A revision of 45 South African species of Argyrolobium is presented comprising nomenclature, typification, recorded distributions and full descriptions. In an attempt to reduce confusion between subtly different species, diagnostic characters are listed and comprehensive illustrations are provided. The dissertation includes numerous taxonomic and nomenclatural changes. Currently two sections are recognised based on fruit morphology. This is replaced by a system of five new sections based on morphological, micromorphological and phytogeographical evidence. Constraints and advantages of floral dimorphism and monomorphism are discussed with respect to the fecundity and distribution of sections. It is proposed that the occurrence of facultative cleistogamy in two sections has enhanced their success and distribution. Generic and specific phylogeny is interpreted using models generated through cladistic methodology. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1994.

A double sandwich enzyme-linked immunosorbent assay (ELISA) was developed to quantify the proteinaceous inhibitor of tissue plasminogen activator (t-PA) which occur in the tissue of Erythrina caffra Thunb. Using the ELISA the t-PA inhibitor could be detected in nanogramme quantities on the micro titer plate. The concentration of the t-PA inhibitor was determined in different tissues of Erythrina caffra. t-PA inhibitor concentrations in the order of 1 000 microgrammes per gramme protein were found in the seeds. Relatively small quantities of t - PA inhibitor, in the order of 10 to 50 microgrammes per gramme protein, occurred in root, shoot, leaf and living bark material. The t-PA inhibitor was found to accumulate in a similar way to the storage proteins in developing seeds. The accumulation of the inhibitor is at a relatively low level during the early period of seed development but increases exponentially just before the seeds reach their maximum size. The t-PA inhibitor content of the cotyledons decreased drastically during the process of germination and subsequent seedling development. The disappearance of the inhibitor be the result of total degradation of the molecule can or partial proteolysis with the modified molecule still being present in the tissue. An attempt was made to increase the t-PA inhibitor content of excised leaves of Erythrina caffra with protein inducing substances such as polyamines, precursors of ethylene and phytic acid. The protein inducing compounds included cell wall hydrolysates of Erythrina caffra, the marine alga Ecklonia maxima Osbeck (Papenfuss) as well as Lycopersicon esculentum Mill which induced the, synthesis of proteinase inhibitors suggested to be involved in the defense mechanism of plants. None of the substances used, increased the t-PA inhibitor content of excised leaves or in vitro cultures of Erythrina caffra. It is suggested that the t-PA inhibitor is probably not involved in a defense mechanism of Erythrina caffra. A callus and suspension culture derived from shoot tissue was developed to determine the occurrence of the t-PA inhibitor in vitro. The optimal nutrient medium for the growth of callus was the salts and vitamins of MURASHIGE and SKOOG (1962). The medium was supplemented with 3 % sucrose, 0. 1 gramme per litre meso - inositol, 10 micromoles per litre benzyl adenine and 5 micromoles per litre 2,4- dichlorophenoxyacetic acid . Different auxins and cytokinins had a similar growth stimulatory effect on the growth of callus derived from a number of organs of Erythrina caffra. The callus from different organs did however, grow at different rates on the same nutrient medium. Callus derived from leaf, shoot, and cotyledonary tissue grew at similar rates on the nutrient media of MURASHIGE and SKOOG (1962), SCHENK and HILDEBRANDT (1972) and B5 (GAMBORG, MILLER and OJIMA, 1968) despite large differences in the concentration of the nutrients in the three nutri.ent media. The source of nitrogen and ratio of nitrate to ammonium was critical to the growth of callus cultures . The optimal concentration of nitrate and ammonium was 30 millimoles per litre . The growth of callus from different organs was significantly affected by the concentration of sucrose in the nutrient medium. A concentration of 3% was optimal for callus growth. Temperature had a significant effect on the growth of callus. The optimal temperature for callus growth was 25 °C. A shoot cell suspension culture was established and maintained at the same temperature and on the same medium as the callus cultures but with a ten times lower concentration of growth regulators. A low shake speed was essential for the growth of the suspension culture. Maximum growth was obtained at a shake speed of 60 rpm. Relatively low quantities of t-PA inhibitor, in the order of 1 to 5 microgrammes per gramme protein, was detected in the suspension cultures. An attempt was made to increase the t-PA inhibitor content of the suspension cultures with the pro te in i nduc i ng compounds used on excised leaves, but without success. However, the t-PA inhibitor content of the suspension culture was significantly increased with a ten times increase in the sulphate content of the nutrient medium. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1990.