from the north, the Sinnemahoning and Moshannon waters from the west, and the Bald Eagle waters from the south. But before the ice invaded New York from Canada, and spread over the northeastern counties of Pennsylvania as far south as the Terminal moraine, Kettle creek was a branch of the Genesee river, and Pine creek was a branch of the Tioga river; consequently most of the rainfall of Potter and Tioga counties formerly flowed off into Lake Ontario; whereas now it swells the volume of the West Branch Susquehanna.
The East Branch Susquehanna, which joins the West Branch at Northumberland, and is much the larger river of the two, was formerly as much inferior to it; for it was only fed by the Lackawanna waters in the east, the Fishing creek and Tunkhannock waters on the north, and the Catawissa and Shamokin waters on the south. All the rest—the greater part—of the rainfall of the New York–Pennsylvania state-line region, which now pours its vast volume (often in disastrous floods) down the valley of the North Branch Susquehanna into Chesapeake bay, then—before the Ice age—poured through Canandaigua lake into Lake Ontario at Sodus Point.
Mr. John F. Carll first placed this whole subject of old Lake country drainage on a sound basis by his observations in the oil regions, as referred to in foot-note to page 106, Report I, 1875, and given fully with illustrations in his third report (III, 1880), where he proved that the upper Allegheny river water basin drained through Lake Chantauqua into Lake Erie, and gave reasons for suspecting what is now pretty certain, that the whole Monongahela, Allegheny and upper Ohio water basin was also drained into Lake Erie by the Beaver–Mahoning–Grand River valley.
Mr. Carll’s summary of his late observations on the Susquehanna waters was given me in a letter dated March 23, 1890, in the following words:
“During the last winter, while doing some field work in central New York, and the counties of Tioga, Bradford and Susquehanna, Pennsylvania, I incidentally made some discoveries in relation to pre-glacial water ways which seem
