Yield component analysis and recombinative heterosis of complex characters in QPM maize

Abstract

Component analysis is the analysis of variation of complex character based on the variation of its components. For assessing the variation,
components should be arranged in ontogenetical order. This allows determination of the mutually independent contributions of the components to
the variation of complex character. Component analysis provides the necessary data for the exploitation of recombinative heterosis in plant
breeding. Recombinative heterosis is the phenomenon where in, the progeny value of a complex character exceeds the mid parental value as a
result of the multiplicative relationship between the complex character and its component traits. It is suggested that this form of heterosis may be
an important cause of specific combining ability. As such, it may be involved in the heterosis of complex characters in F1 hybrids and in the
hybrid- vigor encountered in interspecific hybrid populations. It is demonstrated how recombinative heterosis may be explained by a quantitative
genetical model involving additive inheritance of the component traits. Current study was focused to predict progeny performance for the
complex character from parental data for the component traits. This requires regression of individual components on the preceding primary
characters. The contribution of the coefficient for yield component C3 (single seed weight) to variance of log yield was maximum in the parents
CML 189, UMI 814, CML 145, UMI 524, CML 141 followed by the coefficient C2 (number of kernels per row) which was maximum in the
parents UMI 427 and UMI 814. In the present investigation, 35 hybrid combinations were tested for progeny prediction value. Application of this
procedure revealed that among the 35 crosses, five crosses exceeded the predicted mid parental value. The predicted values can serve as a basis
for the selection of promising combinations. The crosses that exhibited higher values for grain yield per plant was predicted in the combination of
CML 142 x UMI 426, CML 143 x UMI 427, CML144 x UMI 189, CML 144 x UMI 426 and CML 147 x UMI 426.