intervals of rest are plainly marked by the beach lines, of which a series is presented near the borders of each of the lakes, outlining their ancient limits. In the Maumee valley four of these periods of rest are distinctly recorded, the water levels being marked at 220, 195, 165 and 90 to 65 feet, above Lake Erie, along the line of the Air Line branch of the Lake Shore and Michigan Southern Railway, running west from Toledo. Two of these beach lines have been traced for some distance across the country, and their courses are marked on the accompanying map. The lower of the two is the one at 165 feet. Lake Erie then stretched sixty miles up the valley to the site of the town of Defiance, covering all that country now known as the Black Swamp. The upper beach is the highest of the series, and records a lake level at Adrian, Bryan and Columbus Grove of 220 feet above the present, of 210 feet. at Delphos, and of 205 feet at New Haven, six miles east of Fort Wayne. Toward the latter point the beach lines converge, coming from the northeast and southeast; but, instead of uniting, they become parallel, and are continued as the sides of a broad water-course, through which the Great Lake Basin then discharged its surplus waters. At New Haven this channel is not less than a mile and a half broad, and has an average depth of 20 feet, with sides and bottom of drift. For twenty-five miles this character continues, and there is no notable fall. Three miles above Huntington, Indiana, however, the drift bottom is replaced by a floor of Niagara limestone, and the descent westward becomes comparatively quite rapid. At Huntington the valley is contracted in width to one mile, and is walled, on one side at least, by rock in situ. In the eastern portion of this ancient river bed the Maumee and its branches have cut channels fifteen to twenty-five feet deep, without meeting the underlying limestone. Most of the interval from Fort Wayne to Huntington is occupied by a marsh, over which meanders Little river, an insignificant stream whose only claim to the title of river seems to lie in the magnitude of the deserted channel of which it is sole occupant. At Huntington the Wabash river emerges from a narrow cleft of its own carving, and takes possession of the broad trough to which it was once but a humble tributary. The limestone above Huntington is the rocky rim or dam which determined the altitude of the overflow at this point, and is 170 feet above the present level of Lake Erie. Above it the stream must have resembled the Detroit, bearing a smooth surface, but with enough current to excavate its soft bottom somewhat deeply where the marsh and prairie of the Little river now are spread; below, it was more comparable to the Niagara at Buffalo, where it rushes over the outcrop of the Corniferous limestone. At Fort Wayne the St. Joseph and St. Marys contributed their waters. Their mouths were more than fifty feet higher than now, and the flood-plains of gravel and i sand, which they then formed, now flank their valleys as terraces, and can be traced for forty miles toward their sources. When, by the retiring of the lake, they were united, but slight i cause was needed to turn them eastward along the level bottom of the deserted channel, and they have now cut their beds so deeply in the drift, that the highest freshets do not connect them with the Little river. In addition to its general interest, the fact of this ancient, southwesterly lake-discharge is an important element in the study of the nature of the changes, in virtue of which the lakes have stood at so many different levels since the iceperiod. The idea suggested by Dr. Newberry (in a paper before the Boston Society of Natural History in 1862), that they may have been due to local, rather than continental, upheaval and depression, receives here strong confirmation. At the point where Lake Michigan once found outlet southward through the Desplaines river,* the rocky barrier is but four feet above the present level of that Lake, and seventeen feet above Lake Erie. A barrier of drift may have existed one hundred feet higher. At the Wabash outlet the rocky rim is 170 feet above Lake Erie, and was probably covered by fifty feet of drift. And at Lewiston, where the Niagara commenced its work, the elevation of the rim is 38 feet higher than the present Lake.+ The order in which these channels were opened and deserted, is not yet known, but whatever sequence is supposed, it is equally evident that the changes that produced it must have involved the tilting and warping of the land. At whatever time the Wabash valley received the discharge, the barriers, east and west, must have been relatively much higher than at present. To restore now the old water level and current at Fort Wayne, we would need, not merely to fill the gorge of the Niagara, and renew the escarpment at Lewiston, but to construct on that escarpment a retaining wall 170 feet high and many miles in length; and, after filling the valley of the Desplaines to the height-one hundred feet-of the adjacent drift hills, another hundred feet would be needed to complete the dam. Just what has been the warping of the basin to produce this contortion of the rim, I am not prepared to say, but the *This outlet is described by Dr. E. Andrews in his recent paper on "The Nort American Lakes considered as chronometers of post-glacial time" (Trans. Chicag Acad. Sci., vol. ii, p. 14); and the additional data here given in regard to it were kindly furnished me, by the same gentleman, in a private letter. Nat. Hist. of New York, Geology of Fourth District. By James Hall, p. 384 It is interesting to note, in passing, that the containing rock was in each cas the Niagara limestone. and that a deep gorge has been cut at Niagara only because there alone was an underlying softer rock exposed. problem is not insoluble. The lines of bluffs, shingle, ridges and dunes, that mark the limits of the water at its various stages, are plain and faithful records, and their tracing cannot fail to throw light on the character and the order of the undulations which have affected the valley of the lakes in post-glacial time. The field for exploration is large, and the data already accumulated inconsiderable, but we may hope the geological surveys now in progress in Ohio, Indiana and Michigan, and in Canada, will in a few years develop the whole subject. ART. LI-Memorandum on the Amphipleura pellucida; by J. J. WOODWARD, Assistant Surgeon, U. S. Army. THE attention of microscopists has frequently been directed, of late years, to the Amphipleura pellucida or Navicula acus, as a test object well suited to try the defining powers of the very best object glasses. The length of this diatom is stated by Pritchard as ranging from th to th of an inch. The average length is given by the Micrographic Dictionary at 0044 of an inch. The striæ, which are exceedingly difficult, were first described by Messrs. Sollitt and Harrison, who estitimated them at from 120,000 to 130,000 to the inch. Their estimate has been adopted by the Micrographic Dictionary and by the majority of modern writers who have referred to this test; but so many difficulties beset the resolution that few microscopists appear to have attempted to verify the original estimates. Indeed most observers would seem to have been unsuccessful in their efforts to resolve the Amphipleura even with the best objectives and some have gone so far as to deny the existence of any striæ upon the frustules of this species. Among the microscopists who claim to have seen the striæ, several would seem to differ from the original estimates of Sollitt and Harrison as to their fineness. Dr. Royston Pigott, whose papers on "high-power definition" in the Monthly Microscopical Journal have recently attracted much attention, sets down their number at 150,000 to the inch. Dr. Carpenter, on the other hand, in the 4th edition of "The Microscope and its Revelations," expresses the opinion that even the estimates of Messrs. Sollitt and Harrison are too high: and we are told by Mr. Lobb (Monthly Microscopical Journal, vol. iii, p. 104) that Mr. Lealand has recently "succeeded in counting the Amphipleura lines and finds them 100 in 'th of an inch." A few months ago two slides of Amphipleura pellucida were received at the Army Medical Museum from Messrs. Powell and Lealand, and I succeeded in obtaining excellent resolution AM. JOUR. SCI-THIRD SERIES, VOL. I, No. 5.-May, 1871. th by the immersionth of these makers. The frustules on the two slides were found to measure from th to of an inch in length. Resolution could be satisfactorily effected and the striæ counted on any of them. I took eight successful negatives from medium size and small frustules, and verified the counts made in the Microscope by counting the striæ on the glass negatives. I found the striæ on medium sized frustules, sayth of an inch in length, counted usually from 90 to 93 striæ to the Tath of an inch; in that selected for the two photographs which accompany this memorandum the number was 91 to the rath of an inch. Larger frustules exhibited rather coarser, smaller ones rather finer striæ. On the smallest frustules at my disposal, several of them only th of an inch in length, I found no example in which the number of striæ exceeded 100 to the th of an inch. The striæ of these smallest and most difficult frustules do not then rival in fineness the nineteenth band of the Nobert's plate, as has been asserted by some; they compare rather with the sixteenth and seventeenth bands. After making the photographs I extended my observations to a number of other slides of Amphipleura pellucida including two of the original specimens from Hull, kindly sent to the Museum some time since by Mr. W. S. Sullivant, of Columbus, Ohio, and the example in the First Century of Eulenstein. I found that different slides varied considerably in the ease with which I could resolve them, chiefly as I think on account of the thickness of the glass covers, which in several instances did not permit the best work of the immersion th. Perhaps, however, the markings on some frustules may be shallower than on others whose striæ count the same number to the Taath of an inch. In any event I have found, as yet, no slides the covers of which permit theth to be appproximately adjusted, on which it was impossible to resolve the frustules, and no frustules the stria of which exceeded 100 to the Tath of an inch. The best resolution I was able to obtain by ordinary lamp light was not very satisfactory. I used therefore, during the investigation, direct sunlight, rendered monochromatic by passage through the solution of ammonio-sulphate of copper. A parallel pencil of such light was concentrated by the achromatic condenser, which was suitably decentred to attain obliquity. The same illumination was employed in making the photographs. I have since had the pleasure of exhibiting the resolution in quite as satisfactory a manner to several microscopists by monochromatic light obtained from the electric lamp. War Department, Surgeon General's Office, Army Medical Museum, Feb. 1, 1871. ART. LII.-Memorandum on the Surirella gemma; by J. J. WOODWARD, Assistant Surgeon, U. S. Army. THE Surirella gemma has been recommended by Hartnack as a test for immersion objectives of high powers. I have not gained access to his original description, but find accounts of his views, with figures, in the works of Drs. Carpenter and Frey. (The Microscope and its Revelations," 4th edition, p. 182. "Das Mikroskop," 3d edition, p. 40.) Hartnack observed fine longitudinal striæ in addition to the fine transverse ones previously known to exist between the large transverse ribs; he supposed the true markings to have the form of elongated hexagons. Two handsome slides of this diatom were received at the Army Medical Museum a few months since, from Bourgogne of Paris. A careful study of these by monochromatic sunlight inclines me to the opinion that Hartnack's interpretation is erroneous, and that the fine striæ are in reality rows of minute hemispherical bosses; from which, as in the case of other diatoms, the appearance of hexagons would readily result if the frustule was observed by an objective of inferior defining power to that I used, or if the illumination was unsuitable. This memorandum is accompanied by two photographs exhibiting what I saw; one is magnified 1,034, the other 3,100 diameters. The principal frustule shown in these photographs is th of an inch in length. (The mean length of S. gemma is stated by the Microscopic Dictionary atth of an inch.) The fine transverse striæ counted longitudinally at the rate of 72 to theth of an inch. Transversely these were resolved into beaded appearances which counted laterally 84 to the of an inch. If the structure consists, as I suppose it does, of fine hemispherical bosses, projecting from the surface of the frustules, the fact that these bosses are set together more closely in the transverse direction than in the longitudinal would account for the elongated form of the pseudo-hexagons when seen. th Some parts of the photographs closely approach Hartnack's description, but it is easy to observe that these are not the parts which are most nearly in focus. I have also resolved this diatom by monochromatic light derived from the electric lamp. The appearances obtained were identical with those above described. War Department, Surgeon General's Office, |