May 26, I 



NATURE 



81 



Nos. I to 3, 3 to 5, &c., represent intervals between 

 successive minimum epochs, and Nos. 2 to 4, 4 to 6, &c., 

 intervals between successive maximum epochs. These 

 are shown graphically in the annexed figure. 





:FiG. 2. — Length of sun-spot and magnetic periods compared. The thick 

 line shows the variation in length of successive sun-spot periods, 

 and the thin line that between successive magnetic periods. Odd 

 numbers indicate periods from minimum to minimum, and even 

 numbers periods from maximum to maximum. 



Examining further the collected curves, it is seen that 

 the several maximum points have at different epochs 

 very different degrees of intensity. Arranged in order 

 of intensity these are as follows : — 

 Order of Epochs. 



Sun-spot curve 1870 1848 i860 1894 1884 



Declination cuive ... 1870 1848 i860 1894 1884 

 Horizontal force curve... 1870 i860 1848 1894 1884 



The agreement is complete, excepting that in horizontal 

 force the epochs 1848 and i860 are transposed, although 

 •otherwise falling in with the order of the other curves. 



The paper goes on to point out that — considering how 

 the irregularities in the length of the sun-spot and 

 magnetic periods, and also the order of epochs as regards 

 elevation or depression of the maximum points of the 

 curves, so entirely synchronise, and, further, the usually 

 sharp rise, from minimum to maximum and the more 

 gradual fall again to minimum, a characteristic of all 

 three curves—" there would appear to be no escape from 

 the conclusion that such close correspondence, both in 

 period and activity, indicates a more or less direct 

 relation between the two phenomena, or otherwise the 

 existence of some common cause producing both." 

 Reference is also made to the question of the supposed 

 lagging of the magnetic epoch, as referred to the sun- 

 spot epoch, which the results presented do not appear 

 «iuch to confirm. 



The paper concludes with an inquiry as to how far the 

 practice of including in the Greenwich tabulation of 

 magnetic elements all days (except those of extreme 

 •disturbance) may have affected the results presented, for 

 which purpose diurnal ranges were deduced for the years 

 1889 to 1896 from five selected quiet days in each month 

 — days free from magnetic disturbance — with result that 

 the diurnal ranges so found show the same variation 

 with sun-spots as do the diurnal ranges of the ordinary 

 tabulation. William Ellis. 



MONOGRAPHS OF THE UNITED STATES 

 GEOLOGICAL SURVEY. 

 The Glacial Lake .A.gassiz.i 

 /^NCE upon a time in North America the continental 

 ^^ ice-sheet attained an area of about four million 

 square miles, while its maximum thickness, in the central 

 portion, was probably from one to two miles. It extended 



1 By Warren Upham. (" MonoKrapbs of the U.S. Geological Survey," 

 vol. XXV. Pp. xxiv -f 658 ; 38 maps, and 35 other illustrations.) 



from the Atlantic to the Pacific, and from the northern 

 United States to the Arctic Sea. During the closing 

 stage of this glaciation there existed an immense lake, 

 whose area is estimated to have been about 110,000 

 square miles ; a lake which extended 700 miles in length, 

 and attained a width of 250 miles. Its maximum depth 

 was 700 feet above the present level of Lake Winnipeg. 



That the idea of the former existence of this great lake 

 is no romance of modern days, no " glacial nightmare," 

 is indicated by the fact that so long ago as 1823 the 

 traces of it were recognised by Keating. Not, however, 

 until 1879 was its present name applied to it in honour of 

 Louis Agassiz. 



Placed almost in the centre of North America, and 

 occupying what is now part of the plain of the Red River 

 and Lake Winnipeg, together with the Lake of the 

 Woods and other smaller sheets of water, it lay for the 

 most part in territory now Canadian, but a fifth part 

 occurring within the United States. In the southern 

 region, however, its ancient shore-lines have been more 

 exactly explored. A very large part of its area in Canada, 

 besides a considerable tract within its limits in northern 

 Minnesota, is covered by forest, which makes it im- 

 practicable to trace there the beach-ridges and deltas, 

 usually but a few feet high, the low escarpments of erosion, 

 which range from 10 to 30 feet, and the other evidences 

 of this lake, which in the prairie region could far more 

 readily and definitely be followed. 



It was evident that the scientific study of this interesting 

 region should not be restricted by national geographical 

 circumstances, and it is pleasing to note that arrangements 

 were made between Director Powell (,of the U.S. Geo- 

 logical Survey) and Director Selwyn (of the Canadian 

 Survey), that the work of mapping the shores of Lake 

 Agassiz should be continued by Mr. Upham through the 

 prairie region of south-western Manitoba. Altogether 

 this work comprises the results of field-observations 

 carried on during six years. 



Over the greater part of the old lacustrine area there is 

 boulder clay from 100 to 300 feet thick. A series of 

 terminal moraines marks the stages of retreat of the ice- 

 sheet. For a while the lake gradually increased in size 

 northwards, finding an outlet to the south in the "glacial" 

 River Warren, whose channel was cut to a depth of 90 

 feet, and whose course is now occupied by Lakes Traverse 

 and Big Stone and by the Minnesota River. As the 

 ice-front retreated the lake was eventually drained by the 

 natural slope of the land to the north-east, excepting in 

 those areas which now form the lakes of Manitoba. 



The steady uplifting of the area of Lake Agassiz, 

 resulting from the unburdening of the land by the 

 recession of the ice-sheet, gave to its beaches a northward 

 ascent, and caused the several shores of its southern part 

 to become double or multiple as they are traced north- 

 ward. 



The author gives a full and particular account of the 

 beaches formed at different stages in the history of the 

 lake, and discusses various earth-movements, some of 

 which were independent of glaciation. 



His work is by no means devoid of practical value. A 

 chapter is devoted to artesian and common wells, and 

 to the distribution and origin of certain saline waters ; 

 and another chapter is given to the agricultural and 

 material resources of the area. 



The Flora of the Amboy Clays.^ 

 This work, which was very nearly finished in 1890, by 



the late Dr. Newberry, was placed in the hands of Mr. 



HoUick in 1892. His task has not been unattended with 



difficulty, but he has carefully indicated his necessary 



alterations and additions. 



The Amboy Clays take their name from Perth Amboy 



1 By John Strong Newberry. Edited by Arthur HoUiclc. (" Monographs 

 of the U.S. Geological Survey," vtl. xxvi. Pp. x -(- 137, with 58 plates.) 



NO. 1 49 I, VOL. 58] 



