is a principle of beauty, when we consider a piece laterally, that is, parallel to the horizontal line: and the same rule obtains when we consider a landscape in its recession from the eye. It is divided (as Hogarth has observed) into three distances, which are called, the fore-ground, the middle-ground, and the off-skip. The objects on the fore-ground are distinct in their lines, and strong in light and shade. Those on the middle-ground are somewhat fainter: and those in the back-ground partake of that blue colour which the intermediate air gives to all distant objects. But here again the measures should vary as before, because equality produces no harmony.

This tripartite disposition may be regarded at first as a source of beauty which is arbitrary and fanciful 1; but I have so often found myself struck with it, before I had considered it critically, that if I were to lay out an advantageous piece of ground, I would introduce it wherever I had an opportunity, and trust for the event to the taste of the spectator. If you have less than three objects presented to the eye, the composition is deficient and empty: if you have more the sight is dissipated, and you must find some way of reducing, or, as Hogarth calls it, massing them. I suspect that the celebrated statue of the Laocoon, however excellent in other respects, strikes every eye with more pleasure because it consists of three figures, all contributing to the same effect.

In the use of perspective, regard should always be had to the rule of making unequal divisions. The cen

figures. It was Albano who told our author this, and from his mouth I had it. The reasons which he gave, were, first, that he believed there ought not to be above THREE GROUPES of figures in any pieture." See Fresnoy on Painting, page 102.

tre of the object should never be in the centre of the piece. This is the case with the plans and elevations of builders, which have therefore no merit to the eye as pictures. There must be an obliquity in the lines, which produces harmony and variety; and hence a good painter never gives you the full face of a building, nor places a street or an avenue receding directly from the eye, and vanishing into the middle of the picture: all his measures run obliquely; and it will be found that his distribution is never so pleasing as when the sight has three principal points to rest upon.

If we make a transition to architecture, there the three dimensions of length, breadth, and height, which are common to all solid bodies, will never strike us so much with a sense of beauty as when they are accommodated to one another in some proportions deduced by analogy from the theory of music: and such measures, whether they are applied in the external elevation, or the internal divisions, will have a pleasing effect, though the spectator is ignorant of the cause; for musical sounds please the ears of those who know nothing about their proportions. Thus, for example, if we would proportion the dimensions of a room in the best manner, let us take the measures from the harmonic divisions of a musical string, called a monochord; whatever note the whole string sounds, two thirds of that whole (the tension remaining the same) will sound a fifth; three-fourths will sound a fourth; one half will sound an octave, or eighth. To apply these to our present purpose, let the length of a room betwenty-four feet, the breadth sixteen, and the height twelve; then will the breadth be to the length in the ratio of two to three, which is that of the diapente or fifth, a most perfect concord; the height will be to the breadth in the ratio of three to four, which is that of

the diatessaron, or fourth; and to the length in the ratio of one to two, which is that of the diapason, or octave. Every person that has eyes will pronounce such a room to be finely proportioned, and feel the harmony of the dimensions without knowing them. The numbers 36, 24, and 18, having the same ratios to each other, may answer as well. Utility and convenience may require very different dimensions; but still, if we study elegance, we must have regard to the same rule. It may be necessary that the length should be to the breadth in the ratio of two to one, which is that of the octave; or three to one, which is that of the twelfth; or four to one, which is that of disdiapason, or the double octave.

If you would try, by a simple experiment, what proportion will do, only make the figure of a cross with two plain right lines, in which let the breadth be to the length as two to three, and let the point of transection, or distance of the arms from the bottom, compared with the whole length, be also as two to three; such a figure will strike the eye with its symmetry, and perhaps be the most beautiful of the kind that can be constructed; while other inharmonious measures might be introduced, which would be as ungrateful to the sight as discords are to the ear.

But to return to our great principle of tripartition, (if I may be allowed to make a new term for a new thing) the propriety and effects of it are so extensive, that it meets us almost every where. What is said of the sight, when compared with the hearing, will hold good also of the intellect, which is another kind of sight, the sight of the mind. In oratory, does not experience teach us, that the association of three ideas satisfies the mind, as the union of three sounds satisfies the ear? No scholar is a stranger to the fulness

and beauty of those three words, when set together, veni, vidi, vici; the effect of which is increased by a consonance of alliteration, each word beginning with the same letter.

In the art of reasoning, every syllogism consists of three propositions, all of which have a mutual consonance, if they make good logic. But here I am sensible that the parallel may raise a very ridiculous idea in the mind of a musical reader, if he imagines himself to hear a logical concert, by one person repeating the major proposition, another the minor, and a third the conclusion, and all speaking their parts at once. However, it is certainly true, and to our purpose, that as in musical concord two extremes have consent with the mean, and with one another, so in logic two ideas agree with a third, which is called the middle term, and all make good harmony together in the conclusion.

The principle of tripartition, as deducible from music, seems on the whole to be an actual source of pleasure to the judgment; and it is supported by such a variety of instances, that it must be founded in nature. When we are upon a right scent, truth will seem to run along before us of its own accord. There is one remarkable example which I have omitted; and it is this; that the beauty of the light, which gives beauty to all visible objects, is itself constituted by three colours, into which it divides itself, the red, the yellow, and the blue, which are the only original colours, all others being compounded of these; and a pure brightness is the result of them when their effects are united. These strange coincidences between the elements of different arts have often filled my mind with wonder. All I would infer from this uniformity is, that the principle I have proposed is not imaginary, but real, in nature;

and if so your taste will be certainly improved by the application of it: for nature is the ground of art, and a sure rule of pleasure to the judgment.

With regard to composition in painting, which was the art I had chiefly in view from the beginning of this letter, as a polite subject in which every gentleman should have some discernment; the beauties of it, when considered at large, consist in propriety of action; grace of attitude, which is also called ease; truth of proportion; and anatomical perspective. It would require another letter to explain this particularly: I shall only say, that all these beauties concur in the pieces of Sir Joshua Reynolds perhaps more truly than they were ever found together in the works of any other master. It is now very fashionable to see faults in his pictures; but I think chiefly with those who are slow in distinguishing real excellence. Look at the best family pictures of Vandyke, you generally see all the figures standing inanimate, like kings and queens, with nothing to do, but to look at you from their frames but Sir Joshua strikes out a general design, to which every figure in the composition contributes something; instead of looking at you they are engaged in some business of their own; and while you look at them you become interested in it yourself. Thus his family pictures, instead of losing their value with age, like an almanac, will retain as long as they can last, and that even in the eyes of strangers to the family, the merit of historical compositions.

In this copious subject I might have descended to many other particulars: but if you read Hogarth's book carefully, and attend to the few observations I have here added to it, you will acquire what Aristotle calls SEUTEOOV оμpa, a second sight; that sight with