84 PLEISTOCENE OF INDIANA AND MICHIGAN. 



The esker consists chiefly of gravel, but it has some till-covered slopes and seems to be 

 made up largely of till for a short distance north of Warford Fork, though maintaining the 

 form presented by the gravelly portion. The drift under the esker as exposed by Clear Creek 

 and Warford Fork is a clayey till. 



ST. PAUL-HARTSVILLE BOWLDER BELT. 



Bowlders are distributed in small numbers throughout the border district and are especially 

 numerous in certain well-defined narrow belts. Of these belts one of the best defined in south- 

 eastern Shelby and the adjacent parts of Decatur and Bartholomew counties is known as the 

 St. Paul-Hartsville bowlder belt, since it runs from one of these villages to the other and extends 

 but little beyond either village. It is about 15 miles long and 2 or 3 miles wide and has some- 

 what definite limits, there being several times as many bowlders for a given area within it as 

 are found on the bordering districts. At its northern end this bowlder belt is closely associ- 

 ated with the undulating till tract leading southwest from St. Paul to Hope, but within a short 

 distance it turns southward into the plain and ends several miles away from the undulating 

 tract. It is possible that both the bowlder belt and the undulating tract mark positions held 

 by the ice margin, but if they do the margin must have remained stationary at St. Paul while 

 it withdrew several miles from Hartsville to Hope. This appears somewhat unnatural and 

 leaves the interpretation of the relation of the bowlder belt to the undulating tract and of each 

 to the ice margin unsettled. In this bowlder belt, as well as elsewhere in the border region, the 

 bowlders are mainly granite and are from a foot or less up to several feet in diameter. 



Bowlders are very numerous in a small tract in Nineveh Township in southern Johnson 

 County. They are especially numerous in sees. 8, 9, 15, and 16, on the brow of a sandstone 

 hill which stands 75 to 100 feet or more above the plane tracts north and east, but they are also 

 common on these plane tracts. 



In the western part of the border district bowlders are conspicuous at only a few places. 

 Perhaps the most conspicuous occurrence is in the vicinity of Groveland and near Marysville, 

 hi Putnam County, where they abound over an area of several square miles. They are not 

 scarce in any part of the till plain in northeastern Putnam, western Hendricks, and southern 

 Montgomery counties, and generally are plentiful enough to supply the needs of the residents 

 for building in localities where limestone and sandstone- are not present. 



STRIDE NEAR BORDER. 



Stria? that seem referable to the Wisconsin ice invasion have been observed in several 

 places in Indiana inside the limits of the Wisconsin drift within a few miles of the border. They 

 bear toward the Wisconsin border and cross older stria? at large angles. In the district between 

 Kankakee and Wabash rivers four occurrences with westward bearing are referred to the Wis- 

 consin, and others with southward bearing are referred to earlier movements. Stria? that bear 

 southwestward along the Sugar Creek Valley in northern Montgomery County are referred to 

 the latest movement and others that bear southeastward and southward are referred to earlier 

 movements. Near Greencastle several occurrences with southwestward bearing are probably 

 referable to the Wisconsin stage and others with southward bearing seem referable to an earlier 

 stage of glaciation. In one place on the top of Cemetery Hill, south of Greencastle, southward- 

 bearing stria? occur above the level reached by the Wisconsin drift. Southward-bearing stria? 

 noted by Collett in a railway cut north of Greencastle seem likely to be the product of glaciation 

 earlier than the Wisconsin. Stria? a few miles northwest of Greencastle in sec. 30, T. 15 N., 

 It. 5 W., stand at the very border of the Wisconsin drift. They bear about S. 20° W., which 

 is nearly at right angles with the trend of the part of the Wisconsin border in which they occur 

 and is somewhat divergent from the southward-bearing stria? of the earlier invasion found in 

 that region. Some local deflection of the earlier movement may, however, have given them, 

 this bearing. Stria? in the bluff of Clifty Creek, west of Hartsville, Bartholomew County, are 

 directed southeastward toward the Wisconsin border and seem likely to have been formed at 



