504 



SCIENCE 



[N. 8. Vol. XXXVII. No. 953 



Some of these points of similarity are 

 brought out by a comparison of the mol- 

 lusks from various deposits. Thus the 

 Aftonian beds of Harrison and Monona 

 counties, Iowa, have yielded 28 aquatic 

 species and 9 terrestrial species;- the Don 

 beds, 42 species, of which all but two are 

 aquatic;^ the Florencia formation, 30 

 aquatic and swamp species and 19 terres- 

 trial species ;* the buried silt in the Illinois 

 Central cut in Sioux Falls, 14 aquatic and 

 2 terrestrial species, one of the latter 

 doubtfully belonging to the deposit;^ the 

 more modern river alluvium of Harrison 

 County yielded 18 aquatic and swamp spe- 

 cies and 19 terrestrial species;^ the lacus- 

 trine alluvium of West Lake Okoboji ex- 

 posed by recent canal excavations, yielded 

 27 aquatic and swamp species, and 1 ter- 

 restrial species; modern drift along the 

 Big Sioux River opposite Canton, South 

 Dakota, contained 18 aquatic and 5 terres- 

 trial species ; similar drifted material along 

 the Missouri River at Rulo, Nebraska, 

 showed 12 pond and swamp species, and 

 15 terrestrial species; and drifted material 

 on the north shore of Miller's Bay, West 

 Lake Okoboji, Iowa, contained 14 aquatic 

 and marsh species and 14 terrestrial spe- 

 cies. 



These lists indicate that terrestrial mol- 

 lusks are not uncommon in the various 

 alluvial deposits. It should be remem- 

 bered, however, that as a rule the terres- 

 trial species are represented by very few 

 individuals which accidentally drifted in 

 from adjacent land surfaces, while the 

 aquatic species belonging to the genera 



= See Iowa Geol. Survey, Vol. XX., 1910, pp. 

 395-6. 



= Journal of Geology, Vol. IX., 1901, pp. 291-2. 



* Am. Jour, of Science, 4th series, Vol. IV., 

 1897, p. 96. 



° For description of this out see Bull, of the 

 Geol. Soc. of Am., Vol. 23, 1912, pp. 141-3. 



' Iowa Geol. Survey, ibid. 



Sphcerium, Pisidium, Campeloma, Valvata, 

 Amnicola, the larger species of Lymncea, 

 PJiysa, Planorbis, Segmentina and An- 

 cyl'US oe> 'V in larger numbers. 



In those case in which larger numbers 

 of terrestrial forms occur, as in the Miller's 

 Bay and Rulo lists, they may be traced to 

 nearby wooded bluffs, and it is evident that 

 the relatively large number of terrestrial 

 forms in the Harrison County alluvium 

 and the Florencia formation may be traced 

 to the same source. 



It is evident that the shells do not drift 

 far in any case. The Missouri River con- 

 tains few aquatic mollusks, but its smaller 

 tributaries usually show an abundance of 

 them. Yet rarely are fluviatile shells 

 found in modern river drift along the Mis- 

 souri, and the aquatic species are of the 

 pond type prevailing in small ponds along 

 the main river, as is shown in the Rulo list. 

 Shells of terrestrial species which are lim- 

 ited in their distribution are also seldom 

 found at any distance from their habitat. 

 This is strikingly illustrated by Helicina 

 occulta, which is now restricted to limited 

 and widely separated areas. Thus at Iowa 

 City a colony inhabits a half-acre of 

 wooded bluff, and in thirty years your 

 essayist has found but one fresh shell along 

 the creek at the foot of the slope, and none 

 along the nearby river to which the creek 

 is tributary. 



A recently discovered colony of this spe- 

 cies in a similar situation along the Cedar 

 River above Cedar Rapids, lowa,^ gives 

 like results. Various students of mollusks 

 have collected in this region, yet no fresh 

 shells of this species were ever observed in 

 the drifted material along the Cedar River. 

 The Miller's Bay list shows that 14 species 

 of terrestrial shells were carried across the 

 bay from the wooded southern shore, yet 

 repeated dredging in the bay has brought 



' By Mr. E. G. Grissel, of Cedar Eapids, Iowa. 



