506 



SCIENCE 



[N. S. Vol. XXXVII. No. 953 



for our purposes, because they frequently 

 contain fossils and offer by far the best 

 opportunity for the study of Pleistocene 

 terrestrial mollusks. In these deposits ter- 

 restrial forms vastly predominate, and 

 fluviatile forms are wholly wanting. So 

 much has been written on this feature of 

 the subject that only reiteration is here 

 possible. 



Fresh-water shells in the loess are very 

 few. They belong to species which inhabit 

 small ponds and boggy places. They are 

 not of the types found in streams and 

 lakes. They are local in distribution, and 

 in a number of cases clearly associated 

 with buried ponds. Ponds are not rare in 

 high places in loess regions. They fre- 

 quently contain the smaller Lymncsa, etc., 

 which are sometimes found in the loess, 

 aquatic birds and insects probably being 

 responsible for their introduction. Such 

 ponds, if buried by subsequent depositions 

 of loess, would present exactly the condi- 

 tions under which aquatic shells are usu- 

 ally found in the loess. The vastly pre- 

 dominating forms are terrestrial — wpland 

 terrestrial at that. Some have become ex- 

 tinct in the loess region, but occur west- 

 ward and southwestward in the drier part 

 of the continent. Such are Pupa mus- 

 corum, P. Mandi, Sphyradimn edentulum 

 var. alticola, Pyramidula shimeMi and 

 Oreohelix iowensis. Others like Succinea 

 grosvenorii and Vallonia gracillicorta are 

 still found in the loess region, but they 

 prefer dry, often open grounds. The land 

 species which prefer wet grounds are con- 

 spicuously absent from the loess. 



The usual type of the fauna of the 

 northerly loess is well illustrated by the fos- 

 sil and modern fauna of King Hill, South 

 St. Joseph, Missouri. This elevation rises 

 225 feet above the Missouri River bottoms 

 and is capped by a thick bed of yellow loess 

 which is quite fossilif erous to the very sum- 



mit. The northeastern slope is covered 

 more or less with low shrubs and stunted 

 trees from twenty to eighty feet below the 

 summit, merging below into a native grove. 

 The loess at the summit yielded nineteen 

 terrestrial species. Of these seven were 

 also found living on the shrub-covered 

 northeast surface in a relatively very dry 

 habitat. With the latter were associated 

 six additional terrestrial species which 

 were not found in the summit loess, but all 

 of which are known from the loess of the 

 Missouri Valley. The nineteen fossil and 

 thirteen modern species are all strictly ter- 

 restrial and the six species which are com- 

 mon to both well characterize the habitat 

 of the entire fauna. A more careful ex- 

 amination of both the loess and modern 

 surface faunas near the summit of King 

 Hill would no doubt reveal a larger num- 

 ber of species common to both. 



The fossil mollusks do not enable us to 

 determine the age of any of the Pleistocene 

 formations. The fossils of the Aftonian 

 are not sufficiently distinct from those of 

 modern alluvium to permit the drawing any 

 conclusion other than that the conditions of 

 deposition were much the same. They do not 

 enable us to distinguish between the loesses, 

 for the fossils of the gray and the yellow 

 loesses are, in larger series, essentially the 

 same. But they give us an excellent meas- 

 ure of the conditions which prevailed at 

 the time of the deposition of the various 

 fossil-bearing Pleistocene strata. 



The fact that several ice sheets advanced 

 into the upper Mississippi Valley has been 

 well established. The advance and retreat 

 of these several ice sheets were in all proba- 

 bility very slow, resulting in a gradual 

 transition to and from a glacial climate. 

 This suggested the desirability of search 

 for evidence of such gradual transition 

 among the mollusks of the several inter- 

 glacial periods. But no such evidence has 



