Thermal inactivation of Mycobacterium paratuberculosis, a suspected human pathogen, was determined in ultrahigh-temperature whole milk. Three strains of M. paratuberculosis were examined for survival at temperatures from 55 to 75°C using a submerged glass capillary tube method. Clumped and declumped suspensions of the cultures were used to determine the rate of heat inactivation and survival at pasteurization temperatures. Methods for declumping M. paratuberculosis included the use of glass beads, vortexing, and passing the cells through a 26-gauge needle. The latter procedure was found to be superior over other methods and did not affect the viability of cells. Capillary tubes filled with milk containing 4 × 106 to 3 × 107 CFU/ml were heated at temperatures ranging from 55 to 75°C. At 55°C, minimal thermal inactivation was observed for clumped and declumped cells. At 58°C, thermal inactivation ranging from 0.3 to 0.7 log reduction was observed for both clumped and declumped suspensions. D values at 60°C ranged from 8.6 to 11 min and 8.2 to 14.1 min for clumped and declumped cells, respectively. At 63°C, the D values ranged from 2.7 to 2.9 and 1.6 to 2.5 min for clumped and declumped cells, respectively. Survival of M. paratuberculosis at initial levels ranging from 44 to 105 CFU/ml at pasteurization treatment (63°C for 30 min and 72°C for 15 s) was also determined. No survivors were observed after incubating plates for up to 4 months on Middlebrook 7H11 agar and up to 2 months on Herrold's egg yolk medium. The sensitivity of the plating method was 1 CFU/250 μl. These results demonstrate that low levels of M. paratuberculosis, as might be found in raw milk, will not survive pasteurization treatments.