Parasitic¯owering weeds of the genus Striga (Scrophulariaceae) cause substantial losses in sorghum [Sorghum bicolor (L.) Moench] production in sub-Saharan Africa. Striga-resistant sorghum cultivars could be a major component of integrated striga management, if resistance was available in adapted, productive germplasm. In this paper we review methodologies for breeding striga-resistant sorghums. The agar-gel assay is an excellent tool to screen host genotypes in the laboratory for low production of the striga seed germination stimulant. Further laboratory assays are needed which allow the non-destructive, rapid and inexpensive evaluation of individual plants for additional resistance mechanisms. Field screening for striga resistance is hampered by high microvariability in African soils, heterogeneity of natural infestations, and concomitant large environmental effects on striga emergence. An improved ®eld testing methodology should include one or several of the following practices: ®eld inoculation with striga seeds; appropriate experimental design including elevated replication number; speci®c plot layout; use of appropriate susceptible and resistant checks; evaluation in adjacent infested and uninfested plots; and the use of selection indices derived from emerged striga counts, striga vigor, and grain yield or a host plant damage score. Due to the extreme variability of the parasite and signi®cant genotypeÂenvironment interaction effects, multi-locational screening is recommended to obtain materials with stable performance. Additional strategies include: careful de®nition of the target environments; determination of the most important selection traits in each target environment; characterization of crop germplasm and improvement of available sources of resistance for better agronomic performance; transfer and pyramiding of resistance genes into adapted, farmer-selected cultivars; development of striga-resistant parent lines for hybrid or synthetic cultivars; and development of random-mating populations with multiple sources of resistance. The development of markerassisted selection techniques for broad-based, polygenic striga resistance is underway. This approach is particularly promising because striga resistance tests are dif®cult, expensive, and sometimes unreliable; the parasite is quarantined; and some resistance genes are recessive. Transgenic, herbicide-tolerant sorghums could contribute to an immediate, cost-effective control of striga by herbicides, but such cultivars are not yet available. The selection of sorghum cultivars with speci®c adaptation to integrated striga management approaches could contribute to sustainable sorghum production in striga-infested areas of sub-Saharan Africa. #

The stay-green trait is a reported component of tolerance to terminal drought stress in sorghum. To map quantitative trait loci (QTLs) for stay-green, two sorghum recombinant inbred populations (RIPs) of 226 F(3:5) lines each were developed from crosses (1) IS9830 x E36-1 and (2) N13 x E36-1. The common parental line, E36-1 of Ethiopian origin, was the stay-green trait source. The genetic map of RIP 1 had a total length of 1,291 cM, with 128 markers (AFLPs, RFLPs, SSRs and RAPDs) distributed over ten linkage groups. The map of RIP 2 spanned 1,438 cM and contained 146 markers in 12 linkage groups. The two RIPs were evaluated during post-rainy seasons at Patancheru, India, in 1999/2000 (RIP 2) and 2000/2001 (RIP 1). The measures of stay-green mapped were the green leaf area percentages at 15, 30 and 45 days after flowering (% GL15, % GL30 and % GL45, respectively). Estimated repeatabilities for % GL15, % GL30 and % GL45 amounted to 0.89, 0.81 and 0.78 in RIP 1, and 0.91, 0.88 and 0.85 in RIP 2, respectively. The number of QTLs for the three traits detected by composite interval mapping ranged from 5 to 8, explaining 31% to 42% of the genetic variance. In both RIPs, both parent lines contributed stay-green alleles. Across the three measures of the stay-green trait, three QTLs on linkage groups A, E and G were common to both RIPs, with the stay-green alleles originating from E36-1. These QTLs were therefore consistent across the tested genetic backgrounds and years. After QTL validation across sites and verification of the general benefit of the stay-green trait for grain yield performance and stability in the target areas, the corresponding chromosomal regions could be candidates for marker-assisted transfer of stay-green into elite materials.

Molecular markers for resistance of sorghum to the hemi-parasitic weed Striga hermonthica were mapped in two recombinant inbred populations (RIP-1, and -2) of F(3:5) lines developed from the crosses IS9830 x E36-1 (1) and N13 x E36-1 (2). The resistant parental lines were IS9830 and N13; the former is characterized by a low stimulation of striga seed germination, the latter by "mechanical" resistance. The genetic maps of RIP-1 and RIP-2 spanned 1,498 cM and 1,599 cM, respectively, with 137 and 157 markers distributed over 11 linkage groups. To evaluate striga resistance, we divided each RIP into set 1 (116 lines tested in 1997) and set 2 (110 lines evaluated in 1998). Field trials were conducted in five environments per year in Mali and Kenya. Heritability estimates for area under the striga number progress curve (ASNPC) in sets 1 and 2 were respectively 0.66 and 0.74 in RIP-1 0.81 and 0.82 in RIP-2. Across sites, composite interval mapping detected 11 QTL (quantitative trait loci) and nine QTL in sets 1 and 2 of RIP-1, explaining 77% and 80% of the genetic variance for ASNPC, respectively. The most significant RIP-1 QTL corresponded to the major-gene locus lgs (low stimulation of striga seed germination) in linkage group I. In RIP-2, 11 QTL and nine QTL explained 79% and 82% of the genetic variance for ASNPC in sets 1 and 2, respectively. Five QTL were common to both sets of each RIP, wtih the resistance alleles deriving from IS9830 or N13. Since their effects were validated across environments, years and independent RIP samples, these QTL are excellent candidates for marker-assisted selection.

Semi‐arid and subhumid West Africa is characterized by high inter‐annual rainfall variability, with variable onset of the rainy season, somewhat more predictable endings, and drought or excess water occurrence at any time during the growing season. Climate change is predicted to increase this variability. This article summarizes options for plant breeders to enhance the adaptation of pearl millet (Pennisetum glaucum [L.] R. Br.) and sorghum (Sorghum bicolor [L.] Moench) to climate variability in West Africa. Developing variety types with high degrees of heterozygosity and genetic heterogeneity for adaptation traits helps achieving better individual and population buffering capacity. Traits that potentially enhance adaptive phenotypic plasticity or yield stability in variable climates include photoperiod‐sensitive flowering, plastic tillering, flooding tolerance, seedling heat tolerance and phosphorus efficiency. Farmer‐participatory dynamic gene pool management using broad‐based populations and diverse selection environments is useful to develop new diverse germplasm adapted to specific production constraints including climate variability. For sustainable productivity increase, improved cultivars should respond to farmer‐adoptable soil fertility management and water harvesting techniques. Larger‐scale, on‐farm participatory testing will enable assessments of varietal performance under evolving climatic variability, provide perspective on needs and opportunities and enhance adoption. Strengthening seed systems will be required to achieve sustainable impacts.

Although sorghum [Sorghum bicolor (L.) Moench] in West Africa (WA) is generally cultivated with limited or no fertilization on soils of low phosphorus availability, no assessments of the genetic variation among WA sorghum varieties for adaptation to low soil P are known. We assessed grain yields of 70 diverse sorghum genotypes under -P (no P fertilization) and +P conditions at two locations in Mali over 5 yr. Genetic variation for grain yield under -P conditions and the feasibility and necessity of sorghum varietal testing for grain yield under -P conditions were evaluated. Delayed heading dates (0-9.8 d) and reductions of grain yield (2-59%) and plant height (13-107 cm) were observed in -P relative to the +P trials. High estimates of genetic variance and broad-sense heritabilities were found for grain yield across both -P (^2 = 0.93) and +P (h^ = 0.92) environments. The genetic correlation for grain yield performance between -P and +P conditions was high (r^ = 0.89), suggesting that WA sorghum varieties generally possess good adaptation to low-P conditions. However, genotype x phosphorus crossover interaction was observed between some of the highest yielding genotypes from the -P and +P selected sets, with the variety IS 15401 showing specific adaptation to -P. Direct selection for grain yield in -P conditions was predicted to be 12% more efficient than indirect selection in +P conditions. Thus, selection under -P conditions should be useful for sorghum improvement in WA.

No abstract

BackgroundThe distribution area of pearl millet in West and Central Africa (WCA) harbours a wide range of climatic and environmental conditions as well as diverse farmer preferences and pearl millet utilization habits which have the potential to lead to local adaptation and thereby to population structure. The objectives of our research were to (i) assess the geographical distribution of genetic diversity in pearl millet inbreds derived from landraces, (ii) assess the population structure of pearl millet from WCA, and (iii) identify those geographical parameters and environmental factors from the location at which landraces were sampled, as well as those phenotypic traits that may have affected or led to this population structure. Our study was based on a set of 145 inbred lines derived from 122 different pearl millet landraces from WCA.ResultsFive sub-groups were detected within the entire germplasm set by STRUCTURE. We observed that the phenotypic traits flowering time, relative response to photoperiod, and panicle length were significantly associated with population structure but not the environmental factors which are expected to influence these traits in natural populations such as latitude, temperature, or precipitation.ConclusionsOur results suggested that for pearl millet natural selection is compared to artificial selection less important in shaping populations.

a b s t r a c tPearl millet (Pennisetum glaucum (L.) R. Br.) is an important hybrid crop in India. However, to date limited pearl millet hybrid development has been undertaken in West Africa (WA), which is the center of pearl millet origin and diversity and where this crop is most important outside India. Using a diverse set of WA pearl millet germplasm, objectives of this study were to determine the superiority of population hybrids over open-pollinated varieties for agro-morphological and agronomic traits in WA pearl millet germplasm; and (ii) to derive strategies for pearl millet hybrid breeding in WA, based on quantitativegenetic parameters, combining ability and heterotic patterns among geographically close versus distant pearl millet populations. A 10 × 10 factorial mating design was performed with four parental OPVs from each of five WA countries. The 100 population hybrids and their parents were tested for 14 traits at six locations in one year, thereby using contrasting locations to indirectly sample the rainfall variability inherent to WA pearl millet production environments. Grain yield showed an average panmictic midparent heterosis (PMpH) of 16.7%, ranging from −26 to 73%. The mean grain yield of hybrids based on inter-country crosses did not differ significantly from intra-country crosses. Geographic distance between parents was positively correlated with hybrid grain yield (r = 0.31), but not with PMpH . Some crosses between accessions from Niger/Nigeria and Senegal were outstanding. Predictability of population hybrid performance for grain yield was moderate based on midparent values (r = 0.43) and slightly better based on general combining ability (GCA) (r = 0.56). Overall, pearl millet hybrid breeding in WA seems very promising, but there do not seem to be clear "natural" heterotic groups among WA pearl millet landraces. Such heterotic groups as the basis of sustainable hybrid breeding need rather to be created systematically, by building on existing combining ability patterns and aiming to maximize combining ability between the groups.