blende are well defined, and commonly have a more or less parallel arrangement; here and there are bands in which these minerals are more abundant than elsewhere. The quartz and the feldspar are granular in form; the boundaries of these minerals are not rectilinear, but curved, wavy, or lobate; small grains of the one sometimes appear to be inclosed in larger grains of the other. Though the structure of this rock has a superficial resemblance to that of a granite of similar coarseness, it differs from it in this respect, as we can see from the next instance, a true granite, where the rectilinear outline of the feldspar is conspicuous. Here, then, is one of our problems. This difference of structure is too general to be without significance. What, then, does it mean? Among the agents of change known to geologists, three are admittedly of great importance: these are water, heat, and pressure. The first effect of pressure due to great earth-movements is to flatten somewhat the larger fragments in rocks, and to produce in those of finer grain the structure called cleavage. This, however, is a modification mainly mechanical. It consists in a rearrangement of the constituent particles; mineral changes, so far as they occur, being quite subordinate. But in certain extreme instances the latter are also conspicuous. From the fine mud, generally the result of the disintegration of feldspar, a mica, usually colorless, has been produced, which occurs in tiny flakes, often less than one hundredth of an inch long. In this process a certain amount of silica has been liberated, which sometimes augments pre-existing granules of quartz, sometimes consolidates independently as micro-crystalline quartz. Simultaneously carbonaceous and ferruginous constituents are converted into particles of graphite or of iron oxide. As to the effects of pressure when it acts upon a rock already crystalline, there are, as it seems to me, differences in the resultant structures which are dependent upon the mode in which pressure has acted. They are divisible into two groups; one indicating the result of simple direct crushing, the other of crushing accompanied by shearing. In the former case, the rock mass has been so situated that any appreciable lateral movement has been impossible; it has yielded like a block in a crushing-machine. In the latter, a differential lateral movement of the particles has been possible, and it has prevailed when (as in the case of an overthrust fault) the whole mass has not only suffered compression, but also has traveled slowly forward. Obviously, the two cases can not be sharply divided, for the crushing up of a non-homogeneous rock may render some local shearing possible. Still, it is important to separate them in our minds, and we shall find that in many cases the structure, as a whole, like the cleavage of a slate, results from a direct crush; while in others the effects of shearing predominate. The latter, accordingly, ex-
