The Lcess and its Fossils. 07 



comparatively warm, and the glaciers had alread}' retreated 

 far to the north when the deposition commenced. 



II. During at least a part of the summers a large portion 

 of the area now covered b}^ the Loess was elevated above the 

 surface of water, as indicated by the presence of the predomi- 

 nating land shells. 



III. These shells, too, indicate that the surface was not 

 entirely unlike our present prairies in Iowa and Nebraska, 

 though perhaps more moist, and more nearly level. 



IV. The presence of shells of the genera LimncBa^ Physa, 

 Planorbis, and Pisidium^ which are principally pond species, 

 indicates that over this prairie surface were scattered numer- 

 ous ponds and that it was traversed b}- quiet, sluggish 

 streams.* 



V. The distribution of the shells as well as the homogene- 

 ity and fineness of the material forming the deposit indicate 

 that the deposition was unaccompanied by violent distur- 

 bances, but that it took place quietly, and very slowly during 

 a long period, 



VI. The deposit itself was probably partly formed from 

 sediment carried over portions of the surface b}'^ quiet over- 

 flows of the sluggish streams which had not yet deeply cut 

 their channels." The numerous ponds, however, fed by the 

 drainage in their immediate vicinity, were also receiving with 

 this the finer material gathered from the glacial drift sur- 

 rounding them.^ This material, being like that gathered by 



1 This is all the more plausible if we add the fresh-water shells reported 

 from the Loess of the Missouri bv Prof. Swallow. The additional genera re- 

 ported by him are Valvata, Amnicohi and SpIicBrium^ and these, too, are most 

 commonly found in ponds and sluggish streams. 



2 This idea has already been suggested by James E. Mills in the American 

 Geologist, Vol. III., No. 6, and also by Warren Upham (in the gfh Annual 

 Report of tha Geol. and Nat. His. Survey of Minn) and others, but the extent, 

 duration, and violence of the flood suggested b_v the latter are entirely incom- 

 patible with the facts presented in this paper. 



3 The existence of ponds directly in or upon the sand and pebbles of the 

 glacial drift is not remarkable, or extraordinary. In the " Sand Hills" of Ne- 

 braska hundreds of ponds are now found in the sand, the fine sediment at the 

 bottoms of which alone seems to prevent their draining away through the 

 sand. 



