I04 



SCIENCE. 



[Vol. X. No. 238 



the land is northward. As soon as the border of the ice had re- 

 ceded beyond the watershed dividing the basins of the Minnesota 

 and the Red Rivers, it is evident that a lai<e, fed by the glacial 

 melting, stood at the foot of the ice-fields, and extended northward 

 as they withdrew along the Red River valley to Lake Winnipeg, 

 filling this valley and its branches to the height of the lowest point 

 over which an outlet could be found. Until the ice-barrier was 

 melted upon the area now crossed by the Nelson River, thereby 

 draining this glacial lake, its outlet was along the present course of 

 the Minnesota River. At first its overflow was upon the nearly 

 level, gently undulating surface of the drift, about i,ioo feet above 

 the sea; but in process of time this cut a channel 125 to 150 feet 

 deep and from one to two miles wide, in which lie Traverse and 

 Big Stone Lakes, respectively 970 and 962 feet above the sea. From 

 this outlet the plain of the Red River valley, 30 to 50 miles wide, 

 stretches 315 miles north to Lake Winnipeg, which is 710 feet 

 above the sea. Along this entire distance there is a very uniform 

 continuous descent of a little less than one foot per mile. The 

 drift deposited by the ice-sheet upon this area, together with that 

 which may have been dropped by floating ice borne on the waters 

 of the lake, and the silt brought in by glacial rivers and by those of 

 the surrounding land, were here received in a lake, shallow near its 

 mouth, but becoming gradually deeper northward. 



Beyond our national boundary. Lake Agassiz covered a broad ex- 

 panse, including the basins of Lake Winnipeg, Red and Rainy 

 Lakes, and the Lake of the Woods. Its breadth varied from 100 to 

 200 miles, with an extreme length of at least 600 miles and an area, 

 at the time of its greatest extent, exceeding that of Lake Superior. 



The most interesting geological features of the basin of this 

 ancient lake now observable are the terraces or beaches formed 

 along its shores at different levels as its outlet was gradually lowered 

 by erosion. These beaches are continuous ridges of sand and 

 gravel, unbroken, save where crossed by modern streams or ex- 

 panded into the deltas of the ancient lake, whose outlines are thus 

 accurately traced at four distinct levels. The highest or Herman 

 beach is, at the southern end of the lake, 1,045 f^st above the sea, 

 or 85 feet above Lake Traverse, and the lowest beach. 



Mr. LTpham's careful determinations of the altitudes of the beaches 

 have fully established the remarkable fact that the beaches are not 

 level, but have a gradual ascent northward, as compared with the 

 present level line or the surface which a body of water would have 

 now if confined in this valley. The rate of ascent of the highest or 

 Herman beach increases gradually from six inches a mile at Lake 

 Traverse, to above sixteen inches a mile near the national boundary, 

 the total ascent in this distance being 185 feet. 



The several beaches are not parallel, the rate of ascent diminish- 

 ing from the highest to the lowest beach. Thus the second beach 

 is 120 feet, the third 65 feet, and the fourth or lowest 35 feet, higher 

 at the national boundary than at Lake Traverse. 



The altitude of the beaches is a function of the longitude as well 

 as the latitude ; for a comparison of these beaches in Dakota and 

 Minnesota at the same latitude reveals an ascent from west to east 

 similar to that from south to north, but of less amount, and dimin- 

 ishing in a similar ratio between the successive stages of the lake. 



Various causes for these interesting phenomena are suggested 

 and reserved for future discussion ; but Mr. Upham indicates his 

 adoption provisionally of the view that the divergence of these an- 

 cient shore-lines from the present level line was produced by the 

 gravitation of the water of the lake toward the ice-sheet. At first 

 this attraction would have been relatively large, because of the 

 nearness of the great mass of ice on the north-east in Minnesota 

 and northward in British America; but, as the ice retreated, it 

 must have been gradually diminished, and reduced to a compara- 

 tively small influence by the time the ice-sheet had withdrawn so as 

 to permit the northward drainage of the lake. 



NOTES AND NEWS. 

 Two new methods of determining the density of the earth are 

 being experimented upon at Berlin. The one, by Dr. F. Richarz 

 and Dr. A. Konig, has been referred to in Science (v. 217). These 

 gentlemen apply a sensitive balance with a double pair of scales, 

 one swinging above, the other below, a heavy parallelopipedic 

 mass of lead, which consists of a number of blocks which are ex- 



actly measured and weighed. The blocks are perforated, and the 

 wires connecting the upper and lower scales pass through the shaft 

 formed by these perforations. By an ingenious arrangement, the 

 weights, which consist of spheres of lead, can be changed from the 

 upper to the lower scales without opening the case in which the 

 balance is enclosed. The principle on which the experiment is 

 founded is, that, if one of two equal weights is below, the other 

 above, the mass of lead, its attraction will diminish the weight of 

 the former and increase that of the latter. The proportion between 

 this increase and the total weight gives the means for determining 

 the' proportion between the attraction and masses of the lead and 

 the earth. PreHminary experiments made with this balance show 

 that a great exactness of the definite measurements may be ex- 

 pected. These experiments are being earned on under the auspices 

 of the Berlin Academy of Sciences in a casemate of Spandau. At 

 the same time J. Wibring is experimenting by another method in 

 the astrophysical observatory at Potsdam. He uses a pendulum 

 made of a brass tube one metre in length and four centimetres in 

 diameter, with spheres of cast iron weighing five hundred and fifty 

 grams at the two ends. A knife-edge of agate six centimetres in length 

 passes through the centre of the tube, and swings on an agate rest. 

 Two small mirrors are attached to the knife-edge, and the oscilla- 

 tions of the pendulum are observed through a telescope. The time 

 of oscillation of the pendulum may be so nicely regulated that 

 oscillations of five minutes length are perfectly regular. Near the 

 iron spheres and opposite to one another, two iron cylinders 

 weighing 325 kilograms are placed, the lower one attracting the 

 lower end of the pendulum to one side, the upper one the upper 

 end to the opposite side. The attraction of these masses affects the 

 oscillations of the pendulum. The result of these observations for 

 the mean density of the earth is 5.594 ± 0.032. The mean of former 

 reliable observations being 5.57, the new figure corresponds well 

 with these. Both experiments will result in a more accurate and 

 trustworthy determination of the mass of the earth. 



— In a recent paper on literary catalogues, Mr. Samuel H. Sin- 

 deler makes some suggestive remarks about the system of cata- 

 loguing now so much in favor. To quote his words, " Paradoxical 

 as it may sound, the very excellence of his [Dewey's] plan is one 

 objection to it. Mr. Dewey multiplies co-operative advantages to 

 those who use his system to such an extent, that if he lives long 

 enough he will make it so much to the advantage of newly forming 

 or growing libraries to use it, that none will be independent enough 

 to modify it. And why should they wish to modify it } Simply 

 because, less than fifty years ago, the present scheme could not 

 have been formed. There was not knowledge enough in the world. 

 There could not now be found, in any scheme then formed, place 

 for a long range of subjects which appear in his actual classifica- 

 tion. This is especially true in science, and who shall say that 

 history will not repeat itself in the next fifty years ? Let us rather 

 work out the problem of the decimo-mnemonic system on different 

 lines, each library or group of libraries for itself, according to the 

 special needs of the same. Then new Deweys will arise and ply 

 their ingenious arts, and in the millennium the fittest will survive. 

 At present there is danger that the fittest will be handicapped. To 

 give the fittest, when it comes, an earlier chance of survival is one 

 purpose of this paper." 



— Cumming's ' Electricity Treated Experimentally,' which was 

 reviewed in the last number of Science, is published m America by 

 D. Van Nostrand. 



LETTERS TO THE EDITOR. 



*.,.* The attention o/ scientific vien is called to the advantages ofithe correspondence 

 cohimns o/'^'Ci^^Q.E. for placing promptly on record brief preliTninary notices of 

 their investigations. Tiventy copies of the number containing his communication 

 will be furtiished free to any correspondent on request. 



The editor ivill be glad to Publish any queries consonant •with the character of 

 thejournal. 



Correspondents are requested to be as brief as possible* The "writer's na7ne is 

 £« a^ll cases regu ired as Proof of good fa ith . 



Chrome considered as a Poison. 



My attention has been called to an article by Mr. William Glenn 

 of Baltimore in Science (x. 58), entitled * Chrome considered as a 

 Poison,' criticising a paper of mine in the Boston Medical ana 

 Surgical Journal on the same subject. Were it not that the criti- 

 cism appears in a scientific journal of high standing, I should 



