266 
and gentler slopes and is invaded near the east end by a flat spur project- 
ing like a dam from the north side nearly to the bluff on the south. Near 
the south end of the dam the stream has cut a narrow gap thirty feet deep. 
The lower ten feet exposed in the gap is shale with a thin cap of flinty 
limestone. The upper twenty feet of the dam is glacial clay with the ex- 
ception of a little poorly stratified gravel and sand at the south end of the 
dam near the creek. These features present on their face the charac- 
teristics of a drift-dammed lake drained by the down cutting of its outlet. 
The supposed lake bottom is underlaid, so far as discoverable, by six feet 
of alluvium on top of the flinty limestone. The flat-topped dam appears 
to be a terrace and is accordant in level with other terrace fragments in 
the valley of West Little Sugar creek above. In the south bluff the 
boulder clay is interrupted for a few rods by ten feet of finely laminated 
lacustrine silts, the bottom of which is at about the same level as the ter- 
race tops. 
Various hypotheses may be entertained; (1) The expansion of the 
valley is due to the lateral corrasion and shifting of the preglacial stream 
over the surface of the resistent flinty limestone. The whole valley was 
filled with glacial clay. The interglacial stream had a temporary base 
level fifty feet higher than at present and cleaned out the filling down to 
the terrace level (510 feet above the sea). During this process a tribu- 
tary stream from the south cut a valley out of the boulder clay down to 
the terrace level, which was afterward filled with silt. By a lowering of 
the base level in the Wabash the stream was enabled to cut down the dam 
and to clear out the valley to the present level, draining the lake and leay- , 
ing fragments of the valley filling as terraces. 
(2) The preglacial and interglacial course of the stream was north- 
eastward into the valley of East Little Sugar Creek. A readvance of the 
ice left a till dam across the former course and the portion of the present 
valley through the narrows is entirely postglacial. The complete sequence 
of events is not so clear as could be wished. 
(3 The resistance of the two limestone strata which outcrop along 
the Wabash bluff and over a belt about two miles wide, west of the bluff, 
may account for the southward course of Hast Little Sugar Creek, for 
the narrows of the lower end of the valley and for the single gap in the 
bluff through which the waters of the system escape to the Wabash plain. 
Period of Ravine Cutting.—The valley slopes of the ultimate tribu- 
