Evaluation of Population Improvement in the Kenya Maize Breeding Methods Study

No Thumbnail Available
Journal Title
Journal ISSN
Volume Title
The maize (Zea mays L.) breeding methods study in Kenya was initiated into compare the efficiency of various methods for improving yield. A mininum of ten years of selection was completed in each experiment. and some procedures continue today because of their efficacy in line or population development. Following selection. there were three years of evaluation. including both direct and indirect effects of selection. Intra population improvement was conducted. using three variations of mass selection. including two plant densities and two levels of selection intensity. Five variations of ear-to-row selection included comparison of number of generations of random mating prior to beginning selection, plant density. male plant elimination and number of entries in the selection trial. Other methods compared were half-sib. 51 and full-sib recurrent selection. The most consistent improvement was found with ear-to-row selection. with rates of gain of nearly 3% per cycle. Inter population improvement was studied. using reciprocal recurrent selection. 51 and ear-to-row selection in the parental populations. Gains in the variety crosses were evaluated by making up the series over all cycles of selection. Gains using reciprocal recurrent selection were near 7% per cycle. Ear-to-row selection. in contrast. resulted in little gain in the variety crosses. although improvement was realized in the parental populations. Inbred lines were developed from cycles 2. 3 and 4 of the reciprocal recurrent selection experiments in Kitale Synthetic II and Ecuador 573. The lines of cycle were advanced to 53, when they were crossed in all possible combinations. Yields of the best three-way crosses were predicted, and the bf'st 36 or so were actually made and evaluated. The best three-way crosses were entered into the East African Maize Variety Trials, where they generally out yielded other entries at mid- and high-altitude site's. Significant gains were noted in lodging resistance. Two lines from cycle 2 and one from cycle 3 are now involved in production in Kenya. and three cycle 2 lines are used commercially in Tanzania. A singularly high-yielding cycle 4 line was identified from Ecuador 573, but is probably not yet found in commercial hybrids. Gains in the best three-way crosses exceeded gains in the population crosses over the time of development by 2 to 5%; uniformity in seed production and in farmers' fields enhanced the attractiveness of the three-way cross over the population cross. A highly prolific hybrid, such as EAH6302. should be quite useful in inter-cropping conditions when planted at densities of 25.000 plants. At the recommended densities of 44,000 plants l ha. it had fewer barren plants that the older hybrids.
Evaluation Review (No 5), pp. 150-175