Seedling root and coleoptile length as selection criteria for field emergence in pearl millet

Pearl millet (Pennisetum glaucum [L.] R. Br.) production in semi-arid regio ns is limited by poor seedling emergence that results in poor stand establi shment and, low yield. Six pearl millet populations were created by diverge nt selection within the Tift #2 S-1 (TIFT2) pearl millet population for see dling root length in sand in the greenhouse or seedling root or coleoptile length in paper germination towels in the laboratory. The TIFT2 population (TIFT2) and the six selected populations were grown in the Field in a rando mized complete block design at Mississippi State, MS (MSU) and at Katumani and Kiboko, Kenya in 1991 and 1992. Populations were evaluated fur seedling emergence and shoot height every 3 d for the first 15 d after planting. Th e greatest differences in emergence occurred at the most water-stressed loc ations in both years. The population selected for longer seedling root leng th in sand (LR-S) generally showed the highest field seedling emergence and shoot height during the first 9 d. Emergence of the LR-S population (18.8, 11.3, and 9.8 seedlings) at 15 d was significantly greater than TIFT2 (11. 3, 5.7, and 4.3 seedlings) in the more moisture-stressed locations, Katuman i and MSU in 1991 and Kiboko in 1992, respectively. Emergence of the LR-S p opulation exceeded that of TIFT2 in every case except day 6 at MSU in 1992. The LR-S population also exhibited the fastest rates of emergence (1.1 to 2.1 seedlings d(-1)) at the most water-stressed locations. The LR-S populat ion was able to tolerate or avoid moisture stress better than the other pop ulations. There was little or no increase in seedling emergence or shoot he ight with selection for coleoptile length or seedling root length in paper germination towels. Selection for seedling root length in sand culture in t he greenhouse demonstrated promise as an indirect selection criterion for i ncreased seedling emergence and shoot height.