The large geophytic monocotyledon, Hypoxis colchicifolia Baker, has been identified for the importance of its corm extracts in the development of a potential non-toxic prodrug for the treatment of inflammation, certain malignancies and HIV-infection. The underground corms of this plant are also commonly used for therapeutic applications in traditional medicine in Kwazulu-Natal where it primarily occurs. A review of published literature revealed, however, that H. colchicifolia plants are currently harvested in an unsustainable manner from traditional collecting sites due largely to population growth, increased land use for urban development and agriculture, and the popularisation of Hypoxis plants for herbal remedies. A further search of historical records established that H. colchicifolia plants were dominant in grassland vegetation prior to 1950, but had rapidly declined since then. Quantitative data subsequently gathered in this study from comparative surveys of both H. colchicifolia and H. hemerocallidea populations from sites with near-pristine, disturbed, burnt and mown grassland vegetation showed for the first time that exposure to human activity and the grassland management practices of mowing and burning incurred not only a 75% reduction in plant density of both these Hypoxis species, but also the total destruction of mature plants of H. colchicifolia in frequently mown and burnt areas. Flowering data recorded in these surveys, and confirmed by monitoring field performance of cultivated H. colchicifolia plants, showed that a contributing factor to the plant's inability to withstand these pressures was that juvenile forms only reached flowering maturity after three to four years growth, thus adversely affecting seedling recruitment. It was concluded therefore that, since Hypoxis species responded differently to mowing and burning, geophytic plants should be considered individually and not as "forbs" during the planning of grassland management programmes for natural conservation areas. The need to cultivate H. colchicifolia to ensure its survival was also established using the new field data gathered in this study. Methods to propagate this species have, however, not been established. Data gathered on all the plants comprising a single population confirmed that mature plants survive to an estimated 20 years and longer in natural areas. Greatest hypoxoside yields were also obtained from corms with a fresh mass of 350g to 400g. Since these corms were estimated to be 10-years-old and older, propagation and cultivation methods that could sustain plant production and survival for long periods, and therefore increased hypoxoside yields, would have to be developed. Several micropropagation systems suitable for the mass production of H. colchicifolia and from which phenotypically normal plantlets were recovered, were therefore established via organogenesis, embryo culture and somatic embryogenesis. The latter cultures have not been reported previously for Hypoxis. In the former culture the toxic effects of phenolic leachates and browning were controlled, and improved plantlet regeneration achieved, by adding polyvinyl pyrrolidone to the medium and introducing distinct sequential aseptic steps into the micropropagation procedure developed. Defined protocols for the different phases of in vitro somatic embryogenesis are not readily available for monocotyledons, however, neither are the factors controlling embryogenesis and organ regeneration known. In this study the process of somatic embryogenesis from excised zygotic embryos of H. colchicifolia was shown to be complex and the resultant cultures very heterogeneous. Although the stage of development of the zygotic embryo explants was important at the time of inoculation, data showed that the induction and regulation of the processes of embryo culture and somatic embryogenesis were ultimately determined by the exogenously applied plant growth regulators. By comparing the different pathways leading to plantlet regeneration, and the morphological stages of development of the structures produced both on solid and in liquid media, not only photographically, but also quantitatively and schematically, the repeated formation of pseudoembryonic structures and neomorphs confirmed that they form an integral part in the in vitro somatic embryogenic pathway of H. colchicifolia. Evidence suggested not only that two types of somatic embryos are produced in the embryogenic cultures of H. colchicifolia, but that the pseudoembryonic structures produced resemble the pseudobulbils produced in polyembryonic cultures of Citrus. The success of the somatic embryogenic cultures was confirmed by the estimation that 28 112 somatic embryos and embryo clusters of H. colchicifolia could be obtained from 16 ml of somatic embryogenic liquid culture. Furthermore phenotypically normal plantlets regenerated from all of the micropropagation procedures developed were successfully transplanted from the laboratory, acclimatized under greenhouse conditions and their horticultural and field performances evaluated. / Thesis (Ph.D.)-University of KwaZulu- Natal, Pietermaritzburg, 2004.