354 
The Ohi<> Naturalist. 
[Vol. VIII, No. T, 
sions, the following conclusions may be reached. During pre¬ 
glacial time a system of valleys was here developed—valleys 
which had reached maturity. This maturity was more marked 
in and near the master valleys but became less and less strong 
as one pushed back up the little valleys. Many of the latter, 
especially south of the large east and west valley, gradually 
widened northward showing that the line of headwater divides 
must have lain near but south of this large valley. 
The advancing ice would move more easily into the broad, 
mature valleys and into those lying more nearly in the direction 
of ice movement. The ice pushed in from the north, as it came 
into rougher and rougher topography it broke up into tongue¬ 
like dependencies,* which extended into the valleys. While 
the ends of the tongues were fairly stationary, melting freed 
rock waste which accumulated in moraines. Any stream flow¬ 
ing toward the advancing ice would of necessity find its'course 
closed during the ice advance by ice and moraine. The "water, 
augmented by the melting of the ice, would accumulate in a 
lake between the advancing ice on the north, the valley walls 
on the sides and the divides on the south. Its outlet would be 
over the lowest places, whether of moraine or rock, whether in 
valleys or over low cols between ridges at the very heads of the 
streams. Thus was begun the discharge of water over the 
divides. As the ice continued to advance it crossed one of the 
divides and may have lowered it some by ice erosion, but this 
if done, is not very apparent. 
While the ice front lay along the large east and west valley 
studied, it built masses of moraine in the valley and water rose 
in the tributary valley leading southward from Loudonville 
until it was pushed over the divide at Spellacy. When the ice 
withdrew, it became more markedly a series of valley depend¬ 
encies, and the water accumulated in front of their noses until it 
went over the col in Muddy Fork northwest of Lakeville, but 
because of ice obstructions it could not go along the large valley 
leading to Big Prairie. Thus was started the stream over the 
divides between the narrow valleys and not along the earlier 
mature valleys. 
Once started it was easier for the stream, when the ice with¬ 
drew, to maintain its course over the divides than to seek its 
old routes. Hence the cutting down continued and the little 
narrows were finally cut to present dimensions. The cutting 
down of the divide gave all side streams near the divide a chance 
to deepen their courses near their mouths which the}’ proceeded 
*Camev, F. (Jour. Geol. XV, 1907 pp. 488 ff.) very happily applies 
this term to peripheral protuberances, which extended from the great 
ice sheet into preglacial rock valleys. 
