna of the loess, and the vegetation which was necessary to 

 maintain it. To show the possibility of the existence of plants, 

 etc., even in close proximity to perennial ice, attention has 

 been called to the fact that in alpihe regions trees and other 

 plants sometimes grow on masses of earth which were carried 

 over glaciers by land slides. The conditions in the glacier- 

 covered Mississippi valley, however, must have been entirely 

 different. If the glacial mass was so thin and in a climate so 

 warm that it melted away during each season to expose the 

 necessary land surfaces, then there could have been no steady 

 advance of the ice-mass by which enormous quantities of ma- 

 terial were carried hundreds of miles from the northern ledges 

 of rock to which they can be traced. If on the other hand 

 the greater part of the ice-mass persisted year after year, the 

 climate must have been such that snails of the species which 

 we find in the loess, and the plants which they required for 

 food, could not have existed. Glaciers in mountainous regions 

 are not to be compared with such a mass, and do not neces- 

 sarily indicate a cold climate. A difference of a few feet in 

 altitude, or the protection offered by a sheltered valley or ra- 

 vine, may be sufficient to preserve a small glacier even in a 

 climate in which plants may grow abundantly. The writer 

 has seen a profuse mass of summer flowering plants growing 

 against the side of an ice-house within two feet of a great mass 

 of ice! But imagine plants and snails growing in such a region 

 as is pictured by McGee in pi. LX, p. 575 of his great report! 

 The third assumption is untenable. It would be remark- 

 able indeed if over the greater part of the loess-covered region 

 the assumed loess-silt originally filling the valleys should have 

 been removed with such nice exactness that a practically uni- 

 form thickness remains, covering the irregularities of the drift 

 surface beneath, on sides and tops of slopes, even on opposite 

 sides of the same ridge, and hence in different drainage areas ! 

 Moreover, loess frequently shows a lamination parallel with 

 the present surface, and hence not often horizontal. This is 

 especially noticeable in northeastern Iowa, where the loess- 



