About Us | Help Videos | Contact Us | Subscriptions

Members of ASA, CSSA, and SSSA: Due to system upgrades, your subscriptions in the digital library will be unavailable from May 15th to May 22nd. We apologize for any inconvenience this may cause, and thank you for your patience. If you have any questions, please call our membership department at 608-273-8080.


Institutional Subscribers: Institutional subscription access will not be interrupted for existing subscribers who have access via IP authentication, though new subscriptions or changes will not be available during the upgrade period. For questions, please email us at: queries@dl.sciencesocieties.org or call Danielle Lynch: 608-268-4976.



This article in CS

  1. Vol. 29 No. 6, p. 1423-1427
    Received: July 11, 1988

    * Corresponding author(s):
Request Permissions


Feasibility of Mass Selection in Pearl Millet

  1. H. F. Rattunde ,
  2. Pheru Singh and
  3. J. R. Witcombe
  1. U niv. of Hohenheim, (350/3 Postf. 70-05 62, D-7000 Stuttgart 70, West Germany
    P .O. Box 910064, Mongu, Zambia
    I nt. Crops Res. Inst. for the Semi-Arid Tropics (ICRISAT), Patancheru P.O. Box 502 324 A.P., India



Mass selection is a technique widely used in both population and pedigree breeding of pearl millet, Pennisetum glaucum (L.) R. Br. To determine the feasibility of mass selecting for 19 agronomic traits of pearl millet, we (i) estimated trait heritabilities on both a single-plant and a progeny-mean basis and (ii) observed responses of S1 progenies to divergent selection on parental S0 plants. Fourteen hundred and forty S0 spaced plants from each of three pearl millet composites were self-pollinated and evaluated for 19 traits at Patancheru, India on a Udic Rhodostalf soil. Random samples of 289 S1 progenies from each S0 population were evaluated for these same traits in triple-lattice experiments at the same location. Heritabilities estimated on a progeny-mean basis were all significantly (P < 0.01) larger than zero. Heritabilities estimated on a single-plant basis were highest for traits such as panicle length (0.64), plant height (0.58), and seed weight (0.52); they were intermediate straw yield (0.40); and they were lowest for grain yield (0.29), threshing ratio (0.24), and harvest index (0.23), averaged across composites. Divergent selection of the highest and lowest decile of S0 plants identified S1 progenies with significantly (P < 0.05) increased and decreased means, respectively, for panicle and seed characteristics in all composites and for grain yield in two of the composites. Selection for increased efficiency of dry matter partitioning, however, was ineffective. The observed heritability values suggest that mass selection of pearl millet can be effective for all traits examined, with the rate of gain being proportional to the magnitude of those values.

Journal Article no. JA 792 of ICRISAT and Journal Article no. J-13089 of the Iowa Agric. and Home Econ. Exp. Stn.

  Please view the pdf by using the Full Text (PDF) link under 'View' to the left.

Copyright © .