Three groups of 16 red deer weaners, 3 months of age, were orally dosed with High (10⁹), Medium (10⁷) or Low (10³) doses of "bovine" strain M. paratuberculosis and a group of 16 weaners received Medium (10⁷) doses of "ovine" strain M. paratuberculosis. There was a fifth Control group of 17 undosed weaners. The High bovine and Control groups were grazed together, the Medium bovine and Low bovine groups were grazed together and the Medium ovine group was grazed alone in separate small paddocks on a quarantined "Johne's infected" area of the farm.Five of the 16 animals in the High bovine group developed early signs of Johne's disease (JD) 6-10 months after challenge and were euthanased. The surviving deer in the five groups, which were all clinically normal, were killed and necropsied 11 months after challenge. Three animals (two High and one Medium "bovine" strain) had small caseous lesions in the intestinal lymph nodes. The remaining animals had no grossly visible lesions. Histological sections of samples of intestines and gut lymph nodes were examined and showed for the "bovine strain" a gradation of lesion severity that was related to the size of the challenge dose of M. paratuberculosis (mean lesion grade = 4.8, 2.9, 0.9 for High, Medium and Low). The Medium "ovine" strain group had a mean score of 0.9, which was the same as the Low "bovine" group. The Control group, which ran with the High "bovine" strain group, had a mean lesion grade of 2.2 indicating natural transmission of infection. These results confirm the prediction that the level of challenge in young animals is a major factor determining the clinical outcome in rising yearling red deer and it also suggests that the "ovine" strain is less pathogenic for red deer than "bovine" strain M. paratuberculosis.