In a preceding paper(1) were deduced some of the consequences of assuming that water movement in a heavy clay soil is a process akin to diffusion. The only idealisation of the medium was the assumption that the ratio of k, the diffusion coefficient, to σ, the apparent specific gravity of the dry matter in the soil, is constant, independent of depth, and we shall see that this is justified by the experimental evidence. As long ago as 1907, Buckingham(2) showed that the quantity of moisture moving in a given time from a block of wet soil to a drier block in contact with it was dependent mainly on the difference between the respective water contents, and that there was no definite dependence on the actual water contents themselves. Veihmeyer(3), in 1927, placed blocks of wet soil between blocks of drier soil, arranged in a vertical column. Subsequent moisture profiles were typical of a redistribution obeying diffusion laws, and were nearly the same both for upward and downward movement. He demonstrated furthermore that the movement of the water in these conditions was extremely slow, and that the rate of movement was related to the difference in moisture content between the dry and wet soil. Unfortunately insufficient numerical data were given to apply diffusion theory to Veihmeyer's experimental results.