is the general mode, and one that is subject to some modifications, there is another mode by which this difference of level is produced. Miners often find the strata, any distance from a foot to more than 40 fathoms asunder, in point of altitude, on opposite sides of bended-tabular rents, in places where these strata are not bent; but straight, as the stratum a b c d, Plate XXXVIII, fig. 1, and the stratum c, along with the strata B A, fig. 2, are represented. In the first mode, the distances which the strata are asunder, on opposite sides of these rents, are obtained by the bending of the strata, and always continue, and sometimes increase, during many temporary suspensions of this bending; but in the second mode, it takes place without any bending of the strata. I will give one example to illustrate each mode. Let fig. 1 be representative of the first mode. The stratum ad is horizontal, notwithstanding the part ab is the distance b c below the part c d. Above this stratum, at the rent A, the strata are straight for a certain distance, and then they are bent; and below this stratum they are also bent, at first slightly, but with a gradually increasing ratio, that reaches its maximum at the stratum kn; whose two parts kl and m n are the distance m l asunder, which is equal to the distance fg, and to the distance bc, and which is acquired by the bending of the part m n above the line 7 n. The strata close to the side pg cat this part of the rent are thicker than close to the side lfb: they are also thicker at mp than at qn, and that additional thickness throughout the whole of the rent, below the stratum kn, gave rise to the bending of the part m n of that stratum, in the manner which has been shown in my first communication on rents; but the strata above the stratum m n where close to the side m c of the rent, though thicker than the opposite strata on the side lb, are not so thick as they are at the line n d; in consequence, the bending, as seen at m n, gradually decreases upwards till it ceases. Let us take, by way of illustration, the effect on the stratum c h, of this alteration in the thickness of the strata: as much as the strata which are situated between the strata kn and ch are thicker at the line nh than at ke, so much is the distance n h greater than the distance ke, (say by the distance k o,) and so much is the stratum g h bent less than the stratum m n, say by the distance ir. The bending of the strata above the stratum c h also diminishes upwards, from the same cause, till it ceases at the stratum a d, which is straight. At first sight the position of the stratum a d, considering how much one part is higher than the other, appears to be irreconcileable with that arrangement which I have considered the general one; but when its connexion with that of the strata below is traced as we have now done, its difference from that which is the common one is easily accounted for. In fact, the arrangement of the elementary matters in this part is such, that the strata have contracted less, instead of more, at the line nd than close to the side m c of the rent; and by doing so have gradually given the straight, instead of the bended position to the strata of this part: while these strata close to the rent have contracted, as usual, less on the under side pc than on the upper side lb. Hence also the distance between the straight strata on opposite sides of this rent is equal to that between similar parts of the bent strata. An example of the second mode is that which follows. Sometimes the strata near bended-tabular rents are all, or nearly all, straight, notwithstanding they are situated at different levels on opposite sides. Thus the strata A B C, fig. 2, and those which lie between them, are straight on both sides of the rents DE; but the parts of the strata bf k and e in are higher on the under sides than the parts cgl and d h m on the upper sides of these rents. The strata in this figure, as well as those near all rents of this shape, are thicker on the upper than on the under sides; and by this greater thickness the stratum A is higher at be than at c d, the stratum B at fi than at g h, and the stratum C at k n than at 1 m. Now, as has been before shown, this difference in the thickness of the strata is a consequence of the unequal contraction of the stratified matter; that is to say, the strata have contracted more near the upper sides than near the under sides of these rents. But although they have contracted with different ratios on different sides, yet in the example before us the ratio on any one side has been uniformly the same throughout the strata, instead of being, as in general, the least near the rents, and the greatest at given distances from them. In consequence, then, of this uniformity in the ratio of contraction of the strata, when taken on one side only, they are straight on both sides of some rents, although they are situated at different levels on opposite sides of such rents. It may be proper to remark here that, though the strata are straight, and higher on one side of a rent than on the other, when seen in a cross section, as in fig. 2; yet when a view is taken at right angles to this section, or when a person faces the rent, every stratum then separates at one horizontal extremity into two parts, one inclining very gently upwards, and the other downwards, till opposite the middle of the rent; then the higher part dips downwards, and the lower part rises upwards, till they meet again at the other horizontal extremity of the rent. 2. Observations on the Upper Extremities of large Bended-Tabular Rents. The upper extremities of some rents are altogether situated in the solid rock, and at considerable distances below the surface. Many large rents extend downwards from the surface of the solid rock, or that of the solid strata, to great depths; but some of them reach above the solid, through the alluvial matter, to within a few inches of the earth's surface. Some of the rents which reach nearly to the surface are precisely of the same dimensions in the alluvial clay, as in the solid rock below and what is more remarkable in the lower half of the alluvial clay, they are sometimes filled with spar and the usual contents of the rents, and in the upper half, with clay deeply tinged with iron; and sometimes opposite the whole height of the alluvial matter they are filled with iron tinged clay: in both instances the rents are covered with only thin strata of soil. In the Shropshire and Cumberland coal formations I have seen rents so circumstanced at the earth's surface; and at Lead Hills in Scotland, in company with Mr. Martin of that place, I met with two such rents, that are situated in the north side of the valley and to the west of the Susannah vein. Other rents that reach to only a few inches below the surface are as wide in the alluvial clay just above the top of the hard rock as they are below; but upwards, they increase in width in such a ratio that each side deviates from 20° to 30° from a perpendicular line. Opposite the alluvial matter they contain clay, mixed throughout with large cobbles, which last are very numerous at the bottom. The contents in these parts appear as if they had been washed into the rents. I have seen such rents in Cornwall. Rents reaching through the alluvial matter exist most abundantly in low and smooth mountainous districts, such as Cornwall and Lead Hills. The existence of rents in alluvial matter, though new to men of science, is a very important fact. It shows us that the alluvial matter must have been formed before these rents; otherwise, after reaching the surface of the present rocks, the rents could not have passed through the alluvial matter. It also shows us that the alluvial matter was formed from the matter below, when this matter was the least able to resist a disintegrating force: and by it we know that the alluvial matter has not been removed since then. Hence the rocks or strata underneath such parts have not been in the least wasted by the elements. IT. ON STRATIFICATION. I have said that the phenomenon of stratification, in one point of view, is an effect of the unequal contraction of the earth's matter. I will now give my reasons for this assertion. But perhaps it may be previously necessary to give a definition of the term. Stratifica tion consists in that assemblage of tabular masses, wherein any one mass is parallel to that next above, and to that next below it. A formation that is entitled to be called stratified must have this arrangement of parts every where. According to this definition, all, or nearly all, the red and white sand-stone, and some of the limestone formations, are stratified; but the formations of granite, micaslate, &c. are not stratified, unless they lie in hollows, as they sometimes de, on the primitive and unstratified mass. Mountains divided, in a few places, into tabular distinct concrétions, have sometimes been called stratified; but to possess this structure they must be every where divided into tabular masses, which have the |