Shani and Or [1995] propose estimating hydraulic conductivity K in terms of the soil-water pressure by measuring the pressure P within and the discharge Q out of a spherical cavity under steady state conditions. The authors have already pointed out some of the difficulties inherent with the method, for example, the assumption of a soil which follows the model of Gardner [1958], defining an equivalent spherical cavity radius r(o) and the eventual presence of cracks. Possible methods to alleviate some of those difficulties, as well as additional potential sources of errors, are discussed.