vol. 196, no. 1 the american naturalist july 2020 Correlated Evolution of Sex Allocation and Mating System in Wrasses and Parrotfishes Jennifer R. Hodge,* Francesco Santini, and Peter C. Wainwright Department of Evolution and Ecology, University of California Davis, 1 Shields Avenue, Davis, California 95616 Submitted June 8, 2019; Accepted December 5, 2019; Electronically published May XX, 2020 Online enhancements: supplemental PDF, Excel tables. Dryad data: https://doi.org/10.25338/B8GC91. q1 abstract: In accordance with predictions of the size-advantage sequential hermaphroditism is known only in teleosts (Todd model, comparative evidence confirms that protogynous sex change et al. 2016), where sex change can be male to female (pro- is lost when mating behavior is characterized by weak size advan- tandry), female to male (protogyny), or serial bidirectional. tage. However, we lack comparative evidence supporting the adap- Each form of sequential hermaphroditism has evolved mul- tive significance of sex change. Specifically, it remains unclear whether tiple times within teleosts, demonstrating the lability of fish increasing male size advantage induces transitions to protogynous sex sex-determination mechanisms (Smith 1975; Charnov 1982; change across species, as it can within species. We show that in wrasses fi Policansky 1982; Mank et al. 2006). and parrot shes (Labridae) the evolution of protogynous sex change fi is correlated with polygynous mating, and that the degree of male The dominant theory describing the adaptive signi - size advantage expressed by polygynous species influences transitions cance of sequential hermaphroditism is the size-advantage between different types of protogynous sex change. Phylogenetic re- model (SAM; Ghiselin 1969; Warner 1975; Leigh et al. constructions reveal strikingly similar patterns of sex allocation and 1976; Charnov 1982). The model contends that sex change mating system evolution with comparable lability. Despite the plastic- is favored when the rate of increase in reproductive value ity of sex-determination mechanisms in labrids, transitions trend to- with size and age differs between the sexes. Correspondingly, ward monandry (all males derived from sex-changed females), with gonochorism (the existence of separate, fixed sexes) is pre- all observed losses of protogyny accounted for by shifts in the timing fi of sex change to prematuration. Likewise, transitions in mating system dicted when size-speci cmaleandfemalereproductiveout- trend from the ancestral condition of lek-like polygyny toward greater comes do not differ (Warner 1975; Muñoz and Warner male size advantage, characteristic of haremic polygyny. The results of 2003, 2004). A range of complex, interacting factors ca- our comparative analyses are among the first to confirm the adaptive pable of contributing to differences in reproductive value q2 significance of sex change as described by the size-advantage model. between the sexes, including aspects of population demog- Keywords: Labridae, polygynous mating, phylogenetic comparative raphy, life history, social system, and the local environ- method, protogynous hermaphroditism, size-advantage model, ment, have been integrated into the SAM (Charnov 1982; Teleostei. Warner 1988; reviewed in Avise and Mank 2009). Of these, mating behavior has emerged as an important determinant of size-related differential reproductive outcomes (Shapiro Introduction 1987; Ross 1990; Munday et al. 2006a). This is because Sequential hermaphroditism is a reproductive strategy with certain mating systems are also contingent on male size multiple evolutionary origins distributed sporadically across advantage. the tree of life (Policansky 1982; Sadovy de Mitcheson and For example, protogyny, the most prevalent form of se- fi Liu 2008). It is characterized by a change in the functional quential hermaphroditism in shes (Sadovy de Mitcheson expression of sex, from one to the other. Among vertebrates, and Liu 2008; Todd et al. 2016), is predicted to be adap- tive when reproductive value increases with size faster in males than in females (Warner 1975; Leigh et al. 1976). * Corresponding author. Present address: Department of Biological Sciences, Protogynous species can be either monandric, in which case Clemson University, 108 Jordan Hall, Clemson, South Carolina 29634; email: all males are derived from sex-changed females, or diandric, [email protected]. in which case males are either born into the population or ORCIDs: Hodge, https://orcid.org/0000-0003-1603-8559; Santini, https:// fi fi orcid.org/0000-0002-4526-7855. derived from sex-changed females (see g. 1 for de nitions). Am. Nat. 2020. Vol. 196, pp. 000–000. q 2020 by The University of Chicago. Sex-based size asymmetry is also characteristic of polygy- 0003-0147/2020/19601-59300$15.00. All rights reserved. nous mating, where males use their size advantage to monop- DOI: 10.1086/708764 olize access to females by guarding them or the resources 59300.proof.3d 1 Achorn International 05/11/20 21:15 000 The American Naturalist Mating system Male size Sex allocation advantage Haremic polygyny Monandric protogyny Terminal phase males monopolize one or more All males are derived via sex change from females within a permanent territory. functional females (secondary males). Lek-like polygyny Diandric protogyny Terminal phase males establish temporary Males are either born into the population territories visited by females for the purposes (primary males) or derived from sex-changed of reproduction. females (secondary males). Promiscuity Gonochorism Terminal phase males do not defend territories Individuals reproduce exclusively as either with the purpose of attracting mates. male or female throughout their lives. Figure 1: Types of mating and sex allocation systems expressed by labrid fishes, their definitions, and predicted associations with the degree of male size advantage. As male size advantage increases from promiscuous mating to lek-like polygyny, selection is predicted to favor tran- sition from gonochorism to diandric protogyny. Likewise, as male size advantage increases from lek-like to haremic polygyny, selection should favor transitions from diandric to monandric protogyny. Definitions are from Warner and Robertson (1978), Colin and Bell (1991), and Sadovy de Mitcheson and Liu (2008). on which they depend (Ghiselin 1969; Taborsky 1998). This Population demographic studies and observations of behavior results in large males having a reproductive advan- mating dynamics within species provide empirical support tage over females and small males, thereby enabling selec- for the influence of mating behavior on sex allocation as tion for protogynous sex change (Warner 1975, 1984, 1988). predicted by the SAM (Robertson 1972; Warner and Rob- Polygynous mating can include haremic systems, where a ertson 1978; Warner and Hoffman 1980a; Fukuda et al. single male monopolizes and mates with one (Pitcher 1993) 2017). Phylogenetic comparative studies have shown that or more females within a defined permanent territory, and the loss of protogyny is contingent on weak size advan- lek-like systems, where males establish temporary territories tage (Kazancıoğlu and Alonzo 2010) and group spawning that are visited by females for the purpose of reproduction. (Erisman et al. 2009). However, no comparative studies As mating becomes more promiscuous, sperm competition have supported predictions about the adaptive significance increases, and the reproductive advantage of large males de- of protogynous sex change. Specifically, we lack compara- creases because of the dilution of gametes by other males, con- tive evidence showing that as size advantage increases, and sequently reducing selection for protogyny (Warner 1975). large males have greater opportunity to monopolize mating, When large males have strong social control over fe- protogynous sex change evolves. Many aspects of an organ- males (i.e., male size reflects dominance or social status), ism’s biology, demography, and ecology have the potential as in haremic systems, monandric protogyny is predicted to affect the expression of complex traits such as mating (Robertson and Choat 1974; Robertson and Warner 1978; system and sex allocation. Moreover, behavioral traits tend Warner and Robertson 1978; Warner 1984; Nemtzov 1985). to be more evolutionarily labile than life-history traits In lek-like systems where the social control of large males (Blomberg et al. 2003). It remains unknown whether vari- is reduced, primary males are able to realize reproductive ation in the degree of male size-advantage characteristic of success, and selection should favor diandric protogyny (Rob- specific types of polygynous mating induces evolutionary ertson and Choat 1974; Emlen and Oring 1977; Robertson transitions to and within types of protogynous sex change and Warner 1978; Warner and Robertson 1978). Corre- in the context of other influential factors. Do the effects of spondingly, gonochorism is predicted when males lose so- mating behavior on sex allocation observed within species cial control, as in promiscuous mating behaviors such as scale up to macroevolutionary patterns? group spawning (Robertson and Warner 1978; Warner The wrasses and parrotfishes, along with cales and weed- 1984; Hoffman 1985). Qualitative assessments of mating whitings (Labridae) provide an ideal opportunity to evalu- behavior support its role as a primary determinant of the ate the evolutionary synergy between sex allocation and degree of size advantage