The _zikkurat_ of the palace at Khorsabad, placed to the east of the seraglio buildings, has still at the present day three complete steps and the beginning of a fourth; the first describes on the ground a square of 141 ft. each way; each stage is 20 feet high, which gives us reason to believe that the structure was as high as it was broad at the base--a peculiarity already noted by Herodotus and Strabo in the temple of Bel. The stages laid bare by the French excavations were still partly coloured by means of enamelled stucco, the lowest stage white, the second black, the third reddish purple, the fourth blue. Among the ruins of the tower were found numerous fragments of enamelled bricks, coloured vermilion, silver grey and gold, which proves that the tower had seven stages of different colours. It has been remarked that Herodotus (i. 98), gives to the fortress of Ecbatana, in Media, the arrangement of a gigantic tower in stages, the colours of which are similar to those of the _zikkurat_ of Khorsabad. There were, according to him, seven concentric enclosures, the most spacious being as large as Athens, while the battlements of each enclosure rose higher than those outside them. “The battlements of the first wall are of white stone; those of the second of black stone; those of the fourth blue; those of the fifth vermilion.... The two last walls are plated, the one with silver, the other with gold.” [Illustration: Fig. 53.--The staged tower of Khorsabad (restoration by V. Place).] The explorers of Mugheir thought that they recognised, in spite of the bad state of the ruins, that the _zikkurat_ of Ur was constructed in such a way that the stages did not rise exactly in the middle of the square platform of the lower stage which served as their base; they were nearer to one of the sides, so that they present on one side much narrower terraces than on the other three. This observation is confirmed by a bas-relief in the British Museum, unfortunately very rough, in which, however, we distinguish clearly the greater width of the terraces on one side and their corresponding narrowness on the other. On the other hand the slope of each terrace proves that it ascended like a screw, and that there was no staircase cut in each of the stages to put them in communication with each other. This is, moreover, what is observed at Khorsabad: the ascent to the summit of the ruins of the fourth stage is by a quadrangular sloping path which mounts gently as it winds round in a spiral form.