1884.] Vestiges of Glacial Man in Minnesota. 699 
rock affording permanent support to fragments detached. Even — 
_ were it such, however, the quartzes would still have been subject 
= toeolic influences whence we might look to see the successive 
_ products of ages of weathering widely disseminated and occupy- 
_ ing different levels in the neighboring sand. 
: Second: it was a further peculiarity of the quartz layer that it 
did not thin out toward the west, in the direction of the 
fiver, as it should have done had its area been still intact. 
Contrawise, the quartzes extended in full thickness quite up to 
the extreme edge of the terrace-plain, where the deposit termi- 
fated abruptly in an obscuring layer of surface soil derived from 
the drift above. Previous to the inception of grading operations 
upon the notch, by which this ancient cache seems to have been 
nearly and perhaps quite obliterated, it was no infrequent thing, 
heavy rains, to find quartzes, plainly extruded from the 
others, lying along the bottom-land west of their original bed. 
The terrace consists properly of three sections, one above the 
| other, which must have had their beginnings at three different 
_ Periods. First, we have the lowest and oldest, comprising the 
body of till and drift below the quartzes; second, the quartz- 
Yielding stratum itself ; and third, and newest, the accumulations 
of modified drift above the latter. This classification is based on 
existing geological conditions, and formulates the necessary 
= “quence of glacial events. It is, therefore, by no means an ab- 
or an arbitrary division. It is moreover not affected by 
the question whether or not the quartz layer is artificial in char- 
acter. Even were it a proven fact that these fragments are wholly 
onal forms resulting from the wearing down of quartz veins, it 
Nay still be true that the quartzes must, at the time of their 
deposit, have occupied a surface now no longer existing as such, 
: the present plane of the middle section. The truncated 
: Pe of the western edge of the quartz layer, coinciding with 
tk k the terrace-bank above it, points us back to a time when 
mt of these sections had as yet no existence, 
: Pared to lower one, projected west of its present limits, was pre- 
Aka "eceive upon its surface, as in time it actually did danik 
hag now above it. The middle section, that is the 
eek ts es ae 
Ward. Wartzes, must in its turn have had a like west- 
Hi „ Prolongation, The superior section, deposited later still, 
in the same general extension. When the Mis- 
