In these texts another sanctuary included in the same building and placed half way between the base and summit is mentioned. This was the sepulchral chamber of Bel-Merodach in which his oracle was consulted; in M. Chipiez's restoration the entrance to this sanctuary is placed in the middle of the fifth story.

The vast esplanade about the base of the temple was suggested by the description of Herodotus. It is borne by two colossal plinths flanked and retained by buttresses. In our plate the lower of these two plinths is only hinted at in the two bottom corners. In the distance behind the temple itself may be seen one of those embattled walls which divided Babylon into so many fortresses, and, still farther away, another group of large buildings surrounded by a wall and the ordinary houses of the city.

This double-ramped type is at once the most beautiful and the most workmanlike of those offered by these staged towers. With a single ramp we get a tower whose four faces are repetitions of each other, but here we have a true façade, on which a happy contrast is established between the unbroken stages and those upon which the ramps appear—between oblique lines and lines parallel with the soil. The building gains in repose and solidity, and its true scale becomes more evident than when the eye is led insensibly from base to summit by a monotonous spiral.

Figs. 180-182.—Square Assyrian temple. Longitudinal section, horizontal section and plan.

We cannot positively affirm that the architects of Mesopotamia understood and made use of the system just described; there is no positive evidence on the point.[469] It contains, however, nothing but a logical development from the premises, nothing but what is in perfect keeping with Mesopotamian habits, nothing that involves difficulties of execution or construction beyond those over which we know them to have triumphed. Besides, we have proofs that they were not content to go on servilely reproducing one and the same type for twenty centuries; their temples were not all shaped in the same mould. The type of the Mugheir temple differed sensibly from that of the Khorsabad Observatory. One of the Kouyundjik sculptures reveals a curious variant of the traditional theme, so far as Assyria was concerned, in an arrangement of the staged tower that we should never have suspected but for the survival of this relief ([Fig. 34]). The picture in question is no doubt very much abridged and far from true to the proportions of the original, but yet it has furnished M. Chipiez with the elements of a restoration in which conjecture has had very little to say. This we have called the Square Assyrian Temple (see [Plate IV.] and [Figs. 180-182]).