298 THWAITES AND TWENHOFEL— THE WINDROW FORMATION 



Ploat gravel is found in the road in southeast 14? southeast 14, section 

 12, township 7, range 5 west, about 12 miles northeast of Prairie du 

 Chien. The pebbles are quartz and local chert; they are well rounded 

 and range iip to an inch in diameter. The elevation, as measured by 

 aneroid barometer, is 975 feet. The bedrock is the Saint Peter sandstone, 

 which contains some chert conglomerate. 



Float pebbles of water-worn quartz and black chert are abundant in the 

 early Pleistocene terrace gravels of Trempealeau and Jackson counties. 

 The chert pebbles carry fossils, of which all so far discovered have been 

 too poorly preserved to be identified. The only place where any quartz 

 pebbles have been found in these counties on the Oneota upland is south 

 of Arcadia, in section 16, township 20, range 9 west, at an elevation of 

 about 1,200 feet. 



The bluff's of Buffalo County are covered by water-worn pebbles of 

 quartz, jasper, chert, and quartzite. Some of the quartz pebbles are sev- 

 eral inches in length and some of the quartzites reach 15 inches in diam- 

 eter. The quartz pebbles are not so well rounded and polished as are 

 those typical of the Windrow formation. The cherts are mainly gray or 

 bright red ; few are black. The jasper is mainly of the bright red oolitic 

 type. The quartzites can be divided into two types, (a) dark red or 

 purple vitreous quartzite, which is unquestionably not local, and (6) 

 much less cemented pink and gray quartzitic sandstone, which may be 

 local. The first type occurs as rounded stones, the second mainly in sub- 

 angular forms which average considerably larger than the water-worn 

 pebbles. The bedrock in this region is the Saint Peter sandstone and the 

 Oneota dolomite. No rock like the pink quartzites has been seen in place 

 in the Saint Peter, but many ledges resemble the gray type. On the other 

 hand, chert pebbles from the Oneota are rare. The elevation ranges from 

 1,200 to 1,250 feet. The correlation of these deposits with the Windrow 

 formation is uncertain. 



It is probable that there are many localities of the Windrow formation 

 in A¥isconsin that have not yet been discovered, since many are probably 

 concealed by the covering of loess. None of the many geologists Avho have 

 examined the region south of the Wisconsin River have reported any 

 quartz pebbles on the hilltops over that region, so that it is probable that 

 none exist. 



There are limonite occurrences in western AVisconsin which do not 

 belong to this forniatio]i. The great majority of the known iron ore de- 

 posits of the Driftless i\.rea described by Strong (25) are clearly altered 

 siilphides and are not at all like the concretionary, manganiferous limon- 

 ite of the ^^lndrow formation. The brown ore deposit at Spring Valley 



