The early arches in the tomb-passages in Egypt are supposed to owe their origin to the removal of the lower part of the buttress-walls erected to keep the side walls of the passages from collapsing: such buttress-walls would of course fulfil their function in preventing the side walls from falling in, but they would frustrate their own ends by completely blocking the passage, thus rendering it perfectly useless. Accordingly the lower portion of the buttress-wall was removed, the upper part being allowed to remain, and forming in fact a rudimentary arch, and it is possible that the Babylonian arch owes its origin to like fortuitous circumstances. It is perhaps more probable, however, that the origin of the arch-shaped structure, if not the discovery of the principle of the arch, is to be traced to the peculiar form assumed by the native reed-huts, which doubtless bore a close resemblance to those commonly used in the Euphrates valley to-day. This view is advocated by Heuzey, and is the one which Hilprecht is disposed to favour. [Illustration: Fig. 15.—Early T-shaped Arch at Nippur. (Cf. Hilprecht, _Explorations_, p. 399.)] [Illustration: Fig. 16.—Arch at Tellô. (Cf. _Déc. en Chald._, Pl. 57 (bis), 1.)] Most of the ancient buildings of Babylonia have succumbed to the concurrent ravages of time and climate, and have consequently bequeathed to us very little material for the study of Babylonian architecture; the roofs of buildings, and of the chambers comprised therein, have long since ceased to be, and we can thus only theorize as to the general mode of roofing adopted, but the drains and aqueducts constructed beneath the buildings have luckily survived to tell their tale, and we owe our knowledge of the early existence of the arch in Babylonia chiefly to these comparatively insignificant remains.