mp being parallel to DC'. Consequently, the distance of the physical point p from the axis of the cone, will not be altered by the elevation; and since the same holds for every physical point in the circumference of the horizontal circle whose radius is pn, there can be no tension at any point of the physical line forming that circumference, in the direction of its tangent at that point. This is consistent with our assumption of the equable extension of every part of the line 'C', which will therefore be true*. Similarly, if we conceive the whole mass AA'B'B to be formed by the superposition of similar conical shells, it is easily seen that the same result will hold for every horizontal circle concentric about the axis of the cone. Hence it follows, that if any vertical plane be drawn through the axis of the cone, there will be no tension at any point of the mass in this plane in a direction perpendicular to it. The tension will be entirely in the plane, and parallel to the slant side of the cone.

If, then, a fissure which should pass through any proposed point P, were formed according to the greatest tendency of the tensions of the unbroken mass to form it, it would manifestly coincide with the surface of an inverted cone, whose base would be the circle of which the radius is pn, and whose axis would coincide with that of the elevated cone. If p should coincide with C', an orifice would be formed along the axis C'C; and if we consider that the force will act, according to our hypothesis, with the greatest intensity at C, it seems highly probable that the first dislocation will usually take place along, or very near to that axis. For the greater distinctness, suppose this to be the case.

47. The instant this has occurred, the conditions of the problem will be entirely altered. The force at C' maintaining every such line as A'C' and B'C' in its state of tension, being now destroyed, the

* Suppose a tension T to exist along the physical line forming the circumference of the T circle whose radius is pn. This would produce a force acting at p in the direction pn, pn

the resolved part of which in the direction pC' would increase the tension of A'p. In such case the extension of A'C' would be greatest at A', and our assumption of the uniform extension of that line would not be true.

extremities of those lines at C' will separate from each other by the contraction of 'C' and B'C'; and the same will be true for every similar pair of lines. An extension of the orifice at C'' will thus be produced, and consequently a tension of the mass contiguous to it in the direction of a tangent to a horizontal section of it, while the tension in the direction of such lines as C'A' will be entirely destroyed near to C', and much lessened at lower points. The whole tension therefore in the upper part of the mass, will be in the directions of the tangents of horizontal circles concentric about the axis; and the tendency to form a fissure there, will be entirely in a vertical plane passing through the axis of the cone. It is easily seen also that the tension at the vertex will be greater than in any other part. Consequently, if fissures be formed under these circumstances, they will commence at the vertex, and be in positions such as that just mentioned.

48. Let us now suppose the elevatory force to act with additional intensity beneath the point C of the annexed diagram, (which represents a horizontal section,) so as to superimpose on the general elevation

a conical one, having its apex at C. In addition to the tension (F) acting at any point P within the bounds of the cone, and in the direction perpendicular to the general axis of elevation, we shall also have another tension (f) acting at P, in the direction PQ perpendicular to CP, (taking the case of Art. 47.) and the tendency of these tensions will be to form a fissure deviating from perpendicularity with PQ, in a degree depending on the relative intensities of ƒ and F. Consequently, a fissure A'PB' will deviate from parallelism with the line of general elevation, approximating towards C in the manner above represented.

49. If the partial elevation instead of approximating to the conical form, be more nearly spherical, without any such rupture at C, as VOL. VI. PART I.

G

above supposed, the principal tension due to it will be in the direction CP, instead of being perpendicular to that line, in which case the deviation in the direction of the fissure will be the contrary of that above represented.

§. Formation of Longitudinal and Transverse Faults-Anticlinal LinesLongitudinal Valleys-Transverse Valleys-Comparative Effects of subsequent Movements on the Width of Longitudinal and Transverse Fissures-Throw of a Vein.

It appears then, that in the case we have considered, and under the conditions assumed, the elevating forces will produce two systems of fissures with a general approximation (subject to certain modifications) to rectilinearity, and perpendicular to each other. Let us further consider what positions the different portions of the mass may assume subsequently to the formation of these fissures.

50. The diagram in page 45, represents a transverse section of the elevated range, immediately after the contemporaneous formation of the complete fissures MN, CC', &c. It does not appear probable that the effects of the continued action of the elevatory force will afterwards follow any general law; for the subsequent movements of the different portions of the mass, now rendered in some degree independent of each other by the fissures which separate them, must be constantly influenced by that irregularity in the action of the elevatory force, and those accidental and local causes of which it is now impossible to form any estimate. If the elevatory force be produced by an expansive vapour, or act through the medium of any fluid, as we have supposed it to do, its intensity must decrease after a certain time, thus causing subsidencies in the elevated mass, the degree of which in different portions will probably be in general determined by accidental circumstances. One consequence, however, of these irregular causes, would appear to be necessarily a very general one, viz. a difference of elevation in the adjoining parts of different portions of the mass separated by the fissures, whether longitudinal or trans

verse, thus producing systems of longitudinal and transverse faults, such as described in the Introduction, (1. a, ß.)

51. Sections of longitudinal faults which may be thus produced, are shewn in the annexed diagram, which represents one of the forms which, it is manifest, the uplifted mass represented in page 45, may ultimately assume from the causes above mentioned (Art. 50). In such case we shall have an anticlinal line through N", running parallel to the general one through C'' in the central part of the elevation; and a synclinal line through N' parallel to the two former ones. The existence also of these longitudinal fissures and consequent irregularities of surface, will obviously tend to direct the action of superficial

agents of denudation along longitudinal courses, and thus to facilitate the formation of longitudinal valleys, particularly in the case in which the relative elevation of two adjoining portions of the mass is such as represented at N. If this kind of elevation be continued for a considerable distance longitudinally, a distinct longitudinal valley must be the necessary consequence.

52. It not unfrequently happens that we observe in anticlinal lines. a degree of deviation from approximate rectilinearity, which might at first sight appear inconsistent with the mode of formation which this theory would assign to them, assuming that great predominance of general over partial and accidental causes, throughout an extensive area, with which very irregular deviations in the direction of a fissure would not be accordant. It seems, however, highly probable, that this character of anticlinal lines would not, in fact, be the unfrequent consequence of the general causes we are considering. In the first place, we may observe that longitudinal fissures are not necessarily continuous for any great distance, as we have explained in Art. 33, and

therefore an anticlinal line formed along one fissure, may easily be conceived to be continued along another, not exactly in the same line. If we conceive several transferences of this kind to take place from one fissure to another, we shall have a discontinuous anticlinal line, each portion of which will be as rectilinear as the fissure with which it coincides; but if the physical structure of the mass should be placed under that disguise so frequently spread over it by superficial agencies, the geologist, instead of detecting this discontinuous line, consisting of a number of straight ones having parallel directions, will probably only recognize a somewhat ill defined anticlinal line of irregular curvature, and apparently destitute, in a considerable degree, of those characters of rectilinearity and parallelism with the general axis of elevation which this theory might appear to assign to such lines. It may also be observed, that since on the opposite sides of a transverse fissure the movements of the adjoining masses will be in some degree independent of each other, it is easy to conceive that this cause also may sometimes facilitate the transference of an anticlinal line from one longitudinal fissure to another, and thus destroy its apparent rectilinearity.

Similar observations will equally apply to the directions of longitudinal valleys, as far as their formation may be referrible to the causes above mentioned.

53. It has been stated how much the ultimate position of the dislocated mass may generally depend on accidental causes. In particular cases however, and especially with respect to those portions of the mass adjoining the lateral boundaries of the general elevation, there appears reason to expect that the phenomena would, according to our theory, frequently follow a certain law. Suppose the diagram, page 51, to represent the portion of the mass bounded by two parallel transverse fissures, produced as described in Art. 43, by a greater intensity of the elevatory force acting at the point C. For the greater simplicity, we may also suppose this force to act symmetrically with respect to the two transverse bounding fissures. Then, after the general elevation has proceeded as far as represented in the diagram, page 45, and the fissures have been formed,