evo_1563

Document Sample
evo_1563 Powered By Docstoc
					evo_1563     evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012   :959


                                                                                                         EVO     evo_1563 Dispatch: 1-14-2012 CE: AFL
                                                                                                         Journal MSP No. No. of pages: 14     PE: Sonia
      1       O R I G I NA L A RT I C L E
      2
      3                                                                                                               doi:10.1111/j.1558-5646.2011.01563.x

      4
      5
      6
      7
              SIBLING COMPETITION ARENA: SELFING
      8
      9
              AND A COMPETITION ARENA CAN COMBINE
     10
     11
              TO CONSTITUTE A BARRIER TO GENE FLOW
     12
     13       IN SYMPATRY
     14
     15       A. K. Gibson,1,2,3,4 M. E. Hood,5 and T. Giraud1,2
     16       1
                                            ´                               ´
                  Laboratoire Ecologie, Systematique et Evolution, Universite Paris Sud, 91405 Orsay, France
     17       2
                  CNRS, 91405 Orsay, France
     18            3
                       E-mail: amakgibs@indiana.edu
     19
              4
     20 Q1
                  Department of Biology, Indiana University, Bloomington, Indiana
              5
     21           Department of Biology, Amherst College, Amherst, Massachusetts
     22
     23
     24
     25            Received May 16, 2011
     26            Accepted December 9, 2011
     27            Data Archived: Dryad: doi: 10.5061/dryad.rg148qj4
     28
     29            Closely related species coexisting in sympatry provide critical insight into the mechanisms underlying speciation and the mainte-
     30            nance of genetic divergence. Selfing may promote reproductive isolation by facilitating local adaptation, causing reduced hybrid
     31            fitness in parental environments. Here, we propose a novel mechanism by which selfing can further impair interspecific gene
     32            flow: selfing may act to ensure that nonhybrid progeny systematically co-occur whenever hybrid genotypes are produced. Under a
     33            competition arena, the fitness differentials between nonhybrid and hybrid progeny are then magnified, preventing development
     34            of interspecific hybrids. We investigate whether this “sibling competition arena” can explain the coexistence in sympatry of closely
     35            related species of the plant fungal pathogens (Microbotryum) causing anther-smut disease. The probabilities of intrapromycelial
     36            mating (automixis), outcrossing, and sibling competition were manipulated in artificial inoculations to evaluate their contribution
     37            to reproductive isolation. We report that both intrapromycelial selfing and sibling competition significantly reduced rates of hy-
     38            brid infection beyond that expected based solely upon selfing rates and noncompetitive fitness differentials between hybrid and
     39            nonhybrid progeny. Our results thus suggest that selfing and a sibling competition arena can combine to constitute a barrier to
     40            gene flow and diminish selection for additional barriers to gene flow in sympatry.
     41
     42            KEY WORDS:       Assortative mating, fungi, hybridization, prezygotic isolating mechanism, postzygotic reproductive isolation,
     43            selection arena, Silene, speciation.
     44
     45
     46       The emergence of reproductive barriers that preserve locally                been gained from studying closely related species occurring in
     47       adapted gene complexes is critical to the genetic divergence or             sympatry. Studies of sympatric sister taxa, for example, have
     48       coexistence of sympatric populations and, more broadly, to the              consistently shown the importance of premating mechanisms in
     49       process of speciation. Understanding the origins of reproductive            barring gene flow (Husband and Sabara 2004; Kay 2006; Martin
     50       isolation in sympatry, however, has proven difficult. Frequently,                             a            e
                                                                                          and Willis 2007; S´ nchez-Guill´ n et al. 2011).
     51       multiple isolating mechanisms are observed between well-                         Although tremendous variation in mating systems exists
     52       established species, obscuring the primary barriers to gene flow            across all major groups of sexual eukaryotes, selfing as a pre-
     53       (Ramsey et al. 2003; de Vienne et al. 2009). Some insights have             mating barrier to gene flow is studied infrequently and almost

                                   C 2012 The Author(s).

                              1    Evolution
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012      :959




    1      A. K. GIBSON ET AL.

    2
    3

    4      exclusively in plant systems (Levin 2010). Frequent cleistogamy            relative to nonhybrid progeny (Table 1, 1.4). These factors in com-
    5      (i.e., nonopening, self-pollinating flowers), for example, has             bination may then function as a true isolating barrier, with gene
    6      been found to significantly reduce hybridization between sym-              flow reduced in association with interspecific hybridization but
    7      patric Mimulus species (Martin and Willis 2007). The reduc-                not intraspecific outcrossing, provided that outcrossed progeny
    8      tion in outcrossing accompanying selfing can protect reproduc-             do not suffer reduced viability relative to selfed progeny.
    9      tive investment by preventing the formation of hybrid progeny                   The reproductive traits of many taxa, most notably plant
   10      with maladaptive genetic combinations (Antonovics 1968;                    and fungal taxa, suggest that the sibling competition arena may
   11      Allard 1975).                                                              constitute an isolating barrier across many systems. Importantly,
   12            Selfing is considered to be an unusual isolating barrier, how-       such a barrier does not absolutely require sibling competition:
   13      ever. Because it usually isolates intra- and interspecific individuals     competition between any hybrid and nonhybrid individuals could
   14      equivalently, it has been argued that selfing cannot be regarded           suffice. However, mixed broods of hybrid and nonhybrid siblings
   15      as directly promoting speciation (Coyne and Orr 2004, p. 212).             automatically yield an early competitive arena, and this early com-
   16      Nonetheless, selfing commonly facilitates speciation indirectly            petition is predicted to be a more powerful isolating mechanism
   17      by generating other isolating barriers; for example, the reduction         in those systems in which the production of hybrids is inherently
   18      in gene flow from maladapted populations promotes local adap-              coupled to the production of nonhybrids. Thus, siblings are the
   19      tation, thereby accelerating genetic divergence (Coyne and Orr             most relevant competitors in many systems, including this study’s
   20      2004, p. 212).                                                             focal species (Table 1, 3.1). Moreover, the mechanism of isolation
   21            Here, we propose a novel mechanism by which selfing func-            need not be adaptive (Table 1, 4.1), a key distinction between our
   22      tions as a genuine isolating barrier that limits the success of in-        model and Stearns’ selection arena (1987). In fact, sibling com-
   23      terspecific hybrids to a greater extent than the progeny of in-            petition and any resulting reproductive isolation may merely be
   24      traspecific crosses. When reproduction is associated with early,           byproducts of the mating system. For example, fungal pathogens
   25      intense competition between numerous sibling progeny for a lim-            commonly produce large quantities of spores and predominantly
   26      ited resource, hybrids will always compete with selfed nonhy-              perform haploid or diploid selfing (Giraud et al. 2008a). Both of
   27      brids for establishment. If hybrids suffer any degree of fitness           these strategies facilitate reproduction and dispersal to new hosts.
   28      reduction, they will be unable to develop when the available re-           We hypothesize that, in the context of selfing and the sibling com-
   29      sources restrict establishment to only a subset of the competing           petition arena, these isolating mechanisms may additionally help
   30      progeny. Interspecific gene flow is thus directly reduced by the           explain, in an adaptive or nonadaptive manner, the abundance of
   31      combination of selfing and competition. We will henceforth re-             cryptic, host-specific species within fungal pathogen taxa. The
   32      fer to this mechanism as the “sibling competition arena.” The              same hypothesis might be applied to plant taxa in which selfing
   33      use of the term “arena” reflects the requirement for systematic            and the overproduction of seeds are common strategies (Vogler
   34      and intense local competition for successful establishment prior           and Kalisz 2001).
   35      to further development of the zygote. It thus resembles Stearns’                We tested this model using Microbotryum violaceum sensu
   36      selection arena (1987), which proposes early selection of high fit-        lato, a complex of basidiomycete fungi causing anther-smut dis-
   37      ness progeny by maternal choice and resource limitation among              ease on plants of the Caryophyllaceae family (Le Gac et al. 2007a).
   38      abundant offspring. However, several key assumptions are unique            Species of Microbotryum are highly host specific and apparently
   39      to our proposed model, as detailed in Table 1 and below.                   evolve without significant gene flow between closely related taxa
   40            Under this model, the systematic presence of numerous non-           (Le Gac et al. 2007b; Gladieux et al. 2011). Yet, the barriers to
   41      hybrid progeny promotes intense competition, magnifying the                hybridization that have been described for this system thus far are
   42      initial fitness handicap of hybrids. This strong selective sieve re-       insufficient to account for such extensive reproductive isolation.
   43      duces the rate of hybrid production in the population well below           Host and pathogen ranges overlap significantly, with frequent
   44      that predicted solely by the selfing rate and the intrinsic, noncom-       sympatry of diseased hosts (Van Putten et al. 2005; Le Gac et al.
   45      petitive fitness reduction of hybrids relative to nonhybrids. Selfing                   e
                                                                                      2007b; Refr´ gier et al. 2010). These fungi are spread by pollina-
   46      and the production of numerous progeny strongly promote this               tors showing only a partial host specificity that fails to explain
   47      process and may in fact be essential (Table 1, 2.1 and 3.1). First,        the observed species integrity in sympatry (Goulson and Jerrim
   48      these factors generate intense competitive pressure. Second, they          1997; Minder et al. 2007; Van Putten et al. 2007; Karrenberg and
   49      ensure that the production of hybrid progeny is always coupled                                e
                                                                                      Favre 2008; Refr´ gier et al. 2010; Gladieux et al. 2011). More-
   50      with the production of nonhybrid (selfed) progeny, even when               over, mating occurs prior to plant infection, so host specialization
   51      the density of conspecifics is locally reduced. We emphasize that          alone cannot act as a barrier to gene flow, as is suggested for
   52      some degree of divergence between hybridizing genotypes is re-             ascomycete fungi (Le Gac and Giraud 2008; Giraud et al. 2010).
   53      quired, such that hybrid progeny face reduced competitive ability          In addition, there is no evidence for assortative mating in the


                          2      EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012    :959




      1                                                                     S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4     Table 1.   Outline modeling the proposed sibling competition arena. Includes the assumptions and conditions of the sibling competition
      5     arena, their role in reproductive isolation, and their relevance to Microbotryum and other fungal and plant systems.
      6
      7
              Sibling competition arena: components that                  Application to                              Application to
              bar gene flow in combination with selfing                   Microbotryum                                other systems
      8
      9       Required                  1.1 Limited resources:            Occupation of a meristem limited to a       Fungi associated with systemic
     10         1. Competition             space/nutrients restrict          single individual, the unsuccessful        infection: limited number of
     11                                    establishment to a limited        zygotes being unable to persist in         genotypes occupy host
     12                                    number of zygotes at an           the host                                 Plants: limited space, above and
                                           early stage, prior to                                                        below ground, for early
     13
                                           detectable development                                                       establishment/germination
     14
                                        1.2 Intense competition for       The number of diploid teliospores           Fungi: thousands of spores dispersed
     15                                    establishment in the              deposited on a host far exceeds the        to each host individual
     16                                    environment (e.g., host)          number that can ultimately colonize        Plants: number of seeds dispersed
     17                                    through the production of         the host plant                             locally exceeds that which the
     18                                    multiple progeny                                                             environment can support
     19                                 1.3 Mixed population on the       Hybrid hyphae are always produced           Fungi and plants: deposition of
     20                                    required resource: hybrids        simultaneously with nonhybrid              mixed broods of hybrid and
                                           must always compete with          hyphae prior to infection (due to          nonhybrid individuals on/in the
     21
                                           nonhybrids for establishment selfing, 2.1, and to the presence of            same host/environment
     22
                                                                             numerous siblings, 3.1)
     23                                 1.4 Hybrid fitness handicap:      Infection success of hybrids on             Fungi: reduced infection ability of
     24                                    some degree of reduced            parental hosts decreases with              hybrids
     25                                    competitive ability of            genetic distance of species, even        Plants: reduced establishment ability
     26                                    hybrids                           when measured in the absence of            of hybrid seedlings
     27                                                                      competition
     28       Contributes to            2.1 Selfing: ensures the          High intrapromycelial mating                Fungi: selfing (diploid or haploid)
                meeting point 1.3          systematic presence of            (automixis) rates                          frequent
     29
                2. High selfing            nonhybrids, even when                                                      Plants: self-compatibility widespread
     30
                rates                      conspecific density is locally                                               in many plant taxa
     31                                    reduced
     32       Contributes to            3.1 Numerous progeny              Many diploid teliospores of a single        Fungi: thousands of spores produced
     33         meeting point 1.2          commonly compete                  individual deposited on a host             by a single infection
     34         3. Sibling                 intensively, enhancing            plant: nonhybrid and hybrid              Plants: numerous seeds dispersed
     35         competition                competition between               siblings are produced by                   locally
     36                                    siblings, hybrid and              intrapromycelial selfing and
                                           nonhybrid alike                   sporidial mating and compete for
     37
                                                                             infection of the meristem
     38
              Not required              4.1 Selfing and the               The sibling competition arena is likely     Fungi and plants: selfing and
     39         4. Selection to            co-occurrence of numerous         a byproduct of life-history and            overproduction of seeds or spores
     40         avoid                      progeny are not necessarily       reproductive strategies derived from       may serve as adaptations to
     41         hybridization              adaptations for avoiding          ancestry or other selective pressures      facilitate reproduction and
     42                                    hybridization                     (e.g., numerous spores and high            dispersal
     43                                                                      selfing rates may function as
     44                                                                      assurance in host and mate seeking)
     45
     46     form of preferential conjugation with conspecific over heterospe-            ical barriers to hybridization in the field are not consistent with
     47                                                                   e
            cific sporidia (Le Gac et al. 2007b), even in sympatry (Refr´ gier           the near total absence of gene flow in natural populations, which
     48     et al. 2010). Postmating reproductive isolation has been detected,           calls for investigation of additional mechanisms underlying this
     49     with hybrids showing reduced infection ability, but closely related          reproductive isolation (Gladieux et al. 2011).
     50     species can produce viable hybrids (Sloan et al. 2008; de Vienne                  Microbotryum violaceum exhibits a high rate of selfing,
     51     et al. 2009). Overall, the high viability of experimental crosses            which may serve to explain the observed rarity of hybrids in
     52     between closely related pathogen species and the weak ecolog-                                                       e
                                                                                         the field (Giraud et al. 2008b; Refr´ gier et al. 2010; Gladieux
     53



                                                                                                                    EVOLUTION 2012            3
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012    :959




    1      A. K. GIBSON ET AL.

    2
    3

    4      et al. 2011). Conjugation occurs preferentially within the meiotic       tween multiple gene genealogies demonstrated a lack of gene
    5      tetrad (automixis) that comprises the teliospore, the transmissible      flow (Kemler et al. 2006; Le Gac et al. 2007a; Lutz et al. 2008;
    6      stage of Microbotryum (Hood and Antonovics 2000; Giraud et al.           Denchev et al. 2009). The fungus replaces host pollen with dark
    7      2008b; Granberg et al. 2008). This form of selfing is promoted           fungal spores and is transmitted between hosts via pollinators. As
    8      by the development of the meiotic products in a multicellular ba-        previously described, intrapromycelial or intratetrad mating (au-
    9      sidium (e.g., promycelium), where neighboring cells readily con-         tomixis) is the dominant life-history strategy for Microbotryum
   10      jugate. The fungus may also undergo autogamy or outcrossing,                                                                  a
                                                                                    species (Hood and Antonovics 2000, 2004; Sch¨ fer et al. 2010),
   11      primarily through the production and mating of haploid yeast-            although rates of automixis vary between species and populations
   12      like cells (sporidia) (Hood and Antonovics 2000; Giraud 2004;            (Giraud et al. 2005; Granberg et al. 2008). Following mating,
   13      Hood and Antonovics 2004; Giraud et al. 2005, 2008b; Sch¨ fer   a        dikaryotic hyphae form and invade host tissues, with infecting
   14      et al. 2010; Gladieux et al. 2011). Hybridization with other fungal      strains ultimately establishing in a limited region of the host
   15      species occurs through this sporidial mating as well (Le Gac et al.                                                    a
                                                                                    meristems (Audran and Batcho 1982; Sch¨ fer et al. 2010). Upon
   16      2007b; Gladieux et al. 2011). Meiosis and syngamy, through either        flowering, diploid teliospores are formed and expressed in the
   17      intrapromycelial mating and/or sporidial mating, occur following         anthers, for disease transmission via pollinators.
   18      deposition of hundreds of diploid teliospores of Microbotryum                  Natural hybrids have rarely been observed (Gladieux et al.
   19                                                  a
           upon the surface of a new host plant (Sch¨ fer et al. 2010). Some        2011), although Devier et al. (2010) proposed that historic hy-
   20      combination of automixis, autogamy, outcrossing, and hybridiza-          bridization events between moderately distant species were signif-
   21      tion results in the formation of infectious dikayrons on the plant       icant in generating several Microbotryum species. Hybridization
   22      surface. The numerous sibling progeny then compete to occupy             between the closely related species M. lychnidis-dioicae and M.
   23      the host meristem, the seat of Microbotryum infection which is           silenes-dioicae has been most commonly studied. Their respective
   24      limited to colonization by a single diploid individual (Audran           hosts, Silene latifolia and S. dioica, are frequently sympatric (Van
   25                            o                                    a
           and Batcho 1982; L´ pez-Villavicencio et al. 2007; Sch¨ fer et al.                                 e
                                                                                    Putten et al. 2005; Refr´ gier et al. 2010), and hybrids of the two
   26      2010). These life-history traits raise the question of the degree to     fungal species are viable, fertile, and infectious in a laboratory set-
   27      which developmentally promoted selfing and the production of             ting (Van Putten et al. 2003; Le Gac et al. 2007b; de Vienne et al.
   28      multiple progeny may serve as a barrier to interspecific gene flow       2009). In natural populations, however, evidence for hybridization
   29      between sympatric Microbotryum species.                                  is limited. Gladieux et al. (2011) reported only 15 hybrids out of
   30            In this study, we assessed the significance of the combina-        1028 pathogen individuals based upon microsatellite characteriza-
   31      tion of selfing and the sibling competition arena as a mechanism         tion, suggesting strong reproductive isolation in the field between
   32      of reproductive isolation between closely related species of Mi-         these closely related, host-specific pathogen species. Ecological
   33      crobotryum. Artificial inoculations of host plants were designed         isolation, through specialization of habitat or pollinator, may play
   34      to compare rates of overall infection and of hybrid infection in the     some role, but it is far from complete (Goulson and Jerrim 1997;
   35      presence and absence of selfing or the sibling competition arena.        Van Putten et al. 2007). Assortative mating in the form of pref-
   36      First, we tested the effect of developmentally promoted selfing by       erence for conjugation with conspecific sporidial gametes could
   37      intrapromycelial mating on gene flow: we asked if fewer hybrids          not be detected, even between closely related species occurring in
   38      are formed under inoculation with diploid teliospores, undergoing                                                e
                                                                                    sympatry (Le Gac et al. 2007b; Refr´ gier et al. 2010). This find-
   39      mostly automixis, as opposed to inoculation with cultures of hap-        ing holds under simultaneous exposure to both conspecific and
   40      loid sporidia. Second, we tested the effect of the sibling competi-      heterospecific gametes such that hybridization is optional rather
   41      tion arena by comparing hybrid infection rates when hybridization                            e
                                                                                    than forced (Refr´ gier et al. 2010). Selfing has therefore been
   42      was forced to situations where competition with nonhybrids was           proposed as the primary barrier to interspecific gene flow (Giraud
   43      allowed. Under competition, the rate of hybrid infection is ex-          et al. 2008b).
   44      pected to be lower than that based solely upon the noncompetitive
   45      fitness differentials of hybrids and nonhybrids.                         PREPARATION OF FUNGAL STRAINS
   46                                                                               The Microbotryum collections used as inoculum, as well as the
   47                                                                               original location of their collection, are identified in Table 2.
   48      Materials and Methods                                                    Collections were sampled from natural populations of six host
   49      MODEL SYSTEM                                                             species, S. latifolia, S. dioica, S. nutans, S. vulgaris, S. para-
   50      Fungi of the genus Microbotryum (Basidiomycetes: Mi-                     doxa, and Lychnis flos-cuculi. These fungal species are known
   51      crobotryales) cause anther-smut disease on plant hosts of the            respectively as M. lychnidis-dioicae (MvSl), M. silenes-dioicae
   52      Caryophyllaceae. Host-specific lineages have recently been               (MvSd), M. violaceum sensu stricto (MvSn), M. lagerheimii
   53      delineated into species, where the criterion of concordance be-          (MvSv1), M. violaceum sensu lato (MvSp), and M. violaceum


                          4      EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012     :959




      1                                                                      S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4     Table 2.    Identification of the 18 strains of Microbotryum used as inoculum. Presented are the identifying codes, original host plant
      5     from which the strain was isolated, location of the population from which the strain was collected, the date of collection, and the name
      6     of the collector.

      7
              Microbotryum species                        Strain     Plant host species      Original location              Collector
      8
      9       M. lychnidis-dioicae (MvSl)                  729.2     Silene latifolia        Marcillat, France      T. Giraud
     10       M. lychnidis-dioicae (MvSl)                  728.6     Silene latifolia        Foxley Corner, UK      H. Prentice
     11       M. lychnidis-dioicae (MvSl)                  665.2     Silene latifolia        Mt. Biokoyo, Croatia   S. Yakovlev
     12       M. silenes-dioicae (MvSd)                    831.3     Silene dioica           Swiss Alps             S. Karrenberg
              M. silenes-dioicae (MvSd)                    700.3     Silene dioica           Tomtebo, Sweden        B. Giles
     13
              M. silenes-dioicae (MvSd)                    701.3     Silene dioica            a     a o             B. Giles
                                                                                             S¨ vart¨ rn¨ gern, Sweden
     14
              M. violaceum sensu stricto (MvSn)            706.1     Silene nutans                                  M. Fontaine, P. Reignault,
                                                                                             Dunes of l’Escault, France
     15                                                                                                                       e
                                                                                                                       G. Refr´ gier
     16       M. violaceum sensu stricto                   705.3     Silene nutans       Dunes of l’Escault, France M. Fontaine, P. Reignault,
     17         (MvSn)                                                                                                        e
                                                                                                                       G. Refr´ gier
     18       M. violaceum sensu stricto                   719.2     Silene nutans       Dunes of l’Escault, France M. Fontaine, P. Reignault,
     19         (MvSn)                                                                                                        e
                                                                                                                       G. Refr´ gier
     20       M. lagerheimii (MvSv1)                       432.87    Silene vulgaris     Bugnei, Switzerland        B. Devier, D. M. de Vienne,
                                                                                                                       J. Shykoff, L. Salvaudon
     21
              M. lagerheimii (MvSv1)                      C11.1      Silene vulgaris     Swiss Alps                 A. K. Gibson, M. E. Hood
     22
              M. lagerheimii (MvSv1)                       C4.1      Silene vulgaris     Swiss Alps                 A. K. Gibson, M. E. Hood
     23       M. violaceum sensu lato (MvSp)               8A.2      Silene paradoxa     Swiss Alps                 A. K. Gibson, M. E. Hood
     24       M. violaceum sensu lato (MvSp)               4B.1      Silene paradoxa     Swiss Alps                 A. K. Gibson, M. E. Hood
     25       M. violaceum sensu lato (MvSp)              Sp1        Silene paradoxa     Swiss Alps                 A. K. Gibson, M. E. Hood
     26       M. violaceum sensu lato (MvLfc)          6 − 8B        Lychnis flos-cuculi Swiss Alps                 A. K. Gibson, M. E. Hood
     27       M. violaceum sensu lato (MvLfc)          6 − 8E        Lychnis flos-cuculi Swiss Alps                 A. K. Gibson, M. E. Hood
     28       M. violaceum sensu lato (MvLfc)            L F1        Lychnis flos-cuculi Swiss Alps                 A. K. Gibson, M. E. Hood
     29
     30     sensu lato (MvLfc). Abbreviated names indicate the host plant                 dioicae (MvSl), both mating types were available for all three
     31     species. These species of Microbotryum were chosen for their                  strains.
     32     demonstrated ability to hybridize in a laboratory setting with
     33     M. lychnidis-dioicae (MvSl) (Le Gac et al. 2007a). Three field-               INOCULATION AND TREATMENT DESIGN
     34     collected samples were chosen from each pathogen species                      Seeds from a population of S. latifolia in Amherst, Massachusetts,
     35     and were genotyped with microsatellites to verify species                     were chosen for inoculation. This population of S. latifolia is
     36     classification.                                                               known to be highly susceptible to infection. Seeds were sterilized
     37           For each Microbotryum sample, a single anther was taken                 in a solution of deionized water, calcium hypochlorite (12 g/L),
     38     from archived infections and suspended in 200 μL of distilled wa-             and sodium hydroxide (4 g/L) for 20 min, rinsed in a 1:10 dilution
     39     ter. Dilution series of teliospores were spread on GMB2 medium                of the same solution, and left to dry. For germination, seeds were
     40     (Thomas et al. 2003) and incubated at 25◦ C until sporidial colonies          grown on 1% agar-filled petri dishes, with 15 seeds per plate,
     41     derived from single teliospores could be visualized. Sporidial                under fluorescent lights at 22◦ C.
     42     colonies derived from single teliospores were separately subjected                  Seedlings were inoculated upon first emergence of their
     43     to a dilution series and grown on GMB2 medium until colonies                  cotyledons by application of Microbotryum suspensions to the
     44     derived from single haploid sporidia could be visualized. For each            apical meristem. Teliospore inocula were created by suspending
     45     Microbotryum sample, such cultures were isolated and tested for               teliospores from two anthers, one from each of the two fungal
     46     mating type by a conjugation assay to obtain sporidia of both                 species to be crossed, in 1000 μL of distilled water. For sporidial
     47     mating types, a1 and a2 , derived from a common meiosis (Le Gac               inocula, one inoculating loop (10 μL) of each a1 and a2 mating
     48     et al. 2007b). For M. silenes-dioicae (MvSd), M. violaceum sensu              type sporidial culture that had been maintained independently
     49     stricto (MvSn), M. lagerheimii (MvSv1), M. violaceum sensu lato               on GMB2 medium was individually suspended in 300 μL of
     50     (MvSp), and M. violaceum sensu lato (MvLfc), the mating-type                  distilled water. Equal volumes of various individual cultures
     51     ratios were heavily biased (Oudemans et al. 1998; Thomas et al.               were then combined to prepare inocula. Microbotryum crosses
     52     2003), such that samples with only a single mating type were                  were either intraspecific between haploid genotypes of MvSl or
     53     used in certain crosses (Table S1). Importantly, for M. lychnidis-            interspecific between MvSl and one of five other Microbotryum


                                                                                                                     EVOLUTION 2012            5
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012     :959




    1      A. K. GIBSON ET AL.

    2
    3

    4      species. For each species pair, six mating combinations (i.e., a1         moderate under competition with nonhybrids in the absence of
    5      and a2 sporidia) were randomly chosen from among the samples              intrapromycelial selfing (S-mix). Intrapromycelial mating was
    6      available per Microbotryum species. The six inoculum combina-             predicted to reduce rates of hybrid infection to their lowest level,
    7      tions were then applied under four different treatments (S-pair,          even below that projected by intrinsic hybrid fitness and self-
    8      S-mix, T-high, T-low, see below). A total of 144 individual crosses       ing rates. The rate of hybrid infection was therefore expected to
    9      were performed (six pairs of haploid genotypes for each of six            be lower in the presence of selfing by intrapromycelial mating
   10      species pairs × four treatments). These 144 crosses are detailed in       (T-high), and lower still under reduced teliospore concentrations
   11      Table S1. Between 15 and 25 plants were inoculated for each               (T-low), due to less frequent contact between teliospores and a
   12      cross.                                                                    decreased density of hybrids relative to nonhybrids, yielding a
   13           The four treatments were designed to contrast the rates of           stronger sibling competition arena.
   14      selfing with intraspecific outcrossing or interspecific hybridiza-
   15      tion under conditions of either (1) forced hybridization or in-           DATA COLLECTION AND GENOTYPING
   16      traspecific outcrossing, (2) hybridization or selfing possible via        After 2–4 days of incubation, seedlings were transplanted to soil
   17      sporidia (autogamy), or (3) hybridization or selfing possible via         in the greenhouse. Upon flowering, plants were visually assessed
   18      sporidia and intrapromycelial mating (automixis) (Fig. 1). The            for symptoms of anther-smut disease. All flowering plants were
   19      treatment S-pair (for Sporidial pair) comprised inoculation with          removed from the flowerbeds as soon as the first flower appeared
   20      a1 sporidia from one haploid genotype and a2 sporidia from a              to avoid secondary contamination. The number of plants that
   21      second haploid genotype, such that hybridization or intraspecific         flowered for each cross is reported in Table S1. The identity of
   22      outcrossing could be forced. The treatment S-mix consisted of in-         healthy and diseased plants was noted, and 1–2 flowers of diseased
   23      oculation with equal quantities of four distinct sporidial types, the     plants were retained for genetic analysis. Anthers were desiccated
   24      a1 and the a2 from each of two fungal individuals (except when a          on silica gel (Silica gel blue 2–5 mm Prolabo) and stored at 4◦ C.
   25      second mating type was unavailable for the species to be crossed               DNA was extracted, using the Chelex (Bio-Rad) method
   26      with MvSl; see Table S1). Under the S-mix treatment, competi-             (Bucheli et al. 2001), from —one to two anthers from a single
   27      tion between outcrossed/hybrid and nonhybrid (selfed) progeny             flower derived from each diseased sample. Artificial inoculation
   28      could occur because selfing was possible; however, selfing via in-        at the single meristem stage of seedlings largely prevents coex-
   29      trapromycelial mating was absent. In the T-high (for Teliospore,          istence of multiple infections, resulting in systemic infection by
   30      high concentration) and T-low treatments, the inoculum consisted          the single pathogen genotype that persists in subsequently derived
   31      of a suspension of teliospores from two Microbotryum diploids,            meristems (Hood 2003; Gold et al. 2009). Anthers from a single
   32      which allowed intrapromycelial mating. Because teliospores are            flower were therefore considered accurate for genetic typing of
   33      the transmissible stages of Microbotryum and intrapromycelial             an infection under our inoculation protocol. Even in the unlikely
   34      mating is the more common form of mating in nature, these                 event of multiple infections, they would segregate in different
   35      teliospore treatments are most reflective of processes in nat-                                                      o
                                                                                     stems (Hood 2003; Gold et al. 2009; L´ pez-Villavicencio et al.
   36      ural populations (Hood and Antonovics 2000, 2004; Sch¨ fer      a         2011), and our genotyped strains would represent an unbiased
   37      et al. 2010). With teliospores, outcrossing or hybridization was          sample of the infecting strains.
   38      not forced, and competition between outcrossed/hybrid and non-                 Microsatellite genotyping was conducted as described in
   39      hybrid progeny could occur. The T-high and T-low treatments               Giraud (2004). The microsatellite markers SVG8 and SL16 (Gi-
   40      differed in that the T-low inoculum was diluted 100-fold relative         raud et al. 2008c) were used to identify interspecific and intraspe-
   41      to the T-high inoculum to assess the role of teliospore density           cific hybrids, respectively. The majority of strains used in this
   42      in the balance of selfing and outcrossing/hybridization. For each         study were homozygous at the SVG8 microsatellite marker, and
   43      treatment, three intraspecific crosses, consisting of selfing MvSl        different species carried discriminating alleles. Therefore, het-
   44      genotypes from three populations, were conducted. These crosses           erozygotes at this marker indicated infection by a hybrid pathogen.
   45      were designed to obtain a baseline infection rate by selfed progeny       The marker SL16 was additionally used to distinguish two M.
   46      for estimations of the reduction in fitness of hybrid progeny. To         lychnidis-dioicae (MvSl) strains (728.6, 729.2), which were not
   47      simplify the description of the results, the term “hybrid” will           distinguishable using the marker SVG8.
   48      henceforth be used as an umbrella term for both interspecific                  All raw data are deposited in the Dryad repository:
   49      hybrid and intraspecific outcrossed progeny.                              doi:10.5061/dryad.rg148qj4.
   50           Under the hypothesis that selfing, and in particular the de-
   51      velopmental propensity for intrapromycelial mating, plays a role          DATA ANALYSIS
   52      in reproductive isolation, the rate of hybrid infection was pre-          To assess variation in the overall infection rate, logistic regressions
   53      dicted to be highest in the absence of competition (S-pair) and           were performed with JMP 3 (SAS Institute Inc., Cary, NC). For


                          6      EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012     :959




      1                                                                     S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4
      5
      6
      7
      8
      9
     10
     11
     12
     13
     14
     15
     16
     17
     18
     19
     20
     21
     22
     23
     24     Figure 1.   Illustration of the four treatments conducted to compare rates of hybrid infection to assess the combined effect of selfing and
     25     sibling competition as a barrier to hybridization. S-pair: inoculation with the a1 sporidia from one strain and the a2 sporidia from a second
     26     strain. Hybridization or outcrossing is forced. S-mix: inoculation with the a1 and the a2 sporidia from both strains of the cross. Selfing
     27     via sporidial mating and hybridization are both possible. T-high and T-low: inoculation with diploid teliospores from both strains at high
     28     (T-high) or low (T-low) concentrations. Selfing via sporidial and intrapromycelial mating and hybridization are all possible. Each treatment
            is defined in terms of hybridization (forced or by choice) and the possibility of selfing, intrapromycelial mating, and competition. The
     29
            processes of sporidial selfing, intrapromycelial mating, and hybridization are diagrammed.
     30
     31
            each inoculated plant, presence or absence of infection was treated           results were robust to the variation in outcrossing rate observed
     32
            as the dependent variable, and treatment (S-pair, S-mix, T-high,              across different pairings of Microbotryum strains; a rate of 0.30
     33
            T-low) and genetic distance between the fungal parents of the cross           was chosen to reflect the outcrossing rate of a different strain of
     34
            were treated as predictor values. A similar logistic regression was           MvSl examined in a prior study by Giraud et al. (2005). Finally,
     35
            then performed using presence or absence of a hybrid genotype                 to further examine whether the sibling competition arena reduces
     36
            for each infected plant as the dependent variable.                            the rate of hybrid infection below that predicted by fitness reduc-
     37
                 The intrinsic reduction in hybrid fitness relative to nonhy-             tions and observed rates of selfing, the expected hybrid infection
     38
            brids (i.e., without any competition effect) was determined in                rates in the T-high treatment for MvSl × MvSd was computed
     39
            the treatment S-pair by the percent reduction in hybrid infec-                as the expected rate of hybrid infection in the S-mix treatment
     40
            tion rate relative to the mean infection rate of the three selfed             corrected by a factor of [1 − reduction in hybrid infection rate
     41
            crosses (i.e., between a1 and a2 sporidia of the same MvSl sample;            due to intrapromycelial mating]. The reduction in hybrid infec-
     42
            Table S1). The mean infection rate of the three selfed crosses was            tion of MvSl × MvSd due to intrapromycelial mating is reported
     43
            estimated to be 0.617. To test whether the sibling competition                in the third section of the results (76%). A χ2 -test was used to
     44
            arena further reduced the rate of hybrid infection below what was             compare expected and observed rates of hybrid infection in the
     45
            expected based upon intrinsic fitness of hybrids and the observed             T-high treatment.
     46
            rate of intraspecific outcrossing, expected hybrid infection rates
     47
            in the S-mix treatment were computed as the rate of intraspe-
     48
     49
            cific outcrossing in S-mix (MvSl × MvSl) corrected by a factor                Results
            of [1 − reduction in intrinsic hybrid fitness]. The expected and              OVERALL INFECTION RATES
     50
            observed rates of hybrid infection in the S-mix treatment were                The four treatments (S-pair, S-mix, T-high, T-low) differed sig-
     51
            compared using χ2 -tests. The same analysis was also performed                nificantly in overall infection rates (Fig. 2; Table 3). Broadly, the
     52
            using a lower rate of intraspecific outcrossing, to determine if the          S-pair and T-low treatments resulted in lower infection rates than
     53



                                                                                                                     EVOLUTION 2012           7
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012    :959




    1      A. K. GIBSON ET AL.

    2
    3

    4
    5
    6
    7
    8
    9
   10
   11
   12
   13
   14
   15
   16
   17      Figure 2.  Overall infection rate according to treatment, cross, and genetic distance. The proportion of inoculated plants that became
   18      infected is shown for each cross over the four treatments; see Figure 1 and the text for treatment definitions. The genetic distance
   19      between MvSl and the hybridizing pathogen species increases from left to right within each treatment. Error bars show the standard
           error of the proportion.
   20
   21
   22
           the S-mix and T-high treatments (Fig. 2). This is in agreement           became infected with hybrid pathogens (Fig. 3; Table 4). As
   23
           with our expectations that (1) the forced hybridizations in S-pair       expected, the highest hybrid infection rates were observed un-
   24
           would yield lower infection rates than treatments with similar in-       der forced hybridization (S-pair) (Fig. 3). Sporidial mixtures
   25
           oculum concentration but selfing allowed (S-mix and T-high),             (S-mix), where selfing was allowed but there was no possibil-
   26
           and that (2) lower teliospore concentrations (T-low) would               ity of intrapromycelial mating, showed some interspecific hybrid
   27
           yield lower infection rates than higher teliospore concentrations        infection, while the lowest interspecific hybrid infection rates
   28
           (T-high). Genetic distance between the a1 and a2 parents and the         were seen in teliospore treatments (T-high, T-low), where rapid
   29
           interaction of genetic distance with treatment also significantly        intrapromycelial mating promoted selfing. Additionally, the ge-
   30
           affected infection rates (Table 3). As shown in Figure 2, the sig-       netic distance between the two fungal species being crossed was
   31
           nificant interaction term between treatment and genetic distance         a significant predictor of hybrid infection rate (Table 4): hybrid
   32
           likely results from the impact of genetic distance being limited to      fitness decreased with increasing genetic distance between hy-
   33
           the treatment S-pair, where hybridization was forced. Consistent         bridizing species. Interestingly, there was no reduction in fitness
   34
           with previous studies, the fitness of hybrids is shown to decrease       under MvSl × MvSl outcrossing in comparison to selfing. Under
   35
           with genetic distance between hybridizing species (Le Gac et al.         the MvSl × MvSl S-mix treatment, 67% of infections were at-
   36
           2007b). Accordingly, the infection rate was influenced little by         tributed to outcrossed pathogens and only 33% to selfed pathogen
   37
           genetic distance when selfing was allowed, because infections            genotypes. The deviation of this ratio from a 50:50 null hypothesis
   38
           were dominated by selfed progeny (see below). This is consistent         for the comparison of selfing and outcrossing rates was marginally
   39
           with the low hybrid infection rates reported below.                      significant (χ2 = 3.667, df = 1, P = 0.0555).
   40
           HYBRID INFECTION RATES                                                   BARRIERS TO HYBRIDIZATION: SELFING AND
   41
           The four treatments also differed significantly in hybrid infec-         INTRAPROMYCELIAL MATING
   42
           tion rates, that is, the proportion of all inoculated plants that        The significant differences in hybrid infection rates between treat-
   43
                                                                                    ments indicate that the potential for selfing and intrapromycelial
   44
   45      Table 3.  Results of logistic regression for overall infection rate.
   46      Treatment (S-pair, S-mix, T-high, T-low) and genetic distance be-        Table 4.  Results of logistic regression for hybrid infection rate.
           tween crossed species are examined as predictor variables (Whole         Treatment (S-pair, S-mix, T-high, T-low) and genetic distance be-
   47
           model: P < 0.0001, r 2 = 0.2652).                                        tween crossed species are examined as predictor variables (Whole
   48
                                                                                    model: P < 0.0001, r 2 = 0.4256).
   49
            Source                                df      χ2         P-value
   50                                                                                    Source                 df          χ2               P-value
   51       Treatment                             6       98.624     <0.0001
   52       Genetic distance                      2       69.258     <0.0001             Treatment              6           62.748           <0.0001
            Treatment × genetic distance          6       84.237     <0.0001             Genetic distance       2          129.777           <0.0001
   53



                          8      EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012    :959




      1                                                                     S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4
      5
      6
      7
      8
      9
     10
     11
     12
     13
     14
     15
     16
     17
     18     Figure 3.   Hybrid infection rate according to treatment, cross, and genetic distance. The hybrid infection rate is shown for each cross
     19     across all four treatments; see Figure 1 and the text for treatment definitions. Hybrid infection rate indicates the proportion of all
     20     inoculated plants that became infected with hybrid pathogens. The genetic distance between MvSl and the hybridizing pathogen species
     21     increases from left to right within each treatment. Note that MvSl × MvSl crosses represent outcrossing rather than hybridization and
     22     are used for comparison to hybrid crosses. Error bars show the standard error of the proportion.

     23
     24
            mating influences the probability of hybridization between Mi-               BARRIERS TO HYBRIDIZATION: ADDITION OF THE
     25
            crobotryum species (Table 4, Fig. 3). In comparing only treat-               SIBLING COMPETITION ARENA
     26
            ments with forced hybridization (S-pair) to those with the pos-              The intrinsic fitness reduction of hybrid progeny relative to selfed
     27
            sibility of sporidial selfing (S-mix), the potential for selfing             progeny, without accounting for competition, was estimated based
     28
            reduced the rate of hybrid infection by 39% for intraspecific                upon the deviation of infection rates in the forced mating treat-
     29
            crosses (MvSl × MvSl) and by 75% for interspecific hybrid                    ment (S-pair) from the mean infection rate of the three selfed
     30
            crosses with the closest species (MvSl × MvSd). The possibility              MvSl × MvSl crosses, which was calculated to be 0.617. Using
     31
            of selfing reduced the rate of hybrid infection by 77% for the               this estimate of deviation, a significant negative correlation be-
     32
            next most distant cross (MvSl × MvSn) and by 100% for the                    tween intrinsic hybrid fitness and genetic distance between MvSl
     33
            three most distant interspecific crosses (Fig. 3). When compar-              and the hybridizing species was found (r = 0.85, P = 0.031)
     34
            ing treatments in which selfing was possible between sporidia                (S-pair in Fig. 3). The hybrid infection rate in the S-mix treat-
     35
            (S-mix) to those in which intrapromycelial selfing was also pos-             ment for MvSl × MvSd was significantly lower than expected
     36
            sible (T-high), the potential of intrapromycelial mating reduced             based upon the outcrossing rate measured in intraspecific S-mix
     37
            the rate of hybrid infection by 78% for intraspecific crosses and            crosses and the intrinsic reduction in MvSl × MvSd hybrid fit-
     38
            by 76% for the closest interspecific crosses (MvSl × MvSd)                   ness as measured by the S-pair hybrid infection rate (χ2 = 13.333,
     39
            (Fig. 3). This comparison was not informative for the four more              df = 1, P = 0.0003) (Fig. 4). Thus, competition generated by the
     40
            distant interspecific crosses (MvSl × MvSn, MvSl × MvSp,                     presence of selfed progeny further impeded hybrids from suc-
     41
            MvSl × MvSv1, MvSl × MvLfc) because hybrid infection rates                   cessfully infecting beyond their intrinsic fitness reduction and
     42
            were 5% or less under forced hybridization (S-pair) and were                 the selfing rate. This difference remained marginally significant
     43
            already dramatically reduced to 0–1% in the S-mix treatment                  when the intraspecific outcrossing rate was reduced from 0.478
     44
            (Fig. 3).                                                                    to 0.30 to test for robustness to observed variation in outcrossing
     45
                  The concentration of teliospores used in inoculation affected          rates (χ2 = 3.411, df = 1, P = 0.0647) (Giraud et al. 2005).
     46
            hybrid infection rates (Fig. 3, Table 4). As predicted, fewer hy-            The most relevant outcrossing rate here, however, remains the
     47
            brids resulted at low teliospore densities (T-low) relative to high          prior value of 0.478, as it was obtained directly from the exper-
     48
            teliospore densities (T-high) for intraspecific crosses (MvSl ×              imental conditions applied across all treatments. Moreover, the
     49
            MvSl) and the closest interspecific crosses (MvSl × MvSd).                   hybrid infection rate in the T-high treatment for MvSl × MvSd
     50
            This comparison was not possible for more distant interspe-                  was marginally lower than expected based upon the expected
     51
            cific crosses due to the absence of hybrid infections in both                S-mix hybrid infection rate under this intraspecific outcrossing
     52
            treatments.                                                                  rate and the 76% reduction in MvSl × MvSd hybrid infection
     53



                                                                                                                    EVOLUTION 2012            9
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012   :959




    1      A. K. GIBSON ET AL.

    2
    3

    4                                                                              selfed and hybrid progeny, yields nearly complete reproductive
    5                                                                              isolation between species of Microbotryum. Results of artificial
    6                                                                              inoculations confirm our prediction that fewer hybrids success-
    7                                                                              fully establish infection when selfing and intrapromycelial mating
    8                                                                              are possible, in comparison to the forced hybridization treatment.
    9                                                                              Furthermore, our results confirm the prediction that fewer hybrids
   10                                                                              successfully establish infection under competition with nonhybrid
   11                                                                              siblings than expected based upon selfing rates and intrinsic fit-
   12                                                                              ness reductions. Moreover, we demonstrate that low teliospore
   13                                                                              density further reduces rates of hybrid infection, potentially by
   14                                                                              reducing the frequency with which hybrids are generated. Repro-
   15                                                                              ductive isolation in this fungal system thus appears to be strongly
   16                                                                              influenced by selfing and the associated competition between
   17                                                                              nonhybrid and hybrid genotypes, which we refer to as the sibling
   18                                                                              competition arena.
   19                                                                                    Treatments differing in the potential for selfing and in-
   20                                                                              trapromycelial mating yielded significantly different rates of hy-
   21                                                                              brid infection. Intrapromycelial selfing is the most common form
           Figure 4. Reduction of hybrid infection rate resulting from selfing
   22                                                                              of mating in nature and is promoted by the development of phys-
           and the sibling competition arena. For the forced mating treatment
   23                                                                              ically linked gametes upon germination of the teliospore, the
           (S-pair), infection rates are shown for MvSl × MvSl (outcrossed)
   24                                                                              transmissible stage of Microbotryum in natural populations (Hood
           and MvSl × MvSd (hybrid) crosses. For the hybridization choice
   25      treatment (S-mix), the MvSl × MvSd hybrid infection rate is shown                                           a
                                                                                   and Antonovics 2000, 2004; Sch¨ fer et al. 2010). The potential
   26      relative to the expected infection rate based on the results in         for intrapromycelial selfing alone reduced hybrid infection rates
   27      the forced mating treatment (S-pair). Assuming no reduction in          by 76–78% across both the MvSl × MvSl and MvSl × MvSd
   28      hybrid success due to the sibling competition arena, the expected       species pairs, most likely indicating a constant average rate of
   29      hybrid infection rate of MvSl × MvSd in sporidial mixtures (S-mix)
                                                                                   intrapromycelial mating within the MvSl pathogen species. In
           is estimated as the rate of intraspecific outcrossing (MvSl × MvSl)
   30                                                                              natural populations, rates of selfing vary but have been estimated
           in S-mix corrected by the reduction in MvSl × MvSd hybrid fitness
   31                                                                              to be as high as 88–94% (Gladieux et al. 2011), which is higher
           observed under forced hybridization (S-pair). The observed rate
   32                                                                              than the rate observed in our experimental crosses. In nature,
           of hybrid infection is significantly less than that predicted based
   33      upon hybrid fitness and selfing rate alone (χ2 = 14.070, df = 1,          teliospore concentrations and/or the frequency of occurrence of
   34      P = 0.0002), indicating a role for competition between selfed and       different mating partners on a given plant may be lower or more
   35      hybrid progeny (i.e., sibling competition arena). Error bars show       variable than they are under artificial inoculation, influencing es-
   36      the standard error of the proportion for observed rates.                timations of selfing rates. Moreover, Oudemans et al. (1998) and
   37                                                                              Thomas et al. (2003) reported that haplolethal alleles, which limit
   38      rate attributable to intrapromycelial mating (χ2 = 3.190, df = 1,       outcrossing for the sporidia of one mating type, may be as fre-
   39      P = 0.0741).                                                            quent as 100% in some field populations. This was not the case
   40           As noted above, the hybrid infection rates in the S-mix and        for the three MvSl strains used in this study.
   41      T-high treatment for the remaining four crosses were too low for an           It is possible that mechanisms promoting high frequencies
   42      informative statistical test comparing the observed and expected        of selfing represent adaptations to limit gene flow, but identify-
   43      hybrid infection rates. However, the reduction of hybrid infection      ing such adaptations has proven notoriously problematic (Ramsey
   44      rates to zero when competition is introduced is consistent with         et al. 2003; Martin and Willis 2007; Matsubayashi and Katakura
   45      our hypothesis.                                                                       e
                                                                                   2009). Refr´ gier et al. (2010) found no evidence for a mating
   46                                                                              preference for conspecifics in sympatric versus allopatric popula-
   47                                                                              tions of MvSl and MvSd. They attributed this to the presence of a
   48      Discussion                                                              powerful prezygotic isolating mechanism (i.e., selfing) that limits
   49      Selfing in combination with the competition arena is shown here         selection for additional reproductive barriers, such as assortative
   50      to be an important barrier to gene flow between closely related,        mating by mate choice. That study, however, also found no ev-
   51      sympatric species of the phytopathogenic fungi Microbotryum.            idence for adaptive reinforcement in the form of higher rates of
   52      Intrapromycelial mating, a developmentally promoted form of au-         selfing in sympatric populations. Likely, alternative hypotheses
   53      tomixis, in combination with early, intense competition between         for the observed frequencies of selfing are facilitation of mating


                        10       EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012        :959




      1                                                                     S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4     (e.g., reproductive assurance) or acceleration of hyphal formation               development of hybrid individuals. This mechanism requires sys-
      5     (Baker 1955; Lloyd 1992; Granberg et al. 2008); the resulting                    tematic competition between numerous progeny for a limited re-
      6     reproductive isolation may therefore have arisen from selective                  source in which only a small subset of zygotes can ultimately
      7     pressures unrelated to assortative mating.                                       establish (Table 1). In the fungus Microbotryum, this hypothesis
      8           Although its adaptive significance is difficult to clarify, self-          takes the shape of hundreds of sibling, dikaryotic hyphae compet-
      9     ing is shown here to be a promising explanation for the striking                 ing for invasion and infection of a host plant meristem, in which
     10     absence of interspecific gene flow in natural populations of Mi-                 only a single individual can persist. In the event that teliospores
     11     crobotryum. While hybrid pathogens readily form under experi-                    of two different pathogen species are deposited on the surface of a
     12     mental conditions (Le Gac et al. 2007b) and phylogenetic anal-                   single plant, a likely event in sympatric populations, they will self,
     13     ysis supports ancient hybridization events (Le Gac et al. 2007a;                 via intrapromycelial mating, and hybridize, via sporidial mating.
     14     Devier et al. 2010), there is negligible evidence of hybridiza-                  Consistent with previous findings, we assume here that hybrid
     15     tion events occurring in the field, even between closely related,                infection is initially limited by the propensity to self at a cer-
     16                               e
            sympatric species (Refr´ gier et al. 2010; Gladieux et al. 2011).                tain rate and by intrinsic fitness reductions in hybrid progeny but
     17     The prominent role of selfing as a barrier to gene flow is con-                  not by preferential conjugation with conspecific over heterospe-
     18     sistent with studies in plants, where selfing is commonly found                                                              e
                                                                                             cific gametes (Le Gac et al. 2007b; Refr´ gier et al. 2010). Our
     19     to facilitate reproductive isolation under sympatric or parapatric               results indicate that the remaining hybrid progeny then fail to
     20     conditions. Matallana et al. (2010) recently reported that self-                 survive the sibling competition arena due to direct competition
     21     compatible species of the Bromeliaceae family are more likely to                 with their superior, nonhybrid siblings. This finding strongly sug-
     22     overlap with close relatives in their geographic range or blooming               gests a role for early, postzygotic competition between selfed
     23     period than are self-incompatible species. Likewise, populations                 and hybrid progeny in depressing gene flow between sympatric
     24     in parapatry and sympatry with close relatives have been shown                   pathogen species well below that predicted solely by the intrinsic
     25     to have significantly higher rates of selfing than allopatric popu-              fitness handicap of hybrid genotypes. Importantly, this mecha-
     26     lations. This has been interpreted as defending against “gametic                 nism of reproductive isolation is a genuine isolating barrier in
     27     wastage” on unfit hybrid progeny and promoting local adaptation                  that it limits gene flow in the event of hybridization but not in-
     28     (Antonovics 1968; Petit et al. 1997; Fishman and Wyatt 1999;                     traspecific outcrossing, assuming that outcrossed progeny face no
     29     Grossenbacher and Whittall 2011). Although selfing has been                      reduction in fitness relative to selfed progeny. This assumption is
     30     studied almost exclusively in plants, it offers a potentially impor-             supported by our finding that, within a single pathogen species
     31     tant mechanism of reproductive isolation in fungi, where speci-                  (MvSl), outcrossed progeny were not less fit than selfed progeny
     32     ation processes are largely undefined. Mating systems of plant                   under competition: outcrossed progeny in fact represented a larger
     33     and human fungal pathogens are increasingly found to incorpo-                    proportion of infections in the S-mix treatment than did selfed
     34     rate versions of sexual reproduction that favor inbreeding over                  progeny.
     35     outcrossing in a manner proposed to facilitate host adaptation                         The sibling competition arena bears much similarity to other
     36     (Giraud et al. 2010; Heitman 2010).                                              mechanisms of early postzygotic reproductive isolation and off-
     37           Selfing as a barrier to gene flow has been objected to on                  spring selection in plants and animals, such as Stearns’ “selection
     38     the grounds that mating systems with high levels of selfing re-                  arena” (1987). Hauser and Siegismund (2000) hypothesized that, Q2
     39     duce gene flow within the same species to the same degree that                   in the plant species S. nutans, strong discrimination against in-
     40     they reduce gene flow between species (Coyne and Orr 2004,                       bred progeny could result from the decreased survival of selfed
     41     p. 212). Selfing has, nonetheless, been acknowledged to be a ma-                 seed when developing in competition with outcrossed seeds in
     42     jor, indirect contributor to the speciation process: small effective             the same fruit. Lively and Johnson (1994) proposed that brood-
     43     population sizes, facilitation of the founder effect, and chromoso-              ing organisms are more susceptible to the invasion of partheno-
     44     mal rearrangements are all correlates of selfing that may promote                genetic mutants due to a mother’s ability to establish a “selection
     45     speciation (Lewis 1966; Coyne and Orr 2004, p. 212). Here, we                    arena” to weed low fitness progeny from her brood, either actively
     46     propose another consequence of selfing: it can function to strongly              or through sibling competition. The perspectives of these previ-
     47     reduce the likelihood of propagation of hybrid lineages when self-               ous works differ from that of the model proposed here, which
     48     ing is associated with the production of multiple progeny in a                   is tailored for the unique biology of fungal pathogens and some
     49     competition arena, such that hybrid genotypes must always com-                   plants and bears directly upon reproductive isolation. They do,
     50     pete with superior nonhybrid genotypes prior to establishment                    however, propose mechanisms of early zygotic selection, through
     51     and early development (Table 1).                                                 competition or parental control that may similarly facilitate repro-
     52           The sibling competition arena hypothesizes that early, in-                 ductive isolation and eliminate hybrid fitness. Such studies have
     53     tense competition between hybrids and nonhybrids will prevent                    been predominantly conducted in animal and plant systems. By


                                                                                                                      EVOLUTION 2012          11
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)            1-14-2012        :959




    1      A. K. GIBSON ET AL.

    2
    3

    4      extension to the realm of fungi, developmental competition may                        de Vienne, D. M., G. Refregier, M. E. Hood, A. Guigue, B. Devier, E. Vercken,
    5      be found to be a broadly generalizable mechanism for ensuring                              C. Smadja, A. Deseille, and T. Giraud. 2009. Hybrid sterility and invi-
                                                                                                      ability in the parasitic fungal species complex Microbotryum. J. Evol.
    6      the early elimination of low fitness offspring (Bruggeman et al.
                                                                                                      Biol. 22:683–698.
    7      2004). Likewise, the sibling competition arena should be broadly                      Denchev, C. M., T. Giraud, and M. E. Hood. 2009. Three new species of
    8      generalizable to plants, where selfing and early sibling compe-                            anthericolous smut fungi on Caryophyllaceae. Mycol. Balc. 6:79–84.
    9      tition for establishment in a saturated environment are frequent                      Devier, B., G. Aguileta, M. E. Hood, and T. Giraud. 2010. Using phylogenies
   10      (Table 1).                                                                                 of pheromone receptor genes in the Microbotryum violaceum species
                                                                                                      complex to investigate possible speciation by hybridization. Mycologia
   11            We demonstrate here that selfing in the plant fungal                                 102:689–696.
   12      pathogen Microbotryum dramatically reduces hybridization be-                          Fishman, L., and R. Wyatt. 1999. Pollinator-mediated competition, reproduc-
   13      tween closely related species by two mechanisms: directly by                               tive character displacement, and the evolution of selfing in Arenaria
   14      intrapromycelial mating and indirectly by the sibling competition                          uniflora (Caryophyllaceae). Evolution 53:1723–1733.
                                                                                                 Gibson, A. K., M. E. Hood, and T. Giraud. 2011. Data from: sibling competi-
   15      arena’s influence on competitive exclusion. The magnitude of the
                                                                                                      tion arena: selfing and a competition arena can combine to constitute a
   16      reduction in gene flow attributable to selfing appears to strongly                         barrier to gene flow in sympatry. Dryad Data Repository.                     Q3
   17      diminish natural selection for other prezygotic isolating barriers                    Giraud, T. 2004. Patterns of within population dispersal and mating of the
   18      between sympatric species of Microbotryum, consistent with ob-                             fungus Microbotryum violaceum parasitising the plant Silene latifolia.
   19                                                                                                 Heredity 93:559–565.
                                                   e
           servation of natural populations (Refr´ gier et al. 2010). These
                                                                                                 Giraud, T., O. Jonot, and J. A. Shykoff. 2005. Selfing propensity under choice
   20      results, and the predominance of mating systems that facilitate                            conditions in a parasitic fungus, Microbotryum violaceum, and parame-
   21      inbreeding in fungal pathogen and plant species (Billiard et al.                           ters influencing infection success in artificial inoculations. Int. J. Plant
   22      2011), support the inclusion of mating system and selfing rate                             Sci. 166:649–657.
   23      as critical components of reproductive isolation in the study of                      Giraud, T., J. Enjalbert, E. Fournier, C. Dutech, and F. Delmotte. 2008a.
                                                                                                      Population genetics of fungal diseases of plants. Parasite 15:449–454.
   24      speciation.
                                                                                                 Giraud, T., R. Yockteng, M. Lopez-Villavicencio, G. Refregier, and M. E.
   25                                                                                                 Hood. 2008b. Mating system of the anther smut fungus Microbotryum
   26      ACKNOWLEDGMENTS                                                                            violaceum: selfing under heterothallism. Eukaryot. Cell 7:765–775.
   27      We thank Odile Jonot, Odylle Cudelou, Lionel Saunois, Pierre                          Giraud, T., R. Yockteng, S. Marthey, H. Chiapello, O. Jonot, M. Lopez-
                                       e
           Gladieux, Guislaine Refr´ gier, Michael Fontaine, Benjamin Devier,                         Villavicencio, D. M. de Vienne, M. E. Hood, G. Refregier, A. Gendrault-
   28
           Amandine Dubois, Christian Raquin, Jenny Soupama, and Laetitia                             Jacquemard, et al. 2008c. Isolation of 60 polymorphic microsatellite loci
   29      Giraud for help in the greenhouse, and Odile Jonot for her contribution to                 in EST libraries of four sibling species of the phytopathogenic fungal
   30                                                               e
           genotyping. We thank Jacqui Shykoff and Guislaine Refr´ gier for insight-                  complex Microbotryum. Mol. Ecol. Res. 8:387–392.
   31      ful discussions on statistical analyses. We thank the collectors cited in             Giraud, T., P. Gladieux, and S. Gavrilets. 2010. Linking the emergence of
           Table 1. We thank Chris Jiggins and two anonymous referees for com-                        fungal plant diseases with ecological speciation. Trends Ecol. Evol.
   32
           ments on an earlier version of this manuscript. TG acknowledges the                        25:387–395.
   33      grants ANR 06-BLAN-0201, ANR 07-BDIV-003, and ANR-09-0064-                            Gladieux, P., E. Vercken, M. Fontaine, M. E. Hood, O. Jonot, A. Couloux,
   34      01, and AKG acknowledges a grant from the Fulbright Program and la                         and T. Giraud. 2011. Maintenance of fungal pathogen species that are
   35                              e
           Commission franco-am´ ricaine. MEH acknowledges the grant NSF-DEB                          specialized to different hosts: allopatric divergence and introgression
           0747222.                                                                                   through secondary contact. Mol. Biol. Evol. 28:459–471.
   36
                                                                                                 Gold, A., T. Giraud, and M. Hood. 2009. Within-host competitive ex-
   37
           LITERATURE CITED                                                                           clusion among species of the anther smut pathogen. BMC Ecol. 9.
   38                                                                                                 doi:10.1186/1472-6785-1189-1111.                                             Q4
           Allard, R. W. 1975. Mating system and microevolution. Genetics 79:115–126.
   39      Antonovics, J. 1968. Evolution in closely adjacent plant populations V. Evo-          Goulson, D., and K. Jerrim. 1997. Maintenance of the species boundary be-
   40            lution of self-fertility. Heredity 23:219–238.                                       tween Silene dioica and S. latifolia (red and white campion). Oikos
   41      Audran, J., and M. Batcho. 1982. Comportement d’Ustilago violacea (Pers.)                  79:115–126.
                 Rouss. au sein des tissus vegetatifs et reproducteurs du Silene dioica (L.)     Granberg, A., U. Carlsson-Graner, P. Arnqvist, and B. E. Giles. 2008. Variation
   42
                 Clairv. Rev. Cytol. Biol. Veg. 5:59–63.                                              in breeding system traits within and among populations of Microbotryum
   43      Baker, H. G. 1955. Self-compatibility and establishment after ‘long-distance’              violaceum on Silene dioica. Int. J. Plant Sci. 169:293–303.
   44            dispersal. Evolution 9:347–349.                                                 Grossenbacher, D. L., and J. B. Whittall. 2011. Increased floral divergence in
   45                        o
           Billiard, S., M. L´ pez-Villavicencio, B. Devier, M. E. Hood, C. Fairhead, and             sympatric monkeyflowers. Evolution 65:2712–2718.
                 T. Giraud. 2011. Having sex, yes, but with whom? Inferences from fungi          Hauser, T. P., and H. R. Siegismund. 2000. Inbreeding and outbreeding effects
   46
                 on the evolution of anisogamy and mating types. Biol. Rev. 86:421–442.               on pollen fitness and zygote survival in Silene nutans (Caryophyllaceae).
   47                                                                                                 J. Evol. Biol. 13:446–454.
           Bruggeman, J., A. J. M. Debets, and R. F. Hoekstra. 2004. Selection arena in
   48            Aspergillus nidulans. Fungal Genet. Biol. 41:181–188.                           Heitman, J. 2010. Evolution of eukaryotic microbial pathogens via covert
   49      Bucheli, E., B. Gautschi, and J. Shykoff. 2001. Differences in population                  sexual reproduction. Cell Host Microbe 8:86–99.
                 structure of the anther smut fungus Microbotryum violaceum on two               Hood, M. E. 2003. Dynamics of multiple infection and within-host competi-
   50
                 closely related host species, Silene latifiolia and S. dioica. Mol. Ecol.            tion by the anther-smut pathogen. Am. Nat. 162:122–133.
   51
                 10:285–294.                                                                     Hood, M. E., and J. Antonovics. 2000. Intratetrad mating, heterozygosity, and
   52      Coyne, J. A., and H. A. Orr. 2004. Speciation. Sinauer Associates, Inc.,                   the maintenance of deleterious alleles in Microbotryum violaceum ( =
   53            Sunderland, MA.                                                                      Ustilago violacea). Heredity 85:231–241.



                          12       EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)          1-14-2012          :959




      1                                                                             S E L F I N G A N D S I B L I N G C O M P E T I T I O N P R E V E N T G E N E F L OW

      2
      3

      4     ———. 2004. Mating within the meiotic tetrad and the maintenance of ge-                    Minder, A., C. Rothenbuehler, and A. Widmer. 2007. Genetic structure of
      5          nomic heterozygosity. Genetics 166:1751–1759.                                               hybrid zones between Silene latifolia and Silene dioica (Caryophyl-
            Husband, B. C., and H. A. Sabara. 2004. Reproductive isolation between                           laceae): evidence for introgressive hybridization. Mol. Ecol. 16:2504–
      6
                 autotetraploids and their diploid progenitors in fireweed, Chamerion                        2516.
      7
                 angustifolium (Onagraceae). New Phytol. 161:703–713.                                 Oudemans, P. V., H. M. Alexander, J. Antonovics, S. Altizer, P. H. Thrall, and
      8     Karrenberg, S., and A. Favre. 2008. Genetic and ecological differentiation                       L. Rose. 1998. The distribution of mating-type bias in natural popula-
      9          in the hybridizing campions Silene dioica and S. latifolia. Evolution                       tions of the anther-smut Ustilago violacea on Silene alba in Virginia.
     10          62:763–773.                                                                                 Mycologia 90:372–381.
            Kay, K. M. 2006. Reproductive isolation between two closely related                       Petit, C., P. Lesbros, X. J. Ge, and J. D. Thompson. 1997. Variation in
     11
                 hummingbird-pollinated neotropical gingers. Evolution 60:538–552.                           flowering phenology and selfing rate across a contact zone between
     12     Kemler, M., M. Goker, F. Oberwinkler, and D. Begerow. 2006. Implications                         diploid and tetraploid Arrhenatherum elatius (Poaceae). Heredity 79:31–
     13          of molecular characters for the phylogeny of the Microbotryaceae (Ba-                       40.
     14          sidiomycota : Urediniomycetes). BMC Evol. Biol. 6. doi: 10.1186/1471-                Ramsey, J., H. D. Bradshaw, and D. W. Schemske. 2003. Components of
                 2148-1186-1135.                                                                             reproductive isolation between the monkeyflowers Mimulus lewisii and
     15
            Le Gac, M., and T. Giraud. 2008. Existence of a pattern of reproductive                          M. cardinalis (Phrymaceae). Evolution 57:1520–1534.
     16          character displacement in Homobasidiomycota but not in Ascomycota.                        e
                                                                                                      Refr´ gier, G., M. E. Hood, and T. Giraud. 2010. No evidence of reproductive
     17          J. Evol. Biol. 21:761–772.                                                                  character displacement between two sister fungal species causing anther
     18     Le Gac, M., M. E. Hood, E. Fournier, and T. Giraud. 2007a. Phylogenetic                          smut in Silene. Int. J. Plant Sci. 171:847–859.
     19          evidence of host-specific cryptic species in the anther smut fungus.                  a               e
                                                                                                      S´ nchez-Guill´ n, R., M. Wellenreuther, and A. Cordero Rivera. 2011. Strong
                 Evolution 61:15–26.                                                                         asymmetry in the relative strengths of prezygotic and postzygotic bar-
     20
            Le Gac, M., M. E. Hood, and T. Giraud. 2007b. Evolution of reproductive                          riers between two damselfly sister species. Evolution. Accepted article:
     21          isolation within a parasitic fungal species complex. Evolution 61:1781–                     doi: 10.1111/j.158-5646.2011.01469.x.                                        Q5
     22          1787.                                                                                    a
                                                                                                      Sch¨ fer, A. M., M. Kemler, R. Bauer, and D. Begerow. 2010. The illustrated
     23     Levin, D. A. 2010. Environment-enhanced self-fertilization: implications for                     life cycle of Microbotryum on the host plant Silene latifolia. Can. J. Bot.
                 niche shifts in adjacent populations. J. Ecol. 98:1276–1283.                                88:875–885.
     24
            Lewis, H. 1966. Speciation in flowering plants. Science 152:167–172.                      Sloan, D. B., T. Giraud, and M. E. Hood. 2008. Maximized virulence in a
     25     Lively, C. M., and S. G. Johnson. 1994. Brooding and the evolution of                            sterilizing pathogen: the anther-smut fungus and its co-evolved hosts. J.
     26          parthenogenesis: strategy models and evidence from aquatic inverte-                         Evol. Biol. 21:1544–1554.
     27          brates. Proc. R. Soc. B 256:89–95.                                                   Stearns, S. C. 1987. The selection-arena hypothesis. Experientia Suppl.
            Lloyd, D. G. 1992. Self- and cross-fertilization in plants. II. The selection of                 55:337–349.
     28
                 self- fertilization. Int. J. Plant Sci. 153:370–380.                                 Thomas, A., J. Shykoff, O. Jonot, and T. Giraud. 2003. Sex-ratio bias in
     29
             o
            L´ pez-Villavicencio, M., O. Jonot, A. Coantic, M. E. Hood, J. Enjalbert, and                    populations of the phytopathogenic fungus Microbotryum violaceum
     30          T. Giraud. 2007. Multiple infections by the anther smut pathogen are                        from several host species. Int. J. Plant Sci. 164:641–647.
     31          frequent and involve related strains. PLoS Pathog. 3:e176.                           Van Putten, W. F., A. Biere, and J. M. M. Van Damme. 2003. Intraspecific
     32      o
            L´ pez-Villavicencio, M., F. Courjol, A. K. Gibson, M. E. Hood, O. Jonot, J.                     competition and mating between fungal strains of the anther smut Mi-
                 A. Shykoff, and T. Giraud. 2011. Competition, cooperation among kin,                        crobotryum violaceum from the host plants Silene latifolia and S. dioica.
     33
                 and virulence in multiple infections. Evolution 65:1357–1366.                               Evolution 57:766–776.
     34     Lutz, M., M. Platek, M. Kemler, A. Chlebicki, and F. Oberwinkler. 2008.                   ———. 2005. Host-related genetic differentiation in the anther smut fungus
     35          Anther smuts of Caryophyllaceae: molecular analyses reveal further                          Microbotryum violaceum in sympatric, parapatric, and allopatric popu-
     36          new species. Mycol. Res. 112:1280–1296.                                                     lations of two host species Silene latifolia and S. dioica. J. Evol. Biol.
            Martin, N. H., and J. H. Willis. 2007. Ecological divergence associated with                     18:203–212.
     37
                 mating system causes nearly complete reproductive isolation between                  Van Putten, W. F., J. A. Elzinga, and A. Biere. 2007. Host fidelity of the polli-
     38          sympatric Mimulus species. Evolution 61:68–82.                                              nator guilds of Silene dioica and Silene latifolia: possible consequences
     39     Matallana, G., M. A. S. Godinho, F. A. G. Guilherme, M. Belisario, T. S.                         for sympatric host race differentiation of a vectored plant disease. Int. J.
     40          Coser, and T. Wendt. 2010. Breeding systems of Bromeliaceae species:                        Plant Sci. 168:421–434.
     41          evolution of selfing in the context of sympatric occurrence. Plant Syst.             Vogler, D. W., and S. Kalisz. 2001. Sex among the flowers: the distribution of
                 Evol. 289:57–65.                                                                            plant mating systems. Evolution 55:202–204.
     42
            Matsubayashi, K. W., and H. Katakura. 2009. Contribution of multiple iso-
     43          lating barriers to reproductive isolation between a pair of phytophagous
     44          ladybird beetles. Evolution 63:2563–2580.                                                                                     Associate Editor: C. Jiggins
     45
     46
     47
     48
     49
     50
     51
     52
     53



                                                                                                                                     EVOLUTION 2012               13
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012   :959




    1      A. K. GIBSON ET AL.

    2
    3

    4
    5
             Supporting Information
    6
             The following supporting information is available for this article:
    7
    8        Table S1. Identification of the 144 crosses performed.
    9
   10        Supporting Information may be found in the online version of this article.
   11        Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting information supplied by the
   12        authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
   13
   14
   15
   16
   17
   18
   19
   20
   21
   22
   23
   24
   25
   26
   27
   28
   29
   30
   31
   32
   33
   34
   35
   36
   37
   38
   39
   40
   41
   42
   43
   44
   45
   46
   47
   48
   49
   50
   51
   52
   53



                        14       EVOLUTION 2012
evo_1563   evo2009v2.1.cls (1994/07/13 v1.2u Standard LaTeX class)   1-14-2012   :959




            Queries
            Q1 Author: Please check all affiliations as typeset for correctness also provide the zip code in affiliations 4 and 5.
            Q2 Author: “Hauser et al. (2000)” has been changed to “Hauser and Siegismund (2000)” so that this citation matches the
               Reference List. Please confirm that this is correct.
            Q3 Wiley-Blackwell: Please check “Gibson et al. (2011)” as typeset for correctness.
            Q4 Author: Please provide the page range in Gold et al. (2009) and Kemler et al. (2006).
                                                                    a            e
            Q5 Author: Please provide the volume and page range in S´ nchez-Guill´ n et al. (2011).

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:12
posted:9/7/2012
language:Unknown
pages:15