Drought and low soil fertility are two high priority stresses that impede maize production, food security and economic growth in southern Africa. The unprecedented combination of climatic risk, declining soil fertility, as population growth, high input costs, extreme poverty, and unavailability of credit systems have resulted in smallholder farmers producing maize (and other crops) in extremely low-input and therefore low-risk systems. The Southern African Drought and Low Soil Fertility Project (SADLF) is an effort by the National Agricultural Research Programs (NARSs) of the Southern African Development Community (SADC) and CIMMYT to increase and stabilize maize production in southern Africa by developing and disseminating adapted maize cultivars with improved performance under conditions typical for resource-poor farmers.   Maize germplasm is being improved under carefully-managed stress conditions, specifically drought, nitrogen and low pH stress. Breeding for stress tolerance was initiated by the NARS in Angola, Botswana, Malawi, South Africa, Tanzania, Zambia, Zimbabwe, and by CIMMYT-Zimbabwe. This resulted in a decentralized, collaborative breeding approach between CIMMYT and NARS that fostered integration of existing breeding efforts, genetic diversity, and the benefit of germplasm exchange. Experimental maize germplasm with more than 50% higher grain yields under drought, N and/or low pH stress, at a yield level of about 1-2 t/ha, was identified and several varieties released in these countries.   To highlight the differences in performance of maize varieties under agronomically well-managed conditions, as traditionally used by breeders, and the type of conditions resource-poor farmers face, a regional testing network was established among countries that established stress screening sites. Consolidation with other regional testing efforts has meant that maize cultivars at the release and prerelease stage in all public and private seed sector in the SADC region are now routinely being evaluated for drought, N stress, low pH tolerance, responsiveness to good conditions, and resistance to several important diseases and pests. This system was expanded by involving farmers in the evaluation of maize cultivars under local conditions. By the 2002 season, nine SADC countries (Angola, Botswana, Lesotho, Malawi, Mozambique, South Africa, Tanzania, Zambia, Zimbabwe) had farmer-participatory, on-farm trial schemes (Mother-Baby Trials) on the ground, with the NARS serving as coordinating unit and research stations, extension units, NGO’s, rural development projects, and schools serving as partners. With over 30 institutions and 1000 farmers participating, collaborative trials have become an important mechanism for characterizing and comparing maize cultivars used in the SADC region and for promoting knowledge and the use of best cultivars by private companies, NGO’s, extension and farmers.   As new stress tolerant robust maize cultivars are becoming available, they are expected to open new markets. Market surveys in areas where resource-poor farmers are inadequately supplied with seed have therefore been initiated. At the same time, links have been developed with the SADC seed security network working towards harmonization of procedures for cultivar release, and seed movement. New projects have been approved that target improved seed provision strategies for SADC and upgrade the formal training of NARS scientists.