THE DINFT AND GEOLOGIC TIME 741 
F. B. Taylor of from 75,000 to 150,000 years for the recession 
of the Wisconsin ice-sheet from Cincinnati to Mackinac has been 
already alluded to in the early part of this paper. As against 
this apparently large, but possibly not too large estimate (that 
is, admitting the land-ice formation of the Wisconsin drift), it 
is interesting to quote Wright and Upham’s* dictum that ‘the 
late divisions of the glacial period were far shorter than its 
Kansan, Aftonian and Iowan stages,” and the estimate of Cham- 
berlin? that makes the ratio of time to the present from the 
earliest Wisconsin and from the Kansan stages as 2% and 15 
respectively, leaving an undetermined figure for the still earlier 
portion of the glacial period. That would make, according to 
Taylor’s figuring of 150,000 to 300,000 years for the Wisconsin 
invasion and retreat, a period of somewhere between 900,000 
and 1,800,000 years back to the beginning of the Kansan drift. 
These figures are, it is true, rather staggering, but it is not abso- 
lutely necessary to accept the two estimates and combine them. 
There may be other ways of reckoning the duration of the sepa- 
rate stages of the drift. Certain it is at least that neither of these 
authors is to be held responsible for the estimates of the other, 
or the combination of the two. 
In conclusion, the reasoning of this article may be summar- 
ized as follows: 
The estimates of the duration of the glacial period by promi- 
nent geologists vary almost as widely as possible. It is probably 
useless to attempt to obtain any approximate estimate of its 
maximum duration, but we have in the transportation of erratics 
a simple method by which an ultimate minimum of the time 
occupied may be obtained. Accepting the land-ice hypothesis 
of the deposition of the till, we must from all analogies, and all 
our knowledge of glaciers and ice-caps, admit that the motion of 
the ice-sheet was slow, and that it probably did not exceed a few 
inches a day ; indeed, apart from the evidence of the till and its 
contained erratics, it is hard to find any grounds for belief in its 
motion over large proportions of the occupied territory. Erratics 
t Loe. cit., p. 360. 2 Voc. cit. 
