Incidence and severity of fire blight [Erwinia amylovora (Burr.) Winslow, Broadhurst, Buchanan, Krumwiede, Rogers, and Smith] following field infection were recorded using families resulting primarily from open-pollination of Malus sylvestris (L.) Mill. var.domestica (Borkh.) Mansf. cultivars and a few other Malus Mill. sp. The families were structured as three sublines, planted in three successive years (1992 to 1994), of a diverse population of apple germplasm established at HortResearch, Hawkes Bay, New Zealand. The incidence of fire blight varied among the sublines with the oldest planting exhibiting more fire blight. Flowering trees were more likely to be infected than nonflowering trees, in terms of both incidence and severity. Furthermore, the level of fire blight was related to flowering date, with later flowering trees having higher levels. Thus, family means and narrow-sense heritability estimates were computed after first adjusting the fire blight score for flowering date by fitting a linear model. Provenance of origin of the maternal parent explained little variation except that M. sieversii Lebed. families were more resistant than M. sylvestris var. domestica families in one subline. Family means computed using all trees, and those from only flowering trees were highly correlated. Families from open-pollination of M. honanensis Rehder and M. xhartwiggii Koehne females were among the more susceptible. Those from several European M. sylvestris var. domestica cultivars as well as from M. baccata (L.) Borkh. and M. toringoides (Rehder) Hughes females were among the more resistant families. Narrow-sense heritability estimates ranged from 0.05 to 0.85 depending on the subline, with most estimates between 0.12 and 0.36. They were higher in the two older sublines that consisted primarily of open-pollinated families from M. sylvestris var. domestica, and lower in the younger subline that consisted primarily of M. sieversii, due to lower incidence and severity in the latter subline. Breeders who consider potential complications of juvenility, tree size, and flowering date in relation to infection periods should be able to exploit field epidemics to perform effective selection.
James J. Luby, Peter A. Alspach, Vincent G.M. Bus, and Nnadozie C. Oraguzie
J.M. Quintana, H.C. Harrison, J. Nienhuis, J.P. Palta, K. Kmiecik, and E. Miglioranza
To understand the genetics that control pod Ca concentration in snap beans, two snap bean (Phaseolus vulgaris L.) populations consisting of 60 genotypes, plus 4 commercial cultivars used as checks, were evaluated during Summers 1995 and 1996 at Hancock, Wis. These populations were CA2 (`Evergreen' × `Top Crop') and CA3 (`Evergreen' × `Slimgreen'). The experimental design was an 8×8 double lattice repeated each year. No Ca was added to the plants grown in a sandy loam soil with 1% organic matter and an average of 540 ppm Ca. To ensure proper comparison for pod Ca concentration among cultivars, only commercial sieve size no. 4 pods (a premium grade, 8.3 to 9.5 mm in diameter) were sampled and used for Ca extractions. After Ca was extracted, readings for Ca concentration were done via atomic absorption spectrophotometry. In both populations, genotypes and years differed for pod Ca concentration (P = 0.001). Several snap bean genotypes showed pod Ca concentrations higher than the best of the checks. Overall mean pod Ca concentration ranged from a low of 3.82 to a high of 6.80 mg·g-1 dry weight. No differences were detected between the populations. Significant year×genotype interaction was observed in CA2 (P = 0.1), but was not present in CA3. Population variances proved to be homogeneous. Heritability for pod Ca concentration ranged from 0.48 (CA2) to 0.50 (CA3). Evidently enhancement of pod Ca concentration in beans can successfully be accomplished through plant breeding.
William J. Martin and Dennis P. Stimart
Stomatal density during plant development and inheritance of the trait were investigated with the goal of utilizing stomatal density as a correlated trait to cutflower postharvest longevity in Antirrhinum majus L. Inbred P1 (stomatal index = 0.2) was hybridized to inbred P2 (stomatal index = 0.3) to produce F1 (P1 × P2), which was backcrossed to each parent producing BCP1 (F1 × P1) and BCP2 (F1 × P2). P1, P2, F1, BCP1, and BCP2 were used to examine changes in stomatal density with plant development and early generation inheritance. An F2 (F1 self-pollinated), and F3, F4, and F5 families, derived by self-pollination and single seed descent, were used to obtain information on advanced generation inheritance. Stomatal density was stable over time and with development of leaves at individual nodes after seedlings reached two weeks of age. Therefore, stomatal density can be evaluated after two weeks of plant development from a leaf at any node. Stomatal density is quantitatively inherited with narrow sense heritabilities of h2 F2:F3 = 0.47 to 0.49, h2 F3:F4 = 0.37 ± 0.06 to 0.60 ± 0.07, and h2 F4:F5 = 0.47 ± 0.07 to 0.50 ± 0.07.
Douglas V. Shaw
The heritabilities of, and genetic correlations among, variables that describe internal and external color in fresh strawberry (Fragaria × anarrassa) fruit were estimated using factorial analyses of seedlings from 20 controlled crosses. Hunter L and a values, and a subjective score generated by comparison with color plates were obtained for seedling genotypes and their parents at two locations. Genetic effects were responsible for 33% to 61% of the phenotypic variance for color traits, after correction for location effects. Means for objective color variables differed significantly between locations, but means for subjective color scores did not. Genetic × location interaction variances were usually nonsignificant, and were < 12% of the phenotypic variance for all variables. Phenotypic and genetic correlations between objective and subjective color scores were significant and large (absolute values of r = 0.42-0.69; rg = 0.84-1.00). Multiple regression of subjective scores on L and a explained 69% and 59% of the phenotypic variation for external and internal color, respectively. Genetic correlations between measures of internal and external color were small and mostly nonsignificant, suggesting that separate sets of genes condition these traits.
Douglas V. Shaw, W.D. Gubler, Kirk D. Larson, and John Hansen
Resistance to wilt caused by Verticillium dahliae Kreb. was evaluated for 41 strawberry genotypes from the Univ. of California breeding program and 1000 offspring from crosses among 23 of these genotypes. Runner plants from these genotypes and seedlings were inoculated with a conidial suspension containing a mixture of five isolates of V. dahliae from strawberry. Symptoms were scored as the number of dead or seriously stunted plants per plot, or based on a subjective phenotypic resistance score assigned to each plot on five dates during the spring after planting. Most of the California germplasm is highly susceptible to V. dahliae, with an average resistance score of 2.1 (±0.10) and 84.1% (±2.1) plants stunted or dead compared with a score of 3.2 (±0.24) and 57.4% (±4.9) of plants stunted or dead for a control set of six non-California genotypes identified previously as resistant. However, a broad range of intermediate resistance was detected, and 4 of the 41 California genotypes evaluated had resistance scores superior to the mean score for the non-California resistant checks. Plot-mean heritabilities for resistance and stunting scores estimated using genotypic, full-sib family, and offspring-parent analyses ranged from 0.44 to 0.88. Comparison of different estimates of variance components suggests that half or more of the genotypic variance for resistance traits detected is due to the additive effects of genes. There appears to be sufficient variation within the California population to proceed with an effective selection program, despite the absence of directional selection for resistance during the past 3 decades. However, developing cultivars with adequate resistance will ultimately depend on the recovery of transgressive segregants from superior parents, as even the most resistant genotypes from all sources showed some disease symptoms.
The effectiveness of directional phenotypic selection to improve tomato (Lycopersicon esculentum Mill.) seed germination under salt-stress was investigated. Seed of F2 and F3 progeny of F1 hybrids between a salt-tolerant (PI174263) and a salt-sensitive (UCT5) tomato cultivar were evaluated for germination response at three stress levels of 100 (low), 150 (intermediate), and 200 mm (high) synthetic sea salt (SSS). At each salt-stress level, the most tolerant individuals, as determined by the germination speed, were selected. Selected individuals (F2s or F3s) were grown to maturity and self-pollinated to produce F3 and F4 progeny families. The selected progeny from each experiment were evaluated for germination at four treatment levels of 0 (nonstress), 100, 150, and 200 mm SSS and were compared with unselected populations. The results indicated that selections were equally effective at all three salt-stress levels and in F2 and F3 generations and significantly improved seed germination of progeny under salt-stress and nonstress treatments. Estimates of realized heritability for rapid germination under the various salt-stress levels ranged from 0.67 to 0.76. Analysis of response and correlated response to selection indicated a genetic correspondence of up to 100% between germination at different salt-stress levels. Genotypic family correlations between germination at the low, intermediate, and high salt-stress levels ranged from 0.67 to 0.89, and those between nonstress and salt-stress conditions ranged from 0.25 (between 0 and 200 mm) to 0.71 (between 0 and 100 mm salt). The results indicated that similar or identical genes with additive genetic effects contributed to rapid germination response of tomato seeds at different salt-stress levels. Thus, selection at one stress level resulted in progeny with improved germination at diverse salt-stress levels. The results also indicated that to improve tomato seed germination, selection can be based on individual seed performance and early segregating generations.
Tommy E. Thompson and L.J. Grauke
Thirty-six cultivars and 948 seedlings from 15 controlled crosses in the U.S. Dept. of Agriculture-Agricultural Research Service pecan [Carya illinoinensis (Wangenh.) K. Koch] breeding program at Brownwood, Texas, were rated for susceptibility to nut scab [Cladosporium caryigenum (Ell. et Lang.) Gottwald] to determine heritability of this trait. Differences between parents and progenies, and within progenies, were highly significant. Within most families, a complete range of resistance reactions were evident, from fully susceptible to fully resistant. Heritability of resistance was determined by regressing individual progeny values on female, male, and midparent values, with the midparent heritability estimate being the highest (0.54). This moderate level of additive gene action and the identification of superior parents in this study will contribute to the efficiency of breeding resistant cultivars.
Yiran Yu, James Harding, and Thomas Famula
Additive genetic components of variance and narrow-sense heritabilities were estimated for flowering time and cut-flower yield for generations 8-13 of the Davis population of gerbera, using the least squares (LS) and restricted maximum likelihood (REML)
methods. Estimates of heritability for flowering time were 0.54 and 0.50 using REML and LS, respectively, indicating a close agreement between the two methods. However, estimates of heritability for cut-flower yield were 0.30 and 0.46 from REML and LS. This may result from the fact that cut-flower yield was selected in each generation; flowering time was not. Realized heritability for cut-flower yield was estimated to be 0.26 which agreeded more closely with the heritability estimated from REML. The advantages of REML, and its applications in the estimation of components of genetic variance and heritability of plant populations are discussed.
Joseph M. Kemble and Randolph G. Gardner
The heritability of shortened fruit maturation (SFM) period in Cornell 871213-1, an inbred cherry tomato [Lycopersicon esculentum var. cerasiforme (Dunal.) A. Gray] line, was estimated from a greenhouse experiment. Cornell 871213-1 was crossed with the cherry tomato line NC 21C-1. Mean fruit maturation period (FMP) (days from anthesis to the breaker stage of fruit color) was 40.8 days for NC 21C-1 and 32.0 days for Cornell 871213-1. Parental, F1, F2, and backcross generations all differed in mean FMP and yielded, estimates of broad- and narrow-sense SFM heritabilities of 72% and 40%, respectively, on a single-plant basis. A test for midparent heterosis showed significance. Genetic control of SFM was quantitative in nature and highly dominant. A field study of an F2 population developed from the cross Cornell 871213-1 × NC 84173, the latter a large-fruited tomato line (Lycopersicon esculentum Mill.), gave a mean FMP of 48.4 and 31.2 days for NC 84173 and Cornell 871213-1, respectively. The F1 and F2 generations had FMP of 33.1 and 34.7 days, respectively. The parents, F1, and F2 generations all differed in FMP. Parental, F1, and F2 generations yielded an estimate of broad-sense SFM heritability of 64% on a single-plant basis. F3 progenies from selected F2 s were grown in a greenhouse, and F3-F2 regression analysis gave a narrow-sense SFM heritability of 39%. Parental means differed from each other and from the F1 and F2 means for period from sowing to anthesis, fruit weight, and locule number. F1 and F2 means did not differ for any trait and were far below the midparent values, approaching Cornell 871213-1 for each trait except for the number of days from sowing to anthesis. Significant correlations existed in the F2 generation between FMP and fruit weight (0.61) and between fruit weight and locule number (0.69). Significant correlations existed between selected F2s and their F3 progeny for FMP (0.53), fruit weight (0.78), and days from sowing to anthesis (0.78). In the F3 generation, a significant correlation occurred between FMP and fruit weight (0.48). F3-F2 regression and realized heritabilities were used as two estimates of narrow-sense heritability (29% and 31%, respectively) for days from sowing to anthesis.
Paul C. St. Amand and Todd C. Wehner
Heritability of resistance to gummy stem blight (Didymella bryoniae (Auersw.) Rehm.) was measured in two diverse cucumber (Cucumis sativus L.) populations [North Carolina elite slicer 1 (NCES1) and North Carolina wide base pickle (NCWBP)]. Heritability was estimated using parent-offspring regression and half-sib family analysis in North Carolina field tests during 1991 and 1992. NCES1 is a slicing cucumber population with a narrow genetic base, and NCWBP is a pickling cucumber population with a wide genetic base. Heritability estimates were low to moderate ranging from 0.12 to 0.49 for the gummy stem blight leaf rating and from -0.03 to 0.12 for stem rating. Estimates of gain from selection were at least two times larger for selection based on half-sib families than for mass selection for all traits in both populations. Approximately three to five cycles of selection would be required to improve the NCES1 population mean for gummy stem blight leaf resistance by one rating scale unit, and three to four cycles of selection would be required to improve the NCWBP population mean for gummy stem blight leaf resistance by one rating scale unit, based on half-sib family selection. One rating scale unit decrease is equivalent to an 11% reduction in susceptibility. Gain would be slower if selecting for stem, or leaf and stem resistance. A moderate amount of additive genetic variation exists in both populations for gummy stem blight leaf resistance, but estimates for additive genetic variation for stem resistance indicate little to no additive genetic variation. Development of populations specifically for greater initial resistance and greater additive variance than found in these populations should aid in selection for resistance.