An earliness per se gene, designated Eps-A m 1, was mapped in diploid wheat in F 2 and single-seed descent mapping populations from the cross between cultivated (DV92) and wild (G3116) Triticum monococcum accessions. A QTL with a peak on RFLP loci Xcdo393 and Xwg241, the most distal markers on the long arm of chromosome 1A m , explained 47% of the variation in heading date (LOD score 8.3). Progeny tests for the two F 2:3 families with critical recombination events between Xcdo393 and Xwg241 showed that the gene was distal to Xcdo393 and linked to Xwg241. Progeny tests and replicated experiments with line #3 suggested that Eps-A m 1 was distal to Xwg241. This gene showed a large effect on heading date in the controlled environment experiments, and a smaller, but significant, effect under natural conditions. Eps-A m 1 showed significant epistatic interactions with photoperiod and vernalization treatments, suggesting that the different classes of genes affecting heading date interact as part of a complex network that controls the timing of flowering induction. Besides its interactions with other genes affecting heading date, Eps-A m 1 showed a significant interaction with temperature. The effect of temperature was larger in plants carrying the DV92 allele for late flowering than in those carrying the G3116 allele for early flowering. Average differences in heading date between the experiments performed at 16°C and 23°C were approximately 11 days (P < 0.001) for the lines carrying the Eps-A m 1 allele for early flowering but approximately 50 days (P < 0.0001) for the lines carrying the allele for late flowering. The large differences in heading time (average 80 days) observed between plants carrying the G3116 and DV92 alleles when grown at 16°C, suggest that it would be possible to produce very detailed maps for this gene to facilitate its future positional cloning.

Polymerase chain reaction fragment length polymorphisms and nucleotide sequences for a cytochrome P450 gene encoding flavonoid-3',5'-hydroxylase, Hf1, were studied in 19 natural taxa of Petunia. Natural Petunia taxa were classified into six groups based on major insertion or deletion events that occurred only in intron II of the locus. The maximum parsimony method was used to calculate strict consensus trees based on nucleotide sequences in selected regions of the Hf1 locus. Petunia taxa were divided into two major clades in the phylogenetic trees. Petunia axillaris (including three subspecies), P. exserta, and P. occidentalis formed a clade with 100% bootstrap support. This clade is associated with a consistently inflexed pedicel, self-compatibility in most taxa, and geographical distribution in southern and western portions of the genus range. The other clade, which comprised the remainder of the genus is, however, less supported (up to 71% bootstrap); it is characterized by a deflexed pedicel in the fruiting state (except P. inflata), self-incompatibility, and a northeastern distribution. A nuclear gene, Hf1, seems to be a useful molecular marker for elucidating the phylogeny of the genus Petunia when compared with the nucleotide sequence of trnK intron of chloroplast DNA.

Petunia axillaris occurs in temperate South America and consists of three allopatric subspecies: axillaris, parodii, and subandina. Previous studies have revealed that subsp. axillaris is self-incompatible (SI), subsp. parodii is self-compatible (SC) in Uruguay, and subsp. subandina is SC in Argentina. The SI/SC status over the entire distribution range is not completely understood, however. The objective of this study was to examine the overall SI/SC status of the respective subspecies in comparison with floral morphology. The results confirmed that subsp. parodii and subsp. subandina were SC throughout the distribution range, and that subsp. axillaris was also SC in Brazil and in most of the Argentinean territory. The SI P. axillaris occurs in the natural population only between 34 and 36 degrees S, along the eastern shore of South America. The Brazilian and Uruguayan subsp. axillaris differed in SI/SC status and floral morphology. We discuss the cause of this difference.

The effects of Triticum monococcum glutenin loci on cookie making quality and predictive tests for bread making quality were evaluated in recombinant substitution lines (RSLs) between chromosome 1A m from T. monococcum and chromosome 1A from Chinese Spring. All four combinations of high molecular weight (HM r -GS) and low molecular weight glutenin alleles (LM r -GS) were studied in a factorial design to evaluate their interactions. Grain protein content was used as a covariable to evaluate the effect of these loci independently of the variation in protein content among lines. No significant interactions were detected indicating an additive effect. RSLs carrying the HM r -GS from T. monococcum showed a 13·6% increase in SDS sedimentation volume (p=0·004) and a significant reduction in cookie diameter (−5·2%, p=0·02), and cookie quality (−6·8%, p=0·02). RSLs carrying the LM r -GS from T. monococcum showed a significant decrease in the proportion of polymeric protein (−2·8%, p<0·0001), SDS sedimentation volume (−8·1%, p=0·03) and gluten strength (−16·5%, p=0·01), and a significant increase in cookie quality (5·9%, p=0·05). The T. monococcum LM r -GS allele has potential value to be used in soft wheat breeding programs. These results suggest that diploid T. monococcum could be a valuable source for new allelic variation for storage proteins loci and new quality characteristics.