Speaking generally, there can be no study of the architecture in elevation, as the ruins afford no examples of building in stone. We are forced to confine ourselves to examining the ground plans of the buildings, the pavements, and the foundations. Everything else has fallen to pieces, or been reduced to powder. In some of the tells of Chaldea, however, remains of temples and palaces have been found with the lower courses still in position. It is in consequence of this that we have been able to give M. de Sarzec’s reconstruction of the plan of the palace of Gudea at Tello (fig. 2, p. 9). The American Archæological Mission has also discovered at Niffer (Nippur) the lower courses of a _zikkurat_, or staged tower, in excellent preservation. In this region the building materials were frequently kiln-baked bricks, and mortar made of bitumen of such exquisite quality as to render the walls of such consistency that at the present day it is necessary to use powder to demolish them. At Susa, so far as investigations have been carried at present, it appears that crude bricks were usually employed in building, and without the bitumen mortar, with the result that the walls were easily demolished, both by the pick-axe of the intentional devastator and by the corrosive action of the weather. “Thus on all sides,” says M. de Morgan, “reigns the greatest of confusion of piled-up materials.” One exception has thus far been found, a small temple of the god Shushinak, where the plan can be traced, owing to the basement having been constructed of baked bricks, with revetments of glazed sandstone. Large numbers of tiles have also been found, enamelled with yellow or pale green and bearing the name of king Shutruk-Nakhunta. This is the class of decoration which developed during the Achæmenid period, of which we have previously given some specimens.[105] The hiding-places found under the pavement of the temple yielded a number of votive objects, which are exceedingly interesting and valuable. Of these we shall speak later; we must now only mention the brick columns, the principle of which has been studied, and of which we give a representation (fig. 245). [Illustration: Fig. 245.--Brick Column, Susa.] This column is composed of a number of bricks, all of which bear the name and protocol formula of the royal builder, Shutruk-Nakhunta. Some of these bricks are square, others round, and others are segments of circles. The figure here given sufficiently indicates how they are arranged, and it will be seen that the principle is precisely the same as in the similar constructions at Tello (see p. 10, fig. 3).