Response of a maize synthetic to selection for components of partial resistance to Exserohilum turcicum

A synthetic population of maize (Zea mays) was created from five inbred lines of varying levels of partial resistance to northern leaf blight (NLB). This synthetic was subjected to three cycles of recurrent phenotypic selection with pollen control for either increased latent period or decreased lesion length. A selection intensity of ca. 10% was used in each selection cycle. The original synthetic and three advanced selection cycles for each of the two components of partial resistance were evaluated in field trials in the summers of 1999 and 2001, and in greenhouse trials. Selection for increased latent period was more effective in improving resistance to NLB (20 to 27% gain/cycle) (as measured by area under the disease progress curve [AUDPC]) than was selection for decreased lesion length (14 to 18% gain/cycle). Responses in AUDPC to selection for either component of resistance were linear in the 1999 field trial, but were quadratic (decreased response in advanced cycles) in the 2001 trial. Selection for increased latent period in the field resulted in a 0.6-day increase in latent period per selection cycle when measured in the greenhouse and a 2-day increase per selection cycle when measured in the field. Selection for decreased lesion length in the field did not significantly alter latent period in the greenhouse. These results support using selection for increased latent period as an effective means of improving partial resistance to NLB in maize populations. Decreased lesion length was more difficult to measure and selection based on this criterion was less effective in improving partial resistance.