No abstract

Extensive livestock is a basic socio‐economic feature of the Mediterranean region whose environmental and economic sustainability depends on the ability of forage resources to withstand climatically stressful conditions. Perennial forages such as tall fescue can be a valuable alternative to annuals, if they can survive across successive summer droughts. Three‐year dry matter yield and plant survival of five cultivars of Mediterranean‐type tall fescue were evaluated in six sites of Algeria, France, Italy, Morocco and Portugal, with the following objectives: (i) modelling adaptive responses and targeting cultivars as a function of environmental factors associated with genotype × location interaction; and (ii) defining plant ideotypes, adaptation strategies and opportunities for international co‐operation for regional breeding programmes. Site mean yield and winter temperatures were positively correlated, whereas sward persistence was positively correlated to lower site heat and drought stress. Cultivar adaptation was adequately modelled by factorial regression as a function of site spring–summer (April–September) drought stress (long‐term potential evapotranspiration minus actual water available) for yield, and annual drought stress for final persistence. Specific‐adaptation responses to high‐ or low‐stress environments emerged which were consistent with drought‐stress levels of cultivar selection environments. However, the wide‐adaptation response of cultivar Flecha suggested that breeding for wide adaptation can be feasible.

Understanding of adaptation targets, selection environments, genetic resources, and plant types is required in breeding tall fescue [Lolium arundinaceum (Schreb.) S.J. Darbysh.] and orchardgrass (Dactylis glomerata L.) for Mediterranean environments prone to drought stress. Fourth‐year forage yield of seven orchardgrass and five fescue cultivars grown in Algeria and Sardinia revealed (i) crossover cultivar × location interaction in orchardgrass, with dormant germplasm performing best in the drier Algerian site, and nondormant Mediterranean material performing best in Sardinia; and (ii) the advantage of tall fescue over orchardgrass in both sites. Yield of Algerian populations of orchardgrass and tall fescue grown in Algeria was generally lower than the best control (‘Flecha’ tall fescue). Other studies in Italy showed (i) the possibility to select orchardgrass from Mediterranean germplasm that combines summer survival under moderate stress with response to summer rain events; (ii) the nil effect in a Mediterranean site, and the slightly positive effect in a subcontinental climate site of endophyte infection on tall fescue survival; and (iii) the adaptation of Mediterranean and Continental fescue germplasm to their respective climatic areas.

In order to increase genetic variability for chickpea improvement, the Kabuli genotype, variety Ghab4, was treated with 280 Grays of gamma rays (Cobalt 60). Field characterization began with the M2 generation. A total of 135 M2 families were sown in the field resulting in approximately 4,000 plants. Traits related to phenology (days to flowering, days to maturity), plant morphology of vegetative parts (plant height, height of first pod, number of primary branches per plant) and yield (number of seeds per pod, total number of pods per plant, total number of seeds per plant, seed yield and hundred seed weight) were recorded and analyzed to evaluate genetic variability. An evaluation of the efficacy of low-cost TILLING (Targeting Induced Local Lesions IN Genomes) to discover mutations in the M2 generation was undertaken. Mutation screening focused on genes involved in resistance to two important diseases of chickpea; Ascochyta blight (AB) and Fusarium wilt (FW), as well as genes responsible for early flowering. Analysis of variance showed a highly significant difference among mutant families for all studied traits. The higher estimates of genetic parameters (genotypic and phenotypic coefficient of variation, broad sense heritability and genetic advance) were recorded for number of seeds per plant and yield. Total yield was highly significant and positively correlated with number of pods and seeds per plant. Path analysis revealed that the total number of seeds per plant had the highest positive direct effect followed by hundred seed weight parameter. One cluster from nine exhibited the highest mean values for total number of pods and seeds per plant as well as yield per plant. According to Dunnett’s test, 37 M2 families superior to the control were determined for five agronomical traits. Pilot experiments with low-cost TILLING show that the seed stock used for mutagenesis is homogeneous and that small mutations do not predominate at the dosage used.