268 



SCIENCE 



[N. S. Vol. XXXIX. No. 999 



stream as late as just before the Wisconsin 

 stage of the Pleistocene has been clearly 

 traced from a little south of its present 

 junction with the Missouri, northeast past 

 Springfield and Tabor, to the James a few 

 miles northeast of Yankton. There is little 

 doubt that its preglacial course was the 

 same so far. Its altitude was probably 100 

 to 150 feet above the present, i. e., 1,300 to 

 1,350 A. T. It may have been 100 feet 

 higher still. That it belonged to a different 

 system from those hitherto considered is 

 favored by the following facts : (1) there is 

 trace of an old divide crossing the James 

 Eiver south of Mitchell. The Sioux Quartz- 

 ite rises to more than 1,200 A. T. and Cre- 

 taceous rocks overlive to a height of more 

 than 1,300. As this is in the center of the 

 valley it is reasonable to think that 100 

 feet may have been carried away by glacial 

 erosion. (2) There is a line of high ridges 

 lying across the James valley north of the 

 Niobrara, as though they were a remnant 

 of a divide. I refer to the Choteau Creek 

 Hills, James Ridge and Turkey Ridge. 

 (3) The few traces of preglacial surface in . 

 eastern Nebraska and western Iowa seem to 

 call for a lower drainage level than is indi- 

 cated farther north, and also farther south. 

 In short there seems to be need of finding a 

 drainage outlet eastward to the Des Moines, 

 or else to postulate a recently formed syn- 

 cline for which we know no other evidence. 



We conclude therefore that the Niobrara 

 turned south and followed the courses of 

 the James and Missouri to the vicinity of 

 Onawa, Iowa, thence east and northeast 

 through Ida and Sac counties past Wall 

 Lake and thence southeast along the Rac- 

 coon River. 



This conclusion rests on a few apparently 

 reliable reports from wells which show that 

 the preglacial surface indicates a valley 

 whose bottom is less than 900 A. T., in some 

 cases less than 850. It has not been possi- 



ble to outline the valley throughout, but 

 half a dozen observations through Ida and 

 Sac counties are most easily explained by 

 such a theory. The fact that Wall lake 

 lying on the summit formerly drained into 

 Boyer River and now into the Raccoon, 

 and another fact that the Boyer rises east 

 of the crest of the divide, has first a course 

 east of south and at this point turns south- 

 west, are also most easily explained by the 

 theory given. 



There was a descent of about 350 feet 

 from Sioux City to Wall Lake. 



7. Platte River of Nebraska. — This 

 stream presents difficulties which have not 

 yet been satisfactorily solved. The lowest 

 surface overlain with glacial deposits in 

 their original position seems to be about 

 970 feet A. T. in the vicinity of Omaha 

 and Council Bluffs. The lowest striated 

 rock surface is about 1.000 A. T., except a 

 ledge at Omaha reported by Dr. C. A. White, 

 which was about 6 feet above the river, and 

 may have been the result of the action of 

 river ice. It is conceivable that till may 

 have slumped 30 feet into more recent exca- 

 vations without clearly showing the fact. 

 There are numerous evidences of such 

 slumping elsewhere along the trough of the 

 Missouri. 



Now there seems to be quite clear evi- 

 dence, as will be explained in the next sec- 

 tion, that in early Pleistocene there was a 

 divide at least 900 feet A. T. near Leaven- 

 worth and that south of it drainage levels 

 were considerably higher than north of it. 

 There would, therefore, have been insuffi- 

 cient slope in that direction to have exca- 

 vated down to 970 feet A. T. 150 miles 

 away. We seem driven therefore to find 

 some other outlet. 



Could it have been through to the Des 

 Moines by the old channel in Sac county? 

 That would have had a steeper grade than 

 bv Leavenworth. We have not been able 



