The variability in fresh and processed fruit flesh color of six clingstone processing peach [Prunus persica (L.) Batsch] genotypes was measured using CIELAB color variables. The genotypes were selected based on the relative fruit concentrations of β-carotene and β-cryptoxanthin. Significant (p < 0.0001) differences were found among the genotypes for the L*, a*, and b* color variables of fresh and processed fruit. Mean color change during processing, as measured by ΔELAB, was greatest for `Ross' and least for `Hesse'. A plot of the first two principal components (PCs) obtained from PC analysis of the L*, a*, and b* variables for fresh and processed fruit revealed three clusters of genotypes that match groupings based on the relative concentrations in fresh fruit of carotenoid pigments. Path analysis showed that variation in β-cryptoxanthin concentration was more precisely determined from color data than β-carotene concentration. Chemical names used: β-β-carotene (β-carotene), (3R)-β-β-caroten-3-ol (β-cryptoxanthin).

The development of gerbera (Gerbera jamesonii H. Bolus ex. Hooker) as a floricultural crop is traced from its collection as a botanical novelty in South Africa to its establishment as a commercial crop in the 1930s. The origin of the cultivated germplasm, G. jamesonii and G. viridifolia (DC) Schultz- Bipontinus, is discussed, as well as breeding work that occurred in Europe and the United States. The contributions of the two species to the cultivated germplasm is unknown. Early breeding in Europe was conducted by RI. Lynch at the Cambridge Botanic Gardens in England, R. Adnet at La Rosarie in Antibes, France; and by C. Sprenger in Italy. In the United States, early work was done at estates in New Jersey by Herrington and Atkins, and by the commercial growers Jaenicke and the J.L. Childs' Seed Co. Establishing the cold hardiness of the crop for temperate climates was an early goal of horticulturists and breeders. Much of the cultivated germplasm can be traced to material that passed through Cambridge and Antibes.

The distribution of hue (CIELAB colour notation) classes among flowers of the Davis population of gerbera (Gerbera jatnesonii H. Bolus ex Hooker) appears bimodal. This suggests that the genetic control of hue is determined by the segregation of a gene with large effect modified by additional genes with smaller effects. Complex segregation analysis (CSA), routinely employed in human genetic epidemiology, was used to study both qualitative and quantitative variation. CSA applies pedigree analysis through the consideration of transmission probabilities to optimize likelihood functions of various genetic models. Applying this technique to study flower hue of a sample representing generations 14, 15 and 16 of the Davis population, allowed identification of a putative dominant major gene with genotypic values for the dominant homozygote, heterozygote and recessive homozygote of 32, 32 and 71 degrees, respectively. This corresponds to the modes of the hue frequency distribution for the population. The putative major gene represents 0.66 of the total variation. The residual parent-offspring correlation (p0 = 0.2) measures the genetic contribution to the remainder of the variance.

The frequency distribution of gerbera flower hue in the Davis population of gerbera appears continuous and bimodal. This suggests that a gene of large effect may be segregating in a background of polygenic variation. CSA is a statistical technique developed in genetic epidemiology for investigating such complex traits without the need of inbred lines. The REGC program of SAGE (Elston, LSU Medical Center, New Orleans) uses the regressive models of G. Bonney (1984) through pedigree analysis to provide estimates of major gene parameters and residual correlations among relatives. Pedigrees obtained from generations 14, 15, and 16 indicate that a major dominant gene for hue is segregating and accounting for -0.66 of the total variation. The genotypic means are 32 degrees and 71 degrees for the aa and bb genotypes, respectively. The a allele is dominant to the b allele and has a frequency of 0.55. The residual parent-offspring correlation estimate is 0.2 and measures the genetic contribution to the remainder of the variance.

Ligule color of a Gerbera jamesonii H. Bolus ex Hooker population was analyzed with a reflectance spectrophotometer having a spectral capability of 400 to 700 nm and a cv <3% for the variables hue, chroma, and value. The observations for each variable had a continuous distribution; these broad distributions are possibly bimodal. The repeatability of hue, chroma, and value, determined as the correlation between measurements made on plants in December and those made on the same plants the following April, are 0.83, 0.82, and 0.86, respectively. The phenotypic correlations between value and chroma, value and hue, and chroma and hue are -0.35, 0.73, and 0.11, respectively. Some possible biochemical implications concerning the interaction of anthocyanin and carotenoid pigments are discussed. Reflectance spectroscopy and Commission International de l'Eclairage 1976 (L* a* b*) color space notation provide an objective and precise method for incorporating color into a recurrent selection program.