ACADEMY OF SCONCES] qjjjq g^yEY 13 



appears to belong to C. A. White, who, as State geologist of Iowa, had somewhat earlier pub- 

 lished an account of "two well marked but slight elevations in the general [drift-covered] 

 surface of the country," both of which "seem at least to be accumulations of drift material 

 which mark periodical arrests of the recedence, by melting, of the glaciers to the northward 

 as the glacial epoch was drawing to a close. 1 And it is also true that priority in the detection 

 of local lobation with divergent striations on the ice-sheet margin should be credited to M. C. 

 Read, an associate of Gilbert's on the. Ohio survey, although the lobation that he described 

 was due to a northward-opening embayment in the conglomerate-capped uplands of north- 

 eastern Ohio into which a salient of the ice fitted, 2 rather than to a more abundant advance 

 of the ice along a broad depression, such as the floor of Lake Erie. Hence even if Gilbert 

 should not be credited with absolute priority of statement, his views concerning moraines and 

 ice-margin lobation must certainly be regarded as better defined and of broader reach than 

 those of his contemporaries. Yet in certain respects his views were incorrect, as he himself 

 later acknowledged; for there was no land barrier by which, as he supposed, lacustrine waters 

 could have been held at so high a level as to have submerged the morainic swells which now 

 guide the rivers; and instead of their being covered by a "subsequent aqueous and iceberg 

 deposit," it is the clayey moraines themselves that form the broad swells of the surface. 3 



As to the lacustrine barrier, Gilbert made a singular error which he afterwards righted. 

 He erroneously assumed that the higher level reached by the expanded Lake Erie was due 

 to an uplift of the land in the region of the St. Lawrence Eiver, an idea which he held 

 with sufficient confidence to mention it briefly again five years later in his report on the Henry 

 Mountains, even though Newberry had added a corrective footnote to the Maumee Valley 

 report reading as follows : 



It should be remembered that the retreating glacier must have, for ages, constituted an ice dam that 

 obstructed the natural lines of drainage, and may have maintained a high surface level in the water basin that 

 succeeded it. 



When Gilbert was 15 years older and greatly matured by his experience in the Far West, 

 he returned to the investigation of the Great Lakes region, and then, if not sooner, recognizing the 

 correctness of Newberry's good guess, brought out his masterful essay on the history of Niagara 

 River, as will be further told below. 



Another item in the Maumee Valley report deserves mention for its bearing on later studies, 

 as well as for the evidence that its final statement gives of Gilbert's cautious manner of dealing 

 with his problems. He records : 



It is noteworthy that the small streams [which flow from the clay plain into the southwestern end of Lake 

 Erie] . . . occupy, near their mouths, larger channels than it seems natural that they should have opened under 

 the existing conditions. ... If we suppose that the present water level of the upper [southwestern] end of 

 Lake Erie was immediately preceded by a lower level, we have an easy explanation of the phenomena. 



In other words, he recognized that the broadened stream mouths should be explained 

 as slightly drowned valleys, although neither he nor anyone else had at that time used the sug- 

 gestive phrase, "drowned valleys," in this sense. Similarly, the brief statement: "There is 

 evidence that Lake Ontario at Rochester, N. Y., has stood seventy feet lower than now," 

 suggests that he had recognized the neighboring Irondequoit Bay, familiar in his boyhood, as 

 a partly submerged valley. Then after noting that the upland at the eastern end of Lake Erie, 

 through which Niagara River has cut its gorge, is 38 feet above present lake level," and that 

 wave-work ought to have formed beaches corresponding to that outlet level all around the 

 lake shores, he adds regarding the southwestern end of the lake : 



We must look for the record of this work considerably above, or somewhat below the present coast; the 

 present data do not indicate which is the more probable position. 



The problem thus opened he completely solved later. No one else seems to have examined 

 it in the interval. 



' Report, Oeol. Survey Iowa, I, 1870, 98. 



Qeol. Survey of Ohio. Report of Progress, 1870, 471; repeated in Vol. I, 1873, 639. 

 > U. S. Oeol. Survey, Monogr. XLI, 1902, 666. 



