544 J. E. TODD PLEISTOCENE PROBLEMS IN MISSOURI. 



lakes, the waters from the west and north drained. A ridge possibly^ 

 separated it from the Gasconade and the present lower Missouri, which 

 we will suppose were then one stream flowing north to the Illinois river 

 near Alton. The conception that a body of water depositing the loess over 

 Missouri was continuous with the gulf of Mexico or even with a similar 

 body to itself in southern Illinois seems to be positively forbidden by the 

 marked differences in altitude of loamy deposits in the different areas, as 

 has been already stated. The lake drained probably northward, perhaps 

 into the Des Moines, which may have found its way across in an east- 

 erly direction to the Illinois river. The high land, which seems to be, a 

 continuation of this ridge in Pike county, Missouri, extends between the 

 Mississippi and Illinois as far north as Adams county, Illinois, if not 

 farther, and in Calhoun and Pike counties. It is reported by Professor 

 Salisbury to be driftless.f In this lake the bowldery drift may have 

 been deposited by the transportation of floating bergs and river ice de- 

 rived from the many active glaciers farther north. As the outlet was 

 dammed by the Illinois lobe of ice or by northward depression of the 

 region, or by both, it became filled with loess from the glaciers which 

 may have ceased to flow and therefore contributed little coarse material. 

 We may also suppose that some of the loess was derived from the abun- 

 dant Tertiary deposits of silt on the western plains. The upper Missis- 

 sippi, before this flowing through the buried channel, reported by Mr Lev- 

 erett as extending from the mouth of the Wapsipinnicon to the Illinois at 

 Hennepin, was also dammed and emptied its waters for a time into this 

 same lake. About this time w^e may suppose there was an overflow 

 down the present course of the Mississippi above the mouth of the Illi- 

 nois, which before may have been that of a smaller stream. The ice 

 having greatly diminished and a northward elevation having begun, and 

 the lake bed having become filled, the western streams naturally found 

 their way along its southern border and so over the ridge bounding it 

 on the southeast into the Gasconade. Being still swollen by waters 

 from northern glaciers, they rapidly cut down the Missouri to its present 

 slope. 



*The evidence that this ridge was complete at that time is considerably short of demonstration. 

 Along the line of the Wabash railroad which runs on the summit of the divide between the Cuivre 

 and Dardenne rivers, there is quite a regular slope from Warrenton, 853 feet above tide, to Gil- 

 more, where at an altitude of 600 feet above tide a thin sprinliling of northern crystalline pebbles 

 are found under 10 or 12 feet of gray loamy clay as elsewhere. How this could have been formed 

 is as puzzling a question as any presented by these problematic formations. The slope to this 

 point from Warrenton averages 16 feet to the mile, and from Tuque 25 feet. Either the rim of the 

 plain was incomplete here from the beginning of the Pleistocene, or in some way, not ea:^ily con- 

 ceived, it was broken by erosion so early that the drift and loess both overflowed, possiioly in a 

 semi-fluid condition, or there may have been a recent tilting of this immediate region to the east. 



fProc. Am. Assoc. Adv. Sci., Washington meeting, 1891, p. 251. 



