palace; and to the westward a fourth suite, more ruined, however, than the rest, owing to its being situated so near the edge of the terrace. As far as can be made out, the rooms on this face seem to have been arranged three deep: the outer opening on the terrace by three portals, the central one of which had winged bulls, but the lateral seem to have been without these ornaments; the whole façade being about 330 ft. in extent, north and south. [Illustration: 59. Plan of Palace at Khorsabad, showing the excavations as they were left by M. Botta. No scale.] All these apartments were lined with sculptured slabs, representing mostly either the regal state of the sovereign, his prowess in war, or amusements during peace, but many of them were wholly devoted to religious subjects. Beyond these apartments were many others, covering at least an equal extent of ground, but their walls having been only plastered and painted, the sun-burnt bricks of which they were built have crumbled again to their original mud. It is evident, however, that they were inferior to those already described, both in form and size, and applied to inferior purposes. The mound at Nimroud was so much extended after this palace was built, and so covered by subsequent buildings, that it is now impossible to ascertain either the extent or form of this, which is the only palace of the older dynasty known. It will therefore perhaps be as well to turn at once to Khorsabad, which, being built wholly by one king, and not altered afterwards, will give a clearer idea of the position and arrangements of an Assyrian palace than we can obtain from any one on the Nimroud mound. It has besides this the advantage of being the only one so complete and so completely excavated as to enable us to form a correct idea of what an Assyrian palace really was and of all its arrangements. KHORSABAD.[80] The city of Khorsabad was situated about fifteen miles from Nineveh, in a northerly direction, and was nearly square in plan, measuring about an English mile each way. Nearly in the centre of the north-western wall was a gap, in which was situated the mound on which the palace stood. It seems to have been a peculiarity common to all Assyrian palaces to be so situated. Their builders wisely objected to being surrounded on all sides by houses and walls, and at the same time sought the protection of a walled enclosure to cover the gateways and entrances to their palaces. At Koyunjik and Nimroud the outer face of the palace was covered and protected by the river Tigris; and here the small brook Kausser flows past the fort, and, though now an insignificant stream, it is by no means improbable that it was dammed up so as to form a lake in front of the palace when inhabited. This piece of water may have been further deepened by excavating from it the earth necessary to raise the mound on which the palace stood. [Illustration: 60. Terrace wall at Khorsabad.]