BEHAVIOUR OF DIVERGENT BACKCROSSES AND CROSSES BETWEEN BACKCROSSES FOR NATIVE AND IMPROVED MAIZE VARIETIES

Abstract

Limited backcross has been proposed as a methodology to improve maize (Zea mays L.) landraces. Two theoretical reports for the case when heterosis between parents exists, claim that: a) the magnitude of residual heterosis in the F₁ backcross is considerable, b) for the crosses between divergent backcrosses, as the number of backcrosses increases a greater genetic mean is obtained. We propose that the superior edge of the genetic mean dispersion can come near the F₁ heterosis. In this research we evaluated the behavior of the F₁ first backcross to both parents, and sampled the dispersion of the yield mean for the crosses between first backcrosses. In 1997, in  Panindicuaro and Ixtlan de los Hervores, and in 1998 in Tangancicuaro, we evaluated four pairs of landraces – improved parents selected for yield heterosis, its cross and the F₁ backcross to both parents, under a randomized complete block design with two replications and irrigated conditions. In 1999, in Panindicuaro and Jiquilpan, under irrigation and rainfed conditions respectively, we evaluated crosses between backcrosses obtained by hand pollination between lower height segregants of the F₂ first backcrosses and crosses between plants of the original parents in three of the four pairs of parents, under a randomized complete block design with three replications. Yield of the improved parent backcross was the same (P ≥ 0.05) as in the parent F₁ cross, and yield of the landrace one was lowest. Crosses between backcrosses and original parent crosses showed mean standard deviations of 32.2 and 31.2 % in relation to the mean yield; thus, the expected yield was 64 % above the mean for the condition µ+2σ.