EVOLUTION OF IRONDEOUOIT VALLEY. 143 



sharply and crosses abruptly from the east to the west side of the 

 old silt-filled rock valley. The narrow portal through which it 

 passes is exactly midway of this "dugway'' section. ' From this 

 point north to the Bay, at the Sodus railway powerhouse, the east 

 half of the old valley is being redeveloped by a companion stream 

 parallel to the Irondequoit, while to the south of it the opposite half 

 of the old valley is less plainly delineated by a small gully and then 

 by Allen's creek flowing in a reversed direction to find the main 

 stream. These facts suggest a resistant central core to the valley 

 fillings that might serve' as an axis for the great tongue-like remnant 

 of the silt plain extending north down the middle of the valley from 

 Zarges ]\Iills to Float Bridge and in the other direction well toward 

 Despatch. For an analogous case we need go no farther than the 

 upper Irondequoit valley from Bushnell's Basin to Fisher's, which 

 is similarly bifid by a great esker up the middle. A ruler laid 

 along the general course of that esker, which must indicate pretty 

 closely the direction of local ice-flow while the esker was building, 

 coincides almost perfectly with this hypothetical axis in the lower 

 valley. A visit in 1914 to the narrow portal above Zarges Mills 

 revealed there as had been anticipated a cross section of this buried 

 esker, and further exploration has located other exposures of its 

 gravels where it crosses Allen's creek. Presumably this is the same 

 esker whose delta sands in Lake Dawson lie between Allen's Creek 

 and East Rochester (formerly Despatch) as referred to on a pre- 

 vious page ( I2g). Its supposed course is represented on map A 

 ^ Plate IV), and on Plate III. 



In the section of the Irondequoit nver from Zarges Mills north 

 to the mouth of Palmer's Glen the present excavation in the silts 

 has three widenings and three constrictions {x, y, s on Plate III). 

 In each of these constrictions the river is against the west bank 

 and is apparently unable to make much lateral progress into it, 

 whereas the expanded sections are developed on the opposite or 

 right bank and portray plainly old meander swings into soft silts, 

 now being slowly drowned as the Ontario level rises. It would 

 seem that these meander loops so nearly closed in behind the inter- 

 vening spurs as to leave only insignificant necks capable of being 

 overtopped by the rising waters ; and through those narrow necks 



