This study was undertaken in the Waterberg Plateau Park in northern Namibia. Field work took place between January 1988 and September 1989; however relevant information collected since, has been included. The roan antelope population in the reserve stems from re introduced animals. A total of 93 were brought in between 1975 and 1981. By 1984 the population peaked at 275 individuals. 186 roan have since been captured and, sold. The main objective of the study was to develop a better understanding of the management requirements for this population of rare antelope. Approximately one-fifth (32 individuals) of the population was marked with eartags or plastic collars or both. Five animals were radio-collared. This is the first detailed study of. roan ante lope in such a large study area (40 000 ha), with such a large population and so many marked individuals. Chapter 3 deals with population dynamics. It is believed that the entire population was accurately counted, aged and sexed. The adult sex ratio was skewed in favour of females, despite the absence of predators large enough to kill adults. The majority of calves were born during the wet season, from the end of August to March. A positive relationship between early season rainfall and the calf crop in the subsequent year was found. The ratio of juveniles per 100 adult females averaged 4.1 over seven years. Recorded mortality was too small to identify significant factors. Two deterministic population models were developed for exploratory simulations and are presented in Chapter 4. Both simulations were fairly successful, giving good correlations with the actual counts. Nutrition is covered in Chapter 5. Simple nutrient budgets revealed that protein and phosphate requirements were probably not met in the dry season. Calculations based on hand selected forage samples and nutrient requirements of cattle, show that adult roan obtained 15-28% of their crude protein requirements and approximately 10% of the phosphate in the dry season. The social system of the roan is described in Chapter 6. Four different clans were recognized, each with a distinct home range. The breeding groups in each clan were unstable, splitting up and rejoining at random. Adult males exhibited territoriality during the breeding season, which is in contrast to most other studies where harem groups were observed. Outside the breeding season, dominant males rejoined bachelor groups or remained solitary. Home range size and configuration were described using different methods. Home ranges averaged 4800 ha for two clans _'0.-_. and the territory size of one male was estimated to be 1200 ha within a much bigger home range. Distinct wet - and_dry - season ranges could be distinguished. Habitat selection is discussed, and was determined primarily by feeding requirements. Forage quality was identified as the major limiting factor and thus the primary determinant of habitat use. In Chapter 7 management recommendations and research proposals are made to ensure sustainable utilisation, by both live capture and game viewing.