The most important of these buildings, and the most interesting, is the ruin at Warka, which Loftus calls _Wuswas_ (Fig. 172, Vol. I., letter B on the plan).[34] Unfortunately his explorations were very partial and his description is very summary, while his plan of the ruin only gives a small part of it (Fig. 9). There is, however, enough to show the general character of the structure. The latter stood upon a rectangular mound about 660 feet long and 500 wide. In spite of the enormous accumulation of rubbish, Loftus succeeded in making out an open door in the outer wall, and several chambers of different sizes communicating with a large court. There was the same thickness of wall and the same absence of symmetry as at Khorsabad; the openings were not in the middle of the rooms. In the long wall, decorated with panels and grooves, which still stands among the ruins to a height of about twenty-four feet and a length of about 172 feet, the posterior façade, through which there was no means of ingress and egress, may be recognized. We have already copied Loftus’s reproduction of this façade for the sake of its decoration (Fig. 100, Vol. I.). [Illustration: FIG. 9.--Plan of a palace at Warka; from Loftus.] The building at Sirtella (_Tello_) in which M. de Sarzec discovered such curious statues, was less extensive; it was only about 175 feet long by 102 wide. The faces of the parallelogram were slightly convex, giving to the building something of the general form of a terra-cotta tub (Fig. 150, Vol I.). Here the excavations were pushed far enough to give us a better idea of the general arrangement than we can get at Warka. A great central court, about which numerous square and oblong apartments are arranged, has been cleared; there is a separate quarter, which may be the harem; at one angle of the court the massive stages of a _zigguratt_ may be recognized. The walls are entirely of burnt brick. They are decorated only on the principal façade, where the ornaments belong to the same class as those of Wuswas--semi-columns mixed with grooves in which the elevation of a stepped battlement is reproduced horizontally. In none of the ruins of habitations found in this district by the English explorers, were the chambers other than rectangular. Taylor cleared a few halls in two buildings at Mugheir (Fig. 10) and Abou-Sharein (Figs. 11 and 12) respectively. Both of these stood on artificial mounds, and it is difficult to believe that they were private dwellings. The walls of several rooms at Mugheir seemed to have been decorated with glazed bricks; at Abou-Sharein there was nothing but roughly painted stucco. In one chamber the figure of a man with a bird on his fist might yet be distinguished. [Illustration: FIG. 10.--Plan of chambers at Mugheir; from Taylor.] [Illustration: FIG. 11.--Plan of chambers at Abou-Sharein; from Taylor.] [Illustration: FIG. 12.--Plan of chambers at Abou-Sharein; from Taylor.]