conceal the flexures, and disguise the true order of superposition of the rocks.

Waves of the Crust, both Straight and Curvilinear.

In the much corrugated belts, the crust waves are both straight and curvilinear. In the Appalachians, there are groups of both these classes, retaining their special features throughout their entire length, which, in some instances, exceeds 100 miles. Some of the crescent-shaped waves present their convex curvature towards the region of maximum dislocation and metamorphism, while other groups are concave toward the same quarter. These different systems of waves seem to have been generated some of them from straight, others from curvilinear fractures in the earth's crust.

The Appalachian chain, regarded in the light of a long zone, or chain of groups of parallel, straight, and curving waves, consists of eleven sections, six of which are straight, and five curvilinear, three of the latter form being convex towards the N.W., and two convex towards the S.E., the whole zone having a length of 1500, and a maximum breadth of 150 miles. Certain of the straight divisions have their anticlinal axes, or the crest lines of the undulations trending N. 15° E.; other divisions, theirs trending N. 70° E., while some of the curving sections of the chain show a deflection in the direction of their individual axes of as much as 40°. Indeed, in particular instances, the change of trend amounts to as much as 60°. So remarkable a bending without disruption, of groups of parallel anticlinals, seems incompatible with the inferences of some eminent geologists, who conceive that there prevails a general relation throughout the globe between the directions of the lines and the epochs of crust elevation; for we here find that the selfsame axis, generated throughout its whole length, not merely in one geological period, but in one brief interval of time, alters its direction to coincide successively with sundry of the different assumed systems of crust elevation.

GRADATIONS IN FLEXURES.

Every broad belt of undulated strata exhibits certain gradations in the form of its flexures starting from the side of maximum igneous action, as this is displayed in plutonic eruptions, or in dislocations and metamorphism. Crossing the zone, the flexures first met with are invariably of the closely plicated class, their axis planes dipping often at a low angle towards the igneous border. To these succeed more and more open waves, until, from being perpendicular, the steep far sides of the undulations become flatter and flatter in their dips, till at last they assume a slope equal and symmetrical with those of the gentler flanks. Parallel with this gradation is a progressive widening of the waves themselves, and a corresponding sinking or flattening down of the summits, until they finally disappear in imperceptible undulations. All these phenomena of gradation may be clearly discerned in every section across the Appalachian chain, traced from the S.E. towards the N. W., and a perfectly identical structure will be found to exist in the great plicated belt ranging through the Rhenish Provinces and Belgium. In truth, there is no great corrugated zone that does not display a similar law of gradation in its flexures, when these are properly traced and generalized.

FRACTURES IN UNDULATED ZONES.

Two classes of dislocations abound in all the belts of the crust,

where the strata are greatly undulated. The least conspicuous, but most numerous, are comparatively short faults, transverse more or less perpendicularly to the strike of the anticlinal and synclinal axes. These abound in the Appalachians and other corrugated mountain chains, and are a principal cause of the deep transverse ravines and mountain notches which intersect their ridges, and give passage to their streams. The more obvious dislocations are the great longitudinal ones coincident either with the anticlinal and synclinal axis planes, or with the steep or inverted sides of the anticlinals. A distinctive character of these great fractures is their parallelism to the axis planes, whether they are coincident with them or not. Many of the more extensive longitudinal dislocations of the Appalachians are traceable to the rupturing of the anticlinals along their most wrenched inverted slopes. These waves are entire at their extremities, but so broken along all their intervening portions as to present only one-half of the wave form, the other half being profoundly buried with inversion under the unbroken part. Generally, in these great dislocations, the gently-dipping uninverted slope of the waves has been shoved-in the inclined plane of the fault-forward and upward upon the other inverted and crushed half, and in some instances through a great distance.

The up-driven parts having been extensively removed by erosive action, the upper strata of the overturned buried half of the wave are seen to be immediately overlapped in nearly conformable altitude of dip by the denuded lower strata of the uninverted side. Similar phenomena of the plunging of newer formations under older ones, with approximately conformable dips, meet us continually in the Alps, and other much plicated districts, and can be demonstrated to have arisen from the same cause, the upward and forward propulsion of the uninverted halves upon the inverted sides of the anticlinal waves along the great sloping planes of dislocation, into which the flexures have snapped at the time of their sudden bending. These several laws of crust undulations, consisting of those which relate to the parallelism, form, gradation in distance, shape, and dislocation of the waves, are exemplified in detail in the paper, and by appeals to the phenomena of some of the more conspicuously currugated tracts of Europe. Viewing, as one such zone, the undulated districts of southern Belgium, the Rhenish Provinces, the Westphalian coal-field, the chain of the Ardennes, and the Hundsruck, Taurus, and Hartz ranges, and referring for proofs to the descriptions and maps of M. Dumont and other geologists, who have described these provinces in more or less detail, the author shows that this belt displays all the phenomena of structure and gradation described by him as so conspicuous in the Appalachians of America. Sections transverse to this region from S.E. to N.W. will be found to exhibit precisely the same succession, from closely folded flexures with metamorphism through steep normal waves, to broad, open, and approximately symmetrical ones.

The structure of the Jura chain of Switzerland likewise exhibits proofs of the same laws. There the crust waves closely resemble those of the Appalachians-the whole chain is composed of several groups of flexures, differing in their direction or strike; but the waves of each group display a remarkable parallelism among themselves. Very few of the flexures exhibit actual inversion of their steeper sides. It is remarkable that the steep slopes of the great waves of the Jura face the Alps; and those nearest the Alps, or on the borders of the valley of Switzerland, are more compressed than those on the far side of the chain,-their more inclined flanks, for example in the Weissenstein, dipping even perpendicularly, or a little past this, into partial inversion. This southward thrust of the crests of the Jura anticlinals would seem to imply a movement from

the north, and not from the igneous axis of the Alps, or probably from both quarters, at the period of the production of the flexures.

The Alps themselves show the same general structural phenomena as the other plicated zones described, but under more complex conditions. This much convulsed mountain system contains but few waves of the open or normal type, consisting, except in its outer flanks, of many very close plications of the strata. When these foldings are carefully studied and structurally connected with each other, the whole chain appears to be composed of two or more central parallel igneous crests, and each flank of these mountain ranges of a belt of closely compressed waves. Each of these plicated zones or Alpine slopes displays the axis planes of its flexures dipping in towards the centre of its own chain, the flexures nearest the igneous axis plunging at a lower or flatter inclination than those more remote. High in the slopes of the chain, where denudation has removed the largest part of the originally present upper formations, only the synclinal folds of these remain preserved. These are the so-called V's of the tertiary and jurassic beds, pinched in between the closely folded anticlinals of the gneissic, and other older rocks. The inward dip of nearly all the beds of both slopes of the Alps, thus occasioned by the completeness of the folding and the outward thrusting of the anticlinal parts of the flexures, is the obvious cause of that fan-like feature of dip of the entire chain, which has recently excited so much discussion among geologists. Cleavage of the rocks, and a superinduced crystallization parallel to the cleavage planes, contribute not a little, the author conceives, to the illusive appearance of a general inward dip of all the strata, even the newest, under the older formations of the high igneous crests of the chain; for both the cleavage planes and the crystalline foliation observe a very constant parallelism in the direction of their dip to the dip of the axis planes of the flexures.

Slaty Cleavage.

It is now a good many years since Professor Sedgwick and other geologists announced the important general fact, that the structure called slaty cleavage pervades the altered strata affected by it in directions independent of their bedding or laminæ of deposition; that these planes of cleavage are approximately parallel to each other over large spaces of country, however contorted the dip of the rocks; and that where the cleav. age is well developed in a thick mass of slate rock, the strike of this cleavage is nearly coincident with the strike of the beds. Professor Phillips, in 1843, added to this rule a still more comprehensive and exact expression-that the cleavage planes of the slate rocks of North Wales were always parallel to the main direction of the great anticlinal axes. Since 1837, these phenomena of the close parallelism of the cleavage planes with each other, and with the main axes of elevation, have been observed and recorded by Professor W. B. Rogers and the author of this communication; and in 1849 the author submitted to the American Association for the Advancement of Science a communication on the analogy of the ribbon structure of glaciers to the slaty cleavage of rocks, in which he stated what he deems the true law of cleavage of a district of undulated and plicated strata,-namely, that the cleavage dip is parallel to the average dip of the anticlinal and synclinal axis planes, or those planes which bisect the flexures. The generality of this rule was shown by sections exhibiting the flexures and cleavage in the Appalachians, in the Alps, and in the Rhenish Provinces. Subsequent observations in other localities have confirmed the universality of this law; and the recent description of the Devonian strata in the south-west of Ireland by

Professors Harkness and Blyth still farther tend to illustrate and establish it. In their paper in the Edinburgh New Philosophical Journal, (October 1855), they not only recognise an agreement between the strike of the cleavage planes and that of the several rolls (or anticlinals) which affect the island of Valentia, but they show, that while the cleavage dip is northerly, the anticlinal "curves have been pushed over in a more or less southerly direction," inverting the carboniferous limestones and coal measures. Their general statement is, that the cleavage structure of rocks does not result from the simple rolling of the strata, but from this cause combined with a considerable amount of pressure, and this latter force acting from the south, has pushed over the strata in a series of oblique curves to the north, and given to the inclined cleavage more or less of its southern dip. They further support the deductions of Mr Sharpe, "that there has been a compression in the mass in a direction everywhere perpendicular to the planes of cleavage, and an expansion of the mass along these planes in the direction of a line at right angles to the line of incidence of the planes of bedding and cleavage." But from this view of the mechanical nature and the direction of the force engendering cleavage the author of this communication begs leave to dissent.

A second general law is, that where the cleavage is fully developed, and the anticlinal and synclinal flexures are also conspicuous and very sharp, the cleavage planes immediately adjoining these bendings are not parallel to the axis planes, but radiate partially from them, in a fan-like arrangement, upward in the anticlinals, and downward in the synclinals. This aberration from the normal direction is, furthermore, not symmetrical upon the opposite sides of the geometric axis planes, but is usually greatest upon the inverted or steep sides of the waves.

A third prevailing relation of the cleavage planes is their tendency to deviate from the normal direction of parallelism to the axis planes, in order to conform partially to the direction or dip of the strata; and as in every belt of uniform flexures closely plicated with inversions, the uninverted, or normal dips, greatly exceed the inverted ones in breadth, there prevails a lower inclination in the planes of cleavage than belongs to the planes bisecting the flexures.

There is yet another law modifying cleavage, dependent upon the mechanical texture, and possibly the chemical composition, of the strata. In formations composed of alternations of the coarser mechanical rocks, such as siliceous grits and conglomerates, with the finer grained argillaceous beds, such as slates, shales, or marls, the coarser beds are unaffected by cleavage, while the finer-grained ones are often pervaded by it. Indeed, there appears a strict proportion between the degree of intimate fissuring of the rocks by cleavage and the degree of comminution of the particles. Connected probably with this interruption in the propagation of the cleavage, the author has observed another modification of the cleavage planes, -namely, that they tend to curve a little from the normal direction, in the finer-grained argillaceous beds, approximating to parallelism with the surfaces of bedding of the adjoining coarser mechanical deposits, as they approach them, showing in a transverse section, a kind of gentle sigmoid flexure. This fact is well illustrated in the cleavage-traversed rocks at the base of the anthracite coal-formation of Pennsylvania, where the red shales alternate with the lower beds of the coal-sustaining conglomerates and coarse sandstones. These remarkable facts seem sufficient of themselves to refute the hypothesis, somewhat in favour at present, of the purely mechanical origin of the cleavage-producing force; for we cannot conceive how a mechanical force either of compression, or of tension, transmitted, as necessarily it must have been, very equally, through pa

rallel layers of coarse and fine material, should have exerted no fissuring action the moment it reached the surface of the coarser beds, and yet have been able to cleave into thin parallel slaty lamina the whole body of the finer-grained argillaceous strata. One would more naturally suppose that the less finely-aggregated softer mud rocks or shales would have been even less easily fissured into sharp-cleavage joints than the more massive and better cemented grits.

Foliation.

The relations of the foliation or crystalline lamination of metamorphic strata to the cleavage planes and the planes of stratification, are next dwelt on. Two facts may be stated of foliation, which possess perhaps the constancy of general laws. One of them is, that this structure, as it is seen in gneiss and mica schist, observes, when the strata are not traversed by cleavage, an approximate parallelism with the original bedding. The author of this paper has beheld apparent exceptions to this rule in several localities near Philadelphia and elsewhere in the United States; and others have been noticed in Europe by Mr D. Sharpe and other good observers, but all of them can be reconciled to the general fact, and reduced, it is conceived, to the one comprehensive law,—that the planes of foliation, or the lamina formed by the crystalline constituents of the foliated rocks, are parallel to the planes or waves of heat which have been transmitted through the strata. Whenever large tracts of the gneissic rocks retain a nearly horizontal undisturbed position, the foliation is almost invariably coincident with the stratification; and in this case the wave of heat producing the crystalline structure can only have flowed upwards through the crust, invading stratum after stratum in parallel horizontal planes. Again, when injections of granite have lifted the gneissic strata, the crystalline lamination is generally seen to be parallel to the plane of outflowing temperature.

The other general rule is, that the foliation is parallel, or approximately so, to the cleavage, wherever these two structures occur in the same mass of rocks. This fact, recorded by Darwin, of the gneissic rocks and clay slates of South America, has been noticed likewise by Mr D. Sharpe, Mr David Forbes, Mr Sorby, and other geologists in Great Britain, and by the author in many localities in southern Pennsylvania. An interesting instance of such parallelism of the foliation to the cleavage, in the lastnamed region, tending to show convincingly that both phenomena are the consequences of one species of force, or but different degrees of development of the same molecular or crystallizing agency, is presented in the great synclinal trough of the lower Appalachian limestone, north of Philadelphia. On the north side of this trough, the primal and auroral rocks, Cambrian or Lower Silurian, dip S., over a wide outcrop, at a very regular angle of about 45°. On the south side, they have been lifted into, and even a little beyond, the perpendicular position, so that the synclinal axis plane of the belt dips at an angle of 65% or 70° to the south. Neither formation shows cleavage structure on the northern side of the valley, the limestone being there of an earthy texture, and in thick massive beds; but on the south, or upturned side, this limestone is altered into a mottled blue and white crystalline marble, and is pervaded with cleavage planes, dipping at angles of 70° and 80° southward. Many parts of the rock are like a foliated calcareous gneiss, thin laminæ of mica and tale dividing the slate-like plates of the marble. What is especially worthy of notice is, that the foliation of the mica and talc, composing some of the thin