In this research, the CGP 35348 influence of Rw on Ks is questioned and a procedure is developed to demonstrate it. The published pedotransfer functions for homogenous fine soils are used: ten PTFs to predict Ks (Ks-PTFs), from which three depend on OM and soil texture and seven depend on BD, OM and texture; and thirty two PTFs to estimate the bulk density (BD-PTFs) based on measured data of soil texture and OM. It is generally recommended that the PTFs should only be applied to land similar to the soil they were based on, and should not be directly exported to other geological and soil environments unless checking their validity or even calibrating them by direct measurements. Here, a detailed methodology is developed for applying these existing PTFs in the case of a stony soil. We don\'t calibrate any of them but search for the better coupling with regard to a few (15) measurements of Ks. Mainly, the combinations of 32 BD-PTFs with 7 Ks-PTFs are tested by statistical criteria for the proximity of predicted Ks to measured Ks at each of 15 sampling points. The discrepancy put into evidence the need of taking into account the rock fragment content. Then, considering RNA polymerase the 32 BD-PTFs give possible values of BDm0 at each point, the relation of Torri et al. (1994) is used to estimate 32 values of the bulk density BDm (Eq. (2)). Finally, some combinations (Torri-BD-Ks-PTFs) are selected to be used in stony soils.