Environmental effects on pollen–pistil compatibility between Phlox cuspidata and P. drummondii (Polemoniaceae): implications for hybridization dynamics

Abstract

Postpollination mechanisms of reproductive isolation can critically influence the amount of gene flow between hybridizing species. While much evidence exists for genetically based pollen-pistil incompatibility, we show that environmental variation also influences the postpollination performance of heterospecific pollen in the annual Phlox hybrid system. Thus, the environmental segregation of species can influence hybridization dynamics. We found that P. cuspidata was restricted to soils of low Ca concentrations in the field and performed better under experimentally low Ca; P. drummondii was able to inhabit high-Ca soils and sometimes performed better in this environment. To determine whether soil Ca influenced pollen-pistil compatibility in a manner that alters pollen siring success, single-donor pollinations were performed in a completely factorial crossing design between species, maternal Ca environments, and paternal Ca environments. Maternal and paternal environments interacted in their effects on pollen-pistil compatibility for both inter- and intraspecific crosses, such that pollen performance was highest when mothers and fathers were grown in different soil Ca environments. These results suggest that when Phlox species predictably inhabit different environments, environmental heterogeneity can impede the processes of speciation and local adaptation by enhancing the performance of pollen dispersed across species and environments.

Pollen competition and environmental effects on hybridization dynamics between Phlox drummondii and Phlox cuspidata

“…In a recent study of the annual Phlox hybrid system, we found that, in the absence of interspecific pollen competition, edaphic variation in soil Ca influenced in vivo performance of both conspecific and heterospecific pollen (Ruane and Donohue 2007). While traits expressed early after pollen deposition (i.e.…”

“…By identifying a diversity of isolating mechanisms responsible for decreased gene flow between species, past research has helped us understand how closely related species maintain separate identities in the absence of geographical isolation. The efficiency with which these mechanisms maintain species integrity, however, depends on biotic and abiotic variation that pervades natural systems (Ruane and Donohue 2007). Here, we examine the effects of edaphic variation on pollen siring success within the context of interspecific pollen competition.…”

Mating system and the hybridization between self-compatible Phlox cuspidata and self-incompatible Phlox drummondii

“…Because pollen of P. cuspidata grows more slowly than P. drummondii pollen in P. cuspidata pistils (Ruane and Donohue 2007), I assessed whether P. drummondii pollen grains can still outcompete P. cuspidata when P. cuspidata pollen is deposited first, as is likely in the field. The degree to which weak SI in P. cuspidata influences hybridization frequency will depend on the timing of self-pollination.…”

“…Since P. cuspidata pollen also grows more slowly than P. drummondii pollen in P. drummondii pistils (Ruane and Donohue 2007), I assessed whether P. drummondii pollen grains can still outcompete P. cuspidata when faced with a time deficit. To determine the effect of the earlier deposition of heterospecific pollen on the proportion of hybrid progeny produced by P. drummondii maternal plants, I performed the following two pollination treatments: conspecific and heterospecific pollen grains were deposited simultaneously and heterospecific pollen grains were deposited 5 h before conspecific pollen grains.…”

“…I chose 5 h because preliminary trials revealed that this time span typically allowed enough time for pollen germination and sometimes allowed enough time for ovule fertilization. When styles were harvested 5 h after pollen deposition, I quantified pollen germination, the proportion of germinated pollen tubes that reached the base of the style, the proportion fruit set, and the number of seeds sired per fruit (see Ruane and Donohue 2007 for in vivo methods). When styles were not harvested, the proportion fruit set and the number of seeds sired per fruit were quantified.…”