Johnson believes, by between 6 and 24 inches of earth. The tops of the broken-down walls of a caldarium or bath, 9 feet in depth, were likewise covered up with nearly 2 feet of earth. The summit of an arch, leading into an ash-pit 7 feet in depth, was covered up with not more than 8 inches of earth. Whenever a building which has not subsided is covered with earth, we must suppose, either that the upper layers of stone have been at some time carried away by man, or that earth has since been washed down during heavy rain, or blown down during storms, from the adjoining land; and this would be especially apt to occur where the land has long been cultivated. In the above cases the adjoining land is somewhat higher than the three specified sites, as far as I can judge by maps and from information given me by Dr. Johnson. If; however, a great pile of broken stones, mortar, plaster, timber and ashes fell over the remains of any building, their disintegration in the course of time, and the sifting action of worms, would ultimately conceal the whole beneath fine earth. * * * * * _Conclusion_.—The cases given in this chapter show that worms have played a considerable part in the burial and concealment of several Roman and other old buildings in England; but no doubt the washing down of soil from the neighbouring higher lands, and the deposition of dust, have together aided largely in the work of concealment. Dust would be apt to accumulate wherever old broken-down walls projected a little above the then existing surface and thus afforded some shelter. The floors of the old rooms, halls and passages have generally sunk, partly from the settling of the ground, but chiefly from having been undermined by worms; and the sinking has commonly been greater in the middle than near the walls. The walls themselves, whenever their foundations do not lie at a great depth, have been penetrated and undermined by worms, and have consequently subsided. The unequal subsidence thus caused, probably explains the great cracks which may be seen in many ancient walls, as well as their inclination from the perpendicular. CHAPTER V. THE ACTION OF WORMS IN THE DENUDATION OF THE LAND.