Oxytenanthera abyssinica, the African savanna bamboo, was studied in natural populations, and experimentally, in Malawi. In addition, the state-of-knowledge of the species was comprehensively reviewed. The field research addressed silvicultural and management aspects and was conducted in Malawi from October 2001 to October 2003. Oxytenanthera clumps were characterised as management units and vegetative propagation procedures were tested. Seed propagation was also conducted - in open stands in Lilongwe, and in controlled conditions in Bangor. Clump characterisation was undertaken, in terms of culm numbers and ages, over two seasons after applying different culm harvesting intensities. Mapping and monitoring of the treated clumps were also carried out. Differences in clump characteristics between localities (Lilongwe, Nsawagi and Zomba) were observed. A harvesting intensity of 40% of young (7-12 months) and 40% of mature (> 12 months) culms every two years was found to be suitable for clump management in a harvesting intensity experiment at Lilongwe. The survival of detached culm sections for vegetative propagation was affected by the position in the culm and source locality. No effect of culm age, hormone application or the nwnber of intemodes per cutting was observed. A soaking treatment before sowing was effective for seed propagation in the open. In a tetrazolium test, after three months stored at ambient temperatures, all seeds tested were viable. In a direct germination test, cumulative gennination percentage was 53% under a 32/22°C (8h/16h) regime and 40% under a 36/26°C regime 13 days after sowing. Silvicultural and management information is very limited. Most literature provides only descriptions of the plant or reports distribution records. To broaden the knowledge base, literature on the other five indigenous African bamboos was also reviewed - to underline contrasts and similarities of other species with Oxytenanthera. Herbariwn information and ecological literature was used to draft distribution maps and a GIS exercise was then conducted to identify broad relations with environment. It was concluded that with suitable clump management practices the resource can be sustainably managed. It is recommended that seed collection should be carried out whenever flowering provides the opportunity, and enables new propagation research with material raised from seed under controJled conditions. Studies on fertiliser application, which have been undertaken in managing other useful bamboos, should be carried out using permanent research plots.