Lessons on Evolution from the Study of Edaphic Specialization

Abstract

Plants adapted to special soil types are ideal for investigating evolutionary processes, including maintenance of intraspecific variation, adaptation, reproductive isolation, ecotypic differentiation, and the tempo and mode of speciation. Common garden and reciprocal transplant approaches show that both local adaptation and phenotypic plasticity contribute to edaphic (soil-related) specialization. Edaphic specialists evolve rapidly and repeatedly in some lineages, offering opportunities to investigate parallel evolution, a process less commonly documented in plants than in animals. Adaptations to soil features are often under the control of major genes and they frequently have direct or indirect effects on genes that contribute to reproductive isolation. Both reduced competitiveness and greater susceptibility to herbivory have been documented among some edaphic specialists when grown in ‘normal’ soils, suggesting that a high physiological cost of tolerance may result in strong divergent selection across soil boundaries. Interactions with microbes, herbivores, and pollinators influence soil specialization either by directly enhancing tolerance to extremes in soil conditions or by reducing gene flow between divergent populations. Climate change may further restrict the distribution of edaphic specialists due to increased competition from other taxa or, expand their ranges, if preadaptations to drought or other abiotic stressors render them more competitive under a novel climate.

Full text

Lessons on Evolution from the Study
of Edaphic Specialization
Nishanta Rajakaruna
1,2,3
1
Biological Sciences Department, California Polytechnic State University, San Luis Obispo, CA 93407, USA
2
Unit for Environmental Sciences and Management, North-West University, Private Bag X6001,
Potchefstroom 2520, South Africa
3
Author for Correspondence; e-mail: nrajakar@calpoly.edu
#The New York Botanical Garden 2017
Abstract Plants adapted to special soil types are ideal for investigating evolutionary
processes, including maintenance of intraspecific variation, adaptation, reproductive isola-
tion, ecotypic differentiation, and the tempo and mode of speciation. Common garden and
reciprocal transplant approaches show that both local adaptation and phenotypic plasticity
contribute to edaphic (soil-related) specialization. Edaphic specialists evolve rapidly and
repeatedly in some lineages, offering opportunities to investigate parallel evolution, a
process less commonly documented in plants than in animals. Adaptations to soil features
are often under the control of major genes and they frequently have direct or indirect effects
on genes that contribute to reproductive isolation. Both reduced competitiveness and greater
susceptibility to herbivory have been documented among some edaphic specialists when
grown in ‘normal’soils, suggesting that a high physiological cost of tolerance may result in
strong divergent selection across soil boundaries. Interactions with microbes, herbivores, and
pollinators influence soil specialization either by directly enhancing tolerance to extremes in
soil conditions or by reducing gene flow between divergent populations. Climate change
may further restrict the distribution of edaphic specialists due to increased competition from
other taxa or, expand their ranges, if preadaptations to drought or other abiotic stressors
render them more competitive under a novel climate.
Keywords Ecological Speciation .Edaphic Endemism .Harsh Environments .Cost of
Tol era nce .Serpentine .Metal Tolerance .Parallel Speciation .Geobotany.Plant-Soil
Relations .Local Adaptation
BThe red-rock forest may seem hellish to us, but it is a refuge to its flora...it is the
obdurate physical (and chemical) adversity of things such as peridotite bedrock
which often drives life to its most surprising transformations^(Wallace, 1983)
Introduction
Over a century and a half after Darwin’sOn the Origin of Species, there is little doubt
that ecological adaptation plays a crucial role in the diversification of species (Nosil,
Bot. Rev.
https://doi.org/10.1007/s12229-017-9193-2

2012). The study of ecological speciation marks a central theme in evolutionary
biology, with a major focus on understanding the role of adaptation in reproductive
isolation (Baack et al., 2015) and species divergence (Schemske, 2010).
Plants found on ‘extreme’soils, those characterized by unusual chemical (pH, ionic
strength, nutrients, or heavy metals) or physical conditions (soil moisture, temperature,
texture, structure, or depth), provide model systems to examine the role of edaphic (soil-
related) adaptation in ecological speciation (Rajakaruna, 2004; Kay et al., 2011)and
adaptive radiation (Ellis & Weis, 2005;Paunetal.,2016;Shimizu-Inatsugietal.,2016).
The landmark papers BEvolution in closely adjacent plant populations^(Antonovics,
2006 and references therein) focused on the evolution of metal tolerance in the grasses
Anthoxanthum odoratum and Agrostis tenuis on mine tailings in Europe. These classic
papers, along with early studies of ecotypes on western North American serpentine soils
(Anacker, 2014 and references therein), illustrated how strong selection can maintain
distinct sub-populations in the face of gene flow, highlighting three key lessons from
edaphic specialization pertinent to evolution: first, edaphic transitions often have direct or
indirect effects on reproductive isolation between ecologically-divergent populations
(Christie & Macnair, 1987;Wrightetal.,2013), suggesting a link between adaptation
and speciation; second, some populations can evolve (or lose) tolerance to edaphic
conditions rapidly and with relative ease, suggesting that major gene loci (i.e., loci with
large phenotypic effects) could be driving adaptive evolution (Macnair, 1983; Bratteler
et al., 2006,2006; Lowry et al., 2012); and third, populations within some species can
independently evolve tolerance to edaphic conditions (i.e., tolerance evolves multiple
times within a species), suggesting a tendency towards parallel evolution (Rajakaruna &
Whitton, 2004; Ostevik et al., 2012;Rodaetal.,2013). Edaphic specialists, especially
plants growing in ultramafic (Harrison & Rajakaruna, 2011;Anacker,2014), gypsum
(Moore et al., 2014; Escudero et al., 2015; Bolukbasi et al., 2016), mine tailing (Ernst,
2006), seleniferous (Schiavon & Pilon-Smits, 2017), and saline (Saslis-Lagoudakis et al.,
2014; Cheeseman, 2015) habitats, are now at the forefront of plant evolutionary research,
providing model systems for exploring factors and mechanisms driving evolutionary
processes from the species to community level (Cheplick, 2015).
In this review, I highlight studies of edaphic specialists from serpentine outcrops,
saline soils, gypsum and other carbonate deposits, and heavy metal-rich mine tailings,
and explore such questions as:
&How do edaphic endemics evolve (i.e., what are the major pathways to edaphic
specialization)?
&Are edaphic endemics new or old species?
&How does adaptation contribute to reproductive isolation?
&Is the transition to edaphic endemism directional, and are edaphic endemics
evolutionary dead-ends?
&Does parallel evolution of traits conferring adaptations and reproductive isolation
occur frequently in edaphic specialists?
&How does tolerance to edaphic extremes influence competitiveness in ‘normal’
soils, i.e., what is the cost of tolerance?
&Do interactions with herbivores or microbes drive the evolution of edaphic
specialization?
&Are edaphic specialists more or less sensitive to climate change?
N. Rajakaruna

&What is the genetic basis of edaphic specialization and speciation? Is there evidence
for horizontal gene flow conferring edaphic adaptation or for genetic similarity
among the mechanisms used by different species under any given edaphic stress?
Finally, I outline directions for future research on the use of edaphic specialists as
model systems to study evolutionary processes across cellular, organismic, population,
and community levels.
How do Edaphic Endemics Evolve?
Colonization of new environments, including edaphic islands, can be achieved by
ecotypic differentiation via local adaptation or by phenotypic plasticity (Hereford,
2009; Palacio-López et al., 2015). Early studies in Sweden (Turesson, 1922) and
California (Clausen et al., 1940) demonstrated that local adaptation is common in
species distributed across altitudinal and climatic gradients. Similarly, elegant transplant
experiments conducted by the British Ecological Society from the 1920s to the 1930s
highlighted edaphic factors as potent agents of selection (Marsden-Jones & Turrill,
1938). The long-term Park Grass Experiment in Rothamsted, UK (Silvertown et al.,
2006), in which the grass Anthoxanthum odoratum was subjected to different fertilizer
and lime treatments across multiple plots, showed local adaptation to soil types over a
relatively short period. These studies exemplified how experimental methods such as
common garden and reciprocal transplant experimentation can be used to examine how
phenotypic plasticity and local adaptation contribute to habitat specialization (Wright &
Stanton, 2011).
Species living in harsh environments, including on edaphic islands, have genotypes
either broadly tolerant to wide-ranging environmental conditions (Bradshaw, 2006)or
locally adapted to specific ecological factors characteristic of the ‘home’environment,
i.e., ecotypes (Lowry, 2012). Ecotypes are a critical stage in the speciation process, and
the recognition of such populations is vital for evolutionary studies (Via, 2009). The
existence of sharp habitat boundaries between ecotypes is a compelling phenomenon
that has provided some of the best experimental evidence of natural selection in the
wild (Rajakaruna & Whitton, 2004). Although there is much support for ecotypic
differentiation and species-level endemism among flowering plants found on edaphic
islands (O’Dell & Rajakaruna, 2011), such edaphic specialization is not as well-
documented among ferns, gymnosperms, mycorrhizal fungi, lichens, or bryophytes
(Rajakaruna et al., 2014).
It is important to examine factors and mechanisms driving rapid evolution in
response to environmental pressures, such as those influencing plants on edaphic
islands. Still, it is equally critical to pay attention to ‘genostasis,’in which the lack of
sufficient genetic variation may result in evolutionary failure (Bradshaw, 1991). Early
work on mine tailings (Antonovics, 2006 andreferencestherein)providesanexample:
at edges of mine tailings, some species maintain heavy metal-tolerant and -intolerant
genotypes located only a few centimeters apart. In many other adjacent species, such
edaphic differentiation is not found. Most species found off of mine tailings are unable
to colonize metal-contaminated soil because they do not have the necessary alleles for
adaptive variation (Antonovics, 1976). Additionally, a meta-analysis of studies of local
Lessons on evolution from the study of edaphic specialization

adaptation showed that only 45% of population pairs (out of 1,032 compared) had
better performance of the ‘home’population relative to the ‘away’population at both
sites (Leimu & Fischer, 2008). This suggests that habitat-mediated selection may not be
as common in plants as generally assumed. Lack of local adaptation may result from
reasons other than the lack of necessary genetic variance (Bell, 2012; Cheplick, 2015),
including insufficient time for adaptive evolution, a high degree of phenotypic plastic-
ity, extensive gene flow, and random genetic drift due to founder events. Further, lack
of local adaptation may be a statistical/methodological artifact of choosing study
habitats that are too similar in their major selection pressures (Cheplick, 2015).
Landmark papers on factors and mechanisms driving habitat specialization
(Mason, 1946a,b; Raven, 1964) provided an early framework for understanding
the role played by soil-based selection in the origin of new species. These ideas
offered a context to envision a set of stages that may lead to the establishment of an
edaphically endemic species, from the initial tolerance of an edaphic condition by
preadapted genotypes to clear-cut species formation (Kruckeberg, 1986). While this
sequence represents a relatively precise picture of the evolutionary pathways to
endemism, it does not indicate why some genotypes evolve through all the steps
and others do not. It is possible that there is a relatively higher cost of broad
tolerance in edaphic endemic species relative to the cost of tolerance found in
widespread taxa with broad edaphic tolerance (Kazakou et al., 2008;Maestri
et al., 2010). Thus, only lineages experiencing a high cost of broad tolerance will
be subjected to strong divergent selection across an edaphic boundary, leading to
evolution of ecotypes and edaphic endemic species. The decisive step
distinguishing an ecotype from an endemic species is likely the acquisition of
complete reproductive isolation between the ancestral population and the ecotype,
allowing an independent gene pool to evolve (Macnair & Gardner, 1998;Lowry,
2012). How this occurs, especially when populations are not geographically isolat-
ed, is a fundamental question in speciation research (Kay et al., 2011;O’Dell &
Rajakaruna, 2011).
Populations that become geographically isolated can gradually acquire genetic dif-
ferences through selection and drift, and eventually become reproductively isolated
through the process of allopatric speciation. Past climate fluctuations and other dramatic
environmental changes may have led to the extinction of widespread populations due to
massive vegetation shifts such as those that occurred during postglacial reforestation
(Slovák et al., 2012). Those surviving relict populations, especially those isolated on
edaphic islands, may have become differentiated by means of allopatric speciation.
Further, chance dispersal into novel edaphic habitats, particularly during strong climatic
oscillations of the Quaternary, would have also provided conditions of relaxed compe-
tition, facilitating the evolution of edaphic specialization (Dillenberger & Kadereit,
2013). Thus, allopatric speciation may be a particularly important pathway for the origin
of edaphically restricted paleoendemic species (Table 1). Abrupt and sharply delineated
edaphic habitats often provide opportunities for the divergence of edaphic specialist
species along contact zones through the process of parapatric speciation (Table 1). There
are major challenges, however, in demonstrating a parapatric origin for a species, as
speciation must occur without a geographic barrier to gene flow. First, it is possible that
N. Rajakaruna

Tab l e 1 Proposed modes of speciation and substrate type for some taxa having presumably diversified under
edaphic selection
Species Mode of Speciation Type of Substrate Reference
Adiantum viridimontanum Hybrid, Polyploid Serpentine Paris & Windham , 1988;
Paris, 1991
Arabidopsis spp. Hybrid Serpentine Arnold et al., 2016
Arctostaphylos spp. Hybrid Serpentine Gottlieb, 1968
Armeria spp. Hybrid Serpentine Feliner et al., 1996
Caulanthus amplexicaulis
var. barbarae
Allopatric Serpentine Pepper & Norwood, 2001
Galium spp. Hybrid Serpentine Krahulcová & Štěpánková,
1998
Geonoma macrostachys Sympatric Ter r a f i rme and moist
flood plain
Borchsenius et al., 2016
Harmonia spp. Allopatric Serpentine Baldwin, 1999
Helianthus spp. Hybrid, Polyploid,
Peripatric
Desert, Sand Dune,
Salt Marshes
Lai et al., 2005; Yakimowski
& Rieseberg, 2014 and
references therein
Hesperolinon spp. Peripatric Serpentine Schneider et al., 2016
Howea spp.Sympatric Volcanic and carbonate Savolainen et al., 2006
Iris spp. Hybrid Flooded soils Martin et al., 2006
Knautia arvensis Allopatric Serpentine Kaplan, 1998
Knautia spp. Hybrid/Polyploid Serpentine Kolářet al., 2014
Lasthenia californica
complex
Allopatric Ionically-harsh soils
such as saline,
serpentine, etc.
Rajakaruna, 2003;
Kay et al., 2011
Layia discodea Parapatric,
Peripatric
Serpentine Baldwin, 2005
Leptosiphon parviflorus Parapatric Serpentine Kay et al., 2011
Mentzelia monoensis Peripatric Silica-rich rhyolite tephra Brokaw et al., 2015
Metal-tolerant grasses Peripatric Metal mines Antonovics, 2006
Mimulus spp. Peripatric Saline, Serpentine,
Dry/Wet
Grossenbacher et al., 2014;
Selby et al., 2014;Sobel,
2014
Minuartia laricifolia
subsp. ophiolitica
Allopatric Serpentine Moore et al., 2013
Potentilla spp. Hybrid Serpentine Paule et al., 2015
Protium spp. Parapatric White-Sand Fine et al., 2013 and
references therein
Quercus spp. Hybrid Serpentine Forde & Faris, 1962
Silene dioica Allopatric Serpentine Westerbergh & Saura, 1992
Streptanthus glandulosus
complex
Allopatric Serpentine Mayer & Soltis, 1994,1999
Lessons on evolution from the study of edaphic specialization

the initial evolution of traits driving reproductive isolation may have occurred under
allopatric conditions (Pettengill & Moeller, 2012); a new species that evolved under
allopatry may subsequently come into contact along an edaphic boundary and appear
identical to parapatric species in being relatively young, ecologically divergent, and
reproductively isolated. Further, reinforcement of reproductive isolation (Pfennig &
Pfennig, 2009), a process by which selection can minimize maladaptive hybridization,
may take place upon contact with those populations that initially diverged under
allopatry, thereby completing the speciation process. Genetic analyses can help distin-
guish whether divergence first occurred in allopatry and speciation was completed by
reinforcement (Widmer et al., 2009), although extensive introgression and hybridization
following secondary contact may make such determinations difficult (Barraclough &
Vo g l e r, 2000; Ortiz-Barrientos et al., 2009; Strasburg et al., 2012). How speciation can
proceed in the face of gene flow between potentially interbreeding populations, includ-
ing among parents and newly formed hybrid species (Andrew et al., 2012; Renaut et al.,
2013), is especially revealing of the challenges associated with habitat specialization
under parapatry. Gene flow can dilute the adaptive differences that selection generates
between divergent populations; this may be especially problematic when one population
is significantly larger than the other, as in the case of a newly divergent population found
adjacent to an ancestral population, or in the presence of asymmetric gene flow
(Ellstrand, 2014). However, if the edaphic condition confers strong selection pressures
on one side of the edaphic barrier and there is strong selection against the tolerance genes
on the other side of the edaphic boundary (i.e., low ‘Haldanian’fitness), then one could
expect evolution to occur even in the presence of gene flow.
Similarly, case studies of sympatric speciation (i.e., evolution from a single ancestral
species in the same location) must demonstrate species co-existence, sister relation-
ships, reproductive isolation, and that a former allopatric phase is highly unlikely. In
two palm species of Howea, endemic to the remote Lord Howe Island, Australia,
divergence in blooming periods is strongly correlated with soil type; only a few genetic
loci are more divergent between the two species than expected under neutrality, a
finding consistent with sympatric speciation involving disruptive or divergent selection
(Savolainen et al., 2006; Fig. 1a). Morphotypes of the palm Geonoma macrostachys in
Peruvian lowland forests provide another example of sympatric speciation likely
occurring between ecologically divergent (via soil and light) and reproductively isolat-
ed (via phenology and pollinators) populations (Borchsenius et al., 2016).
Budding or peripatric speciation is a diversification process through which new
species form at the periphery of an ancestral population. In this case, a few preadapted
genotypes encounter and colonize an edaphic island resulting in edaphic specialization.
Budding speciation appears to be especially common in plants (Crawford, 2010), as
high rates of selfing (Antonovics, 1968) and polyploidization (Soltis et al., 2004)result
in abrupt reproductive isolation between the ancestral and derivative populations. As
the initial peripheral populations are small, they can often quickly acquire reproductive
isolation via catastrophic selection, i.e., differential survival during an episode of mass
extinction (Gottlieb, 2004), or strong divergent selection (Grossenbacher et al., 2014).
At the time of speciation, sister taxa will exhibit highly asymmetrical and completely or
partially overlapping distributions, yet occupy distinct microhabitats (Gottlieb, 2004),
such as edaphically contrasting microsites. Eighty percent of the sister species of
Mimulus show patterns consistent with budding speciation: pairs are broadly sympatric,
N. Rajakaruna

have highly asymmetric distributions, and occupy distinct habitats (Grossenbacher
et al., 2014), suggesting that ecologically divergent taxa in this genus (Lowry et al.,
2008) may have arisen from budding or peripatric speciation (see Table 1for additional
case studies).
Fig. 1 Examples of edaphic influence on speciation and reproductive isolation. AHowea foresteriana is
largely restricted to limestone, while H. belmoreana is largely restricted to basalt (Savolainen et al., 2006;
Hipperson et al., 2016). Their geographic separation is accompanied by reproductive isolation via a difference
in flowering time: peak flowering in H. foresteriana (H. f.) occurs around two months before H. belmoreana
(H. b.). BHelianthus annuus (H. a.) and H. petiolaris are salt-intolerant species (glycophytes), but
H. paradoxus (H. pa.), which originated as a hybrid between the former two species, is a halophyte (Lexer
et al., 2003;2004). Genetic studies suggest that this may be due to changes in the expression of genes for ion
transporters, with increased calcium uptake limiting sodium uptake in the halophyte (Edelist et al., 2009). C
Mimulus guttatus has both copper-tolerant and -intolerant populations in California. The alleles for copper
tolerance (Tol 1 ) and hybrid lethality (Nec1) are at closely linked genetic loci; as selection for copper tolerance
occurred, both alleles rose to high frequency in the population on copper-contaminated soil (Wright et al.,
2013). DCollinsia sparsiflora has serpentine and nonserpentine ecotypes. Progeny of within-ecotype crosses
have high fertility; between-ecotype hybrids do occur, but the genetic distance between the ecotypes is
maintained because these hybrids have reduced fertility (ca. 26% reduction in seed viability in F1; Moyle
et al., 2012). Image Credit: Ian D. Medeiros
Lessons on evolution from the study of edaphic specialization

A meta-analysis of phylogenies available for plants in the California Floristic
Province (CFP) showed that of 71 sister pairs from 12 families, 57 pairs (80%) were
broadly sympatric (Anacker & Strauss, 2014). Range sizes of sisters differed by a mean
of 10-fold and range overlap and range asymmetry were greatest in younger sisters,
suggesting that in the CFP, budding speciation has been an important pathway for the
origin of neoendemic species. Interestingly, of the 71 sister pairs they examined, 29
(41%) showed complete edaphic shifts whereas 20 (28%) showed partial edaphic shifts,
strongly implying that budding speciation via edaphic specialization is commonplace in
the edaphically-heterogeneous landscape of the CFP. Similarly, in the Cape Floristic
Region (CFR) of South Africa, 87% of sister species show partial or complete
ecological shifts; however, complete edaphic shifts were found in only 17% of CFR
pairs (van der Niet & Johnson, 2009) versus in 41% of the CFP pairs. This difference
may be attributable to lower levels of edaphic and topographic heterogeneity in the
CFR compared to the CFP, likely making edaphic transitions less significant in
speciation (Anacker & Strauss, 2014).
Hybridization, with or without polyploidy, can play a critical role in the origin and
establishment of edaphic specialists (Abbott et al., 2013; De Storme & Mason, 2014),
especially under parapatric, sympatric, and peripatric modes of speciation. Studies of
the relative importance of hybridization and polyploidization on the origin of edaphic
endemics show that introgression between close relatives and subsequent adaptation of
hybrid offspring to novel edaphic conditions, when directly or indirectly resulting in
reproductive isolation with parental taxa, can result in the origin and establishment of
hybrid species (Table 1). Edaphic divergence of the hybrid species from its parental
taxa appears to be central to the successful origin of several hybrid species, occurring
even in the absence of post-zygotic isolation caused by chromosomal and other genetic
incompatibilities (Abbott et al., 2010). There is strong evidence, via reciprocal trans-
plant and common garden approaches, for greater fitness of a hybrid in a novel edaphic
habitat relative to that of parental taxa for the sand dune specialist Helianthus anomalus
and the salt marsh specialist H. paradoxus (Fig. 1b), although such conclusive evidence
is lacking for another hybrid species, H. deserticola (Edelist et al., 2009;Donovan
et al., 2010). Similarly, a Salix hybrid complex from the European Alps has greater
tolerance to cold temperatures and nutrient-poor/low pH soils than either parent
(Gramlich et al., 2016). The hybrids appear to have originated after glaciers retreated
and established persistent populations within a few decades. A major factor contribut-
ing to hybrid establishment under sympatric settings is the ability to occupy more
extreme edaphic niches relative to either parent species (Abbott et al., 2010;Gramlich
et al., 2016).
Are Edaphic Endemics New or Old Species?
Edaphic specialists, formed through the multiple modes described above, have often
been used to compare the relative importance of rapid versus gradual speciation in the
origin of endemic species (Pepper & Norwood, 2001; Anacker, 2014). Edaphic special-
ists can arise under both of these scenarios (i.e., rapid versus gradual). Recently evolved
neoendemics can evolve from nearby, non-specialized relatives via rapid and local
speciation, including via parapatric, sympatric, and peripatric modes (Anacker &
N. Rajakaruna

Tab l e 2 Some case studies of edaphic differentiation where soil type shifts between close relatives (sister taxa, congeners, populations) have been confirmed via ecological,
physiological, phylogenetic, or population genetic studies. Traits identified via experimental studies as key to adaptation and reproductive isolation are listed along with any direct
evidence for their genetic bases. For a detailed list of adaptive and reproductive trait differences between close relatives found on and off serpentine soils see O’Dell and Rajakaruna
(2011). Unknown=results inconclusive, no evidence, or not studied
Species Soil Type Adaptive Traits Reproductive Barriers Speciation Genes
Agrostis tenuis, A.
capillaris,
and A. stolonifera
Zinc and Copper mine soil (Wu et al.,
1975; Nichols & McNeilly, 1982;
Al-Hiyali et al., 1993); Saline soil
(Venables & Wilkins, 1978)
Multiple mechanisms, including
nicotianamine, L-cysteine,
superoxide dismutase production,
and HSP70 overexpression (Hego
et al., 2016)
Unknown Unknown
Alyssum serpyllifolium Serpentine (Sobczyk et al., 2017) Nickel hyperaccumulation (Sobczyk
et al., 2017)
Unknown Unknown
Anthoxanthum odoratum Mine soil, Park Grass Experiment
(nutrient and lime treatments)
(Silvertown et al., 2005;
Antonovics, 2006)
Metal exclusion from shoot (Qureshi
et al., 1985); nitrogen tolerance
(Freeland et al., 2010)
Flowering time difference; self-fertility
(Antonovics, 1968,2006;
Silvertown et al., 2005)
Unknown
Arabidopsis halleri Metal soils (Hunter & Bomblies, 2010) Zinc and Cadmium tolerance (Roux
et al., 2011)
Reduced hybrid fitness (Hunter &
Bomblies, 2010)
Unknown; HMA4 (heavy metal
ATPase 4) provides metal
tolerance (Roux et al., 2011)
Arabidopsis lyrata Serpentine (Turner et al., 2008) Possibly, heavy metal and Ca:Mg
tolerance (Turner et al., 2008,2010)
Genetic variation for flowering time
(Hunter & Bomblies, 2010)
Unknown; QTL associations
between local adaptation and
flowering time(Leinonen
et al., 2013)
Armeria maritima Calamine soils (Baumbach & Hellwig,
2007)
Role of malate in metal translocation,
reduced oxidative stress (Olko
et al., 2008; Parys et al., 2014)
Partial self-fertility (Vekemans &
Lefèbvre, 1997)
Unknown
Caulanthus amplexicaulis
var. barbarae
Serpentine (Pepper & Norwood, 2001) Nickel tolerance (Burrell et al., 2012) Unknown; possibly, geographic
isolation via biotype depletion as it
is completely interfertile in artificial
Unknown
Lessons on evolution from the study of edaphic specialization

Tab l e 2 (continued)
Species Soil Type Adaptive Traits Reproductive Barriers Speciation Genes
crosses with sister taxon (Pepper &
Norwood, 2001)
Ceanothus roderickii Gabbro (Burge & Manos, 2011) Low nutrient and Ca:Mg tolerance
(Burge & Manos, 2011)
Selection against hybrids in parental
habitats and possible genetic
incompatibilities (Burge et al.,
2013)
Unknown
Cerastium alpinum Serpentine (Nyberg Berglund et al.,
2004)
Magnesium and nickel tolerance
(Nyberg Berglund et al., 2004)
Unknown; possibly, selection against
hybrids (Ostevik et al., 2012)
Unknown
Collinsia sparsiflora Serpentine (Moyle et al., 2012) Possibly, Ca:Mg tolerance (Moyle
et al., 2012)
F1 hybrid sterility (Moyle et al., 2012) Unknown
Deschampsia cespitosa Metal soils (Hayward et al., 2013) Increased nicotianamine and
phytochelatin synthesis (Hayward
et al., 2013)
Unknown Unknown
Geonoma macrostachys Well-drained vs poorly-drained soil
(Borchsenius et al., 2016)
Unknown Flowering time and pollinators
(Borchsenius et al., 2016)
Unknown
Helianthemum spp. Gypsum (Moore et al., 2014) Accumulate excess calcium as oxalate
(Palacio et al., 2014); sulphate ac-
cumulation (Moore et al., 2014)
Unknown Unknown; phylogeny available
(Parejo-Farnés et al., 2013)
Helianthus exilis Serpentine (Sambatti & Rice, 2007) Magnesium and sodium exclusion,
water use efficiency (Sambatti &
Rice, 2007)
Unknown Unknown
Helianthus paradoxus Salt (Lexer et al., 2003) Increased calcium, potassium uptake
and sodium exclusion (Lexer et al.,
2003,2004; Edelist et al., 2009)
Habitat segregation due to salt
tolerance; differences in
photoperiodicity; hybrid sterility
(Lai et al., 2005; Henry et al., 2014;
Palacio et al., 2014)
Karyotypic differences leading
to hybrid sterility (Lai et al.,
2005)
Flowering time differences and
selection against hybrids
Pleiotropy or linkage between
loci contributing to flowering
N. Rajakaruna

Tab l e 2 (continued)
Species Soil Type Adaptive Traits Reproductive Barriers Speciation Genes
Howea forsteriana and
H. belmoreana
Volcanic soil and calcareous soil
(Savolainen et al., 2006)
Tolerance to salinity,drought,pH
(Dunning et al., 2016)
(Savolainen et al., 2006;Hipperson
et al., 2016)
time and adaptation to water,
salt, and pH stress (Dunning
et al., 2016)
Lasthenia californica
complex
Serpentine/Saline/Dry (Rajakaruna
et al., 2003a,b)
Enhanced shoot magnesium and
sodium tolerance (Rajakaruna et al.,
2003a); Early and increased
flowering under drought
(Rajakaruna et al., 2003b)
Flowering times, pollen-stigma in-
compatibilities (Rajakaruna &
Whitton, 2004)
Unknown
Lasthenia maritima Guano (Crawford et al., 1985) Elevated leaf nitrogen levels (Ornduff,
1965)–exact mechanism of toler-
ance Unknown
Self-compatibility (Crawford et al.,
1985)
Unknown
Layia discoidea Serpentine (Baldwin, 2005) Unknown Unknown; possibly selection against
hybrids in parental habitats
Unknown
Leptosiphon parviflorus Serpentine (Kay et al., 2011) Unknown Flowering time difference, possibly,
flower color (Kay et al., 2011)
Unknown
Mimulus guttatus Dry(inland)/Wet(coastal) (Lowry et al.,
2009)
Elevated leaf sodium levels (Lowry
et al., 2009)–exact mechanism of
tolerance unknown
Flowering time (Lowry & Willis,
2010)
Salt tolerance loci and an
inversion locus show fitness
effects in contrasting habitats;
colocalization of loci for
molybdenum and cadmium
accumulation (Lowry &
Willis, 2010; Lowry et al.,
2012; Twyford & Friedman,
2015)
Mimulus guttatus,
M. congdonii,
M. douglasii,
M. floribundus,
M. glaucescens,
Serpentine (Selby et al., 2014) Tolerance to low soil Ca:Mg and high
nickel; Greater magnesium
accumulation in shoot (Selby et al.,
2014;Palmetal.,2012)
Self-compatibility, Flowering times,
hybrid lethality, pollinator
differences (Gardner & Macnair,
2000; Selby et al., 2014)
Unknown
Lessons on evolution from the study of edaphic specialization

Tab l e 2 (continued)
Species Soil Type Adaptive Traits Reproductive Barriers Speciation Genes
M. kelogii,
M. layneae,
M. mephiticus,
M. nudatus,
M. pardalis
Mimulus guttatus,
M. cupriphilus
Copper (Macnair & Christie, 1983;
Wright et al., 2013)
Copper tolerance (Macnair, 1983;
Macnair et al., 1993;Smith&
Macnair, 1998)
Self-compatibility, flowering time
differences, hybrid lethality
(Christie & Macnair, 1984,1987;
Macnair et al., 1989;Friedman&
Willis, 2013;Wrightetal.,2013)
Linkage between Copper
tolerance locus and hybrid
incompatibility locus (Wright
et al., 2013; also see Christie
&Macnair,1987)
Mimulus laciniatus Granite (Peterson et al., 2013) Drought tolerance (Peterson et al.,
2013)
Flowering time differences (Ferris
et al., 2016)
Unknown
Noccaea caerulescens Serpentine and other metal-enriched
soils (Mandáková et al., 2015)
Enhanced metal homeostasis gene
expression (Mandáková et al.,
2015)
Possible role of inversion
polymorphism (Mandáková et al.,
2015); lower selfing rates in
metallicolous populations (Dubois
et al., 2003; Mousset et al., 2016)
Possibleroleofmetalion
homeostasis genes and
chromosomal rearrangements
(Mandáková et al., 2015)
Senecio lautus Sandy dunes and rocky headlands
(Melo et al., 2014)
Salinity tolerance, potassium transport
(Roda et al., 2013)
Immigrant inviability and extrinsic
postzygotic isolation via herbivores
and soil type (Melo et al., 2014
Unknown; however, loci for ion
homeostasis and
reproduction are known
(Roda et al., 2013)
Silene vulgaris Serpentine (Bratteler et al., 2006,b);
Copper soils (van Hoof et al., 2001)
Nickel tolerance (Bratteler et al.,
2006); Copper tolerance (van Hoof
et al., 2001)
Unknown Unknown; however,
constitutively higher
SvMT2b expression
increases the level of
tolerance produced by one or
more epistatic primary
tolerance genes (van Hoof
et al., 2001)
N. Rajakaruna

Strauss, 2014;Ferrisetal.,2014; Schneider et al., 2016). In contrast, ancient
paleoendemics may have evolved via gradual speciation through biotype depletion,
including via allopatric (vicariant) speciation, as has been shown for the Streptanthus
glandulosus complex (Mayer & Soltis, 1994), Caulanthus amplexicaulis var. barbarae
(Pepper & Norwood, 2001), and for Minuartia spp. (Moore et al., 2013;Moore&
Kadereit, 2013; Table 1). There is ample evidence from studies of edaphic specialists
that large-scale geographic isolation is not always critical for diversification; total
reproductive isolation can be achieved even in the face of gene flow (Antonovics,
2006). In fact, recent phylogenetic investigations suggest that neoendemism may be
the norm in cases of edaphic specialization, not the exception (Baldwin, 2014).
Peripatric or budding speciation is an especially common mode of speciation among
neoendemic edaphic specialists, with high rates of selfing and polyploidization causing
immediate reproductive isolation between adjacent yet edaphically distinct populations
(Gottlieb, 2004). Strong directional and disruptive selection that often occurs across
edaphic boundaries, combined with parallel changes contributing to reproductive isola-
tion between divergent populations, can lead to rapid and local speciation (Levin, 1993;
Crawford, 2010). In peripatric speciation, the progenitor species isinitially paraphyletic;
however, lineage sorting and extinction will eventually result in monophyly (Crawford,
2010). Case studies of edaphic differentiation in which there are phylogenetic or
population genetic data to support relatedness between edaphically divergent taxa are
listed in Table 2along with additional evidence for edaphic divergence via ecological,
physiological, reproductive, or genetic studies (when available).
How Does Adaptation Contribute to Reproductive Isolation?
Establishing a direct or indirect link between traits that confer edaphic adaptation and
traits that contribute to reproductive isolation is strong evidence for the role of edaphic
factors in the divergence of species (Hopkins, 2013; Wright et al., 2013). Classic
studies of plants on mine tailings in Europe (Antonovics, 2006 and references
therein) clearly demonstrated that rapid and repeated evolution of metal tolerance is
often associated with the evolution of prezygotic isolating mechanisms, including the
evolution of both self-compatibility (Mousset et al., 2016 and references therein) and
earlier flowering in metal-tolerant populations (Antonovics, 2006 and references
therein). Both of these traits significantly limit gene flow between divergent popula-
tions, aiding in the evolution of independent gene pools under strong edaphic selection.
Flowering time shifts in Anthoxanthum odoratum in the Park Grass Experiment, at the
boundaries between plots vastly varying in soil pH, have resulted in reduced gene flow
(Silvertown et al., 2005). Adjacent populations that had a common origin at the start of
the experiment in 1856 have now diverged at neutral marker loci, suggesting that
reinforcement (i.e., selection against maladaptive hybridization) has been key to com-
pleting speciation. Establishing a direct link between pH tolerance and flowering time
is necessary to confirm the role of such reinforcement in speciation. Interestingly, there
was no indication of reinforcement at the boundaries of plots with smaller pH differ-
entials, suggesting a link between pH preference and flowering time. Earlier flowering
times have often been documented for serpentine-associated taxa, including those in the
Lasthenia californica complex (Yost et al., 2012), Helianthus exilis (Sambatti & Rice,
Lessons on evolution from the study of edaphic specialization

2007), Mimulus spp. (Macnair & Gardner, 1998; Selby et al., 2014), Gilia capitata
(Kruckeberg, 1951), and Leptosiphon spp. (Kay et al., 2011). Near complete reproduc-
tive isolation between coastal and inland populations of Mimulus guttatus has been
achieved by selection against immigrants and flowering time differences (Lowry et al.,
2008). The annual, inland populations exposed to more summer drought flower early
compared to the perennial populations along the coast on more moist soils.
Diversification within the stone plant genus Argyroderma of the Western Cape
Province in South Africa also appears to have occurred through allopatric adaptive
speciation, under strong edaphic-mediated divergent selection along with reproductive
isolation achieved through divergence in flowering times (Ellis et al., 2006). In the well
documented case of sympatric speciation in Howea, the shift in flowering phenology
between species on volcanic and calcareous soils dramatically reduced interspecific
gene flow (Hipperson et al., 2016; Fig. 1a). There also appears to be a pleiotropy
between flowering time and adaptation to drought, salinity, and pH, driving speciation
(Dunning et al., 2016). Temporal isolation is a key factor in facilitating plant speciation;
however, flowering time differences between close relatives can arise as a flexible
response to divergent edaphic conditions (Levin, 2009) or through direct or pleiotropic
genetic effects of edaphic adaptations (Hunter & Bomblies, 2010; Selby et al., 2014).
Soil constituents, such as heavy metals, are directly implicated in causing reproduc-
tive isolation between close relatives. Copper tolerance in Mimulus guttatus is
expressed in the sporophyte as well as in the pistil (Searcy & MacNair, 1993).
Copper in the pistils acts as a selective agent since seed production is reduced when
pollen donors are maladapted to copper-rich soils (Searcy & Mulcahy, 1985; Searcy &
MacNair, 1990). Metal accumulation in reproductive organs generates a strong
prezygotic isolating mechanism by decreasing reproductive output in crosses between
edaphically divergent parents. Therefore, edaphic factors directly act as reproductive
barriers, favoring reproduction between plants growing under similar edaphic condi-
tions and minimizing gene flow between edaphically divergent taxa.
Serpentine tolerant Brassicaceae endemics can limit nickel uptake into the shoot,
particularly into reproductive organs, while non-endemics accumulate significantly
more nickel into reproductive organs relative to shoots (Meindl et al., 2014). In the
selenium hyperaccumulating Stanleya pinnata, selenium is preferentially allocated to
flowers, while the non-accumulating relative Brassica juncea has higher selenium
concentrations in leaves relative to flowers (Quinn et al., 2011a). Selenium had no
effect on pollen germination in the hyperaccumulator, but it impaired pollen germina-
tion in the relative, showing a possible evolutionary cost for selenium accumulation
through decreased pollen germination in the non-accumulating taxon.
Pollinators can distinguish between ‘clean’nectar and that contaminated with
metals, and avoid nectar enriched with metal because of toxicity (Meindl & Ashman,
2013). Pollinator preference of nectar, therefore, can potentially limit gene flow
between close relatives found on distinct substrates. Additional trait differences be-
tween edaphically divergent taxa can also act as reproductive barriers, including flower
size differences leading to differential pollinator preferences (Gardner & Macnair,
2000) and pollen-stigma incompatibilities leading to reduced potential for fertilization
(Rajakaruna & Whitton, 2004).
Edaphic specialists also provide opportunities for examining how post-zygotic
isolating mechanisms contribute to edaphic divergence. In the Mimulus guttatus
N. Rajakaruna

complex, there is evidence for inviability in hybrids between metal-tolerant and -
intolerant parents (Gardner & MacNair, 2000), although such post-zygotic isolation is
not a factor between coastal and inland ecotypes of the same taxon (Lowry et al., 2008).
Work on copper tolerance in M. guttatus (Macnair & Christie, 1983; Wright et al.,
2013) has revealed a genetic association between edaphic tolerance and hybrid invia-
bility (Fig. 1c). In Collinsia sparsiflora, ecotypic adaptation to serpentine is associated
with the expression of hybrid sterility between serpentine and nonserpentine ecotypes,
over spatial scales of less than 1 km (Wright & Stanton, 2007; Moyle et al., 2012;
Fig. 1d). Further, allozyme (Wright & Stanton, 2011) and microsatellite data (Moyle
et al., 2012) suggest there is little genetic differentiation between ecotypes, indicating
that hybrid sterility may have recently evolved.
A potentially fascinating example of the relationship between adaptation and repro-
ductive isolation yet to be studied lies in the progenitor-derivative species pair of Layia
glandulosa subsp. lutea and L. discoidea (Baldwin, 2005). Unlike Mimulus, in which
hybrid lethality has been documented between copper-tolerant and -intolerant taxa
(Wright et al., 2013), there is complete fertility between artificial crosses of the
serpentine-intolerant and serpentine-endemic Layia species; however, spontaneous
hybrids are not found in nature, suggesting that partially or completely reduced hybrid
fitness in one or both parental habitats may be responsible for the diversification.
Additionally, it is unknown whether the flower shape (discoid- vs. ray-flowered) plays
a role in pollinator preferences or whether potential differences in flowering time,
especially when populations occur along contact zones, contribute to pre-zygotic
isolation. Similarly, factors contributing to reproductive isolation between the guano-
endemic Lasthenia maritima and its close relative, the coastal bluff-specialist L. minor,
are little-known (Rajakaruna, 2004). There is evidence for greater self-compatibility
within the guano-endemic (Crawford et al., 1985), resembling the trends shown for
some metal-tolerant species (Mousset et al., 2016 and references therein); however, it is
unclear whether other isolation barriers, such as differences in pollinators or flowering
times, or hybrid lethality and reduced fitness of hybrids in parental habitats, also
contribute to preserving isolation. Further, the physiological bases for guano adaptation
are also unknown. Whether the high nutrient content (nitrogen, phosphorus), low pH
(due to uric and other acids), or some other chemical or physical factor of guano drives
selection on these substrates needs further investigation. For both Layia and Lasthenia
case studies, QTL-based comparative genome mapping and transcript profiling ap-
proaches (Sun & Schliekelman, 2011) can be used to determine whether any traits
conferring adaptation to substrate have effects on traits contributing to reproductive
isolation between the divergent taxa.
Are Edaphic Endemics Evolutionary Dead-Ends?
The growing availability of molecular phylogenies has provided us with means to
investigate the evolutionary dynamics of habitat specialization (Vamosi et al., 2014),
including directionality of evolution of edaphic endemism (Anacker et al., 2011).
Population genetic and phylogenetic analyses of serpentine-tolerant plants (Mengoni
et al., 2003a,b; Nyberg Berglund et al., 2004;Anacker,2011), metallophytes (Mengoni
et al., 2001; Pauwels et al., 2005), halophytes (Bennett et al., 2013), calciphiles (Klein
Lessons on evolution from the study of edaphic specialization

& Kadereit, 2015), and gypsophiles (Bresowar & McGlaughlin, 2015) have demon-
strated that tolerance to edaphic conditions has been gained numerous times within
various groups of angiosperms and has even been lost in some groups, suggesting that
traits conferring edaphic tolerance are evolutionarily labile. Further, differences in the
evolutionary dynamics (i.e., gains vs. losses) of edaphic tolerance documented among
different lineages may result from association of edaphic tolerance traits with other
ecophysiological traits related to tolerance of environmental stresses, with edaphic
tolerance being more easily gained in preadapted lineages (Saslis-Lagoudakis et al.,
2014; Selby et al., 2014). The direction of serpentine endemism among genera with
serpentine-tolerant species in California is mostly from non-tolerant and -tolerant
species to endemic species, with a few reversals from the tolerant or endemic state to
the non-tolerant state (Anacker et al., 2011). This suggests that the evolution of edaphic
endemism may represent an evolutionary dead-end. This may not always be the case,
however, as endemics are often able to influence adjacent populations by repeated
episodes of introgression and polyploidization. In the diploid-tetraploid complex of
Knautia arvensis, recurrent polyploidization within serpentine populations followed by
hybridization with the neighboring non-serpentine tetraploid populations, has allowed
serpentine tetraploids to escape from their serpentine refuge and spread further into
non-serpentine habitats (Kolářet al., 2012). Similarly, endemic Streptanthus clades are
often quite diverse, with endemism having evolved numerous times and having been
lost at least once (Cacho et al., 2014). In Arabidopsis, there is also evidence for
serpentine-tolerant loci being introgressed from one serpentine-endemic to another
(Arnold et al., 2016) and for hybridization between two sympatric and diploid lime-
stone endemics, giving rise to a tetraploid siliceous soil endemic (Schmickl & Koch,
2011). These findings suggest that edaphic endemism does not always represent an
evolutionary dead-end and that endemics have the potential to further diversify within
(Cacho et al., 2014)oracrosssubstrates(Schmickl&Koch,2011).
How Does Tolerance to Edaphic Extremes Influence Competitiveness
in ‘Normal’Soils?
Edaphic islands have long been considered a refuge from competition (Kruckeberg,
1954), suggesting that tolerance to edaphic extremes has a cost. However, the idea of a
trade-off of competitive ability for tolerance (i.e., trade-off hypothesis) needs close
scrutiny via experimentation (Harper et al., 1997a). Edaphic specialists, compared to
non-specialists, often appear to be less competitive (Milla et al., 2011; Moore et al.,
2013;Anacker,2014; Bastida et al., 2014) and more susceptible to herbivory when
found on ‘normal’soils (Dechamps et al., 2008; Kay et al., 2011;Strauss&Boyd,
2011). However, plants that specialize in a particular edaphic setting (e.g., a metal
mine) may have adapted not only to the increased metal concentrations but also to other
stressors such as drought, low nutrients, high light, and poor soil structure character-
izing the mine site. Therefore, it is unclear if a reduction in fitness or competitive ability
results from the direct cost associated with metal tolerance per se (i.e., trade-off
hypothesis) or with other correlated traits conferring adaptation to the mine site. To
demonstrate that a negative correlation between traits is the result of pleiotropy (and not
via gene/s in linkage disequilibrium with metal tolerance gene/s) it is essential to
N. Rajakaruna

experimentally demonstrate that the traits are genetically correlated (Partridge & Sibly,
1991). Experimental studies on copper tolerance in Mimulus guttatus (Harper et al.,
1997a), however, provide no clear evidence in support of a trade-off hypothesis.
Additionally, low fitness of metal tolerant plants in ‘normal’environments can also
result from an increased metabolic requirement for the metal (i.e., metal requirement
hypothesis). Studies conducted on M. guttatus, however, show no such increased
requirement for copper (Harper et al., 1997b,1998), suggesting that neither the trade-
off hypothesis nor the metal requirement hypothesis can fully explain the cost of copper
tolerance in M. guttatus. Additional experimentation will likely reveal that it is not one
trait but a suite of adaptive traits (i.e., genes, their modifiers, and the interactions
between them; see Smith & Macnair, 1998)thathasacostina‘normal’habitat.
Similarly, there is little evidence of a cost for tolerance to either low soil calcium to
magnesium ratios or high concentrations of heavy metals, two key selective pressures
experienced by plants growing on serpentine soils (Brady et al., 2005; Kay et al., 2011).
Serpentine endemics, however, are typically slow-growing and stress-tolerant, rather
than fast-growing and what is typically considered as competitively superior (Anacker
&Harrison,2012; Fernandez-Going et al., 2012), consistent with the notion of a trade-
off between competitive ability and tolerance to serpentine. Indigenous resistant plants
on serpentine soil have less competition from invasive species when growing on
serpentine soil (Harrison, 1999;Grametal.,2004; Going et al., 2009), but are poor
competitors on high-nutrient soils (Rice, 1989; Jurjavcic et al., 2002; Vallano et al.,
2012). Regardless, the evidence that competition is reduced on serpentine soils (Moore
& Elmendorf, 2011) and that edaphic endemics are competitively inferior is not
substantial (Powell & Knight, 2009; Fernandez-Going & Harrison, 2013).
Furthermore, these hypotheses of a cost of tolerance have not been tested in numerous
lineages, nor have they been examined across distinct substrate types.
Comparing pairs of sister taxa could help reveal the evolutionary costs of habitat
specialization relative to broad tolerance and address why some species become
edaphic specialists whereas others maintain general-purpose genotypes (sensu Baker,
1965) that can tolerate wide-ranging habitat conditions. A widely-distributed Aquilegia
subspecies has greater adaptive phenotypic plasticity and higher competitive ability
than a related taxon that is a narrow endemic (Bastida et al., 2014). This suggests that
adaptive plasticity, competitive ability, and lineage history all contribute to phenotypic
divergence, edaphic niche specialization, and range distribution.
The hypothesis of a greater cost of tolerance and resulting trade-offs in competitive
ability is, however, gaining recognition as a potentially important driving force in the
evolution of edaphic endemism. The integration of molecular phylogenies with de-
scriptive and experimental ecological data reveals that occupation of bare habitats is a
precursor for serpentine specialization in Streptanthus and close allies (Cacho &
Strauss, 2014) and may be central to soil specialization. When comparing ancestor-
derivative species pairs, it was evident that shifts onto serpentine soils were not from
soils with a low calcium to magnesium ratio or high nickel content (like those
characteristic of serpentine soils) but from bare microhabitats. Preadaptations to fea-
tures characterizing bare habitats, including water stress, combined with inherently
lower competitive ability, perhaps due to increased investment for defense (Strauss &
Cacho, 2013), may have enabled the colonization of serpentine (also, see von Wettberg
et al., 2014), leading to multiple origins of serpentine adaptation among Streptanthus
Lessons on evolution from the study of edaphic specialization

and allies (Armbruster, 2014). Whether such preadaptation to bare habitats is a
precursor for adaptation to other rock outcrop plant communities can be tested via
similar approaches, using a combination of molecular phylogenetic and ecological data
(Cacho & Strauss, 2014).
Does Parallel Evolution Occur Frequently in Edaphic Specialists?
Parallel (or repeated) evolution provides strong evidence for the role of ecology in
driving divergence (Nosil, 2012). Parallel edaphic adaptation may be achieved via the
evolution of novel alleles contributing to specialization or via parallel increases in the
frequency of alleles (Faria et al., 2014). Repeated evolution of traits may not be unusual
in clades that have undergone ecological radiation, particularly in those taxa adapted to
environments imposing strong divergent selection such as harsh edaphic environments
(Levin, 2001; Mengoni et al., 2003a,b). When changes in edaphic tolerance are
associated with changes in reproductive compatibility, a case for parallel ecological
speciation can be made (Rajakaruna & Whitton, 2004). Contrary to these findings,
there are only a limited number of studies showing the repeated evolution of traits
conferring adaptation and reproductive isolation in plants, i.e., parallel ecological
speciation (Ostevik et al., 2012). Interestingly, of the 23 potential cases for parallel
ecological speciation cited by Ostevik et al. (2012), 10 are from plants adapted to harsh
edaphic conditions (see Table 1in Ostevik et al., 2012). The prevalence of cases of
edaphic adaptation among the limited number of studies of parallel ecological specia-
tion suggests that parallel evolution appears to occur more often in, or is better known
from, edaphic specialists relative to other species. Genomic methods can play an
important role in determining whether parallel (via independent evolution of the
same or different genes; Nichols & McNeilly, 1982;Levin,2001; Rajakaruna et al.,
2003c) or horizontal gene flow of clustered genes (via introgression; Arnold et al.,
2016) are responsible for the parallel evolution of edaphic tolerance.
Do Interactions with Herbivores or Microbes Drive the Evolution
of Edaphic Specialization?
Plant interactions with other organisms, including those that are mutualistic and
antagonistic, appear to have a role in edaphic specialization (Strauss & Boyd, 2011;
Van Nuland, 2016). Harsh soils, which are often nutrient-poor, water-stressed, trace-
element rich, and found in barren and open landscapes, may increase physiological
costs associated with herbivore damage as well as make plants more conspicuous to
herbivores (Strauss & Boyd, 2011). Edaphic specialists may be more susceptible to
herbivory than their non-specialist relatives and plant-herbivore interactions can some-
times be central to edaphic specialization (Fine et al., 2004; Van Zandt, 2007; Lau et al.,
2008). Increased herbivore pressure due to increased noticeability on rock outcrops can
drive increased investment in defense strategies among edaphic specialists (Strauss &
Cacho, 2013). The high costs of defense can result in trade-offs in plant competitive-
ness (Fine et al., 2006) and restrict plant distributions on harsh soils, contributing to
high rates of edaphic endemism. Similarly, the pathogen refuge hypothesis
N. Rajakaruna

(Kruckeberg, 1992;Springer,2007) suggests that plants will experience lower disease
pressure on serpentine and other edaphically harsh environments, either through
reduced disease transmission rates in sparsely populated serpentine plant communities
(Thrall et al., 2007) or through lowered disease-associated damage, as has been shown
for serpentine-adapted Hesperolinon (Springer, 2009). Serpentine specialists within
Hesperolinon are less susceptible to fungal infection, perhaps due to their ability to
selectively uptake calcium, which is required to initiate an effective immune response.
Less specialized species appear to be more susceptible to fungal disease on serpentine
soil (Springer, 2007,2009). Such a pathogen-refuge effect might also reinforce edaphic
associations among plants and promote greater edaphic specialization. In parapatric
coastal populations of Senecio lautus, herbivores have a central role in preventing gene
flow in the field via differential seedling predation of both migrants and hybrids,
whereas soil elemental content contributes to divergent selection on populations both
locally and on a geographic scale (Melo et al., 2014).
Specialization in plant host-symbiont-soil interactions may also facilitate plant
adaptation to edaphic stress and subsequent specialization (see Table 2in Van
Nuland, 2016). Arbuscular mycorrhizal fungi (AMF) are common root symbionts that
can increase a plant hosts' establishment and growth in edaphically-stressful environ-
ments (Palacio et al., 2012; Schechter & Branco, 2014; Southworth et al., 2014). There
are plant and microbial symbionts that enhance plant adaptation to edaphic stress
(Thrall et al., 2007), showing that co-adapted interactions may play a role in edaphic
specialization. For example, arsenic-tolerant AMF suppress high-affinity arsenate/
phosphate transport in the roots and decreases arsenic uptake by the arsenic-tolerant
grass, Holcus lanatus, showing that arsenic-tolerant AMF may contribute to enhanced
arsenic-tolerance in the metal-tolerant populations of the grass (Gonzalez-Chavez et al.,
2002). However, salt-tolerant Acacia species from a high-salinity environment are less
dependent on Rhizobium bacteria than salt-sensitive hosts from less stressful environ-
ments, suggesting plant–rhizobial mutualisms may be less important in plant adaptation
to salinity (Thrall et al., 2008). Arbuscular mycorrhizal fungal assemblages associated
with serpentine and non-serpentine ecotypes of Collinsia sparsiflora showed that plant
ecotypes are associated with distinct AMF assemblages (Schechter & Bruns, 2008);
however, a common garden test of host-symbiont specificity provided no evidence of
adapted host ecotype-AMF specificity, suggesting that soil type, not microbial speci-
ficity, is key to determining AMF assemblage structure in C. sparsiflora (Schechter &
Bruns, 2013). Findings on plant host-microbial symbiont specificity and the role of
AMF on edaphic stress tolerance in higher plants are variable (Thrall et al., 2008;
Glassmand & Casper, 2012); some AMF, however, have the capacity to confer metal
tolerance in some plant host species either via sequestration or by contributing to
growth enhancement via increased nutrient acquisition (Schechter & Branco, 2014).
Such mutualistic interactions are clearly beneficial for some plant species, allowing
them to colonize harsh soils and thus setting the stage for potential divergence.
The ecological and evolutionary significance of metal hyperaccumulation has re-
ceived much attention (Boyd, 2014). The selective pressure for resistance to pathogens
and herbivores may also result in fixation of metal hyperaccumulating trait(s) in plant
populations (Boyd, 2007;Hörgeretal.,2013). Metal (and metalloid)
hyperaccumulation has a profound effect on cross-kingdom interactions (El Mehdawi
& Pilon-Smits, 2012; Boyd, 2014) as metal-hyperaccumulating plants offer a
Lessons on evolution from the study of edaphic specialization

specialized niche for metal-tolerant herbivores, pollinators, microbes, and neighboring
plants. For example, high levels of metal in plant tissues reduce herbivory (Quinn et al.,
2010) and pathogen infection (Fones et al., 2010) and result in increased litter decom-
position by specialized microbes (Quinn et al., 2011b), causing an elemental allelo-
pathic effect on metal-intolerant neighbors (El Mehdawi et al., 2011); all these effects
can influence the fitness of plants accumulating the metal, thereby contributing to
edaphic specialization. Plant-microbe interactions may also influence gene flow be-
tween divergent populations due to microbe-assisted edaphic differentiation. A study of
an Artemisia hybrid zone in central Utah showed that the microbial community
influenced the performance of parental and hybrid plants in native and nonnative soils,
segregating the distinct populations and likely contributing to their edaphic divergence
(Miglia, 2007).
Metal-hyperaccumulator plants found on serpentine and other metal-rich environ-
ments are hosts to insects that appear to have co-evolved with them (Boyd, 2014).
Some of these insects do accumulate heavy metals to toxic levels (Boyd, 2007)and
may use them as defense against natural enemies (i.e. elemental defense hypothesis;
Boyd, 2007). Although only one study supports this hypothesis (Boyd & Wall, 2001),
little research regarding this hypothesis has been conducted to date. There is also no
overwhelming evidence for a cost of hyperaccumulation in terms of reproductive
functions or pollinator visitation. For example, pollinators of Stanleya pinnata, a Se-
hyperaccumulator, showed no visitation preference between high- and low-Se plants,
suggesting greater tolerance to the metalloid (Quinn et al., 2011a). Whether the
presence of aluminum and nickel in nectar of Impatiens capensis alters foraging
behavior by bumblebees has also been investigated (Meindl & Ashman, 2013). The
presence of aluminum in nectar did not influence foraging patterns; however, flowers
containing nickel-laden nectar solutions had shorter visits relative to controls, and even
resulted in reduced visitation of nearby nickel-contaminated flowers. Reduced visita-
tion by native pollinators has also been documented for Streptanthus polygaloides
(Meindl & Ashman, 2014) which accumulates nickel in pollinator rewards such as
nectar and anthers. Although heavy metal effects on plant-pollinator interactions are
clearly context-dependent, edaphic features, such as heavy metal concentrations, have
the capacity to influence gene flow and reproductive isolation between close relatives
through pollinator selection.
Are Edaphic Specialists More or Less Sensitive to Climate Change?
Species can respond to climate-associated stressors via both phenotypic plasticity and
local adaptation (Valladares et al., 2014). Edaphic specialists, many of which display
largely disjunct but widespread distributions, are ideal models for exploring how
climate interacts with edaphic factors to influence plant distribution (Damschen et al.,
2010; Sánchez et al., 2017). It is easy to envision how an altered climate regime can
have differential effects on populations of an ecologically widespread taxon. For
example, serpentine populations are generally better-adapted to low moisture condi-
tions typical of their mostly exposed, rocky, and shallow soils (Wu et al., 2010)and
may be able to withstand increased aridity and high temperatures. Therefore, under a
drying climate, as is predicted for some parts of the world, nonserpentine populations
N. Rajakaruna

may be extirpated, whereas serpentine populations, already preadapted to aridity, could
survive and become endemic due to the extinction of the more widespread
nonserpentine populations (Moore et al., 2013; but see Anacker, 2014). Pronounced
climatic oscillations of the Quaternary period appear to have forced populations of
Sempervivum,Jovibarba,andAdenostyles to respond via latitudinal or altitudinal shifts
into edaphically unsuitable but relatively competition-free habitats (Klein & Kadereit,
2015). This set the stage for the divergence of edaphically specialized species via
allopatric speciation. A similar scenario is presented for the evolution of the edaphically
specialized Lilium pyrophilum (Douglas et al., 2011).
There is increasing evidence that climatic and edaphic factors interact to control
local adaptation and substrate endemism (Fernandez-Going, 2014). Serpentine ende-
mism peaks in wet regions both in California and some other parts of the world
(Anacker, 2011; Fernandez-Going, 2014), likely due to increased competition on
‘normal’substrates under more productive climates. At the community level, there is
also more species, functional, and phylogenetic turnover across edaphic boundaries in
mesic relative to arid regions of California (Anacker & Harrison, 2012). Furthermore,
serpentine endemics in California generally occupy wetter regions than even their
closest relatives (Fernandez-Going, 2014), but this is only true in the case of
neoendemics arising from serpentine-intolerant ancestors; more productive climates
may have been more suitable for the persistence of small neoendemic taxa with novel
adaptations during and after speciation. Paleoendemics, on the other hand, occupy
similar climates to those of their closest relatives, suggesting that range contractions
leading to biotype depletion may have been similar in arid and mesic regions (Anacker,
2014). It is as of yet unclear whether climate can alter soil-specific competitiveness in
edaphic endemics (see Fernandez-Going & Harrison, 2013) and whether
paleoendemics and neoendemics differ in their sensitivities to competition.
Climate change will clearly have differential effects on the ranges of edaphic
specialists. In California, future climate conditions may extend the potential range of
many endemics (Loarie et al., 2008); however, whether edaphic endemics can naturally
disperse to suitable habitat is questionable (Damschen et al., 2012;Spasojevicetal.,
2014). Preadaptations to dry conditions may allow edaphic specialists to expand their
current range (Fernandez-Going et al., 2012;Fernandez-Going,2014) and even be
more competitive with non-specialists off of their special edaphic niche (Moore et al.,
2013). Tolerance of edaphic stress appears to go hand-in-hand with tolerance of
climatic stress for plants growing on serpentine soils, linking soil infertility to a
stress-tolerant functional trait syndrome that confers unusually high resistance to
climate change upon plant species and communities (Harrison et al., 2015).
Therefore, in the absence of nutrient deposition, stress-tolerant invasive species, and
habitat loss, edaphic specialists may perform better under future warmer/drier climates
relative to plants adapted to more fertile soils. With possible range expansions onto
‘new’habitat, however, they will no longer be restricted to their special edaphic niche,
thereby losing their edaphic endemic status (Anacker, 2014). Habitat patchiness may
also contribute to invasiveness, with competitive and dispersal abilities evolving more
rapidly on island-like habitats (Williams et al., 2016), suggesting that fragmentation can
select for more rapid invasion velocity. How these findings are influenced by different
climatic regimes, or by habitat patches with different levels of edaphic stress, are also
important avenues for research. Naturally fragmented edaphic islands, such as
Lessons on evolution from the study of edaphic specialization

serpentine outcrops, are model habitats for exploring such questions and for testing the
roles of patch size, shape, type (soil), and isolation in ecological and evolutionary
processes (Gil-López et al., 2014).
What is the Genetic Basis of Edaphic Specialization and Speciation?
In the era of genomics, there are many tools (e.g., Bratteler et al., 2006a,b;von
Wettberg & Wright, 2011; Selby et al., 2014) at our disposal to explore the genetic basis
of adaptation to harsh environments, including the nature of adaptation to dissimilar
habitats with similar abiotic stressors (e.g., arid and saline habitats contributing to water
stress, saline and serpentine soils contributing to ionic stress, or serpentine and carbon-
ate soils contributing to high pH). Phylogenetic analyses show that there is a significant
association between halophytes and metal hyperaccumulators, although within each
group there is low phylogenetic clustering (Moray et al., 2016), suggesting that salinity
and metal tolerance can vary among close relatives. In Asteraceae, Amaranthaceae,
Fabaceae, and Poaceae, halophytes and hyperaccumulators are more closely related
than would be expected if the two traits had evolved independently, suggesting a strong
genetic association between the abilities to tolerate salt and heavy metals. Studies of
evolutionary ecology now routinely use high-throughput sequencing and other geno-
mic approaches to examine the genetic basis of speciation (Tiffin & Ross-Ibarra, 2014)
and these methods have clearly promoted the study of edaphic adaptation (Turner et al.,
2010; Porter et al., 2017).
Genomic polymorphisms that are differentiated between edaphically contrasting
populations can reveal candidate loci for edaphic adaptation. Gene pool sequencing
of serpentine and granitic populations of Arabidopsis lyrata from the USA shows
that the polymorphisms that are most strongly associated with soil type are loci for
heavy-metal detoxification and calcium and magnesium transport, providing several
candidate loci for serpentine adaptation (Turner et al., 2010), including CAX7,
which appears to be responsible for adaptation to low calcium to magnesium ratios
characteristic of serpentine soils (Turner et al., 2008). Moreover, the same associ-
ations between individual loci and serpentine tolerance were observed in the
European subspecies of A. lyrata, further implicating edaphic selection. The se-
quencing of three candidate loci in the European subspecies indicated parallel
differentiation of the same polymorphism at one locus, confirming edaphic special-
ization, and different polymorphisms at two other loci, suggesting convergent
evolution. Whether these loci also contribute to reproductive isolation between
progenitor-derivative taxon pairs is unknown.
Adaptation to low soil calcium to magnesium ratios has often been suggested as key
to serpentine tolerance (Palm & Van Volkenburgh, 2014). In serpentine-intolerant
Arabidopsis thaliana, an induced loss of function mutation in the Ca
2+
-H
+
antiporter,
CAX1, enhances survival on soils with a low calcium to magnesium ratio (Bradshaw,
2005). CAX1 is a high-capacity Ca
2+
-H
+
antiporter that maintains cytoplasmic ion
homeostasis by pumping excess calcium from the cytoplasm into the vacuole (Cheng
et al., 2003). Mutation of CAX1 effectively acts as a magnesium exclusion strategy,
enabling the plant to maintain adequate cytoplasmic calcium concentrations under the
low calcium to magnesium ratios encountered on serpentine soils.
N. Rajakaruna

Additional candidate genes have recently been identified for serpentine adaptation in
autotetraploid Arabidopsis arenosa, providing evidence that some selected alleles were
introgressed from the diploid serpentine-endemic A. lyrata, whereas others were
independently involved in separate adaptation events in the two species (Arnold
et al., 2016).TheresearchbyArnoldetal.(2016) shows strong evidence for selection
on genes that control specific ion homeostasis-related traits, as well as drought adap-
tation. Although several alleles under selection appear to have been introgressed from
A. lyrata, others have been independently selected following serpentine colonization in
both species. The significant overlap between selected genes in serpentine-endemic
A. arenosa and A. lyrata suggests that adaptations to serpentine are not qualitatively
different between diploid and tetraploid species. The work described above on
Arabidopsis (Turner et al., 2008,2010; Arnold et al., 2016) advances our understanding
of the polygenic basis of multi-trait adaptation to serpentine soils and its repeatability
across species. Further, the work suggests that edaphic divergence can occur even in the
presence of high levels of inter- and intraspecific gene flow. Serpentine-tolerant
Arabidopsis taxa, along with their close relatives, can provide a useful system for
investigating the genetics of speciation under the serpentine influence.
The genetic tools at our disposal (Sun & Schliekelman, 2011;vonWettberg&
Wright, 2011; Tiffin & Ross-Ibarra, 2014) can reveal the genetic architecture of
adaptive traits that confer reproductive isolation (i.e., speciation genes; Nosil &
Schluter, 2011) and whether the same alleles or different alleles contribute to parallel
evolution (Ostevik et al., 2012), revealing new insights into the mechanisms by which
natural selection can bring about adaptation, reproductive isolation, and speciation.
Mechanisms resulting in reproductive isolation may be incidental byproducts of natural
selection on linked loci or on loci with pleiotropic effects. For example, plant popula-
tions on serpentine and other harsh edaphic settings often flower early (O’Dell &
Rajakaruna, 2011). Such phenological isolation could be a byproduct of adaptation to
harsh soils or a result of an independent set of genetic changes that were favored by
natural selection (Baack et al., 2015). A detailed genetic analysis of the alleles
responsible for reproductive barriers can separate these hypotheses, particularly when
traits are correlated and contribute to reproductive isolation. In Mimulus, differences in
flowering time are often associated with differences in soil water availability (Wu et al.,
2010; Friedman & Willis, 2013). In wet coastal and dry inland populations of Mimulus
guttatus (Lowry et al., 2008), an inversion polymorphism appears to be a major
quantitative trait locus (QTL) for flowering time and growth-related traits, explaining
20–30% of the variation observed between populations (Lowry & Willis, 2010).
Reciprocal transplant experiments have demonstrated strong local adaptation in traits
that map to the inversion and differ between the coastal and inland ecotypes (Lowry
et al., 2008;Halletal.,2010).
In Mimulus guttatus, the locus that contributes to copper tolerance also leads to
hybrid lethality in crosses between copper-tolerant and -intolerant populations (Macnair
&Christie,1983). High-resolution genome mapping show that copper tolerance and
hybrid lethality are not caused by the same gene, but are separately controlled by two
tightly linked loci (Wright et al., 2013; Fig. 1c). Selection on the copper tolerance locus
indirectly causes the hybrid incompatibility allele to increase in frequency in the mine
population via hitchhiking. Therefore, hybrid incompatibilities can evolve as a by-
product of an adaptation to a novel edaphic environment. Whether there are similar
Lessons on evolution from the study of edaphic specialization

bases for hybrid lethality between edaphically divergent ecotypes of serpentine tolerant
species such as Collinsia sparsiflora (Moyle et al., 2012;Fig.1d) remains to be seen.
Current model systems are not representative of all life history strategies and are
mostly restricted to herbs, and we need to continue to add new model systems to allow
comparative approaches for exploring the intricacies of adaptive evolution (Selby et al.,
2014; von Wettberg et al., 2014). In this regard, the extensive work on the genetics of
metal tolerance (Verbruggen et al., 2009) in metal-accumulating and -
hyperaccumulating plants used in phytoremediation (Rascio & Navari-Izzo, 2011),
and work on the genetics of abiotic stress tolerance and phenology in crops (Roy
et al., 2011), are important for examining the genetic bases for edaphic adaptation and
species divergence.
Conclusions and Future Directions
Edaphic specialists are key players in the study of local adaptation and ecological
speciation (Crawford et al., 2014). Ecological, physiological, phylogenetic, and popu-
lation genomic studies of plants growing on serpentine, saline, and metal-contaminated
soils continue to shed light on factors and mechanisms driving ecological speciation
(Table 2). Studies such as those on Mimulus on copper-enriched soil (Macnair &
Christie, 1983; Macnair et al., 1993; Wright et al., 2013; Fig. 1c], Collinsia and
Arabidopsis on serpentine soil (Moyle et al., 2012; Turner et al., 2010;Fig.1d),
Helianthus on saline soils (Edelist et al., 2009; Fig. 1b], and Senecio on sand dunes
and rocky headland soils (Melo et al., 2014), are key to establishing a direct link
between adaptation, reproductive isolation, and subsequent speciation. Ongoing re-
search on Mimulus,Helianthus,andArabidopsis (and others; Tables 1and 2) will
continue to unravel how edaphic conditions can directly or indirectly contribute to
species divergence.
Lessons on Evolution
There are several key lessons on evolutionary processes from the study of edaphic
specialization:
a. Both adaptive phenotypic plasticity and local adaptation contribute to edaphic
divergence. Edaphic specialization is key to diversification in some groups; how-
ever, in others, there appears to be lack of variation for adaptation to harsh soils.
Although edaphic specialization is important in the diversification of flowering
plants and ferns, there is little evidence for its significance in other plant groups such
as gymnosperms and bryophytes or in lichens, soil algae, and cyanoprokaryotes.
b. Although edaphic specialists can evolve under allopatric, parapatric, and sympatric
modes of speciation, there is strong evidence pointing to budding or peripatric
speciation being central to the origin of neoendemic species (Anacker & Strauss,
2014). Additionally, an increasing body of work shows that speciation is possible
in the face of gene flow, especially in lineages where there is considerable
ecological (including edaphic) divergence. Further, introgression between close
relatives can also lead to the acquisition of edaphic tolerance (Arnold et al., 2016).
N. Rajakaruna

c. Edaphic adaptations have direct or indirect effects on reproductive isolation;
adaptive and reproductively isolating traits can evolve rapidly and repeatedly in
some lineages, making edaphic specialists key to studying parallel speciation.
d. There is mixed evidence for a physiological cost of edaphic tolerance driving
habitat specialization, although both reduced competitive ability and greater sus-
ceptibility to herbivory have been documented among some edaphic specialists
when grown in ‘normal’soils.
e. Mutualistic and antagonistic interactions with other organisms can mediate edaphic
specialization either by directly enhancing tolerance to edaphic extremes via the
action of microbes or by reducing gene flow between divergent populations via the
action of pollinators, herbivores, and pathogens.
f. Climate and soils can interact to influence edaphic specialization. Future climates
may further restrict the distribution of edaphic specialists due to increased compe-
tition from invasive species. However, edaphic specialists could also expand their
ranges, even outside their edaphic niche, if their preadaptations to drought make
them more competitive than non-specialists under a drying climate.
g. QTL mapping studies, population genomic approaches, and high-throughput phe-
notyping assays, when combined with reciprocal transplant studies using hybrid
mapping populations (including, F2, backcross, near isogenic lines, and recombi-
nant inbred lines), can identify major loci contributing to edaphic adaptation to
serpentine, saline, and heavy metal-rich soils (Selby et al., 2014). These loci often
have direct or indirect effects on loci that contribute to reproductive isolation
(Table 2).
Information Gaps and Future Directions
There are several information gaps that need to be addressed to better develop current
model systems as well as to incorporate additional taxa for the study of edaphic
specialization. The following are key areas for future research:
a. The identification of specific soil factors that are central to edaphic specialization
is important for demonstrating the role of adaptation in speciation. Although most
studies state only the general soil type (e.g., serpentine versus non-serpentine,
saline versus non-saline, gypsum versus non-gypsum, etc.) being investigated, a
handful of studies have attempted to examine the exact soil chemical or physical
factor contributing to the specialization and/or divergence (Rajakaruna et al.,
2003a,b;Bradshaw,2005; Bratteler et al., 2006,b; Palacio et al., 2007;Edelist
et al., 2009; Selby et al., 2014). Identifying the key soil variable or variables is
more effective when comparing sister populations found across substrates that
differ in a distinct way (e.g., with or without a heavy metal as in the case of copper
tolerance in Mimulus). Plant ionomic approaches can also be used to investigate
ecological functions of ionomic alleles in the adaptation to distinct soils (Lowry
et al., 2012;Huang&Salt,2016).
b. Better characterization of functional traits (e.g., morphological, physiological, and
reproductive traits) between sister species that differ in their edaphic niche is also
critical for determining which traits may be contributing to the divergence. Whether
Lessons on evolution from the study of edaphic specialization

traits characteristic of the edaphic habitat are plastic responses or are genetically
based should also be confirmed. Once those traits targeted by edaphic selection are
identified, the physiological/genetic basis of those traits should be determined.
Much of what we know about edaphic specialization comes from studies on plants
found on metal mines, serpentine soils, or saline environments of temperate or
Mediterranean climes, and it is essential that we expand our studies to identify and
characterize systems found on other types of substrates and in other climatic regions
of the world. Such a broad survey can reveal similarities and differences in trends
associated with speciation resulting from edaphic specialization.
c. Because multiple reproductive barriers can contribute to reduced gene flow
between divergent taxa (Baack et al., 2015; Ostevik et al., 2016), it is important
to identify and characterize the relative contribution of all pre- and post-zygotic
reproductive isolating mechanisms to the total reproductive isolation between
edaphically divergent pairs. The order in which barriers originate and their genetic
bases and associations with traits characterizing edaphic adaptations may shed
light on the relative importance of each barrier to different stages of a divergence
event and at different levels of habitat specialization.
d. Why edaphic tolerance only sometimes leads to speciation is still unknown. It is
possible that speciation is more likely when an adaptation or a suite of adaptations
comes with a large fitness trade-off between edaphic habitats, contributing to
stronger divergent selection across edaphic boundaries. A greater cost of tolerance
may lead to adaptations that have a direct or indirect effect on reproductive
isolation, leading to the evolution of endemic species. This fundamental gap in
our understanding of edaphic specialization can be answered by using multiple,
progenitor-derivative taxon pairs that have adaptively diverged along the same
edaphic axis but which vary in their accommodation to the substrate (i.e., tolerator
versus endemic). Field surveys to characterize the habitat differences (i.e., are
endemics found on harsher soils than tolerator species?), reciprocal transplant
studies to estimate the fitness trade-offs (i.e., is there a greater cost of tolerance in
species pairs with endemics relative to those with tolerators?), and greenhouse
studies to assess the extent of reproductive isolation between pairs (i.e., is there
greater reproductive isolation in species pairs with endemics relative to those with
tolerators) can reveal answers to this critical question.
e. Edaphic specialists offer opportunities, via methods of comparative genomic and
population genetic analyses, to examine the relationships among genetic loci that
are responsible for adaptations and the degree to which they influence reproduc-
tive isolation (Crawford et al., 2014). In addition to the elegant work done on
model plants such as Helianthus (Andrew & Rieseberg, 2013), Mimulus (Selby
et al., 2014), and Arabidopsis (Turner et al., 2008,2010;Arnoldetal.,2016), the
genetic architecture of traits conferring adaptation and reproductive isolation
should be explored in other species with good ecological and biological data
supporting edaphic differentiation. Serpentine-tolerant plants such as Silene
(Bratteler et al., 2006,b), Lasthenia (Rajakaruna, 2003), Layia (Baldwin, 2005),
Leptosiphon (Kay et al., 2011), Streptanthus (Pope et al., 2014), and Collinsia
(Moyle et al., 2012); metal-tolerant plants including Alyssum,Caulanthus,and
Thalspi (Noccaea)(Mengonietal.,2003a,b; Burrell et al., 2012;Gilletal.,
2012); selenium-tolerant plants such as Stanleya and Astragalus (Schiavon &
N. Rajakaruna

Pilon-Smits, 2017); and gypsophiles such as Mentzelia,Nama, and
Helianthemum (Moore et al., 2014 and references therein) are ideal for testing
how findings from the model plants may be applicable across a wider range of
wild plants experiencing a broader array of edaphic pressures. Taxa within these
genera show a range of divergence stages and degrees of edaphic endemism and,
in many cases, their natural history and ecology are well documented via both
descriptive and experimental methods. For some of these taxa (or their close
relatives), there are extensive genetic tools already in place. Genome scans, when
combined with candidate gene approaches, comparative transcriptomics, QTL
mapping, functional analysis, and experimental manipulation, can reveal loci with
high levels of differentiation between edaphically divergent populations, enabling
the discovery of genes driving adaptation and reproductive isolation (Selby et al.,
2014; Faria et al., 2014; Hoban et al., 2016). In species for which there is evidence
for parallel evolution of traits, it will be critical to test whether the same or
different alleles are responsible for the parallel speciation events (Ostevik et al.,
2012; Roda et al., 2013).
f. Edaphic specialists can help determine how prezygotic and postzygotic reproduc-
tive isolating mechanisms limit gene flow between close relatives at various
geographical scales. The often widely disjunct and patchy distribution patterns
of edaphic specialists provide settings in which to test whether pre-zygotic
isolating mechanisms are stronger in populations in parapatry/sympatry relative
to those in allopatry. Enhanced reproductive isolation between edaphically diver-
gent taxa found under parapatry/sympatry, relative to those in allopatry, may
reflect the action of reinforcement, i.e., selection for reduced gene flow to avoid
maladaptive hybridization (Widmer et al., 2009). There are only a few well-
documented cases of reinforcement in plants (Baack et al., 2015) and edaphic
specialists may provide good model systems for testing the importance of this
process in plant speciation (see Silvertown et al., 2005).
g. Comparative genomic and population genetic analyses can reveal how habitat
specialization is achieved and maintained in the face of gene flow (Papadopulos
et al., 2011). Hybridization and polyploidization are ubiquitous forces in the
diversification of plants, and edaphic endemic species can help reveal how these
two forces influence divergence (Kane, 2009).
h. Chromosomal rearrangements can contribute to reproductive isolation through
their effects on hybrid fertility, as well as by reducing interspecific recombination
and gene flow (Baack et al., 2015). In Mimulus, a chromosomal inversion
polymorphism contributes to differences in flowering time and other traits, as
well as multiple reproductive barriers between annual and perennial ecotypes of
M. guttatus (Lowry & Willis, 2010). Whether karyological anomalies drive
isolation between edaphically divergent individuals must be tested on species
pairs showing post-mating and post-zygotic isolation.
i. If soil elements, like heavy metals, have a direct impact on reproductive isolation
of edaphically divergent taxa, it is easy to demonstrate that adaptation to a specific
soil factor can directly contribute to isolation and subsequent divergence. The
work exploring how heavy metal accumulation in reproductive tissue can influ-
ence reproductive success (Searcy & MacNair, 1993; Quinn et al., 2011a), includ-
ing pollinator-mediated isolation (Meindl & Ashman, 2013,2014,2015), is
Lessons on evolution from the study of edaphic specialization

important for further exploration, especially in sister taxa in which one taxon is a
metal hyperaccumulator.
j. Metal hyperaccumulators have received much attention as candidates for
green technologies such as phytoremediation and phytomining (Rascio &
Navari-Izzo, 2011). However, they also provide opportunities for exploring
the cost of edaphic tolerance hypothesis on habitat specialization. In genera
with metal-tolerant, metal-hyperaccumulating, and metal-intolerant species
(such as in Alyssum,Thlaspi/Noccaea,Streptanthus,andSilene), sister taxa
can be used to test whether there is a trade-off in terms of competitive
ability between metal-intolerant versus metal-tolerant and metal-intolerant
versus metal hyperaccumulator sisters (prediction: greater trade-off in the
metal-intolerant versus metal-hyperaccumulator pair). Further, to explore if
the cost of tolerance to edaphic specialization drives speciation, one could
test whether there is a greater trade-off between metal-intolerant versus
edaphic endemic metal hyperaccumulator pairs or metal-intolerant versus
nonendemic metal hyperaccumulator pairs (prediction: greater trade-off in
competitive ability in the metal intolerant versus edaphic endemic metal
hyperaccumulator pair).
k. Plants found in stressful environments tend to share a suite of traits (i.e.,
stress resistance syndrome or SRS traits) that provides cross-tolerance to a
range of low productivity habitats (Brady et al., 2005; von Wettberg et al.,
2014 and references therein). It is unclear if the same SRS loci or different
loci are responsible for edaphic shifts seen across distinct substrates (e.g.,
between serpentine and limestone, serpentine and saline, serpentine and mine
tailing, limestone and dolomite/gypsum, etc.) characterized by overlapping
selection pressures (e.g., drought, pH, ionic strength, ions, etc.). If SRS traits
are shared (with or without a similar genetic basis), one can envision how an
adaptation to one ‘harsh’soil type can preadapt a lineage to another ‘harsh’
soil type. In Mimulus,Arabidopsis,Helianthus, and others such as Lasthenia
(Rajakaruna, 2003), Silene (Mengoni et al., 2001), and Minuartia (Moore &
Kadereit, 2013)therearemultiplespeciesshowing adaptation to numerous
soil types, including those characterized by similar selection pressures: ions,
pH, drought, etc. Ecological, phylogenomic, and population genomic studies
of plants showing such cross-tolerance to distinct soil types can provide
insights on common physiological traits and trends associated with edaphic
specialization.
l. Climate change is clearly a major stressor influencing plant diversity. How
edaphic specialists will respond to climate change may depend on whether they
are preadapted to future climate conditions, cross-tolerant to edaphic-climatic
stressors (e.g., salt-drought), can compete under novel climates, or can rapidly
evolve climate tolerance, as well as how quickly, how far, and how effectively
they can disperse into suitable edaphic niches under a changing climate
(Fernandez-Going, 2014). Further, how cross kingdom interactions (both mutual-
istic and antagonistic) might play into these possible outcomes is unknown.
Studies of the interaction between climate and edaphic factors in determining
the distribution of edaphic specialists should incorporate species from multiple
lineages, growth forms, substrate types, and varying climates to gain a better
N. Rajakaruna

understanding of how climate-associated threats will impact edaphically special-
ized plants.
The study of edaphic specialization continues to provide unique perspectives on the
processes of evolution and will continue to lead the way in understanding the important
role ecology plays in the origin of species and assembly of special plant communities.
The fast-growing field of molecular ecology will undoubtedly play a key role in this
regard, especially by aiding in the discovery of genes responsible for adaptation and
reproductive isolation under edaphic influences. Edaphically specialized plants will
also provide model systems for exploring little-studied phenomena in plant evolution
such as the importance of reinforcement of reproductive isolation in speciation and the
nature of parallel speciation. Lastly, research on edaphic specialization, particularly the
study of adaptation to edaphic factors such as nutrient levels, salt, heavy metals, and
drought, will help counter growing challenges faced in fields such as agriculture
(Rozema & Schat, 2013) and habitat restoration (O’Dell, 2014), and contribute to
advancing green technologies, including phytoremediation and phytomining (Chaney
et al., 2014).
Acknowledgments I would like to thank Bob Boyd, Ian Medeiros, Elizabeth Farnsworth, Susan Harrison,
Tanner Harris, and Jonathan Gressel for constructive comments on earlier drafts of the manuscript. Additional
comments from Mark R. Macnair and an anonymous reviewer greatly improved the manuscript. Funding from
the US-Sri Lanka Fulbright Commission during the writing of this review is gratefully acknowledged.
Literature Cited
Abbott, R. J., M. J. Hegarty, S. J. Hiscock & A. C. Brennan. 2010. Homoploid hybrid speciation in action.
Tax on 59: 1 3 75 –1386.
———,D.Albach,S.Ansell,J.W.Arntzen,S.J.E.Baird,N.Bierne,J.Boughman,A.Brelsford,C.A.
Buerkle, R. Buggs, R. K. Butlin, U. Dieckmann, F. Eroukhmanoff, A. Grill, S. H. Cahan, J. S.
Hermansen, G. Hewitt, A. G. Hudson, C. Jiggins, J. Jones, B. Keller, T. Marczewski, J. Mallet, P.
Martinez-Rodriguez, M. Möst, S. Mullen, R. Nichols, A. W. Nolte, C. Parisod, K. Pfennig, A. M.
Rice, M. G. Ritchie, B. Seifert, C. M. Smadja, R. Stelkens, J. M. Szymura, R. Väinölä, J. B. W. Wolf
&D.Zinner.2013. Hybridization and speciation. Journal of Evolutionary Biology 26: 229–246.
Al-Hiyali, S. A. K., T. McNeilly, A. D. Bradshaw & A. M. Mortimer. 1993. The effect of zinc contam-
ination from electricity pylons. Genetic constraints on selection for zinc tolerance. Heredity 70: 22–32.
Anacker, B. L. 2011. Phylogenetic patterns of endemism and diversity. Pp. 49–79. In:S.P.Harrison&N.
Rajakaruna (eds.), Serpentine: the evolution and ecology of a model system. University of California
Press, Berkeley.
———, J. Whittall, E. Goldberg & S. P. Harrison. 2011. Origins and consequences of serpentine
endemism in the California flora. Evolution 65: 365–376.
——— &S.P.Harrison.2012. Climate and the evolution of serpentine endemism in California.
Evolutionary Ecology 26: 1011–1023.
——— 2014. The nature of serpentine endemism. American Journal of Botany 101: 219–224.
——— & S. Y. Strauss. 2014. The geography and ecology of plant speciation: Range overlap and niche
divergence in sister species. Proceedings of the Royal Society B. 281: 20132980.
Andrew, R., K. L. Ostevik, D. P. Ebert & L. H. Rieseberg. 2012. Adaptation with gene flow across the
landscape in a dune sunflower. Molecular Ecology 21: 2078–2091.
Andrew,R.L.&L.H.Rieseberg.2013. Divergence is focused on few genomic regions early in speciation:
incipient speciation of sunflower ecotypes. Evolution 67: 2468–2482.
Lessons on evolution from the study of edaphic specialization

Antonovics, J. 1968. Evolution in closely adjacent plant populations V. Evolution of self-fertility. Heredity 23:
219–238.
——— 1976. The nature of limits to natural selection. Annals of the Missouri Botanical Gardens 63: 224–247.
——— 2006. Evolution in closely adjacent plant populations X: long-term persistence of prereproductive
isolation at a mine boundary. Heredity 97: 33–37.
Armbruster, W. S. 2014. Multiple origins of serpentine-soil endemism explained by preexisting tolerance of
open habitats. Proceedings of the National Academy of Sciences of the United States of America 111:
14968–14969.
Arnold, B. J., B. Lahner, J. M. DaCosta, C. M. Weisman, J. D. Hollister, D. E. Salt, K. Bomblies & L.
Ya nt . 2016. Borrowed alleles and convergence in serpentine adaptation. Proceedings of the National
Academy of Sciences of the United States of America 113: 8320–8325.
Baack, E., M. C. Melo, L. H. Rieseberg & D. Ortiz-Barrientos. 2015. The origins of reproductive isolation
in plants. New Phytologist 207: 968–984.
Baker, H. G. 1965. Characteristics and modes of origins of weeds. Pp. 147–172. In: H. G. Baker & G. L.
Stebbins (eds.), The genetics of colonizing species. Academic Press, New York.
Baldwin, B. G. 1999. New combinations and new genera in the North American tarweeds (Compositae-
Madiinae). Novon 9: 462–471.
——— 2005. Origin of the serpentine-endemic herb Layia discoidea from the widespread L. glandulosa
(Compositae). Evolution 59: 2473–2479.
——— 2014. Origins of plant diversity in the California Floristic Province. Annual Review of Ecology
Evolution and Systematics 45: 347–369.
Barraclough, T. G. & A. P. Vogler. 2000. Detecting the geographical pattern of speciation from species-level
phylogenies. American Naturalist 155: 419–434.
Bastida, J. M., P. J. Rey & J. M. Alcántara. 2014. Plant performance and morpho-functional differentiation
in response to edaphic variation in Iberian columbines: cues for range distribution? Journal of Plant
Ecology-UK 7: 403–412.
Baumbach, H. & F. H. Hellwig. 2007. Genetic differentiation of metallicolous and non-metallicolous
Armeria maritima (Mill.) Willd. taxa (Plumbaginaceae) in Central Europe. Plant Systematics and
Evolution 269: 245–258.
Bell, G. 2012. Evolutionary rescue and the limits of adaptation. Philosophical Transactions of the Royal
Society London B Biological Sciences 368: 20120080.
Bennett, T. H., T. J. Flowers & L. Bromham. 2013. Repeated evolution of salt-tolerance in grasses.
Biological Letters 9: 20130029.
Bolukbasi, A., L. Kurt & S. Palacio. 2016. Unravelling the mechanisms for plant survival on gypsum soils:
an analysis of the chemical composition of gypsum plants from Turkey. Plant Biology 18: 271–279.
Borchsenius, F., T. Lozada & J. T. Knudsen. 2016. Reproductive isolation of sympatric forms of the
understorey palm Geonoma macrostachys in western Amazonia. Botanical Journal of the Linnean Society
182: 398–410.
Boyd, R. S. 2007. The defense hypothesis of elemental hyperaccumulation: status, challenges and new
directions. Plant and Soil 293: 153–176.
——— & M. A. Wall. 2001. Responses of generalist predators fed high-Ni Melanotrichus boydi (Heteroptera:
Miridae): Elemental defense against the third trophic level. American Midland Naturalist 146: 186–198.
——— 2014. Ecology and evolution of metal-hyperaccumulator plants. Pp. 227–241. In: N. Rajakaruna, R. S.
Boyd & T. B. Harris (eds.), Plant ecology and evolution in harsh environments. Nova Science Publishers,
Hauppauge, New York.
Bradshaw, A. D. 1991. The Croonian Lecture, 1991. Genostasis and the limits to evolution. Philosophical
Transactions of the Royal Society London B Biological Sciences 333: 289–305.
——— 2006. Unravelling phenotypic plasticity–why should we bother? New Phytologist 170: 644–648.
Bradshaw, H. D. 2005. Mutations in CAX1 produce phenotypes characteristic of plants tolerant to serpentine
soils. New Phytologist 167: 81–88.
Brady, K. U., A. R. Kruckeberg, H. D. Bradshaw JR. 2005. Evolutionary ecology of plant adaptation to
serpentine soils. Annual Review of Ecology Evolution and Systematics 36: 243–266.
Bratteler, M., M. Baltisberger & A. Widmer. 2006. QTL analysis of intraspecific differences between two
Silene vulgaris ecotypes. Annals of Botany 98:411–419.
———,C.Lexer&A.Widmer.2006. Genetic architecture of traits associated with serpentine adaptation of
Silene vulgaris. Journal of Evolutionary Biology 19: 1149–1156.
Bresowar, G. E. & M. E. McGlaughlin. 2015. Morphological and genetic discrepancies in populations of
Oreocarya par adoxa and O. revealii: The impact of edaphic selection on recent diversification in the
Colorado Plateau. American Journal of Botany 102: 1647–1658.
N. Rajakaruna

Brokaw, J. M., T. A. Johnson & C. H. Hofsommer. 2015. Edaphic Specialization in the Cryptic Species
Mentzelia monoensis (Loasaceae). Madroño 62: 88–100.
Burge, D. O. & P. S. Manos. 2011. Edaphic ecology and genetics of the gabbro-endemic shrub Ceanothus
roderickii (Rhamnaceae). Madroño 58: 1–21.
———, R. Hopkins, Y.-H. E. Tsai & P. S. Manos. 2013. Limited hybridization across an edaphic
disjunction between the gabbro-endemic shrub Ceanothus roderickii (Rhamnaceae) and the soil-
generalist Ceanothus cuneatus. American Journal of Botany 100: 1883–1895.
Burrell, A. M., A. K. Hawkins & A. E. Pepper. 2012. Genetic analyses of nickel tolerance in a North
American serpentine endemic plant, Caulanthus amplexicaulis var. barbarae (Brassicaceae). American
Journal of Botany 99: 1875–1883.
Cacho, N. I. & S. Y. Strauss. 2014. Occupation of bare habitats, an evolutionary precursor to soil
specialization in plants. Proceedings of the National Academy of Sciences of the United States of
America 111: 15132–15137.
———, A. M. Burrell, A. E. Pepper & S. Y. Strauss. 2014. Systematics and the evolution of serpentine
tolerance in the California Jewelflowers (Streptanthus) and its allies. Molecular Phylogenetics and
Evolution 72: 71–81.
Chaney, R. L., R. D. Reeves, I. A. Baklanov, T. Centofanti, C. L. Broadhurst, A. J. M. Baker, A. van der
Ent & R. J. Roseberg. 2014. Phytoremediation and phytomining: using plants to remediate contaminated
or mineralized environments. Pp. 365–392. In: N. Rajakaruna, R. S. Boyd, T. B. Harris (eds.), Plant
ecology and evolution in harsh environments. Nova Science Publishers, New York.
Cheeseman, J. M. 2015. The evolution of halophytes, glycophytes and crops, and its implications for food
security under saline conditions. New Phytologist 206: 557–570.
Cheng, N. H., J. K. Pittman, J. B. Bronwyn, T. Shigaki & K. D. Hirchi. 2003. The Arabidopsis cax1
mutant exhibits impaired ion homeostasis, development, and hormonal responses and reveals interplay
among vacuolar transporters. Plant Cell 15: 347–364.
Cheplick,G.P.2015. Approaches to Plant Evolutionary Ecology. Oxford University Press, New York.
Christie, P. & M. R. Macnair. 1984. Complementary lethalfactors in two North-American populations of the
yellow monkey flower. Journal of Heredity 75: 510–511.
——— & M. R. Macnair. 1987 The distribution of postmating reproductive isolating genes in populations of
the yellow monkey flower, Mimulus guttatus. Evolution. 41: 571–578.
Clausen,J.,D.D.Keck&W.M.Hiesey.1940. Experimental studies on the nature of species. I. Effect of varied
environment on Western North American plants. Carnegie Institution of Washington, Washington, DC.
Crawford , D. J. 2010. Progenitor-derivative species pairs and plant speciation. Taxon 59: 1413–1423.
———, R. Ornduff & M. C. Vasey. 1985. Allozyme variation within and between Lasthenia minor and its
derivative species, L. maritima (Asteraceae) 72: 1177–1184.
———, J. J. Doyle, D. E. Soltis, P. S. Soltis & J. F. Wendel. 2014. Contemporary and future studies in plant
speciation, morphological/floral evolution and polyploidy: honouring the scientific contributions of Leslie
D. Gottlieb to plant evolutionary biology. Philosophical Transactions of the Royal Society London B
Biological Sciences 369: 20130341.
Damschen, E. I., S. P. Harrison & J. B. Grace. 2010. Climate change effects on an endemic-rich edaphic
flora: resurveying Robert H. Whittaker’s Siskiyou sites (Oregon, USA). Ecology 91: 3609–3619.
———, S. P. Harrison, D. D. Ackerly, B. Fernandez-Going & B. L. Anacker. 2012. Endemic plants on
serpentine soils: Early victims or hardy survivors of climate change? Journal of Ecology 100: 1122–1130.
De Storme, N. & A. Mason. 2014. Plant speciation through chromosome instability and ploidy change:
Cellular mechanisms, molecular factors and evolutionary relevance. Current Plant Biology 1: 10–33.
Dechamps, C., N. Noret, R. Mozek, J. Escarré, C. Lefèbvre, W. Gruber, P. Meerts. 2008. Cost of
adaptation to a metalliferous environment for Thlaspi caerulescens: a field reciprocal transplantation
approach. New Phytologist 177: 167–177.
Dillenberger, M. S. & J. W. Kadereit. 2013. The phylogeny of the European high mountain genus
Adenostyles (Asteraceae-Senecioneae) reveals that edaphic shifts coincide with dispersal events.
American Journal of Botany 100: 1171–1183.
Donovan, L. A., D. R. Rosenthal, M. Sanchez-Velenosi, L. H. Rieseberg & F. Ludwig. 2010. Are hybrid
species more fit than ancestral parent species in the current hybrid species habitats? Journal of
Evolutionary Biology 23: 805–816.
Douglas, N. A., W. A. Wall, Q.-Y. (J) Xiang, W. A. Hoffmann, T. R. Wentworth, J. B. Gray & M. G.
Hohmann. 2011. Recent vicariance and the origin of the rare, edaphically specialized Sandhills lily,
Lilium pyrophilum (Liliaceae): evidence from phylogenetic and coalescent analyses. Molecular Ecology
20: 2901–2915.
Lessons on evolution from the study of edaphic specialization

Dubois, S., P.-O. Cheptou, C. Petit, P. Meerts, M. Poncelet, X. Vekemans, C. Lefèbvre & J. Escarré.
2003. Genetic structure and mating systems of metallicolous and nonmetallicolous populations of Thlaspi
caerulescens. New Phytologist 157: 633–641.
Dunning,L.T.,H.Hipperson,W.J.Baker,R.K.Butlin,C.Devaux,I.Hutton,J.Igea,A.S.T.
Papadopulos, X. Quan, C. M. Smadja, C. G. N. Turnbull & V. Savolaine. 2016. Ecological speciation
in sympatric palms: 1. Gene expression, selection and pleiotropy. Journal of Evolutionary Biology 29:
1472–1487.
Edelist, C., X. Raffoux, M. Falque, C. Dillmann, D. Sicard, L. H. Rieseberg & S. Karrenberg. 2009. Differential
expression of candidate salt-tolerance genes in the halophyte Helianthus paradoxus and its glycophyte
progenitors H. annuus and H. petiolaris (Asteraceae). American Journal of Botany 96: 1830–1838.
El Mehdawi, A. F., C. F. Quinn & E. A. Pilon-Smits. 2011. Effects of selenium hyperaccumulation on plant-
plant interactions: evidence for elemental allelopathy? New Phytologist 191: 120–131.
——— &E.A.H.Pilon-Smits.2012. Ecological aspects of plant selenium hyperaccumulation. Plant
Biology 14: 1–10.
Ellis, A. G. & A. E. Weis. 2005. Coexistence and differentiation of ‘flowering stones’: the role of local
adaptation to soil microenvironment. Journal of Ecology 94: 322–335.
———, A. E. Weis & B. S. Gaut. 2006. Evolutionary radiation of ‘stone plants’in the genus Argyroderma
(Aizoaceae): unraveling the effects of landscape, habitat and flowering time. Evolution 60: 39–55.
Ellstrand, N. C. 2014. Is gene flow the most important evolutionary force in plants? American Journal of
Botany 101: 737–753.
Ernst, W. H. O. 2006. Evolution of metal tolerance in higher plants. Forest Snow and Landscape Research
80: 251–274.
Escudero, A., S. Palacio, F. T. Maestre & A. L. Luzuriaga. 2015. Plant life on gypsum: a review of its
multiple facets. Biological Reviews 90: 1–18.
Faria, R., S. Renaut, J. Galindo, C. Pinho, J. Melo-Ferreira, M. Melo, F. Jones, W. Salzburger, D.
Schluter & R. Butlin. 2014. Advances in Ecological Speciation: an integrative approach. Molecular
Ecology 23: 513–521.
Feliner, G. N., A. Izuzquiza & A.R. Lansac. 1996. Natural and experimental hybridization in Armeria
(Plumbaginaceae): A. villosa subsp. carratracensis. Plant Systematics and Evolution 201: 163–177.
Fernandez-Going, B. M., B. L. Anacker & S. P. Harrison. 2012. Temporal variability in California grasslands:
Soil type and species functional traits mediate response to precipitation. Ecology 93: 2104–2114.
——— & S. P. Harrison. 2013. Effects of experimental water addition depend on grassland community
characteristics. Plant Ecology 214: 777–786.
——— 2014. Climate change and the future of edaphic floras. Pp. 297–312.In: N. Rajakaruna, R. S. Boyd &
T. B. Harris (eds.), Plant ecology and evolution in harsh environments. Nova Science Publishers,
Hauppauge, New York.
Ferris, K. G., J. P. Sexton & J. H. Willis. 2014. Speciation on a local geographic scale: the evolution of a
rare rock outcrop specialist in Mimulus. Philos. Philosophical Transactions of the Royal Society London
B Biological Sciences 369: 20140001.
———, L. L. Barnett, B. K. Blackman & J. H. Willis. 2016. The genetic architecture of local adaptation
and reproductive isolation in sympatry within the Mimulus guttatus species complex. Molecular Ecology
26: 208–224.
Fine, P. V. A., I. Mesones & P. D. Coley. 2004. Herbivores promote habitat specialization by trees in
Amazonian forests. Science 305: 663–665.
———, Z. J. Miller, I. Mesones, S. Irazuzta, H. M. Appel, M. H. H. Stevens, I. Sääksjärvi, J. C. Schultz
& P. D. Coley. 2006. The growth-defense trade-off and habitat specialization by plants in Amazonian
forests. Ecology 87: S150–S162.
———, F. Zapata, D. C. Daly, I. Mesones, T. M. Misiewicz, H. F. Cooper & C. E. A. Barbosa. 2013. The
importance of environmental heterogeneity and spatial distance in generating phylogeographic structure
in edaphic specialist and generalist tree species of Protium (Burseraceae) across the Amazon Basin.
Journal of Biogeography 40: 646–661.
Fones,H.,C.A.R.Davis,A.Rico,F.Fang,J.A.C.Smith&G.M.Preston.2010. Metal
hyperaccumulation armors plants against disease. PLOS Pathogens 6: e1001093.
Forde, M. B. & D. G. Faris. 1962. Effect of introgression on the serpentine endemism of Quercus durata.
Evolution 16: 338–347.
Freeland, J. R., P. Biss, K. F. Conrad & J. Silvertown. 2010. Selection pressures have caused genome-wide
population differentiation of Anthoxanthum odoratum despite the potential for high gene flow. Journal of
Evolutionary Biology 23: 776–782.
N. Rajakaruna

Friedman, J. & J. H. Willis. 2013. Major QTLs for critical photoperiod and vernalization underlie extensive
variation in flowering in the Mimulus guttatus species complex. New Phytologist 199: 571–583.
Gardner, M. & M. R. Macnair. 2000. Factors affecting the co-existence of the serpentine endemic Mimulus
nudatus Curran and its presumed progenitor, Mimulus guttatus Fischer ex DC. Biological Journal of the
Linnean Society 69: 443–459.
Gil-López, M. J., J. G. Segarra-Moragues & F. Ojeda. 2014. Population genetic structure of a sandstone
specialist and a generalist heath species at two levels of sandstone patchiness across the Strait of Gibraltar.
PLOS One 9: e98602.
Gill, S. S., N. A. Anjum, I. Ahmad, P. Thangavel, G. Sridevi, M. Pacheco, A. C. Duarte, S. Umar, N. A.
Khan & M. E. Pereira. 2012. Metal Hyperaccumulation and Tolerance in Alyssum,Arabidopsis and
Thlaspi: An Overview. Pp. 99–137. In: N.A. Anjum, I. Ahmad, M.E. Pereira, A.C. Duarte, S. Umar, N.A.
Khan (eds.), The plant family Brassicaceae: contribution towards phytoremediation. Springer Science &
Business Media, Dordrecht.
Glassmand, S. I. & B. B. Casper. 2012. Biotic contexts alter metal sequestration and AMF effects on plant
growth in soils polluted with heavy metals. Ecology 93: 1550–1559.
Going, B. M., J. Hillerislambers & J. M. Levine. 2009. Abiotic and biotic resistance to grass invasion in
serpentine annual plant communities. Oecologia 159: 839–847.
Gonzalez-Chavez, C., P. J. Harris, J. Dodd & A. A. Meharg. 2002. Arbuscular mycorrhizal fungi confer
enhanced arsenate resistance on Holcus lanatus. New Phytologist 155: 163–171.
Gottlieb, L. D. 1968. Hybridization between Arctostaphylos viscida and A. canescens in Oregon. Brittonia 20:
83–93.
——— 2004. Rethinking classic examples of recent speciation in plants. New Phytologist 161: 71–82.
Gram, W. L., E. T. Borer, K. L. Cottingham, E. W. Seabloom, V. L. Boucher, L. Goldwasser, F. Micheli,
B. E. Kendall & R. S. Burton. 2004. Distribution of plants in a California serpentine grassland: Are
rocky hummocks spatial refuges for native species? Plant Ecology 172: 159–171.
Gramlich, S., P. Sagmeister, S. Dullinger, F. Hadacek & E. Hörandl. 2016. Evolution in situ: hybrid origin
and establishment of willows (Salix L.) on alpine glacier forefields, Heredity 116: 531–541.
Grossenbacher, D. L., S. D. Veloz & J. P. Sexton. 2014. Niche and range size patterns suggest that speciation
begins in small, ecologically diverged populations in North American monkeyflowers (Mimulus spp.).
Evolution 68: 1270–1280.
Hall, M. C., D. B. Lowry & J. H. Willis. 2010. Is local adaptation in Mimulus guttatus caused by trade-offs at
individual loci? Molecular Ecology 19: 2739–2753.
Harper, F. A., Smith, S. E. & M. R. Macnair. 1997a. Where is the cost in copper tolerance in Mimulus
guttatus? Testing the trade-off hypothesis. Functional Ecology 11: 764–774.
———,Smith,S.E.&M.R.Macnair.1997b. Can an increased copper requirement in copper-tolerant
Mimulus guttatus explain the cost of tolerance? I. Vegetative phase. New Phytologist 136: 455–467
———, Smith, S. E. & M. R. Macnair. 1998. Can an increased copper requirement in copper-tolerant
Mimulus guttatus explain the cost of tolerance? II. Reproductive phase. New Phytologist 140: 637–654.
Harrison,S.P.1999. Native and alien species diversity at the local and regional scales in a Native and alien
species diversity at the local and regional scales in a grazed California grassland. Oecologia 121: 99–106.
——— & N. Rajakaruna (eds.). 2011. Serpentine: the evolution and ecology of a model system. University
of California Press, Berkeley.
———, E. Damschen, B. Fernandez-Going, A. Eskelinen & S. Copeland. 2015. Plant communities on
infertile soils are less sensitive to climate change. Annals of Botany 116: 1017–1022.
Hayward, A. R., K. E. Coates, A. L. Galer, T. C. Hutchinson & R. J. Emery. 2013. Chelator profiling in
Deschampsia cespitosa (L.) Beauv. reveals a Ni reaction, which is distinct from the ABA and cytokinin
associated response to Cd. Plant Physiology and Biochemistry 64: 84–91.
Hego, E., S. Vilain, A. Barré, S. Claverol, J.-W. Dupuy, C. Lalanne, M. Bonneu, C. Plomion, M. Mench.
2016. Copper stress-induced changes in leaf soluble proteome of Cu-sensitive and tolerant Agrostis
capillaris L. populations. Proteomics 16: 1386–1397.
Henry,L.P.,R.H.B.Watson&B.K.Blackman.2014. Transitions in photoperiodic flowering are common and
involve few loci in wild sunflowers (Helianthus; Asteraceae). American Journal of Botany 101: 1748–1758.
Hereford, J. 2009. A quantitative survey of local adaptation and fitness trade-offs. American Naturalist 173:
579–588.
Hipperson,H.,L.T.Dunning,W.J.Baker,R.K.Butlin,I.Hutton,A.S.T.Papadopulos,C.M.Smadja,
T. C. Wilson, C. Devaux & V. Savolainen, V. 2016. Ecological speciation in sympatric palms: 2. Pre-
and post-zygotic isolation, Journal of Evolutionary Biology 29: 2143–2156.
Lessons on evolution from the study of edaphic specialization

Hoban, S., J. L. Kelley, K. E. Lotterhos, M. F. Antolin, G. Bradburd, D. B. Lowry, M. L. Poss, L. K.
Reed, A. Storfer & M. C. Whitlock. 2016. Finding the genomic basis of local adaptation: pitfalls,
practical solutions, and future directions. American Naturalist 188: 379–397.
Hopkins, R. 2013. Reinforcement in plants. New Phytologist 197: 1095–1103.
Hörger, A. C., H. N. Fones & G. M. Preston. 2013. The current status ofthe elemental defense hypothesis in
relation to pathogens. Frontiers in Plant Science 4: 1–11.
Huang, X. & D. E. Salt. 2016. Plant ionomics: from elemental profiling to environmental adaptation.
Molecular Plant 9: 787–797.
Hunter, B. & K. Bomblies. 2010. Progress and promise in using Arabidopsis to study adaptation, divergence,
and speciation. Arabidopsis Book 8: e0138.
Jurjavcic, N. L., S. P. Harrison & A. T. Wolf. 2002. Abiotic stress, competition, and the distribution of the
native annual grass Vulpia microstachys in a mosaic environment. Oecologia 130: 555–562.
Kane, N. C. 2009. Comparative genomic and population genetic analyses indicate highly porous genomes and
high levels of gene flow between divergent Helianthus species. Evolution 63: 2061–2075.
Kaplan, Z. 1998. Relict serpentine populations of Knautia arvensis s. l. (Dipsacaceae) in the Czech Republic
and an adjacent area of Germany. Preslia 70: 21–31.
Kay, K. M., K. L. Ward, L. R. Watt & D.W. Schemske. 2011. Plant Speciation. Pp. 71–96. In:S.P.
Harrison & N. Rajakaruna (eds.), Serpentine: the evolution and ecology of a model system. University of
California Press, Berkeley.
Kazakou, E., P.G. Dimitrakopoulos, A.J. Baker, R.D. Reeves & A.Y. Troumbis. 2008. Hypotheses,
mechanisms and trade-offs of tolerance and adaptation to serpentine soils: from species to ecosystem
level. Biological Reviews 83: 495–508.
Klein, J. T. & J. W. Kadereit. 2015. Phylogeny, biogeography and evolution of edaphic association in the
European oreophytes Sempervivum and Jovibarba (Crassulaceae). International Journal of Plant Sciences
176: 44–71.
Kolář,F.,T.Fér,M.Štech,P.Trávníček, E. Dušková, P. Schönswetter & J. Suda. 2012. Bringing together
evolution on serpentine and polyploidy: spatiotemporal history of the diploid-tetraploid complex of
Knautia arvensis (Dipsacaceae). PLoS ONE 7: e39988.
———,M.Dortová,J.Lepš,M.Pouzar,A.Krejčová & M. Štech. 2014. Serpentine ecotypic differen-
tiation in a polyploid plant complex: shared tolerance to Mg and Ni stress among di- and tetraploid
serpentine populations of Knautia arvensis (Dipsacaceae). Plant and Soil 374: 435–447.
Krahulcová,A.&J.Štěpánková. 1998. Serpentine and polyploid differentiation within Galium pumilum
agg. (Rubiaceae) in Eastern C. Europe. Folia Geobotanica 33: 87–102.
Kruckeberg, A. R. 1951. Intraspecific variability in the response of certain native plant species to serpentine
soil. American Journal of Botany 38: 408–419.
——— 1954. The ecology of serpentine soils. III. Plant species in relation to serpentine soils. Ecology 35:
267–274.
——— 1986. An essay: The stimulus of unusual geologies for plant speciation. Systematic Botany 11: 455–463.
——— 1992. Plant life of western North American ultramafics. Pp. 31–73. In:B.A.Roberts&J.Proctor
(eds.), The ecology of areas with serpentinized rocks. Springer Netherlands. Dordrecht.
Lai, Z., T. Nakazato, M. Salmaso, J. M. Burke, S. Tang, S. J. Knapp & L. H. Rieseberg. 2005. Extensive
chromosomal repatterning and the evolution of sterility barriers in hybrid sunflower species. Genetics.
171: 291–303.
Lau, J. A., A. C. McCall, K. F. Davies, J. K. McKay & J. W. Wright. 2008. Herbivores and edaphic factors
constrain the realized niche of a native plant. Ecology 89: 754–762.
Leimu, R. & M. Fischer. 2008. A meta-analysis of local adaptation in plants. PLoS ONE 3(12): e4010.
Leinonen, P. H., D. L. Remington, J. Leppälä & O. Savolainen. 2013. Genetic basis of local adaptation and
flowering time variation in Arabidopsis lyrata. Molecular Ecology 22: 709–723.
Levin, D. A. 1993. Local speciation in plants: the rule not the exception. Systematic Botany 18: 197–208.
——— 2001. The recurrent origin of plant races and species. Systematic Botany 26: 197–204.
——— 2009. Flowering-time plasticity facilitates niche shifts in adjacent populations. New
Lexer, C., M. E. Welch, J. L. Durphy & L. H. Rieseberg. 2003. Natural selection for salt tolerance
quantitative trait loci (QTLs): Implications for the origin of Helianthus paradoxus, a diploid hybrid
species. Molecular Ecology 12: 1225–1235.
———,Z.Lai&L.H.Rieseberg.2004. Candidate gene polymorphisms associated with salt tolerance in
wild sunflower hybrids: implications for the origin of Helianthus paradoxus, a diploid sunflower hybrid
species. New Phytologist 161: 225–233.
Loarie, S. R., B. E. Carter, K. Hayhoe, S. McMahon, R. Moe, C. A. Knight & D. D. Ackerly. 2008.
Climate Change and the Future of California's Endemic Flora. PLOS One 3: e2502.
N. Rajakaruna

Lowry, D. B., R. C. Rockwood & J. H. Willis. 2008. Ecological reproductive isolation of coast and inland
races of Mimulus guttatus. Evolution 62: 2196–2214.
———,M.C.Hall,D.E.Salt,J.H.Willis.2009. Genetic and physiological basis of adaptive salt tolerance
divergence between coastal and inland Mimulus guttatus. New Phytologist 183: 776–788.
——— &J.H.Willis.2010. A widespread chromosomal inversion polymorphism contributes to a major life
history transition, local adaptation, and reproductive isolation. PLOS Biol. 8: e1000500.
——— 2012. Ecotypes and the controversy over stages in the formation of new species. Biological Journal of
the Linnean Society 106: 241–257.
———,C.C.Sheng,Z.Zhu,T.E.Juenger&B.C.Lahner.2012. Mapping of ionomic traits in Mimulus
guttatus reveals Mo and Cd QTLs that colocalize with MOT1 homologues. PLOS One 7: e30730.
Macnair, M R. 1983. The genetic control of copper tolerance in the yellow monkey flower, Mimulus guttatus.
Heredity 50: 283–293.
——— & P. Christie. 1983. Reproductive isolation as a pleiotropic effect of copper tolerance in Mimulus
guttatus? Heredity 50: 295–302.
———,V.E.Macnair&B.E.Martin.1989. Adaptive speciation in Mimulus: an ecological comparison of
M. cupriphilus with its presumed progenitor, M. guttatus. New Phytologist 112: 269–279.
———, S. E. Smith & Q. J. Cumbes. 1993. Heritability and distribution of variation in degree of copper
tolerance in Mimulus guttatus at Copperopolis, California. Heredity: 71: 445–455.
——— & M. Gardner. 1998. The evolution of edaphic endemics. Pp. 157–171. In: D. J. Howard & S. H.
Berlocher (eds.), Endless forms: species and speciation. Oxford University Press, New York.
Maestri, E., M. Marmiroli, G. Visioli & N. Marmiroli. 2010. Metal tolerance and hyperaccumulation: Costs
and trade-offs between traits and environment. Environmental and Experimental Botany 68: 1–13.
Mandáková, T., V.Singh, U. Krämer & M. A. Lysak. 2015. Genome structure of the heavy metal
hyperaccumulator Noccaea caerulescens and its stability on metalliferous and nonmetalliferous soils.
Plant Physiology 169: 674–689.
Marsden-Jones, E. M. & W. B. Turrill. 1938. Transplant experiments of the British Ecological Society at
Potterne, Wiltshire. Summary of results: 1928–1937. Journal of Ecology 26: 380–389.
Martin, N. H., A. C. Bouck & M. L. Arnold. 2006. Detecting adaptive trait introgression between Iris fulva
and I. brevicaulis in highly selective field conditions. Genetics 172: 2481–2489.
Mason, H. L. 1946a. The edaphic factor in narrow endemism. I. The nature of environmental influences.
Madroño 8: 209–226.
——— 1946b. The edaphic factor in narrow endemism. II. The geographic occurrence of plants of highly
restricted patterns of distribution, Madroño 8: 241–257.
Mayer, M. S. & P. S. Soltis. 1994. The evolution of serpentine endemics: A chloroplast DNA phylogeny of
the Streptanthus glandulosus complex (Cruciferae). Systematic Botany 19: 557–574.
——— &P.S.Soltis.1999. Intraspecific phylogeny analysis using ITS sequences: Insights from studies of
the Streptanthus glandulosus complex (Cruciferae). Systematic Botany 24: 47–61.
Meindl,G.A.&T.-L.Ashman.2013. The effects of aluminum and nickel in nectar on the foraging behavior
of bumblebees. Environmental Pollution 177: 78–81.
——— & T. -L. Ashman. 2014. Nickel accumulation by Streptanthus polygaloides (Brassicaceae) reduces
floral visitation rate. Journal of Chemical Ecology 40: 128–135.
———,D.J.Bain&T.-L.Ashman.2014. Nickel accumulation in leaves, floral organs and rewards varies
by serpentine soil affinity. AoB PLANTS 6: plu036.
———& T. -L. Ashman. 2015. Effects of floral metal accumulation on floral visitor communities:
Introducing the elemental filter hypothesis. American Journal of Botany 102: 379–389.
Melo,M.C.,A.Grealy,B.Brittain,G.M.Walter&D.Ortiz-Barrientos.2014. Strong extrinsic reproductive
isolation between parapatric populations of an Australian groundsel. New Phytologist 203: 323–334.
Mengoni, A., C. Gonnelli, F. Galardi, R. Gabbrielli & M. Bazzicalupo. 2001. Genetic diversity of heavy
metal tolerant populations in Silene paradoxa L. (Caryophyllaceae): a chloroplast microsatellite analysis.
Molecular Ecology 10: 1909–1916.
———, A. J. M. Baker, M. Bazzicalupo, R. D. Reeves, N. Adigüzel, E. Chianni, F. Galardi, R.
Gabbrielli & C. Gonnelli. 2003a. Evolutionary dynamics of nickel hyperaccumulation in Alyssum
revealed by its nrDNA analysis. New Phytologist 159: 691–699.
———, C. Gonnelli, E. Brocchini, F. Galardi, S. Pucci, R. Gabbrielli & M. Bazzicalupo. 2003b.
Chloroplast genetic diversity and biogeography in the serpentine endemic Ni-hyperaccumulator
Alyssum bertolonii. New Phytologist 157: 349–356.
Miglia, K. J. 2007. Genotype, soil type, and locale effects on reciprocal transplant vigor, endophyte growth,
and microbial functional diversity of a narrow sagebrush hybrid zone in Salt Creek Canyon, Utah.
American Journal of Botany 94: 425–436.
Lessons on evolution from the study of edaphic specialization

Milla,R.,A.Escudero&J.M.Iriondo.2011. Congruence between geographic range distribution and local
competitive ability of two Lupinus species. American Journal of Botany 98: 1456–1464.
Moore, A. J. & J. W. Kadereit. 2013. The evolution of substrate differentiation in Minuartia series
Laricifoliae (Caryophyllaceae) in the European Alps: In situ origin or repeated colonization? American
Journal of Botany 100: 2412–2425.
———, D. Merges & J. W. Kadereit. 2013. The origin of the serpentine endemic Minuartia laricifolia
subsp. ophiolitica by vicariance and competitive exclusion, Molecular Ecology 22: 2218–2231.
Moore, K. A. & S. C. Elmendorf. 2011. Plant competition and facilitation in systems with strong
environmental gradients. Pp. 223–236. In: S. P. Harrison & N. Rajakaruna (eds.), Serpentine: the
evolution and ecology of a model system. University of California Press, Berkeley.
Moore, M. J., J. F. Mota, N. A. Douglas, H. F. Olvera & H. Ochoterena. 2014. The ecology, assembly, and
evolution of gypsophile floras. Pp. 97–128. In: N. Rajakaruna, R. S. Boyd & T. B. Harris (eds.), Plant
ecology and evolution in harsh environments. Nova Science Publishers, Hauppauge, New York.
Moray, C., E. W. Goolsby & L. Bromham. 2016. The phylogenetic association between salt tolerance and
heavy metal hyperaccumulation in angiosperms. Evolutionary Biology 43: 119–130.
Mousset, M., P. David, C. Petit, J. Pouzadoux, C. Hatt. E. Flaven, O. Ronce & A. Mignot. 2016. Lower
selfing rates in metallicolous populations than in non-metallicolous populations of the pseudometallophyte
Noccaea caerulescens (Brassicaceae) in Southern France. Annals of Botany 117: 507–519.
Moyle, L.C., M. Levine, M. L. Stanton & J. W. Wright. 2012. Hybrid sterility over tens of meters between
ecotypes adapted to serpentine and non-serpentine soils. Evolutionary Biology 39: 207–218.
Nichols, M.K. & McNeilly, T. 1982. The possible polyphyletic origin of copper tolerance in Agrostis tenuis
(Gramineae). Plant Systematics and Evolution 140: 109–117.
Nosil, P. & D. Schluter. 2011. The genes underlying the process of speciation. Trends in Ecology and
Evolution 26: 160–167.
——— 2012. Ecological Speciation. Oxford University Press, Oxford.
Nyberg Berglund, A. B., S. Dahlgren & A. Westerbergh. 2004. Evidence for parallel evolution and site-
specific selection of serpentine tolerance in Cerastium alpinum during the colonization of Scandinavia.
New Phytologist 161: 199–209.
O’Dell, R. E. & N. Rajakaruna. 2011. Intraspecific variation, plant adaptation, and evolution. Pp. 71–96. In:
S.P. Harrison & N. Rajakaruna (eds.), Serpentine: the evolution and ecology of a model system.
University of California Press, Berkeley.
——— 2014. Conservation and restoration of chemically extreme edaphic endemic flora in the western US.
Pp. 3313–3364. In: N. Rajakaruna, R. S. Boyd, T. B. Harris (eds.), Plant ecology and evolution in harsh
environments. Nova Science Publishers, New York.
Olko, A., A. Abratowska, J. Żyłkowska, M. Wierzbicka & A. Tukiendorf. 2008. Armeria maritima from a
calamine heap—Initial studies on physiologic–metabolic adaptations to metal-enriched soil.
Ecotoxicology and Environmental Safety 69: 209–218.
Ornduff, R. 1965. Ornithocoprophilous endemism in Pacific Basin angiosperms. Ecology 46: 864–867.
Ortiz-Barrientos, D., A. Grealy & P. Nosil. 2009. The genetics and ecology of reinforcement: implications
for the evolution of prezygotic isolation in sympatry and beyond. Annals of the New York Academy of
Sciences 1168: 156–182.
Ostevik, K. L., B. T. Moyers, G. L. Owens & L. H. Rieseberg. 2012. Parallel Ecological Speciation in
Plants? International Journal of Ecology 939862.
———, R. L. Andrew, S. P. Otto & L. H. Rieseberg. 2016. Multiple reproductive barriers separate recently
diverged sunflower ecotypes. Evolution 70: 2322–2335.
Palacio, S., A. Escudero, G. Montserrat-Martí, M. Maestro, R. Milla & M. J. Albert. 2007. Plants living
on gypsum: beyond the specialist model. Annals of Botany 99: 333–343.
———, D. Johnson, A. Escudero & G. Montserrat-Martí. 2012. Root colonisation by AM fungi differs
between gypsum specialist and non-specialist plants: Links to the gypsophile behavior. Journal of Arid
Environments 76: 128–132.
———, J. Azorín, G. Montserrat-Martí & J. P. Ferrio. 2014. The crystallization water of gypsum rocks is
a relevant water source for plants. Nature Communications 5: 4660; doi:https://doi.org/10.1038
/ncomms5660.
Palacio-López, K., B. Beckage, S. Scheiner & J. Molofsky. 2015. The ubiquity of phenotypic plasticity in
plants: a synthesis. Ecology and Evolution 5: 3389–3400.
Palm, E., K. Brady & E. Van Volkenburgh. 2012. Serpentine tolerance in Mimulus guttatus does not rely on
exclusion of magnesium. Functional Plant Biology 39: 679–688.
N. Rajakaruna

——— & E. Van Volkenburgh. 2014. Physiological adaptations of plants to serpentine soil. Pp. 129–148. In:
N. Rajakaruna, R. S. Boyd, T. B. Harris (eds.), Plant ecology and evolution in harsh environments. Nova
Science Publishers, Hauppauge, New York.
Papadopulos, A. S. T., W. J. Baker, D. Crayn, R. K. Butlin, R. G. Kynast, I. Hutton & V. Savolainen.
2011. Speciation with gene flow on Lord Howe Island. Proceedings of the National Academy of Sciences
of the United States of America 108: 13188–13193.
Parejo-Farnés, C., R. G. Albaladejo, J. Arroyo & A. Aparicio. 2013. A phylogenetic hypothesis for
Helianthemum (Cistaceae) in the Iberian Peninsula. Botanica Complutensis 37: 83−92.
Paris, C. A. & M. D. Windham. 1988. A biosystematic investigation of the Adiantum pedatum complex in
eastern North America. Systematic Botany 13: 240–255.
——— 1991. Adiantum viridimontanum, a new maidenhair fern in eastern North America. Rhodora 93: 105–122.
Partridge, L. & R. Sibly. 1991. Constraints in the evolution of life histories. Philosophical Transactions of the
Royal Society London, B 332: 3–13.
Parys, E., W. Wasilewska, M. Siedlecka, M. Zienkiewicz, A. Drożak & E. Romanowska. 2014. Metabolic
responses to lead of metallicolous and nonmetallicolous populations of Armeria maritima. Archives of
Environmental Contamination and Toxicology 67: 565–577.
Paule, J., F. Kolář& C. Dobeš.2015. Arctic-alpine and serpentine differentiation in polyploid Potentilla
crantzii. Preslia 87: 195–215.
Paun, O., B. Turner, E. Trucchi, J. Munzinger, M. W. Chase & R. Samuel. 2016. Processes Driving the
Adaptive Radiation of a Tropical Tree (Diospyros, Ebenaceae) in New Caledonia, a Biodiversity Hotspot.
Systematic Biology 65: 212–227.
Pauwels, M., P. Saumitou-Laprade,A.C. Holl, D. Petit & I. Bonnin. 2005. Multiple origin of metallicolous
populations of the pseudometallophyte Arabidopsis halleri (Brassicaceae) in central Europe: the cpDNA
testimony. Molecular Ecology 14: 4403–4414.
Pepper, A. E. & L. E. Norwood. 2001. Evolution of Caulanthus amplexicaulis var. barbarae (Brassicaceae),
a rare serpentine endemic plant: a molecular phylogenetic perspective. American Journal of Botany 88:
1479–1489.
Peterson, M. L., K. J. Rice & J. P Sexton. 2013. Niche partitioning between close relatives suggests trade-
offs between adaptation to local environments and competition. Ecology and Evolution 3: 512–522.
Pettengill, J. B. & D. A. Moeller. 2012. Phylogeography of speciation: allopatric divergence and secondary
contact between outcrossing and selfing Clarkia. Molecular Ecology 21: 4578–4592.
Pfennig, K. S. & D. W. Pfennig. 2009. Character displacement: ecological and reproductive responses to a
common evolutionary problem. The Quarterly Review of Biology 84: 253–276.
Pope, N., M. Fong, R. S. Boyd & N. Rajakaruna. 2014. The role of elevation and soil chemistry in the
distribution and ion accumulation of floral morphs of Streptanthus polygaloides Gray (Brassicaceae), a
Californian nickel hyperaccumulator. Plant Ecology and Diversity 7: 1–39.
Porter, S. S., P. L. Chang, C. A. Conow, J. P. Dunham, M. L. Friesen. 2017. Association mapping reveals
novel serpentine adaptation gene clusters in a population of symbiotic Mesorhizobium. The ISME Journal
11: 248–262
Powell,K.I.&T.M.Knight.2009. Effects of nutrient addition and competition on biomass of five Cirsium
species (Asteraceae), including a serpentine endemic. International Journal of Plant Sciences 170: 918–925.
Quinn, C. F., J. L. Freeman, R. J. B. Reynolds, J. J. Cappa, S. C. Fakra, M. A. Marcus, S. D. Lindblom,
E. K. Quinn, L. E. Bennett & E. A. H Pilon-Smits. 2010. Selenium hyperaccumulation offers
protection from cell disruptor herbivores. BMC Ecology 10: 1–11.
———,C.N.Prins,J.L.Freeman,A.M.Gross,L.J.Hantzis,R.J.B.Reynolds,S.Yang,P.A.Covey,
G. S. Bañuelos, I. J. Pickering, S. C. Fakra, M. A. Marcus, H. S. Arathi, & E. A. H. Pilon-Smits.
2011a. Selenium accumulation in flowers and its effects on pollination. New Phytologist 192: 727–737.
———, K. A. Wyant, A. L. Wangeline, J. Shulman, M. L. Galeas, J. R. Valdez, J. R. Self, M. W.
Paschke & E. A. H. Pilon-Smits. 2011b. Enhanced decomposition of selenium hyperaccumulator litter
in a seleniferous habitat—evidence for specialist decomposers? Plant and Soil 341: 51–61.
Qureshi, J. A., D. A. Thurman, K. Hardwick, H. A. Collin. 1985. Uptake and accumulation of zinc, lead
and copper in zinc and lead tolerant Anthoxanthum odoratum L. New Phytologist 100: 429–434.
Rajakaruna, N. 2003. Edaphic differentiation in Lasthenia: A model for studies in evolutionary ecology.
Madrono 50: 34–40.
——— 2004. The edaphic factor in the origin of species. International Geology Review 46: 471–478.
———, M. Y. Siddiqi, J. Whitton, B. A. Bohm & A. D. M. Glass. 2003a. Differential responses to Na
+
/K
+
and Ca
2+
/Mg
2+
in two edaphic races of the Lasthenia californica (Asteraceae) complex: A case for
parallel evolution of physiological traits. New Phytologist 157: 93–103.
Lessons on evolution from the study of edaphic specialization

———,G.E.Bradfield,B.A.Bohm&J.Whitton.2003b. Adaptive differentiation in response to water stress
by edaphic races of Lasthenia californica (Asteraceae). International Journal of Plant Sciences 164: 371–376.
———, Baldwin, B.G., R. Chan, A. M. Desrochers, B. A. Bohm, & J. Whitton. 2003c. Edaphic races and
phylogenetic taxa in the Lasthenia californica complex (Asteraceae: Heliantheae): An hypothesis of
parallel evolution. Molecular Ecology 12: 1675–1679.
——— & J. Whitton. 2004. Trends in the evolution of edaphic specialists with an example of parallel
evolution in the Lasthenia californica complex. Pp. 103–11 0. In:Q.C.B.Cronk,J.Whitton,R.Ree&I.
E. P. Taylor (eds.), Plant adaptation: molecular genetics and ecology. NRC Research Press, Ottawa.
———,R.S.Boyd&T.B.Harris.2014. Synthesis and future directions: What have harsh environments taught
us about ecology, evolution, conservation and restoration. Pp. 393–409. In:N.Rajakaruna,R.S.Boyd&T.
B. Harris (eds.), Plant ecology and evolution in harsh environments. Nova Science Publishers, Inc., NY.
Rascio, N. & F. Navari-Izzo. 2011. Heavy metal hyperaccumulating plants: how and why do they do it? And
what makes them so interesting? Plant Science 180: 169–181.
Raven, P. H. 1964. Catastrophic selection and edaphic endemism. Evolution 18: 336–338.
Renaut, S., C. J. Grassa, S. Yeaman, Z. Lai, N. K. Kane, B. T. Moyers, J. E. Bowers, J. M. Burke & L.
H. Rieseberg. 2013. Genomic islands of divergence are not affected by geography of speciation in
sunflowers. Nature Communications 1827: doi:https://doi.org/10.1038/ncomms2833.
Rice, K. 1989. Competitive interactions in California annual grasslands. Pp. 59–72. In:L.F.Huenneke&H.
A. Mooney (eds.), Grassland structure and function: California annual grassland. Kluwer, Dordrecht.
Roda, F., L. Ambrose, G. M. Walter, G,. H. L. Liu, L. Schaul, A. Lowe, P. B. Pelser, P. Prentis, L. H.
Rieseberg & D. Ortiz-Barrientos. 2013. Genomic evidence for the parallel evolution of coastal forms in
the Senecio lautus complex. Molecular Ecology 22: 2941–2952.
Roux, C., V. Castric, M. Pauwels, S. I. Wright, P. Saumitou-Laprade & X. Vekemans. 2011. Does
speciation between Arabidopsis halleri and Arabidopsis lyrata coincide with major changes in a
molecular target of adaptation? PLoS ONE 6: e26872.
Roy, S. J., E. J. Tucker & M. Tester. 2011. Genetic analysis of abiotic stress tolerance in crops. Current
Opinion in Plant Biology 14: 232–239.
Rozema, J. & H. Schat. 2013. Salt tolerance of halophytes, research questions reviewed in the perspective of
saline agriculture. Environmental and Experimental Botany 92: 83–95.
Sambatti, J. B. & K. J. Rice. 2007. Functional ecology of ecotypic differentiation in the Californian
serpentine sunflower (Helianthus exilis). New Phytologist 175: 107–119.
Sánchez, A. M., P. Alonso-Valiente, M. J. Albert & A. Escudero. 2017. How might edaphic specialists in
gypsum islands respond to climate change? Reciprocal sowing experiment to infer local adaptation and
phenotypic plasticity. Annals of Botany 120: 135–146.
Saslis-Lagoudakis, C. H., C. Moray & L. Bromham. 2014. Evolution of salt tolerance in angiosperms: a
phylogenetic approach. Pp. 77–95. In: N. Rajakaruna, R. S. Boyd & T. B. Harris (eds.), Plant ecology and
evolution in harsh environments. Nova Science Publishers, Hauppauge, New York.
Savolainen, V., M.-C. Anstett, C. Lexer, I. Hutton, J. J. Clarkson, M. V. Norup, M. P. Powell, D.
Springate, N. Salamin & W. J.Baker. 2006. Sympatric speciation in palms on an oceanic island. Nature
441: 210–213.
Schechter, S. P. & T. D. Bruns. 2008. Serpentine and non-serpentine ecotypes of Collinsia sparsiflora
associate with distinct arbuscular mycorrhizal fungal assemblages. Molecular Ecology 17: 3198–3210.
——— &T.D.Bruns.2013. A common garden test of host-symbiont specificity supports a dominant role
for soil type in determining AMF assemblage structure in Collinsia sparsiflora. PLOS One 8: e55507.
——— & S. Branco. 2014. The ecology and evolution of mycorrhizal fungi in extreme soils. Pp. 33–52. In:
N. Rajakaruna, R. S. Boyd & T. B. Harris (eds.), Plant ecology and evolution in harsh environments.
Nova Science Publishers, New York.
Schemske, D. W. 2010. Adaptation and the origin of species. American Naturalist 176: S4–S25.
Schiavon, M. & E. A. H. Pilon-Smits. 2017. The fascinating facets of plant selenium accumulation –
biochemistry, physiology, evolution and ecology. New Phytologist 213: 1582–1596.
Schmickl, R. & M. A. Koch. 2011. Arabidopsis hybrid speciation processes. Proceedings of the National
Academy of Sciences of the United States of America 108: 14192–14197.
Schneider, A. C., W. A. Freyman, C. M. Guilliams, Y. P. Springer & B. G. Baldwin. 2016. Pleistocene
radiation of the serpentine-adapted genus Hesperolinon and other divergence times in Linaceae
(Malpighiales). American Journal of Botany 103: 221–232.
Searcy, K. B. & D. L. Mulcahy. 1985. Pollen selection and the gametophytic expression of metal tolerance in
Silene dioica (caryophyllaceae) and Mimulus guttatus (Scrophulariaceae). American Journal of Botany
72: 1700–1706.
N. Rajakaruna

——— &M.R.Macnair.1990. Differential seed production in Mimulus guttatus Fisch, ex DC in response
to increasing concentrations of copper in the pistil by pollen from copper tolerant and sensitive sources.
Evolution 44: 1424–1435.
——— & M. R. Macnair. 1993. Developmental selection in response to environmental conditions of the
maternal parent in Mimulus guttatus. Evolution 47: 13–24.
Selby, J. P., A. L. Jeong, K. Toll, K. M. Wright, D. B. Lowry. 2014. Methods and discoveries in the pursuit
of understanding the genetic basis of adaptation to harsh environments in Mimulus. Pp. 243–265. In:N.
Rajakaruna, R. S. Boyd & T. B. Harris (eds.), Plant ecology and evolution in harsh environments. Nova
Science Publishers, New York.
Shimizu-Inatsugi, R., A. Terada, K. Hirose, H. Kudoh, J. Sese & K. K. Shimizu. 2016. Plant adaptive
radiation mediated by polyploid plasticity in transcriptomes. Molecular Ecology 26: 193–207.
Silvertown, J., C. Servaes, P. Biss & D. Macleod. 2005. Reinforcement of reproductive isolation between
adjacent populations in the Park Grass Experiment. Heredity 95: 198–205.
———, P. Poulton, E. Johnston, G. Edwards, M. Heard & P. M. Biss. 2006. The Park Grass Experiment
1856–2006: its contribution to ecology. Journal of Ecology 94: 801–814.
Slovák, M., J. Kučera, P. Turis & J. Zozomová-Lihová. 2012. Multiple glacial refugia and postglacial
colonization routes inferred for a woodland geophyte, Cyclamen purpurascens: patterns concordant with
the Pleistocene history of broadleaved and coniferous tree species. Biological Journal of the Linnean
Society: 105: 741–760.
Smith, S. E. & M. R. Macnair. 1998. Hypostatic modifiers cause variation in degree of copper tolerance in
Mimulus guttatus. Heredity 80: 760–768.
Sobel, J. M. 2014. Ecogeographic isolation and speciation in the genus Mimulus. American Naturalist 184:
565–579.
Sobczyk, M. K., J. A. C. Smith, A. J. Pollard & D. A. Filatov. 2017. Evolution of nickel hyperaccumulation
and serpentine adaptation in the Alyssum serpyllifolium species complex. Heredity 118: 31–41.
Soltis, D. E., P. S. Soltis & J. A. Tate. 2004. Advances in the study of polyploidy since plant speciation. New
Phytologist 161: 173–191.
Southworth, D., L. E. Tackaberry & H. B. Massicotte. 2014. Mycorrhizal ecology on serpentine soils. Plant
Ecology and Diversity 7: 445–455.
Spasojevic, M. J., E. I. Damschen & S. P. Harrison. 2014. Patterns of seed dispersal syndromes on
serpentine soils: examining the roles of habitat patchiness, soil infertility and correlated functional traits.
Plant Ecology and Diversity 7: 401–410.
Springer, Y. P. 2007. Clinal resistance structure and pathogen local adaptation in a serpentine flax-flax rust
interaction. Evolution 61: 1812–1822.
——— 2009. Edaphic quality and plant-pathogen interactions: Effects of soil calcium on fungal infection of a
serpentine flax. Ecology 90: 1852–1862.
Strasburg, J. L., N. A. Sherman, K. M. Wright, L. C. Moyle, J. H. Willis & L. H. Rieseberg. 2012. What
can patterns of differentiation across plant genomes tell us about adaptation and speciation? Philosophical
Transactions of the Royal Society London B Biological Sciences 367: 364–373.
Strauss, S. Y. & R. S. Boyd. 2011. Herbivory and other cross-kingdom interactions on harsh soils. Pp. 181–
199. In: S. P. Harrison & N. Rajakaruna (eds.), Serpentine: the evolution and ecology of a model system.
University of California Press, Berkeley.
——— &N.I.Cacho.2013. Nowhere to run, nowhere to hide: The importance of enemies and apparency in
adaptation to harsh soil environments. American Naturalist 182: E1–E14.
Sun, G. & P. Schliekelman. 2011. A genetical genomics approach to genome scans increases power for QTL
mapping, Genetics 187: 939–953.
Thrall, P. H., M. E. Hochberg, J. J. Burdon & J. D. Bever. 2007. Coevolution of symbiotic mutualists and
parasites in a community context. Trends in Ecology & Evolution 22: 120–126.
———,J.D.Bever&J.F.Slattery.2008. Rhizobial mediation of Acacia adaptation to soil salinity:
evidence of underlying trade-offs and tests of expected patterns. Journal of Ecology 96: 746–755.
Tiffin, P. & J Ross-Ibarra. 2014. Advances and limits of using population genetics to understand local
adaptation. Trends in Ecology & Evolution 29: 673–680.
Turesson, G. 1922. The species and the variety as ecological units. Hereditas 3: 100–113.
Turner, T. L., E. J. von Wettberg, S. V. Nuzhdin. 2008. Genomic analysis of differentiation between soil
types reveals candidate genes for local adaptation in Arabidopsis lyrata. PLOS One 3: e3183.
———, E. C. Bourne, E. J. Von Wettberg, T. T. Hu, S. V. Nuzhdin. 2010. Population resequencing reveals
local ada ptation of Arabidopsis lyrata to serpentine soils. Nature Genetics 42: 260–263.
Twyford, A.D. & J. Friedman. 2015. Adaptive divergence in the monkey flower Mimulus guttatus is
maintained by a chromosomal inversion. Evolution 69: 1476–1486.
Lessons on evolution from the study of edaphic specialization

Valladares, F., S. Matesanz, F. Guilhaumon, M. B. Araújo, L. Balaguer, M. Benito-Garzón, W.
Cornwell,E.Gianoli,M.vanKleunen,D.E.Naya,A.B.Nicotra,H.Poorter&M.A.Zavala.
2014. The effects of phenotypic plasticity and local adaptation on forecasts of species range shifts under
climate change. Ecology Letters 17: 1351–1364.
Vallano, D. M., P. C. Selmants & E. S. Zavaleta. 2012. Simulated nitrogen deposition enhances the
performance of an exotic grass relative to native serpentine grassland competitors. Plant Ecology 213:
1015–1026.
Vamosi, J. C., W. S. Armbruster & S. S. Renner. 2014. Evolutionary ecology of specialization: insights
from phylogenetic analysis. Proceedings of the Royal Society B: Biological Sciences 281: 20142004.
van der Niet, T. & S. D. Johnson. 2009. Patterns of plant speciation in the Cape floristic region, Molecular
Phylogenetics and Evolution 51: 85–93.
van Hoof, N. A. L. M., V. H. Hassinen, H. W. J. Hakvoort, K. F. Ballintijn, H. Schat, J. A.C. Verkleij, W.
H.O. Ernst, S. O. Karenlampi & A. I. Tervahauta. 2001. Enhanced copper tolerance in Silene vulgaris
(Moench) Garcke populations from copper mines is associated with increased transcript levels of a 2b-
type metallothionein gene. Plant Physiology 126: 1519–1526.
Van Nuland, M. E. 2016. Plant–soil feedbacks: connecting ecosystem ecology and evolution. Functional
Ecology 30: 1032–1042.
Van Za n d t , P. A . 2007. Plant defense, growth, and habitat: a comparative assessment of constitutive and
induced resistance. Ecology 88: 1984–1993.
Vekemans, X. & C. Lefèbvre. 1997. On the evolution of heavy-metal tolerant populations in Armeria
maritima: evidence from allozyme variation and reproductive barriers. Journal of Evolutionary Biology
10: 175–191.
Venables,A.V.&D.A.Wilkins.1978. Salt tolerance in pasture grasses. New Phytologist 80: 613–622.
Verbruggen, N., C. Hermans & H. Schat. 2009. Molecular mechanisms of metal hyperaccumulation in
plants. New Phytologist 181: 759–776.
Via, S. 2009. Natural selection in action during speciation. Proceedings of the National Academy of Sciences
of the United States of America 106: 9939–9946.
von Wettberg, E. J. & J. W. Wright. 2011. Genomic approaches to understanding adaptation. Pp. 139–153.
In: S.P. Harrison & N. Rajakaruna (eds.), Serpentine: the evolution and ecology of a model system.
University of California Press, Berkeley.
———,J.Ray-Mukherjee,N.D’Adesky, D. Nesbeth & S. Sistla. 2014. The evolutionary ecology and
genetics of stress resistance syndrome (SRS) traits: revisiting Chapin, Autumn and Pugnaire (1993).
Pp.201–226.In: N. Rajakaruna, R. S. Boyd & T. B. Harris (eds.), Plant ecology and evolution in harsh
environments. Nova Science Publishers, New York.
Wallace, D. R. 1983. The Klamath Knot. University of California Press, Berkeley.
Westerbergh, A. & A. Saura. 1992. The effect of serpentine on the population structure of Silene dioica
(Caryophyllaceae). Evolution 46: 1537–1548.
Widmer, A., C. Lexer & S. Cozzolino. 2009. Evolution of reproductive isolation in plants. Heredity 102: 31–38.
Williams,J.L.,R.E.Snyder&J.M.Levine.2016. The influence of evolution on population spread
through patchy landscapes. American Naturalist 188: 15–26.
Wright,J.W.&M.L.Stanton.2007. Collinsia sparsiflora in serpentine and nonserpentine habitats: Using F2
hybrids to detect the potential role of selection in ecotypic differentiation. New Phytologist 173: 354–366.
——— &M. L. Stanton. 2011. Local adaptation in heterogeneous landscapes–reciprocal transplant exper-
iments and beyond. Pp. 155–180. In: S.P. Harrison & N. Rajakaruna (eds.), Serpentine: the evolution and
ecology of a model system. University of California Press, Berkeley.
Wright, K. M., D. Lloyd, D. B. Lowry, M. R. Macnair & J.H. Willis. 2013. Indirect evolution of hybrid
lethality due to linkage with selected locus in Mimulus guttatus. PLOS Biology 11: e1001497.
Wu, C. A., D. B. Lowry, L. I. Nutter & J. H. Willis. 2010. Natural variation for drought-response traits in
the Mimulus guttatus species complex. Oecologia, 162: 23–33.
Wu, L., A. D. Bradshaw & D. A. Thurman. 1975. The potential for evolution of heavy metal tolerance in
plants. I. The rapid evolution of copper tolerance in Agrostis stolonifera. Heredity 34: 165–178.
Yakimowski, S.B. & L.H. Rieseberg. 2014. The role of homoploid hybridization in evolution: A century of
studies synthesizing genetics and ecology. American Journal of Botany 101: 1247–1258.
Yost, J. M., T. M. Barry, K. M. Kay & N. Rajakaruna. 2012. Edaphic adaptation maintains the coexistence
of two cryptic species on serpentine soils. American Journal of Botany 99: 890–897.
N. Rajakaruna