292 
PETER SANDSTONE 
down, and it is positively certain that little of the sand could 
have come from that source. Southward, the land-mass south of 
the Ouachita geosyncline was a source of important sediments, 
but these were largely shales, and it is inconceivable that the 
muds rested nearest the land, and the sands moved north across 
the geosyncline. Neither Roubidoux nor Peter sands seem to 
have come from this source. 
To the west, in the Rocky Mountain region, nearly pure lime¬ 
stone was forming in Colorado during Beekmantown time, and it 
does not seem reasonable to look there for the Roubidoux sands. 
Since the Peter sands seem to have come from the same area that 
yielded the earlier sands, the north is the only plausible source. 
The general history of the central Mississippi Valley region 
during Early Ordovicic times is believed to be about as follows: 
Sands from a northern land-mass were delivered by rivers, and 
possibly directly by winds, to the Roubidoux-New Richmond seas, 
which extended north into Minnesota and Wisconsin; and were 
spread out by waves and currents. Limestone and shale-lenses 
were interbedded, and local emergences, with warping, allowed 
sun-cracking. As the land-mass to the north wore lower and 
lower, less and less sand was contributed, and the Jefferson and 
Cotter (Shakopee to the north) dolomites were laid down, with 
only occasional sand-lenses. A retreat of the sea, with the north¬ 
ern land-mass still low and furnishing little sediment, allowed the 
erosion of the surface of the Cotter formation, and a slight re¬ 
advance brought in the Powell materials over the southern half of 
the region without clastic sediments. 
This movement was followed by a still more extensive retreat, 
producing the pre-Peterian unconformity. During this emergence, 
a relief of 200 feet or more was carved out upon the old Cam¬ 
bric surface, with the removal of more material to the north than 
to the south. 
The Peter sands which rest upon this erosion surface could 
not in any manner have been derived from the underlying rocks, 
and it has been shown that they probably came from the north, 
as did the Roubidoux sands. It is not believed possible that these 
sands were brought in as a great series of drifting dunes in an 
extensive interior desert. The rounding and frosting which are 
cited as evidence of this hypothesis are just as well-developed in 
