Skip to main content
ARS Home » Southeast Area » Tifton, Georgia » Crop Genetics and Breeding Research » Research » Publications at this Location » Publication #105810


item Wilson, Jeffrey - Jeff
item Hanna, Wayne

Submitted to: Phytopathology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/3/2000
Publication Date: 10/1/2000
Citation: Wilson, J.P., Hess, D.E., Hanna, W.W. 2000. Resistance to striga hermonthica in wild accessions of the primary gene pool of pennisetum glaucum. Phytopathology 90:1169-1172.

Interpretive Summary: Cereals and grain legumes are the principal food crops grown in West Africa and provide the main source of human nutrition. Twenty one million ha of land in Africa are estimated to be infested by Striga, resulting in an annual grain yield loss of 4.1 million tons and 7 billion US dollars in revenues to Africa. Striga hermonthica can infect corn, sorghum, pearl millet, rice, and sugarcane. Host plant resistance is the most economic control measure since resistant cultivars can be grown without additional inputs. Most efforts at breeding for resistance to Striga focus on number of Striga plants parasitizing the host. Selection for low Striga number has not resulted in significant improvements in crop resistance These experiments showed that several factors unrelated to genetic resistance affect Striga numbers. Because S. hermonthica is an obligate parasite, relationships exist between Striga infection, host physiology, and other variables, which confound resistance assessments. Observations of Striga number alone tends to overestimate resistance. In order to more accurately identify resistance, a resistance index is proposed to analytically correct for factors affecting host physiology and Striga number.The index accounts not only for low observed Striga numbers, but also is weighted to identify accessions that have a lower than expected number of Striga.

Technical Abstract: Resistance to Striga hermonthica in 274 wild Pennisetum glaucum accessions was evaluated at Samanko, Mali in 1997 and 1998, and at Cinzana, Mali and Sadore, Niger in 1998. Data recorded included number of Pennisetum plants, Pennisetum anthesis date, downy mildew incidence, date of Striga emergence and number of Striga plants/plot at least 3 times during the season. Across strials ave. number of Striga/host plant ranged from 0.9-8.3. Days to Strig emergence ranged from 54-68 d. Downy mildew incidence ranged from 10-32%. Across all trials maximum Striga emergence was correlated positively with Pennisetum flowering date and negatively with downy mildew incidence and date of Striga emergence. Predicted maximum Striga within each trial was calculated by linear and multiple regression equations of d to Pennisetum flowering date, downy mildew incidence and d to Striga emergence.Resistance indices were developed to account for the relationships between maximum Striga counts and the other variables. When observed Striga counts were compared by Fisher's lsd, 84% of the germplasm collection did not differ from the most resistant accession. When comparing resistance indices, the percentage of the germplasm collection that did not differ from the most resistant accession ranged from 37-47%. When all 4 resistance indices were used for comparisons, only 31% of the collection could be considered resistant. When observed Striga counts were used, 144 accessions were falsely classified as resistant, while 1 accession was falsely classified as susceptible. Because several variables not directly related to genetic resistance are related to Striga emergence, use of the resistance indices may allow more consistent identification of resistance to Striga.