78 NEW YORK STATE MUSEUM 



tion in volume would lead to a great decrease in cutting power, and 

 that the resultant gorge would hence be much narrower and shal- 

 lower than the one cut when the water supply was as large as at 

 present. The Nipissing-Mattawa outlet was finally closed, as we 

 have seen, by the elevation of the land on the north, and the upper 

 lakes assumed their modern outlet by way of Port Huron. As a 

 result the water supply of Niagara was greatly increased, and the 

 broad and deep gorge, which extends from south of the railway 

 bridges to the present falls, was cut by a cataract of the size of the 

 present Horseshoe falls, which in addition carried the water now 

 passing over the American falls. This correlation between change 

 in drainage of the upper lakes and change in size of the gorge of 

 Niagara is certainly very suggestive, and seems admirably to ac- 

 count for many features observed in the gorge. For example, it 

 explains satisfactorily the sudden widening of the gorge just before 

 reaching the whirlpool, forming what Taylor has called the Eddy 

 basin, from the strong eddy which characterizes this portion of the 

 river. This wider part of the gorge Taylor believes was formed by 

 the same large-volume river which cut out the broad channel north 

 of the whirlpool, and he farther thinks, that the sudden change from 

 this broad channel to the narrow one of the whirlpool rapids marks 

 the reduction in volume of water on the opening of the Nipissing- 

 Mattawa channel, which had hitherto been blocked by the remnant 

 of the Laurentian glacier. There are however several features 

 which must be satisfactorily explained before this theory (which 

 Upham rejects on grounds already stated) can be accepted. It is 

 highly probable that the gorge of St Davids was worn back beyond 

 the whirlpool. From the great depth of the whirlpool basin, and 

 the presence of the quartzose sandstone bed at the inlet to it, it 

 seems probable that a fall existed here in the ancient stream which 

 carved the St Davids channel. That channel has probably a depth 

 similar to or greater than that of the part now constituting the 

 whirlpool basin. Now, if, as we have reason to believe, this old 

 channel was formed by an obsequent stream of moderate volum 1 

 flowing northward to the Ontario lowland, it can hardly be assumed 

 that there was but one continuous fall of from four hundred to five 



