Statistical discrimination between pollen tube growth and seed set in establishing self incompatibility in Gaura lindheimeri

The University of Minnesota Gaura breeding program is developing USDA Z3-4 winter-hardy genotypes via interspecific hybridization of G. lindheimeri (Z5-6) and G. coccinea (Z2-4). Prior to commencing interspecific hybridization, the reproductive barriers operating in both parental species need to be characterized. The objective of this research was to determine the type and stability of reproductive barriers operating in G. lindheimeri by statistical comparisons between pollen tube growth and seed set in a full-sib diallel. Slowed or aborted pollen tube growth in the style indicated the presence of a gametophytic self incompatibility (SI) system. A statistical method, female (FCC) and male (MCC) coefficients of crossability, was used to verify that a stable SI system was operating and that other reproductive barriers were present. Several genotypes also expressed stage-specific inbreeding depression and incongruity. The simple linear regression equation for FCC/MCC, using pollen tube growth, was Y = 0.0124 + 0.974X, which was much closer to the expected Y = 0.0 + 1.0X (indicating a stable SI system) than the equation for seed set, Y = 0.012 + 0.910X. Using pollen tube length, both general combining ability (GCA) and specific combining ability (SCA) values were highly significant for G. lindheimeri (P ≤ 0.001). Histograms were used to delineate cut-offs to identify intra-incompatible/inter-compatible classes and S allele groups. Four possible classes were identified, but further research is needed to verify S allele genotypes.