Wild Anagallis monelli exhibits blue or orange flower colors in geographically isolated populations. A new red flower color was developed through breeding, and a three-gene model was proposed for the inheritance of flower color in this species. In this study, blue and orange wild diploid accessions were used as parents to develop six F2 populations (n = 19 to 64). Sexual compatibility between blue and orange wild individuals was low with only 29% of the hybridizations producing F1 individuals. Six of 14 cross combinations between F1 siblings produced fruits, and fruiting success ranged from 55% to 90%. The number of seeds per fruit averaged 14.1 and germination rates for the F2s were low (16.8% to 30.7%). In three of six F2 populations obtained, flower color segregation ratios for orange, blue, and red were not significantly different from the expected ratios under a previously proposed three-gene model. White flower color was obtained as a fourth color variant in two of the remaining F2 populations. For one of these populations, segregation ratios were not significantly different from expected ratios for an expanded four-gene model. White flowers did not contain anthocyanidins, suggesting that there was a mutation in the early stage of the anthocyanin pathway. Orange flower color was found to be primarily the result of pelargonidin, blue to malvidin, and red to delphinidin. These three pigments may be present simultaneously, and their ratios play a significant role in determining flower color. Other factors such as copigments, metal ions, or a different molecular conformation of the anthocyanin could also be involved in flower color determination.
This study was designed to compare snap and dry beans (Phaseolus vulgaris L.) for pod Ca concentration, and to identify genetic resources that might be useful in breeding programs directed to increase Ca concentration in bean pods. Pods from eight snap bean and eight dry bean cultivars were evaluated for Ca concentration during 1995 and 1996 at Hancock, Wis. A randomized complete-block design was utilized with three replications in 1995 and six in 1996. Beans were planted in June and hand-harvested in August for both experiments. Soil Ca at planting time was 580 mg·kg–1 in 1995 and 500 mg·kg–1 in 1996. No additional Ca was added. Plots consisted of 10 plants each. At harvest, a pooled sample of 10 to 15 size no. 4 pods was collected from each plot. Atomic absorption spectrophotometry was used to determine Ca content. Significant differences (P ≤ 0.01) were detected among and within bean types (dry and snap). Although bean type × year interaction was nonsignificant, a strong year effect was observed (P ≤ 0.01). Snap beans (4.6 ± 0.7 mg·g–1 dry weight) had significantly higher pod Ca concentration than did dry beans (4.2 ± 0.6 mg·g–1 dry weight). Within snap beans, `Checkmate' had the highest pod Ca concentration (5.5 ± 0.3 mg·g–1 dry weight) and `Nelson' the lowest (3.8 ± 0.3 mg·g–1 dry weight). Within dry beans, `GO122' had the highest (5.1 ± 0.4 mg·g–1 dry weight) and `Porrillo 70' the lowest pod Ca concentration (3.6 ± 0.3 mg·g–1 dry weight). Six cultivars had pod Ca concentrations significantly (P ≤ 0.01) higher than the overall mean (4.4 ± 0.3 mg·g–1 dry weight).
To measure the effect of added Ca fertilizer on the Ca concentration of snap bean pods, four snap bean cultivars were grown during Summer 1996 and 1997 at Hancock, Wis. Fertilizer treatments consisted of 80 kg of Ca per hectare applied as Ca sulfate (CaSO4·2H2O) or Ca nitrate [Ca(NO3)2], and the control (no Ca applied. The experimental design was a randomized complete block with a factorial set of treatments (4 × 3). Calcium sulfate was applied at planting, whereas Ca nitrate was split applied four times at weekly intervals starting 1 week before flowering. Yield and Ca concentration in pods were determined. The statistical analysis showed no significant effect of Ca fertilizers on pod Ca concentration or yield. A strong cultivar effect was detected for both parameters measured. `Evergreen' (5.47 mg Ca per gram dry weight) had the highest pod Ca concentration and `Labrador' (4.10 mg Ca per gram dry weight) the lowest. No significant fertilizer × cultivar interactions were observed. Results for pod Ca concentration remained consistent, even when significant year effects were found for both parameters.