The comet 1862 III (sometimes called 1862 II by not counting Encke's comet), which seems thus to have such interesting relations with the August meteors, was discovered on the evening of the 18th of July, 1862, by Mr. H. P. Tuttle at Cambridge, Mass., and a little later on the same evening by Mr. Thomas Simons at Albany, N. Y.* It was first seen in Europe on the 22d of July, and remained visible more than two months. At its brightest its nucleus was equal to a star of the second or third magnitude, and its tail, according to some observers, was as much as 25° in length. The changes that took place in the coma and tail were quite remarkable, and were carefully observed. They will now possess a double interest.

The telescopic comet 1866 I was discovered by Tempel on the 19th of December, 1865, and was visible about a month. Its minimum distance from the earth's orbit was 00660, about two and a half times the distance from the earth to the moon. This distance for Tuttle's comet is 00472, or about 430,000 miles.

4. Age of the November group of shooting stars.

In the Paris Academy of Sciences, Jan. 21st, 1867, LeVerrier spoke of the November meteors (Comptes Rendus, lxiv, 94). Inasmuch as the group is not a complete ring, he argues that it is of comparatively recent formation, having come into the solar system, and been turned into its present orbit within a few centuries.

Now a body coming from a great distance and so having a great velocity in the vicinity of the earth could not be thrown into an orbit nearly circular by the feeble action of the lower planets. Computation leads to this result, which is fully confirmed by the fact that the swarm passes every 33 years near the earth and yet returns at regular intervals.

Assuming then an orbit whose period is 331 years, whose perihelion distance is 0.989, viz., the earth's distance from the sun on the 14th of November, and assuming the position of the radiant to be long. 142°, N. lat. 81°, he computes corresponding elements.

The group, when it came into the system, could not be thrown into its present orbit except by a powerful perturbing cause, as was the case with the comet of 1770. Moreover, comets so acted upon that the newly acquired orbit has a small perihelion distance, return necessarily to the orbit of disturbing body, just as the comet of 1770 returned to Jupiter. We cannot help then being struck with the circumstance that the November group extends to the orbit of Uranus and a very little farther; and

Mr. Swift, of Marathon, N. Y., claims to have seen it two or three days earlier, but he made no announcement of the discovery, supposing it to be another comet.

that these orbits intersect, very nearly, just after the group passes its aphelion, and above the plane of the ecliptic.

The question then arises whether the group and Uranus have ever been together at this point. By calculation it is found that no such meeting could have taken place since the year 126 of our era, and that by a change of the computed node for that epoch by 1° 48', and by placing the perihelion 4° from the descending node in November, the group would then actually strike the planet Uranus. These two changes are not greater than the possible errors of our observations.

LeVerrier's researches farther show that a globular group onethird of the diameter of Uranus (more or less) might at that time have been then thrown into a shape and an orbit which should by this time give all the phenomena of the November meteors. Its previous orbit might have been an ellipse, a parabola, or a hyperbola. Its motion might even have been direct in an elliptic or parabolic orbit.

In the course of future time, he argues, the phenomena will extend over a larger and larger number of consecutive years, diminishing, at the same time, in intensity. But no change in perihelion distance will make them disappear entirely. Even if this group again meets Uranus, the planet can act only upon a part of its matter, and cannot throw it all into a new orbit as Jupiter did the comet of Lexell.

These reasonings of Schiaparelli and LeVerrier have certainly great force, and make it probable that of the five possible periodic times of the November meteors (this Journal, xxxviii, 57), that of 33-25 years is the true one. The strongest objection to this conclusion is that the radiant in November does not seem to be a point, but rather a small area. This area cannot be of great breadth in latitude since the ring is only twenty-five or fifty thousands of miles in thickness. If, as observations seem to require, the radiant extends in longitude two or three degrees only, then the lines of apsides of the orbits of the several members of the group differ considerably. In this case it is more reasonable to suppose the orbits themselves grouped about an exact circle than about a long ellipse.

If upon examination it shall be found that the center of the radiant area was decidedly more or less than 891° from the sun, on the morning of the 14th of November last, then this objection will lose much of its force. Again, if there shall not be seen on the morning of May 12th, 1867, between 1 o'clock and dawn, a few scattering members of the November group, radiating from a point 180° from Leo, this also will tend, to a certain extent, to strengthen Schiaparelli's reasonings.

ART. XXXII.—Observations upon the Drift phenomena of Southwestern Iowa; by C. A. WHITE, M.D.

[In advance of his final report upon the Geology of Iowa.]

IN the year 1858 I discovered distinct glacial scratches upon an exposed layer of the Upper Burlington limestone (Subcarboniferous), and made full notes and drawings of the same, which having been unfortunately destroyed by fire within a year afterward, no account of the observations was ever published. No opportunity has since presented itself to verify those observations, but I think I am not mistaken in the recollection that there was but one set of scratches, which were straight, distinct, and rather numerous; having a direction south, about twenty-two degrees east. This, so far as I am aware, was the first observation of glacial scratches upon rocks in situ in the state of Iowa, although boulders with similar scratches upon them are often seen in various parts of the state.

During my official labors last season, although considerable attention was given to the drift deposit, no similar traces of gla cial phenomena were ever discovered until I reached the Missouri river in Mills county, where, on section 16, township 71, range 43, west of the fifth principal meridian, very distinct glacial scratches were found upon limestone of the Upper Coalmeasures, not far from the middle of the series.

The locality is upon the western abrupt slope of the bluffs which border the bottom land of the Missouri river. The river being distant nearly three miles to the westward, the exact height of the scratches above it was not definitely ascertained, but it is probably not much less than one hundred feet above the ordinary stage of water. About four feet in thickness of ordinary drift material rested upon the striated surface. This had been partially removed by the quarrymen, exposing the scratches to view. Resting upon this light deposit of drift, and sloping upward to the high lands, are about one hundred and fifty feet of that peculiar lacustrine deposit called by Dr. Owen "siliceous. marl," and by Prof. Swallow the "Bluff formation," which deeply covers the drift and underlying rocks of this region, except where they have been exposed by fluviatile denudation.

The boulders and pebbles contained in the drift material of this locality were both granitic and metamorphic. They were well rounded and worn, and striated faces were observed upon quite a number of them, thus as nearly as possible detecting them in the very act of scoring the rocks in situ.

The scratches here are in two sets, a coarser and a finer; those of the latter more numerous than the former, but those AM. JOUR. SCI.-SECOND SERIES, VOL. XLIII, No. 129.-MAY, 1867.

of both sets being perfectly parallel with their fellows, distinct and straight. The surface of the rock had been ground level and smooth, removing all unevenness of the natural bedding surface. The directions of the striæ were determined by a very good pocket compass. That of the coarser set (No. 1) was found to be S. 20° E., and that of the finer set (No. 2) S. 51° E. No allowance was made in either case for the variation of the magnetic needle, which the local surveyors calculate at about eleven degrees east of north.

At an exposure of the same limestone one mile below Omaha, the capital of Nebraska, immediately upon the right bank of the Missouri river, and only some six or eight feet above the ordinary stage of water, other scratches of a similar character were observed. They were found upon the upper surface of a firm layer, which the workmen had exposed and were removing for building purposes. Here, however, there is but one set of scratches, their direction being S. 41° W. (set No. 3), not allowing for variation of the magnetic needle. The surface of the rock is nearly or quite level, but the roughness of the natural bedding surface has not been entirely removed, yet the stria were so distinct that no difficulty was found in ascertaining their true direction.

The drift at this locality is principally composed of a dark colored, stiff, clayey material, intermixed with sand, gravel and boulders, and varies from one foot to eight feet in thickness. Upon this rests the bluff formation as before described. The face of the bluff at the locality where the latter observations were made is nearly at right angles with the direction of the striæ.

Considering that the whole region surrounding these localities, and for a long distance to the northward of them, is an entirely open country; that the present prominently uneven features of the region had their origin at a period subsequent to the drift; and the extreme simplicity and uniformity of the strata over which the glaciers must have moved, the direction of their currents which these scratches apparently indicate seems very remarkable. We not only see at the Mills county locality that the scratches upon one and the same surface prove that two separate currents existed there during some portions of the glacial period, having a divergence of thirty-one degrees with each other, but also, only about twenty miles to the northward, we find the evidence that another current moved in a direction. which formed an angle of ninety-three degrees with one of those in Mills county.

Since we see no evidence of the cotemporaneous existence of obstructions which might have deflected the current of a regular southerly moving glacier, and thus have produced the scratches

in the various directions shown, it seems necessary that we should seek for some other explanation of them. Observations thus far made certainly afford very inadequate data upon which to base a definite theory concerning the real direction of glacial currents over this part of the continent, but the coincidence of the direction of those scratches which have been observed with the general course of the drainage of the region in which they occur, is worthy of careful consideration. By reference to the mention of the locality, near Burlington, at the commencement of this article, and to a map of Iowa, it will be seen that the direction of the scratches observed there, coincides pretty nearly with the general direction of the drainage of the eastern watershed of the state.

That the close of the drift period left the surface of our state unmarked by strong features, and with shallow, longitudinal depressions which gave initial direction to the courses of the streams, and that these subsequently cut out their own valleys by erosion, there seems to be no reason to doubt. It seems not improbable, also, that these initial depressions in the surface, whether primarily caused by flexures of the earth's crust, as Whitney has suggested, or not, may be regarded to some extent as indices of the general direction of ancient glacial currents.

There is another interesting matter in connection with these observations of drift phenomena along the Missouri river, and the existence of the important lacustrine deposit there. The close of the Drift period evidently left a large depression of the general surface in the region, a portion of which is now occupied in part by the counties of Fremont, Mills, Pottawatomie, Harrison and Monona. This depression became a large freshwater lake, the borders of which have not yet been definitely ascertained, but no satisfactory evidence of its existence eastward of the East Nishnabotany river has been observed in Iowa. The Missouri river evidently emptied into this lake, and flowed from it, until it became filled with the peculiar deposit of fine, siliceous, marly material, known as the Bluff formation, the character of which is very much the same as that of the muddy material which would now be deposited from the waters of the river if it were possible to throw a permanent obstruction across it. As the valley of the river was gradually deepened during the Terrace epoch, its waters found no difficulty in sweeping out the fine homogeneous material which they had before deposited, leaving those high peculiar bluffs upon each side of its broad bottom. Fresh-water and land shells of existing species, principally Gasteropods, are often found in this deposit, from base to top, and its thickness sometimes reaches more than two hundred feet above the drift material upon which it rests.

Seventy or eighty miles to the eastward of the Missouri river,