Mating system and effective pollen dispersal were studied in a natural stand of knobcone pine (Pinus attenuata Lemmon.) using 11 isozyme loci. Analyses were performed by fitting neighbourhood and mixed-mating models to multilocus genotypic arrays of offspring from four mother trees. Neighbourhoods consisted of all potential outcross males within 11 m of each mother tree (44, on average). Average outcrossing rates of the mother trees were estimated to be 0.97 and 0.96 for the respective models, whereas the population-wide outcrossing rate based on the mixed-mating model and a broader sample of mother trees was 0.92. The estimated proportion of offspring sired by males outside the neighbourhood of each mother tree was 0.56. Thus about 41 per cent of matings resulted from outcrossing with nearby males (within neighbourhoods). Distance and direction of individual males from mother trees and the size (tree height) of males played significant roles in determining outcross mating patterns within neighbourhoods. Male mating success increased with both proximity and tree size, although males east of mother trees accounted for more offspring than males in other directions.Despite the role of proximity, directionality and tree size in determining mating success within neighbourhoods, the effective number of males mating with each female seems to be large.Keywords: gene flow, mating success, mating Pinus attenuata.
IntroductionMating systems and patterns of pollen dispersal play a central role in plant population genetics (Levin & Kerster, 1974;Hamrick, 1989). They influence effective population size and the degree of population subdivision resulting either from drift or differential selection. In addition, they affect levels of inbreeding and genetic structure within populations. Knowledge of mating systems and gene flow is also important for managing breeding populations (Levin & Kerster, 1974;Adams & Birkes, 1991) and for designing efficient genetic conservation strategies (Ledig, 1986;Hamrick et al., 1991 (Clegg, 1980;Brown, 1990; Mitton, 1992). Mating pattern analysis, however, has usually been restricted to estimation of parameters in the mixedmating model, which subdivides all mating events into two classes: self-fertilization and random outcrossing (Fyfe & Bailey, 1951). The mixed-mating model makes several assumptions, including that the probability of an outcross is independent of maternal genotype, and outcross pollen-pool allele frequencies are homogeneous across maternal parents (Brown et a!., 1985;Neale & Adams, 1985).Evidence in forest trees, however, indicates that these and other assumptions of the mixed-mating model are often violated (Mitton, 1992). In addition, this mating model does not describe patterns of outcross mating within populations, such as the extent to which male mating success is a function of proximity to females or pollen fecundity.Methods of estimating fertilities of individual males within populations, based on genotypic arrays 252 J. BURCZYK ETAL.of offspring of individual mothers, wer...