ture of the climate. It ranges from 37° to 42° F., never rising above the latter temperature excepting in shallow places near shore. The average depth of the lake is estimated by Bayfield at 900 feet. Its height above the sea is 600 feet, hence its bottom is 300 feet below sea level.

The shores of this lake are much more elevated than those of the other great lakes, and high table lands extend far back into the interior and are thickly wooded. The coast, especially on the north side of the lake, is abruptly precipitous directly to the water's edge. And the air on the surface of the lake rarely is of a higher temperature than 50°, while that in the forest at noon is frequently as high as 90°, or even 94°. It is obvious, then, that during a summer's day the air in the forests becomes highly rarified by heat, and takes up a proportional quantity of water in the state of invisible moisture. When this current of warm air slides from the precipices over the surface of the lake, the warm air by its specific levity from rarification floats upon the cooler air of the lake, and does not directly mingle with it. The consequence necessarily is, that a film of moisture is condensed at the surfaces of contact of the warm and cold air, and thus a screen is produced on which the objects reflected from below are seen as in a mirror. Meanwhile, by refraction, this image is seen higher up than it is really painted on the mist. This was obviously the cause of the strange phantoms which we have witnessed on Lake Superior.

It is no uncommon thing on other parts of the lake to see vessels inverted in the air before their hulls become visible above the horizon; and it is well known that similar appearances very rarely occur on our seacoast, and have given rise, in former times, to strange and superstitious tales.

Prof. AGASSIZ mentioned an additional phenomenon which has been frequently witnessed by himself and his party upon Lake Superior. Not only did the shores and islands, with all their vegetation, appear repeated, higher up and in an inverted position, but that above this inverted landscape, there was sometimes still another, in which every thing was upright, so that the picture was twice repeated above the surface of real nature once inverted, and above that, the same erect. This fact must be explained by any theory which professes to account for similar phenomena; but it may be simply the image of the landscape, inverted upon the surface of the lake, reproduced with the inverted image of the landscape itself.

Dr. JACKSON remarked that he had observed the same thing frequently; but, by some oversight, he had made no minute of it.

Prof. ADAMS exhibited specimens of polished shells, showing that they were susceptible of a very high polish, by which their colors would be brought out, and they would be much more ornamental.

REMARKS ON THE DISTRIBUTION OF THE TESTACEOUS MOLLUSCA OF JAMAICA. BY PROF. C. B. ADAMS.

THE great number of species is remarkable. A few miles of coast, without the aid of storms, and without dredging, yielded 420 species. In the small bay of Port Royal, 350 marine species were found. A pint of sand, taken from a surface three yards long, contained 110 species. In Manchester, a district four miles by one mile, contains 100 terrestrial species. A small part of the island afforded 220 species of land shells. Probably there are 350 or 400 species of land shells, and two or three times as many of marine species. Extensive districts occur, however, which are nearly destitute of land or marine shells. They are accumulated in favorable stations.

The difference in the extent of the distribution of the marine and of the terrestrial species is remarkable. A majority of the marine species are known to occur in the other islands; probably not more than 10 or 15 per cent. of them will be found to be peculiar to Jamaica. But of the land shells, 95 per cent. are peculiar to the island.

The limited distribution of the terrestrial species is remarkable. A few are generally distributed, but a large number are limited to districts of a few miles in diameter, and several, although occurring abundantly, could be found only within the space of a few rods.

Only 17 fresh water species were found. Seven belong to Planorbis, and the other ten to nine genera. There are no Naiadae, and but one very small and rare Cyclas was found. Favorable stations for fresh water species are rare.

In respect of the number of individuals of Mollusca in Jamaica, as compared with more northern latitudes, the rule so obvious in the class of fishes is not applicable to the same extent. Of fishes the species are much more numerous, but the individuals much less so. Of the Mollusca, the total number of individuals is about the same as in this latitude, and the number of species represented by a profusion of individuals is about the same. But the number of species not

occurring abundantly is much greater, so that the average of individuals to all the species is less than in this latitude.

From a comparison of the laws of distribution of the marine and terrestrial species in the Antilles, it follows that the number of the latter must exceed that of the former.

With the insular distribution of the terrestrial species may be associated the fact that the coral reefs are all fringing, for both facts are connected with the geological fact that these islands are in a process of elevation.

ON THE RIVER TERRACES OF THE CONNECTICUT VALLEY, AND ON THE EROSIONS OF THE EARTH'S SURFACE. BY PRESIDENT EDWARD

HITCHCOCK.

I OUGHT perhaps to say that my paper consists of suggestions and a few facts, or numerical results, rather than any finished theory. In fact, I follow my field notes chiefly, which have not yet been completed or reduced. I had hoped, ere this time, to present a carefully prepared paper upon these subjects, having been interested in them for twenty years, although I have never, until the past year, carefully measured the height of these terraces. I will be as brief as possible, as my leading object is, to elicit from others deductions from the facts which I have observed; and hope I shall be pardoned for presenting the subject in so crude a state.

I have examined the Connecticut River almost from its mouth to Turner's Falls; having carefully measured the heights of the terraces. But these measurements I shall not introduce, in detail, till I have carried the work farther towards the source of the river, which I hope to do soon.

During the last year, a work of considerable size has been published by Mr. Chambers, of Edinburgh, who considers most of the terraces in Great Britain as ancient sea-margins, which have become elevated to their present position by successive uplifts of the continent. Now, I suppose that our terraces are of three kinds; 1. The ancient sea-margins; 2. The margins of ancient fresh-water lakes; and 3. River-margins.

I have undertaken to examine the terraces of the Connecticut val. ley, which extends some 300 miles, nearly north and south, embracing the Connecticut and its tributaries. As you approach the river, you find plains of sand, gravel, or loam, terminated by a slope, sometimes as steep as 35°, and a second plain, then another slope and

another plain, and so on, sometimes to a great number. Sometimes, upon the other side of the river, you can perceive similar terraces, which seem, at that distance, to correspond in height with those upon which you stand, but which is not exactly the case.

I find that these terraces occur in successive basins, formed by the approaches of the mountains upon the banks, at intervals. Sometimes the basin will be 15 or 20 miles in width, but usually, much narrower; and it is upon the margins of these basins that the terraces are formed. The first basin which I examined, extends from Middletown northward to Mount Holyoke, where there is a barrier through which the river seems to have cut. North of that you have another basin which extends from Holyoke to Turner's Falls; another which extends to Brattleboro; another to Bellows' Falls; another to Charleston; another to Windsor, &c.

On the tributaries of the Connecticut River are several very fine terraces, and especially upon Deerfield River and Westfield River are remarkable ones which will throw a great deal of light upon the manner in which these terraces are formed.

I have rarely found terraces more than 200 feet above the river; which would be, in Massachusetts, about 300 feet above the ocean, and at Hanover, New Hampshire, about 560 feet. Nowhere do they exist, along any river, unless that river has basins; unless the mountains close down upon the river, and it has cut through the barrier at some time or other. In such basins, if they have considerable width, you will find terraces. As to the materials of which they are formed, they appear exceedingly artificial. The outer, or highest terrace, is generally composed of coarser materials than the inner terraces. They are all composed of materials which are worn from the rocks; but the outer terrace oftener is full of pebbles, some of them as large as 12 inches, while the inner seem reduced to an impalpable powder, like the soil of a meadow which is overflowed during high water.

Whence did these materials originate? How did these valleys fill up to the height of 200 feet in many places? It must have required an immense amount of materials, brought in by some agency or other; for the materials were first worn from solid rocks, and afterwards brought into these valleys. The outer terrace appears to have been often in part the result of the drift agency. Afterwards, the river agency sorted the materials, and gave them a level surface, the successive basins having at that time barriers. The inner terrace appears to have been, at least, in its upper part, the result of deposi

tion from the river itself. It is composed of fine sand and clay, or loam. Such, for the most part, I conceive to have been the origin of the materials.

I will now mention a few of the facts which I have observed.

First, the terraces do not generally agree in height upon the opposite sides of the valley. I do not find any such correspondence, except in the case of deltas at the mouth of the smaller streams. The higher ones oftener agree, perhaps, than the lower ones. To an eye unaccustomed to judge of heights, a difference of 20, 30, or 40 feet, may be imperceptible at the distance of a mile or two; but measurement corrects the error. If formed, as I suppose, from the rivers, we should not expect them to agree in height upon the opposite sides of the river.

Another fact is, that the terraces slope downward in the direction of the stream. The same terrace which, near South Hadley, is 190 feet above the river, slopes until, at East Hartford, it is only 40 feet above the river, thus sloping 150 feet more than the slope of the river itself, in a distance of 40 or 50 miles. On the smaller rivers there are terraces which slope so rapidly that you can see it distinctly. This is a very important fact, as it shows that they cannot have been formed by the sea; for, in that case, they would have been horizontal; nor could they have been formed by a lake, for the same reason. It proves then, that they were formed by the river. The greatest number of terraces which I have observed upon the rivers is eight or nine; generally, there are but two or three.

I

agree with my friend, Mr. Dana, who has published some very able articles upon the terraces, in the American Journal of Science, in almost every thing that he has advanced. I agree with him in the position which he has taken against that of Mr. Chambers, that we must require some decisive evidence in the materials besides their arrangement of a marine origin, before we adopt the conclusion that the sea produced them, and deny the agency of the river. Now in this whole valley of the Connecticut, 300 miles in length, in all these deposits, not a single organic relic of any kind has yet been discovered. The only things that look like organic remains, are two kinds of concretions; the most remarkable of which are clay-stones, which I believe the geologists of the north of Europe have considered as originating in molluscs, with which opinion I do not at all agree; they are formed, I think, as crystals are, rather than as the petrifaction, by which I mean the substitution of mineral for animal matter. I do not