The black shank (Phytophthora parasitica var. nicotianae) fungus is a very destructive tobacco disease which is responsible for great losses to farmers worldwide. This disease is also a problem in South Africa, as the most popular South African air-cured tobacco cultivar, CDL28, is very susceptible to black shank. This diallel study focussed on finding the most suitable black shank resistance source to include in a resistance breeding programme with CDL28. Four cultivars were crossed in all possible combinations and planted as an F1 field and greenhouse trial. The F1's were selfed to obtain a segregating F2 population, planted in a greenhouse, to be compared with the mean values of the F1 trials. The field trial was exposed to natural infection while the greenhouse trials were root inoculated. The general combining ability effects of the four parent cultivars differed significantly from each other. The specific combining ability effects of the F1 trials were non significant. These experimental results suggest that additive genetic effects were involved in black shank resistance. The Beinhart 1000-1 source of resistance was significantly better than the Florida 301 source. Burley 37, which possesses the Florida 301 as well as another source of resistance performed better than Domkrag with only the Florida 301 source of resistance. In order to incorporate black shank resistance in CDL28, Beinhart 1000-1 and Burley 37 can be used in a backcross breeding programme with CDL28 which can solve the problems encountered in the cultivation of CDL28 in the presence of black shank. / Dissertation (MSc (Genetics))--University of Pretoria, 2006. / Genetics / unrestricted