OCTOBEB 16, 1908] 



SCIENCE 



529 



directions, and continued northward. The 

 Dale would thus turn an angle of over 135 

 degrees, coming from the northwest, and 

 turning north. Truly a remarkable course 

 for a well-adjusted stream, though not an im- 

 possible one. On the hypothesis of a south- 

 ward drainage, the Dale from the northwest 

 and the Oatka from the north, united opposite 

 Warsaw, and continued southeastward after 

 the approved manner of normal streams. On 

 this hypothesis we must look for a southeast- 

 ward continuation of the Warsaw River 

 formed by the junction of the Dale and Oatka. 

 For ten miles the country is drift-covered, 

 then we come to the Glen Iris Valley, which 

 is exactly in line with the course of the War- 

 saw, and which, as we have seen, requires a 

 northwestward continuation. This then is 

 furnished after an interval by the Warsaw 

 Valley, while conversely the Warsaw Valley 

 finds its southeastward continuation in the 

 Glen Iris Valley. Each finds its complement 

 in the other, the requirements of each are 

 fulfilled by the other. The only unknown 

 quantity is the drift-filled interval between 

 Glen Iris and Warsaw. Thus a southward 

 drainage for the Preglacial Warsaw and its 

 two tributaries, the Dale and the Oatka, and 

 their junction with the Genesee at Portage- 

 ville, explains all the known phenomena, and 

 there are none known which oppose such an 

 explanation. On the hypothesis of a north- 

 ward flow of the Warsaw, many known facts 

 are unexplained, many are opposed to it, and 

 the drainage takes on a most complicated 

 manner, instead of the simple direct course of 

 a southward-draining system. 



One of the chief objections to the hypothesis 

 of southward drainage of the Warsaw and its 

 tributaries is the greater depth of the Warsaw 

 over the Glen Iris Valley. - This difference 

 finds a very simple explanation in the 

 hypothesis of glacial deepening of the Warsaw 

 Valley. Is there any evidence of such deepen- 

 ing ? There is the most direct evidence. The 

 sides of the Warsaw Valley are much steeper 

 than those of ancient valleys not deepened. 

 At the aforementioned bend of the valley 

 opposite Warsaw, the west or concave bank 

 is almost precipitous. It is here that a valley 



glacier would produce the greatest erosion. 

 The drift in the southern part of the valley 

 and between it and Glen Iris, is local, com- 

 posed of the material gouged out of the valley. 

 But the most conclusive evidence of glacial 

 deepening of the Warsaw Valley is the junc- 

 tion of the Dale Valley. This, as has been 

 said, is as broad and deep at its junction with 

 the Warsaw Valley, as anywhere above. But 

 its rock bottom is nearly 200 feet above the 

 rock bottom of the Warsaw Valley. It is thus 

 clearly a hanging valley, and the only ex- 

 planation of this relation would seem to be 

 glacial deepening of the Warsaw valley by 

 ice. This removes the only known objection 

 to the southward drainage of the Warsaw 

 system. 



Considering the southward drainage of the 

 Warsaw system into the Genesee at Portage- 

 ville as proved, we must next note that the 

 stream which formerly occupied the St. 

 Helena-Lower Falls valley, and into the 

 floor of which the modern Genesee has in- 

 cised its bed, was also a southward-flowing 

 stream. This is generally accepted, and was, 

 I believe, first pointed out by Fairchild. It 

 could indeed be interpreted in no other way, 

 since it was very evidently not the path of the 

 ancient Genesee. As has been shovra, a third 

 ancient valley of smaller size joined the 

 Genesee from the west. Thus three south- 

 ward and eastward flowing streams united at 

 Portageville. If the Genesee was a north- 

 ward-flowing stream, and the combined system 

 flowed out to the Canaseraga Valley by way 

 of the Nunda Valley, a complicated system 

 existed, with the streams turning back in 

 direction upon themselves. Moreover, in that 

 case the Canaseraga Valley was deepened by 

 ice to the extent of 650 feet. That it was 

 glacially deepened to some extent will be 

 shown later to be the case, but the evidence 

 is clear that such deepening was not more 

 than half the amount required by the hypoth- 

 esis of northward drainage, and probably 

 much less. 



On the hypothesis of southward drainage 

 of the Genesee, the Nunda Valley must be 

 regarded as that of a southwestward-flowing 

 tributary of the Genesee. Its junction with 



