§ 38. And the fact is, generally, that natural cliffs are very cautious how they overhang, and that the artist who represents them as doing so in any extravagant degree entirely destroys the sublimity which he hoped to increase, for the simple reason that he takes away the whole rock-nature, or at least that part of it which depends upon weight. The instinct of the observer refuses to believe that the rock is ponderous when it overhangs so far, and it has no more real effect upon him than the imagined rocks of a fairy tale.
Though, therefore, the subject sketched on this page is sufficiently trifling in itself, it is important as a perfect general type of the overhanging of that kind of precipices, and of the mode in which they are connected with the banks above. Fig. 88 shows its abstract leading lines, consisting of one great parabolic line _x y_ falling to the brow, curved aqueous lines down the precipice face, and the springing lines of its vegetation, opposed by contrary curves on the farther cliff. Such an arrangement, with or without vegetation, may take place on a small or large scale; but a bolder projection than this, except by rare accident, and on a small scale, cannot. If the reader will glance back to Plate +37+, and observe the arrangement of the precipices on the right hand, he will now better understand what Turner means by them. But the whole question of the beauty of this form, or mode of its development, rests on the nature of the bank above the cliffs, and of the aqueous forces that carved it; and this discussion of the nature of banks, as it will take some time, had better be referred to next chapter. One or two more points are, however, to be stated here.
§ 39. For the reader has probably been already considering how it is that these overhanging cliffs are formed at all, and why they appear thus to be consumed away at the bottom. Sometimes if of soft material they actually _are_ so consumed by the quicker trickling of streamlets at the base than at the summit, or by the general action of damp in decomposing the rock. But in the noblest instances, such cliffs are constructed as at c in Fig. 73, above, and the inward retirement of the precipice is the result of their tendency to break at right angles to the beds, modified according to the power of the rock to support itself, and the aqueous action from above or below.