We used data from a six-year (2005-2010) study of a nine-banded armadillo (Dasypus novemcinctus) population in Mississippi, USA, to examine patterns and consequences of exposure to Mycobacterium leprae, the causative agent in producing leprosy. No juveniles or yearlings tested seropositive, implying there was no vertical transmission of infection. Older adults were more likely to be seropositive than younger adults and, in some years, females were more likely to be seropositive than males. Prevalence of infection varied from year to year but there was no spatial clustering of seropositive individuals. The latter result suggests transmission of infection was via some form of homogeneous mixing process (e.g. contaminated soil). Population modeling of host and pathogen dynamics showed that leprosy reduced annual survival of adult armadillos by ~15% and population growth rate by ~13%. The probability that a non-leprous adult armadillo would become seropositive from one year to the next was 0.18, with no possibility of recovery. The net reproductive number (R0) was estimated to be 1.36, which suggests a 36% increase in seroprevalence per leprosy generation. Assuming a leprosy generation time of 3-5 years, we predict that M. leprae will spread within the population at the rate of 7-12% per year. Nine-banded armadillos are the only known non-human animal hosts of leprosy in the Western Hemisphere, and current evidence indicates likely zoonotic transmission. Thus, our results are important not just for understanding the dynamics of disease in a wildlife population, but may have important implications for human public health as well.