In Chaldaea and Assyria the conditions were different. Supposing the architecture of those two countries to be yet entire, should we find in it vaults rivalling in age the arch in a tomb at Abydos which Mariette attributes to the sixth dynasty?[294] Probably not. So far as we can judge, Chaldaean civilization does not date from so remote a past as that of Egypt, but it appears certain that the principles of the vault were discovered and put in practice by the Chaldees long before the comparatively modern times in which the segmental and pointed arches of Nineveh were erected. The latter alone are preserved because they have been hidden during all these centuries under the heaped-up ruins of the buildings to which they belonged, while those of Chaldaea have been carried away piece by piece, and their materials used again and again by the modern population of Mesopotamia. In spite, however, of the absence of such direct evidence, we may affirm without fear that the Chaldaean architects soon discovered the principle of the arch, and used it at least in its simplest and least complex forms. We are led to these conclusions not only by their restriction to small units of construction--a restriction which is sure, sooner or later, to lead to the discovery in question--but also by induction from the monuments we have just been studying. The arches under the hanging gardens of Babylon, the vaults of the sewers and gateways, the domes that covered the great square chambers in the Ninevite palaces--all these were derived, we may be sure, from the ancient civilization. We cannot believe that such consummate skill in the management of a difficult matter was arrived at in a day. The purely empiric knowledge of statics it implies could only have been accumulated by a long series of more or less happy experiments.