In early 2001, tuberculosis-like lesions were detected in three hunter-killed red deer (Cervus elaphus) in the Brotonne Forest (Normandy, France), and Mycobacterium bovis was isolated. In subsequent hunting seasons, two surveys were conducted in the area. In the first survey (2001–02 hunting season), nine (13%) of 72 red deer sampled were positive for M. bovis. In the 2005–06 hunting season, the prevalence of M. bovis infection increased to 24% (χ²=3.85, df=1, P=0.05; 33 positive among 138 sampled). The prevalence remained stable in juveniles, but it increased significantly in adults: from 13% in 2001–02 to 32% in 2005–06 (χ²=5.13, df=1, P=0.02). Wild boar (Sus scrofa) were heavily infected in both surveys. One roe deer (Capreolus capreolus) and one red fox (Vulpes vulpes) also tested positive in the second survey. Mycobacterium bovis was not isolated from Eurasian badgers (Meles meles). Spoligotyping and mycobacterial interspersed repetitive unit-variable number tandem repeat analysis demonstrated that all M. bovis strains isolated from wildlife were of the same genotype. Thus, the wildlife outbreak involved only a single strain, and this strain was the same as that circulating in nearby cattle herds since 1995. Sensitivity, specificity, and predictive values of the presence of macroscopic lesions as a diagnostic criterion were evaluated from the data obtained from red deer. Necropsy seems to be satisfactory as a routine tool to monitor the disease in wild red deer populations in which bovine tuberculosis has become established.