2 12 



NA TURE 



[December 30. 1897 



uplift caused the water to slowly recede from its northern shores. 

 This change after a time separated Lake Superior from the other 

 lakes, bringing the St. Marys river into existence, and eventually 

 the present condition was reached. 



These various changes are so intimately related to the history 

 of the Niagara river that the Niagara time estimates, based on 

 the erosion of the gorge by the cataract, can be applied to them. 

 Lake Erie has existed approximately as long as the Niagara 

 river, and its age should probably be reckoned in tens of 

 thousands or hundreds of thousands of years. Lake Ontario 

 is much younger. All that can be said of the beginning of 

 Great Lake Nipissing is that it came long after the beginning 

 of Lake Erie, but the date of its ending, through the transfer 

 of outlet from the Mattawa to the St. Clair, is more definitely 

 known. That event is estimated by Taylor to have occurred 

 between 5000 and 10,000 years ago. 



The lake history thus briefly sketched is characterised by a 

 progressive change in the attitude of the land, the northern and 

 north-eastern portions of the region becoming higher, so as to 

 turn the waters more and more towards the south-west. The 

 latest change, from Great Lake Nipissing to Great Lakes 

 Superior, Michigan and Huron, involving an uplift at the north 

 of more than 100 feet, has taken place within so short a period 

 that we are naturally led to inquire whether it has yet ceased. 

 Is it not probable that the land is still rising at the north, and 



'Clevel&nd 



Fig. 



t.^Map of the Great Lakes, showing pairs of gauging stations and isobases of outlets. 

 The isobases are marked by full lines. Broken lines show the pairs of stations. 



ihe lakes are still encroaching on their southern shores ? Dr. 

 J. W. Spencer, who has been an active explorer of the shore- 

 lines of the glacial lakes, and has given much study to related 

 problems, is of opinion that the movements are not complete, 

 and predicts that they will result in the restoration of the 

 Chicago outlet of Lake Michigan and the drying of Niagara. 



Measurements of Changes in the Shore-lines. 



The importance of testing this question by actual measure- 

 ments was impressed upon me several years ago, and I 

 endeavoured to secure the institution of an elaborate set of 

 observations to that end. Failing in this, I undertook a less 

 expensive investigation, which began with the examination of 

 existing records of lake height as recorded by gauge readings, 

 and was continued by the establishment of a number of gauge 

 stations in 1896. 



If the volume of a lake were invariable, and if its water were 

 in perfect equilibrium under gravity, its surface would be con- 

 stant and level, and any variation due to changes in the height 

 of the land could be directly determined by observations on the 

 position of the water surface with reference to the land ; but 

 these conditions are never realised in the case of the Great 

 Lakes, where the volume continually changes and the water 

 is always in motion. The investigator therefore has to arrange 

 his measurements so as to eliminate the effect of such changes. 



The various oscillations of the water, though differing widely 



NO. 1470, VOL. 57] 



in amplitude, rate and cause, yet coexist, and they make the 

 actual movement of the water surface highly complex. The 

 complexity of movement seriously interferes with the use of the 

 water plane as a datum level for the measurement of earth 

 movements, and a system of observations for that purpose needs 

 to be planned with much care. The main principles of such a 

 system are, however, simple, and may readily be stated. The 

 most important is that the direct measurement of the heights of 

 individual points should not be attempted, but comparison should 

 always be made between two points, their relative height being 

 measured by means of the water surface used as a levelling 

 instrument. 



It will not be necessary to give here the details of observation 

 and computation, as they are fully set forth in a paper soon to 

 be printed by the U.S. Geological Survey, but the general scope 

 of the work may be briefly outlined. As the tilting shown by the 

 geological data was towards the south-south-west, stations were, 

 so far as possible, selected to test the question of motion in that 

 direction. The riiost easterly pair were Sacketts Harbour and 

 Charlotte, New York, connected by the water surface of Lake 

 Ontario (Fig. i). From observations by the U.S. Lake Survey 

 in 1874, it appeared that a bench mark on the old lighthouse in 

 Charlotte was then 18 '531 feet above a certain point on the 

 Masonic Temple in Sacketts Harbour. In 1896 the measure- 

 ment was repeated, and the diff"erence found to be 18 "470 feet, 

 the point at Sacketts Harbour having gone up, 

 as compared to the point at Charlotte, o'o6i 

 foot, or about three-fourths of an inch. Simi- 

 larly it was found that between 1858 and 1895 

 a point in Port Colborne, at the head of the 

 Welland Canal, as compared to a point in Cleve- 

 land, Ohio, rose 0*239 foot, or nearly three 

 inches. Between 1876 and 1896 a point at 

 Port Austin, Michigan, on the shore of Lake 

 Huron, as compared to a point in Milwaukee, 

 on the shore of Lake Michigan, rose 01 37 

 foot, or one and one-half inches ; and in the 

 same period a point in Escanaba, at the north 

 end of Lake Michigan, as compared to the same 

 point in Milwaukee, rose 0'i6i foot, or about 

 two inches. 



There is not one of these determinations that 

 is free from doubt ; buildings and other struc- 

 tures on which the benches were marked may 

 have settled ; mistakes may have been made in 

 the earlier levelling, when there was no thought 

 of subjecting the results to so delicate a test ; 

 and there are various other possible sources of 

 error to which no checks can be applied. But 

 the fact that all the measurements indicate 

 tilting in the direction predicted by theory, 

 inspires confidence in their verdict. 



The stations of the several pairs are at different 

 distances apart, the directions of the lines con- 

 necting them make various angles with the theoretical direction 

 of tilting, and the time intervals separating the measurements 

 are different. To reduce the results to common terms, I have 

 computed from each the rate of tilting it implies in the theoretical 

 direction, S. 27' W., and determined the change in relative 

 height of the ends of a line 100 miles long during a. century. 



Compared in this way, the results are remarkably harmonious, 

 the computed rates of tilting ranging only from 0*37 foot to 

 0*46 foot per 100 miles per century ; and in view of this 

 harmony it is not easy to avoid the conviction that the buildings 

 are firm and stable, that the engineers ran their level lines with 

 accuracy, that all the various possible accidents were escaped, 

 and that we have here a veritable record of the slow tilting of 

 the broad lake-bearing plain. 



The computed mean rate of tilting, 0*42 foot per 100 miles 

 per century, is not entitled to the same confidence as the fact of 

 tilting. Its probable error, the mathematical measure of pre- 

 cision derived from the discordance of the observational data, is 

 rather large, being one-ninth of the whole quantity measured. 

 Perhaps it would be safe to say that the general rate of tilting, 

 which may or may not be uniform for the whole region, falls 

 between 0*30 and 0*55 foot. 



Future of the Great Lakes. 

 The geographical effects of the tilting are of scientific and 

 economic importance. Evidently the height of lake water at 



