troit it has been sunk completely through, and found to have a thickness of one hundred and eighteen feet.

This lower deposite of clay, in the southeastern part of the state, is usually overlayed by a stratum, varying from one to five feet in thickness, of an exceedingly fine marly clay. This clay, when sufficiently free from lime, is well adapted to the manufacture of bricks and earthern ware.

On the northwestern side, bordering on Lake Michigan, the upper clays are much more largely developed than upon the more southeasterly portions of the peninsula. These deposites of clay alternate with beds of sand and gravel, the whole sometimes attaining a thickness of from one hundred to four hundred feet. The separate beds vary considerably in character; the upper usually containing a much larger proportion of lime than the lower ones, yet they usually agree in possessing an extreme fineness of texture. Many portions of these clays appearing in the abrupt shores of Lake Michigan, are well adapted to the manufacture of bricks and earthern ware, but they usually contain so large a proportion of lime as to render them unfit for use for those purposes. Some portions of these clays, in which lime enters largely as an ingredient, rather deserve the name of marls, and they are admirably adapted for use upon the sandy lands of the northern part of the peninsula.

Shell Marl.

Several beds of shell marl were noticed upon the northwesterly side of the peninsula; and upon the eventual settlement of the country they will prove of great value to the agriculturist, as well as for the manufacture of lime. In consequence of the unsurveyed condition of the country, it is impossible, at this time, to designate the localities.

White river of Lake Michigan takes its name from the occurrence of a bed of shell marl, of a very white color, directly at i's mouth. The marl composing this bed would appear to have been deposited in an old channel of the river, which had been shut up by the action of the winds and waves upon the sand at its mouth, and afterwards to have been buried many feet in depth by drifting sands. In process of time the river returned to its former place of embouchure, thus laying bare the marl in question. It will prove a valuable material for the manufacture of lime, as well as for application to the light sandy lands in the immediate vicinity. The bed is not extensive.

Gypsum.

Gypsum occurs associated with the northern limestone, but for the most part under circumstances that will effectually prevent its being obtained in any considerable quantities.

Gypsum of a beautiful white color occurs in the bed of the lake a little north from Point au Grais river, but to what extent it is impossible to determine, for it is covered by several feet of water, which will effectually prevent the working of the bed.

On the St. Martin's group of islands, near Mackinac, gypsum also occurs, chiefly in loose pieces, scattered over the islands. A bed of gypsum is said to be associated with the limerock in the immediate vicinity of these islands, and in such a situation that during the low stages of water, it appears above the surface; but at the time of my examination it was covered by several feet of water. I am informed that some years ago several ship loads of gypsum, collected in loose masses upon the St. Martin's islands, were transported to the lower lakes. Nearly all which appeared upon the surface has been removed, and the low level of the islands will effectually prevent any considerable explorations for

more.

Gypsum also occurs on the northern peninsula, between Green Bay and Mackinac, but to what extent has not yet been determined. Small quantities have also been collected and shipped from this part of the coast.

Change of Elevation in the Waters of the Great Lakes.

Intimately connected with the geological changes which are taking place from the deposite of detrital matter at the mouths of streams, and in the deeper portions of the lakes, together with the degradation of the lake and river coasts, are the changes in the relative level of the waters of the lakes; a subject to which the attention of our citizens has been more particularly called within the past two years.

The great interest which this subject possesses in connection with our lake harbors, as well as with those agricultural interests situated upon the flat lands bordering the lakes and rivers, may be a sufficient apology for the introduction, in this report, of the accompanying facts and reflections upon the subject. An accurate and satisfactory determination of the total rise and fall of the waters of the lakes is a subject, the importance of which, in connection with some of our works of internal improvement and harbors, can, at this time, scarcely be appreciated.

Much confusion is conceived to have arisen, in the minds of a portion of our citizens, in consequence of a confounding of the regular annual rise and fall to which the waters of the lakes are subject, with that apparently irregular elevation and subsidence, which only appears to be completed in a series of years; changes that are conceived to depend upon causes so widely different that while the one can be calculated with almost the same certainty as the return of the seasons, the other can by no means be calculated with any degree of certainty.

It is well known to those who have been accustomed to notice the relative height of the water of the lakes, that during the winter season, while the flow of water from the small streams is either partially or wholly checked by ice, and while the springs fail to discharge their accustomed quantity, the water of the lakes is invariably low.

As the spring season advances, the snow that had fallen during the winter is changed to water, the springs receive their accustomed supply, and the small streams are again opened, their banks being full in proportion to the amount of snow which may have fallen during the winter, added to the rapidity with which it has been melted.

The water of the lakes, in consequence of this suddenly increased quantity received from the immense number of tributaries, commences rising with the first opening of spring, and usually attains its greatest elevation, (at least in the upper lakes,) some time in the month of June or July. As the seasons advance, or during the summer and a large portion of the autumnal months, evaporation is increased, and the amount of water discharged by the streams lessened, in consequence of which the water of the lakes falls very gradually until winter again sets in, when a still greater depression takes place from the renewed operation of the causes already mentioned.

The extreme variation in the height of water from winter to summer is subject to considerable change, according as the winters may vary from cold and dry to warm and wet; but during the past eight years, it may be estimated at two feet.

This annual rise and fall of the water of the lakes, dependent as it manifestly is, upon causes which are somewhat uniform in their operation, must not be confounded with that elevation and depression to which the waters are subject, independent of causes connected with the seasons of the year. These latter changes which take place more gradually, sometimes undergoing but little variation for a series of years, are least liable to be noticed, unless they be very considerable; but with respect to consequences they are of vastly more importance, since they are subject to a larger and more permanent range.

That the waters of the lakes, from the earliest settlement of the country, have been subject to considerable variation in relative height, is well known. At one time the belief was very general that these changes take place at regular intervals, rising for a space of seven years, and subsiding for a similar length of time; a belief which would appear to be in consonance with that of the Indians upon the peninsula, and with whom it no doubt originated. It is not wonderful that a subject, the causes of which are so little comprehended by our natives, should be invested with an air of mystery, or that an error once propagated (in consequence of the

long series of years required to bring about any considerable change.) could scarcely be eradicated.

While the idea of the septennial rise and fall must be regarded as founded in error, it is nevertheless true that from the earliest records, the height of the lakes has been subject to a considerable variation, usually rising very gradually and irregularly for a series of years, and after this falling in a like manner.

Our old inhabitants agree in stating that the waters were high from 1800 to 1802; in proof of which it is stated that the roads which had before been in use upon the banks of the Detroit river, were so completely inundated as to be rendered impassable A similar circumstance is related to have occurred in the vicinity of Chicago, a broad sandy beach forming the immediate shore of the lake near that place having been wholly overflowed.

I have been unable to obtain authentic information respecting the changes which took place between the years just mentioned and 1814, but from the latter year to the present time, we have a more connected series of facts relating to the subject.

"It is now a matter of record, that in 1814 and 1815 the Detroit and St Clair rivers were unusually high; that the foundations of the houses, and much land that had long been under dry cultivation, were submerged. These buildings had been erected many years before, and of course under the belief that they were aloof from all but extraordinary and temporary inundations. No observations appear to have been made upon the progress of the elevation, whether it were gradual or abrupt, or whether there were any preceding seasons of a character to produce it.

"In 1820, or about that time, the rivers had resumed their usual level. Several wharves were built at Detroit, between that year and 1828, at a height, as was supposed, sufficiently above the general level for all purposes of convenience and safety. At the latter date the rivers had again attained the elevation of 1815, and remained so until 1830, with only such occasional depressions as might be caused by strong winds, being nearly upon a level with the wharves."*

From 1830, when my attention was first drawn to this subject, to the present year, I have been enabled to make a somewhat connected series of observations, under circumstances peculiarly favorable, having during that time followed the complete line of coast, from the foot of Lake Huron to the head of Lake Superior by canoe, and having traversed portions of the coast several times, thus being enabled to renew observations at points where they had been previously made. During the time of these examinations, I have been enabled to fix, with a considerable degree of

*The above extracts are from the pen of Col. Henry Whiting, U. S. Army, and their value is much enhanced from the fact that they embrace only such portions of the subject as were the result of his personal observation.

certainty, upon the height at which the waters of the lakes stood in 1819 and '20, when they were at their lowest level; a step which was conceived to be one of the first necessary in determining the complete range between high and low water.

For the last two years my attention has been more particularly called to the coasts of lake Huron and Michigan, and I feel confident in asserting that the water of these lakes has, during the last year, (1838,) attained a greater elevation than has before occurred in a very great number of years; a fact which is conclusively shown by the renewed degradation of banks covered with debris, that had long remained undisturbed, as well as by the great number of forest trees, sometimes covering many acres of ground, that have been destroyed in consequence of inundation. Many of these forest trees may be estimated to have attained an age of from one to two centuries.

In order to arrive as nearly as possible at correct conclusions as to the variation in the height of the water of the lakes from 1820 to 1838, I have carefully compared my own observations with those contained in an invaluable register, kept in this city by Col. Henry Whiting, U. S. Army, as also with the valuable data contained in the report of the State Topographer, hereto appended. It should be noted that the height of the water in the Detroit river is much more subject to fluctuation from slight causes, such as the effects of the winds and ice, than that in the open lakes: causes for the operation of which, it is sometimes difficult, if not impossible, to make the proper allowance. In fact, slight causes are productive of such changes as to render it absolutely impossible to arrive at accurate conclusions, except by simultaneous observations, made at points widely separated.

Assuming June, 1819 and '20, as zero, or the point of low water, the following table will not vary very far from an accurate statement of the relative height for several of the subsequent years.

In examining this table of relative heights, it should be borne in mind that this estimate does not include the regular yearly variation to which the waters of the lakes are subject. The estimates, it will be seen, are made from June of each year, or that month in which the waters are invariably high; but it is conceived the result would not be varied were the calculations made from any other month in the year, provided the same month were