Frozen St feu nix of Iowa Drift Border. — Wilson. 369 
Whatever causes may be assigned for the immunity of the 
Driftless Area, the absence of ice there was evidently not due 
to a higher temperature in that area. At the time when the 
ice front in the Kansan and Iowan stages moved over the 
channel of the South Maquoketa river, the ice field must have 
stretched for 100 miles or more in a southwesterly direction 
from that part of the margin.* The Illinois lobes were at the 
.same time probably nearly as far south as those in Iowa: so 
that, if we suppose the snow line to have been even forty or 
fifty miles back of the southern limit of the glacier, it would 
be as far south as the region in question. 
Seclusion from the sun's rays was secured at first by the 
depth and narrowness of the river channels ; and as the ice 
front in this part may be assumed to have sloped in a north- 
easterly direction, this itself would aid in excluding the sun's 
heat in the hottest portion of the day. 
This river receives the greater part of its water from springs 
along its own channel and those of its tributaries. Some of 
these springs have an altitude equal to, or greater than, that 
of these relics of water carving. The numerous joints in the 
Niagara limestone allow free penetration of water to consid- 
erable depths; and so the upper limits of the Maquoketa 
shales must have been an horizon of numerous springs in pre- 
glacial days in this border belt, as it is now in the Driftless 
Area; and the filling of the channels to that level, or higher, 
would cause the springs to break out at other points still 
higher. It is worth noting that the erosion towers above men- 
tioned are all in the near vicinity to springs now existing. 
The idea of frozen streams is certainly consistent with that 
of the deep freezing of the soil in the immediate front of the 
ice-sheet, as affirmed by various writers. f 
The question may arise whether the weight and motion of 
the glacial ice above would noi cause a movement in the river 
ice itself that would amount to glacial action in its effects on 
the water carved rocks. To this it may be replied that, it' 
the granular structure in glacial ice can be considered to con- 
*See plates 55 and 56 in Eleventh Annual Report, U. S. Geol. Survej : 
also plate 5 in the American Geologist, Aug., 1895. 
fSee McGee's memoir, Eleventh Annual Report, U. S. Geol. Survey, 
p. 5(17; also an article by Prof. Calvin in the Am. (Jkoukust. Kelt. Isn't;. 
p. 76. 
