170 FULLER AXD CLAPP — MARL-LOESS OF LOWER WABASH VALLEY 



the 500-foot level in the same region, no loess flats being found at alti- 

 tudes higher than 500 feet above tide. The tj^pe of accumulation on the 

 till pkin is not therefore that which would naturally be expected from 

 a gradual and therefore necessaril}^ uniform settling of an exceedingb^ 

 fine dust on an originally undulating plain, or from silts swept along 

 the ground by the winds as in the case of the not ver)^ uncommon loess 

 dunes, but is rather that which characterizes aqueous accumulation. In 

 this connection it may be of interest to note that a single exposure of 

 fossiliferous material of the marl-loess type was observed at the base of 

 the common loess at the edge of the bluff facing the White River flats, 

 3 or 4 miles west of Petersburg, indicating that the base of the silt mantle 

 is almost certainly of aqueous origin. 



Loess-covered aqueous deposits. — During the maximum extension of the 

 Illinoian ice-lobe a glacial lake of some size, known as lake Patdka, 

 occupied the valley and lowlands bordering the present Patoka river, in 

 Dubois and Pike counties, to the southeast of Petersburg. In this lake 

 deposits accumulated to an elevation a trifle less than 500 feet, the ele- 

 vation probably being fixed by one of the divides, of which there are 

 several, to the south or southwest of the lake. These deposits are now 

 covered by loess. Although sufficient well data have not been obtained 

 to determine with certainty the details of the variations in thickness of 

 this loess mantle, the roadside exposures appear to indicate variations 

 of several, if not man)^, feet. The loess surface, leaving out of account 

 the post-Iowan cutting, is as nearlj^ an absolute flat as can be readily 

 imagined, there being considerable areas where not the slightest inequal- 

 ities can be detected by the e3^e. The irregularities in the thickness of 

 the loess represent therefore the presence of depressions in the under- 

 lying surface, due either to original inequalities or to stream erosion in 

 Sangamon times. As in the case of the mantle of the till plain described 

 above, the perfect loess plain is such as would appear to be more char- 

 acteristic of aqueous than of eolian deposition. 



Silt flats. — There are many points both within and without the glacial 

 boundary in the Petersburg and Boonville quadrangles where there are 

 somewhat extensive and approximately level crests standing at the 500- 

 foot level. These, though in the main dependent on the underlying 

 surface, are more regular than the latter and have the appearance of 

 being due to the same agency as those producing the flat surfaces of the 

 silts overlying the till plains and the Lake Patoka deposits. 



Silted divides. — Throughout southwestern Indiana and outside of the 

 marl-loess area there are a considerable number of rather broad and flat 

 divides of loess of the common type at an altitude of 500 feet above 

 tide. These are especially numerous in the region within a radius of 5 



