Selecting desired agronomic traits may lead to a loss of genetic diversity in crop species. A molecular investigation was conducted to determine how well a set of olive (Olea europaea L.) accessions sampled in Moroccan traditional orchards represented the entire Moroccan olive diversity range. We therefore collected, in traditional agroecosystems from northern and central Morocco, a total of 88 olive trees chosen for their agronomic traits based on local farmers' knowledge. Using 12 SSR loci, 45 trees (51.1%) had a genotype identical to the ‘Picholine Marocaine’ variety, while the remaining samples were classified into 27 different SSR profiles. Two categories of genotypes were identified: (i) genotypes closely related to the ‘Picholine Marocaine’ variety and probably resulting from intensive vegetative propagation from a limited number of clones, and (ii) genotypes displaying a high number of dissimilar alleles which may have originated from selected spontaneous seedlings. A significant difference in allelic richness was revealed between the 28 on-farm selected genotypes and the overall olive diversity, represented by 57 local genotypes, indicating that the on-farm selected trees represented a subsample of Moroccan genetic diversity. This could be explained by the prevalence of ‘Picholine Marocaine’ in traditional orchards, while some original genotypes with favourable agronomic traits resulting from local farmers' selection were also identified. Applying an ethnobotany approach combined with criteria to fulfil farmers' household needs could be particularly relevant and better explain the obtained results.

Flavour is one of the most highly demanded consumer traits of tomato at present; poor flavour is one of the most commonly heard complaints associated with modern varieties of tomato. In order to combine flavour with other desirable fruit traits in improved cultivars, it is important to determine how much variability exists in the crucial compounds that contribute most to flavour. The objective of the present study was to determine the variability of flavour-contributing components including total soluble solids (TSS) and total titratable acids (TTA) among other subjective traits related to flavour in a core collection of tomato accessions. The core collection was comprised of 173 tomato accessions with a wide genetic background from the United States Department of Agriculture (USDA), Agricultural Research Services (ARS) Plant Genetic Resources Unit repository. The TTA varied from 0.20 to 0.64%, whereas the TSS ranged from 3.4 to 9.0%, indicating the availability of broad variation for these traits. Rinon (PI 118783), Turrialba, Purple Calabash and LA2102 were among the high TTA (>0.45%) containing accessions, whereas those with high TSS (>7.0%) were AVRDC#6, Sponzillo and LA2102. A positive correlation of overall flavour with TTA (r= 0.33; P< 0.05) and TSS (r= 0.37; P< 0.05) indicated that these two components play an important role in determining the overall flavour in tomato. Subjectively measured other traits including fruity odour and fruity flavour had positive correlations with overall flavour. Overall flavour is discussed in the context of other traits including fruit firmness. Information obtained from this study may be useful for tomato breeders aiming to improve tomato flavour.

Improvement in multi-cut forage sorghum varieties that can perform better than SSG 59-3, a sweet Sudan grass released in 1974, has been a challenge. Efforts were made to create variability in SSG 59-3 through mutagenesis using both physical and chemical mutagens. Fifteen such mutants were evaluated for 2 years in different locations in India. Considerable variability was observed for important fodder yield and quality traits. SSG 237 flowered 10 d early than SSG 59-3. SSG 231, SSG 260, SSG 232 and SSG 237 had high protein percentage and in vitro dry matter digestibility (IVDMD) values compared with SSG 59-3. Hydrocyanic acid (HCN), a major anti-nutritional factor, was low in six mutants, the lowest being in SSG 226. This variability can be used in different breeding programmes aimed at improving multi-cut forage sorghum varieties. The lines with improved fodder quality (low HCN, high protein and IVDMD) can be used in the breeding programme for the improvement in multi-cut forage sorghum varieties.

India is the second most populous nation in the world after China, and its plant genetic resources are an important basis for crop improvement to meet human population needs. An important legume in the diet of the Indian population is common bean (Phaseolus vulgaris L.). Common beans are one of the many important legumes grown in India, but unlike others, its centre of origin is not in Asia but in the Americas. The objective of this study was to evaluate two collections of Indian common beans: one for an internationally available collection of Food and Agriculture Organization (FAO)-protected accessions and one from the north-western Himalayan region. In total, 149 Indian landraces were evaluated with a total of 24 microsatellites across the two collections, and these represented all common bean-growing states of India. A population structure analysis was used to find groups in each collection, and this was compared across the collections. The genetic analysis of the two sets of Indian accessions with neighbour-joining trees and principal component analysis categorized the landraces into Andean and Mesoamerican gene pool groups. The Andean genotypes dominated the north-western Himalayan collection while the FAO-protected accessions were predominantly Mesoamerican. The Indian subcontinent can be considered as a region of high bean diversity; however, very little introgression was observed between the gene pools in both the germplasm sets. Gene pool identity was further substantiated by the comparison of seed traits, particularly seed size. The role of the landraces in plant breeding programmes is discussed.

The evolutionary diversity of wild potato species makes them excellent materials for improving the narrow genetic basis of the cultivated potato Solanum tuberosum. Understanding their genetic diversity is important not only to choose the best parents for breeding, but also to design proper crossing schemes and selection strategies. The objectives of this study were to determine the resistance response to Ralstonia solanacearum, Potato virus Y and low temperatures of 21 clones of 12 potato species, and to determine their genetic diversity through simple sequence repeat (SSR) markers. Sources of resistance have been found for all the investigated traits, with high resistance variability not only between but also within species. Combined resistances were also identified, with positive implications for efficient breeding. SSR analysis allowed the detection of 12 loci and 46 alleles across all genotypes, with an average value of 3.8 alleles per locus. Both unique and rare alleles useful for marker-assisted selection were found. SSR-based cluster analysis revealed that resistant genotypes were distributed among all clusters, suggesting that genetically different resistant genotypes were identified. The information obtained in this study is discussed from a breeding perspective.

The narrow genetic base of existing commercial oil palm cultivars has prompted oil palm breeders to give increased importance to augmenting these genetic resources because the sustainable development of the crop depends largely on the availability of genetic diversity and its use. Therefore, the purpose of this study was to conduct a molecular characterization of an ex situ collection of oil palm Elaeis guineensis Jacq. populations from the Republic of Cameroon using microsatellite molecular markers. Overall, 31 simple sequence repeats were polymorphic, with a total of 223 alleles, 78.4% of which were found at low frequency. The total genetic diversity was relatively high (HT= 0.673). The genetic differentiation between geographical regions was low (GST= 0.023, P= 0.001), and between families it was high (GST= 0.166, P= 0.001), showing greater variation between families than among geographical regions. The molecular data indicate that genetic diversity among the genotypes evaluated is mainly distributed within regions, suggesting that there is no isolation by geographical distance and that all the sampled individuals form a single diverse population. Therefore, it was concluded that a relatively low number of accessions (120 in the analysed case) that includes at least one representative of each family would allow us to efficiently collect almost the entire genetic diversity of Cameroon within the collection studied. This will allow for the efficient use of genetic resources and a reduction in morpho-agronomic characterization costs.

Data on intra-specific variability for seed oil content, physical characteristics and fatty acid composition in Cucurbita moschata and Cucurbita argyrosperma are lacking in the scientific literature. We examined 528 genebank accessions of C. moschata and 166 accessions of C. argyrosperma – which included members of both subsp. argyrosperma and subsp. sororia – for seed oil content, oil physical characteristics and fatty acid composition. The oil of both species had near-identical viscosities, viscosity indices, colour and oxidative stabilities while the oil of C. argyrosperma had a slightly higher pour point, cloud point, percentage of free fatty acids and acid value when compared with C. moschata. Mean oil content values of the two species were similar at 28.7 ± /2.7 and 29.8 ± /2.6% for C. moschata and C. argyrosperma, respectively. The mean seed oil content of C. argyrosperma subsp. argyrosperma var. palmeri (32.1%) was significantly higher than that of the other taxa examined. The average (mean) percentage of total seed weight attributable to the kernel was 77.2% in C. moschata (n= 34) and 74.5% in C. argyrosperma (n= 46). The percentage of total seed weight attributable to the hull was correlated with seed oil content, in both species. Linoleic was the predominant fatty acid in all the samples analysed. Means for individual fatty acids in C. moschata were linoleic 48.5%, oleic 22.6%, palmitic 20.7% and stearic 7.5%. Means for individual fatty acids in C. argyrosperma were linoleic 47.3%, oleic 27.5%, palmitic 16.5% and stearic 8.0%.

This is the first report of a pasture plant collecting mission to the highlands of Eritrea and a preliminary examination of the potential of species for both Eritrea and southern Australia. In 2004, seeds from 53 legume species were collected from 58 locations in the southern highlands between Keren, Adi Quala and Senafe. Strains of Rhizobium from 18 species were also collected. Seed collections of 11 species with Rhizobium were established in germplasm nurseries at the Medina Research Station, Western Australia between 2005 and 2010. Observations on their growth, flowering and seed production were recorded. Based on a climate match analysis and observations from germplasm nurseries, it was suggested that species with most promise for parts of southern Australia include the annual legume Biserrula pelecinus ssp. leiocarpa and the perennial shrub Colutea abyssinica. The greatest potential, however, is reserved for the highlands of Eritrea where germplasm is well adapted. Species found low in the landscape including from the genera Lotus, Trifolium and Medicago appeared well utilized. Different species found higher in the landscape including from the genera Indigofera, Tephrosia, Crotalaria, Trifolium schimperi, B. pelecinus ssp. leiocarpa and C. abyssinica were much less common, appeared under-utilized and may be under threat from genetic erosion. Animal production on the non-arable dry hillsides of the highlands would benefit from better utilization of these species through replanting some areas, careful grazing management and demonstration of the benefits of increasing the native legume component of these wild pastures.

Studies to quantify genetic variation in cassava germplasm, available within the national breeding programmes in Africa, have been limited. Here, we report on the nature and extent of genetic variation that exists within 1401 cassava varieties from seven countries: Tanzania (270 genotypes); Uganda (268); Kenya (234); Rwanda (184); Democratic Republic of Congo (DRC; 177); Madagascar (186); Mozambique (82). The vast majority of these genotypes do not exist within a formal germplasm conservation initiative and were derived from farmers' fields and National Agricultural Research Systems breeding programmes. Genotypes were assayed using 26 simple sequence repeat markers. Moderate genetic variation was observed with evidence of a genetic bottleneck in the region. Some differentiation was observed among countries in both cultivars and landraces. Euclidean distance revealed the pivotal position of Tanzanian landraces in the region, and STRUCTURE analysis revealed subtle and fairly complex relationships among cultivars and among landraces and cultivars analysed together. This is likely to reflect original germplasm introductions, gene flow including farmer exchanges, disease pandemics, past breeding programmes and the introduction of cultivars from the International Institute of Tropical Agriculture – Nigeria. Information generated from this study will be useful to justify and guide a regional cassava genetic resource conservation strategy, to identify gaps in cassava diversity in the region and to guide breeding strategies.

Grapevine cultivar identification is based mainly on two complementary methodologies: microsatellite (simple sequence repeat (SSR)) DNA analysis and traditional ampelography. Here, we report a direct multiplex PCR approach that allows the simultaneous amplification of 11 SSR loci from crude samples, i.e. bypassing DNA extraction, by using an engineered DNA polymerase improved to tolerate plant PCR inhibitors. Many different plant tissues were successfully amplified: leaf, root, wood, berry flesh and skin, stalk and must, from wine and table grape varieties, and rootstocks. The direct multiplex PCR that we propose is quicker and cheaper than the methodologies used until now, and provides accurate results. Thus, SSR DNA analysis becomes economically more accessible to a larger number of potential users in addition to research institutes.