In this study, host-specific forms of the blast pathogen Magnaporthe oryzae in sub-Saharan Africa (SSA) were characterised from distinct cropping locations using a combination of molecular and biological assays. Finger millet blast populations in East Africa revealed a continuous genetic variation pattern and lack of clonal lineages, with a wide range of haplotypes. M. oryzae populations lacked the grasshopper (grh) element (96%) and appeared distinct to those in Asia. An overall near equal distribution (47-53%) of the mating types MAT1-1 and MAT1-2, high fertility status (84-89%) and the dominance of hermaphrodites (64%) suggest a strong sexual reproductive potential. Differences in pathogen aggressiveness and lack of cultivar incompatibility suggest the importance of quantitative resistance. Rice blast populations in West Africa showed a typical lineage-based structure. Among the nine lineages identified, three comprised ~90% of the isolates. Skewed distribution of the mating types MAT1-1 (29%) and MAT1-2 (71%) was accompanied by low fertility. Clear differences in cultivar compatibility within and between lineages suggest R gene-mediated interactions. Distinctive patterns of genetic diversity, sexual reproductive potential and pathogenicity suggest adaptive divergence of host-specific forms of M. oryzae populations linked to crop domestication and agricultural intensification.

Effect of plant density (37,037, 44,444 and 55,555 plants/ha) on the increase of northern leaf blight in time and space on two maize cultivars planted at spacings of 90, 75 and 60 between rows and 30 cm within rows was investigated. Exserohilum turcicum infested maize residue was used as an inoculum source. Maize density did not significantly affect the disease indices, but significantly influenced the grain yield of the two cultivars. In contrast, the two cuitivars differed significantly (P < 0.01) in disease severity, but not grain yield. Higher disease severities and grain yields were associated with higher plant densities, whereas the apparent infection rate was lower in higher plant density. Distance from the maize residue (inoculum source) significantly influenced disease severity. The percentage leaf area blighted, area under disease progress curve and disease progress curve intercept decreased with distance from the maize residue area, but the apparent infection rate on EV8429-SR appeared to increase with distance. Disease gradients (b) were higher in closely planted maize and flattened with time in one location only.

Finger millet is a staple, high-quality food, important to the livelihoods of millions of smallholder farmers in East Africa. It has been neglected by major donors to agricultural research. This paper reports recent investment by the UK Department for International Development (DFID) in several projects on blast disease that has not only led to successful promotion of sound blast management strategies to farmers, but has also fostered partnerships in an evolving finger millet innovation system in East Africa. A key entry point has been created to address other constraints to finger millet production and utilization, such as ineffective weed management, poor grain quality, inefficient seed systems and production-supply chain problems, notably through 'spill-in' and adaptation of relevant technologies developed elsewhere. Further donor investment in the finger millet sector is likely to make a significant contribution to fighting malnutrition and poverty in East Africa.

No abstract