The action of sub-tropical rains for centuries has destroyed whole lengths of walls. For instance, a trench which occupied half a dozen labourers two days to excavate was, after a heavy shower lasting but an hour, completely filled up by mud streams from a higher level. There is hardly a wall on the Acropolis Hill that has not had to bear some added weight of silted soil from higher levels, and these in places have been so extensive that when accumulated on the upper sides of walls the effect has been to push the wall bodily over. In this way the terraces of enclosures round the north, west, and south faces of the Acropolis have in most instances been entirely filled up and buried, while in others the outer and down-side wall has been burst through and destroyed. Streams of water during storms of real African violence have worn deep channels along the bases of some of the walls, exposing the foundations which bridge across the holes, the water causing the decomposition of the cement bed of the foundations and making the wall throughout its complete height to sway downwards and to bulge threateningly outwards. Some of these water-made holes up against the bases of the walls contained damp and moisture all through the dry season, especially those on the south side of walls where the holes were protected from the sun. In as many cases as possible for the time engaged on the preservation work (1902–4) these spots were levelled, and catchment areas were made, so that for the future no rain-water can lodge there, but the waving lines of the courses in the walls still show where these holes existed. In a similar way block débris falling from higher levels has lodged behind lower walls and eventually pressed them over. In some instances on the Acropolis a mass of walling has fallen from a great height and completely demolished walls below. These were no gradual dilapidations, but instances where sections of the higher wall had gone completely over _en masse_. Such falls almost entirely explain the damage done to the outer walls of the South-East Ancient Ascent, lengths of which have evidently been made good by Kafirs of a very old period, as the well-built ancient foundations can be seen below the later walls.