The most satisfactory evidence of a fault is the discovery of a rock bed which may be easily identified, and which is actually seen displaced on a plane of fracture which intersects it (Fig. 42, p. 59). When such an easily recognizable layer is not to be found, the plane of displacement may perhaps be discovered as a narrow zone composed of angular fragments of the rock cemented together by minerals which form out of solution in water. Such a fractured rock zone which follows a plane of faulting is a _fault breccia_. If the fault breccia, or vein rock, is much stronger than the rock on either side, it may eventually stand in relief at the surface like a dike or wall. At other times the displacement produces little fracture of the walls, but they slide over each other in such a manner as to yield either a smoothly corrugated or an evenly polished surface which is described as “slickensides.” It may be, however, that during the movement either one or both of the walls have “dragged”, and so are curled back in the immediate neighborhood of the fault plane (Fig. 44). When, as is quite generally the case, the actual plane of displacement of a fault is not open to inspection, the movement may be proven by the observation of abrupt, as contrasted with gradual, changes in the strikes and dips of neighboring exposures (Fig. 45); or by noting that some easily recognized formation has been sharply offset in its outcrops (Fig. 46). [Illustration: FIG. 45.—Map to show how a fault may be indicated in abrupt changes of the strike and dip of neighboring exposures.] [Illustration: FIG. 46.—A series of parallel faults indicated by successive offsets in the course of an easily recognizable rock formation.] There are in addition many indications rather than proofs of the presence of faults, which must be taken account of in every general study of the geology of a district. Thus the outcrops of all neighboring formations may terminate abruptly upon a straight line which intersects all alike. Deep-seated fissure springs may be aligned in a striking manner, and so indicate the course of a prominent fracture, though not necessarily of a fault. Much the same may be said of the dikes of cooled magma which have been injected along preëxisting fractures.