222 Carney — Possible Overflow Channel of Ponded Waters. 



ice abutted the rock salient at channel No. 1, thus holding the 

 Keuka valle} r lake up to the level of the channel, the lobe of 

 ice in Seneca valley extended south many- miles, covering the 

 site of the Himrod delta, which was not built till the ice-lobe 

 had retreated several miles north. In case this stream did 

 mark its union with Lake Newberry by a delta, we would look 

 for it southwest of Himrod. 



That this overflow channel correlates with the Newberry 

 level is supported (1) by the position of drift on the north 

 slope of the outlier immediately north, showing that ice was 

 present and hindered the passage of water by a lower contour ; 

 (2) by the distribution of moraine laterally from this position 

 evincing a continuous ice-front ; (3) the evidence of at least 

 some stream work in the direction of Lake Newberry ; (1) the 

 certainty that when the ice in its retreat had reached this point, 

 discovering a lower level, 1000 feet, than the Wayne outlet, 

 1100 feet (both levels corrected for differential movement), 

 Hammondsport Lake would be diverted into Seneca valley. 



But other facts must be considered before deciding that a 

 pro-Wisconsin (recessional) ponded lake was connected with the 

 genesis of channel No. 1 : (1) The walls of this channel have 

 a more aged appearance in slope and the consequent covering 

 of vegetation than have unquestionable post- Wisconsin chan- 

 nels of this vicinity cut in the same formation. (2) There is 

 glacial drift in the bed of the channel; it is evidently mod- 

 ified so far as may be judged in the absence of sections. (3) 

 The proportions of this channel imply a time factor that should 

 be represented by erosional work in the path the stream must 

 have taken. While the altitude of the channel in reference 

 to the Newberry level is not contradictory to their being 

 associated, nevertheless it seems improbable that a stream hav- 

 ing a gradient of approximately 100 feet in less than three 

 miles, and having performed so much degradation work at the 

 outlet of the lake which it drained, should not have developed 

 a channel elsewhere in its course. If such a channel exists, it 

 is buried ; if buried, it was not genetically related to Lake 

 Newberry. 



An Alternate Hypothesis. 

 As the Wisconsin ice-sheet moved into and over the Alle- 

 ghany plateau, its front was fringed by lakes, the ponded north- 

 flowing streams. These lakes, save the very earliest, over- 

 flowed southward, probably doing some erosional work on 

 their outlets. Tarr, however, concludes, after having studied 

 a wide area, that "ice-born stream-erosion," particularly that 

 of marginal lakes, was very slight.* The ice later passed over 



•Bull. Geol. Soc. Am, vol. xvi, pp. 239-240, 1905. 



