in the form of a block capped by a pyramid or are excavated in the rock. The former class offers little interest from the architectural point of view. But some of the rock-cut tombs of Beni-hasan, belonging to the Twelfth Dynasty, exhibit a feature which calls for mention. These tombs have been so made as to leave pillars of the living rock standing, both at the entrance and in the chapel. The simplest of these pillars are square in plan and somewhat tapering. Others, by the chamfering off of their edges, have been made eight-sided. A repetition of the process gave sixteen-sided pillars. The sixteen sides were then hollowed out (channeled). The result is illustrated by Fig. 6. It will be observed that the pillar has a low, round base, with beveled edge; also, at the top, a square abacus, which is simply a piece of the original four-sided pillar, left untouched. Such polygonal pillars as these are commonly called proto-Doric columns. The name was given in the belief that these were the models from which the Greeks derived their Doric columns, and this belief is still held by many authorities.

With the New Empire we begin to have numerous and extensive remains of temples, while those of an earlier date have mostly disappeared. Fig. 7 may afford some notion of what an Egyptian temple was like. This one is at Luxor, on the site of ancient Thebes in Upper Egypt. It is one of the largest of all, being over 800 feet in length. Like many others, it was not originally planned on its present scale, but represents two or three successive periods of construction, Ramses II., of the Nineteenth Dynasty, having given it its final form by adding to an already finished building all that now stands before the second pair of towers. As so extended, the building has three pylons, as they are called, pylon being the name for the pair of sloping-sided towers with gateway between. Behind the first pylon comes an open court surrounded by a cloister with double rows of columns. The second and third pylons are connected with one another by a covered passage—an exceptional feature. Then comes a second open court; then a hypostyle hall, i.e., a hall with flat roof supported by columns; and finally, embedded in the midst of various chambers, the relatively small sanctuary, inaccessible to all save the king and the priests. Notice the double line of sphinxes flanking the avenue of approach, the two granite obelisks at the entrance, and the four colossal seated figures in granite representing Ramses II.—all characteristic features.

Fig. 8 is taken from a neighboring and still more gigantic temple, that of Karnak. Imagine an immense hall, 170 feet deep by 329 feet broad. Down the middle run two rows of six columns each (the nearest ones in the picture have been restored), nearly seventy feet high. They have campaniform (bell-shaped) capitals. On either side are seven rows of shorter columns, somewhat more than forty feet high. These, as may be indistinctly seen at the right of our picture, have capitals of a different type, called, from their origin rather than from their actual appearance, lotiform or lotus-bud capitals. There was a clerestory over the four central rows of columns, with windows in its walls. The general plan, therefore, of this hypostyle hall has some resemblance to that of a Christian basilica, but the columns are much more numerous and closely set. Walls and columns were covered with hieroglyphic texts and sculptured and painted scenes. The total effect of this colossal piece of architecture, even in its ruin, is one of overwhelming majesty. No other work of human hands strikes the beholder with such a sense of awe.

Fig. 9 is a restoration of one of the central columns of this hall. Except for one fault, say Messrs. Perrot and Chipiez,[Footnote: "Histoire de l'Art Egypte," page 576. The translation given above differs from that in the English edition of Perrot and Chipiez, "Art in Ancient Egypt," Vol. II., page 123.] "this column would be one of the most admirable creations of art; it would hardly be inferior to the most perfect columns of Greece." The one fault—a grave one to a critical eye—is the meaningless and inappropriate block inserted between the capital and the horizontal beam which it is the function of the column to support. The type of column used in the side aisles of the hall at Karnak is illustrated by Fig. 10, taken from another temple. It is much less admirable, the contraction of the capital toward the top producing an unpleasant effect.

Other specimens of these two types of column vary widely from those of Karnak, for Egyptian architects did not feel obliged, like Greek architects, to conform, with but slight liberty of deviation, to established canons of form and proportion. Nor are these two by any means the only forms of support used in the temple architecture of the New Empire. The "proto-Doric" column continued in favor under the New Empire, though apparently not later; we find it, for example, in some of the outlying buildings at Karnak. Then there was the column whose capital was adorned with four heads in relief of the goddess Hathor, not to speak of other varieties. Whatever the precise form of the support, it was always used to carry a horizontal beam. Although the Egyptians were familiar from very early times with the principle of the arch, and although examples of its use occur often enough under the New Empire, we do not find columns or piers used, as in Gothic architecture, to carry a vaulting. In fact, the genuine vault is absent from Egyptian temple architecture, although in the Temple of Abydos false or corbelled vaults (cf. page 49) do occur.

Egyptian architects were not gifted with a fine feeling for structural propriety or unity. A few of their small temples are simple and coherent in plan and fairly tasteful in details. But it is significant that a temple could always be enlarged by the addition of parts not contemplated in the original design. The result in such a case was a vast, rambling edifice, whose merits consisted in the imposing character of individual parts, rather than in an organic and symmetrical relation of parts to whole.

Statues of the New Empire are far more numerous than those of any other period, but few of them will compare in excellence with the best of those of the Old Empire. Colossal figures of kings abound, chiseled with infinite patience from granite and other obdurate rocks. All these and others may be passed over in order to make room for a statue in the Louvre (Fig. 11), which is chosen, not because of its artistic merits, but because of its material and its subject. It is of bronze, somewhat over three feet in height, thus being the largest Egyptian bronze statue known. It was cast in a single piece, except for the arms, which were cast separately and attached. The date of it is in dispute, one authority assigning it to the Eighteenth Dynasty and another bringing it down as late as the seventh century B.C. Be that as it may, the art of casting hollow bronze figures is of high antiquity in Egypt. The figure represents a hawk-headed god, Horus, who once held up some object, probably a vase for libations. Egyptian divinities are often represented with the heads of animals— Anubis with the head of a jackal, Hathor with that of a cow, Sebek with that of a crocodile, and so on. This in itself shows a lack of nobility in the popular theology. Moreover it is clear that the best talents of sculptors were engaged upon portraits of kings and queens and other human beings, not upon figures of the gods. The latter exist by the thousand, to be sure, but they are generally small statuettes, a few inches high, in bronze, wood, or faience. And even if sculptors had been encouraged to do their best in bodying forth the forms of gods, they would hardly have achieved high success. The exalted imagination was lacking.

Among the innumerable painted bas-reliefs covering the walls of tombs and temples, those of the great Temple of Abydos in Upper