SUMMARY 311 



(3) The surface of the underlying rocks is weathered and is locally 

 stream- worn. 



(4) The distribution of the Windrow formation is discontinuous and 

 there is little to suggest that the present isolated patches were ever parts 

 of a continuous sheet. 



(5) The bedding is poorly defined, the assortment is extremely poor, 

 the quartz pebbles are well rounded, and the chert pebbles are poorly 

 rounded. All the characteristics of the formation show that it was de- 

 posited in streams of considerable velocity. The iron oxides were depos- 

 ited contemporaneously in bogs along the stream courses. 



( 6 ) Judging from the size of the pebbles, the deposits were made under 

 conditions of some relief. No evidence has been observed which necessi- 

 tates connecting them with a period of peneplanation. 



(7) Since the origin of the Windrow formation the region of their 

 occurrence has experienced great erosion; divides have migrated and the 

 streams entrenched themselves to a maximum depth of nearly 900 feet; 

 hundreds of feet of strata have been stripped off from areas of thousands 

 of square miles. 



(8) The Windrow formation closely resembles the Cretaceous con- 

 glomerates of western Iowa. The stream courses in which it was depos- 

 ited appear to have led to the Cretaceous sea to the west. 



(9) The chert pebbles were derived from Ordovician, Silurian, and 

 possibly Devonian strata, which must have once had a much wider dis- 

 tribution than at present. The quartz pebbles were probably derived 

 from the veins of the pre-Cambrian. 



(10) All lines of available evidence connect the Windrow formation 

 with the Cretaceous period.* 



^ Since the above paper was sent to press an article entitled "The erosional history of 

 the driftless area," by Dr. A. C. Trowbridge, has appeared as number 3 of volume ix, 

 University of Iowa Studies — Studies in Natural History. In this paper the gravels of 

 the Windrow formation are discussed under the title of upland fluvial deposits (pages 

 78, 79, 111-113, and 121-123). Trowbridge concludes that they are the product of stream 

 deposition on a peneplain. He places their age as probably late Tertiary (page 123). 

 This conclusion is based mainly on the fact that he finds chalky particles in the Cre- 

 taceous gravels of Minnesota which are absent in the gravels of the Driftless Area, which 

 latter "are strikingly similar to the Tertiary gravels of the Gulf region" (pages 122, 

 123). He also concludes that the surface on which the Windrow formation rests, when 

 projected, fits better with the surface beneath the Tertiary than with that beneath the 

 Cretaceous (page 122). This correlation makes the age of the valleys of the Driftless 

 Area mainly post-Nebraskan — a conclusion which the author thinks is backed by the 

 absence of ice erosion and stream diversions within the valleys of northeastern Iowa 

 where Nebraskan drift is present on the divides (pages 124, 125), 



There is, however, no direct evidence to show how much older the gravels are than 

 the valleys. If the latter are in large part post-Nebraskan, it requires either a longer 

 duration of the Pleistocene than is currently accepted or a greater rate of erosion in the 

 early part of that period than in the later part, or both. 



