We have, therefore, to consider, lastly, what are the limitations under which this form is to be given. 1st. Let the height of the picture be a fixed line = a, in fig.ニニ. Draw AB, at right angles, to a. With centre c, distance 2 a, describe circle, cutting AB in B.:. ACB=60. AB is the utmost length of the picture which can be admitted; and AB = v(BC2-a2) = v((2a)2-a2) = av3. And such a length of picture as this is very rarely admissible; two-thirds of it are about the best average distance. Hence it appears, that all such paintings as Stothard's Canterbury Pilgrimage, are panoramas, not pictures. In the Royal Academy, two years ago, there was a very sweet bit by Landseer-Highland drovers crossing a bridge; and if the picture had been confined to the breadth of the bridge itself, and a white Shetland pony looking over into the water, which was the chief light, all had been well; instead of this, we had a parallelogram of about seven feet by one, with a whole procession of figures, extending from one end to the other, the bridge in the centre, and the picture was altogether ruined. 2nd. The corners of the picture, as we have seen, are out of the ellipse, and, therefore, beyond the limit of sight. Accordingly, they might be vague and subdued in colour, and totally without objects; but as this would draw too much attention to them, the artist continues his proximate colour into them, generally, however, keeping his brush in circular sweeps, indicating the form of the ellipse. Copley Fielding's management of the angles of a breezy sea-piece is, perhaps, the best instance that can be given. Lastly. The true distance at which the eye ought to be placed is the length of the minor axis of the ellipse; but as this minor axis is usually a little diminished, the best standard is the vertical of the equilateral ? (triangle) whose side is the major axis, or the greatest dimension of the picture. In those drawings where the composition is good, and the attention very much confined, this distance may even be exceeded.