Maintaining their genetic distance: Little evidence for introgression between widely hybridizing species of Geum with contrasting mating systems.

Within the plant kingdom, many genera contain sister lineages with contrasting outcrossing and inbreeding mating systems that are known to hybridize. The evolutionary fate of these sister lineages is likely to be influenced by the extent to which they exchange genes. We measured gene flow between outcrossing Geum rivale and selfing Geum urbanum, sister species that hybridize in contemporary populations. We generated and used a draft genome of G. urbanum to develop dd-RAD data scorable in both species. Coalescent analysis of RAD data from allopatric populations indicated that the species diverged 2-3 Mya, and that historical gene flow between them was extremely low (1 migrant every 25 generations). Comparison of genetic divergence between species in sympatry and allopatry, together with an analysis of allele frequencies in potential parental and hybrid populations, provided no evidence of contemporary introgression in sympatric populations. Cluster- and species-specific marker analyses revealed that, apart from four early-generation hybrids, individuals in sympatric populations fell into two genetically distinct groups that corresponded exactly to their morphological species classification with maximum individual admixture estimates of only 1-3%. However, we did observe joint segregation of four putatively introgressed SNPs across two scaffolds in the G. urbanum population that was associated with significant morphological variation, interpreted as tentative evidence for rare, recent interspecific gene flow. Overall, our results indicate that despite the presence of hybrids in contemporary populations, genetic exchange between G. rivale and G. urbanum has been extremely limited throughout their evolutionary history.

Patterns of mating, generation of diversity, and fitness of offspring in a Geum hybrid swarm.

To understand the evolutionary consequences of hybridization between the outcrossing plant Geum rivale (Rosaceae) and the selfer Geum urbanum, we tested the predictions of two simple models that assume either (A) low or (B) high pollen fitness in hybrids. Model A predicts only four genotypic classes (G. rivale, G. rivale backcross [BC(R)], F1, and Geum urbanum) and asymmetric introgression from inbreeding to outbreeding species. Model B predicts additional genotypic classes and potential generation of novel inbreeding lines in the hybrid swarm. Amplified fragment length polymorphism (AFLP) analysis of adults revealed only the four genotypes predicted by model A. However, microsatellite analysis of parent-progeny arrays demonstrated production of selfed offspring by F1 and BC(R) maternal parents and contribution of these genotypes to outcross pollen pools, as predicted by model B. Moreover, AFLP and morphological analysis showed that the offspring generation comprised genotypes and phenotypes covering the entire spectrum of variation between the two parental species, in line with model B. A common garden experiment indicated no systematic reduction in fitness of offspring derived from hybrid parents. The genetic structure of the adults in the Geum hybrid swarm cannot be explained by restricted mating patterns but may result from ecological selection acting on a diverse offspring population.

Gas exchange, growth, and defense responses of invasive Alliaria petiolata (Brassicaceae) and native Geum vernum (Rosaceae) to elevated atmospheric CO2 and warm spring temperatures.

PREMISE OF STUDY: Global increases in atmospheric CO2 and temperature may interact in complex ways to influence plant physiology and growth, particularly for species that grow in cool, early spring conditions in temperate forests. Plant species may also vary in their responses to environmental changes; fast-growing invasives may be more responsive to rising CO2 than natives and may increase production of allelopathic compounds under these conditions, altering species' competitive interactions. METHODS: We examined growth and physiological responses of Alliaria petiolata, an allelopathic, invasive herb, and Geum vernum, a co-occurring native herb, to ambient and elevated spring temperatures and atmospheric CO2 conditions in a factorial growth chamber experiment. KEY RESULTS: At 5 wk, leaves were larger at high temperature, and shoot biomass increased under elevated CO2 only at high temperature in both species. As temperatures gradually warmed to simulate seasonal progression, G. vernum became responsive to CO2 at both temperatures, whereas A. petiolata continued to respond to elevated CO2 only at high temperature. Elevated CO2 increased thickness and decreased nitrogen concentrations in leaves of both species. Alliaria petiolata showed photosynthetic downregulation at elevated CO2, whereas G. vernum photosynthesis increased at elevated temperature. Flavonoid and cyanide concentrations decreased significantly in A. petiolata leaves in the elevated CO2 and temperature treatment. Total glucosinolate concentrations and trypsin inhibitor activities did not vary among treatments. CONCLUSIONS: Future elevated spring temperatures and CO2 will interact to stimulate growth for A. petiolata and G. vernum, but there may be reduced allelochemical effects in A. petiolata.

The consequences of mating over a range of parental genetic similarity in a selfing allopolyploid plant species.

In diploids, F(1) offspring performance is expected to increase with increasing genetic dissimilarity between the parents until an optimum is reached because outbreeding mitigates inbreeding depression and maximizes heterosis. However, many flowering plant species are derived through allopolyploidization, i.e. interspecific hybridization with genome doubling. This mode of plant speciation can be expected to considerably alter the consequences of inbreeding and outbreeding. We investigated the F1 fitness consequences of mating over a range of (genetic) distances in the allohexaploid plant species Geum urbanum. Offspring was raised under controlled conditions (632 plants). The performance of outcrossed progeny was not significantly better than that of their selfed half-siblings and did not increase with parental genetic dissimilarity (0-0.83). Our findings support low, if any, inbreeding depression and heterosis. We attribute this to the peculiar state of quasi-permanent heterozygosity in allopolyploids and frequent selfing.

Phenotypic variation in nurse traits and community feedbacks define an alpine community.

Much is known about facilitation, but virtually nothing about the underlying genetic and evolutionary consequences of this important interaction. We assessed the potential of phenotypic differences in facilitative effects of a foundation species to determine the composition of an Alpine community in Arizona. Two phenotypes of Geum rossii occur along a gradient of disturbance, with 'tight' competitive cushions in stable conditions and 'loose' facilitative cushions in disturbed conditions. A common-garden study suggested that field-based traits may have a genetic basis. Field experiments showed that the reproductive fitness of G. rossii cushions decreased with increasing facilitation. Finally, using a dual-lattice model we showed that including the cost and benefit of facilitation may contribute to the co-occurrence of genotypes with contrasting facilitative effects. Our results indicate that changes in community composition due to phenotypic differences in facilitative effects of a foundation species may in turn affect selective pressures on the foundation species.

Contrasting effects of hemiparasites on ecosystem processes: can positive litter effects offset the negative effects of parasitism?

Hemiparasites are known to influence community structure and ecosystem functioning, but the underlying mechanisms are not well studied. Variation in the impacts of hemiparasites on diversity and production could be due to the difference in the relative strength of two interacting pathways: direct negative effects of parasitism and positive effects on N availability via litter. Strong effects of parasitism should result in substantial changes in diversity and declines in productivity. Conversely, strong litter effects should result in minor changes in diversity and increased productivity. We conducted field-based surveys to determine the association of Castilleja occidentalis with diversity and productivity in the alpine tundra. To examine litter effects, we compared the decomposition of Castilleja litter with litter of four other abundant plant species, and examined the decomposition of those four species when mixed with Castilleja. Castilleja was associated with minor changes in diversity but almost a twofold increase in productivity and greater foliar N in co-occurring species. Our decomposition trials suggest litter effects are due to both the rapid N loss of Castilleja litter and the effects of mixing Castilleja litter with co-occurring species. Castilleja produces litter that accelerates decomposition in the alpine tundra, which could accelerate the slow N cycle and boost productivity. We speculate that these positive effects of litter outweigh the effects of parasitism in nutrient-poor systems with long-lived hemiparasites. Determining the relative importance of parasitism and litter effects of this functional group is crucial to understand the strong but variable roles hemiparasites play in affecting community structure and ecosystem processes.

Novel weapons: invasive plant suppresses fungal mutualists in America but not in its native Europe.

Why some invasive plant species transmogrify from weak competitors at home to strong competitors abroad remains one of the most elusive questions in ecology. Some evidence suggests that disproportionately high densities of some invaders are due to the release of biochemicals that are novel, and therefore harmful, to naive organisms in their new range. So far, such evidence has been restricted to the direct phytotoxic effects of plants on other plants. Here we found that one of North America's most aggressive invaders of undisturbed forest understories, Alliaria petiolata (garlic mustard) and a plant that inhibits mycorrhizal fungal mutualists of North American native plants, has far stronger inhibitory effects on mycorrhizas in invaded North American soils than on mycorrhizas in European soils where A. petiolata is native. This antifungal effect appears to be due to specific flavonoid fractions in A. petiolata extracts. Furthermore, we found that suppression of North American mycorrhizal fungi by A. petiolata corresponds with severe inhibition of North American plant species that rely on these fungi, whereas congeneric European plants are weakly affected. These results indicate that phytochemicals, benign to resistant mycorrhizal symbionts in the home range, may be lethal to naïve native mutualists in the introduced range and indirectly suppress the plants that rely on them.

On net reproductive rate and the timing of reproductive output.

Abstract: Understanding the relationship between life-history patterns and population growth is central to demographic studies. Here we derive a new method for calculating the timing of reproductive output, from which the generation time and its variance can also be calculated. The method is based on the explicit computation of the net reproductive rate (R0) using a new graphical approach. Using nodding thistle, desert tortoise, creeping aven, and cat's ear as examples, we show how R0 and the timing of reproduction is calculated and interpreted, even in cases with complex life cycles. We show that the explicit R0 formula allows us to explore the effect of all reproductive pathways in the life cycle, something that cannot be done with traditional analysis of the population growth rate (lambda). Additionally, we compare a recently published method for determining population persistence conditions with the condition R0 > 1 and show how the latter is simpler and more easily interpreted biologically. Using our calculation of the timing of reproductive output, we illustrate how this demographic measure can be used to understand the effects of life-history traits on population growth and control.

Landscape genetics of the self-compatible forest herb Geum urbanum: effects of habitat age, fragmentation and local environment.

To investigate the role of habitat fragmentation, fragment age and local environment in shaping the genetics of plant populations, we examined the genetic structure of the self-compatible forest herb Geum urbanum using microsatellite markers. A historical land-use reconstruction assigned the studied populations to two age classes: populations in primary forest fragments, and populations in secondary fragments. Local environmental conditions were quantified on the basis of the herb-layer community composition. A stepwise general linear model revealed that levels of within-population genetic diversity were best explained by population size, landscape connectivity and the interaction between both. Connectivity was positively correlated with the genetic diversity of small populations, but did not significantly affect the diversity of large populations. Contrary to what we expected, secondary-forest populations showed lower divergence relative to populations located in primary patches. Small populations were genetically more diverged compared to large populations. Mantel tests showed no significant isolation by distance and no significant correlation between habitat similarity and genetic differentiation. We conclude that gene flow has probably prevented founder events from being reflected in the present genetic structure of G. urbanum. Gene flow towards low-connectivity populations, however, seemed to be insufficient to counteract the effects of drift in small populations.