the church as you can, without losing sight of the top of it. Look at the capitals of the columns on the second story. You see that they are exquisitely rich, carved all over. Tell me their patterns: You cannot. Tell me the direction of a single line in them: You cannot. Yet you see a multitude of lines, and you have so much feeling of a certain tendency and arrangement in those lines, that you are quite sure the capitals are beautiful, and that they are all different from each other. But I defy you to make out one single line in any one and falsehood of of them. Now go to Canaletto's painting of this church, in the Palazzo Manfrini, taken from the How much has he represented of

§14. Vacancy

Canaletto.

very spot on which you stood. all this? A black dot under each capital for the shadow, and a yellow one above it for the light. There is not a vestige nor indication of carving or decoration of any sort or kind.

Very different from this, but erring on the other side, is the ordinary drawing of the architect, who gives the principal lines of the design with delicate clearness and precision, but with no uncertainty or mystery about them; which mystery being removed, all space and size are destroyed with it, and we have a drawing of a model, not of a building. But in the capital lying on the foreground in Turner's Daphne hunting with Leucippus, we have the perfect truth. Not one jag of the acanthus leaves is absolutely visible, the lines are all disorder, but you feel in an instant that all are there. And so it will invariably be found through every portion of detail in his late and most perfect works.

in landscape fore

grounds.

But if there be this mystery and inexhaustible finish merely in the more delicate instances of architectural decoration, how much more in the ceaseless and incomparable decoration of nature. The detail of a single weedy bank laughs § 15. Still greater fulness and finish the carving of ages to scorn. Every leaf and stalk has a design and tracery upon it, every knot of grass an intricacy of shade which the labor of years could never imitate, and which, if such labor could follow it out even to the last fibres of the leaflets, would yet be falsely represented, for, as in all other cases brought forward, it is not clearly seen, but confusedly and mysteriously. That which is nearness for the bank, is distance for its details; and however near it may be,

the greater part of those details are still a beautiful incomprehensibility.*

* It is to be remembered, however, that these truths present themselves in all probability under very different phases to individuals of different powers of vision. Many artists who appear to generalize rudely or rashly are perhaps faithfully endeavoring to render the appearance which nature bears to sight of limited range. Others may be led by their singular keenness of sight into inexpedient detail. Works which are painted for effect at a certain distance must be always seen at disadvantage by those whose sight is of different range from the painter's. Another circumstance to which I ought above to have alluded is the scale of the picture; for there are differ ent degrees of generalization, and different necessities of symbolism, belonging to every scale: the stipple of the miniature painter would be offensive on features of the life size, and the leaves with Tintoret may articulate on a canvas of sixty feet by twenty-five, must be generalized by Turner on one of four by three. Another circumstance of some importance is the assumed distance of the foreground; many landscape painters seem to think their nearest foreground is always equally near, whereas its distance from the spectator varies not a little, being always at least its own calculable breadth from side to side as estimated by figures or any other object of known size at the nearest part of it. With Claude almost always; with Turner often, as in the Daphne and Leucippus, this breadth is forty or fifty yards; and as the nearest foreground object must then be at least that distance removed, and may be much more, it is evident that no completion of close detail is in such cases allowable, (see here another proof of Claude's erroneous practice ;) with Titian and Tintoret, on the contrary, the foreground is rarely more than five or six yards broad, and its objects therefore being only five or six yards distant are entirely detailed.

None of these circumstances, however, in any wise affect the great principle, the confusion of detail taking place sooner or later in all cases. I ought to have noted, however, that many of the pictures of Turner in which the confused drawing has been least understood, have been luminous twilights; and that the uncertainty of twilight is therefore added to that of general distance. In the evenings of the south it not unfrequently happens that objects touched with the reflected light of the western sky, continue even for the space of half an hour after sunset, glowing, ruddy, and intense in color, and almost as bright as if they were still beneath actual sunshine, even till the moon begins to cast a shadow but in spite of this brilliancy of color all the details become ghostly and ill-defined. This is a favorite moment of Turner's, and he invariably characterizes it, not by gloom, but by uncertainty of detail. I have never seen the effect of clear twilight thoroughly rendered by art; that effect in which all details are lost, while intense clearness and light are still felt in the atmosphere, in which nothing is distinctly seen, and yet it is not darkness, far less mist, that is the cause of concealment. Turner's efforts at rendering this effect (as the Wilderness of En

Hence, throughout the picture, the expression of space and size is dependent upon obscurity, united with, or rather resultant from, exceeding fulness. We destroy both space and size, either by the vacancy, which affords us no measure

§ 16. Space and

alike by distinctness and by vacancy.

false one.

size are destroyed of space, or by the distinctness, which gives us a The distance of Poussin, having no indication of trees, nor of meadows, nor of character of any kind, may be fifty miles off, or may be five; we cannot tell-we have no measure, and in consequence, no vivid impression. But a middle distance of Hobbima's involves a contradiction in terms; it states a distance by perspective, which it contradicts by distinctness of detail.

§17. Swift execution best secures perfection of details.

§ 18. Finish is far
more necessary
in landscape than

A single dusty roll of Turner's brush is more truly expressive of the infinity of foliage, than the niggling of Hobbima could have rendered his canvas, if he had worked on it till doomsday. What Sir J. Reynolds says of the misplaced labor of his Roman acquaintance on separate leaves of foliage, and the certainty he expresses that a man who attended to general character would in five minutes produce a more faithful representation of a tree, than the unfortunate mechanist in as many years, is thus perfectly true and well founded; but this is not because details are undesirable, but because they are best given by swift execution, and because, individually, they cannot be given at all. But it should be observed (though we shall be better able in historical sub- to insist upon this point in future) that much of harm and error has arisen from the supposition and assertions of swift and brilliant historical painters, that the same principles of execution are entirely applicable to landscape, which are right for the figure. The artist who falls into extreme detail in drawing the human form, is apt to become disgusting rather than pleasing. It is more agreeable that the general outline and soft hues of flesh should alone be given, than its hairs, and veins, and lines of intersection. And even the most rapid and generalizing expression of the human body,' gedi, Assos, Chateau de Blois, Caer-laverock, and others innumerable,) have always some slight appearance of mistiness, owing to the indistinctness of details; but it remains to be shown that any closer approximation to the effect is possible.

jects.

if directed by perfect knowledge, and rigidly faithful in drawing, will commonly omit very little of what is agreeable or impressive. But the exclusively generalizing landscape painter omits. the whole of what is valuable in his subject,-omits thoughts, designs, and beauties by the million, everything, indeed, which can furnish him with variety or expression. A distance in Lincolnshire, or in Lombardy, might both be generalized into such blue and yellow stripes as we see in Poussin; but whatever there is of beauty or character in either, depends altogether on our understanding the details, and feeling the difference between the morasses and ditches of the one, and the rolling sea of mulberry trees of the other. And so in every part of the subject. I have no hesitation in asserting that it is impossible to go too fine, or think too much about details in landscape, so that they be rightly arranged and rightly massed; but that it is equally impossible to render anything like the fulness or the space of nature, except by that mystery and obscurity of execution which she herself uses, and in which Turner only has followed her.

We have now rapidly glanced at such general truths of nature as can be investigated without much knowledge of what is beautiful. Questions of arrangement, massing, and generalization, I prefer leaving untouched, until we know some

$19. Recapitulation of the section.

thing about details, and something about what is beautiful. All that is desirable, even in these mere technical and artificial points, is based upon truths and habits of nature; but we cannot understand those truths until we are acquainted with the specific forms and minor details which they affect, or out of which they arise. I shall, therefore, proceed to examine the invaluable and essential truths of specific character and form-briefly and imperfectly, indeed, as needs must be, but yet at length sufficient to enable the reader to pursue, if he will, the subject for himself.

SECTION III.

OF TRUTH OF SKIES.

$1. The peculiar adaptation of the

sky to the pleasing and teaching of

man.

CHAPTER I.

OF THE OPEN SKY.

It is a strange thing how little in general people know about the sky. It is the part of creation in which nature has done more for the sake of pleasing man, more for the sole and evident purpose of talking to him and teaching him, than in any other of her works, and it is just the part in which we least attend to her. There are not many of her other works in which some more material or essential purpose than the mere pleasing of man is not answered by every part of their organization; but every essential purpose of the sky might, so far as we know, be answered, if once in three days, or thereabouts, a great ugly black rain cloud were brought up over the blue, and everything well watered, and so all left blue again till next time, with perhaps a film of morning and evening mist for dew. And instead of this, there is not a moment of any day of our lives, when nature is not producing scene after scene, picture after picture, glory after glory, and working still upon such exquisite and constant principles of the most perfect beauty, that it is quite certain it is all done for us, and intended for our perpetual pleasure. And every man, wherever placed, however far from other sources of interest or of beauty, has this doing for him constantly. The noblest scenes of the earth can be seen and known but by few; it is not intended that man should live always in the midst of them, he injures them by his presence, he ceases to feel them if he be always with them; but the sky is for all; bright as it is, it is not "too bright, nor good, for human nature's daily food ;"