DRAWN BY J. W. McKECHNIE.

See pages 103, 126, 279.

these divisions are of equal length. Yet it is impossible to look at them without suspecting that the divisions nearest the ends of the lines are the longest. This is the same as to say that to cause these end divisions to appear of the same length as the others, they should be made shorter. The reason why this is so is owing, of course, to the roundness of the eye. When we look at the middle of a horizontal line, there is actually more eyesurface covered by the divisions at the sides than there is by the divisions seen directly in front, which latter divisions are opposite that part of the eye which is most nearly flat. As the eye is rounded vertically as well as horizontally, a similar principle applies sometimes to vertical measurements.

In order to produce the differences in measurement of corresponding factors in different buildings, an architect need merely apply to architecture the same methods of carrying out the laws of perspective that are known to be applied in painting. In this latter art, it is seldom considered necessary to apply these laws with mathematical exactness. Each draughtsman, in arranging his outlines, feels at liberty to stand off from his drawing, and, as a result of repeated examinations and experiments, to use his own ingenuity. Indeed, he must do this, in any cir cumstances, because the required measurements differ with every foot by which he stands nearer to his product, or farther from it. Precisely so in architecture. Let the man in Fig. 16, page 105, step a few feet farther away from the building, and, in order to preserve the same effect, not only would the curve in the cornice have to be lessened, but the columns at either end of the colonnade would have to be brought nearer together. Let a temple placed upon the brow of a hill be intended to produce a

certain effect upon those ascending it and the pediment would have to be higher than if it were intended to produce the same effect upon those on a level plain. No wonder, then, that we find such variations in the measurements, and such apparent lack of meaning in the variations, as are indicated in the following, taken from Penrose:

In the age in which the Greek temples were constructed, other artists believed-and why not the architect?—that a man should study upon a product, if he intended to have it remain a model for all the future. Is it not natural to suppose that in such an age the structural arrangements intended to counteract optical defects, or to produce optical illusions, or, as some think, to produce, in connection with these, effects of variety or of vagueness in line or outline (see page 89), were largely the results of the individual experiments of individual

very

nearly

builders? If not such results, why were they invariably different in different buildings? But if they were such, the predominating motive in the mind of the artist was not to imitate any particular form that he had seen before, so much as to represent its general effect. Thus, from the beginning of architecture in which we see the builder taking suggestions from primitive huts or from the trunks and branches of trees in nature, to the highest stage of its development, where we see him taking suggestions from the works of previous architects we find him, in the degree in which he is a great artist, representing rather than imitating.