Selection and inertia in the evolution of holocentric chromosomes in sedges (Carex, Cyperaceae)

Selection and inertia in the evolution of holocentricchromosomes in sedges (Carex, Cyperaceae)

Marcial Escudero1,2, Andrew L. Hipp2,3, Thomas F. Hansen4, Kjetil L. Voje4 and Modesto Luceno1

1Department of Molecular Biology and Biochemical Engineering, Pablo de Olavide University, Ctra. Utrera km 1, 41013-Seville, Spain; 2The Morton Arboretum, 4100 Illinois Route 53, Lisle,

IL 60532-1293, USA; 3Department of Botany, The Field Museum, 1400 S Lake Shore Dr, Chicago, IL 60605, USA; 4Centre for Ecological and Evolutionary Synthesis, Department of Biology,

University of Oslo, PB 1066, Blindern, 0316, Oslo, Norway

Author for correspondence:Marcial Escudero

Tel: +1 630 373 1979Email: [email protected]

Received: 31 January 2012

Accepted: 24 February 2012

New Phytologist (2012)doi: 10.1111/j.1469-8137.2012.04137.x

Key words: adaptation, Cariceae,chromosome number, holocentricchromosomes, Ornstein–Uhlenbeck model,recombination rate.

Summary

• Changes in chromosome number as a result of fission and fusion in holocentrics have direct

and immediate effects on the recombination rate. We investigate the support for the classic

hypothesis that environmental stability selects for increased recombination rates.

• We employed a phylogenetic and cytogenetic data set from one of the most diverse angio-

sperm genera in the world, which has the largest nonpolyploid chromosome radiation (Carex,

Cyperaceae; 2n = 12–124; 2100 spp.). We evaluated alternative Ornstein–Uhlenbeck models

of chromosome number adaptation to the environment in an information-theoretic frame-

work.

• We found moderate support for a positive influence of lateral inflorescence unit size on

chromosome number, which may be selected in a stable environment in which resources for

reproductive investment are larger. We found weak support for a positive influence on

chromosome number of water-saturated soils and among-month temperature constancy,

which would be expected to be negatively select for pioneering species. Chromosome number

showed a strong phylogenetic signal.

• We argue that our finding of small but significant effects of life history and ecology is com-

patible with our original hypothesis regarding selection of optima in recombination rates: low

recombination rate is optimal when inmediate fitness is required. By contrast, high recombina-

tion rate is optimal when stable environments allow for evolutionary innovation.

Introduction

Stebbins (1958) and Grant (1958) argued that the function ofgenetic recombination is to bring about a workable compromisebetween the contradictory demands for immediate fitnessand evolutionary flexibility. In stable communities, already-established species have little to lose through recombination,because rare allelic combinations may have extreme fitness and theloss of gametes is outweighed by those with increased fitness. Bycontrast, unstable environments may select for lower recombina-tion rates, as those individuals that survive necessarily carrysuccessful allelic combinations and combinations and selectionfavours high reproductive potential to build up a population asquickly as possible (Mather, 1943; Stebbins, 1958; Grant, 1958;Barton, 1995; Burt, 2000). While this theory has not been studiedin a large number of taxa, studies in a handful of plant genera havefound correlations between recombination systems and ecologicalstrategies (e.g. Australian Senecio, Lawrence, 1985; Erythrina,Forni-Martins & da Cruz, 1996; and Carex, Bell, 1982).

The sedges, Carex (Cyperaceae), with c. 2100 species, compriseone of the largest angiosperm genera in the world and the largestin the northern temperate regions. The genus exhibits a

remarkable chromosome number radiation, ranging from 2n =12 to 124 chromosomes (Roalson, 2008), and intraspecies varia-tion can cover a range of > 10 chromosomes (Luceno &Castroviejo, 1991; Roalson, 2008). Carex species have holo-centric (or holokinetic) chromosomes, which are characterized bythe absence of localized centromeres. Holocentric chromosomesare found in all studied species of Cyperaceae and its sister familyJuncaceae (Greilhuber, 1995) as well as several unrelated angio-sperm genera, green algae, and mosses from Bryopsida. In addi-tion, holocentric chromosomes may be found in Rhizaria, velvetworms, nematodes, and several orders of arthropods (reviewed inMola & Papeschi, 2006). Fragments from fissions of holocentricchromosomes segregate normally during meiosis (reviewed inCope & Stace, 1985), and single-chromosome fission or fusionevents appear not to be underdominant (Faulkner, 1972;Luceno, 1993; Hipp et al., 2009). As a consequence, holocentryallows rapid evolution of chromosome number, mainly fromagmatoploid (chromosome fission; Davies, 1956) and symploid(chromosome fusion; Luceno & Guerra, 1996) events.

The drivers of among-species differences in chromosome num-ber distribution within the genus are unknown. Previous worksat the section level in Carex have produced contradictory results.

Research

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012) 1www.newphytologist.com

Chromosome number exhibits clade-specific shifts in equilibriumvalues, evolutionary rate, and phylogenetic heritability within ac. 5 million-yr-old clade (Carex section Ovales; Hipp, 2007),suggesting that the dynamics of chromosome rearrangement arenonuniform across the genus. The older (c. 13 million-yr-old)section Spirostachyae exhibits uniformly high phylogenetic herita-bility and no clade-specific shifts in rate or equilibrium value(Escudero et al., 2010). It remains to be seen whether there aredetectable ecological or life-history correlates of cladogeneticshifts in the dynamics of chromosome evolution. Chromosomedivergence has a demonstrated effect on the rate of hybridizationand gene flow within species (Escudero et al., 2010; Hipp et al.,2010), suggesting that karyotype evolution plays a role in speciesdiversification within the genus.

In addition to affecting rates of gene flow among populationsor lineages, chromosome number variation in Carex affects thearrangement and constitution of linkage groups (Faulkner, 1972)and recombination rates within species (Bell, 1982). Recombina-tion in organisms with holocentric chromosomes is likely to bedictated by chromosome number; holocentric species display oneto two chiasmata per chromosome, irrespective of chromosomesize (Nokkala et al., 2004). Bell (1982) used a comparative dataset of chromosome numbers in Carex to test for correlationsbetween recombination rate and habitat, predicting that lowerchromosome number would be associated with open, xeric, mon-tane, novel and northerly habitat and a higher rate of productionof small propagules. He utilized categorical assignments of speciesto habitats in a nonphylogenetic comparative study to demon-strate a strong correlation between recombination rate andpotential reproductive output as well as significant correlationsbetween chromosome number and a few of the habitat variables.However, his results were ‘at best highly equivocal’ (pp. 429)with respect to his predictions.

The goal of this study was to tease apart the effects of phylo-genetic inertia (i.e. resistance to adaptation) and adaptation in theevolution of among-species differences in chromosome numberin sedges (Carex), a cytogenetically hypervariable group. We eval-uate the hypothesis that environmental stability selects for highrecombination rates, and that unstable habitats select for lowrecombination rates. We used for this purpose a phylogeneticcomparative method (the stochastic linear Ornstein–Uhlenbeckmodel; Hansen et al., 2008) that simultaneously estimates theoptimal or equilibrium relationship between a trait and its envi-ronment and the rate of evolution or adaptation towards theselective optimum or equilibrium; phylogenetic inertia can beunderstood as the inverse of this rate. In doing so, we provideevidence for a link between the evolution of chromosomenumber and the ecological diversification of the temperate zone’slargest angiosperm genus.

Description

Phylogenetic data and analysis

The phylogenetic data analysed for this project are the most com-prehensive Carex-wide phylogeny published to date (Waterway

et al., 2009; Supporting Information Notes S1). Molecularsampling for that study included 139 accessions (one of twopopulations of Cymophyllus fraserianus was excluded) sequencedfor two nuclear ribosomal DNA regions (the internal transcribedspacer (ITS), excluding the 5.8S gene, which lies between ITS1and ITS2; and the external transcribed spacer (ETS)) and twoadjacent chloroplast regions (the trnL intron and the trnT-trnLspacer) (Notes S1). Genus Carex is paraphyletic with respect tothe other genera of Cyperaceae tribe Cariceae (Cymophyllus,which is monospecific, Kobresia with c. 50 spp., Schoenoxiphiumwith c. 20 spp. and Uncinia with c. 60 spp.). Consequently, allfive Cariceae genera are represented in the present study.Sequences were downloaded from NCBI GenBank(http://www.ncbi.nlm.nih.gov/genbank/).

Phylogenetic trees with branches proportional to time wereestimated using the uncorrelated log-normal relaxed clock model(Drummond et al., 2006) as implemented in BEAST v. 1.5.4(Drummond & Rambaut, 2007). The GTR + I + G substitu-tion model was used as it showed the best fit for all three data sets(ITS (excluding 5,8S), ETS and trnL intron – trnL-F) based onthe Akaike Information Criterion in MRMODELTEST 1.1b(http://www.abc.se/~nylander/). The analysis was conducted onthe combined data set using several independent MCMC(Markov chain Monte Carlo) runs of 10 000 000 generationseach to assess convergence, assuming the Yule tree prior with themean substitution rate set at 1.0. A consensus of 9000 trees (1tree per 1000 generations was retrieved) from one of the runs wasobtained with maximum clade credibility and mean heightsoptions of TREEANNOTATOR v. 1.4.5 implemented in BEAST with aposterior probability of 0.95 and a burn-in of 1000 trees (1 000000 generations).

For phylogenetic comparative analyses, we pruned 39 of 139tips (the five outgroups and 34 ingroup species without cytoge-netic counts; Supporting Information Table S1) for a total of100 species with at least one cytogenetic sample each. Four extratips were pruned (accordingly our final data set included 96species): two species without georeferenced specimens to estimatethe climatic envelope, and the two species of the Siderostictagroup, which is sister to the remainder of the genus and evolvesprimarily by polyploidy (Yan-Cheng & Qiu-Yun, 1989) ratherthan chromosome fission and fusion, as is typical in the rest ofthe Carex genus (Luceno & Castroviejo, 1991). The crown ofthis large clade (Carex excluding the Siderostictae group) dates tothe Late Eocene and Oligocene 30.8 million yr ago (95% highestposterior density = 21.8–41.07; Escudero et al., 2012).

Chromosome data and predictors

Cytogenetic data for the 96 species were taken from a recentcompendium of chromosome numbers for the genus (Roalson,2008). Chromosome means were estimated for each species.Standard errors to be used in the phylogenetic analysis were notestimated separately for each species, as sampling intensity variedwidely among species. Instead, we took the sample-size weightedaverage of sample variances across all species as the global esti-mate of average within-species variance, then we estimated the

2 Research

NewPhytologist

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

squared standard error of the mean separately for each species bydividing the global variance estimate by the sample size of theindividual species (as recommended in Labra et al., 2009 andHansen & Bartoszek, 2012); the sample size for each speciesranged from one to 39 counts (Table S1). Initial analyses wereperformed on log-transformed and raw data. These trials demon-strated that data transformation had negligible effects on ourresults and conclusions, and consequently only the results fromanalysis of raw (untransformed) data are reported.

We characterized climate envelopes for the 96 species in ourdata set based on climatic conditions at localities of georeferencedspecimens reported in the GBIF network (The GlobalBiodiversity Information Facility; http://www.gbif.org). Dupli-cated and imprecise localities (a minimum precision of 0.0001latitude and longitude degrees was required) were excluded usingR software (R Development Core Team, 2010). In addition, geo-graphic ⁄ climatic outliers (localities outside of reported distribu-tion ⁄ habitat in the literature; Kukenthal, 1909; Chater, 1980;Ball et al., 2002) were excluded by visual inspection of the dataset. Georeferenced specimen records (N = 1892 samples perspecies on average, SD = 3187; range (2–) 30–19 000 samplesper species; Table S1) were then used to extract estimatedbioclimatic variables for each locality from the WorldClimclimate database (at 2.5 min scale; http://www.diva-gis.org/climate). Five of the 19 BIOCLIM variables were used (measure-ment units and BIOCLIM abbreviations in parentheses): annualmean temperature (�C, BIO1); temperature seasonality (SD ·100, BIO4); temperature annual range or continentality (�C,BIO7); annual precipitation (mm, BIO12); and precipitationseasonality (coefficient of variation, BIO15). These five variableswere chosen among the 19 available BIOCLIM variables becausethey are ecologically meaningful in terms of stable vs unstablehabitats. For example, we expect stable habitats to be correlatedpositively with annual mean temperature (BIO1) and annual pre-cipitation (BIO12), and negatively with seasonality of tempera-ture (BIO4) and precipitation (BIO15), and temperaturecontinentality (BIO7) (for the relationship between BIO1 andBIO12 and the different habitats on Earth, see Sadava et al.,2007). Means and standard deviations over georeferenced speci-mens were estimated for each species in R (R Development CoreTeam, 2010) using the package dismo (Hijmans et al., 2010).

Morphological data were taken from (in priority order): Floraof North America (Ball et al., 2002), Flora Europaea (Chater,1980), and the only world-wide monograph of tribe Cariceae(Kukenthal, 1909). Four measurements were used to characterizethe relevant aspects of vegetative and reproductive morphology inCarex: culm length (cm), leaf width (mm), lateral inflorescenceunit length (the single peduncle unit of an inflorescence (mm),following Reznicek, 1990), and utricle length (mm). For unispi-cate species, we report the length of the single inflorescence unit.A proxy of the number of utricles per spike was estimated aslength of the lateral inflorescence unit length ⁄ utricle length,where the length of the lateral inflorescence unit length wasdivided by the utricle length. Because observations in floras aregiven as ranges of variation, following Bell (1982), midpoints ofthe frequent ranges of variation (excluding low and high outliers)

instead of means values were taken. Culm length and leaf widthcharacterize the studied species vegetatively; lateral inflorescenceunit length, utricle length and number of utricles per lateral inflo-rescence unit characterize them reproductively. We interpretlateral inflorescence unit length as a surrogate for the total invest-ment of plants in each inflorescence unit (Bogdanowicz et al.,2011), and utricle length and number of utricles per inflorescenceunit as estimates of the rate of production of propagules per inflo-rescence unit and their sizes (see r ⁄ K strategies in Pianka, 1970).

Soil moisture and light habitat were coded followingWaterway et al. (2009) as follows: soil moisture: (1) water-saturated soil, (2) moist to dry upland soils, or (3) either inter-mediate positions on the moisture gradient or variation in habitatwith respect to moisture; light habitat: (1) shaded conditions inforest habitats, (2) open habitats, or (3) intermediate positions onthe insolation gradient or variation in habitat with respect to theinsolation. Data for the core subgenus Carex clade were directlytaken from Waterway et al. (2009) and for the remaining speciesdata were inferred from floras (Chater, 1980; Ball et al., 2002).These two categorical variables are also ecologically meaningfulin terms of stability vs instability of the habitats. For example,extreme and unstable habitats such as arctic tundra (Sadava et al.,2007) will be characterized by dry soil and high insolation cate-gories. In contrast, more stable conditions such as temperaterainforest (Sadava et al., 2007) will be characterized by water-saturated soil and low insolation categories.

In this study, all analyses were conducted at the level of species.Accordingly, mean diploid chromosome numbers (response vari-able) and bioclimatic parameters were estimated at species levelbased on data from individuals. We did, however, include uncer-tainty attributable to within-species variation as measurementerror in the comparative analyses, as described in Hansen &Bartoszek (2012). Separate microevolutionary studies (e.g. at thepopulation or individual level; cf. Felsenstein, 2008; Stone et al.,2011) may complement the results presented here, but wouldaddress different research questions and need a different studyapproach and methodology.

Phylogenetic comparative analysis

In our analyses, the evolution of the response trait (mean diploidchromosome number) was modelled as an Ornstein–Uhlenbeckprocess, which is often used as a model of adaptive evolution thatexplains the evolution of a character as a result of the action ofnatural selection or other deterministic forces (Hansen, 1997;Butler & King, 2004; Hansen et al., 2008). This model has pre-viously been used in comparative studies of chromosome numberevolution by Hipp (2007) and Escudero et al. (2010), and evolu-tion of recombination rates by Dumond & Payseur (2008). TheOrnstein–Uhlenbeck model includes two components: a deter-ministic component, a tendency to evolve towards a ‘primary’optimal state that can be fixed or varying on the tree; and a sto-chastic component that represents unknown secondary factorsaffecting the trait. This model is described by the equationdY ðt Þ ¼ aðh� Y ðt ÞÞdt þ rdBðt Þ, where dY(t) is the change ofour trait (chromosome number) over an infinitesimal time

NewPhytologist Research 3

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

interval dt, a determines the rate of change towards a primaryoptimum or selection equlibrium, h, Y(t) is our trait at time t, ris the standard deviation of the secondary stochastic changes anddB(t) are independent random variables that are normally distrib-uted with mean zero and unit variance (i.e. white noise). Themodel parameters can be translated into parameters more easilyinterpreted in terms of trait evolution. The phylogenetic half-life,t1 ⁄ 2 = loge (2) ⁄ a, is an estimate of the time it takes for a speciesto evolve halfway from its ancestral state towards its primary opti-mum in a new niche, and is thus a quantification of phylogeneticinertia (Hansen, 1997). Phylogenetic inertia is the resistance toadaptive change (i.e. it comes about when related species inheritan inert trait from a common ancestor), which then resistsadaptive change to new optima (Hansen et al., 2008). Thesecond estimated parameter, the stationary variance, of theOrnstein–Uhlenbeck process (my = r2 ⁄ 2a) is an estimate of theresidual trait variance expected among species that have fullyadapted to their niches, and is directly proportional to the vari-ance of the stochastic change and inversely proportional to therate of change towards the primary optimum.

The phylogenetic Ornstein–Uhlenbeck model as originally for-mulated assumed categorical optima for different niches mappedonto the phylogeny (Hansen, 1997; Butler & King, 2004);recently it was extended to continuous predictor variables that donot need to be mapped on the phylogeny (Hansen et al., 2008;see also Labra et al., 2009 for some further developments). Thisextension retains the Ornstein–Uhlenbeck process as the modelof adaptation, but assumes that the primary optimum, h, is nowa linear function of randomly changing predictor variables. Themethod returns an estimate of the regression of the primary opti-mum on the predictor variables. This optimal regressiondescribes the optimal adaptive relationship free of ancestral influ-ence or phylogenetic inertia, that is, the predicted relationship wewould see if the species were given enough time to complete theiradaptation to their current environments. The method assumesthat the predictor variables evolve as a Brownian-motion processand that a linear model appropriately describes the relationshipbetween the optimum or equilibrium and the predictor variables.Nevertheless, there are no specific assumptions about the state ofany variable along the phylogeny farther than the tip values (i.e.the approach does not rest on any particular internal node recon-structions, although the model does imply a distribution for eachinternal node).

For our study (evolution of chromosome number in holocen-tric organisms), it is more correct to refer to the evolutiontowards an optimal state as an approach to a selective equilibriumstate in chromosome number. This karyotypic equilibriumwould be determined by the rates of fixed fusion and fission, andwe used the model to test whether this equilibrium, and hencethe rates of (fixed) fusion and fission, are dependent on the envi-ronment (i.e. on the environmental predictor variables we test).Accordingly, we will refer to chromosome-number or karyotypicequilibria rather than optima (see also Hipp, 2007).

In the current study, we utilized a combination of the linearmodelling approaches described above for continuous predictorsand an ANOVA and ANCOVA extension of the method

implemented in SLOUCH (Hansen et al., 2008; Labra et al., 2009)to evaluate the effect of two categorical ecological predictors, soilmoisture and insolation, on the evolution of chromosomenumber as well as new methods for the incorporation of measure-ment error and computation of the intercept. For this approach,categorical states were mapped onto the phylogeny using severalreconstruction methods (when the predictor variable is categori-cal, there are assumptions about the state along the phylogeny).We report results from parsimony reconstruction, but as resultsmay depend on ancestral character estimations, we also evaluatedseveral other hypotheses for ancestral characters (e.g. using ances-tral character estimations based on the all-rates-different likeli-hood model using the ace function as implemented in the ape Rpackage (Paradis et al., 2004; R Development Core Team,2010)). All analyses yielded results that are qualitatively the same;thus, only parsimony analyses are reported in this study. We usedAkaike’s Information Criterion (AIC) and AIC weights (AICw)to compare models to the ‘no-specific-adaptation model’, inwhich chromosome evolution evolves towards a single karyotypicequilibrium (this is a regression of form Y = b + e, where Y is avector of the trait of interest (in this case, diploid chromosomenumber), b is the intercept, and e is the residuals vector). Use ofAICw is a way to standardize between 0 and 1 the weight ofevidence in favour of each alternative model (Burnham &Anderson, 2002). We also report AIC weights for each parameterrelative to all models tested, as an assessment of the total eviden-tial support for that parameter relative to the complete set ofmodels evaluated. Because the plausible set of models consideredin our analysis was selected a posteriori, AIC weights formultiple-predictor models are included as a means of evaluatingwhether combined systems of environment, habitat, or morpho-logical variables are significantly better at predicting chromosomenumber than single predictors, not as absolute estimates of theposterior support for each model.

Results

Phylogeny and environmental predictors

The topology of the consensus tree (Fig. 1; Notes S2 for paren-thetical format; matrix in Notes S3) is congruent with thephylogeny published in Waterway et al. (2009); the few minortopological differences can be explained by methodologicaldifferences in phylogeny reconstruction between our study andWaterway et al.’s. The tree is scaled to 1.0 total length (from theroot to the tip of any single leaf in the ultrametric tree) to facili-tate interpretation of parameter estimates. The response and pre-dictor variables are summarized in Table S1. The total variancein climatic variables explained by among-species differencesranges from 30% (BIO12 and BIO15) to 60% (BIO1), andaccordingly, 40–70% by within-species differences. Among thefive climatic predictors, one pair exhibits correlations of |r | >0.70 (BIO4:BIO7, R2 = 0.94). Among the four morphologicalpredictors, no pair exhibits correlations of |r | > 0.70 (maximumR2 = 0.18, utricle length:lateral-inflorescence unit length). Corre-lations are also low between morphological and climatic predictor

4 Research

NewPhytologist

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

pairs (maximum R2 = 0.18, BIO1:leaf width). Climatic andmorphological variables exhibit a strong phylogenetic signal; onlyin one case was t1 ⁄ 2 < 0.5 (0.30 for BIO15). A t1 ⁄ 2 = 0.5 in unitsof tree height means that a species entering a new niche wouldneed a time span equal to half the tree length before it has losthalf the influence of its ancestral state. Regarding habitat charac-terization (see Fig. 1 for ancestral character estimation of soilmoisture), Waterway et al. (2009) suggested clustered distribu-tion on the tree for these categorical variables, which may reflectniche conservatism. A priori, diploid chromosome number maytrack all the examined predictors (bioclimatic, morphologicaland categorical habitat), as all of them display strong phyloge-netic signals. In addition, our analyses do not strongly violate theassumption of SLOUCH that predictor variables follow Brown-ian motion (all include t1 ⁄ 2 = ¥ in their support set).

Phylogenetic effects in chromosome number for the ‘singleequilibrium O–U model’

In a no-predictor (single equilibrium) O–U model, an estimateof t1 ⁄ 2 = 0 (a = ¥, instantaneous adaptation) implies that thereis no influence of the past on trait value (no phylogenetic effect),and all species represent independent draws from the trait dis-tribution. By contrast, if t1 ⁄ 2 = ¥ (a = 0, no adaptation orBrownian-motion model), phylogeny is a strong predictor of traitvalue. In our case, the point estimate of t1 ⁄ 2 for the no-predictorO–U model suggests a strong phylogenetic effect (t1 ⁄ 2 = 0.60 inunits of tree length; Fig. 2a), with supported values (values with alog-likelihood until two units lower than the maximum log-

likelihood; Edwards, 1992) ranging from a moderate to verystrong phylogenetic effect and the hypothesis of species indepen-dence strongly rejected (support interval over t1 ⁄ 2 = 0.26–¥;Fig. 2a).

Adaptation and inertia in chromosome number

None of the predictor variables explained a lot of the variation inchromosome number, but we still found clear evidence for weakeffects from several variables. As we will argue in the Discussion,we judge these effects to be biologically important. Mean chro-mosome number is more strongly predicted by inflorescence unitlength than by any other predictor (R2 = 0.063, AICw = 0.945relative to a no-predictor O–U model) (Table 1). We interpretinflorescence unit length as a proxy for total resource investmentin each inflorescence unit. Chromosome number is negativelycorrelated with temperature seasonality (BIO4) but withmarginal support (R2 = 0.038, AICw = 0.778; Table 1; Figs 2b, 3a).The correlations between chromosome number and the remain-ing continuous predictors are weak (R2 = 0.000–0.038; Table 1).For categorical habitat predictors, the correlation is overall weak(R2 = 0.035–0.054; Table 1). The best-supported model withonly categorical habitat predictors is a model with soil moistureas the sole predictor (R2 = 0.054, AICw = 0.660; Table 1;Fig. 3c), in which chromosome number is positively correlatedwith soil moisture (dry soil = 51.5 ± 4.7 chromosomes, interme-diate = 63.5 ± 6.2 chromosomes, and water-saturated soil =73.7 ± 7.8 chromosomes). The SLOUCH confidence intervals forthe regression parameters and primary optima are conditional on

0.0 0.2 0.4 0.8

backii

rupestriscurvulamyosuroides

pedunculata

hirtifoliaarctata

gracillima

depauperata

filicina

pensylvanicapeckiicommunis

lanceolataglacialis

boottianadigitata

blandaalbursina

digitalisplantagineaeburnea

sparganioides

bicknellii

rosea

nigricans

simpliciuscula

capillaris

sylvaticapendula

flava

pumila

lupulina

intumescensgrayi

scirpoidea

debilis

pallescens

ischnostachyabrownii

doniana

mollicula

foliosissimafernaldianaflacca

baccansmedia

panicea

filipesparciflora

aureagranularis

gibba

deweyanaprojecta

maritima

remota

canescens

radiatamacrocephala

0.6Time

pulicaris

paucifloraleptalea

phleoidesuncinata

punctata

michauxianafolliculata

oligosperma

riparialasiocarpa

retrorsa

hystericinacomosa

gynandranigratyphinasquarrosamacrochaeta

magellanicalimosastylosa

acutiformis

dispalata

vaginata

livida

crus−corvibromoides

diandravulpinoidea

echinata

chordorrhiza

gynocratesstipata

trispermaheleonastes

1.0

20=2n 30 40506070 8090100

Fig. 1 Ultrametric phylogenetic tree from the consensus of 9000 trees from BEAST analysis of 139 accessions of Cyperaceae. Forty-three accessions werepruned after analysis. Ancestral character estimation (based on Fitch parsimony) of categorical variables of soil moisture are shown on the branch tree. Blue,water-saturated soils; green, intermediate; red, dry uplands. Diploid chromosome number range is indicated.

NewPhytologist Research 5

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

the alpha and sigma parameters of the Ornstein–Uhlenbeck pro-cess and they are local confidence intervals. For BIO4, lateralspike unit length and soil moisture predictor, we have taken somealternative values of alpha and sigma (at the edges of the supportintervals, and a few internal points) to estimate global confidenceintervals. Our conclusions are also supported by the global confi-dence intervals for the regression parameters and primary optima(results not shown).

Multiple-predictor models explain up to 23% of the variancein chromosome number (R2 = 0.084–0.232; Table 1). However,all strongly supported multiple-predictor models include lateralinflorescence unit length as a predictor, and the supports for thesemodels as estimated using AIC weights do not exceed the AICsupport of the model with only lateral inflorescence unit lengthas a predictor (AICw = 0.194 relative to all models considered;multiple-predictor models range from AICw = 0.168 to 0.208).

–0.5

0.00.0

0.0

0.5

1.0

1.5

vy

50

100

150

200

250

300

350

-2.0

–1.5

–1.0

–0.5

0.0

Half-lifeHalf-life

Half-lifeHalf-life

0.0

0.5

1.0

1.5vy

50

100

150

200

250

300

350

Log-likelihood

Log-likelihood

Log-likelihood

Log-likelihood

–2.0

–1.5

–1.0

–0.5

0.0

0.0

0.5

1.0

1.5vy

50

100

150

200

250

300

350

–2.0

–1.5

–1.0

0.0

0.5

1.0

1.5

vy

50

100

150

200

250

300

350

–2.0

–1.5

–1.0

–0.5

(a) (b)

(c) (d)

Fig. 2 The support (log-likelihood) difference from the best model is shown on the ordinate relative to the different parameter combinations. (a) Phyloge-netic effects on chromosome number in Cyperaceae based on a model including only an intercept (2n ~ 1; best estimate: t1 ⁄ 2 = 0.60 (0.26–¥) in units oftree height; vy = 216.5 chromosome number squared). (b) Phylogenetic effects and inertia in chromosome number based on a model including tempera-ture seasonality as predictor (best estimate: 0.52 (0.22–7.5); vy = 176). (c) Phylogenetic inertia in chromosome number based on a model with lateralinflorescence unit length as predictor (best estimate: t1 ⁄ 2 = 0.50 (0.22–6.75); vy = 171.5). (d) Phylogenetic inertia in chromosome number based on amodel including lateral inflorescence unit length, soil moisture and temperature seasonality (best estimate: t1 ⁄ 2 = 0.38 (0.14–1.85); vy = 140).

6 Research

NewPhytologist

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

Table 1 Models, phylogenetic half life (t1 ⁄ 2) scaled in tree-length units with 2-unit support interval in parentheses, stationary variance (vy) in units ofchromosome number squared, intercept (± SE) in units of chromosome number and slope (± SE) from phylogenetic regression, R-squared (R2 in %),Akaike Information Criterion (AIC) and Akaike Information Criterion weights (AICw) in comparison with the no-specific-adaptation model(single-equilibrium O–U model: 2n ~ 1) are shown. In all models the response variable is 2n, where 2n is diploid chromosome number. Predictor variablesare given to the left of the tilde, and a ‘1’ means that the model has only an intercept.

Model [K] t1 ⁄ 2 vy Intercept (± SE) Slope (± SE) R2 AICSimpleAICw**

GlobalAICw*

2n ~ 1 (single-equilibrium O–U) 0.60 (0.26–¥) 217 58.042 ± 3.876 – 0.000 752.099 – 0.01132n ~ 1 (Brownian motion) [K = 2] – – 57.165 ± 6.794 – 0.000 755.612 0.1473 0.00202n ~ mean temperature (BIO1) 0.76 (0.28–¥) 244 55.535 ± 4.776 0.311 ± 0.275 #chr. ⁄ �C 0.013 752.705 0.5752 0.00842n ~ temperature seasonality(BIO4)

0.52 (0.22–7.5) 176 69.917 ± 6.094 ) 1.646 ± 0.707 #chr. ⁄ �K 0.038 749.593 0.7778 0.0396

2n ~ temperature range (BIO7) 0.52 (0.22–8.9) 190 66.171 ± 6.416 ) 0.272 ± 0.181 #chr. ⁄ �C 0.018 752.200 0.5126 0.01082n ~ mean precipitation (BIO12) 0.62 (0.26–¥) 221 58.381 ± 5.331 ) 0.000 ± 0.003 #chr. ⁄ mm 0.000 754.082 0.2707 0.00422n ~ precipitation seasonality(BIO15)

0.80 (0.28–¥) 239 53.549 ± 5.704 0.140 ± 0.127 #chr. ⁄ mm 0.015 751.311 0.5973 0.0168

2n ~ culm length 0.62 (0.26–¥) 217 57.031 ± 4.397 0.027 ± 0.055 #chr. ⁄ cm 0.003 753.843 0.2949 0.00472n ~ leaf width 0.64 (0.26–¥) 217 59.708 ± 4.103 ) 0.476 ± 0.303 #chr. ⁄ mm 0.021 751.992 0.5134 0.01192n ~ LIUL 0.50 (0.22–6.75) 172 54.858 ± 3.526 0.169 ± 0.065 #chr. ⁄ mm 0.063 746.413 0.9450 0.19442n ~ utricle length 0.62 (0.26–¥) 221 58.422 ± 4.486 ) 0.092 ± 0.499 #chr. ⁄ mm 0.000 754.068 0.2720 0.00422n ~ utricles per LIUL 0.54 (0.22–¥) 194 56.731 ± 3.726 0.294 ± 0.178 #chr. ⁄ utricles 0.027 751.421 0.5840 0.0159

2n ~ soil moisture (3 categories) 0.52 (0.20–¥) 185 D = 51.493 ± 4.725I = 63.496 ± 6.212S = 73.658 ± 7.842

– 0.054 750.780 0.6597 0.0219

2n ~ light (3 categories) 0.50 (0.22–¥) 190 I = 59.807 ± 5.892F = 51.348 ± 5.828O = 69.075 ± 6.904

– 0.035 752.764 0.4712 0.0081

2n ~ light and soil moisture(9 categories)

0.32 (0.16–1.02) 136 DF = 43.008 ± 5.117DI = 55.527 ± 6.896DO = 60.602 ± 8.113IF = 59.9467 ± 9.070II = 59.192 ± 6.013IO = 79.800 ± 9.117SF = 69.070 ± 12.501SI = 78.909 ± 9.422SO = 68.283 ± 5.235

– 0.163 752.933 0.3973 0.0075

2n ~ LIUL +temperature seasonality

0.46 (0.20–3.1) 158 64.264 ± 6.44 LIUL = 0.138 ± 0.064#chr. ⁄ mm

TS = ) 1.237 ± 0.724#chr. ⁄ �K

0.084 746.276 0.9484 0.2082

2n ~ temperatureseasonality + soil moisture

0.40 (0.16–2.45) 149 D = 61.778 ± 6.629I = 69.736 ± 6.433S = 78.739 ± 7.340

) 2.432 ± 1.274 #chr. ⁄ �K 0.093 749.155 0.8134 0.0494

2n ~ LIUL + soilmoisture

0.40 (0.16–2.45) 149 D = 50.061 ± 4.109I = 58.877 ± 4.927S = 65.884 ± 6.271

0.251 ± 0.103 #chr. ⁄ mm 0.111 746.486 0.9430 0.1875

2n ~ LIUL + soilmoisture + temperatureseasonality

0.38 (0.14–1.85) 140 D = 57.650 ± 6.806I = 65.000 ± 6.655S = 72.268 ± 7.639

TS = ) 1.734 ± 1.267#chr. ⁄ �K

LIUL = 0.204 ± 0.101#chr. ⁄ mm

0.132 746.705 0.9369 0.1680

2n ~ LIUL +soil moisture andlight + temperatureseasonality

0.26 (0.12–0.72) 113 DF = 43.530 ± 8.292DI = 55.570 ± 8.035DO = 58.969 ± 7.911IF = 57.8418 ± 9.356II = 55.951 ± 7.215IO = 75.263 ± 8.340SF = 65.781 ± 11.701SI = 65.645 ± 10.522SO = 67.005 ± 6.719

TS = ) 0.446 ± 1.103#chr. ⁄ �K

LIUL = 0.204 ± 0.085#chr. ⁄ mm

0.232 750.504 0.6895 0.0251

Significant and marginally significant AICw are in bold. D, dry soil; W, water-saturated soil; I, intermediate; O, open environment; F, forest; TS, temperatureseasonality; LIUL, lateral inflorescence unit length; #chr, chromosome number.*Global AICw is the AIC weight for each model relative to all models tested. As 20 models were evaluated, the prior expectation for AICw is 0.05.**Simple AICw is the AIC weight for each model relative to the no-predictor O–U model.

NewPhytologist Research 7

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

These models differ, however, in their estimates of phylogeneticinertia. Phylogenetic half-life estimates vary from t1 ⁄ 2 = 0.26 to0.80 in units of tree length (Table 1), and confidence intervals

for many of the models reject the pure Brownian-motion model(i.e. the upper bound on t1 ⁄ 2 < ¥; Table 1). Many of these samemodels are favoured over the single-equilibrium (no-predictor)O–U model (Table 1; simple AICw > 0.5), although only inflo-rescence unit length strongly rejects the single-equilibrium model(simple AICw = 0.945). For the most strongly supportedmodels, half-life estimates vary from t1 ⁄ 2 = 0.38 to 0.52, withconfidence intervals rejecting both instantaneous adaptation (t1 ⁄ 2= 0) and Brownian motion (t1 ⁄ 2 = ¥, or approx. t1 ⁄ 2 ‡ 6)(Table 1).

Discussion

Adaptation, inertia, or unpredictable chromosome numbervariation?

In general, selection on recombination rates is a weak, second-ary force (Barton, 1995; Otto & Barton, 2001). Moreover, theeffects of small changes in chromosome number through fusionor fission will only lead to small changes in overall recombina-tion rates. This means that we do not expect to find strongdirect selection and rapid adaptation of chromosome number toexternal variables. Instead, evolution of chromosome numbermust be erratic and influenced by genetic drift and ⁄ or stochasticindirect selection as a result of coincidental associations of thechromosomal mutations to other aspects of the individualphenotype. Accordingly, Dumond & Payseur (2008) concludedthat the genomic rate of recombination in mammals follows aneutral evolution (Brownian-motion) model. Hence, it wouldbe unrealistic to predict a strong, dominant association betweenchromosome number and environmental predictors. Instead,selection for varying recombination rate can at most be one rel-atively weak force among many others that affect chromosomenumber variation (reviewed in Hipp et al. (2009) for Carex).To be able to detect such a systematic, but weak primary effect,we need a large number of species to see the pattern amongstrong ancestral effects and variation from the many secondaryeffects.

In this study of 96 species of sedges, we were able to detect sys-tematic associations of chromosome number to life-history andhabitat variables. Specifically, we found that sedges with higherchromosome numbers tend to have larger lateral inflorescenceunits, and, to a lesser extent, occur in habitats with water-satu-rated soils and low temperature seasonality. The total amount ofvariance in chromosome number explained by these predictorswas low. Separately it ranged from 3.8 to 6.3%, and in combina-tion these traits explained 13% of chromosome variation in

2 4 6 8 10

3040

5060

7080

9030

4050

6070

8090

3040

5070

6080

90

Temperature seasonality (°K)

Mea

n di

ploi

d ch

rom

osom

e nu

mbe

r

D I S

Moisture

Mea

n di

ploi

d ch

rom

osom

e nu

mbe

r

0 50 100 150Lateral inflorescence unit length (mm)

Mea

n di

ploi

d ch

rom

osom

e nu

mbe

r

(a)

(b)

(c)

Fig. 3 (a) Optimal (in grey) and phylogenetic (in black) regressions ofdiploid mean chromosome number on temperature seasonality in �K. (b)Optimal (in grey) and phylogenetic (in black) regressions of diploid meanchromosome number on lateral inflorescence unit length in mm. (c) Meanchromosome number and the standard deviation (wide boxes) of the dataas well as mean ± SE of primary equilibria (thin boxes) for the three soilmoisture categories (D, dry uplands; I, intermediate; W, water-saturated).For more details, see Table 1.

8 Research

NewPhytologist

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

Carex, leaving 87% unexplained. As explained above, this is to beexpected, and in the light of strong phylogenetic inertia, explain-ing even a small component of total variation in chromosomenumber can be considered a strong effect. The study also demon-strates that phylogenetic inertia is significant component of everymodel evaluated (viz., t1 ⁄ 2 > 0). Hence, although the majority ofvariation in chromosome number remains unexplained by ouranalyses, we regard our results as consistent with the generalhypothesis that chromosomal fusion and fission are influenced byselection to adapt recombination rates to the life history of theorganism.

Phenotypic and genetic consequences of chromosomenumber variation in holocentric organisms

The physiological and ecological implications of polyploidy havebeen studied a great deal (reviewed in Otto & Whitton, 2000;Balao et al., 2011). Because ploidy changes entail changes inDNA content and gene number, expression differences associatedwith polyploidization can have dramatic effects on phenotype(Otto & Whitton, 2000). The phenotypic and physiologicalimplications of small cytogenetic changes, however, wherechanges in DNA content are minimal (quantitative aneuploidy)or null (qualitative aneuploidy: agmatoploidy or symploidy), arenot well understood. In Carex, where polyploidy is rare andchromosome evolution proceeds primarily by fission, fusion, andtranslocations, there is no expected adaptive significance of chro-mosome number per se (Hipp, 2007). However, chromosomeevolution in the genus is expected to affect recombination rates(Bell, 1982) and constitution of linkage groups (Faulkner, 1972).As Bell (1982) observed, chromosome number is an appropriateproxy for recombination rate only if chiasma frequency is fairlyconstant per chromosome and chromosome rearrangements donot entail chromosome duplication or deletion. Recent researchdemonstrates that, in organisms with holocentric chromosomes,among-chromosome variation in chiasma frequency is very lim-ited (one or two chiasmata per chromosome, irrespective of chro-mosome size; Nokkala et al., 2004; M. Luceno, pers. obs. inRhynchospora, Cyperaceae), and recombination rates have beendemonstrated to be dependent on the number of chromosomearms (proper homologue disjunction at meiosis requires at leastone chiasma per chromosome arm; Pardo-Manuel de Villena &Sapienza, 2001). Moreover, chromosome number changes inCarex are not associated with correlated changes in DNA content(Chung et al., 2011, 2012).

For example, in our data set, chromosome number ranges from2n = 26 (n = 13) in Carex pedunculata to 2n = 96 (n = 48) inUncinia phleoides and, accordingly, one might expect 13–26chiasmata in C. pedunculata and 48–96 in U. phleoides duringmeiosis. Such differences in chiasma number during meiosiscertainly entail differences in recombination rates. Thus Bell’s(1982) argument that chromosome number is an appropriateproxy for recombination rates in Carex is supported by the lastthree decades of additional research in holocentric organisms. Wedo not expect additional consequences from changes in chromo-some number in holocentric organisms. The exception may be

positional effects (i.e. after fission or fusion, a gene that was inthe middle of a chromosome may be at the extreme and viceversa, which may have consequences in gene expression). Never-theless, while we expect a systematic effect in recombination ratesfrom changes in chromosome number, we do not expect it ingene expression.

Bell (1982) tested whether recombination rates are lower inspecies occupying novel, disturbed, or marginal environmentsand ⁄ or species with larger numbers of small propagules. Heformulated the habitat hypothesis by investigating whether lowerchromosome numbers are characteristic of open, xeric, montane,novel and northerly habitats, and he concluded that habitatcorrelations were equivocal. In addition, he reported a significantpositive correlation between any measure of potential reproduc-tive output and chromosome number. Bell (1982) viewed thisfinding as a violation of his expectations. Our study, however,refines Bell’s finding and suggests a different interpretation. First,the approach we have taken provides insights into both thephylogenetic component of chromosome variability and rate ofevolution towards an equilibrium influenced by inflorescenceunit length (and, to a lesser extent, seasonality and soil moisture).Secondly, our results of a correlation between chromosome numberand bioclimatic variables, although only marginally significant(Table 1), are congruent with our initial hypothesis, as low chro-mosome number (low recombination rates) are related to extremeand unstable habitat (dry soil and unstable temperature). Thirdly,we found that variables strongly related to the number of propa-gules produced per plant and propagule sizes (utricles per inflo-rescence unit and utricle length) explained less chromosomenumber variation than inflorescence unit length. This suggeststhat shorter inflorescences may correlate with lower dependenceon local resources, which is characteristic of plants growing inextreme and unstable habitats (e.g. Sonesson & Callaghan,1991). Thus, our results suggest in aggregate a support for theadaptive significance of recombination rates.

Implications for patterns of lineage diversification

Chromosome rearrangements have previously been argued to beadaptatively significant in Carex (Faulkner, 1972; Luceno &Castroviejo, 1991; Bell, 1982), as they have been for numerousother organisms (e.g. Alvarez-Castro & Alvarez, 2005; Butlin,2005; Alvarez-Castro & Carlborg, 2008; Hoffmann &Rieseberg, 2008). The current study is the first to demonstratethat chromosome number itself – which may be a result of manyindependent and different rearrangements – may be in part aproduct of correlation with life-history traits or (more weaklysupported) adaptation to the environment. While the directionof chromosome evolution has long been of interest in Carex(Heilborn, 1924; Davies, 1956; Luceno & Castroviejo, 1991),chromosome evolution in the genus is clearly not directional(Reznicek, 1990; Hipp et al., 2009; Escudero et al., 2010). Thecurrent study provides a nonstochastic explanation for thisobservation and for previous findings that pure phylogeneticexplanations of chromosome variance are inadequate (Hipp,2007).

NewPhytologist Research 9

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

Carex is the most diversified angiosperm genus in the temper-ate regions of the Northern Hemisphere (Reznicek, 1990) andexhibits one of the most exceptional chromosome radiations inangiosperms. The relationship among chromosome number,reproductive traits, and environment reported here may well haveplayed a significant role in the diversification of the genus. In oneof the largest sections of the genus, for example, origins of aneastern North American clade from western North Americanancestry entailed a significant decrease in equilibrium chromo-some number, compatible with a shift in selective regime but notwith a pure phylogenetic (Brownian-motion) model (Carexsection Ovales; Hipp, 2007). The direction of evolution in thisclade – which was largely excluded from Waterway’s data set andconsequently from analyses presented in this paper – is in broadterms compatible with the observations reported here: the shift toa novel environment (from western to eastern North America) isassociated with a significant decrease in chromosome number(lower recombination rates). Similarly, the change in the North-ern Hemisphere from a stable, warm and humid climate duringthe Miocene and beginning of the Pliocene to a more unstable,dryer, colder climate during the end of the Pliocene and Pleisto-cene was essential to the expansion of Carex and diversification inthe northern temperate regions (Escudero et al., 2012). Congru-ently, our results suggest a correlation of low chromosome num-ber (low recombination rate) with short lateral inflorescence unitlength (presumably associated with lower dependence on localresources), and with dry soil and unstable temperature (Table 1).The capacity to evolve rapid changes in chromosome numbermay have facilitated the spread of Carex in the temperate zone, asincreased potential for evolution of recombination rates mayameliorate the effects of drift in the highly selfing populationstypical of Carex (Friedman & Barrett, 2009) or allow for morerapid rates of adaptation to novel environments (Burt, 2000;Betancourt et al., 2009). The diversity of sedges may thus dependon their chromosomal diversity.

Acknowledgements

We thank Dr Francisco Balao, Dr Kyong-Sook Chung, EnriqueMaguilla, Dr Richard Abbott and three anonymous reviewers forvaluable comments on the manuscript; Dr Jason Pienaar, Dr IanPearse and Dr Luis Villagarcıa for help with some analyses; andPaco Fernandez, Monica Mıguez, Bethany Brown and MarleneHahn for technical support. This research was supported by theNILS mobility project (UCM-EEA-ABEL-02-2009 to M.E. andT.F.H.); the Spanish Government (CGL2009-09972 to M.L.(PI) and M.E.); and the USA National Science Foundation(NSF-DEB Award #0743157 to A.H.).

References

Alvarez-Castro JM, Alvarez G. 2005. Models of general frequency-dependent

selection and mating-interaction effects and the analysis of selection patterns in

Drosophila inversion polymorphisms. Genetics 170: 1167–1179.

Alvarez-Castro JM, Carlborg O. 2008. The yank of Dobzhansky’s bequest.

Evolutionary Biology 35: 176–181.

Balao F, Herrera J, Talavera S. 2011. Phenotypic consequences of polyploidy

and genome size at the microevolutionary scale: a multivariate morphological

approach. New Phytologist 192: 256–265.

Ball PW, Reznicek AA, Murray DF. 2002. Cyperaceae Jussieu. In: Flora of

North America Editorial Committee, eds. Flora of North America north ofMexico, vol. 23. New York, NY, USA: Oxford University Press, 3–608.

Barton N. 1995. A general model for the evolution of recombination. GeneticalResearch 65: 123–144.

Bell G. 1982. The masterpiece of nature: the evolution and genetics of sexuality.Berkeley, CA, USA: University of California Press.

Betancourt AJ, Welch JJ, Charlesworth B. 2009. Reduced effectiveness of

selection caused by a lack of recombination. Current Biology 19: 655–660.

Bogdanowicz AM, Olejniczak P, Lembicz M, Zukowski W. 2011. Costs of

reproduction in life history of a perennial plant Carex secalina. CentralEuropean Journal of Biology 6: 870–877.

Burnham KP, Anderson DR. 2002. Model selection and multimodel inference: apractical information-theoretic approach. New York, NY, USA: Springer.

Burt A. 2000. Perspective: sex, recombination, and the efficacy of selection – was

Weismann right? Evolution 54: 337–351.

Butler M, King AA. 2004. Phylogenetic comparative analysis: a modelling

approach for adaptive evolution. American Naturalist 164: 683–695.

Butlin RK. 2005. Recombination and speciation. Molecular Ecology 14: 2621–2635.

Chater AO. 1980. Carex L. In: Tutin TG, Heywood VH, Burges NA, Moore

DM, Valentine DH, Walters SM, Webb DA, eds. Flora Europaea, vol. 5.

Cambridge, UK: Cambridge University Press, 290–323.

Chung K-S, Hipp AL, Roalson EH. 2012. Chromosome number evolves

independently of genome size in a clade with non-localized centromeres

(Carex: Cyperaceae). Evolution. doi: 10.1111/j.1558-5646.2012.01624.x.

Chung K-S, Weber JA, Hipp AL. 2011. Dynamics of chromosome number and

genome size variation in a cytogenetically variable sedge (Carex scoparia var.

scoparia, Cyperaceae). American Journal of Botany 98: 122–129.

Cope TA, Stace CA. 1985. Cytology and hybridization in the Juncus bufonius L.

aggregate in Western Europe. Watsonia 15: 103–320.

Davies EW. 1956. Cytology, evolution and origin of the aneuploid series in the

genus Carex. Hereditas 42: 349–365.

Drummond AJ, Ho SYW, Phillips MJ, Rambaut A. 2006. Relaxed

phylogenetics and dating with confidence. PLoS Biology 4: 699–710.

Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evolutionary analysis by

sampling trees. BMC Evolutionary Biology 7: 214.

Dumond BL, Payseur BA. 2008. Evolution of the genomic rate of recombination

in mammals. Evolution 62: 276–294.

Edwards AWF. 1992. Likelihood, 2nd edn. Baltimore, MD, USA: Johns Hopkins

University Press.

Escudero M, Hipp AL, Luceno M. 2010. Karyotype stability and predictors of

chromosome number variation in sedges: a study in Carex section Spirostachyae(Cyperaceae). Molecular Phylogenetics and Evolution 57: 353–363.

Escudero M, Hipp AL, Waterway M, Valente LM. 2012. Diversification rates

and chromosome evolution in the most diverse angiosperm genus of the

temperate zone (Carex, Cyperaceae). Molecular Phylogenetics and Evolution.

doi:10.1016/j.ympev.2012.02.005.

Faulkner JS. 1972. Chromosome studies on Carex section Acutae in northwest

Europe. Botanical Journal of the Linnean Society 65: 271–301.

Felsenstein J. 2008. Comparative methods with sampling error and within-species

variation: contrasts revisited and revised. American Naturalist 171: 713–725.

Forni-Martins ER, da Cruz ND. 1996. Recombination indices in species of

Erythtina L. (Leguminoseae, Papilionoideae). Botanical Journal of the LinneanSociety 122: 163–170.

Friedman J, Barrett SCH. 2009. The consequences of monoecy and protogyny

for mating in wind-pollinated Carex. New Phytologist 181: 489–497.

Grant VE. 1958. The regulation of recombination in plants. Cold Spring HarborSymposia on Quantitative Biology 23: 337–363.

Greilhuber J. 1995. Chromosomes of the monocotiledons (general aspects). In:

Rudall PJ, Cribb PJ, Cutler DF, Humphries CJ, eds. Monocotiledons: systematicand evolution. London, UK: Royal Botanical Gardens Kew, 379–414.

Hansen TF. 1997. Stabilizing selection and the comparative analysis of

adaptation. Evolution 51: 1341–1351.

10 Research

NewPhytologist

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com

Hansen TF, Bartoszek K. 2012. Interpreting the evolutionary regression: the

interplay between observational and biological errors in phylogenetic

comparative studies. Systematic Biology. doi: 10.1093/sysbio/syr122.

Hansen TF, Pienaar J, Orzack SH. 2008. A comparative method for studying

adaptation to a randomly evolving environment. Evolution 62: 1965–1977.

Heilborn O. 1924. Chromosome numbers and dimensions, species formations

and phylogeny in genus Carex. Hereditas 5: 129–216.

Hijmans RJ, Phillips S, Leathwick J, Elith J. 2010. dismo: species distribution

modeling. R package. [WWW document] URL http://cran.r-project.org/

web/packages/dismo/index.html [accessed 18 November 2011].

Hipp AL. 2007. Nonuniform processes of chromosome evolution in sedges

(Carex: Cyperaceae). Evolution 61: 2175–2194.

Hipp AL, Rothrock PE, Roalson EH. 2009. The evolution of chromosome

arrangements in Carex (Cyperaceae). Botanical Review 75: 96–109.

Hipp AL, Rothrock PE, Whitkus R, Weber JA. 2010. Chromosomes tell half of

the story: the correlation between karyotype rearrangements and genetic

diversity in sedges, a group with holocentric chromosomes. Molecular Ecology19: 3124–3138.

Hoffmann AA, Rieseberg LH. 2008. Revisiting the impact of inversions in

evolution: from population genetic markers to drivers of adaptive shifts and

speciation. Annual Review of Ecology, Evolution, and Systematics 39: 21–42.

Kukenthal G. 1909. Cyperaceae-Caricoideae. In: Engler A (ed.). DasPflanzenreich, vol 4 (20). Leipzig, Germany: Engelmann, 1–814.

Labra A, Pienaar J, Hansen TF. 2009. Evolution of thermal physiology in

Liolaemus lizards: adaptation, phylogenetic inertia, and niche tracking.

American Naturalist 174: 204–220.

Lawrence ME. 1985. Senecio L. (Asteraceae) in Australia: recombination systems

of a polyploid series. Australian Journal of Botany 33: 209–219.

Luceno M. 1993. Chromosome studies on Carex (L.) section Mitratae Kukenth.

(Cyperaceae) in the Iberian Peninsula. Cytologia 58: 321–330.

Luceno M, Castroviejo S. 1991. Agmatoploidy in Carex laevigata (Cyperaceae)

fusion and fission of chromosomes as the mechanism of cytogenetic evolution

in Iberian populations. Plant Systematics and Evolution 177: 149–159.

Luceno M, Guerra M. 1996. Numerical variations in species exhibiting

holocentric chromosomes: a nomenclatural proposal. Caryologia 49:

301–309.

Mather K. 1943. Polygenic inheritance and natural selection. Biological Review18: 32–64.

Mola LM, Papeschi AG. 2006. Holocentric chromosomes at a glance. Journal ofBasic and Applied Genetics 17: 17–33.

Nokkala S, Kuznetsova VG, Maryanska-Nadachowska A, Nokkala C. 2004.

Holocentric chromosomes in meiosis. I. Restriction of the number of

chiasmata in bivalents. Chromosome Research 12: 733–739.

Otto SP, Barton NH. 2001. Selection for recombination in small populations.

Evolution 55: 1921–1931.

Otto SP, Whitton J. 2000. Polyploid incidence and evolution. Annual Review ofGenetics 34: 401–437.

Paradis E, Claude J, Strimmer K. 2004. APE: analyses of phylogenetics and

evolution in R language. Bioinformatics 20: 289–290.

Pardo-Manuel de Villena F, Sapienza C. 2001. Recombination is proportional to

the number of chromosome arms in mammals. Mammalian Genome 12:

318–322.

Pianka ER. 1970. On r and K selection. American Naturalist 104: 592–597.

R Development Core Team. 2010. R: a language and environment for statisticalcomputing. Vienna, Austria: R Foundation for Statistical Computing.

Reznicek AA. 1990. Evolution in sedges (Carex, Cyperaceae). Canadian Journal ofBotany 68: 1409–1432.

Roalson EH. 2008. A synopsis of chromosome number variation in the

Cyperaceae. Botanical Review 74: 209–393.

Sadava D, Heller HC, Orians GH, Purves WK, Hillis DM. 2007. Life: thescience of biology, 8th edn. Sunderland, MA, USA: Sinauer Associates and New

York, NY, USA: WH Freeman.

Sonesson M, Callaghan TV. 1991. Strategies of survival in plants of the

Fennoscandian tundra. Arctic 44: 95–105.

Stebbins GL. 1958. The inviability, weakness and sterility of interspecific hybrids.

Advances in Genetics 9: 147–215.

Stone GN, Nee S, Felsenstein J. 2011. Controlling for non-independence in

comparative analysis of patterns across populations within species. PhilosophicalTransactions of the Royal Society B: Biological Sciences 366: 1410–1424.

Waterway MJ, Hoshino T, Masaki T. 2009. Phylogeny, species richness, and

ecological specialization in Cyperaceae tribe Cariceae. Botanical Review 75:

138–159.

Yan-Cheng T, Qiu-Yun X. 1989. Cytological studies of Carex siderosticta Hance

(Cyperaceae) and its importance in phytogeography. Cathaya 1: 49–60.

Supporting Information

Additional supporting information may be found in the onlineversion of this article.

Table S1 List of species including cytogenetic sampling, meandiploid chromosome number, categorical predictor variables, cli-mate predictor variables and morphological predictor variables

Notes S1 Accession data for voucher specimens for DNAsequences used in this study.

Notes S2 Ultrametric phylogenetic tree in parenthetical formatfrom the consensus (using maximum clade credibility tree andmean heights in TREEANNOTATOR v.1.61) of 9000 trees fromBEAST analysis of 139 accessions of Cyperaceae.

Notes S3 Phylogenetic matrix in nexus format with 139 taxa and2588 characters.

Please note: Wiley-Blackwell are not responsible for the contentor functionality of any supporting information supplied by theauthors. Any queries (other than missing material) should bedirected to the New Phytologist Central Office.

NewPhytologist Research 11

� 2012 The Authors

New Phytologist � 2012 New Phytologist Trust

New Phytologist (2012)

www.newphytologist.com