The work comprises two principal sections; an attempt at ellucidating generic relationships in the South African bulbous Liliaceae, and a revision of the species of selected genera. In the first section the approaches investigated were: 1) A re-evaluation of the nature of the bulb apex, showing a likelihood that certain groups possess axillary and other groups terminal inflorescences. 2) An investigation of leaf anatomy, showing that most genera have very similar leaf structure, but that a few genera do differ from this norm. Scilla (Euscilla) generally possesses a distinctly lignified bundle sheath and the Urginea-Drimia group show a great range in structures - including a palisade. 3) An impression technique to determine patterns in the surface of the leaf cuticle. A close correlation appears to exist between these patterns and leaf shape. 4) Scanning Electron Microscope pictures of pollen revealed a considerable uniformity within the group and between this group and most other members of the Liliaceae. 5) Scanning Electron Microscope pictures of seeds revealed two main groups in the bulbous Liliaceae determined on the appearance of the cell walls of the testa. 6) A simple chromatographic technique was not found to be useful. 7) An investigation of chromosome numbers of. 44 samples of 15 species supplemented published data in allowing an analysis to be made of the application of chromosome numbers to generic concepts. Many genera of the Liliaceae - especially in the bulbous group - have extremely variable chromosome numbers. It is considered that numbers are more applicable to tribal than to generic concepts. Few indications of meiotic abnormalities were detected. Phylogenetic and taxonomic aspects of the genera of the bulbous Liliaceae are discussed and a key to the South African bulbous Liliaceae genera constructed. The following generic alterations are proposed: Ledebouria to be split off Scilla. Schizocarphus to be placed in Scilla. Resnova to be placed in Drimiopsis. The Astemme section (= Neobakeria) to be removed from Polyxena and placed in Massonia. The Cape species of Hyacinthus to be placed in Polyxena. Thuranthos, Urgineopsis and Urginea to be placed in Drimia. It is suggested that Schizobasis and Bowiea are of great importance in explaining the origin of Asparagus - possibly even directly from this group. In the generic revisions, thirteen genera are dealt with including typification, synonymy, distribution, field notes, species descriptions and keys. 67 species are dealt with, of which three are new (all in Ledebouria), and 31 new combinations are proposed.

Medicinal bulbs form part of the diversified flora in South Africa. The plants are used extensively in South African traditional medicine in the treatment of various ailments. Due to the ever-increasing demand and the unrestricted collection of medicinal plants from the wild, many of these slow growing bulbous plant species are driven into over-exploitation and extinction. The main parts collected for use are the underground bulbs, leading to the destructive harvesting of the whole plant. This form of plant harvesting poses threats to the long term sustainability of these plant resources from their natural habitats. Sustainable harvesting of these plants should be within the limits of their capacity for self-renewal. However, this seldom occurs with the often inconsiderate medicinal plant gatherers. Conservation of these plants is therefore necessary. A strategy that would take into consideration the sustainable harvesting and perhaps simultaneously provide similar medicinal benefits, would be the substitution of bulbs with leaves of the same plant. This study was aimed at evaluating the seasonal pharmacological and phytochemical properties in bulbs/corms and leaves of medicinal bulbs with a view of promoting the substitution of bulbs with leaves in traditional medicinal use. Four medicinal bulbous plants, Tulbaghia violacea, Hypoxis hemerocallidea, Drimia robusta and Merwilla plumbea were evaluated for the pharmacological and phytochemical properties in their bulbs/corms and leaves in spring, summer, autumn and winter seasons, with a view of promoting the use of leaves as a conservation strategy. Dried plant materials were sequentially extracted with petroleum ether (PE), dichloromethane (DCM), 80% ethanol (EtOH) and water in each season. The extracts were tested for activities against Gram-positive (Bacillus subtilis and Staphylococcus aureus), Gram-negative (Escherichia coli and Klebsiella pneumoniae) bacteria and the fungus Candida albicans using the in vitro microdilution assays to obtain minimum inhibitory concentrations (MIC) and minimum fungicidal concentrations (MFC). The four plant species were also evaluated for their ability to inhibit cyclooxygenase (COX-1 and COX-2) enzymes. Spectrophotometric methods were used to evaluate saponin and phenolic contents of samples from the four plant species in each season. Antibacterial activity was fairly comparable between bulbs/corms and leaves of H. hemerocallidea, T. violacea, and M. plumbea, with at least one extract showing some good activity (MIC < 1 mg/ml) in most of the seasons. Bulb extracts of D. robusta did not show good antibacterial activity while the leaf extracts showed good activity (0.78 mg/ml) against B. subtilis in spring, summer, and autumn and S. aureus (0.78 mg/ml) in autumn. The best antibacterial activity was recorded in winter, with MIC values as low as 0.195 mg/ml from the DCM bulb extracts of T. violacea against K. pneumoniae and S. aureus and PE corm extracts of H. hemerocallidea (0.195 mg/ml) against B. subtilis. Good antibacterial activity from water extracts were only recorded from corm extracts of H. hemerocallidea in summer, autumn and winter, H. hemerocallidea leaf extracts in autumn and winter, and M. plumbea bulb extracts in autumn. The leaf extracts of all the screened plant species demonstrated good fungicidal activity in autumn, with H. hemerocallidea corm water extracts recording an MFC value as low as 0.39 mg/ml. The leaf extracts of H. hemerocallidea (water), D. robusta (DCM) and M. plumbea (DCM) had good MFC values of 0.78 mg/ml each, in spring. The DCM leaf extracts of T. violacea also showed good fungicidal activity (0.78 mg/ml) in summer, while corm water extracts of H. hemerocallidea had an MFC value of 0.39 mg/ml in winter. There were no fungicidal activities recorded from all the bulb extracts in all the seasons. All the PE and DCM extracts in all the tested plant samples recorded between moderate (40-70%) and high (> 70%) COX-1 and COX-2 inhibition levels across all seasons. The EtOH corm extracts of H. hemerocallidea also demonstrated moderate to high inhibitory activity against COX-1 enzyme across all seasons. Bulb and leaf extracts of T. violacea showed selective inhibitory activity for COX-2 enzyme in all the seasons. The highest COX inhibitory levels were recorded in COX-2 from the PE leaf (spring) and bulb (autumn) extracts of T. violacea, with both recording 100% inhibitory activity. Phytochemical analysis revealed higher total phenolic compounds in bulbs/corms and leaves of all the analysed plant species, to be either higher in spring or winter. Plant material collected in autumn had the least levels of total phenolics. An almost similar trend to that of total phenolics was observed for flavonoids, gallotannins and condensed tannins in most plant samples, with higher levels either in spring or winter. Total saponins were consistently higher in winter than in the other seasons in all the screened plant species. There were in some cases, relationships between the peaks in the levels of some phytochemical compounds and the observed levels of bioactivity in different assays. The results obtained from this study demonstrate that the leaves of the screened plant species may substitute or complement bulbs in the treatment of certain ailments in traditional medicine. Thus, plant part substitution can be sustainably utilised in the conservation of these plant species while retaining the same medicinal benefits. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2010.