Mycobacterium bovis has a wide host range that includes several wildlife species, and this can hamper attempts to eradicate bovine tuberculosis from livestock. The purpose of this study was to determine if common rodent species, namely meadow voles (Microtus pennsylvanicus), house mice (Mus musculus), and Norway rats (Rattus norvegicus), that inhabit the bovine tuberculosis endemic area of Michigan, can be experimentally infected with M. bovis. The objectives of the study were: 1) to determine if these rodent species can be infected, and if so, to document attendant pathologic processes/pathogenesis; 2) to detect any fecal shedding of M. bovis; and 3) to evaluate the relative susceptibility of the three species to M. bovis infection. For each species (n=36) there were two treatment (n=12/group) and one or two control groups depending on species (n=6–12/group); the maximum study duration was 60 days. The meadow vole treatments consisted of high dose inocula that were given by oral or intranasal routes, whereas the house mice and Norway rats were given only oral inocula at either a high or low dose. Of the three species, meadow voles were most susceptible to M. bovis infection. Upon intranasal inoculation, all 12 voles were infected as determined by gross and microscopic lesions and culture of M. bovis from tissue and feces. Seven of the 12 meadow voles inoculated orally were infected. House mice also were susceptible; M. bovis was isolated from 14 of 24 animals. Only one Norway rat in the high dose treatment group was positive by culture and this was the only animal from which minimal attendant lesions were observed. Results of this study indicate that meadow voles and house mice can be infected with M. bovis and might serve as spillover hosts. Concerted efforts should, therefore, be made to reduce or eliminate these rodents on premises where M. bovis-infected livestock are present.