We have no published means of estimating, from this source or otherwise, the history of the development of this ancient capital. But some conjectures, as a working hypothesis, may be made from the probabilities of the case with this date as a basis, awaiting meanwhile further illumination from Dr. Winckler and his colleagues. In the first place, as to the date of the main fortifications, though the period of empire is not often the time of building home defences, yet in this case the deliberate and vast nature of the outer walls conveys no impression of a stricken people hastening to defend themselves, nor even of precipitation. The scheme and details are carried out with dignity, thoroughness, and elaboration. It was the product of a prosperous age, dictated by prudence rather than immediate conscious necessity. Yet the pride of Hittite power soon passed; even while treating on equal terms with the courts of Thebes and Babylon, the shadow of the Assyrian armies already clouded the eastern horizon; and the menace of barbarian northern hordes was probably ever present, particularly as their offensive powers weakened. It may safely be supposed that their city must have been prepared against assault at any rate before the inroads of the Phrygian Muski, in the twelfth century B.C. And secondly, with regard to the palace just considered, built as it is upon the ruins of one which flourished in the time of Rameses the Great, it represents a reconstruction and re-establishment of royal state at some time subsequent. As to the date of this revival there is little evidence. From the plan of the palace it may be conjectured to have preceded any wide spreading of Assyrian influences; and from our own observations it was probably contemporary with a certain class of coloured pottery, which at Sakje-Geuzi[495] was already passing out of vogue at the beginning of the first millennium B.C. Upon this point it is interesting to notice that the difference of axial direction between this and the buried palace, namely, 2½°, would, if astronomically dictated, suggest a difference of date amounting to about two hundred and thirty years,[496] assigning the period of restoration to the eleventh century B.C.