134 



SCIENCE 



[N. S. Vol. XXXVIII. No. 969 



intervening plane of oxidation; but in places, 

 and apparently at lower levels, a line of scat- 

 tered pebbles is sometimes evident. In other 

 places at still lower levels the plane of separa- 

 tion is marked by bowlders and a yellowish 

 oxidized surface of the bowlder-bearing por- 

 tion of the Kansan, the horizon that is so 

 commonly seen in Warren, Madison and Lucas 

 counties, which appearance led Bain to coin 

 the term " f erretto." Here and there the de- 

 posit is replaced by beds of stratified sand 

 revealing places of current action. 



This is the deposit which McGee called the 

 " gumbo " of southern Iowa. Perhaps there is 

 no more important relation brought to light in 

 the entire series of exposures than the relation 

 of this common deposit for this part of the 

 state. It is so free from pebbles, weathers so 

 quickly, and forms a soil so like that formed 

 from loess that it has by some (including 

 myself) been judged to be a modified loess; 

 but these excellent extensive exposures of the 

 deposit in many variations leave no chance to 

 doubt the conclusion that this " gumbo " is 

 not a loess, but is related to the Kansan drift 

 and deposited in the closing stages of the 

 Kansan invasion. 



The writer has thus far looked in vain for 

 evidences of kames and drumlins. He has 

 also in previous years endeavored to trace the 

 boundaries of this same " gumbo " to ascertain 

 whether it thinned out as if in basins, but 

 found it through the upland and dissected by 

 ravines. A main difficulty has been to distin- 

 guish between a low-ground gumbo and an 

 upland gumbo, which were apparently con- 

 nected along the sides of the large ravines. 

 The sides of these new railroad cuts and the 

 various excavations in low ground reveal such 

 mixtute and gradation due to wash and creep, 

 in which stratification due to wash has not 

 persisted, that it now seems necessary to 

 recognize this form of low-ground gumbo as 

 not contemporaneous with the upland gumbo, 

 but largely derived from it. However, gumbo 

 ten to twenty feet above the surface of the 

 river valleys is found banked in against and 

 on the Kansan drift, and apparently identical 

 with the upland gumbo. (Such is the deposit 



at the Siegel Brick and Tile Works at 

 Osceola.) 



In the deep cut east of Sandy ville the de- 

 posits above the bowlder-bearing portion of 

 the Kansan drift are in two portions : a lower 

 portion six feet thick and an upper portion 

 one to two feet thick. The surface of this 

 lower portion contains hemispherical depres- 

 sions three to five feet in diameter filled with 

 clay of the upper portion. It is probable that 

 this irregular surface was due to a slight final 

 movement of the ice before the last of the 

 Kansan ice disappeared. No pebbles are 

 found in the depressions, as might be expected 

 if the depressions were potholes, and the cross 

 sections are too rounded to appear due to 

 stream erosion. The whole appearance sug- 

 gests moulding by overriding ice. 



Hitherto the oxidized portion of the Kansan 

 drift found at a depth of thirty feet from the 

 surface in wells of the upland, seen as the 

 upper level of the " f erretto " at the same dis- 

 tance below the upland on so many hillsides, 

 and marked on others as close to the bottom, 

 of the upland gumbo, was judged to be the 

 oxidized surface of the Kansan plain, so con- 

 spicuous throughout south central Iowa, the 

 gumbo itself being then considered a later de- 

 posit on this plain. Classing this gumbo as 

 related to the Kansan drift rather than to the 

 post-Kansan deposits raises the supposed level 

 of this Kansan ground moraine by an amount 

 equal to the thickness of the " gumbo," twenty 

 to thirty feet, and supplies that much of un- 

 eroded material that in places could well have 

 been surface settlings on the upland of the 

 extensive Kansan plain as the Kansan ice' 

 gradually disappeared; in other places a de- 

 posit in hollows on the surface ; in other places 

 not deposited at all, or eroded since deposition. 



On comparing the evidence revealed in this 

 series of railroad cuts with the description 

 which Professor B. Shimek gives of the 

 " Loveland " found along the Missouri River 

 in the western part of the state, announced in 

 the Bulletin of the Geological Society of 

 America, 1910, in Science, 1910, and very 

 fully described in his " Geology of Harrison 

 and Monona Counties," volume 20, Iowa Geo- 



