the perspective. The entire composition is, however, so harmonious, that this violation of a law does not strike the spectator at the first glance; and if it be excusable, the apology is to be found in the fact that if but one point of sight, which must have been the central one, only, had been used, the openings which show the sky and trees, and a distant dome, must have been shut out, and much that is beautiful lost. Not that these excuses should constitute a precedent, for it is always in the power of the painter to adopt an arrangement that needs them not.

But the example of Masaccio in the treatment of architecture with reference to figures is perhaps a better one, and in this he is sometimes to be preferred even to Raphael; for in the engravings I have seen from his works at Florence his backgrounds are always strictly subordinate.

I have heard the diminutive proportions of the architecture in the cartoons defended on the ground that Raphael's object was to give importance to his figures. But whether it was with this intention or not that he dwarfed his buildings, I have no hesitation in saying that it was not necessary to such an end that he should do so.*—The large architecture which forms the background to Titian's great picture in the Church of the Frari, at Venice, ennobles the composition, with no loss of consequence to the figures, the dignity

* It probably arose from his frequent reference to antique bas-reliefs, in which the diminutive architecture is entirely conventional. The general plan of the "Sacrifice at Lystra," is almost copied from one of these. Yet all in it that is of most interest, is entirely Raphael's.

and importance of which is the first thing that impresses the mind; nor do the figures in the "Sacraments" of Poussin suffer in the least by the magnitude of the architecture, which the simplicity of its forms and the breadth of its light and shadow sufficiently subordinate to the groups.

The

But to return to the consideration of perspective. I know not that I can better point out its great value than by comparing the upper part of Raphael's "Dispute of the Sacrament" with the composition which furnished him with its arrangement. "Last Judgment," by Orcagna, in the Campo Santo, not only suggested the general plan of Michael Angelo's great work, but its lines of the Apostles sitting on each side of the Saviour and the Madonna, furnished the similar arrangement, though not of the same personages, to the "Dispute of the Sacrament." The immense superiority of Raphael's composition need not be pointed out; my object in the comparison is merely to observe how much he gained by his knowledge of perspective, in the elegant semicircular sweep formed by the cloud that supports the figures on each side of the Saviour, and the higher and therefore still more curved line of the angels above. This instance of the great value of perspective is the more striking because the composition is not architectural.

Balance of lines and masses is the great principle of general composition; and whether this be obtained by exact symmetry of parts, as in the "School of Athens," or by the many other more irregular plans of arrangement, depends wholly on the subject-for one form is not more legitimate than another. Nature

delights us in so many different ways, that Art may, and indeed should, follow her variety if it would avoid the stagnation of mediocrity,-the invariable result of too exclusive an attachment to any one system.

The first thing we are taught by perspective is, that all objects are apparently altered in magnitude and shape as they approach or recede from the eye, and that these alterations are infinitely varied as the points of distance and of sight are changed. Objects that are perfectly spherical form the only exceptions I can think of to this law, their apparent changes being of size only, and not of shape.

I remember, when I was a student, hearing it argued that parallel perspective must always be untrue, because perspective makes all horizontal lines tend towards points on the horizon, and all perpendicular lines approach each other to meet in a single point perpendicular to the point of sight, and that, therefore, there is no such thing as parallel perspective in nature. But, in such reasoning, one important fact is entirely overlooked, namely, that the plane of the picture itself is subject to the laws of perspective, and becomes altered, more or less, in shape, according to the point from which we view it, and carries with it all the lines on its surface that are parallel with, or perpendicular to, the horizon, exactly as such lines, in Nature, would be altered by the laws of vision;-parallel perspective being thus made true by Nature herself.

I am inclined to think that the vanishing points of rectangular objects, when represented obliquely to the