The small Cape mountain zebra population in Kammanassie Nature Reserve (KNR) is one of three relict Cape mountain zebra populations. Although the genetic diversity of this population is low, it is genetically distinct from the other two populations, the Gamka Mountain Nature Reserve and Mountain Zebra National Park populations, and thus vital for the conservation of this subspecies. The management of the Cape mountain zebra metapopulation requires the mixing of at least some animals from the three relict populations, but this process has been hampered by the relatively slow growth of the KNR population. We investigated the influence of habitat, fire and rainfall on mountain zebra population growth in KNR and used a diffusion model to perform a population viability analysis on the population. Mountain Fynbos covers more than 80% of the reserve but of this, only the Arid Restioid Fynbos and Waboomveld habitats were preferred by mountain zebra. The suitability of these habitats for mountain zebra is, however, highly variable and dependent on fire. The analysis also indicated that the population growth of mountain zebra is related to the incidence of fire and rainfall, but that the stochastic nature of these factors in space and time, limits the growth of the population. Although the population has a low probability of reaching quasi-extinction in the next 50 years (G = 0.123 × 10⁻⁴), we argue that the growth of this population will continue to be slow under the present ‘hands off’ management policy. A number of options to facilitate the growth of this population are considered;changing fire management in the habitat preferred by mountain zebra;acquiring adjacent land; and the translocation of mountain zebra onto adjacent land. Of these options, the latter two need to be implemented as part of a strategy to promote the management of the Cape mountain zebra metapopulation.