Morphological variation and a cladistic analysis of the large, endemic Southern African genus Lachenalia Jacq.f. ex Murray (Hyacinthaceae: Massonieae) is presented. Its close taxonomic relationship with the small endemic sympatric genus Polyxena Kunth (which has been included in the morphological and cladistic study) is discussed. The inclusion of Polyxena within Lachenalia is supported. One hundred and twenty species (139 taxa), comprising 115 Lachenalia and five Polyxena species are recognised. A wide range of morphological characters were analysed, including macromorphology, micromorphology, anatomy and palynology. A discussion and comparison of karyological data is also presented. A brief historical background, species diversity maps, figures, tables, appendices and illustrations of anatomical, micromorphological and macromorphological characters, and cladistic data, are presented, as well as discussions of pollination biology and phytogeography. This work is based on species studied in their natural habitats as well as under cultivation, and from representative herbarium specimens examined from BOL, NBG, PRE and SAM. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.

Trials were done on four Lachenalia cultivars (Romaud, Robyn, Rolina and Romelia) to improve flower quality and inflorescence stem length and to evaluate vase life, as these criteria are important for the cut flower grower. Flowering size bulbs were grown under five different shade nets (white 18%, green 40%, black 40%, black 55%, black 7O%) and in the open as a control. Under each shade net, bulbs were planted at a low planting density of two bulb diameters apart (56 bulbs/m2), as well as a high planting density of one bulb diameter apart (111 bulbs/m2). The effect (If the growth hormone gibberellic acid (GA3) on plant growth was investigated to determine if longer stems could be obtained. The plants were treated with gibberellic acid at 10 ppm and 0 ppm, which was applied as a foliar spray and as El bulb dip treatment. It is necessary to identify and describe specific flowering stages to be able to establish a standard rating system for Lachenalia vase life. The morphology of Lachenalia cultivar Romaud was described to determine the real succession of opening flowers on the inflorescence. The stages of 'first flower', 'full flower' and '50% wilt' were described. A high photosynthetically active radiation (PAR) measurement of 1250 µ.mol.m-2.s-1 in the control, resulted in the shortest inflorescence stem length of 13 cm for all four cultivars compared to the rest of the shade nets. The longest inflorescence stem length of 24 cm was produced for cultivars Rolina and Romelia by a low PAR of 400 µ.mol.m-2.s-1 under the black 70% shade net. The inflorescence stem length of all four Lachenalia cultivars tend to decrease to a minimum of 13 cm when average temperatures are 30oC and higher and to a maximum of 24 cm when temperatures are in the range from 24 to 27oC. A long vase life of 12 to 14 days for cultivar Rolina was observed under the green 40% and black 40% shade nets, compared to 10 to 11 days in the control. The green 40% shade produced a long vase life of 14 to 16 days for cultivar Romelia, compared to the 12 to 14 days in the control. Planting density significantly increased inflorescence stem length by about 2.5 cm and vase life by 2 days at the high planting density compared to the low planting density for all four cultivars. Inflorescence stem length for cultivars Romaud and Romelia increased significantly by 3 cm for both the GA3 foliar spray and bulb dip treatments. The number of flowers per inflorescence decreased significantly by about 3 to 5 flowers for cultivar Romaud and Romelia when GA3 was applied as a foliar spray or bulb dip treatment. A significant increase in vase life (2 days) of GA3 treated plants was observed. All four Lachenalia c:ultivars are suitable for cut flower production, as inflorescence stem lengths were either just below or above the 20 cm mark and vase life was longer than the five to six days required by the cut flower industry. / Dissertation (MSc (Agric): Horticulture)--University of Pretoria, 2006. / Plant Production and Soil Science / unrestricted

Two field experiments were carried out. In the first experiment, 13 variables were studied for 3 inbred groups as influenced by genetic differences, source of pollen, soil moisture, fertilizers , and size of bulbs. Path-coefficient analysis was made of var ious insect pollinators influencing seed yield in different inbreds, and of components of seed production in all inbreds together. In the second experiment, 4 variables were studied for one male-sterile inbred and one male-fertile inbred as influenced by soil moisture and fertilizers. Genetic differences among inbreds were the most important source of variation in seed yields. This points out that a major portion of the onion seed problem could be solved through plant breeding. The inbreds differed very significantly in t heir attractiveness to pollinating insects, probably as a result of varying genetic vigor which influenced nectar formation. Most of the inbreds had negative correlation or no correlation between the number of honey bee nectar collectors and seed yield. The separation of correlation with honey bee nectar and pollen collectors into components of direct and indirect effects indicates that the number of pollinating insects was excessive for all inbreds in the experiment. Negative direct effects on seed yield and negative indirect effects through percent fertilized flowers and seeds per fertilized flower disclose some important aspects as possible causes for low onion seed yields: 1. Depletion of assimilated food needed for the development of fertilized ovules through an overdraw on nectar by nectar collectors. 2. Water deficit affecting fertilized developing ovules due to increased transpiration resulting from overhandling of flowers by honey bees, especially by pollen collectors, which through nervous search for pollen on male- sterile lines might cause enough physical damage to increase transpiration even at low visitation levels. At high visitation levels, pollen collectors might also harm male-fertile lines. 3 . Removal of pollen from the stigmas of male-sterile flowers by high populations of pollen collecting honey bees. Compatibility of the pollen parents with the male-sterile inbreds was influenced by soil moisture levels. At high moisture, M611C increased seed yields while B12115-2C decreased them. At low moisture the reverse was true. In the first experiment there was no significant yield response to varying soil moisture levels due to contrasting effects of high soil moisture on compatibility with pollen parents. In the second experiment, the pollen parent B2215C was used and a significant favorable response to high soil moisture was obtained. Salinity aspects and the effects of indiscriminate use of fertilizers were tested through the method of fertilizer placement. While fertilizers reduced seed yield , high soil moisture tended to reduce the negative effects of nitrogen and to produce favorable response to phosphorus. Seed yield followed the trend in plant survival as affected by the induced salinity x moisture relationships. The size of bulbs had a great influence on seed yields, mainly through an increased number of umbels per plant. It seems that the number of umbels per plant had more influence on the number of flowers per row than did the number of flowers per umbel. Path-coefficient analysis of components of onion seed production indicates that the percentage of fertilized flowers is the most important criterion for selection to improve seed yield. It would be ideal to combine both high percentage of fertilized flowers and large number of flowers per umbel, but if compromise has to be made, selection would best be considered on the basis of greater percentage of fertilized flowers in the umbel. Plants which have a large percentage of fertilized flowers can be easily detected at the pre-seed harvest stage.

M. Tech. (Biotechnology, Faculty of Applied and Computer Science), Vaal University of Technology / Tulbaghia violacea Harv. (wild garlic) has been used in traditional medicine in Southern Africa for the treatment of various ailments. Despite the widespread use and popularity of this medicinal plant as a herbal medicine, there is contradictory evidence regarding the safety and toxicity of the plant. The phytochemical profiling of the plant has also been neglected in research. The determination of chemical constituents present in plant material as well as the potential toxicity found in plants are preliminary steps necessary for the discovery and development of novel therapeutic agents with improved efficacy. The aim of this study was to evaluate the cytotoxic and genotoxic potential of crude extracts from the leaves, stems and roots of T. violacea. This was performed in vitro using aqueous and ethanol extracts of the leaves, stems and roots. The aim of the study was achieved by three major objectives; (1) to identify the active phytocompounds present in the leaves, stems and roots, (2) to assess the cytotoxicity using the MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide) cell proliferation assay, and (3) to evaluate the genotoxic potential of the leaf, stem and root water extracts using the Allium cepa assay. A total of 14 phytochemicals were each extracted separately with distilled water and 70% ethanol by maceration from the leaves, stem and roots of T. violacea. The results of the qualitative phytochemical analysis showed that pharmacologically active compounds such as tannins, terpenoids, flavonoids, saponins, proteins, steroids, cardiac glycosides, phenols and coumarins were present in some organs of T. violacea. However, phlobatannins, leucoanthocyanins, alkaloids, carbohydrates and anthocyanins were absent in all plant parts. Overall, the leaves of the plant contained more active compounds than those present in the stems and roots when both water and 70% ethanol were used as the extractants. The quantitative phytochemical analysis for the Total Flavonoids Content (TFC) and Total Phenolic Contents (TPC) was also assessed. The water (0.027 mg/g) and 70% ethanol (0.053 mg/g) were most effective in extracting flavonoids from the leaves while the least amounts were obtained from the stems and roots. This observation was similar to the TFC were the water extracts of the leaves were the most effective in extracting phenols followed by the stems and roots. The MTT assay was conducted using two cell lines RAW 264.7 and C2C12. The experiment was conducted in triplicates for the leaf, stem and root extracts (water and ethanol) of T. violacea. The experimental design employed a 23 factorial design where three independent variables (concentration, incubation time and type of extracts) were selected using two levels for each variable (high (+) and low (-)). The results illustrated that both the water and ethanol vi extracts only showed a significant reduction in the number of viable cells at the concentration higher than 250 μg/ml treatment for both RAW 264.7 and C2C12 cells. The ethanol extracts from the leaves, stems and roots were found to be toxic towards the RAW 264.7 cells even at lower concentrations at both 24 and 48 h incubation periods (% cell viability < 50%). The water extracts were non-toxic to RAW 264.7 cells except for the water stem extract which showed toxicity after 48 h incubation (IC50 = 9.475 (4.061 to 23.39)). For the C2C12 cells, the lowest potent toxic concentration was 250 μg/ml for the ethanol extract of the stem after 48 h incubation. Overall, the T. violacea plant extracts were non-toxic as percentage cell viability greater than 50% was noted for both extraction solvents in all the plant parts of T. violacea. No cytotoxic activity was observed in all T. violacea plant parts with the C2C12 cell line (IC50 > 30 μg/ml). For the Allium cepa assay, only the water crude extracts of the leaves, stems and roots of T. violacea were used. A similar trend of potent genotoxic activity in the water stem extracts compared to the leaf and root extracts at the concentration ranges studied. Similar to the MTT assay, it is clear from the study that at higher concentrations, the water crude extracts from the leaves, stems and roots of T. violacea is toxic. From this study, it can be concluded that the extraction of compounds using water is more efficient than using ethanol. Overall, the T. violacea leaf extracts extracted the most phytocompounds and showed the highest percentage of viable cells as well as desirable IC50 values. However, preparation of herbal remedies using T. violacea plant extracts should be done with caution due to their possible genotoxic and cytotoxic potential at higher concentrations. This study raises a need to further conduct in vivo cytogenetic studies to ascertain the possible toxic effects of T. violacea crude extracts.