220 GROVE KARL GILBERT— DAVIS [ME " om \xo"xkt 



He undoubtedly had occasion, as his research advanced, to modify earlier conclusions and 

 perhaps sometimes to reverse them; but he was one who never seems to have had difficulty in 

 maintaining an open mind even with regard to his own earlier opinions; his adoption of "sim- 

 plified spelling" shows that. He wrote characteristically to the same correspondent: 



You have my cordial s\'mpathy in your regret that some of your work must be undone, but there is another 

 side to the matter which has some importance from a personal point of view. The ability to recede from a 

 cherished position is not so common but that it is highly appreciated by the scientific public, and the man who 

 corrects his own mistakes instead of waiting for others to do it for him reaps a positive benefit in reputation. 



This recalls the incident at the first winter meeting of the Geological Society of America, 

 above narrated. A later letter commented on the unsatisfactory methods of a certain observer, 

 who Gilbert thought was — 



injuring himself in an intellectual way by continuing to theorize as to numerous details when he is unable to 

 acquire personal knowledge of the facts. He is certainly unfortunate in not being able to give up a minor factor 

 of a first theory when it is clearly shown that the facts are against him. 



Gilbert regarded his own problems in so objective a manner that no such criticism could 

 have been justly directed against Mm. 



THE FIRST EASTWARD DISCHARGE OF THE PROGLACIAL LAKES 



One of the most gratifying discoveries ever made in connection with the evolution of the 

 Great Lakes and Niagara is to be credited to Gilbert for the season of 1896. It concerned 

 certain minor effects of the change from a westward to an eastward discharge, of the proglacial 

 lake waters, which, as already intimated, took place when the receding ice margin opened a 

 passage along the northern slope of the Appalachian plateau margin in central New York at 

 a lower level than that of the outlet which till then had led westward to the Mississippi. His 

 earlier conception of this change was imperfect. The statement concerning it in the Toronto 

 lecture of 1889 was as follows: 



The next change in the history of the lakes was a great one. The ice, which had previously occupied nearly 

 the whole of the Ontario basin, so far withdrew as to allow the accumulated water to flow out by way of the 

 Mohawk valley. The level of discharge was thus suddenly lowered 550 feet, and a large district previously 

 submerged became dry land. Then for the first time Lake Erie and Lake Ontario were separated, and then 

 for the first time the Niagara river carried the surplus water of Lake Erie to Lake Ontario. 



This brief statement was truly a great advance over earlier vague ideas; but in describing 

 the lowering of the level of discharge as taking place "suddenly" many detailed happenings 

 that accompanied the lowering were overlooked. 



It was these detailed happenings that were discovered in 1896. Gilbert wrote on October 

 15 of that year to the same worker to whom he addressed the several letters from which extracts 

 are quoted in the preceding section: 



In the longitude of Syracuse I find four channels crossing one N-S ridge [a spur of the plateau slope]. They 

 are all in the space of six miles but have a vertical range of about 375 feet. Two of them had [east-falling] 

 cataracts over Helderberg limestones, and one falling 160' dug a hole which now holds a lake 50' deep. I find 

 that this lake already has a bundle of theories tied to it. 



No explanation accompanied this brief statement, but when the explanation was announced 

 it completely replaced the whole bundle of earlier theories. A week later, after he had led a 

 local geologist over some of the channeled spur, Gilbert wrote again : 



I find skeptical about my old channels across this country and am going to take him tomorrow to 



see a fine one that I have just finished tracing. It starts beautifully in the air across a limestone sill and then 

 alternately canyons through shale and builds deltas in transverse [north-south] valleys of the Finger lake type. 



It thus appears that the channels were by no means self-explanatory, even when they were 

 pointed out. The explanation that Gilbert gave them is in essence as follows: 3 



In the first place, the general principle was well understood that when the retreating ice 

 sheet lay upon land that sloped away from the ice margin, the ice-water streams merely re- 

 enforced the normal drainage of the surface and ran away with it; but Gilbert hardly touched 



' Old tracks of Erian waters in western New York. Bull. Oeol. Soc. Amer., viii, 1897, 285-286. 



