Patterns of Reproductive Isolation in Three Angiosperm Genera

Abstract

Abstract. Analyses among animal species have found that reproductive isolation increases monotonically with genetic distance, evolves more quickly for prezygotic than postzygotic traits, and is stronger among sympatric than allopatric species pairs. The latter pattern is consistent with expectations under the reinforcement hypothesis. To determine whether similar trends are found among plant species, patterns of reproductive isolation (postpollination prezygotic, postzygotic, and ''total'' isolation) in three plant genera (Glycine, Silene, Streptanthus) were examined using data from previously published artificial hybridization experiments. In Silene, all measures of reproductive isolation were positively correlated with genetic distance. In contrast, in Glycine and Streptanthus, correlations between reproductive isolation and genetic distance were weak or nonsignificant, possibly due to the influence of biologically unusual taxa, variable evolutionary forces acting in different lineages, or insufficient time to accumulate reproductive isolation. There was no evidence that postpollination prezygotic reproductive isolation evolved faster than postzygotic isolation in Glycine or Silene. We also detected no evidence for faster accumulation of postmating prezygotic isolation between sympatric than allopatric species pairs; thus we found no evidence for the operation of speciation via reinforcement. In Silene, which included six polyploid species, results suggest that changes in ploidy disrupt a simple monotonic relationship between isolation and genetic distance.

Factors contributing to the accumulation of reproductive isolation: A mixed model approach

“…The lack of allopatric pairs for the postzygotic and total reproductive isolation measurements precluded analysis of the effects of geographical context on these measures, although the general relationship between reproductive isolation and genetic distance was positive and significant consistent with Moyle et al. (2004); data not shown.…”

“…(2004). There was a significant relationship between genetic distance and reproductive isolation (${\mathrm{\beta }}_{\mathrm{normalgen}.\mathrm{normaldist}}$), which did not differ between sympatry and allopatry (Table 3).…”

“…These comparisons cannot be made using PICs or Mantel tests, so in many classical studies, the effects of these factors are instead indirectly compared (Coyne & Orr, 1989, 1997; Jewell et al., 2012; Moyle et al., 2004). Using a phylogenetic mixed model framework, and classic datasets on species ability to cross with one another, I formally tested the interactions among these effects on the evolution of reproductive isolation.…”

“…(2004). Prezygotic isolation was a measure of the total number of failed pollinations (likely due to pollen pistil interactions) in interspecific crosses, compared to the reproductive isolation of the parental species.…”

“…Patterns of reproductive isolation inferred from comparative analyses of species willingness or ability to mate (i.e., prezygotic isolation) and produce viable and fertile offspring (i.e., postzygotic isolation) have generated important and influential observations about the evolution of reproductive isolation. These studies have been especially important for making generalizations about comparisons between allopatric and sympatric species pairs, or prezygotic and postzygotic barriers to reproduction (Coyne & Orr, 1989; Funk, Nosil, & Etges, 2006; Malone & Fontenot, 2008; Meiners & Winkelmann, 2012; Mendelson, 2003; Moyle, Olson, & Tiffin, 2004; Presgraves, 2002). These analyses have produced iconic patterns and rules of speciation including that reproductive isolation accumulates more quickly between sympatric species pairs compared to allopatric species pairs and, specifically, that in sympatry prezygotic reproductive isolation appears more quickly than postzygotic reproductive isolation.…”