Microbial infections are a major threat to public health particularly in developing countries due to the relative unavailability of medicine and the emergence of widespread drug resistance. Serious invasive fungal infections caused by Candida albicans, Cryptococcus neoformans and Aspergillus spp. represents an increasing threat to human health. They have increased significantly during the past decade, especially in immunocompromised individuals, due to the increased occurrence of HIV infections and resistance development. The toxicity of available antifungal drugs/agents has contributed greatly to the need for new antifungal drugs. Cryptococcus neoformans is a yeast organism that causes cryptococcosis in both humans and animals. This disease develops following inhalation and dissemination of the organism from the lungs to the central nervous system. Infection with C. neoformans often produces pneumonia and cryptococcal meningitis in HIV-infected patients. The problem with this fungus is that AIDS patients respond poorly to treatment and need lifelong therapy to suppress the infection and the drug treatment may be expensive in developing countries. This indicates an urgent need to develop new specific fungicidal antimicrobial agents for the treatment of cryptococcosis. Plants synthesize a large number of secondary metabolites for protecting themselves against microbe infections caused by bacteria, fungi and viruses. These substances may be useful in the treatment of microbial infections in humans and animals. Plants can be considered as potential sources of therapeutic extracts or active pure chemical compounds for the development of medicines. During this project, ten plant species (Zanthoxylum capenses, Morus mesozygia, Calodendron capenses, Catha transvaalensis, Cussonia zuluensis, Ochna natalitia, Croton sylvaticus, Maytenus undata, Celtis africana and Cassine aethiopica) were screened for activity against C. neoformans using both bioautography and the microdilution assay. The most active plant species was selected for the isolation of active metabolites. The selection of plant species was based on the lowest MIC value, presence of clear zones on bioautograms indicating antifungal activity, and high total activity against C. neoformans. M. undata indicated the presence of clear zones on bioautograms, a low average MIC value of 0.09 mg/ml and high total activity. C. sylvaticus and C. transvaalensis had lower or equal average MIC values to M. undata of 0.07 mg/ml and 0.09 mg/ml respectively. However a lack of clear bands to identify the position of active compounds on bioautography plates disqualified them for further analysis in this study. The leaves of M. undatawere exhaustively extracted with hexane, dichloromethane, acetone and methanol respectively. The hexane extract indicated the lowest MIC value of 0.02 mg/ml and was used for isolation of the active constituents. Column chromatography and bioassay-guided fractionation led to the isolation of six triterpene-like compounds.The structure of the isolated compounds was elucidated using the NMR and MS techniques and the compounds were identified as friedelin (1), epifriedelanol (2), taraxerol (3), 3-oxo-11á-methoxyolean-12-ene-30-oic acid (4), 3-oxo-11á-hydroxyolean-12-ene-30-oic acid (5)&3,11-dihydroxyolean-12-ene-30-oic acid (6). Friedelin (1) and epifriedelanol (2) belong to the friedelane group of triterpenoids, taraxerol (3) belongs to the taraxerane group and 3-oxo-11á-methoxyolean-12-ene-30-oic acid (4), 3-oxo-11á-hydroxyolean-12-ene-30-oic acid (5)&3,11-dihydroxyolean-12-ene-30-oic acid (6) belong to the 12-oleanene group. These groups have been isolated previously from plants that belong to the Celastraceae family. Four of the six isolated compounds 1,3,5 and 6 were isolated in sufficient quantity to be assayed against two fungal species (Candida albicans and Cryptococcus neoformans), two Gram-positive bacterial species (Staphylococcus aureus, ATCC 29213 and Enterococcus faecalis, ATCC 29212) and two Gram-negative bacterial species (Escherichia coli, ATCC 27853 and Pseudomonas aeruginosa, ATCC 25922).Two of the compounds, 3-oxo-11á-hydroxyolean-12-ene-30-oic acid (5)&3,11-dihydroxyolean-12-ene-30-oic acid (6), showed clear bands against all the tested organisms on bioautograms indicating microbial growth inhibition. MIC values ranged from 24 µg/ml to 63 µg/ml except for S. a ureus which was resistant. All the tested microorganisms showed resistance against friedelin (1) and taraxerol (3) with MIC values of >250 µg/ml, except for E. faecalis with an MIC value of 130 µg/ml for taraxerol. The cytotoxicity of the hexane extract and the isolated compounds were investigated using the tetrazolium-based colorimetric assay (MTT) using Vero monkey kidney cells and the hemagglutination assay using formaldehyde-fixed erythrocytes (RBCs). The hexane plant extract indicated toxicity towards the Vero monkey cells with an LC50 of 0.076 mg/ml. Compounds 1 and 3 indicated no toxicity against the cells with an LC50 greater than 200 µg/ml. However compounds 5 and 6 indicated toxicity with an LC50 of 6.16 µg/ml and 3.36 /ml, respectively. Also the hemagglutination assay indicates that hexane extract is toxic towards the RBCs with a HA titer value of 1.6. Both compounds 1 and 3 indicated no agglutination and compounds 5 and 6 indicated HA titer values of 1.33 and 0.67, respectively.