Twenty landraces of cowpea (Vigna unguiculata (L.) Walp.) scattered throughout Algeria were compared through morphological and genetic characterization. At the morphological level, for qualitative characters there was no intra-landrace variation and for quantitative characters the variations were low except for landrace NAG2 Three different cultigroups were located in Algeria: Biflora that was dominant in the Sahara, Melanophtalmus in the North and Unguiculata including one landrace in Kabylia and two in Sahara. The AMOVA analysis indicated that the genetic variation was lower within than among agro-ecological regions. A Mantel test, revealed a correlation between the qualitative morphological data and the geographical data (R = 0.28; P \ 0.01), indicating that the degree of morphological change among landraces was roughly proportional to the geographical distances separating them.Genetic diversity was analyzed by using 11 random amplified polymorphic DNA (RAPD) and 12 intersimple sequence repeat (ISSR) markers. No intralandrace variability was found. The eleven RAPD primers yielded 77 bands, of which 45 (58.44%) were polymorphic; the genetic similarity ranged from 66.0 to 96.7%. The twelve ISSR primers provided a total of 104 bands, of which 65 (62.5%) were polymorphic; the genetic similarity ranged from 62.8 to 97.8%. cluster analysis showed a good match between genetic background and geographical distribution, which was confirmed by the results of the Mantel test. In particular, geographical data and genetic data were found to be correlated: (R = 0.33; P \ 0.01) for RAPD, (R = 0.37; P \ 0.01) for ISSR, and (R = 0.33; P \ 0.01) for a combined RAPD-ISSR dataset. Moreover, despite the absence of significant correlation between morphological and RAPD data (R = 0.14; P = 0.14), significant correlations between morphological data and both ISSR (R = 0.27, P \ 0.05) and a combined RAPD-ISSR dataset (R = 0.22, P \ 0.05) were noted. ISSR markers were better linked to morphological variation than were RAPD markers. However, despite this, genetic distances among these landraces were found to be essentially the same no matter which markers were used.

Exclusive selection for yield raises, the harvest index of self-pollinated crops with little or no gain in total bipmass. In addition to selection for yield, it is suggested that efficient breeding for higher yield requires simultaneous selection for yield's three major, genetically controlled physiological components. The following are needed: (1) a superior rate of biomass accumulation. (2) a superior rate of actual yield accumulation in order to acquire a high harvest index, and (3) a time to harvest maturity that is neither shorter nor longer than the duration of the growing season. That duration is provided by the environment, which is the fourth major determinant of yield. Simultaneous selection is required because genetically established interconnections among the three major physiological components cause: (a) a correlation between the harvest index and days to maturity that is usually negative; (b) a correlation between the harvest index and total biomass that is often negative, and (c) a correlation between biomass and days to maturity that is usually positive. All three physiological components and the correlations among them can be quantified by yield system analysis (YSA) of yield trials. An additive main effects and multiplicative interaction (AMMI) statistical analysis can separate and quantify the genotype × environment interaction (G × E) effect on yield and on each physiological component that is caused by each genotype and by the different environment of each yield trial. The use of yield trials to select parents which have the highest rates of accumulation of both biomass and yield, in addition to selecting for the G × E that is specifically adapted to the site can accelerate advance toward the highest potential yield at each geographical site. Higher yield for many sites will raise average regional yield. Higher yield for multiple regions and continents will raise average yield on a world-wide basis. Genetic and physiological bases for lack of indirect selection for biomass from exclusive selection for yield are explained.

No abstract

a b s t r a c tTo set up a rational collecting strategy for germplasm of the edible-seeded cucurbit Cucumeropsis mannii, a study was conducted using 24 morphological and seven putative enzyme markers to determine the intra-specific variability from 16 and 22 accessions (representing three cultivars), respectively. The analysis of variance, showed a significant difference between the three cultivars. Principal component analysis pointed out a variation among individuals, mainly on the basis of flower, fruit, and seed size. Dendrogram with UPGMA method allowed clustering of the cultivars. Genetic diversity indices estimated equalled: 9.96% for the proportion of polymorphic loci (P), 1.10 for the number of alleles (A) and 0.023 for observed heterozygosity (H o ). The level of the within accessions genetic diversity (H S ¼ 0.078) was higher than among accessions (D ST ¼ 0.042). Nei's genetic distances between the three cultivars were also low (0.079-0.147), indicating a high degree of similarity of the analysed cultivars.Crown

Seed dormancy and its impact on the soil seed bank for wild Lima bean (Phaseolus lunatus) populations were studied in the Central Valley of Costa Rica. Five populations were selected in contrasted environments. In all cases, distribution of seeds in the soil was limited to 3 cm depth. No innate dormancy was observed but combination of hard seed coat and hilum opening controlled by environmental conditions were responsible for an induced dormancy and the constitution of a persistent seed bank. Breaking of this dormancy was obtained by a brief elevation of temperature from 25° to 45°C. Impacts of this phenomenon concern both genetic and demographic aspects of in situ conservation of the species. Consequences on ex situ conservation are mainly related with the regeneration of the seed collection.