Batch type and column experiments were used to study competitive adsorption-desorption and transport for Cd, Cu, Ni and Zn in a mine soil, both un-amended and amended with mussel shell. Batch type experiments showed that adsorption was affected by the added concentration of the metals, generally following the sequence Cu>Zn>Cd≈Ni. Metal desorbed was a function of the dose of metal added, as well as of the dose of shell amendment, being relevant that even when the highest dose of metal (2300 μM) was added, the 24 g kg(-1) shell amendment caused a drastic diminution in the amount of metal desorbed. Column experiments showed that even the lowest dose of the shell amendment (6 g kg(-1)) caused a strong retention of the 4 heavy metals assayed, whereas using the 24 g kg(-1) shell amendment no metal was detected in the effluent during the time of the experiment. The mass of metal retained in the un-amended soil was very different for the various metals assayed, but the amendment with 6 g kg(-1) shell increased this retention in all cases, and the 24 g kg(-1) amendment caused almost 100% retention for all 4 metals. The retardation factor (R) suffered an overall increase as a function of the shell dose; the profile distribution of the 4 heavy metals was homogeneous through the un-amended soil into the column, but the shell amendment clearly decreased the solute transport affecting these metals, causing its concentration in the first centimeters of the soil profile.