Some Leishmania species affect humans in two principal forms: visceral and cutaneous leishmaniosis (CL). Several studies have identified dogs as the main reservoirs of the visceral leishmaniosis (VL) caused by Leishmania infantum. The purpose of this work was to carry out a survey of the canine population associated with human cases of American tegumentary leishmaniosis (ATL), in order to establish the clinical, parasitological, serological and immunological characteristics of the canine disease, in an endemic region for both ATL and Chagas' disease in the province of Salta, in northwestern Argentina. Two hundred and eight dogs from the endemic area were examined and 41 (19.7%) of them presented lesions compatible with leishmaniosis. In order to investigate the presence of antibodies against Leishmania spp. and Trypanosoma cruzi, sera were screened by ELISA using two complex antigens from these parasites and, because of cross-reactions between them, a specific antigen for diagnosis of T. cruzi infection. Sixty-two (29.8%) of 208 dogs were positive for the complex antigen F45 from Leishmania and 50 (24%) were positive for the complex antigen F105 from T. cruzi. Nine dogs (4.3%) were positive for the specific Ag163B6-cruzipain suggesting that these dogs were truly infected with T. cruzi. Furthermore, three of these nine dogs presented Leishmania sp. in their skin lesions and therefore were considered as infected by both, T. cruzi and Leishmania parasites. The prevalence of Leishmania infection detected by lesions and/or positive serology was 27.4% (57/208). On the basis of previous observations regarding the clustered appearance of human ATL, the dog population was divided into two groups: zone A, dogs living within a 100 m radius from houses with human cases, and zone B, dogs living beyond this limit. The prevalence of ATL in dogs was significantly higher in zone A (34.6%) than in zone B (7.3%), suggesting a strong correlation between canine and human cases. The average time required for a parasitological diagnosis by microscopy was six times longer for dog samples than human ones, and the average number of parasites per 100 microscopic fields was 14-fold lower in canine samples. The high prevalence of Leishmania infection and the close association with human cases, demonstrated that dogs are a very susceptible host for Leishmania infection, but the scarcity of parasites in their lesions suggests that they may not be the main reservoir of the parasite in this endemic area.