Correspondence. 129 
made, which would place that retreat within 12,000 years — nor would 
so long a period be assigned to the retreat from the upper part of this 
valley, in Lat. 48 N. Again, 5000 years would be far too long a per- 
iod to assign as the length of post-glacial time in large areas in British 
Columbia, whilst in Alaska, it is manifest that the glaciers of the St. 
Elias group covered areas notably greater when America was discov- 
ered, than they cover now. It is evident also, that in spite of short 
periods of temporary advance, the glaciers of the northern hemisphere 
are yet gradually retreating, and that the Ice age has not yet departed 
from that hemisphere. 
It is manifestly not logical nor correct, to take an average of the 
various estimates of the length of post-glacial time at these various 
latitudes as representing anything except the time at which glacial re- 
treat may have passed an average latitude, somewhere between the ex- 
treme ranges of the stages taken into consideration. This average time 
period could not be the time which has elapsed since the Kansan stage, 
any more than it could be that which has elapsed since the "Sitkan" 
stage, in which region active retreat is yet in progress. But the aver- 
age between an estimate of post-glacial time since the Kansan stage 
and the present, would give a period applicable to some latitude between 
38 N., and 58 N. 
The average of all fairly reliable estimates of the length of post- 
glacial time at various latitudes, is applicable not to the time since the 
end of an age not yet passed, but to the time since the line of retreat 
passed some intermediate region between the extreme stages. If sev- 
eral geologists made separate and different estimates of the time since 
the Kansan stage, an average of these estimates might be nearer the 
truth than either of the extremes. But such average could not properly 
be applied to the length of time which has elapsed since glacial ice 
receded from the site of Yankton, S. D., or from the Canadian rockies. 
To give a still farther range to the progressiveness of glacial retreat, 
the undersigned cites the evidences of ice action in lower California, 
in Lat. 23 ° to 24 N. ; around Puget sound in Lat. 48 to 50 ? N., and 
at the head of Lynn canal, in Lat. 59 N. The retreat of glacial condi- 
tions from the peninsula of Lower California may have been 50,000 
or more years ago; from the Puget Sound region, 10,000; and from the 
head of Lynn canal, 1500. The average of these periods, assuming 
them to be approximately correct, is 30.750 years, which may repre- 
sent the lapse of time since some of the Sierra glaciers extended to the 
base of that range. — But it certainly dues not represent the probable 
length of time since the Ice age. Glacial conditions which existed at 
the culmination of that age, have been, and are yet being progressively 
removed, and this removal has recorded and is recording its progres- 
siveness in both hemispheres toward polar regions and from sea level 
upward. 
The rate of retreat is unquestionably slower as the limit is ap- 
proached — but at all latitudes and altitudes at which glacial ice yet 
rests, this retreat, although slightly fluctuating, is yet in progress. 
