Characterisation of sorghum germplasm and sorghum anthracnose epidemics in Uganda.

Anthracnose caused by Colletotrichum sublineolum is considered one of the most important diseases of sorghum. The disease can cause up to 50% yield loss as it attacks all aerial parts of the plant. The pathogen is highly variable accounting for disease resistance breakdown. Therefore additional sources of host plant resistance are needed for the development of anthracnose resistant cultivars to prevent crop losses. The objective of this study was to characterise the sorghum - C. sublineolum pathosystems in Uganda. Achieving this objective required studing the pathogen, host resistance and epidemiology. Accordingly, the study had three specific objectives: The first specific objective was to evaluate the sorghum germplasm collection from Uganda for reaction to foliar anthracnose. During the first rains of 2007, 200 accessions were collected from the major sorghum growing districts of Uganda and evaluated for reaction to sorghum anthracnose. Forty six (46) accessions were found to be resistant to anthracnose, 64 were moderately resistant and 90 accessions were considered susceptible to anthracnose. The anthracnose resistant germplasm identified from this evaluation were considered as possible additional sources of resistance for sorghum improvement and subjected to further analysis in which 10 accessions were selected for multiple resistances including athracnose, drought and turcicum leaf blight. This process led to the identification of hightly resistant accessions that can serve as sources of resistance for anthracnose breeding. The second specific objective was to characterise the temporal attributes of foliar anthracnose epidemics in two agroecologies in Uganda. The study was conducted at two locations that is Makerere University Agricultural Research Institute, Kabanyolo (MUARIK) representing the Lake Victoria Crescent and the National Semi Arid Resources Research Institute (NaSARRI), Serere, representing the north-eastern central grass-bush-farmlands. Three local land races from the germplasm collection with varying reaction to anthracnose were planted in a randomised complete block design with three replications. The test plants were inoculated at approximately 30 days after planting with anthracnose infested sorghum kernels. Disease assessment commenced 14 days after inoculation and continued until maturity at an interval of 7-10 days. Severity data collected were used to compute areas under disease progress curves. Logistic and Gompertz models were evaluated for describing the epidemics. The logistic model provided the best fit for the temporal spread of epidemics. Host genotype had a significant effect on the progress of foliar anthracnose from the point of inoculation. Generally, the increase of anthracnose severity was only limited to the susceptible and moderately resistant accession. In this study on the progress of anthracnose over time, the level of resistance had a significant effect in the disease progress, with slope steepness increasing with susceptibility. Therefore under favourable weather conditions and with susceptible cultivars, considerable spread of anthracnose over time may occur as exemplified by accession MUC007/081. Thirdly, the spatial attributes of foliar anthracnose epidemics was also characterised in the two agro-ecologies in Uganda. Three local varieties with varying reaction to anthracnose were planted in a randomised complete block design. Sorghum plant residues colonised by C. sublineolum from the previous season were placed in the centre of each plot and varied at 80% soil coverage (covering an area of 2 x 2m), 40% (covering an area of 1 m x 1 m) and 0%, (control). Results obtained showed a decline in the AUDPC with distance from the source of residue during the first rains of 2008 at both MUARIK and NaSARRI. It also showed that distance from inoculum source, cultivar and residue level significantly affected the disease level. The gradients were generally high and increased with time although it was very variable in some cases. The gradients ranged from 0.046 to -0.178 at NaSARRI during the second rains of 2007, 0.016 to -0.009 at NaSARRI during the first rains of 2008 and 0.064 to -0.076 at MUARIK during the first rains of 2008. The gradients were also observed to flatten over time and this was attributed to secondary spread of the pathogrn and background contamination. Overall, the identified novel sources of resistance to anthracnose, as well as characterised both temporal and spatial spread of C. sublineolum in two diverse agroecologies of Uganda. These results will support breeding efforts and design of other disease management options in the country.