OHIO STATE ACADKMT OF SCIENCE 



form a temporary lake. But Brodhead's gravel beds prove to be 

 of local origin^ while the first-mentioned drift reaches but little, if 

 any, more than 100. feet above the stream, and the ice-dam did not 

 hold for a period that allowed any channel to be made to the south 

 of it; so another cause must be sought for the loess deposits in the 

 vicinity. 



In and around St. Louis the loess forms a cap, covering nearly 

 all the early formations. While it is thin on hilltops and thicker 

 in valleys by reason of erosion and re-deposition on uneven ground, 

 yet it is singularly regular over many square miles. Eeports of 

 railway cuttings, wells, and other excavations, contain numerous 

 references to "loess 12 or 14 feet thick." It is from 6 to 8 feet 

 thick on a plateau nearly 350 feet high; and is not more than 20 

 feet in most of the county unless near the foot of a slope. This 

 means a depth of water that would submerge hills at the level 

 indicated, and lowering so rapidly in the end as to uncover all the 

 territor}^ within a comparatively short time. 



Worthen says that in Jackson county (Illinois), the loess 

 occupies only a narrow belt on the top of the river bluffs; and in 

 Union county, next south of Jackson it was found at only one 

 point and that below the top of the bluff. 



Shumard notes the presence at Wittenberg, Missouri, of a 

 mass of graoiite weighing several tons ; and thinks this is evidence 

 of a ledge of eruptive rock in the neighborhood. 



It should be stated that Big Muddy separates Jackson from 

 Union county, and that Wittenberg lies opposite to the old mouth ; 

 being about eighty miles south of St. Louis. Between these two 

 points are Eock Creek, twenty miles south of the city, in whose 

 valley the loess (modified by local drift) is one hundred feet high ; 

 and Plattin creek, forty miles south, where it covers a slope at an 

 elevation of eighty feet. 



It seemed plausible to suppose that a prolongation or spur of 

 the glacier might have reached from the Big Muddy to the Mis- 

 souri side, thus choking the Mississippi and allowing the water 

 to stand at a level sufficient to drown most of the country above. 

 Additional color was given to this supposition by the gorge at 

 Grand Tower, just below Wittenberg. Here, the river flows in a 

 narraw, rock-bound channel, over a solid rock bottom, while on the 

 Illinois side is a valley fully three miles wide, of alluvial silt 

 soibject to overflow in great freshets. But the granite proves to be 

 only a boulder, lying in a small ravine a few feet above the river's 

 ordinary level, and it may have come in with an ice floe at any 

 time. And there is not a trace of evidence on either side of the 

 river, that the glacier had even reached the lowland. This was 

 an additional problem, instead of an explanation; for there was 

 St. Louis and Cairo only sand and silt are found along the valley, 

 a feature that apparently indicates a drainage no more vigorous 



