Let us now penetrate into the interior of the Chaldæan edifice, of which the blind and dumb walls leave in our imagination an impression of gloom and cold uniformity. The walls seem never to have exhibited the smallest architectural decoration; they are entirely bare, and only characterised from time to time by depressions and projections; no traces of mouldings, of plinths, of cornices, and of those devices to which the architects of all countries have recourse in order to break the lines of the walls, and to call forth effects of light and shade. It must be supposed that the interior decoration of the palace consisted entirely of colouring and hanging draperies. The thickness of the wall varies from 8 ft. 6 in. to 2 ft. 7 in. All the partitions cut one another at right angles, forming thirty-six square or rectangular chambers; the largest measures 39 ft. 4 in. by 12 ft. 2 in., and the smallest 10 ft. 11 in. by 9 ft. 9 in. The disproportion which exists, especially in the state saloon, between the length and breadth, the extreme thickness of the walls, even of those which are the least important in the structure, form essential peculiarities to which we shall draw attention later in the Assyrian edifices. At Nineveh it has been proved that it is the thrust of the semicircular vaulting, which roofs the chambers, that has forced the architect to bring the parallel walls near to one another and to give them an enormous thickness. Are we, in the absence of palpable proof, to draw the same conclusion with regard to the palaces of old Chaldæa? Are we authorised to assert that the vault was known three thousand years before our era? In a word, how were the halls of Gudea’s building covered? Was it everywhere by means of transverse rafters supporting a floor and a terrace? or was it oftener by a bricked vault? As far as we have read M. de Sarzec’s narrative, or M. Heuzey’s studies on the excavations of Tello, we have found no direct answer to this question. Perhaps the present state of the ruins or the successive alterations to which the primitive structure has been subjected do not allow a categorical solution of the problem to be given. However, important indications authorise us to believe that the Chaldæans of the time of Gudea already understood the vault and used it for roofing their houses. In several parts even of the palace of Tello, M. de Sarzec found small vaulted passages, 3 ft. 3 in. high and 1 ft. 11 in.[9] thick, in a perfect state of preservation; in one of the secondary mounds he brought to light a small vaulted drain which carried the sewage of the town far away into the plain. Taylor found, in an underground chamber of the necropolis at Mugheir, the most primitive kind of vault that has ever been known--that called the corbelled vault. In this false vault the courses of bricks ascend in parallel rows on each side until they meet one another, every fresh course projecting perceptibly beyond that