Breeding procedures used in pearl millet [Pennisetum glaucum (L.) R. Br.] improvement in India and the USA are aimed at exploitation of hybrid vigor for both grain and forage yields. Selection efforts in Africa have been limited to intrapopulation improvement for grain yield. Knowledge of heterotic patterns and combining ability of the widely cultivated African landraces would be useful in the development of a sound breeding program in Africa. The objectives of this study were to evaluate the combining ability and heterosis among African pearl millet populations and to explore the utility of inter‐population improvement for the development of high yielding varieties and hybrids. Five populations and their 10 interpopulation crosses were evaluated in a randomized complete‐block design with six replications at two locations in Niger during the 1989 and 1990 rainy seasons. Plant height, flowering time, natural incidence of downy mildew, spike length, grain yield, and 1000‐seed weight were measured in different populations. Significant differences among entries for most of the traits were observed. Better‐parent heterosis for grain yield ranged from 25 (Iniari × Ugandi) to 81% (Mansori × Ex.Bornu). Six of the crosses showed significant heterosis and gave 36 to 81% more grain yield than their better parent. All crosses except one were significantly more tolerant to downy mildew than their better parent. Mean squares for general combining ability were significant (P < 0.05) for most traits indicating the importance of additive gene effects for these traits. The large heterotic effects and the magnitude of the additive effects observed in the experiment should be useful in choosing pearl millet landraces for intercrossing in the development of cultivars with improved grain yield.

Pearl millet [Pennisetum glaucum (L.) R. Br.] is widely grown in arid to semi-arid regions of Africa. The crop is particularly adapted to Sahelian West Africa where landraces have evolved in different ecological niches. These landraces have accumulated interpopulation diversity that has not been characterized. Evaluation of genetic diversity is a prerequisite for successful germplasm exploitation through breeding. The objective of this study was to characterize morphological and agronomic variability among African landrace populations of pearl millet. Ten pearl millet landrace populations widely grown in several African countries and two experimental Fi hybrids were evaluated at two locations in Niger during the 1989 rainy season. Thirteen characters (downy mildew [Sderospora graminicola (Sacc.) Schroet] incidence, days to flowering, primary spike length, peduncle exsertion, spike girth, flag leaf width, stem diameter, spike number per plant, non-productive tillers per plant, plant height, spike yield per plot, grain yield per plot, and 1000-seed weight) were measured on six replicates of each landrace populations. In the pooled analysis, all landrace populations were significantly different for one or more of the characters evaluated. The Niger landrace populations showed much less variation than the other African landrace populations for most characters investigated. Ward's cluster and principal component analyses were used to investigate the nature and degree of divergence in the landrace populations. The cluster analyses revealed similarities between Niger and Senegal and between Niger and Nigerian landrace populations. Four principal components were found to explain 92% of the total variation. Days to flowering, plant height, stem diameter, primary spike length, and grain and spike yield per plot were the major sources of diversity among the landrace populations. These results could be useful in choosing potentially heterotic pearl millet populations for intercrossing to develop improved cultivars, synthetics, and hybrids for use in Africa.

No abstract

No abstract

No abstract