hibit phenomena resembling the effects of a tensile stress. Materials of a like character assume a more or less linear arrangement; the rock becomes slightly banded, and exhibits, as has been said, a kind of flexion structure. The mass gradually assumes a fragmental condition under the pressure, and its particles, as they shear and slide under the effects of thrust, behave to some extent like those of a non-uniform mass of rock in a plastic condition, as, for example, a slaggy glass. Illustrations of the effects of direct crushing in a granitoid rock are common in the Alps. Those of a shearing crush are magnificently developed near the great overthrust faults in the northwest Highlands of Scotland. It seems, then, to be demonstrated that by mechanical deformation, accompanied or followed by molecular rearrangement, foliated rocks, such as certain gneisses and certain schists, can be produced from rocks originally crystalline. But obviously there are limits to the amount of change. To get certain results you must have begun with rocks of a certain character. Hitherto we have been dealing with rocks which were approximately uniform in character, though composed of diverse materials—that is, with rocks more or less granular in character. Suppose, now, the original rock to have already acquired a definite structure—suppose it had assumed, never mind how, a distinct mineral banding, the layers varying in thickness from a small fraction of an inch upward. Would this structure survive the mechanical deformation? I can give an answer which will at any rate carry us a certain way. I can prove that subsequent pressure has frequently failed to obliterate an earlier banded structure. In such a district as the Alps we commonly find banded gneisses and banded schists which have been exposed to great pressure. Exactly as in the former case, the new divisional planes are indicated by a coating of films of mica, by which the fissility of the rock in this direction is increased. The mass has assumed a cleavage-foliation. I give it this name because it is due to the same cause as ordinary cleavage, but is accompanied by mineral change along the planes of division, while I term the older structure stratification-foliation, because so frequently, if it has not been determined by a stratification of the original constituents, it is at any rate a most extraordinary imitation of such an arrangement. In many cases the new structure is parallel with the old; but in others, as in the "strain-slip" cleavage of a phyllite, the newer can be seen distinctly cutting across the older mineral banding. To put it briefly, I assert, as the result of examining numbers of specimens, that, though in certain cases the new structure is dominant, a practiced eye seldom fails to detect traces of the older foliation, while in a large number of instances it is still as definite as the stripe in a slate. We have got, then, thus far—that pressure acting on rocks pre-
