EXTENT OF IOWA PORTION OF 10 WAN DRIFT. 145 



of thick drift was discovered along - which rock is seldom encountered at 

 less than 80 feet, and in places the drift reaches a thickness of 200 feet or 

 more. This thick belt of drift has a breadth of 3 or 4 miles and at first 

 was conjectured to be the marginal ridge of the Illinoian drift, though it is 

 somewhat broader than the general width of that ridge in southeastern Iowa. 

 It carries a thick capping of loess (20 to 30 feet or more) which greatly 

 obscures the glacial deposits. A few exposures were found, however, which 

 showed a surface reddening and leaching of the till such as is displayed in 

 the Illinoian or Kansan sheets of drift. Such reddening has not been found 

 in loess-covered portions of the Iowan drift in northwestern Illinois, nor, so 

 far as the writer is aware, has it been noted in the Iowan drift of north- 

 eastern Iowa. These observations were made in the district where only the 

 upper till is represented on McGee's map. The writer, therefore, inferred 

 that the upper till here should be considered Illinoian and proceeded west-, 

 ward, expecting to find the belt of thick drift swing southward to connect 

 with the margin of the Illinoian already traced to western Scott County. 

 It was found that the "Goose Lake Channel," described by McGee as an 

 old course of the Mississippi, cuts through this belt immediately south of 

 the village of Goose Lake (see PL XVIII). From the west side of this broad 

 valley the belt continues nearly due west across northern Washington and 

 Welton townships, occupying its usual width of 3 or 4 miles and constituting 

 the divide between the Maquoketa and Wapsipinicon rivers. It there takes 

 a course north of west, passing- between Elwood and Lost Nation and still 

 constituting the water parting between the two drainage systems. Slight 

 incursions into the district north of this belt showed only thin deposits of 

 drift resting on or mingled with residuary clays. The thick drift was traced 

 to the vicinity of Onslow in Jones County, and there the examination was dis- 

 continued, for it had become evident that this belt of drift could not connect 

 with the Illinoian margin in Scott County. In returning eastward it Avas 

 found that the thick drift, with its heavy coating of loess, lies along the 

 north edge of a plain of Iowan till, strewn with the immense bowlders 

 characteristic of that deposit and presenting only thin or patchy develop- 

 ments of the loess. 



At Dewitt Prof. J. A. Udden joined the writer for a couple of days' 

 study, and a trip was made together from Dewitt to Sabula. The large 



JION XXXVIII 10 



