Examine, then, the plate just given (12 A). The cleavage of the shales runs diagonally up from left to right; note especially how delicately it runs up through the foreground rock, and is insisted upon, just at the brow of it, in the angular step-like fragments; compare also the etching in the first volume. Then note the upright pillars in the distance, marked especially as rhomboidal by being drawn with the cleavage still sloping up on the returning side, as at _a_, Fig. 83, not as at _b_, which would be their aspect if they were square; and then the indication of interruption in the structure at the brow of the main cliff, where, as well as on the nearer mass, exposure to the weather has rounded away the cleavages. This projection, as before mentioned, does exist at the spot; and I believe is partly an indication of the contortion in the beds alluded to by Mr. Phillips; but no one but Turner would have fastened on it, as in anywise deserving special attention. For the rest, no words are of any use to explain the subtle fidelity with which the minor roundings and cleavages have been expressed by him. Fidelity of this kind can only be estimated by workers: if the reader can himself draw a bit of natural precipice in Yoredale shale, and then copy a bit of the etching, he will find some measure of the difference between Turner's work and other people's, and not otherwise; although, without any such labor, he may at once perceive that there is a difference, and a wide one,--so wide, that I have literally nothing to compare the Turnerian work with in previous art. Here, however, Fig. 84, is a rock of Claude's (Liber Veritatis, No. 91, on the left hand), which is something of the shape of Turner's, and professes to be crested in like manner with copse-wood. The reader may "compare" as much as he likes, or can, of it. [Illustration: FIG. 85.]