Apbil 5, 1907] 



SCIENCE 



539 



of this Bection recently published by the state 

 survey. Where the connection is broken, this 

 can generally be shown to be due to drift 



The following drainage systems were tenta- 

 tively outlined, the outline being presented as 

 a report of progress rather than as a final 

 settlement in any one case. 



On the west, the Wyoming (Warsaw) valley 

 probably had the Dale valley, now occupied in 

 part by the Little Tonawanda, as a western 

 branch, joining it north of Warsaw. The 

 Warsaw valley is still believed to have been 

 continuous with the Upper Genesee valley, 

 above Portageville, by way of Glen Iris, as 

 outlined by the speaker in 1894 and earlier. 

 The valley of Silver Lake joined the Warsaw 

 valley somewhere near Silver Springs. A 

 narrower valley, now occupied by the Genesee 

 from Gibsonville to St. Helena, is continued 

 by a buried gorge from that place to Portage- 

 ville, where it joins with the Warsaw-Glen 

 Iris valley and another valley from the north- 

 west, to continue southward in the large valley 

 now occupied by the Upper Genesee. 



The Canasseraga valley, now occupied in 

 part by the Genesee, was cut by an inde- 

 pendent stream. This is the largest valley of 

 the region and was that of the master stream. 

 The Nunda-Cashaqua valley, generally held to 

 have been the former path of the Genesee, is 

 probably only an inner-lowland type of valley, 

 carved on the contact between Portage shales 

 and Chemung sandstones. It may have been 

 in part a tributary of the Genesee at Portage- 

 ville. The Canasseraga, above the junction 

 of the Cashaqua, is as broad and flat-bottomed 

 as below that point, and was certainly con- 

 tinuous throughout, being carved by a single 

 stream, the Tertiary Canasseraga, as sug- 

 gested nearly fifteen years ago by the speaker. 

 This river, flowing southward, received as a 

 tributary the Conesus, the valley of which is 

 broad and open to Scottsburg. Hemlock and 

 Canadice rivers joined southward, receiving 

 another branch near Springwater, the united 

 series joining the Canasseraga by way of Way- 

 land. Honeoye and Canandaigua rivers, joined 

 near Naples having another eastern branch in 

 West Eiver. Originally this series may have 



drained southward by way of Cohocton, but 

 may later have been captured by a branch of 

 the Canasseraga. This proposition, however, 

 needs careful study. Another branch of this 

 system seems to have been the Flint, the 

 valley of which, traceable for twenty miles or 

 more, points toward the Cohocton outlet. 

 Another system is represented by the two 

 branches of Keuka Lake, which have other 

 branches uniting with them southward. 



Other systems are represented by the valleys 

 of the more eastern lakes. So far as the study 

 has proceeded, these valleys could only have 

 been formed by a southward drainage, as out- 

 lined in Bulletin 45, New York State Museum. 



The remainder of the paper consisted of a 

 description of the gorges and falls about Port- 

 age, illustrated with lantern slides. The suc- 

 cessive stages in the development of the lower 

 falls received special attention. 



Dr. George F. Kunz then presented a plan 

 of development of the park as a means for 

 scientific education. Both papers were illus- 

 trated with series of beautiful lantern slides. 

 Alexis A. Julien, 

 Secretary of Section 



THE TORREY BOTANICAL CLUB 



The meeting of December 11, 1906 was 

 called to order at 8 :15 o'clock p.m., at the 

 American Museum of Natural History, with 

 President Eusby in the chair. Eight persons 

 were present. 



The scientific program was as follows : 

 ' Some Hawthorns of the Vicinity of New 



Yorh City': Mr. W. W. Eggleston. 



Species and variations of Cratcegus growing 

 within the vicinity of New York City were 

 described and illustrated by herbarium speci- 

 mens. 



' Centers of Distribution of Coastal Plain 



Plants ' : Eoland M. Harper. 



One of the most familiar phenomena of 

 plant distribution is that neighboring areas of 

 equal extent often differ considerably in the 

 number of species they contain. And it 

 usually happens that a region with a rich 

 flora (if a large enough area be taken into 

 consideration) contains a considerable num- 



