An experimental investigation into the influence of artificially induced trauma in the production of leishmanial metastatic lesions and into the possible role played by Leishmania-reactive T cell populations in the metastatic process was carried out. Trauma was induced by incising a small cut into the shaved rump of Leishmania amazonensis-infected BALB/c mice. Ten days after the trauma, mice were killed to quantify the parasite load in the traumatic lesion or in the equivalent area in nontraumatized mice, by limiting dilution analysis. Results demonstrated that metastatic lesions occurred earlier in traumatized animals and that parasites could be detected sooner in traumatic lesions than in equivalent areas in nontraumatized mice. When lymph node cells from L. amazonensis antigen-immunized BALB/c mice were adoptively transferred intravenously to L. amazonensis-infected syngeneic mice, the parasite load in the metastatic lesions was greater in the animals that received La Ag-reactive T cells than in the controls. When CD4(+)- or CD8(+)-depleted T cell populations from La Ag-immunized mice were adoptively transferred to infected traumatized or nontraumatized animals, we observed that the metastatic lesions in CD4(+)-inoculated animals had a greater number of parasites than the lesions in mice from all other groups. Thus, a new and reliable mouse model for studying the mechanisms involved in leishmanial metastasis is described.