| Ecological investigation and experiments were conducted in the big sagebrush (Artemisia tridentata) hybrid zones in Utah to test the two hybrid zone models. The dynamic equilibrium model posits universal hybrid unfitness resulting from the disruption of coadapted gene complexes, i.e., selection is solely caused by intrinsic genetic interactions and hence is endogenous. This model predicts that hybrid zones should occur at areas of low population densities.;Alternatively, the bounded hybrid superiority model assumes that the interplay of genotype and environment determines fitness and the positions and stability of hybrid zones. In this model, selection is ecological and hence is exogenous. This model predicts that hybrid zones should occur at ecotones.;The elemental concentrations, thickness, and pH of soils at the parental taxa habitats differed significantly from each other, and from those of the hybrid zones. The hybrid zone soils are not just simple blends of the two parental habitat soils. Rather, they possess novel characteristics, and are more variable than the parental habitat soils.;There is no evidence of population density troughs in the big sagebrush hybrid zone, which is clearly at odds with the prediction of dynamic equilibrium model. The big sagebrush parentals are associated with different groups of unique species and occupy markedly different soils. Thus, the big sagebrush hybrid zone in Clear Creek Canyon appears to occur at an ecotone, which is clearly consistent with the prediction of the bounded hybrid superiority model.;Elemental concentrations differed significantly among big sagebrush parental and hybrid taxa. Each of the big sagebrush taxa also responded differentially to edaphic heterogeneity. Significant genotype by environment interactions were shown for elemental uptake, germination, growth, and reproduction, as assumed by the bounded hybrid superiority model. Further, a composite fitness derived from reciprocal transplant experiments showed that big sagebrush hybrids were the most fit in the hybrid garden, but less fit than the parental taxon in the corresponding native parental garden. These results strongly support the bounded hybrid superiority model. |