Breeding for resistance is the primary control measure for brown rust of sugarcane. Resistance screening utilizing natural infection symptom severity ratings provides erratic results. Therefore, a method accomplishing infection and disease expression under controlled conditions was evaluated to determine whether it could provide accurate resistance ratings for seedlings and clones with known and unknown reactions. Seedlings from crosses between parents with different levels of resistance were inoculated with increasing concentrations of urediniospores. Inoculum concentration affected disease severity and the frequency of resistant progeny in crosses. Brown rust resistance is a heritable trait; however, parental reaction was not a consistent determinant of progeny distribution across resistance rating categories. These results suggest that seedling inoculation may be misleading for the evaluation of brown rust resistance. Clone resistance reactions could not be reliably determined for susceptible clones in single inoculations. Ratings for controlled-conditions inoculation and field natural infection severity were not correlated. Multiple inoculations under controlled conditions accurately identified resistant and susceptible clones, with severe infection resulting from any single inoculation indicating susceptibility. Therefore, controlled-conditions inoculation has the potential to be useful in limited studies to characterize parents in a recurrent selection program and for basic studies of resistance to brown rust.