Improvement of Resistance to Rust through Recurrent Selection in Pearl Millet

Abstract

Two pearl millet [Pennisetum glaucum (L.) R. Br. = P. typhoides (Burm.) Staff & Hubb., P. americanum (L.) K. Schum.] bulk populations, Tift #2 and Tift #5, served as base populations for four cycles of recurrent selection against susceptibility to Puccinia substriata Ell. & Barth. var. indica Ramachar & Cumm. A bulk inoculum of the pathogen was used. The objectives were to evaluate the progress achieved regarding overall resistance to the pathogen in the field and resistance to different races of the pathogen, and also to evaluate changes in unselected traits. During selection, the frequency of rust resistant plants continuously increased from about 30% in each base population to more than 85% by the third cycle of selection in both populations. An average increase of about 21 and 18% per cycle was obtained in Tift #2 and Tiff #5, respectively. A continuous increase of the frequency of plants resistant to some races of the pathogen was also obtained. In Tift #5, 80% of the plants were resistant to eight races by the third cycle of selection. The accumulation of resistance observed in the seedlings was manifested in the field, both in 1993 and 1994, by a reduction of the final rust severity from the base population to the fourth selection cycle of both populations. This improvement in resistance to the rust pathogen was accompanied by an increase in the frequency of plants resistant to Pyricularia grisea (Cooke) Sacc. only in Tift #2. Despite the improvement in the selected character, genetic variability for agronomic traits such as plant height, number of culms/plant, flowering date, and panicle length was successfully maintained within each population.