HERITABILITY OF COMPONENTS OF A SIMPLE PHYSIOLOGICAL MODEL FOR YIELD IN GROUNDNUT UNDER SEMIARID RAIN-FED CONDITIONS

Use of physiological models has been suggested as a means to improve e fficiency of breeding for higher yield. Our objectives were to estimat e heritabilities of yield components of groundnut (Arachis hypogaea L. ) identified in a yield model [crop growth rate (C), reproductive dura tion (DR) and partitioning (p)] and determine their predictive value i n early generations. Forty bulk populations and nine parental lines we re evaluated in replicated trials in 1992 (F-2), 1993 (F-3) and 1994 ( F-4) at three contrasting locations in Niger. Physiological components of yield were estimated from final yield and biomass as well as data on flowering and maturity. Differences were observed among populations for pod yield and model components. The effects of locations were sig nificant (P<0.01) for C, p and D-R in F-2 and F-3 but nonsignificant f or yield and C in F-4 Heritabilities were estimated by parent-offsprin g regression of F-3 on F-2 and F-4 on F-3. Heritability estimates for C, p, D-R and yield based on the F2:F3 regression were 0.10, 0.45, 0.1 0 and 0.16, respectively. Heritabilities based on F3:F4 regression wer e 0.20 for C, 0.46 for p, 0.14 for D-R and 0.57 for yield. These resul ts reveal that none of the yield-model traits had larger heritability than yield and that selection for these traits in segregating bulk pop ulations is difficult. (C) 1998 Elsevier Science B.V. All rights reser ved.