Mycorrhizal ecology on serpentine soils

Background: Serpentine ecosystems support different, often unique, plant communities; however, we know little about the soil organisms that associate with these ecosystems. Mycorrhizas, mutualistic symbioses between fungi and roots, are critical to nutrient cycling and energy exchange below ground.

Aims: We address three hypotheses: H1, diversity of mycorrhizal fungi in serpentine soils mirrors above-ground plant diversity; H2, the morphology of mycorrhizas and fungi on serpentine soils differs from that on non-serpentine; and H3, mycorrhizal fungal communities of the same or closely related hosts differ between serpentine and non-serpentine soils.

Methods: This review focuses on whether plant diversity on serpentine soils correlates with the below ground diversity of mycorrhizal fungi.

Results: Studies show that plants and fungi formed abundant ectomycorrhizal and arbuscular mycorrhizal symbioses on and off serpentine soils. No serpentine-endemic fungi were identified. Molecular analyses indicate distinct serpentine isolates for Cenococcum geophilum and for Acaulospora, suggesting adaptation to serpentine soils. While fungal sporocarp assemblages on serpentine sites resembled those off serpentine, fruiting of hypogeous fungi was greatly reduced.

Conclusions: Ectomycorrhizal fungal communities did not differ between soil types; however, arbuscular mycorrhizal communities differed in some cases but not others. The additive response to multiple factors, described as the serpentine syndrome, may explain part of the response by fungi.     

Genetic Diversity of Bacterial Communities of Serpentine Soil and of Rhizosphere of the Nickel-Hyperaccumulator Plant <Emphasis Type="Italic">Alyssum bertolonii</Emphasis>

Serpentine soils are characterized by high levels of heavy metals (Ni, Co, Cr), and low levels of important plant nutrients (P, Ca, N). Because of these inhospitable edaphic conditions, serpentine soils are typically home to a very specialized flora including endemic species as the nickel hyperaccumulator Alyssum bertolonii. Although much is known about the serpentine flora, few researches have investigated the bacterial communities of serpentine areas. In the present study bacterial communities were sampled at various distances from A. bertolonii roots in three different serpentine areas and their genetic diversity was assessed by terminal restriction fragment length polymorphism (T-RFLP) analysis. The obtained results indicated the occurrence of a high genetic diversity and heterogeneity of the bacterial communities present in the different serpentine areas. Moreover, TRFs (terminal restriction fragments) common to all the investigated A. bertolonii rhizosphere samples were found. A new cloning strategy was applied to 27 TRFs that were sequenced and taxonomically interpreted as mainly belonging to Gram-positive and -Proteobacteria representatives. In particular, cloned TRFs which discriminated between rhizosphere and soil samples were mainly interpreted as belonging to Proteobacteria representatives.This revised version was published online in November 2004 with corrections to Volume 48.     

Bryophyte flora and vegetation of serpentine sites in the French Massif Central

Eight major serpentine sites in the French Massif Central have been bryofloristically surveyed and their specific assemblages studied by performing systematic relevés. Seventy bryophyte taxa are reported from these serpentine areas. Archidium alternifolium, Bryum gemmilucens, Cephaloziella stellulifera, Grimmia dissimulata, Racomitrium fasciculare, R. lanuginosum and Riccia subbifurca are dominant and characteristic species. Soil and rock communities harbour typical combinations of specialized and ubiquitous taxa. The occurrence of saxicolous species (Grimmia) on the ground may be a typical feature of serpentine communities. The scarcity of liverworts is underlined, as is the complexity of the controlling factors. Lack of competition and periodic submergence are probably ecological factors of prime importance. Successional pathways are certainly insufficiently known, especially as far as bryophytes are concerned. The causes of negligible succession on skeletal soils would benefit from renewed studies. Grazing of serpentine grasslands, generally considered the most relevant management tool, should be evaluated further.     

Soil calcium and plant disease in serpentine ecosystems: a test of the pathogen refuge hypothesis

Ecologists have long sought mechanistic explanations for the patterns of plant distribution and endemism associated with serpentine soils. We conducted the first empirical test of the serpentine pathogen refuge hypothesis, which posits that the low levels of calcium found in serpentine soils provide associated plants with a refuge from attack by pathogens. We measured the range of soil calcium concentrations experienced by 16 wild population of California dwarf flax (Hesperolinon californicum) and experimentally recreated part of this range in the greenhouse by soaking serpentine soils in calcium chloride solutions of varying molarity. When flax plants grown in these soils were inoculated with spores of the rust fungus Melampsora lini we found a significant negative relationship between infection rates and soil calcium concentrations. This result refutes the pathogen refuge hypothesis and suggests that serpentine plants, by virtue of their association with low calcium soils, may be highly vulnerable to attack by pathogens. This interaction between plant nutrition and disease may in part explain demographic patterns associated with serpentine plant populations and suggests scenarios for the evolution of life history traits and the distribution of genetic resistance to infection in serpentine plant communities.     

Native and Non‐Native Community Assembly through Edaphic Manipulation: Implications for Habitat Creation and Restoration

Chemical and physical (abiotic) conditions can be determining factors of community assembly and invasibility, but can this observation be used as a practical tool for habitat creation? Serpentine soils, in particular, have three abiotic components thought to confer invasion resistance: a low Ca:Mg ratio, low water‐retention capacity, and high concentrations of heavy metals. Consequently, not only do some serpentine‐adapted native plants persist only on serpentine soils, but also the community members that depend upon those plants become dependent upon serpentine as well. In an effort to provide additional habitat for the threatened and serpentine‐restricted Bay checkerspot butterfly (Euphydryas editha bayensis), we experimentally altered a non‐serpentine site to mimic the abiotic conditions of serpentine. Attempts to lower the Ca:Mg ratio of soils through the addition of MgSO₄ were unsuccessful. We then altered soil depth through the addition of gravel beds to determine the effects of water stress on native and non‐native community composition. We found that shallow soils had lower water content and correspondingly had significantly lower non‐native species richness and cover. The results present promising means, but also cautionary information, for habitat creation efforts and demonstrate the possible utility of edaphic manipulation in abating non‐native plant invasions. None of the experimental plots supported communities capable of sustaining E. editha populations, emphasizing that the manipulation of physical conditions is only likely to be successful in coordination with other restoration techniques.     

Serpentine Soils Do Not Limit Mycorrhizal Fungal Diversity

Background

Physiologically stressful environments tend to host depauperate and specialized biological communities. Serpentine soils exemplify this phenomenon by imposing well-known constraints on plants; however, their effect on other organisms is still poorly understood.

Methodology/Principal Findings

We used a combination of field and molecular approaches to test the hypothesis that serpentine fungal communities are species-poor and specialized. We conducted surveys of ectomycorrhizal fungal diversity from adjacent serpentine and non-serpentine sites, described fungal communities using nrDNA Internal Transcribed Spacer (ITS) fragment and sequence analyses, and compared their phylogenetic community structure. Although we detected low fungal overlap across the two habitats, we found serpentine soils to support rich fungal communities that include representatives from all major fungal lineages. We failed to detect the phylogenetic signature of endemic clades that would result from specialization and adaptive radiation within this habitat.

Conclusions/Significance

Our results indicate that serpentine soils do not constitute an extreme environment for ectomycorrhizal fungi, and raise important questions about the role of symbioses in edaphic tolerance and the maintenance of biodiversity.     

Serpentine and non-serpentine Silene dioica plants do not differ in nickel tolerance

Serpentine soils are hostile to plant life. They are dry, contain high concentrations of nickel and have an unfavorable calcium/magnesium ratio. The dioecious plant Silene dioica (L.) Clairv. (Caryophyllaceae) is the most common herb on serpentine soils in the Swedish mountains. It also commonly grows on non-serpentine soils in the subalpine and coastal area. I have compared the germination frequency, plant establishment and growth of serpentine and subalpine non-serpentine populations in serpentine soil under greenhouse conditions. Further more I have studied the specific effect of nickel on root and shoot growth of serpentine and non-serpentine plants from the subalpine and coastal area in solutions with different concentrations of nickel. Plants from serpentine and non-serpentine populations grew well and in a similar fashion in serpentine soil. Moreover, S. dioica plants, irrespective of original habitat, tolerated enhanced concentrations of nickel when grown in solutions. An analysis of metal content in serpentine plants from natural populations shows that S. dioica has a higher nickel concentration in the roots than in the shoots. The growth studies show that S. dioica is constitutively adapted to serpentine, and that all populations have the genetic and ecological tolerance to grow on serpentine.     

THE EFFECT OF SERPENTINE ON THE POPULATION STRUCTURE OF SILENE DIOICA (CARYOPHYLLACEAE)

Serpentine soils are rich in heavy metals and have a distinctive flora. Silene dioica is a member of the Scandinavian serpentine plant community but is also widespread outside serpentine soils. To study the population genetic consequences of serpentine stress and the origin and evolution of serpentine populations we analyzed the isozyme genetic structure of S. dioica. Seventeen populations located in the mountains of Västerbotten and Jämtland, central Sweden, were investigated by starch gel enzyme electrophoresis. About one half of the populations grow in serpentine soils and the rest on adjacent non-serpentine sites. Analyses of allele frequencies show that both serpentine and non-serpentine populations in the northern part of the studied area (Västerbotten) are genetically similar. Evidently serpentine does not exert strong selection acting upon isozyme loci. In the south (Jämtland), however, the serpentine populations exhibit genetic differentiation. This allozyme divergence is probably not due to direct selection but rather represents the effects of isolation and genetic drift. The results suggest that S. dioica has colonized serpentine repeatedly and that the tolerant populations have a multiple origin.     

Ectomycorrhizal communities of Quercus garryana are similar on serpentine and nonserpentine soils

Serpentine soils, rich in iron, magnesium, and heavy metals, select for unique plant communities and for endemic species. Because mycorrhizal fungi mediate the interaction between plants and soil, we hypothesized that distinct ectomycorrhizal fungi would colonize Quercus garryana roots on serpentine and nonserpentine soils. We sampled roots of Q. garryana on serpentine soils at two locations in the Klamath-Siskiyou Mountains of southwestern Oregon and identified ectomycorrhizas by morphological and molecular methods. The same six most abundant and most frequent mycorrhizal species, Cenococcum geophilum, Tuber candidum, Genea harknessii, Tomentella sp., Sebacina sp., and Inocybe sp., were found on serpentine and nonserpentine soils. Based on similarities calculated using the Sørensen index in Non-metric Multidimensional Scaling, mycorrhizal communities on serpentine and nonserpentine soils were not significantly different. This study showed that ectomycorrhizal species associated with Q. garryana exhibit edaphic tolerance and were neither reduced nor excluded by serpentinite or peridotite parent materials.     

Niche partitioning in arbuscular mycorrhizal communities in temperate grasslands: a lesson from adjacent serpentine and nonserpentine habitats

Arbuscular mycorrhizal fungi (AMF) represent an important soil microbial group playing a fundamental role in many terrestrial ecosystems. We explored the effects of deterministic (soil characteristics, host plant life stage, neighbouring plant communities) and stochastic processes on AMF colonization, richness and community composition in roots of Knautia arvensis (Dipsacaceae) plants from three serpentine grasslands and adjacent nonserpentine sites. Methodically, the study was based on 454‐sequencing of the ITS region of rDNA. In total, we detected 81 molecular taxonomical operational units (MOTUs) belonging to the Glomeromycota. Serpentine character of the site negatively influenced AMF root colonization, similarly as higher Fe concentration. AMF MOTUs richness linearly increased along a pH gradient from 3.5 to 5.8. Contrary, K and Cr soil concentration had a negative influence on AMF MOTUs richness. We also detected a strong relation between neighbouring plant community composition and AMF MOTUs richness. Although spatial distance between the sampled sites (c. 0.3–3 km) contributed to structuring AMF communities in K. arvensis roots, environmental parameters were key factors in this respect. In particular, the composition of AMF communities was shaped by the complex of serpentine conditions, pH and available soil Ni concentration. The composition of AMF communities was also dependent on host plant life stage (vegetative vs. generative). Our study supports the dominance of deterministic factors in structuring AMF communities in heterogeneous environment composed of an edaphic mosaic of serpentine and nonserpentine soils.     

Evidence of adaptive tolerance to nickel in isolates of Cenococcum geophilum from serpentine soils

Selection for metal-tolerant ecotypes of ectomycorrhizal (ECM) fungi has been reported in instances of metal contamination of soils as a result of human activities. However, no study has yet provided evidence that natural metalliferous soils, such as serpentine soils, can drive the evolution of metal tolerance in ECM fungi. We examined in vitro Ni tolerance in isolates of Cenococcum geophilum from serpentine and non-serpentine soils to assess whether isolates from serpentine soils exhibited patterns consistent with adaptation to elevated levels of Ni, a typical feature of serpentine. A second objective was to investigate the relationship between Ni tolerance and specific growth rates (μ) among isolates to increase our understanding of possible tolerance/growth trade-offs. Isolates from both soil types were screened for Ni tolerance by measuring biomass production in liquid media with increasing Ni concentrations, so that the effective concentration of Ni inhibiting fungal growth by 50% (EC₅₀) could be determined. Isolates of C. geophilum from serpentine soils exhibited significantly higher tolerance to Ni than non-serpentine isolates. The mean Ni EC₅₀ value for serpentine isolates (23.4 μg ml⁻¹) was approximately seven times higher than the estimated value for non-serpentine isolates (3.38 μg ml⁻¹). Although there was still a considerable variation in Ni sensitivity among the isolates, none of the serpentine isolates had EC₅₀ values for Ni within the range found for non-serpentine isolates. We found a negative correlation between EC₅₀ and μ values among isolates (r = −0.555). This trend, albeit only marginally significant (P = 0.06), indicates a potential trade-off between tolerance and growth, in agreement with selection against Ni tolerance in “normal” habitats. Overall, these results suggest that Ni tolerance arose among serpentine isolates of C. geophilum as an adaptive response to Ni exposure in serpentine soils. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Plant-by-Plant Variations of Bacterial Communities Associated with Leaves of the Nickel Hyperaccumulator Alyssum bertolonii Desv.

Bacteria associated with tissues of metal-hyperaccumulating plants are of great interest due to the multiple roles they may play with respect to plant growth and resistance to heavy metals. The variability of bacterial communities associated with plant tissues of three populations of Alyssum bertolonii, a Ni hyperaccumulator endemic of serpentine outcrops of Central Italy, was investigated. Terminal-restriction fragment length polymorphism (T-RFLP) analysis of bacterial 16S rRNA genes was applied to DNA extracted from leaf tissues of 30 individual plants from three geographically separated serpentine outcrops. Moreover, T-RFLP fingerprinting was also performed on DNA extracted from the same soils from which the plants were collected. Fifty-nine unique terminal-restriction fragments (TRFs) were identified, with more than half of the taxonomically interpreted TRFs assigned to Alpha- and Gamma-Proteobacteria and Clostridia. Data were then used to define the extent of variation of bacterial communities due to single plants or to plant populations. Results indicated a very high plant-by-plant variation of leaf-associated community (more than 93% of total variance observed). However, a core (numerically small) of plant-specific TRFs was found. This work demonstrates that plant-associated bacterial communities represent a large reservoir of biodiversity and that the high variability existing between plants, even from the same population, should be taken into account in future studies on association between bacteria and metal-hyperaccumulating plants.     

Comparison of ectomycorrhizas of Quercus garryana (Fagaceae) on serpentine and non-serpentine soils in southwestern Oregon

The diversity of ectomycorrhizal communities associated with Quercus garryana on and off serpentine soils was compared and related to landscape-level diversity. Serpentine soils are high in magnesium, iron, and heavy metals and low in fertility. In plant communities on serpentine soils, a high proportion of flowering plant species are endemic. At three sites with paired serpentine and nonserpentine soils in southwestern Oregon, we sampled Q. garryana roots and categorized ectomycorrhizas by morphotyping and by restriction fragment length patterns. Ectomycorrhizas were abundant at all sites; no single fungal species dominated in the ectomycorrhizas. Of 74 fungal species characterized by morphotype and pattern of restriction fragment length polymorphisms, 46 occurred on serpentine soils, and 32 were unique to serpentine soil. These species are potentially endemic to serpentine soil. Similarities in species composition between paired serpentine and nonserpentine soils were not significantly lower than among three serpentine sites or among three nonserpentine sites. We conclude that mycorrhizal communities associated with oaks on serpentine soil do not differ in species richness or species evenness from those on neighboring nonserpentine soil.     

Plant communities on infertile soils are less sensitive to climate change

Background and Aims Much evidence suggests that plant communities on infertile soils are relatively insensitive to increased water deficit caused by increasing temperature and/or decreasing precipitation. However, a multi-decadal study of community change in the western USA does not support this conclusion. This paper tests explanations related to macroclimatic differences, overstorey effects on microclimate, variation in soil texture and plant functional traits.Methods A re-analysis was undertaken of the changes in the multi-decadal study, which concerned forest understorey communities on infertile (serpentine) and fertile soils in an aridifying climate (southern Oregan) from 1949–1951 to 2007–2008. Macroclimatic variables, overstorey cover and soil texture were used as new covariates. As an alternative measure of climate-related change, the community mean value of specific leaf area was used, a functional trait measuring drought tolerance. We investigated whether these revised analyses supported the prediction of lesser sensitivity to climate change in understorey communities on infertile serpentine soils.Key Results Overstorey cover, but not macroclimate or soil texture, was a significant covariate of community change over time. It strongly buffered understorey temperatures, was correlated with less change and averaged >50 % lower on serpentine soils, thereby counteracting the lower climate sensitivity of understorey herbs on these soils. Community mean specific leaf area showed the predicted pattern of less change over time in serpentine than non-serpentine communities.Conclusions Based on the current balance of evidence, plant communities on infertile serpentine soils are less sensitive to changes in the climatic water balance than communities on more fertile soils. However, this advantage may in some cases be lessened by their sparser overstorey cover.     

Serpentine soils promote ectomycorrhizal fungal diversity

Serpentine soils impose physiological stresses that limit plant establishment and diversity. The degree to which serpentine soils entail constraints on other organisms is, however, poorly understood. Here, I investigate the effect of serpentine soils on ectomycorrhizal (ECM) fungi by conducting a reciprocal transplant experiment, where serpentine and nonserpentine ECM fungal communities were cultured in both their native and non-native soils. Contrary to expectation, serpentine soils hosted higher fungal richness compared to nonserpentine, and most species were recovered from serpentine soil, suggesting ECM fungi are not overall specialized or strongly affected by serpentine edaphic constraints.     

Patterns of seed dispersal syndromes on serpentine soils: examining the roles of habitat patchiness,soil infertility and correlated functional traits

Background: It is critical to understand the ecological factors shaping seed dispersal in plant communities in order to predict their fate in the face of global change. Communities restricted to patchy habitats may contain more species with ‘directed’ dispersal syndromes that facilitate successful seed dispersal to other patches; however, habitat quality may constrain the presence of and efficiency of dispersal syndromes found within those habitats.

Aims: The aim of this study was to hypothesise that if habitat patchiness is an important filter on dispersal syndromes, ‘directed’ vertebrate dispersal should be more prevalent in serpentine habitats because of their patchiness. Alternatively, if habitat quality is more important, wind dispersal should be more prevalent in serpentine habitats because of their low fertility.

Methods: Using three datasets representing grassland, chaparral and forest vegetation types, we analysed differences in the composition of dispersal syndromes (vertebrate, wind, passive, water and ant) between communities on patchy infertile serpentine soils and on continuous, fertile non-serpentine soils. Our analyses also accounted for correlated functional traits and phylogenetic relatedness.

Results: Across and within all three vegetation types, serpentine communities had significantly higher proportions of wind dispersed and lower proportions of vertebrate-dispersed species. These patterns were not independent of functional traits. Proportions of the other dispersal syndromes did not differ.

Conclusions: Our results suggest that on low-fertility soils, habitat quality may outweigh habitat patchiness as a filter on the availability of dispersal syndromes, potentially adding to the vulnerability of such communities to stochastic extinctions and global change.     

Two Invasive Plants Alter Soil Microbial Community Composition in Serpentine Grasslands

Plant invasions pose a serious threat to native ecosystem structure and function. However, little is known about the potential role that rhizosphere soil microbial communities play in facilitating or resisting the spread of invasive species into native plant communities. The objective of this study was to compare the microbial communities of invasive and native plant rhizospheres in serpentine soils. We compared rhizosphere microbial communities, of two invasive species, Centaurea solstitialis (yellow starthistle) and Aegilops triuncialis (barb goatgrass), with those of five native species that may be competitively affected by these invasive species in the field (Lotus wrangelianus, Hemizonia congesta, Holocarpha virgata, Plantago erecta, and Lasthenia californica). Phospholipid fatty acid analysis (PLFA) was used to compare the rhizosphere microbial communities of invasive and native plants. Correspondence analyses (CA) of PLFA data indicated that despite yearly variation, both starthistle and goatgrass appear to change microbial communities in areas they invade, and that invaded and native microbial communities significantly differ. Additionally, rhizosphere microbial communities in newly invaded areas are more similar to the original native soil communities than are microbial communities in areas that have been invaded for several years. Compared to native plant rhizospheres, starthistle and goatgrass rhizospheres have higher levels of PLFA biomarkers for sulfate reducing bacteria, and goatgrass rhizospheres have higher fatty acid diversity and higher levels of biomarkers for sulfur-oxidizing bacteria, and arbuscular mycorrhizal fungi. Changes in soil microbial community composition induced by plant invasion may affect native plant fitness and/or ecosystem function.     

Edaphic influences of ophiolitic substrates on vegetation in the Western Italian Alps

Background and aims

Soils derived from serpentinite (serpentine soils) often have low macronutrient concentrations, exceedingly low Ca:Mg molar ratios and high heavy metal concentrations, typically resulting in sparse vegetative cover. This combined suite of edaphic stresses is referred to as the “serpentine syndrome.” Although several plant community-level studies have been conducted to identify the most important edaphic factor limiting plant growth on serpentine, the primary factor identified has often varied by plant community and local climate. Few studies to date have been conducted in serpentine plant communities of alpine or boreal climates. The goal of our study was to determine the primary limiting edaphic factors on plant community species composition and productivity (cover) in the alpine and boreal climate of the Western Alps, Italy.

Methods

Soil properties and vegetation composition were analyzed for several sites underlain by serpentinite, gabbro, and calc-schist substrates and correlated using direct and indirect statistical methods.

Results

Boreal forest soils were well-developed and tended to have low pH throughout the soil profile resulting in high Ni availability. Alpine soils, in comparison, were less developed. The distinct serpentine plant communities of the Western Alps are most strongly correlated with high levels of bioavailable Ni associated with low soil pH. Other factors such as macronutrient deficiency, low Ca:Mg molar ratio and drought appear to be less important.

Conclusions

The strong ecological influence of Ni is caused by environmental conditions which increase metal mobilization.     

Local adaptation to serpentine soils in Pinus ponderosa

Local adaptation to serpentine soils is studied using both transplant experiments and molecular genetic techniques. In long-lived species, such as pines, it is unclear how soon after germination local adaptation becomes detectable. Here I present results of a 36-year reciprocal transplant experiment using Pinus ponderosa, along with allozyme analyses from the same trees. Using a repeated measures analysis of variance, there is evidence for adaptation to serpentine soils; however, significant differences between source soil types do not become apparent until 20 years after the start of the experiment. Analysis of allozyme data showed no evidence for differentiation between the serpentine and non-serpentine populations. Comparing the performance of families over the course of the experiment found that there was little correlation between performance after 1 or 4 years of growth in the field and performance after 36 years. This suggests that short-term transplant experiments may not provide definitive evidence for adaptation to serpentine soils. A literature survey of all transplant studies using pine species growing on and off of serpentine soils found that studies that lasted fewer than 2 years showed no evidence for adaptation. However, in the two experiments (this one included) that lasted more than 2 years, both showed evidence for adaptation to serpentine soils. More long-term experiments are required to validate these results.     

Within- and trans-generational plasticity affects the opportunity for selection in barbed goatgrass (Aegilops triuncialis)

• Premise of the study: Environments are composed of selective agents, and environments may also modify the efficacy of these agents. Environments affect the rate of maximum evolutionary change by influencing variation in relative fitness (i.e., the opportunity for selection, or I). Within- and transgenerational plastic environmental responses may affect I, speeding or slowing processes of local adaptation. • Methods: We determined whether environmental factors affected the opportunity for selection (I) in Aegilops triuncialis (barbed goatgrass) by measuring I as a within- and transgenerational plastic response to two maternal glasshouse environments (serpentine/dry and loam/moist). We also determined whether this species’ two most common genetic lineages (determined by DNA microsatellite length polymorphism) varied in response to glasshouse treatments. • Key Results: Opportunity for selection was less for plants grown in the dry serpentine environment than for plants grown in the moist loam environment. This response varied between genetic lineages. The east lineage exhibited a within-generation response to the dry serpentine environment. For both seed mass and average seed weight in this lineage, the opportunity for selection was lower in dry serpentine than in moist loam. The west lineage had a transgenerational response to the dry serpentine such that the opportunity for selection for seed number and seed mass was lower for plants produced by mothers grown in dry serpentine than for plants produced by mothers in moist loam. • Conclusions: Phenotypic variation in relative fitness is constrained by the dry serpentine environment, which leads to lower evolvability in this environment. Within- and transgenerational effects of the environment may slow local adaptation to serpentine soils.