356 W. J. McGce — Geology of the Mississippi Valley. 



comminuted in the latter region, and there is a greater admixture of 

 foreign substances ; and in consequence of this and of its greater 

 tenuity, it is much less compact than the true loess. Still, banks 

 25 feet high and sloping not more than 10° from the vertical, have 

 stood for years with but little change along the Wapsipinnecon 

 river, in Jones county, Iowa ; and steps cut in banks nearly as steep 

 may be used for several consecutive seasons, just as they have been 

 seen to be by the writer in Western Iowa and in Dakota and Nebraska 

 — the loess region of America, par excellence. The greater the thick- 

 ness, too, the more striking is the resemblance to the typical loess. 

 The calcareous concretions characteristic of the loess have not been 

 seen in Northern Iowa and Illinois, however, nor are the tubular 

 rootlet-like pores, lined with carbonate of lime, on which Baron von 

 Eichthofen lays so much stress, very clearly defined. It would seem, 

 though, that this member, wherever found, is a somewhat modified 

 but probably equivalent representative of the loess proper. 



As previously intimated, however, its distribution sometimes 

 seems altogether anomalous. In passing from north to south over 

 the western half of Dubuque county, five or six " divides " are 

 crossed. These seem to be quite independent of the general 

 drainage system, as they all extend in very nearly the same direction 

 (about S. 75° E.), and are in many places cut by the nearly trans- 

 verse water-ways and valleys of erosion. Their width is very 

 irregular, but seldom exceeds a mile or two ; and they are generally 

 much broken and cut up by ravines. Their surfaces consist of the 

 member number one of this paper. Its depth is not great, very 

 rarely exceeding 20 or 80 feet ; and it thins out and finally dis- 

 appears on approaching the lower level of the intervening areas, 

 just as it usually does in approaching the high levels bounding 

 valleys covered by it. These broad intervening valleys may have 

 a width of 4 or 5 miles, and their surfaces are composed of un- 

 modified glacial drift, with exposed boulders. There is no evidence 

 that these areas were ever overspread by the loess-like deposit, and 

 that this was subsequently removed by erosion — indeed, in many 

 instances, the lines of drainage are such as to show conclusively that 

 this could not have been the case. 



The same member is often found, also, covering the summits of 

 kames, which, like the divides, overlook the country for miles 

 around. This has been observed not only in Iowa, but in Illinois, 

 near Clarendon Hills, 20 miles west of Chicago, on nearly the 

 highest land in the State, and in Macon, Christian, Shelby, and 

 Fayette counties, near the southern point of the State, where the 

 surface of the kames graduates directly into the true loess. This 

 will be recurred to later. 



The anomalous fact is that the loess-like member forms the Tapper 

 surface of hills and ridges rising 40 or 50, or even more, feet above 

 the general level, over which the same materials were never deposited. 



Member number two. — The materials composing this member are 

 largely coarse sand, gravel, and well-rounded and water-worn 

 pebbles, with more or less of clay. The pebbles are both local 



