Review of History of the Great Lakes. — Spencer. 299 
age. If the terrestrial elevation of the land shall continue as 
for the last 1,500 years, the barrier across the outlet of lake 
Erie must rise so high as to turn the drainage of the lakes into 
the Mississippi, by way of Chicago; and it is computed that 
the end of the Niagara river and falls, under such conditions, 
will be about 5,000 years hence. 
All these estimates are based upon the rate of recession of 
the falls and the amount of work done in each episode, as dis- 
covered in working out the history of the lakes. In 1842, 
Prof. James Hall made the first instrumental survey of the 
falls. The next wus made in 1875 by the Coast Survey. In 
1886, Prof. P. S. Woodward made the third; and in 1890 Mr. 
Aug. S. Kibbe made the last. From these four surveys, the 
mean rate of recession of the falls (that is, the mean elonga- 
tion of the gorge) was found to be 4.17 feet a year. But tin- 
river in the region of the falls is now crossing a pre-Pleisto- 
cene valley, where the hard surface rocks have been removed 
for 80 or 90 feet in depth beneath the rocky ridge crossing 
the course of the canyon a short distance below the present site 
of the falls. Thus the amount of work now being done by the 
river is much less than the average demand upon the stream 
during the greater part of the life of the river. Before 1N75 
all statements as to the age of the river were pure conjectures, 
but that of Lyell was nearly correct. The estimates made 
upon the retreat of the falls alone have proved to be not even 
so accurate, although the method was better as far as it went; 
but it stopped short of the history of the falls. Again, spec- 
ulations as to the ancient Niagara flowing down by the Whirl- 
pool-St. David's valley have been disproved by the rock which 
crosses that course hundreds of feet above the lake level; in- 
stead, the Niagara here touches a little buried tributary of an 
ancient stream to the west. 
In conclusion, the Niagara falls serve as a chronometer of 
geological time, as they give some idea of the epoch of the 
lakes. If the Ice age ended witli the birth of Warren water, 
then we can roughly estimate it to date back sonic 50,00(1 or 
60,000 } T ears. At the birth of the Niagara river and falls, and 
long before, there was no ice barrier in the Niagara district. 
Lastly, if we regard the Iroquois water as at any time ob- 
structed by ice, such conditions have not existed since the 
