18 The America n Geologist January, 1896 
Gilbert, Wright, Spencer, and Taylor have supposed, receives only 
slight exposition, the opinion being reaffirmed as if under no serious 
doubt. To this debatable element in the history of the Laurentian 
lakes ami Niagara river, Prof. Spencer attributes much the greater pari 
i>f his estimate of 32,000 years as the duration of the Niagara river and 
falls. 
The earliest publication on Niagara falls b) Mr. Gilberl was in the 
Proceedings of the American Association for 1886, when, according to 
the recent and historical rate of recession of the falls, a conjectural du 
ration of aboul 7,000 years was assigned for the time of erosion of the 
Niagara gorge, beginning as soon as the northward retreat of the Pleis- 
tocene ice-sheet uncovered that area, and including the whole Postgla- 
cial period. Numerous qualifying conditions in the history of the river, 
n it susceptible of satisfactory estimate, were then noticed, some of them 
tending to increase and others to diminish the time thus derived from 
the division of the length of the gorge by the rate of its extension at the 
falls. On the whole, however, it seemed to be implied that 7.000 years 
measure the approximate duration of the river ancLits falls and of the 
whole period since the Ice age; and this was generally published as Mr. 
Gilbert's estimate, both in American and foreign journals, as in Nature, 
vol. xxxiv. 1880. p. 560, and vol. xxxv. 1887, p. 176. 
In a more elaborate discussion of this subject, presented in the Sixth 
Annual Report of the Commissioners of the State Reservation at Niag- 
ara, for the year 1889 'also published in the Annual Report of the Smith 
sonian Institution for 1890), Mr. Gilbert refrained from stating any nu- 
merical estimate of the duration of this period. Out gave much attention 
to the uncertainty introduced by the probable or possible outlet east of 
lake Nipissing. In the paper here reviewed, also, he expresses doubt 
that the investigations yet made can supply any reliable time estimate: 
though evidently the time is geologically very short, since the walls of 
the oldest part of the gorge are still scarcely more affected by weather- 
ing than at the distance of only one mile below the cataract. 
The present reviewer, on the other hand, thinks that good reasons are 
found for a distrust of the withdrawal of the drainage of the upper 
lakes to be tributary to the Ottawa instead of the Niagara river. Ac- 
cording to my studies, the elevated shore lines around these lakes and 
the directions of retreat of the ice imply that the country east and 
southeast from lake Nipissing to lake Ontario and the upper part of the 
river St. Lawrence was still enveloped by the ice sheet as a barrier 
againsl the outflow of the Lake Huron basin, turning it into lake Erie 
and the Niagara river, until the Nipissing and Ottawa region was so far 
raised from its Champlain subsidence that the land itself, so soon as the 
ice blockade was removed, became a natural watershed as to-day. This 
view is given in the American Journal of Scienee for January, 1895, and 
in the Twenty third Annual Report of the Geological Survey of Minne- 
sota. The maps presented with these papers, and in the American Ge- 
ologist for last May and August, show how. in the reviewer's opinion. 
the retreating ice barrier turned all the drainage of the upper Lauren 
