186 THE GLACIAL LAKE AGASSIZ. 



10 to 25 rods and in depth from 10 to 30 feet. The top of the excavation 

 is aliout 20 feet below the crest of the hill. As thus exposed to view, the 

 greater part of this deposit is seen to be gravel, much of which is very 

 coarse, containing pebbles and rock fragments of all sizes up to 1^ feet in 

 diameter, many of the smaller being well rounded, liut the larger mostly 

 angular, with only slight marks of water wearing. In some portions near 

 the west end of this section no interbedding of coarser and finer layers of 

 the torrential esker gravel is noticeable for 10 feet or more vertically, the 

 spaces between the larger stones and cobbles being filled with finer gravel 

 and sand. Embedded in this coarse gravel on the south side of the exca- 

 vation I noted a mass of ordinary till, unstratified bowlder-clay, inclosing 

 gravel and bowlders in a solid matrix of somewhat sandy clay, wholly 

 bounded by definite but irregular outlines, its dimension vertically being 

 about 10 feet and its length 20 feet. No other mass of till, of either small 

 or large size, was observed in this entire section. It probably was derived 

 from the ch'ift that was contained within the ice-sheet and finally overspread 

 its surface when the greater part, of the thickness of the ice was melted. 

 From a sheet of drift thus dejjosited on the ice that formed the bank of the 

 glacial river this mass may have fallen into its channel. The eastern half 

 of the section includes much fine gravel and sand irregularly interbedded, 

 and along a considerable extent there the south side of the excavation, from 

 10 to 20 feet below its top, is clear sand. Paleozoic limestones make up 

 about three-fourths of the gravel, the remainder Ijeing Archean granites, 

 gneiss, and schists. Some two hundred bowlders were found scattered 

 upon the area of the excavation; and they occur with nearly the same fre- 

 quency on other portions of this northern slope of the hill, but are rarely 

 found on its top and southern slope. They vary in size from 2 to 8 or 10 

 feet in length; nearly all are Archean, but a few of Paleozoic limestone, 

 up to 5 feet in length, were observed. None was seen inclosed within the 

 gravel and sand of the esker, and the workmen informed me that they 

 occur only on or near the surface. This hill was covered by Lake Agassiz, 

 and its bowlders were doubtless dropped or stranded from bergs and floes 

 on this lake before the border of the ice-sheet had retreated from the 

 vicinity. Indeed, the occurrence of the bowlders chiefly on the northern 



